The Routledge Handbook of Essence in Philosophy (Routledge Handbooks in Philosophy) [1 ed.] 0367442795, 9780367442798

Essences have been assigned important but controversial explanatory roles in philosophical, scientific, and social theor

122 34 8MB

English Pages 532 [533] Year 2024

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
CONTENTS
CONTRIBUTORS
ACKNOWLEDGEMENTS
INTRODUCTION
1 Essence
2 Why This Volume, and Why Now?
3 Primer
4 Future Directions
Note
References
PART 1 History
Introduction
1 ANCIENT
1.1 Socrates and Plato
1.2 Aristotle: The Logic of Essence
1.3 Aristotle: Essence and Form
1.4 Hellenistic Period and Late Antiquity
1.5 Related Topics
Notes
References
2 MEDIEVAL
2.1 Basic Issues: The Essential vs. the Accidental, Whether There are Individual Essences or Only Species Essences, the Knowabil
2.2 Two Debates: Is Matter Part of the Essences of Material Substances? Are Essence and Existence Really Distinct?
2.3 Conclusion
2.4 Related Topics
Notes
References
3 MODERN
3.1 The Traditional Notion of Essence
3.2 Early Modern Traditionalists: Descartes, Hobbes, Cavendish, and Conway
3.3 First Countermovement: Spinoza and Leibniz
3.4 Second Countermovement: Locke
3.5 Related Topics
Notes
References
4 PRAGMATISM
4.1 Pragmatism and Traditional Philosophical Terminology
Essence:
Essentialism:
Realist Essentialism:
Anti-realist Essentialism:
Anti-essentialism (active):
essentialism
Anti-essentialism (passive):
essentialism
4.2 Peirce
4.3 James
4.4 Dewey
4.5 Summary and Conclusion
Notes
References
5 CONTEMPORARY (PHENOMENOLOGICAL TRADITION)
5.1 Introduction
5.2 Essences and Their Connections
5.3 Essence, Generality, Modality and the Apriori
5.4 Essence and Internal Relations
5.5 Rôles
5.6 Essence Guides
5.7 The Functionalization of Knowledge of Essence
5.8 Essence Intrudes
5.9 Knowledge of Essence and Bird’s Eye Views
5.10 The Intuition of General Essences (Ideas) and Conceptual Ladders
5.11 The Essence of Intuition of Essence
5.12 Stigmatic Intuition vs Bird’s Eye Views
5.13 The Intuition of Individual Essences
5.14 Essence vs Family Resemblances
5.15 Related Topics
Notes
References
6 CONTEMPORARY (ANALYTIC TRADITION)
6.1 Introduction
6.2 Essence and the Birth of Analytic Philosophy
6.3 Logical Empiricism and Post-Logical Empiricism
6.4 The Way Towards the Modal View of Essence
6.5 Fine
6.6 The Post-Finean Landscape in Metaphysics
6.7 Related Topics
Acknowledgments
Notes
Bibliography
PART 2 Essence and Essentialisms
Introduction
7 MODAL CONCEPTIONS OF ESSENCE
7.1 Introduction
7.2 Modality
N(
P(
Nis
Pis
Ncaptures
7.3 Classical Modalism
N(
N NN
N(
7.4 Identity Across Worlds
NPt
7.5
Essentialism
N(
N
7.6 Fine’s Objection to Classical Modalism
7.7 Sophisticated Modalism
N(
7.8 Hybrid Modalism
7.9 Conclusions
7.10 Related Topics
Notes
References
8 NON-MODAL CONCEPTIONS OF ESSENCE
8.1 Introduction
8.2 The Modal Conception and Its Problems
8.3 Non-Modal Accounts
8.4 Assessment of the Non-Modal Accounts
8.5 Accounting for Metaphysical Necessity in Terms of Essence
8.6 Objections Against Accounts of Necessity in Terms of Essence
NE.
Source
NT. Source
Grounding
NT
Grounding
Source.
NT
8.7 Further Topics, Further Developments
Reduction
Grounding
8.8 Related Topics
Acknowledgments
Notes
Bibliography
9 ESSENCES OF INDIVIDUALS
9.1 Individuals, Kinds, and Essentialism
9.2 Individual Essences
9.3 Sortal Essentialism
9.4 Conclusion
9.5 Further Readings
9.6 Related Topics
Notes
References
10 NATURAL KIND ESSENTIALISM
10.1 Introduction: What Are Natural Kinds?
10.2 The Kripke–Putnam Framework
10.3 Intrinsic vs. Extrinsic Essences
10.4 Microstructural Essences
10.5 Refined Natural Kind Essentialism
10.6 Related Topics
Acknowledgments
Notes
References
11 ORIGIN ESSENTIALISM
11.1 Introduction
11.2 A Closer Look at Origin Essentialism
11.3 Expansion Arguments for Origin Essentialism
11.4 The Paradox of Flexible Origin Essentialism
11.5 Related Topics
Notes
References
12 SCIENTIFIC ESSENTIALISM
12.1 Introduction
(i)
(ii)
(iii)
(iv)
(v)
(vi)
(vii)
(i)–(vii).
12.2 Ellis’ System: Scientific Essentialism
(i)–(vii),
12.3 Assessing Scientific Essentialism
12.4 Related Topics
Notes
References
13 DISPOSITIONAL ESSENTIALISM
13.1 Introduction
13.2 The Identities of Properties
13.3 The Laws of Nature
13.4 Property Dualism
13.5 The Hybrid Position
13.6 The Challenges from Modern Physics
13.7 Related Topics
Notes
Further Readings
References
14 THE EPISTEMOLOGY OF ESSENCE
14.1 Preliminaries
14.2 Kripke
14.3 Lowe
14.4 Hale
14.5 Oderberg
14.6 Elder
14.7 Kment
14.8 Related Topics
Notes
References
15 LANGUAGE OF ESSENCE
15.1 Representational Essentialism
Hidden and Not Directly Observable:
Inductive Potential and Homogeneity:
Heritability and Mutability:
Explanation:
Discrete Category Boundaries:
15.2 Linguistic Constructions and Expression Types
15.3 Broader Connections: Metaphysics, Methodology, and Social Political Projects
Notes
Works Cited
16 LOGIC OF ESSENCE
16.1 The Language of Essence
LE)
LE
16.2 The Proof Theory of Essence
LE
LE.
K
RN
LE,
K-
RN
LE
RN
K
LE
RN
LE
RN
Rigidity
N-Rigidity
Rigidity
5-
Monotonicity
LE
LE).
Chaining
LE
Localization
LE,
LE;
T.
4.
5.
T-
4
5
4
5-
4
5
LE
4
5.
4
5
4
4-
5.
5
4,
S5,
B
RN
S4
RN
LE
16.3 Semantics
5.
5-
5
RN
16.4 Summary and Further Work
16.5 Related Topics
Notes
References
PART 3 Applications
Introduction
17 ARTIFACTS, ARTWORKS, AND SOCIAL OBJECTS
17.1 Introduction
17.2 Essentialism
17.3 The Essence of Artifacts
17.4 The Essence of Social Objects
17.5 Conclusion
Related topics:
Notes
References
18 BIOLOGICAL SPECIES
18.1 A Primer on Contemporary Biological Systematics
18.2 Essentialist Statements Involving Species
18.3 Species as Natural Kinds vs. Species as Individuals
18.4 From the Death of Species Essences to the Rebirth of Species?
18.5 Related Topics
Notes
References
19 IDENTITY, PERSISTENCE, AND INDIVIDUATION
19.1 Persistence and Persistence Conditions
19.2 Persistence Conditions and Their Riddles
19.3 Essentialist Takes on the Riddles
19.4 Individuation and Individual Essence
Related Topics
Notes
References
20 ESSENCE, GROUNDING, AND EXPLANATION
20.1 Introduction
20.2 Unity: Essence and Grounding as Closely Unified
20.3 Supplementation
20.4 Independence: Grounding and Essence as Two Distinct Modes of Explanation
20.5 Conclusion
Related Topics
Notes
References
21 THE “REDUCTION” OF NECESSITY TO NON-MODAL ESSENCE
21.1 Introduction
21.2 Explanatory Connections Between Essence and Metaphysical Modality
21.3 Explanatory Strategies
21.4 Conclusion
21.5 Related Topics
Notes
References
22 PERSONS
22.1 Three Lockean Ideas
22.2 Distinctive Cognitive Capacities
22.3 Accountability
22.4 Identity
22.5 Philosophical Implications
22.6 Concluding Remarks
22.7 Cross-References
Notes
References
23 PSYCHIATRIC KINDS
23.1 Introduction
23.2 The Mechanistic Property Cluster (MPC) View and Its Problems
23.3 Alternatives to the MPC View
23.4 Hacking on Interactive Kinds
23.5 Psychiatric Kinds as Socially Constructed
23.6 Conclusion
23.7 Related Topics
Notes
References
24 RACE
24.1 Racial Essentialism is a Folk Theory
24.2 Racial Essentialism is False
24.3 Alternatives to Folk Essentialism: Skepticism
24.4 Denying E1: Nonessentialist Race Without Simple Necessary and Sufficient Features
24.5 Denying E2: Nonessentialist Race Without Intrinsic and Natural Features
24.6 Is Nonessentialism Less Bad?
24.7 Combatting Racial Essentialism With
Anti-Essentialism
24.8 Summing Up
24.9 Related Topics
Acknowledgments
Notes
References
25 SEX AND GENDER
25.1 Introduction
25.2 Sex and Essentialism
25.3 The Sex/Gender Distinction: First Pass
25.4 Sex and Anti-Essentialism
25.5 The Sex/Gender Distinction: Second Pass
25.6 Context, Dispositions & Identity: Essentialist & Anti-Essentialist Perspectives on Gender
25.7 The Problems
25.8 Related Topics
Notes
References
26 SOCIAL JUSTICE
26.1 Introduction
26.2 Social Groups and Essences
26.3 Essences and the Unification of Social Categories
26.4 Social Justice and Individual Essences
26.5 Ideological Oppression and Essentializing Beliefs
26.6 Definitions and Essences
26.7 Conclusion
26.8 Related Topics
Notes
References
27 UNITY
27.1 Essence and Unity in Aristotle
27.2 Essence and Unity in Contemporary Metaphysics
27.3 Social Kinds and Uniessentialism
27.4 Related Topics
Notes
References
28 ETHICAL VALUE
28.1 Virtue Theory and the Essence of Human Beings
28.2 Ethics of Enhancement and the Essence of Being Human
28.3 Disability Ethics and Essence
28.4 The Essence and Ethics of Killing
28.5 Essence of Identity and Moral Responsibility
28.6 Essence and the Ethics of Sexual Activity
28.7 Essence and the Ethics of Sex and Gender
28.8 Ontological Oppression
28.9 Conclusion
28.10 Related Topics

Notes
Works Cited
PART 4 Anti-essentialist Challenges
Introduction
29 QUINE ON ESSENCE
29.1
Modality
29.2
Modality
29.3 The Non-strict Modalities
29.4 Essence
29.5 Related Topics
Notes
References
30 CONVENTIONALISM
30.1 Introduction
30.2 Traditional Conventionalism
30.3 Traditional Conventionalism and Necessity
30.4 Conventionalism Post-
Necessity
GPI Schema:
WATER:
P1:
C1:
P2:
C2:
GPI Conventional Schema:
30.5 Other Views
30.6 Potential Challenges to and Questions for a Conventionalist Theory of Essence
(A)
(LINK)
30.7 Conclusion
30.8 Related Topics
Notes
References
31 SOCIAL CONSTRUCTION
31.1 Introduction
31.2 Social Construction
31.3 Essence
31.4 Social Constructionist Challenges to Essentialism
31.5 Social Kind Essentialism
31.6 Conclusion
31.7 Related Topics
Notes
References
32 CONFERRALISM
32.1 Realism vs. Anti-realism about Essence
32.2 What Is a Conferred Property?
32.3 Conferralism about Essence
Essentiality
Ideal Subjects, i.e., Ideal Versions of Us Concept Users
Their Finding It Inconceivable that the Object not Have the Property
At the Limit of Enquiry into How We Use Concepts
32.4 Critical Questions
32.5 Conclusion
32.6 Related Topics
Notes
References
33 WITTGENSTEIN
33.1 Part I: The
in
something—
something
is
nature
33.2 Part II. The Philosophical Investigations
32.3 Related Topics in this Volume
Notes
References
INDEX
Recommend Papers

The Routledge Handbook of Essence in Philosophy (Routledge Handbooks in Philosophy) [1 ed.]
 0367442795, 9780367442798

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

THE ROUTLEDGE HANDBOOK OF ESSENCE IN PHILOSOPHY

Essences have been assigned important but controversial explanatory roles in philosophical, scientific, and social theorizing. Is it possible for the same organism to be first a caterpillar and then a butterfly? Is it impossible for a human being to transform into an insect like Gregor Samsa does in Kafka’s The Metamorphosis? Is it impossible for Lot’s wife to survive being turned into a pillar of salt? Traditionally, essences (or natures) have been thought to help answer such central questions about existence, identity, persistence, and modality. These questions are not only of great philosophical interest, they also are of great interest to society at large. This Handbook surveys the state of the art on essence. Core issues about essence are discussed in 33 chapters, all of them written exclusively for this volume by leading experts. They are organized into the following four major parts, each with its own introduction that provides a summary and comparison of the part’s chapters: • • • •

History Essence and Essentialisms: Themes and Variations Applications Anti-Essentialist Challenges

The volume is accessible enough for students while also providing enough details to make it a valuable reference for researchers. While the notion of essence has been targeted for sustained criticisms since antiquity, recent work has renewed interest in the topic. This Handbook explains and synthesizes much of this current interest, placing essence within its historical context and drawing connections to many contemporary areas of philosophy as well as to scholarly work in other disciplines. With cross-references in each chapter and a comprehensive index, The Routledge Handbook of Essence in Philosophy is a useful resource and essential reading for anyone, whether in or out of academic philosophy, seeking clarification on one of philosophy’s most distinctive and notorious notions. Kathrin Koslicki is Professor of Theoretical Philosophy at the University of Neuchâtel. Koslicki’s research interests in philosophy lie mainly in metaphysics, the philosophy of language and ancient Greek philosophy, particularly Aristotle. In her two books (The Structure of Objects, Oxford UP, 2008; and Form, Matter, Substance, Oxford UP, 2018), she defends a neo-Aristotelian analysis of concrete particular objects as compounds of matter (hulē) and form (morphē). Michael J. Raven is Professor of Philosophy at the University of Victoria and Affiliate Professor of Philosophy at the University of Washington. He is a co-founder and co-editor-in-chief of the journal Metaphysics, and also a co-founder and steering committee member of the Metaphysics Collaborative.

ROUTLEDGE HANDBOOKS IN PHILOSOPHY

Routledge Handbooks in Philosophy are state-of-the-art surveys of emerging, newly refreshed, and important fields in philosophy, providing accessible yet thorough assessments of key problems, themes, thinkers, and recent developments in research. All chapters for each volume are specially commissioned, and written by leading scholars in the field. Carefully edited and organized, Routledge Handbooks in Philosophy provide indispensable reference tools for students and researchers seeking a comprehensive overview of new and exciting topics in philosophy. They are also valuable teaching resources as accompaniments to textbooks, anthologies, and research-orientated publications. Also available: THE ROUTLEDGE HANDBOOK OF THE PHILOSOPHY OF BIODIVERSITY Edited by Justin Garson, Anya Plutynski, and Sahotra Sarkar THE ROUTLEDGE HANDBOOK OF PHILOSOPHY OF THE SOCIAL MIND Edited by Julian Kiverstein THE ROUTLEDGE HANDBOOK OF PHILOSOPHY OF EMPATHY Edited by Heidi Maibom ROUTLEDGE HANDBOOK OF EPISTEMIC CONTEXTUALISM Edited by Jonathan Jenkins Ichikawa THE ROUTLEDGE HANDBOOK OF BRENTANO AND THE BRENTANO SCHOOL Edited by Uriah Kriegel THE ROUTLEDGE HANDBOOK OF EPISTEMIC INJUSTICE Edited by Ian James Kidd, José Medina and Gaile Pohlhaus THE ROUTLEDGE HANDBOOK OF PHILOSOPHY OF TEMPORAL EXPERIENCE Edited by Ian Philips For more information about this series, please visit: https://www.routledge.com/RoutledgeHandbooks-in-Philosophy/book-series/RHP

THE ROUTLEDGE HANDBOOK OF ESSENCE IN PHILOSOPHY

Edited by Kathrin Koslicki and Michael J. Raven

Cover image: Rose Choi, Flight, 2023 First published 2024 by Routledge 605 Third Avenue, New York, NY 10158 and by Routledge 4 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2024 selection and editorial matter Kathrin Koslicki and Michael J. Raven; individual chapters, the contributors The right of Kathrin Koslicki and Michael J. Raven to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN: 978-0-367-44279-8 (hbk) ISBN: 978-1-032-74659-3 (pbk) ISBN: 978-1-003-00875-0 (ebk) DOI: 10.4324/9781003008750 Typeset in Sabon by MPS Limited, Dehradun

To Kit

CONTENTS

Notes on Contributors Acknowledgements

xi xv

Introduction Kathrin Koslicki and Michael J. Raven

1

PART 1

History

15

1 Ancient Marko Malink

19

2 Medieval Gloria Frost

30

3 Modern Anat Schechtman

41

4 Pragmatism Andrew Howat

53

5 Contemporary (Phenomenological Tradition) Kevin Mulligan

67

6 Contemporary (Analytic Tradition) Robert Michels

84

vii

Contents PART 2

Essence and Essentialisms: Themes and Variations

101

7 Modal Conceptions of Essence Alessandro Torza

105

8 Non-Modal Conceptions of Essence Fabrice Correia

124

9 Essences of Individuals Marco Marabello

143

10 Natural Kind Essentialism Tuomas E. Tahko

156

11 Origin Essentialism Teresa Robertson Ishii

169

12 Scientific Essentialism Travis Dumsday

181

13 Dispositional Essentialism Ka Ho Lam

194

14 The Epistemology of Essence Antonella Mallozzi

209

15 Language of Essence Katherine Ritchie

226

16 Logic of Essence Jon Erling Litland

240

PART 3

Applications

257

17 Artifacts, Artworks, and Social Objects Asya Passinsky

261

18 Biological Species Ingo Brigandt

276

viii

Contents

19 Identity, Persistence, and Individuation Maria Scarpati

291

20 Essence, Grounding, and Explanation David Mark Kovacs

305

21 The “Reduction” of Necessity to Non-Modal Essence Kathrin Koslicki

319

22 Persons Annina Loets

333

23 Psychiatric Kinds Danielle Brown

347

24 Race Ron Mallon

361

25 Sex and Gender Esther Rosario

375

26 Social Justice Natalie Stoljar

388

27 Unity Charlotte Witt

401

28 Ethical Value Stavroula Glezakos and Julie Tannenbaum

410

PART 4

Anti-Essentialist Challenges

425

29 Quine on Essence Kit Fine

427

30 Conventionalism Jonathan Livingstone-Banks and Alan Sidelle

437

31 Social Construction Aaron M. Griffith

455

ix

Contents

32 Conferralism Anand Jayprakash Vaidya and Michael Wallner

472

33 Wittgenstein Arata Hamawaki

487

Index

503

x

CONTRIBUTORS

Ingo Brigandt is Professor of Philosophy and Canada Research Chair in Philosophy of Biology at the University of Alberta. His publications span across philosophy of biology, socially relevant philosophy of science, science education, and evolutionary developmental biology. Danielle Brown is a PhD candidate at the University of Alberta. She is interested in the philosophy of psychiatry and the factors influencing the construction of psychiatric ontologies. Fabrice Correia is Professor of Philosophy at the University of Geneva. His main interests are in metaphysics and philosophical logic. He has published extensively on essence, grounding, ontological dependence, identity, modality and time, and has recently discovered the wonders of theorizing about location. Travis Dumsday is Associate Professor of Philosophy at Concordia University of Edmonton. He is the author of Dispositionalism and the Metaphysics of Science (Cambridge University Press, 2019) and Assisted Suicide in Canada: Moral, Legal, and Policy Considerations (University of British Columbia Press, 2021). Kit Fine is University Professor and Silver Professor of Philosophy and Mathematics at New York University. He writes on metaphysics, logic and philosophy of language and his most recent book is Vagueness: A Global Approach, published by Oxford University Press. Gloria Frost is Professor of Philosophy at the University of St. Thomas (St. Paul, MN). She is the author of Aquinas on Efficient Causation and Causal Powers (Cambridge: Cambridge University Press, 2022) and articles which have appeared in journals such as Journal of the History of Philosophy, Ergo and Oxford Studies in Medieval Philosophy. Stavroula Glezakos is Associate Professor of Philosophy at Wake Forest University. Her work is focused on issues at the intersection of philosophy of language, ethics, and social philosophy, and has been published in Linguistics and Philosophy and volumes with MIT and Oxford University Presses.

xi

Contributors

Aaron M. Griffith is an Associate Professor of Philosophy at William & Mary in Williamsburg, Virginia USA. His research is on truth, truthmaking, grounding, social ontology, and the philosophy of race. His work has appeared in Philosophy and Phenomenological Research, Philosophical Studies, Synthese, Erkenntnis, and other journals. Arata Hamawaki is Associate Professor in Philosophy, Auburn University. He has published papers on Kant, Wittgenstein, Aesthetics, Epistemology, and the Nature of Judgment. Andrew Howat is a Professor of Philosophy at California State University, Fullerton. His research focuses on truth, realism, and classical pragmatism (esp. C.S. Peirce), and he has published in journals such as Philosophical Studies, Erkenntnis, Synthese, and Philosophia. Kathrin Koslicki is Professor of Theoretical Philosophy at the University of Neuchâtel. Koslicki’s research interests in philosophy lie mainly in metaphysics, the philosophy of language and ancient Greek philosophy, particularly Aristotle. In her two books (The Structure of Objects, Oxford University Press, 2008; and Form, Matter, Substance, Oxford University Press, 2018), she defends a neo-Aristotelian analysis of concrete particular objects as compounds of matter (hulē) and form (morphē). David Mark Kovacs is a Senior Lecturer at Tel Aviv University, Israel. Ka Ho Lam is currently a lecturer in the Department of Philosophy at the University of Hong Kong. Jon Erling Litland is Associate Professor of Philosophy at the University of Texas at Austin and professorial fellow at the University of Oslo. He has recently published papers about mathematical structuralism, the grounds of identity, propositional fineness of grain, as well as about the puzzles of ground. Jonathan Livingstone-Banks is a lecturer and senior researcher in evidence-based healthcare at the University of Oxford. His philosophical research interests include essentialism, realism/ anti-realism debates, and the philosophy of evidence-based healthcare. Annina Loets holds positions as Assistant Professor at Humboldt-University, Berlin and the University of Wisconsin, Madison. Her research interests lie mainly in metaphysics and the philosophy of language. She has recently published papers on the metaphysics and semantics of qua-qualification, the logic of normality, as well as on the semantics of ability ascriptions. Marko Malink is Professor of Philosophy and Classics at New York University (USA). He is the author of Aristotle’s Modal Syllogistic (Harvard University Press, 2013), and various articles on Aristotle’s logic and metaphysics as well as on the history of logic. Ron Mallon is a Professor of Philosophy and Director of the Philosophy-NeurosciencePsychology program at Washington University in Saint Louis. His research is in social metaphysics and moral psychology. He is the author of the Construction of Human Kinds (2016, Oxford).

xii

Contributors

Antonella Mallozzi is Assistant Professor of Philosophy at Providence College and an associate editor of Analysis Reviews. She works primarily in metaphysics and epistemology. She is the editor of the special issue of Synthese “New Directions in the Epistemology of Modality”, a coauthor of the Stanford Encyclopedia of Philosophy entry on “The Epistemology of Modality”, and the author of several recent articles on essence, explanation, the epistemology of imagination and the a priori. Marco Marabello is a doctoral assistant at the University of Neuchâtel, Switzerland. Robert Michels, since 2022, is FCT assistant researcher and a member of LanCog at the University of Lisbon, Portugal. His research interests include essence, modality, indeterminacy, the laws of nature, and the history of analytic philosophy. Kevin Mulligan is professor of philosophy, University of Italian Switzerland, Director of Research, Institute of Philosophical Studies, Lugano, and honorary professor of analytic philosophy, University of Geneva. He is a member of the Royal Swedish Academy of Letters and the Academia Europaea. Asya Passinsky is an Assistant Professor of Philosophy at Central European University in Vienna, Austria. She specializes in metaphysics, social philosophy, and feminist philosophy, and her work has appeared in academic journals such as Philosophers’ Imprint, Journal of the American Philosophical Association, Inquiry, and Journal of Social Ontology. Katherine Ritchie is an Associate Professor at the University of California, Irvine in the United States. Her research focuses on the metaphysics of social groups and the ways we represent social groups mentally and linguistically. Her articles have appeared in journals including Australasian Journal of Philosophy, Cognitive Psychology, Ethics, Mind & Language, Philosophers’ Imprint, Philosophical Studies, and Philosophy and Phenomenological Research, among others. Teresa Robertson Ishii is Professor of Philosophy at the University of California, Santa Barbara. Her primary academic interests are in analytic metaphysics and the philosophy of language. Charmed by puzzles and paradoxes, she has recently published articles on hylomorphism and the Russell-Myhill paradox; kinds and a puzzle about them; and plenitudinarianism and the paradox of flexible origin essentialism. Esther Rosario is a lecturer in Philosophy and Women’s, Gender, and Sexuality Studies at Dartmouth College. Her research focuses on the metaphysics of sex and gender. She is especially interested in sex and gender kinds. Her current research centers on sex as a biological kind. Her previous research includes conceptual engineering projects for sex, gender, and race concepts. She is also interested in social metaphysics broadly construed, the metaphysics of science, the philosophy of biology, science and values, and metaphilosophy. Maria Scarpati got her PhD at the University of Neuchâtel (Switzerland) with a thesis on metaphysical haecceitism and individual essences. She then worked as a post-doctoral researcher at the University of Geneva and University of Oxford, focusing on the topic of unity relations.

xiii

Contributors

Anat Schechtman is Associate Professor of Philosophy at the University of Texas at Austin. She has written on Descartes, Spinoza, Leibniz, and Locke, publishing in Mind, Philosophical Review, Journal of the History of Philosophy, and elsewhere. She is currently writing a monograph about early modern notions of infinity. Alan Sidelle is Professor of Philosophy at the University of Wisconsin-Madison. He is author of Necessity, Essence and Individuation: A Defense of Conventionalism (1989). Natalie Stoljar is Professor of Philosophy and Director of the Institute for Gender, Sexuality and Feminist Studies at McGill University. She holds a joint appointment in the Department of Equity, Ethics and Policy, Faculty of Medicine. Her research expertise is in social and political philosophy, feminist philosophy, and the philosophy of law. She has published numerous articles and book chapters and is co-editor (with C. Mackenzie) of Relational Autonomy. Feminist Perspectives on Autonomy, Agency and the Social Self (OUP 2000) and (with K. Voigt) of Autonomy and Equality. Relational Approaches (Routledge 2021). Tuomas E. Tahko is Professor of Metaphysics of Science at the Department of Philosophy, University of Bristol, UK. He is the author of Unity of Science (Cambridge UP, 2021) and An Introduction to Metametaphysics (Cambridge UP, 2015), and editor of Contemporary Aristotelian Metaphysics (Cambridge UP, 2012). Julie Tannenbaum is Associate Professor of Philosophy at Pomona College. Her research in ethical theory and bioethics has been published in Ethics, Bioethics, Hastings Center Report, Oxford Studies in Metaethics, Oxford Studies in Normative Ethics, Philosophical Studies, and The Stanford Encyclopedia of Philosophy. Alessandro Torza is Associate Professor of Philosophy at the University of Parma. He is author of Indeterminacy in the World (Cambridge UP), as well as a number of articles on indeterminacy, modality, and (meta)metaphysics at large. Anand Jayprakash Vaidya is Professor of Philosophy at San Jose State University and occasional Director of the Center for Comparative Philosophy. Since 2005 his research has focused on the epistemology of modality, essence, and grounding. Michael Wallner is “Universitätsassistent” (Assistant Professor – fixed term) at the Department of Philosophy at the University of Graz, Austria. He is predominantly working in metaphysics, epistemology, and phenomenology and focuses on issues in modality, essence, grounding, explanation, and fundamentality. Charlotte Witt is Professor of Philosophy at the University of New Hampshire. She is the author of Substance and Essence in Aristotle and Ways of Being in Aristotle’s Metaphysics, both published by Cornell University Press. She is the co-editor of A Mind of One’s Own: Feminist Essays on Reason and Objectivity and three other collections including Adoption Matters: Philosophical and Feminist Essays. Recent work includes a monograph The Metaphysics of Gender (Oxford 2011) and an edited volume Feminist Metaphysics: Explorations in the Ontology of Sex, Gender and the Self (Springer 2011).

xiv

ACKNOWLEDGEMENTS

The initial idea for this volume emerged from our collaborations on the Metaphysics Collaborative (formerly, the Canadian Metaphysics Collaborative). We are grateful for the advice from three anonymous referees, as well as Andrew Beck and Marc Stratton at Routledge, for helping this idea take shape. There were plans to further develop and integrate the chapters by having authors workshop early drafts with one another in Switzerland. The COVID-19 pandemic disrupted these plans (along with so many others). We pivoted, inviting authors to workshop their early drafts virtually in June 2021. We are grateful for their flexibility and for their cooperative and productive engagement with each other’s work. Editing this volume gave us the opportunity to recruit and engage with an extraordinary group of authors. Working with, and learning from, them has been a delight. We thank them all for their excellent contributions to this volume. Marco Marabello and Emilie Pagano provided vital organizational assistance in preparing the volume. We thank them both for their outstanding work. The volume is the culmination of the 5-year project “The Essence of Anti-Essentialism”, which was funded by an Insight Grant from the Social Sciences and Humanities Research Council of Canada (SSHRC). We are grateful to SSHRC for supporting this volume directly, and for its indirect support for Emilie Pagano’s postdoctoral fellowship. Rose Choi provided the original cover art for this volume. Whatever aesthetic virtues the volume has we owe to her artistic talents. We thank her for her contribution. Our final acknowledgement is to Kit Fine. It would be hard to overstate his influence on this volume, or indeed on contemporary metaphysics. Even those who have only glanced at the recent literature on essence will know how profoundly his work has shaped it. Without it, there would be no place in the literature for this volume. Kit also helped foster the volume itself. He provided detailed feedback on many chapters. And he encouraged us throughout. The volume was improved by his generosity and support. But even with all that said, our gratitude goes far beyond. An anecdote may help convey how. Mike wrote his first philosophy term paper on “Essence and Modality” back in spring 2000. Although still a newcomer to philosophy, he recognized it as an iconoclastic work (at least, back then) by a maverick philosopher (still to this day). That paper had an enormous impact on him: not just for its originality and its contrast with the prevailing wisdom, but also in a more personally

xv

Acknowledgements

philosophical way. It was only two decades later while writing an encyclopedia entry “Kit Fine” that he realized just how deeply Kit’s philosophical approach had affected him. That approach is manifest in “Essence and Modality”: inspiringly respectful in its attempt to understand its topic properly for what it is, while boldly resolute in resisting cynical shortcuts that diminish or distort its topic. It was an aspirational model of philosophical virtue. Those fortunate enough to have philosophized with Kit know that he is too.

xvi

INTRODUCTION Kathrin Koslicki and Michael J. Raven

1 Essence Is it possible for one and the same organism to be first a caterpillar and then a butterfly? Is it impossible for a human being to undergo the sort of transformation we encounter in Franz Kafka’s The Metamorphosis, in which Gregor Samsa, a traveling salesman, wakes up one morning only to find himself transformed into a giant insect? Is it impossible for Lot’s wife to survive being turned into a pillar of salt? What (if anything) determines when an entity comes into or goes out of existence? More generally, in virtue of what (if anything) is an entity identical to itself at a time or over time? Through what sorts of changes (if any) can an entity persist? What (if anything) explains what features an entity must have in order for it to exist at all, and which features are compatible with its nature but optional (i.e., its modal profile)? Traditionally, essences (or natures) have been taken to play important roles in helping to answer such central questions about existence, identity, persistence, and modality. Unless responses to these questions can be found, key questions about the world will go unanswered.1 If a certain feature is essential to an entity, then (at the very least) the entity in question cannot exist without having the feature in question; a feature that is not essential to an entity is accidental to it. For example, if Socrates is essentially human but merely accidentally a philosopher, then Socrates must be human if he is to exist at all, but he could have existed without being a philosopher. For example, Socrates could have been an architect. An essence can either be a general or kind essence or an individual essence: a general or kind essence is shared by members of a single kind; an individual essence is distinctive of an individual instance of a certain kind. There has been significant debate about whether kinds are associated with essences and whether it is part of the meaning of certain of our words to connect us to these essences (Bird & Tobin 2012; Khalidi 2013; Koslicki 2008; Kripke 1971, 1980; Putnam 1962, 1975). A statement of the essence, or “real definition”, aims to tell us what it is to be the entity in question; such statements contrast with “nominal definitions” which aim to capture instead what a certain word means or what a certain concept represents. Strictly speaking, nominal definitions may also be viewed as a special case of real definitions; only the entity that is being defined, in this case, is a linguistic expression or a concept, rather than a non-linguistic or non-conceptual entity that is denoted by the expression or concept in question. To illustrate,

DOI: 10.4324/9781003008750-1

1

Kathrin Koslicki and Michael J. Raven

one might think that the statement (real definition), “Gold is a chemical element with atomic number 79”, captures the essence of, or what it is to be, gold, while the statement (nominal definition), “A centaur is a mythological creature with the upper body of a human being and the lower body of a horse”, captures what the English word, “centaur”, means or what the concept, centaur, represents. Essentialists hold that grasping the essence of, or what it is to be, an entity is crucial to an accurate understanding of the world, independently of how we represent or conceptualize the entity in question and independently of what we value relative to a particular context. Antiessentialists, by contrast, interpret our apparent essence ascriptions as in some way reflective of our specifically human interests, goals, or practices, rather than of the world “as such”, i.e., as it would be if we somehow managed to subtract from it our specifically human contributions. When we speak of an entity as being essentially thus-and-so, the anti-essentialist regards such a characterization as being indicative merely of the fact that some feature is being singled out, in a particular context, as especially salient or important from a specifically human point of view. Historically, the doctrine of essentialism traces back at least to Plato, who in his middle period (e.g., Phaedo, Republic, Symposium) in effect reified essences in the guise of immaterial, changeless, and eternal forms (e.g., justice, beauty, goodness) and argued that we can arrive at genuine knowledge or understanding only when we are able to grasp these forms from which sensible particulars derive their characteristics. Aristotle invented a special technical phrase, “to ti ēn einai” (literally “what it was to be”), to refer to the essence of an entity and assigned an enormously important role to definitions (statements of the essence) in his Posterior Analytics, where they function as first principles or axioms (archai) within his theory of demonstrative scientific knowledge (epistēmē). Many of Aristotle’s central metaphysical and epistemological questions concerning essence, e.g., its relation to substance and form, continued to occupy center stage during the medieval period, particularly insofar as they interacted with specific Christian doctrines (e.g., transubstantiation or resurrection). These medieval debates were subjected to an increasingly critical reception during the early modern period by such thinkers as Bacon, Hobbes and Locke, all of whom were skeptically inclined not only towards the existence of essences, but also towards our ability to grasp them. In modern times, essentialism came under serious attack from a variety of directions, as is illustrated by the following criticisms from philosophy, natural science, and social science. Firstly, during the first half of the 20th century, W. V. Quine famously objected that a view he calls “Aristotelian essentialism” is incoherent (Quine 1953). Some of Quine’s concerns were logical and have since been put to rest by developments in modal logic (Kripke 1963; Marcus 1967). But Quine’s other concerns were of a more metaphysical nature and are still debated among metaphysicians today (Della Rocca 1996a, 1996b; Koslicki 2020; Paul 2004, 2006). Essentialists, in Quine’s view, are unable to meet the demand for necessary and sufficient conditions by which to identify the very same entity across different possible scenarios (or “worlds”); and the idea that each entity is associated with an unchanging kernel of essential features is called into question by Quine’s position that “you can change anything to anything by easy stages through some connecting series of possible worlds” (Quine 1976, p. 861). According to Quine, an apparent essence ascription must always be interpreted as displaying some relativity to a particular way of singling out the entities under consideration, e.g., when we say of human beings that they are essentially rational but only accidentally bipedal, we must specify whether we have in mind human beings qua mathematicians or qua bicyclists. Without filling in the alleged ellipsis in such apparent essence ascriptions, Quine could attach no sense to an utterance which purports to express that an entity has a certain nature or essence in itself and independently of its being described in a certain way (Cartwright 1968).

2

Introduction

Secondly, it has been argued by natural scientists and philosophers of science that essentialism is incompatible both with the results of our best scientific theories and with the aims of our best scientific practices. For one thing, the fact that species evolve evidently contradicts Aristotle’s assumption that the essences associated with biological species are eternal and unchanging (Brigandt 2009; Ereshefsky 2004; Hacking 2007; Love 2009; Nanay 2011; Wilson, Baker & Brigandt 2007). In addition, given that the goals and interests driving scientists in their practices are multi-faceted, the recent “turn to practice” in the philosophy of science recommends a pluralistic attitude towards systems of classification in place of essentialism about natural kinds (Dupré 1993; Love & Brigandt 2017; Minelli 2014; Waters 2017; Wilson 1999). Thirdly, essentialism has been a central target in the debate concerning the status of social categories such as race and gender. Whether social categories, such as race and gender, represent real and objective divisions among people who share certain essential characteristics is highly controversial and widely disputed, e.g., by those who take these categories to be fictional, conventional, socially constructed, or reflective of unjust hierarchies. For example, on the assumption that a person has their gender essentially, trans people would not survive their gender transitions (Appiah 1990). According to Simone de Beauvoir, for example, “one is not born, but, rather, becomes a woman” (de Beauvoir 2011, p. 607); Judith Butler argues that gender is “real only to the extent that it is performed” (Butler 1988, p. 527; Butler 2007); and Sally Haslanger proposes that, in an ideal society which lacks the systematic oppression or subordination of some groups by others, such categories as woman or African American would become obsolete and should therefore be abandoned (Haslanger 2000, 2005, 2012).

2 Why This Volume, and Why Now? Given these serious and wide-ranging criticisms to which essentialism has been subjected, one may legitimately wonder whether this doctrine is not simply a vestige left over from ancient and medieval times that should have been abandoned a long time ago. In fact, however, the debate between essentialists and anti-essentialists is alive and well and has, if anything, gained momentum in light of an increased interest among philosophers working across a wide range of subdisciplines in so-called “hyper-intensional” notions, e.g., metaphysical ground (Raven 2020), which attempt to draw more fine-grained distinctions than those that can be captured by standard logical and modal tools. The purpose of this Handbook is to provide a state-of-the-art survey of contemporary philosophical debates concerning essence. There are at least three main motivations for this volume. The first is to collate the often-fragmented literature on essence. Most of the literature on essence belongs to one of several clusters, such as biological essentialism, social essentialism, the metaphysics of essence, the logic of essence, skeptical challenges to essence, and so on. Despite each of these clusters focusing on essence in one way or another, the literature on each has often developed largely independently of the rest. This Handbook seeks to unite these clusters under one banner. We hope that this endeavor in turn will also help to facilitate comparative studies of the role of essence within each cluster. Secondly, the recent literature on essence, in our estimation, has at times been dominated by an unhealthy focus on a handful of examples, such as the relation between Socrates and Socrates’ singleton set, which is cited in Kit Fine’s influential work on essence and modality as a counterexample to modal approaches to essence (Fine 1994). This myopic focus on particular cases may at times encourage the tendency to overgeneralize what are in fact idiosyncratic features. It is our hope that this Handbook can contribute to a reorientation of

3

Kathrin Koslicki and Michael J. Raven

these debates away from an unhelpful preoccupation with specific examples and towards an active engagement with the larger, and ultimately more important, underlying philosophical issues. A third reason for which the present volume strikes us as timely is that essentialists and anti-essentialists often appear to be talking past each other. For one thing, some antiessentialists end up attacking a “strawman” version of the opposing view. These antiessentialists load up the notion of essence with so many constraints that no reasonably minded contemporary essentialist would want to accept the resulting notion of essence. To illustrate, to assume right from the start, as is sometimes done, that essential features must also be intrinsic immediately closes off questions on which in fact no consensus has as of yet been reached. For instance, philosophers disagree about whether an individual’s actual origins are essential to it and to what extent essentialism can be applied within the social domain. By contrast, essentialists may be tempted to respond to anti-essentialist challenges by arguing that the relevant notion of essence need not meet the constraints in question (e.g., intrinsicality, mind-independence, or the demand that essences must supply necessary and sufficient conditions for kind-membership). In the process, however, essentialists risk embracing a notion of essence that is so stripped down that it can no longer perform the relevant explanatory roles that essences were supposed to play. The most persuasive version of anti-essentialism thus will be one which does not build implausible and outdated assumptions into what essences are taken to be. Conversely, the strongest version of essentialism will be one that does not simply ignore important challenges to essentialism that have been put forward by anti-essentialists. Ideally, both essentialists and anti-essentialists alike should be interested in targeting a dialectically strong and appropriate notion of essence that cannot easily be dismissed or discredited by the opposing camp. It is our hope that this Handbook may help to clarify such a target notion, as well as the various debates that engage with it.

3 Primer Discussions of essence often involve making subtle distinctions, drawing intricate connections between neighboring notions, and so on. The distinctions and notions in play have evolved over time. And just which are in play can change: some that were historically prominent remain so while others drift away, and still others only emerged recently. But all these perturbations may be seen as orbiting a single distinction at their center of gravity. This is the distinction between essence and accident. At least in modern times, this is standardly thought of as a distinction among an item’s properties. On the one hand, a property may hold in virtue of an item’s identity, or be part of what it is to be that item. Such a property is essential to the item. On the other hand, a property may hold of an item but not by virtue of its identity. Such a property is inessential or accidental to the item. Before proceeding, a methodological qualification is needed. It is hard to find uncontroversial examples. This is not because examples are hard to find. Most any domain provides many. It is rather that controversies await even those examples that may strike us, at least at first, as most plausible. The examples discussed should not be thought of as comprehensively covering all the bases. Nor should they be thought of as conclusive illustrations immune to challenge. They should, however, at least help to fix ideas. Michelangelo sculpted the statue David. What differentiates a statue from a sculpture is that a statue must be of a person or animal. So, David wouldn’t be what it is—a statue—were it not humanoid. It lies in the nature, or identity, of David to be humanoid. That is, David’s humanoid shape is essential to it.

4

Introduction

But many of its material properties are not. Nothing about the nature, or identity, of David requires it to have its exact material composition. The protective patina Michelangelo polished into David eroded long ago. David’s left arm was repaired after being broken off in a riot in 1527. And in 1991 David’s toe was damaged by a hammer-wielding attacker. These events changed the marble David is made of. These were changes in how David is composed, but not in what David is. The changed material properties are accidental, not essential, to David. In making a claim of essence, our intention is to make a claim about an item itself, and not just our conception of it. Some have thought that this cannot be done. Locke famously distinguished between real essence and nominal essence. The real essence of an item is what it is. By contrast, the nominal essence of an item is, roughly, our abstract conception of what an item is. Locke argued that we do not grasp whatever real essences there may be, and so are stuck with only their nominal essences. Whether or not he was right, we may still sharply distinguish between real and nominal essences and insist that our focus is on the former. The example of David illustrates how the distinction between essential and accidental properties applies to an individual item. But the distinction appears to generalize to kinds of items as well. Suppose there is a certain kind of thing: a statue. Then being a humanoid is essential to being a statue, whereas being made of marble is not. We may later extend our discussion to kinds, if we wish. But, for now, our focus will be on individuals, like David. It may be that some properties are had essentially if at all, or accidentally if at all. For example, perhaps a number is prime only if it is essentially prime. In such cases, if there are any, it may be permissible to speak of the property itself as essential, or accidental. But, usually, a property is essential or accidental only relative to an item which has it essentially or accidentally. The property of being humanoid is essential to David, but it is only accidental to the marble. David is made (mostly) of a certain chunk of marble. This may be just an accidental property of David’s. Perhaps David could have been made from a different chunk of marble, or perhaps not even marble but bronze (although see Kripke 1980 for an opposing view). Be that as it may, it does not seem as if David could be what it is without being made of matter of some or other sort. If so, David is essentially made of matter. We have identified a second essential property of David’s. Each of these, on its own, partially characterizes what David is. So, in identifying both, we more fully characterize what David is. If there is a full characterization of what David is, then it is the essence of David. In general, the essence of an item is a full characterization of what it is. The essence of an item, so understood, may be identified with, or at least taken to determine, all of its essential properties. We have used various locutions to characterize the distinction. This is partly because none is ideal. Each has its own misleading insinuations. We may prevent misunderstandings by flagging them. In saying that an essential property concerns what something is as opposed to how it is, one might think there is an insinuation that essential properties don’t also concern how things are. But that is not so. There is no need to prohibit a sense in which how something is includes what it is. What is required, however, is allowing that how something is does not settle what it is. Similarly, there is no need to prohibit an expansive sense of “identity” that includes all of an item’s properties. What is required is just the recognition of a narrower sense of “identity” that excludes an item’s accidental properties. In saying that it lies in the nature of David to be humanoid, one might think there is an insinuation that essential properties must concern the natural world. And, of course, some claims of essence do. Perhaps, for example, it lies in the nature of butane to have four carbon

5

Kathrin Koslicki and Michael J. Raven

atoms. But even when an essentialist claim is about the natural world, it is not just because of the use of the word ‘nature’ in the phrase “it lies in the nature of”. Essentialist claims needn’t concern the natural world. For example, Descartes thought that he was essentially a thinking thing. He was, notoriously, not making a claim about the natural world. And, for another example, the number 2 is essentially prime. This too is not a claim about the natural world (it may be what some philosophers call an “unworldly” claim). When phrases like “lies in the nature of” express claims of essence, they are not biased for, or against, anything to do with the natural world. In saying what an item is, one might think there is an insinuation against referring to other items. After all, if our aim was to say what David is, why must anything else be referred to? Indeed, one way to expose an alleged essentialist claim as incorrect is to point out that it involves extraneous things. Although David is now located in the Galleria dell’Accademia in Florence, this is not essential to it. (It could be, and has been, moved.) One may be tempted to explain why David’s location is only accidental to it by saying that its location involves things (Florence) external to it. And this may tempt one into assuming that, in general, an essential property of an item must be internal or intrinsic to it. As common as this intrinsicality assumption may be, it is clearly an adjunct. There is nothing latent in the essence/accident distinction requiring essential properties to be intrinsic. Moreover, many prominent examples of the distinction are incompatible with such a requirement. This includes our own example. To illustrate, there is a difference between marble sculpted to be humanoid and a chunk of marble shaped by erosion to be humanoid. The difference is that the sculptor intended to represent an actual or fictional human (or perhaps just some or other human in general). We may accordingly refine what we mean in claiming that David is essentially humanoid: it is essentially sculpted by a sculptor intending it to represent the biblical David. By contrast, a chunk of marble shaped by erosion isn’t essentially humanoid. It’s accidental that it was. This refinement makes plain that the property ascribed to David as essential is not intrinsic in any recognizable sense. It refers, at least implicitly, to its sculptor and the sculptor’s intentions, as well as to the biblical David himself. Further counterexamples to the intrinsicality assumption derive from Kripke’s influential discussions of essence. Most of his examples involve historical essences. He suggests, for instance, that it may be essential to a complex material object that it originate in the matter from which it was formed. He also suggests that it may be essential to a person to originate in the sperm and egg from which they grew. Claims like these ascribe historical properties as essential. Such historical properties may make reference, at least implicitly, to items external to the item in question. For example, the sperm and egg from which I grew are identified by coming from, respectively, my father and my mother. For another example, the prominent phylogenetic conception of a species partly identifies a species’ members by their ancestral lineage. Historical properties, in general, needn’t be intrinsic. So, if historical properties may be essential, then it cannot consistently be required that they also be intrinsic. The preceding considerations were meant to help home in on a target notion of essence. Much more remains to be said about it, and notions in the vicinity. There is no hope of canvassing them all comprehensively here. Still, it will be helpful to survey some of the more prominent of these in one place. Many of the distinctions and issues that follow were introduced, or at least introduced into the contemporary literature, by Kit Fine’s pioneering article “Senses of Essence” (Fine 1995). Further discussions of them, and much more, may be found across the chapters in this volume, especially those in Part II. Objectual/generic. The example above focused on the essential properties of an object: the statue David. This illustrates the objectual notion of essence: essence as it applies to objects.

6

Introduction

But there is also a generic (or predicational) notion of essence: essence as it applies to what is predicated to objects. If we assume that being of a kind is expressed by predication, then we have already seen an illustration: to be a statue is essentially to be a sculpture of a person or animal. This is meant to capture part of what it is to be a statue, as opposed to what it is to be this statue (say, David). The distinction between objectual and generic essence, however, is not the end of the line. If one is theorizing in a setting of higher-order logic, then we may also ask about essences of higher-order types. For example, one may admit a notion of essence as it applies to operations (such as truth-functions) on what is predicated to objects. Partial/full. Earlier we considered two essential properties of David’s. Each partially characterizes what David is. Neither alone, nor perhaps jointly, fully characterizes what David is. We were implicitly appealing to a distinction between partial and full essence. Just how it gets drawn, however, depends on whether it is assumed that there is a full characterization of what an item is. If the assumption is granted, then the full characterization may be identified with its full essence. This will be, or will determine, a complete list of its essential properties. Each of these will be part of its full essence. So, granting the assumption, partial essence may be defined as part of a full essence. Without the assumption, however, partial essence cannot be defined as part of a full essence. It must, instead, be allowed that an essential property of an item may only strictly partially characterize what it is even if there is no full characterization of which it is a part. The preceding distinctions contrast features of the core notion of essence. Several additional distinctions contrast ways in which a conception of essence may be more or less discriminating: Immediate/mediate. There is a distinction between an immediate notion of essence that concerns only what is part of an item’s own identity and a mediate notion that also concerns the identities of any items on which it depends. To illustrate, we supposed that David is essentially humanoid. Now consider the singleton set {David} which, we suppose, essentially contains David. Is it essential to the set {David} that David is humanoid? The answer is “No” on the immediate notion, but “Yes” on the mediate notion. The mediate essence of {David} is therefore “hereditary” in that it includes all that is essential to its “ancestors”, such as David. In general, when an item depends on others, their essential properties must be included in its mediate essence but needn’t be included in its immediate essence. Constitutive/consequential. There is a distinction between a consequential notion of essence that includes essential properties had merely as consequences of others and a constitutive notion that excludes these. This distinction may at first seem to repeat the immediate/mediate distinction. But there is a significant difference. The consequential essence of an item is not limited to including only what is essential to what that item depends on. To illustrate, we supposed that David is essentially humanoid. Nothing whatsoever about David depends on being a turnip. But it is a logical consequence of anything’s being humanoid that it is humanoid or a turnip. Is it essential to David that it is humanoid or a turnip? The answer is “Yes” on the consequential notion, but “No” on the constitutive notion. In general, the constitutive notion of essence captures the “core” of an item’s identity, whereas the consequential notion captures the logical closure of its “core” (for some given notion of logical consequence). Individual/collective. Our examples, so far, focused on the essential properties of individuals. We may also allow for collections of individuals to have essential properties. This may be desirable in cases where there is no other way to distinguish individuals. For example, there are two solutions to the quadratic equation x2 + 1 = 0. The solutions are algebraically indistinguishable. The solutions may be distinguished once one is labeled as the

7

Kathrin Koslicki and Michael J. Raven

imaginary unit i and the other as its inverse -i. But the labeling is arbitrary. So, it may be doubted whether the referent of ‘i’ can be fixed independently of ‘-i’s. If so, it is unclear that ‘i is essentially -i’s inverse’ says anything different than ‘-i is essentially i’s inverse’. And even if it does, the two claims may involve an objectionable circularity: for part of the identity of each solution is stated in terms of the other. The notion of collective essence may help avoid these difficulties. For it may be taken to be essential to the collection i,-i together that they are inverses. (Cases like this are discussed in Litland 2022). Definition and essence. We used the phrase ‘part of what it is’ to express an essential property of an item. This phrase is also often used in connection with definition. For example, in stating that part of what it is to be a triangle is to have three angles, we may be stating that it is part of the definition of a triangle to have three angles. These points are confirmed by a venerable tradition associating definition and essence. It may be that not all essences are stateable. But when an item has a stateable essence, its definition is the statement of its essence. (Confusingly, and as mentioned earlier, a definition is sometimes taken to be an item’s essence itself, not just a statement of its essence.) While a definition is a statement of essence, it is a separate matter whether what is defined is itself a statement, or any other linguistic object. It may be in some cases. For example, in stating that ‘bachelor’ means unmarried man, one is stating the essence of the word ‘bachelor’. By contrast, other definitions do not state the essence of anything linguistic. For example, in stating that David is essentially humanoid, one is stating part of the essence of the statue David. The connection between definition and essence suggests that each of the distinctions for essence has an analogue for definition. For example, given that it is part of David’s essence to be humanoid, the statement that this is so is a partial definition of David. For another example, given that being humanoid is part of the mediate essence of the singleton set {David}, the statement that this is so is a partial mediate definition of {David}. And, for a third example, given that it is essential to the collection i,-i that they are inverses, the statement that this is so is a collective definition of the pair i,-i. Expressing essence. Essentialist claims may be expressed in various ways. The differences in the manner of expression suggest different regimentations of their grammatical form. There are two main approaches. The predicational approach expresses essence by means of a predicate modifier, usually ‘essentially’. This is attached to a predicate to form a complex predicate “essentializing” it. The complex predicate is then applied to a subject term denoting the item whose essence is at issue. This approach is most naturally suggested by locutions of the form ‘x is essentially ϕ’. For example, essentially-humanoid(David) gives the underlying form of ‘David is essentially humanoid’. The sentential approach expresses essence by means of a sentential operator, usually an indexed ‘◻’. The index is a term standing for the item whose essence is in question. The indexed operator is then prefixed to a sentence. This approach is most naturally suggested by locutions of the form ‘it lies in the nature of x that ϕ’. For example, ◻DavidHumanoid(David) gives the underlying form of ‘It lies in the nature of David to be humanoid’.

8

Introduction

Other approaches have also appeared in the literature. Perhaps the most recent of these is Fine’s (2015) oversize arrow notation: x = David �x Humanoid(x) Each of the approaches offers its benefits and costs, with some being better suited for some purposes than others. Nevertheless, the sentential approach has become the dominant approach, and is the most followed in this volume as well. Many of the preceding distinctions largely focused on broadly structural features of essence, or the expression of essence. Other distinctions have more to do with the substantive or contentful features of essence. Among the most important of these is how essence relates to modality. Modality. Essence is often associated with modality: what is necessary, contingent, or impossible. Philosophers have distinguished between various modal notions. Especially since Kripke (1980), many have thought that there is a distinctively metaphysical notion. It is this notion of metaphysical modality that, presumably, is most relevant to essence. Most (although not all) philosophers who have written on the subject have accepted that essence implies necessity. If an item essentially has a property, then it has that property necessarily. The rapid development of modal logic and modal metaphysics in the second half of the 20th century increased enthusiasm for modal conceptions of difficult notions. In particular, it was common to find philosophers also accepting the converse implication: that necessity implies essence. This fits with a modal conception of essence: that for an item to essentially have a feature just is for it to have it of necessity. Kit Fine’s iconoclastic arguments in “Essence and Modality” (Fine 1994) promoted a nonmodal conception of essence. Fine did not dispute that essence implies necessity. But he argued against the converse implication by providing examples of items that necessarily had a property without having it essentially. His most famous example involves Socrates and the singleton set {Socrates} with Socrates as its sole member. Fine claimed that although Socrates necessarily is a member of this set, Socrates is not essentially a member of this set. This, he claims, reveals our inclination towards a conception of essence that takes it to have modal consequences but does not conceive of it as a modal notion. Reversing the direction of explanation, Fine also suggested that such a conception allows for an account of modality in terms of essence (although see Vetter’s 2015 for a competing account of modality in terms of potentiality). In the aftermath of Fine’s work, it has become an important choice point whether to adopt a modal or nonmodal conception of essence. Either choice, in its own way, must confront various other questions about how essence and modality relate. Metaphysical modality is widely (although not uncontroversially) thought to have certain formal features. Specifically, it is widely thought to conform to the modal logic known as S5. Must the logic of essence mirror the logic of metaphysical modality? Another recent debate concerns the metaphysical import of modal logic. Necessitists claim that necessarily everything is necessarily something, whereas contingentists deny this. There are currently unresolved questions about how essence interacts with the debate between necessitists and contingentists (see, for instance, Teitel 2019). Ontology. There are various ontological issues surrounding the notion of essence. One of these concerns a traditional conception of substance. On this conception, a substance is something that exists independently of all else. Some have suggested clarifying this conception in terms of essence: a substance is something whose essence does not prevent its existence.

9

Kathrin Koslicki and Michael J. Raven

There are many other distinctions in the vicinity, especially those involving various senses in which the essence of something may depend on something else. But perhaps the most basic ontological question about essence concerns its existence. Do essences exist? If so, are they entities of some sort? If they are not but still exist, what sort of thing are they? The distinctions and issues surveyed above barely scratch the surface. Too many remain to even list comprehensively here. Our hope, however, is to provide at least a rough orientation of the topics that most animate the literature. The chapters to follow engage with these and many of the remaining topics.

4

Future Directions

One motivation for this volume is to prompt and guide the discipline to reflect on future directions for research. All of the chapters, in one way or another, engage with important topics worthy of further pursuit. But we will conclude with a few of our own guiding suggestions for future research. Our main suggestion is to promote a methodology that resists presupposing reductive tendencies and, instead, adopts as a working hypothesis that essence is its own bona fide notion worthy of study. Some reductive tendencies may derive from skepticism about essence (such as those discussed in Part IV). Others are wolves in sheep’s clothing. Perhaps the prime example is the modal conception of essence. At least on one understanding of it, it aims to reduce essence to necessity (or, at least, necessity of a certain sort). This constrains how essence must be understood. Features of essence that resist expression in modal terms are demoted or dismissed. In our view, this risks distorting the subject. Even if it is discovered that essence and modality ultimately neatly align, this should be the outcome of a profound investigation, not a presupposition of it. This leads to a second suggestion: epicycle avoidance. Some epicycles are born from fetishizing examples. In our view, some of the literature has gotten bogged down by focusing on Fine’s justly famous case of Socrates and his singleton set. As arresting as this case may be, many attempts have been made to tweak the modal conception of essence to better accommodate our judgments about the case. But, in our view, the key insight behind Fine’s case has nothing really to do with persons, sets, or indeed any particular kind of thing at all. It is rather meant to reveal our implicit grasp of a notion of essence that is not purely modal. An endless back-and-forth of tweak and reply risks missing the forest for the trees. This, in turn, leads to a third and related suggestion: myopia avoidance. Once again, Fine’s case may illustrate the point. His case involves persons and sets, two kinds of entities with idiosyncratic features. Some may have been misled into uncritically transferring these idiosyncrasies onto other cases of other kinds, or onto the abstract notion of essence itself. But essence has been applied to a wide range of phenomena: natural kinds, social items, artifacts, and much more. Overreliance on any one sort of case risks irrelevance to others. A deeper understanding of essence is more likely to come from considering a wide range of examples while not losing sight of the idiosyncrasies peculiar to each. Heeding this suggestion may have been hampered by failing to heed our fourth suggestion: rigorous compartmentalization. By this we mean sharply distinguishing between a core notion of essence from peripheral views about it. This is not meant to involve legislating at the outset on whether these views are correct. It rather involves a methodology of treating these views, at least provisionally, as adjuncts to the core notion. It is left open whether they help characterize essence. Of course, it may not always be clear what belongs to the core and what to the periphery. There may be legitimate debates about what belongs where. But, in our

10

Introduction

view, many theorists have foisted upon the core notion what we regard to be, at best, peripheral assumptions and, at worst, outright distortions. To illustrate, many theorists have been under the impression that essences must conform to certain conditions: that they are intrinsic, not relational; or that they are mind-independent or “discovered”, not mind-dependent or “constructed”. Indeed, some of these theorists regard these conditions as analytic, or definitional, of essence. But our view is that it is a substantive question whether any of these conditions, and many like them, are correct. We regard them as peripheral to a core notion of essence. Accordingly, we recommend a methodology that does not uncritically foist the periphery onto the core. Failure to do so, we think, has been a source of endless confusion. For example, biological or social essences are often rejected on the grounds that the biological and the social are relational but that essences are intrinsic. Reasoning like this is often taken to show that research into biological or social essences is a dead end. But, in our view, this reasoning is confused. We think that the reasoning foists upon a core notion of essence controversial assumptions from its periphery: namely, that essence is intrinsic. This peripheral assumption, however, is excess baggage. Retaining the excess baggage has stunted research into biological and social essences. These missed opportunities, however, are avoidable. The peripheral assumptions, such as intrinsicality, may often simply be discarded. In this case, doing so immediately clears a path for fruitful research on biological and social essences. We intend these remarks as mere suggestions for future research. They are not meant as articles of dogma, but rather as defeasible beacons to guide future work. Even so, we conjecture that following them may be productive. For instance, if essence is treated as its own bona fide notion worthy of study, then we may ask whether it has a distinctive epistemology. And if essence is compartmentalized from problematic peripheral assumptions about it, we may better explore its applications to normative domains, ameliorative projects, conceptual engineering, socio-political topics and other areas long thought to be hostile to it. Many uncharted waters remain. We hope this volume, however, may help guide us through where there be dragons and, ultimately, to safe harbor.

Note 1 Here, and in what follows, we use the terms, “essence” and “nature”, interchangeably.

References Appiah, Anthony (1990): “‘But Would That Still Be Me?’ Notes on Gender, ‘Race’, Ethnicity, as Sources of ‘Identity’”, Journal of Philosophy, Vol. 87, No. 10 (October), pp. 493–499 Beauvoir, Simone de (2011): The Second Sex, translated by Constance Borde and Sheila MalovanyChevallier, 1st Vintage Book Edition, Vintage, New York, NY Bird, Alexander and Tobin, Emma (2012): “Natural Kinds”, Stanford Encyclopedia of Philosophy, < http://plato.stanford.edu/entries/natural-kinds/> Brigandt, Ingo (2009): “Natural Kinds in Evolution and Systematics: Metaphysical and Epistemological Considerations”, Acta Biotheoretica, Vol. 57, pp. 77–97 Butler, Judith (1988): “Performative Acts and Gender Constitution: An Essay in Phenomenology and Feminist Theory”, Theatre Journal, Vol. 40, No. 4 (December 1988), pp. 519–531 Butler, Judith (2007): Gender Trouble: Feminism and the Subversion of Identity, Routledge, London, UK Cartwright, Richard (1968): “Some Remarks on Essentialism”, Journal of Philosophy, Vol. 65, No. 20, pp. 615–626

11

Kathrin Koslicki and Michael J. Raven Della Rocca, Michael (1996a): “Essentialism: Part 1”, Philosophical Books, Vol. 37, No. 1 (January 1996), pp. 1–20 Della Rocca, Michael (1996b): “Essentialism: Part 2”, Philosophical Books, Vol. 37, No. 2 (April 1996), pp. 81–89 Dupré, John (1993): The Disorder of Things: Metaphysical Foundations of the Disunity of Science, Harvard University Press, Cambridge, MA Ereshefsky, Marc (2004): The Poverty of Linnaean Hierarchy: A Study of Biological Taxonomy, Cambridge University Press, Cambridge, UK Fine, Kit (1994): “Essence and Modality”, Philosophical Perspectives, Vol. 8 (Logic and Language), pp. 1–16 Fine, Kit (1995): “Senses of Essence”, in Walter Sinnott-Armstrong, Diana Raffman, and Nicholas Asher (eds.), Modality, Morality, and Belief, Essays in Honor of Ruth Barcan Marcus, Cambridge University Press, New York, NY, pp. 53–73 Fine, Kit (2015): “Unified Foundations for Essence and Ground”, Journal of the American Philosophical Association, Vol. 1, No. 2, pp. 296–311 Hacking, Ian (2007): “Natural Kinds: Rosy Dawn, Scholastic Twilight”, Royal Institute of Philosophy Supplement, Vol. 61, pp. 203–239 Haslanger, Sally (2000): “Gender and Race: (What) are They? (What Do We Want Them to Be?”, Noûs, Vol. 34, No. 1, pp. 31–55 Haslanger, Sally (2005): “What are We Talking about?: The Semantics and Politics of Social Kinds”, Hypatia, Vol. 20, No. 4, pp. 10–26 Haslanger, Sally (2012): Resisting Reality: Social Construction and Social Critique, Oxford University Press, Oxford, UK Khalidi, Muhammad Ali (2013): Natural Categories and Human Kinds: Classification in the Natural and Social Sciences, Cambridge University Press, Cambridge, UK Koslicki, Kathrin (2008): “Natural Kinds and Natural Kind Terms”, Philosophy Compass, Vol.3/4 (2008), pp. 789–802 Koslicki, Kathrin (2020): “Essence and Identity”, in Mircea Dumitru (ed.), Metaphysics, Meaning and Modality: Themes from Kit Fine, Oxford University Press, Oxford, UK, pp. 113–140 Kripke, Saul A. (1963): “Semantical Analysis of Modal Logic I: Normal Modal Propositional Calculi”, Mathematical Logic Quarterly, Vol. 9, No. 5-6, pp. 67–96 Kripke, Saul (1971): “Identity and Necessity”, in Milton Munitz (ed.), Identity and Individuation, New York University Press, New York, NY, pp. 135–164 Kripke, Saul (1980): Naming and Necessity, Harvard University Press, Cambridge, MA, originally published in: Donald Davidson and Gilbert Harman (eds.), Semantics of Natural Language, D. Reidel Publishing Co., Dordrecht, Netherlands, 1972, pp. 253–355 & pp. 763–769 Litland, Jon Erling. (2022): “Collective Abstraction”, Philosophical Review, Vol. 131, No. 4, pp. 453–497 Love, Alan C. (2009): “Typology Reconfigured: From the Metaphysics of Essentialism to the Epistemology of Representation”, Acta Biotheoretica, Vol. 57, pp. 51–75 Love, Alan C., and Brigandt, Ingo (2017): “Philosophical Dimensions of Individuality”, in Scott Lidgard and Lynn K. Nyhart (eds.), Biological Individuality: Integrating Scientific, Philosophical, and Historical Perspectives, University of Chicago Press, Chicago, IL, pp. 318–348 Marcus, Ruth Barcan (1967): “Essentialism in Modal Logic”, Noûs, Vol. 1, No. 1, pp. 91–96 Minelli, Alessandro (2014): “Developmental Disparity”, in Alessandro Minelli and Thomas Pradeu (eds.), Towards a Theory of Development, Oxford University Press, Oxford, UK, pp. 227–245 Nanay, Bence (2011): “Three Ways of Resisting Essentialism about Natural Kinds”, in J. K. Campbell and M. H. Slater (eds.), Carving Nature at its Joints, MIT Press, Cambridge, MA, pp. 175–197 Paul, L. A. (2004): “The Context of Essence”, Australasian Journal of Philosophy, Vol. 82, No. 1, pp. 170–184 Paul, L. A. (2006): “In Defense of Essentialism”, in John Hawthorne (ed.), Philosophical Perspectives (Metaphysics), pp. 333–372 Putnam, Hilary (1962): “The Analytic and the Synthetic”, Minnesota Studies in the Philosophy of Science, Vol. 3, edited by Herbert Feigl and Grover Maxwell, University of Minnesota Press, Minneapolis, MN; reprinted in Mind, Language and Reality: Philosophical Papers, Volume 2, Cambridge University Press, Cambridge, UK, 1975, pp. 33–69

12

Introduction Putnam, Hilary (1975): “The Meaning of ‘Meaning’”, Minnesota Studies in the Philosophy of Science, Vol. 7, University of Minnesota Press, Minneapolis, MN; reprinted in Mind, Language and Reality: Philosophical Papers, Volume 2, Cambridge University Press, Cambridge, UK, 1975, pp. 215–271 Quine, Willard van Orman (1953): “Three Grades of Modal Involvement”, Actes du XIème Congrès International de Philosophie, Volume XIV, Volume complémentaire et communications du Colloque de Logique, North-Holland Publishing Company, Amsterdam, and Éditions E. Nauwelaerts, Louvain, 1953, pp. 65–81 Quine, Willard van Orman (1976): “Worlds Away”, Journal of Philosophy, Vol. LXXIII, No. 22 (December), pp. 859–863 Raven, Michael J. (ed.) (2020): Routledge Handbook of Metaphysical Grounding, Routledge, London, UK Teitel, Trevor (2019): “Contingent Existence and the Reduction of Modality to Essence”, Mind, Vol. 128, No. 509, pp. 39–68 Vetter, Barbara. (2015). Potentiality: From Dispositions to Modality, Oxford University Press, Oxford Waters, C. Kenneth (2017): “No General Structure”, in Matthew Slater and Zanja Yudell (eds.), Metaphysics in Philosophy of Science, Oxford University Press, Oxford, UK, pp. 81–108 Wilson, Jack (1999): Biological Individuality: The Identity and Persistence of Living Entities, Cambridge University Press, Cambridge, UK Wilson, Robert A., Barker, Matthew J., and Brigandt, Ingo (2007): “When Traditional Essentialism Fails: Biological Natural Kinds”, Philosophical Topics, Vol. 35, No. 1/2, pp. 189–215

13

PART 1

History Kathrin Koslicki and Michael J. Raven

Introduction Philosophers have discussed essence since antiquity. The notion of essence played a central role in ancient Greek philosophy and occupied center stage during the Middle Ages. A more critical stance towards this notion developed during the early modern era and continues, in some quarters, into the present time. The turn of the millennium saw the notion of essence falling upon especially hard times. The general anti-metaphysical sentiments of the early 20th century often singled out essence as a particularly pernicious relic of a bygone Aristotelian era. But the tides have begun to turn. In the 1950–70s, pioneering work in modal logic and the metaphysics of modality helped create a climate in which essence—or at least its modal aspects—could be respectably studied. Thereafter, particularly starting in the 1990s, the widely accepted modal notion of essence began to be questioned and a distinct non-modal notion of essence emerged which could once again become a reputable subject of investigation in its own right. Since then interest in hyperintensional notions, such as non-modal essence and grounding, has increased substantially in metaphysics and other parts of philosophy. All this suggests that it is high time to reevaluate essence. To arrive at an adequate assessment of where we are, however, it is necessary first to appreciate the history of philosophical discussions of essence that have preceded our current vantage point. Part I of this Handbook thus begins by offering a historical survey of the role of essence in ancient Greek, medieval, and modern philosophy as well as in the contemporary traditions of pragmatism, phenomenology, and early analytic philosophy. Chapter 1, Marko Malink’s “Ancient” surveys the history of philosophical discussions of essence in ancient Greek philosophy, beginning with Plato’s focus on Socratic “What is X?”style questions, whose answers seek to specify an account or definition (logos) stating the substance, nature or essence (ousia) of the phenomenon, X, under discussion. Malink next turns to Aristotle, who speaks of essence as “the what it is” (to ti esti) or “the what it was for something to be” (to ti ēn einai), and assigns a central role to this notion throughout many of his works, e.g., in discussions of dialectical reasoning (Topics); scientific knowledge (Posterior Analytics); causation (Physics); and substance (Metaphysics). Aristotle, as Malink notes, views essence as non-modal, i.e., as not reducible to necessity, and therefore acts as an important precursor to contemporary debates concerning the relation between essence and

DOI: 10.4324/9781003008750-2

Kathrin Koslicki and Michael J. Raven

modality, which come into focus again in Part II of the Handbook. Malink’s chapter ends by briefly remarking on the role of essence during the Hellenistic period and later antiquity. Chapter 2, Gloria Frost’s “Essence in Medieval Philosophy” discusses central approaches and debates concerning essence or quiddity (quiditas) in medieval scholastic philosophy. Frost focuses on a selection of questions concerning essence which were prominently debated by medieval authors, among them the following: how to distinguish essential from accidental features; whether there are only species essences or also individual essences; whether and to what extent essences are epistemically accessible to the human intellect; whether the essences of material substances are exhausted by their forms or also include their matter; and whether there is a real distinction between the essence of a created substance and its existence. Given their shared Aristotelian commitments and theological beliefs, scholastic philosophers engage with the notion of essence in ways that are both continuous with, but also interestingly distinct from, contemporary debates. Chapter 3, Anat Schechtman’s “Essence in Early Modern Philosophy” documents how early modern philosophers gradually came to disassociate themselves from the paradigmatically Aristotelian approaches to essence that had been prevalent during the medieval period. This shift, so Schechtman argues, also brings with it a changed conception of the theoretical roles that are associated with essences. Previously, only certain selective features of an entity were regarded as essential, namely those for example that explain an entity’s modal profile, kind-membership, persistence over time, or unity. While this conception of essence as selective and explanatory is still accepted by some key figures during the beginning of the early modern period (e.g., René Descartes, Thomas Hobbes, Margaret Cavendish, and Anne Conway), other thinkers of this era (e.g., Benedict de Spinoza, G.W. Leibniz, and John Locke) embrace instead a more expansive conception of essence, according to which essences can no longer be relied upon to be both explanatory and kind-determining. As subsequent chapters of the Handbook bring out, many of the considerations which prompted early modern philosophers to move away from Aristotelian essences continue to resonate with skeptically minded philosophers working in later traditions, even up to the current day. Chapter 4, Andrew Howat’s “Pragmatism & Essence” focuses on the movement known as “classical pragmatism”, which includes such figures as Charles Sanders Peirce, William James, and John Dewey. Although it is typically taken for granted that pragmatists are hostile towards essences, Howat suggests that the relationship between pragmatism and essentialism is in fact much more nuanced and interesting. Pragmatism, so Howat argues, should not simply be conflated with anti-essentialism; nor is it even obviously correct to say that its founding members rejected the possibility that essences, particularly when given an antirealist construal, might play philosophically important roles. Considering the sizable influence exerted by this movement on such heavyweights in the analytic tradition as Rudolf Carnap, W. V. O. Quine, and Hilary Putnam, we thus have much to gain by re-evaluating deeply engrained presuppositions concerning the relationship between pragmatism and essentialism. Chapter 5, Kevin Mulligan’s “Contemporary (Phenomenological Tradition)” takes us through the rich, complex, and varied use of essence made by such figures as Brentano, Hartmann, Husserl, Meinong, Reinach, Scheler, and Stein. In particular, Mulligan’s chapter is organized around the following questions: What did the phenomenologists take essences and their connections to be? What philosophical and non-philosophical roles did the phenomenologists assign to essence? And what sort of epistemic and non-epistemic contact did they take themselves to enjoy with essences? Throughout, Mulligan emphasizes the continuity between the phenomenological school and its historical predecessors; as well as the influences exerted by this tradition on other figures such as Wittgenstein.

16

History

Chapter 6, Robert Michels’ “Contemporary (Analytic Tradition)” provides an overview of the history of discussions of essence in 20th century analytic philosophy. In particular, Michels surveys approaches to essence taken by Russell and Moore prior to and following their break with British idealism; the (pre- and post-)logical positivists’ critique of metaphysics and rejection of essence found in the works of philosophers like Wittgenstein, Carnap, Schlick, and Stebbing; attempts to loosen the notion of logical necessity in the service of accommodating certain metaphysical truths by Wittgenstein and others; Quine’s logical rehabilitation of metaphysics and criticisms of de re modality; the modal view of essence adopted by philosophers like C. I. Lewis, Barcan Marcus, and Kripke in the wake of the development of quantified modal logic and direct reference theory; the emergence of the notion of metaphysical necessity, e.g., in the works of Kneale and Kripke; and finally Fine’s reestablishment of a neo-Aristotelian, hyper-intensional notion of essence in contemporary metaphysics.

17

1 ANCIENT Marko Malink

1.1 Socrates and Plato In Plato’s Socratic dialogues, Socrates often asks questions of the form: What is X? Prominent examples include: What What What What What What

is is is is is is

piety? (Euthyphro) courage? (Laches) justice? (Republic I) temperance? (Charmides) beauty? (Hippias Major) virtue? (Meno)

When Meno first attempts to answer the question “What is virtue?,” he lists a number of examples of particular virtues: the ability to manage public affairs, the ability to manage the home well, and so on (71e–72a). In Socrates’ view, however, this is not a satisfactory answer to the question he asked. For, according to Socrates, giving a list of examples of X does not yet tell us what X is. Instead, he argues, one must seek to identify a single form (eidos) common to all instances of X: Even if the virtues are many and various, all of them have one and the same form (eidos) because of which they are virtues, and it is right to look to this when one is asked to make clear what virtue is. (Plato, Meno 72c6–d1) Similarly, when Socrates asks Euthyphro “What is piety?”, he asks Euthyphro to tell him “not one or two of the many pious things, but that very form (eidos) because of which all pious things are pious” (6d9–11). Thus, a proper answer to the question “What is X?” should specify a single form (eidos) which is shared by all the things that are X and which explains, for each of these things, why it is X. When we inquire into a subject X, it is important to know what X is because this is a necessary condition for knowing any of X’s attributes. For, as Socrates puts it in the Meno, “if I do not know what something is, how could I know what attributes it possesses?” (71b3–4).

DOI: 10.4324/9781003008750-3

19

Marko Malink

At the same time, Socrates warns us that the attributes of X do not tell us what X is. For example, when Euthyphro makes a second attempt at answering the question “What is piety?”, he suggests that the pious is that which is loved by all the gods (7a and 9d). Again, Socrates is not satisfied with this answer. For, if being loved by all the gods were the answer to the question “What is piety?”, then anything pious would be pious because it is loved by the gods. But, according to Socrates, this is not the case; instead, what is loved by the gods is loved by the gods because it is pious (10d4–8). Hence, Socrates contends that Euthyphro’s second answer is not successful: I’m afraid, Euthyphro, that when you were asked what piety is, you did not wish to make its essence (ousia) clear to me, but instead you told me an attribute of it, namely, that the pious has the attribute of being loved by all the gods; but you have not yet said what the pious is. (Plato, Euthyphro 11a6–b1) In order to answer the question “What is X?”, Socrates argues, one must specify the essence (ousia) of X.1 Here, “essence” translates the Greek ousia, an abstract noun which is derived from the verb “to be” (einai) in a similar fashion in which the Latin noun essentia is derived from the verb “to be” (esse). The noun ousia was commonly used in Attic Greek in the sense of “possession”. In the present passage, however, Plato uses it in a novel, distinctively philosophical way that corresponds broadly to what philosophers today mean when they speak of “essence”. Socrates maintains that the proper answer to the question “What is X?” is an account (logos) specifying the essence (ousia) of X.2 Accordingly, Aristotle credits Socrates with being the first to seek general definitions specifying what something is (Metaphysics M 4 1078b17–31). According to Aristotle, some Platonists made the mistake of “separating” the universals specified by these definitions, and called these universals “ideas” (ideai, 1078b30–4). Socrates’ own inquiries into the essence of things, however, are free from such ontological commitments to Platonic ideas. In the Republic, Plato lays out his conception of dialectic (dialektikê). One of the key tasks of dialectic is the search for essences. Plato characterizes dialectic as “the enquiry which attempts, for each thing just by itself, to grasp what that thing is” (Republic VII 533b2–3). Accordingly, he maintains that we “call dialectical the person who grasps the account of the essence (ousia) of each thing” (534b3–7). In his late dialogues, Plato highlights the method of division (dihairesis), in which a genus is divided into species and subspecies in such a way that we obtain definitions specifying what each of these species is (Sophist, Statesman). For example, in the Sophist, Plato offers a complex series of divisions with the aim of defining what a sophist is.3 However, it is a matter of debate to what extent Plato took the method of division to provide all the resources needed to identify the essence of a given kind.4 The status of questions of the form “What is X?” was an important topic of discussion among Socratic thinkers. For example, Antisthenes, a pupil of Socrates and opponent of Plato, examined the nature of definitions: Antisthenes was the first to define “definition” (logos), by saying: “A definition is an account revealing what a thing was or is.” (Diogenes Laertius, Lives of the Eminent Philosophers 6.3) However, unlike Socrates and Plato, Antisthenes ended up denying the possibility of defining things by specifying their essence. According to Aristotle, Antisthenes held the view that “one cannot define the essence of a thing since the definition would be a long account;

20

Ancient

but one can specify and teach of what sort a thing is” (Metaphysics H 3 1043b23–7).5 As we will see, Aristotle himself does not share Antisthenes’ skepticism, but contends that we are in fact able to define things by specifying their essence.

1.2 Aristotle: The Logic of Essence The notion of essence plays a central role in Aristotle’s thought, both as a conceptual tool and as a subject of philosophical investigation in its own right. When Aristotle speaks of essence, he refers to it by the nominalized phrases “the what it is” (to ti esti) and “the what it was for something to be” (to ti ên einai). Sometimes he also uses Plato’s terminology, referring to a thing’s essence as its ousia. To express statements of essence, Aristotle employs a predicational syntax to the effect that a given predicate is predicated essentially (katêgoreisthai en tôi ti esti) of a subject. He introduces this syntax in the Topics, an early work designed to offer guidance on how to succeed in dialectical debates. His account of dialectical arguments in this work is based on a distinction between five kinds of predication: definition, genus, differentia, property, and accident (Topics 1.4–5). Typical examples are: rational animal is a definition of human being animal is a genus of human being rational is a differentia of human being capable of learning grammar is a property of human being just is an accident of human being According to Aristotle, these five kinds provide an exhaustive classification of all predications (Topics 1.8). Thus, he maintains that, for any A and B: A is predicated of B iff A is a definition, genus, differentia, property, or accident of B Here, A and B are terms signifying either universals (such as the species human being) or particulars (such as Socrates). The first three kinds of predication are essential. Unlike properties and accidents, definitions, genera, and differentiae specify the essence of their subjects, either their complete essence or part of their essence. For example, being rational is predicated essentially of human being and is part of what it is to be human. By contrast, being just or being capable of learning grammar are not predicated essentially of human being, and are not part of what it is to be human. Thus, Aristotle holds (Topics 1.8): A is predicated essentially of B iff A is a definition, genus, or differentia of B On this account of essential predication, there is a close connection between essence and definition. A definition, for Aristotle, is a predicate which is coextensive with its subject and signifies its complete essence (Topics 1.4 and 1.8). This kind of definition is not a “nominal” definition specifying merely the meaning of a word (Posterior Analytics 2.10). Instead, it is what later came to be called a “real” definition, that is, an account that specifies the essence (to ti ên einai) of the thing defined.6 In the Topics, Aristotle assumes that every such definition consists of a genus and one or more differentiae.7 Thus, essential predication includes all and only predications of definitions and their components.

21

Marko Malink

Given its close connection to real definition, Aristotle’s conception of essence is non-modal (cf. Chapters 7 and 8 in this volume). The essence of a kind does not include all necessary attributes of the kind, but only those that appear in its real definition (i.e., its genera and differentiae). For example, having the sum of interior angles equal to two right angles is a necessary attribute of the kind triangle; yet Aristotle is clear that this attribute is not predicated essentially of triangle but only non-essentially as a property.8 Consequently, Aristotle’s conception of essence cannot be characterized solely in modal terms. Instead, Aristotle adopts the converse strategy, using the notion of essence to characterize the modality of necessity. Thus, for example, he appeals to essential predications to characterize the necessity of premises appearing in scientific demonstrations (Posterior Analytics 1.4 and 1.6). As Jonathan Barnes puts it, “Aristotle starts from the fact that the objects of understanding—and hence of demonstrative understanding—are necessary, and asks, in effect, what is the ground of their necessity. He answers that the necessity is ultimately grounded in essential or definitional connections”.9 In the Topics, Aristotle describes a large number of dialectical rules regarding essential predication. In particular, he provides detailed instructions for how to establish and refute claims concerning genera and differentiae (Topics 4) and claims concerning definitions (Topics 6 and 7). For example, Aristotle argues that a thing must be defined in terms that are “prior and more familiar” than the thing defined, on the grounds that any given thing has no more than one essence and hence no more than one definition: See if the interlocutor has failed to produce the definition through terms that are prior and more familiar. For a definition is produced in order to come to know the thing defined, and we come to know things not through any terms, but through those that are prior and more familiar … . Accordingly, it is clear that one who does not define through such terms has not defined at all. Otherwise, there will be more than one definition of the same thing; for clearly a definition through terms that are prior and more familiar is a better definition, so that both will then be definitions of the same object. This, however, does not seem to be correct, for of each being there is a single essence. If, then, there were a number of definitions of the same thing, the thing defined will be the same as the essences represented in each of the definitions; but these are not the same, inasmuch as the definitions are different. (Aristotle, Topics 6.4 141a26–b1) Since many of the dialectical rules described in the Topics derive from the practice of dialectical debates in Plato’s Academy, this suggests that the non-modal conception of essence and the theory of essential predication presented by Aristotle in the Topics were generally known among the practitioners of dialectical debates in the Academy. Throughout the Topics, Aristotle focuses on essential predications in which the subject is a general term signifying a universal, such as “human being” or “justice”. However, he also countenances essential predications in which the subject is a singular term denoting a particular substance. For example, “animal” is predicated essentially as a genus of “Socrates” (Categories 5 3a2). Thus, essential predications are used by Aristotle to specify the essence of both universal kinds and particular substances (cf. Chapters 9 and 10 in this volume). That being said, it is important to note that Aristotle does not countenance definitions of particular substances such as Socrates. While Socrates has a genus (animal) and a differentia (rational), his genus and differentia do not constitute a definition of Socrates as opposed to other particular substances. Instead, they constitute a definition of what Aristotle calls the “lowest species” (teleutaion eidos) predicated of Socrates.10 A lowest species, also known as infima

22

Ancient

species, is indivisible (atomon) in the sense that it cannot be further divided by differentiae to yield further subspecies of which it is predicated essentially:11 A is a lowest species iff i A is a species of some genus (i.e., A is a universal and there is a B such that B is a genus of A), and ii A does not have any subspecies (i.e., A is not predicated essentially of any universals) Aristotle maintains that every particular substance has exactly one lowest species. Thus, for example, the lowest species of Socrates is human being, and the lowest species of his favorite pet is cat. Of course, there are numerous subclasses of cats, e.g., the classes of curious cats, cute cats, and clumsy cats. But none of these classes constitutes a species of which cat is predicated essentially. For, Aristotle adopts a demanding notion of genus and species such that not every class of particulars instantiating a given genus constitutes a species of that genus. Lowest species provide necessary conditions for the persistence of things that instantiate them (cf. Chapter III.319 in this volume). A particular substance instantiating a lowest species can remain one and the same while undergoing various changes, but it cannot remain one and the same if it ceases to instantiate its lowest species: It is impossible for a thing still to remain the same if it entirely changes out of its species, for example, the same animal cannot at one time be, and at another not be, a human being. (Aristotle, Topics 4.5 125b37–9) Consequently, Socrates cannot exist without his lowest species, human being, existing, whereas the lowest species can exist without Socrates existing. More generally, Aristotle holds that, if A is a genus or differentia of B, then A is ontologically prior to B in the sense that, if A were to perish, B would perish but not vice versa (Topics 6.4 141b25–9). Aristotle divides all beings into ten categories: substance, quantity, quality, relation, place, time, position, possession, activity, and being affected (Categories 4, Topics 1.9). The beings in each category are either particular or universal (Categories 2). For example, in the category of substance, Socrates is a particular while the species human being and the genus animal are universals; in the category of quality, Socrates’ pallor is a particular, while the species pallor and the genus color are universals. Aristotle does not elaborate on the status of non-substantial particulars in his ontology, and there is a debate about whether or not they can be viewed as tropes.12 In any case, Aristotle is clear in the Topics that both substances and non-substances have an essence. For example, just as the genus animal is predicated essentially of human being, the genus color is predicated essentially of pallor.13 In the Metaphysics, Aristotle qualifies this view, stating that only substances have an essence in a primary sense, whereas non-substances have an essence only in a derivative sense (Zeta 4 1030a17–27, 1030b4–12, 5 1031a1–14). For Aristotle, the essence of non-substances is inferior to that of substances because it involves a reference to the kind of substance that serves as the underlying subject of the non-substance under consideration, and therefore involves a sort of complexity that is absent from the essence of substances.14 Aristotle maintains that the relation of essential predication is transitive (Posterior Analytics 2.4 91a18–21): If A is predicated essentially of B, and B is predicated essentially of C, then A is predicated essentially of C.

23

Marko Malink

Thus, for Aristotle, essential predication licenses mediate and immediate essentialist claims alike. Consider, for example, a chain of essential predications starting from a subject, A: … D is predicated essentially of C, C is predicated essentially of B, B is predicated essentially of A. Given that essential predication is transitive, each of the predicates B, C, D, … is predicated essentially not only of its immediate predecessor, but of all its predecessors. Now, Aristotle maintains that any upward chain of essential predicates B, C, D, … is finite and terminates in some highest predicate (Posterior Analytics 1.22 82b37–83a1, 83a39–b8). If the chain were infinite, he argues, the subject A would have infinitely many items predicated essentially of it. In this case, the essence of A would be infinitely complex, hence undefinable and unknowable by us because “it is impossible to go through infinitely many items in thought” (83b6–7). Aristotle rejects this consequence and contends that all essences are finite (Metaphysics α 2 994b16–23). This assumption plays an important role in his argument that scientific demonstrations are finite, and in his refutation of sceptical challenges to possibility of scientific knowledge (Posterior Analytics 1.19–22). Consider an upward chain of essential predications specifying Socrates’ essence (Posterior Analytics 1.22 83b1–5): … animal is predicated essentially of biped animal, biped animal is predicated essentially of human being, human being is predicated essentially of Socrates. Given the finitude of essential predication, any such chain must terminate in a highest essential predicate. Aristotle does not specify what these highest essential predicates are. However, he is clear that they do not include the predicate being (or, existence). Aristotle regards being (on) as a maximally general predicate that is predicated of everything.15 For Aristotle, everything is a being. At the same time, he holds that being is not a genus or differentia of anything, and hence is not predicated essentially of anything.16 The same is true for the predicate one (hen): while it is maximally general and predicated of everything, it is not predicated essentially of anything. Aristotle holds that grasp of essences is indispensable for scientific knowledge. In fact, for Aristotle, to have scientific knowledge of a thing just is to grasp the essence of that thing.17 Following Plato, he also maintains that essential predication is endowed with explanatory power. In particular, he takes genera to explain why their subjects have the attributes they have. If all animals have the power of perception, then horses, being a species of the genus animal, have the power of perception because they are animals. More generally, Aristotle holds:18 If A is predicated of B and B is a genus of C, then C is A because C is B. This shows that, for Aristotle, genera are not arbitrary collections of particulars, but universals that play a central role in scientific explanation. Aristotle employs, as David Charles puts it, an “explanation-involving conception of genus, species, and differentiae”.19

24

Ancient

Accordingly, Aristotle devotes a significant portion of his theory of science to investigating the interrelations between scientific explanation and real definitions specifying essences.20

1.3

Aristotle: Essence and Form

In addition to his essentialism, Aristotle is committed to hylomorphism. He maintains that perceptible objects are hylomorphic compounds consisting of matter (hylê) and form (eidos). In many contexts, Aristotle treats essentialism separately from hylomorphism, investigating the notion of essence without recourse to the distinction between matter and form. This nonhylomorphic approach is found, for example, in the Topics, the Posterior Analytics, and in Metaphysics Z 4–5. Aristotle describes the non-hylomorphic treatment of essence as “logical” (logikôs).21 In other contexts, however, Aristotle goes beyond the “logical” approach and studies essence in connection with the distinction between matter and form. This approach is especially prominent in the Physics, in Metaphysics Z 7–11, and in Metaphysics H. The close relationship between essentialism and hylomorphism in Aristotle’s natural philosophy can be seen, for example, in his discussion of causation in the Physics. Aristotle distinguishes four types of cause (aition): material cause, formal cause, efficient cause, and final cause (Physics 2.3). Using the example of a house, the four causes can be illustrated as follows: Material cause of a house: bricks, wood, and other materials Formal cause of a house: the capacity to shelter bodies and goods Efficient cause of a house: the housebuilder Final cause of a house: sheltering bodies and goods The material cause is “that from which as a constituent something comes to be” (194b24). In the case of a house, these are the materials of which the house is composed. The formal cause of a thing is its essence (to ti ên einai, 195a20). Alternatively, Aristotle describes a thing’s formal cause as its “form” (eidos, 194b26, 195a21). Thus, Aristotle takes form and essence to coincide, with both of them playing the role of formal cause. For example, the capacity to shelter bodies and goods is both the form and the essence of a house. The final cause of a thing is its end (telos). Aristotle claims that a thing’s formal cause and its final cause often “converge upon one thing” (Physics 2.7 198a24–7). However, this should not be taken to mean that the formal and final cause are identical. For example, while a house’s final cause is the activity of sheltering bodies and goods, its formal cause is not this activity but rather the corresponding capacity.22 A house possesses the capacity that is its essence even at those times when it is not performing the activity that is its end. Now, Aristotle maintains that activity is prior in definition to capacity, so that the capacity to perform an activity is to be defined by reference to this activity.23 Thus, the full real definition specifying the essence of a house includes a reference to the end of a house. According to Aristotle, the essence of a house explains why the materials of which it is composed constitute a house: Why are these things here, e.g., bricks and stones, a house? It is clear that what is sought is the cause (aition). And this is, to put it logically, the essence (to ti ên einai), which in some cases is that for the sake of which it is [i.e., the final cause], as presumably in the case of a house or a bed. (Aristotle, Metaphysics Z 17 1041a26–30)

25

Marko Malink

Why do these bricks and stones constitute a house? The answer is: because the essence of a house holds of them (1041b4–6). It is because they are capable of sheltering bodies and goods that these bricks and stones constitute a house. Of course, the capacity to shelter bodies and goods is not the essence of the bricks and stones. Instead, it is the essence of the house. Still, this essence holds non-essentially of the bricks and stones. The same account applies not only to artefacts but to natural substances as well. Why does this body constitute a human being? Answer: because the essence of a human being holds of the body (1041b6–7). Thus, the essence of hylomorphic compounds has explanatory power: for any hylomorphic compound X, the fact that its essence holds of its matter explains why the matter constitutes X (cf. Chapter 27 in this volume). In Metaphysics Z, Aristotle investigates the essence of hylomorphic compounds. In doing so, he identifies essence with form. He maintains that, for any hylomorphic compound, its essence is its form (Z 7 1032b1–2, Z 10 1035b14–16). Now, Aristotle takes the form of animals to be their soul (psychê, Z 10 1035b14–16). Hence, the essence of animals is their soul. As such, the soul is a cause explaining why a given animal is alive and exists (De anima 2.4 415b7–15). For example, Sappho’s essence is her soul, and her soul’s presence explains why Sappho is alive and exists. In Metaphysics Z, Aristotle regards the form of hylomorphic compounds such as Sappho as a “primary substance” (prôtê ousia).24 In doing so, he departs from his practice in the Categories, in which particular hylomorphic compounds are not analyzed into matter and form but instead are themselves regarded as “primary substances” (Categories 5). While Sappho counts as a primary substance in the Categories, in Metaphysics Z it is not Sappho but her soul that counts as a primary substance.25 Thus, in the hylomorphic framework presented in Metaphysics Z, Sappho’s essence, her soul, is a substance over and above Sappho. Given that Sappho’s essence is distinct from Sappho, one may wonder whether Sappho’s essence (her soul) has an essence distinct from itself. If it does, there is a danger of an infinite regress of essences. Aristotle blocks the regress by granting that Sappho’s soul has an essence, but claiming that this essence is the same as Sappho’s soul (Metaphysics Z 6).26 Thus, while Sappho’s essence is distinct from Sappho, the essence of Sappho’s essence is the same as Sappho’s essence. More generally, the essence of every hylomorphic compound (i.e., its form) is distinct from this compound, but is the same as its own essence. Aristotle relies on the identification of essence and form to determine what should be mentioned in the definition of a thing. For example, should the definition specifying the essence of a thing mention the parts of that thing (Metaphysics Z 10 1034b20–4)? In some cases, the answer is yes, and in other cases the answer is no. For instance, the definition of the syllable “BA” should mention the letters “B” and “A,” but the definition of a circle should not mention the various segments of the circle (1034b24–8). Both the syllable and the circle are divided into the letters and segments, respectively; but in the one case the resulting parts are mentioned in the definition of the whole, and in the other case they are not. The reason for this, according to Aristotle, lies in the different hylomorphic structures exhibited by the syllable and the circle (1035a1–22). The letters of a syllable belong to the syllable’s form and are mentioned in an account of this form. The letters should therefore be mentioned in the definition specifying the essence of the syllable. By contrast, the segments of a circle do not belong to the form of the circle but only to its matter. These “material” parts of the circle should therefore not be mentioned in its definition. For the same reason, the fingers of a human being should not be mentioned in the definition of human being (1035b3–14). Nor should other parts of the human body be mentioned in the definition of human being (1035b20-1). At the same time, Aristotle

26

Ancient

warns us against adopting an overly abstract definition of human being that would mislead us into “believing that it is possible for a human being to exist without its parts just as a circle can exist without the bronze from which it is made” (Z 11 1036b26–8).27 Aristotle also appeals to his hylomorphism to explain why men and women do not constitute two different species of humans (Metaphysics I 9). While Sappho differs from Socrates in that she is female and he is male, this does not mean that they belong to two different species of humans. Instead, they both share the same lowest species, namely, human being. More generally, male and female do not constitute different species of any kind of animal.28 According to Aristotle, this is because the difference between male and female does not pertain to an animal’s form but only to its matter and its body (1058b21–3). But matter, Aristotle argues, does not give rise to a differentia, and hence not to a species (1058b6). Just like a wooden circle and a bronzen circle differ in matter but not in species, the same is true for a male human and a female human. Thus, the difference between male and female does not pertain to the essence of human beings. Sappho is not essentially female, and Socrates not essentially male.29 Aristotle maintains that universals are definable, but perceptible particulars are not.30 While the species human being can be defined by genus and differentia as rational animal, no such definition is available for particulars such as Sappho and Socrates. Nor is there any other kind of definition specifying a unique essence for each of these particulars. Instead, Sappho and Socrates share the same definable essence, corresponding to the lowest species human being.31 As far as their definable essence is concerned, Sappho and Socrates are indistinguishable. Aristotle sometimes speaks of the essence of particulars, denoting it by phrases such as “what it is for Kallias to be” (1022a27), “what it is for you to be” (1029b14–15), and “for Socrates to be” (1032a8). However, the fact that Aristotle uses such phrases does not mean that he took Kallias’ essence to be distinct from Sappho’s essence.32 We know that the essence of these particulars is their form (i.e., their soul). At the same time, Aristotle is notoriously silent on the individuation of forms (or souls). As a result, there is considerable scholarly debate about whether Aristotle countenances particular forms, that is, forms which are present in no more than one hylomorphic compound.33 If he did, then Sappho’s form is distinct from Socrates’ form, and hence her essence is distinct from his essence. But even if Sappho has such a particular essence distinct from the essence of every other human being, this particular essence is not definable.34

1.4

Hellenistic Period and Late Antiquity

Definitions played an important role in Stoic philosophy. However, it is not clear whether the Stoics countenanced the idea of a real definition specifying a thing’s essence. Instead, the extant sources suggest that the Stoics did not require definitions to specify the essence of the thing defined.35 Thus, Alexander of Aphrodisias, a leading Peripatetic philosopher of the early 3rd century AD, criticizes Stoic conceptions of definition on the grounds that they lack any essentialist import.36 More generally, essentialism is largely absent from Stoic thought. On the other hand, essentialism was prominent in the work of Peripatetic and Neoplatonic philosophers throughout late antiquity.37 In particular, Aristotle’s theory of essential predication plays a central role in Porphyry’s Isagoge (3rd century AD). In this treatise, Porphyry lays out elements of the theory of predication introduced by Aristotle in the Topics and the Categories, with special attention to the type of essential predication that obtains between a genus and its species.38 Porphyry’s Isagoge served as a standard introduction to

27

Marko Malink

philosophy for more than 1,000 years in Byzantium, the Arabic world, and in the Latin West. It became one of the main ways in which Aristotle’s essentialism was transmitted to the medieval world.39

1.5

Related Topics

Medieval; Contemporary (Analytic Tradition); Non-Modal Conceptions of Essence; Essences of Individuals; Natural Kind Essentialism; Essence, Grounding, and Explanation; Unity

Notes 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

See also Meno 72b1. See Silverman 2002: 28–48. Cf. Vlastos 1969. See Sophist 217b2–3, 218c1–7, 221c6. See Brown 2010, Gill 2010. For Antisthenes’ views regarding essence and definition, see Prince 2015: 445–85; Malink 2024. Topics 7.5 154a31–2, Metaphysics Z 5 1031a11–12. Topics 1.8 103b14–16, 6.1 139a28–9, 6.4 141b25–7, 6.6 143b19–20, 7.3 153b14–15. Metaphysics Δ 30 1025a30–2, De partibus animalium 1.3 643a27–31. Barnes 1993: 120. Moreover, Aristotle’s theory of essential predication can be seen to underlie the system of modal syllogisms developed by Aristotle in Prior Analytics 1.8–22 (see Patterson 1995, Malink 2013). Metaphysics Δ 6 1016a30, Δ 10 1018b4–6. For Aristotle’s use of atomon in connection with lowest species, see Metaphysics I 8 1058a17–22, De partibus animalium 1.3–4 643a1–644a33. Cf. also Metaphysics K 1 1059b36–7. See Corkum 2009, Cohen 2013: 233–40. See Topics 2.2 109a37–8, 4.1 121a7–9, 4.3 123b25–7, Categories 11 14a20–3. Zeta 1 1028a35–6 and 5 1030b14–28. See Loux 1991: 72–108. Metaphysics B 3 998b21, I 2 1053b20–1, Topics 4.6 127a28–30. Metaphysics B 3 998b22–7, Z 16 1040b18, H 6 1045b2–6, I 2 1053b22–3, Topics 4.6 127a26–38. Metaphysics Z 6 1031b6–7 and 1031b20–1. See also Z 1 1028a36–b2. Posterior Analytics 2.14 98a1–12. See Malink 2022: 203–8. Charles 2000: 239–43. Posterior Analytics 2.1–13. See Goldin 1996, Charles 2000, Ferejohn 2013: 98–155, Bronstein 2016: 69–224. Metaphysics Z 4 1029b13. See Burnyeat 2001: 6–8 and 19–25. See Rosen 2014. Metaphysics Θ 8 1049b12–17. Z 7 1032b1–2, Z11 1037a5, 1037a28–29. See Code 1986: 431–9. Metaphysics Z 6 1031b18–1032a11, Z 11 1037a33–b7. See Code 1985: 110–23. Metaphysics Z 11 1036b21–1037a5. For a discussion of this difficult passage, see Meister 2020. Metaphysics I 9 1058a29–32, De generatione animalium 1.23 730b33–731a1. See Connell 2016: 290–1. Metaphysics Z 10 1035b33–1036a5, Z 15 1039b27–1040a7. Code 1984: 15–20; 1985: 112. This has been pointed out by Albritton 1957: 703; Modrak 1979: 377 n. 40. For an excellent overview of this debate, see Gill 2005: 230–3. See Code 1986: 438–9, Crivelli 2010: 396–408. Alexander of Aphrodisias, Commentary on Aristotle’s Topics 42.27–43.8. See Chiaradonna 2023: 91–184. See Barnes 2003. For helpful comments on earlier versions of this article I would like to thank István Bodnár, Kit Fine, Kathrin Koslicki, Jessica Moss, Asya Passinsky, Mike Raven, and Charlotte Witt.

28

Ancient

References Albritton, R. (1957) “Forms of Particular Substances in Aristotle’s Metaphysics,” Journal of Philosophy 54, 699–708. Barnes, J. (1993) Aristotle’s Posterior Analytics, 2nd edn., Oxford: Oxford University Press. Barnes, J. (2003) Porphyry: Introduction, Oxford: Clarendon Press. Bronstein, D. (2016) Aristotle on Knowledge and Learning: The Posterior Analytics, Oxford: Oxford University Press. Brown, L. (2010) “Definition and Division in Plato’s Sophist,” in: D. Charles (ed.) Definition in Greek Philosophy, Oxford: Oxford University Press. Burnyeat, M. F. (2001) A Map of Metaphysics Zeta, Pittsburgh: Mathesis. Charles, D. (2000) Aristotle on Meaning and Essence, Oxford: Oxford University Press. Chiaradonna, R. (2023) Ontology in Early Neoplatonism, Berlin: de Gruyter. Code, A. (1984) “The Aporematic Approach to Primary Being in Metaphysics Z,” Canadian Journal of Philosophy 14, 1–20. Code, A. (1985) “On the Origins of Some Aristotelian Theses About Predication,” in: J. Bogen and J. E. McGuire (eds.) How Things Are: Studies in Predication and the History of Philosophy and Science, Dordrecht: Reidel. Code, A. (1986) “Aristotle: Essence and Accident,” in: R. E. Grandy and R. Warner (eds.) Philosophical Grounds of Rationality: Intentions, Categories, Ends, Oxford: Oxford University Press. Cohen, S. M. (2013) “Accidental Beings in Aristotle’s Ontology,” in: G. Anagnostopoulos and F. D. Miller (eds.) Reason and Analysis in Ancient Greek Philosophy: Essays in Honor of David Keyt, Dordrecht: Springer. Connell, S. M. (2016) Aristotle on Female Animals: A Study of the Generation of Animals, Cambridge: Cambridge University Press. Corkum, P. (2009) “Aristotle on Nonsubstantial Individuals,” Ancient Philosophy 29, 289–310. Crivelli, P. (2010) “The Stoics on Definition,” in: D. Charles (ed.) Definition in Greek Philosophy, Oxford: Oxford University Press. Ferejohn, M. T. (2013) Formal Causes: Definition, Explanation, and Primacy in Socratic and Aristotelian Thought, Oxford: Oxford University Press. Gill, M. L. (2005) “Aristotle’s Metaphysics Reconsidered,” Journal of the History of Philosophy 43, 223–241. Gill, M. L. (2010) “Division and Definition in Plato’s Sophist and Statesman,” in: D. Charles (ed.) Definition in Greek Philosophy, Oxford: Oxford University Press. Goldin, O. (1996) Explaining an Eclipse: Aristotle’s Posterior Analytics 2.1–10, Ann Arbor: University of Michigan Press. Loux, M. J. (1991) Primary Ousia: An Essay on Aristotle’s Metaphysics Z and H, Ithaca, NY: Cornell University Press. Malink, M. (2013) Aristotle’s Modal Syllogistic, Cambridge, Mass.: Harvard University Press. Malink, M. (2022) “The Discovery of Principles in Prior Analytics 1.30,” Phronesis 67, 161–215. Malink, M. (2024) “Antisthenes on Definition: Metaphysics H 3,” in: D. Bronstein, T. Johansen, and M. Peramatzis (eds.) Aristotelian Metaphysics, Ancient and Modern: Essays in Honour of David Charles, Oxford: Oxford University Press. Meister, S. (2020) “Aristotle on the Purity of Forms in Metaphysics Z.10–11,” Ergo 7, 1–33. Modrak, D. K. W. (1979) “Forms, Types, and Tokens in Aristotle’s Metaphysics,” Journal of the History of Philosophy 17, 371–381. Patterson, R. (1995) Aristotle’s Modal Logic: Essence and Entailment in the Organon, Cambridge: Cambridge University Press. Prince, S. (2015) Antisthenes of Athens: Texts, Translations, and Commentary, Ann Arbor: University of Michigan Press. Rosen, J. (2014) “Essence and End in Aristotle,” Oxford Studies in Ancient Philosophy 46, 73–107. Silverman, A. (2002) The Dialectic of Essence: A Study of Plato’s Metaphysics, Princeton: Princeton University Press. Vlastos, G. (1969) “Reasons and Causes in the Phaedo,” Philosophical Review 78, 291–325.

29

2 MEDIEVAL Gloria Frost

Medieval thinkers maintained that things in the world have mind-independent essences. A term they often used for essence is quiddity (quiditas). This term derives from the Latin term quid which means what. The essence or quiddity of a thing is its whatness. The essence of a thing includes the group of features that is most central or important to a thing since these are the features that make it what it is. Accordingly, a thing’s essential features are included in its definition. Essential features are contrasted with accidental features which are those features that are more peripheral or less important to a thing. Socrates’s humanity and rationality, for example, are essential features, while his whiteness or his being non-musical are accidental. This essay introduces some important medieval commitments about essences, as well as some of the central debates regarding essences in the period. The essay focuses in particular on the Latin scholastic tradition, yet it will also consider some earlier influences on their views, particularly from the medieval Islamic tradition. The first section examines medieval views on basic issues about essences. The section begins by discussing medieval conceptions of the distinction between essential and accidental features. Next the section discusses medieval positions on whether there are only species essences or also individual essences. Lastly, the section examines medieval views on the knowability of essences. The second section of the of the essay focuses on two medieval debates about essences. The first debate had to do with which features of material beings were essential to them. The Latin scholastic authors on whom this essay focuses adopted Aristotle’s theory of hylomorphism. This theory maintains that material substances are composed of matter and form. Scholastic authors debated about whether the essences of material substances were constituted merely by their forms or if their essences also contained their matter. The second debate focused on the question of whether there is a real distinction between the essence of a created substance and its existence. The debate was about whether in addition to their essence a creature was composed by a distinct actualizing principle through which the essence was the essence of a real being, rather than a merely possible one. Though this question is foreign to contemporary philosophy, it was one of the most pressing concerns regarding essences for medieval scholastic thinkers.

30

DOI: 10.4324/9781003008750-4

Medieval

2.1

Basic Issues: The Essential vs. the Accidental, Whether There are Individual Essences or Only Species Essences, the Knowability of Essences

In contemporary philosophy, one of the main ways to demarcate the distinction between essential and accidental features is in terms of modality. In this view, a thing’s essential features are whatever features it has necessarily and its accidental features are whatever features it has contingently. According to this view, Socrates’s rationality is essential to him by virtue of there being no possible world in which he is not rational and his non-musicality is accidental to him in so far as there are possible worlds in which he exists as not musical. Medieval thinkers demarcate the difference between essential and accidental features in a very different way. In the medieval period it was standard to think about the difference between essential vs. accidental features in terms of ontological dependence or fundamentality.1 Essential features are those features of a thing which do not depend on any of its prior features or any external cause acting upon it. Accordingly, essential features are fundamental. They are features that a thing has which admit of no further foundation or cause. Accidental features, by contrast, are those features which a thing has in virtue of its own prior features or an external cause’s action upon it. Thomas Aquinas (1224/5–74) provides some examples to help illustrate this view: [W]hatever is in something beyond its essence must be caused, either by the principles of the essence, just as proper accidents which follow from the species as risible follows from human and is caused by the essential principles of the species, or by something exterior, just as heat is caused in water by fire.2 As Aquinas explains, a human being’s risibility is an accidental feature since humans are risible in virtue of being human and more specifically, being rational. There is a further, more foundational feature that explains why humans are risible. There was discussion in the period about how to understand the manner in which accidental features arose from a substance’s essential features. Aquinas insists that it is not by way of efficient causation, but rather as a “natural resultant”.3 This is to say that essence is not an agent that causes further features as fire causes heat water, but rather further features naturally follow from essential features in so far as the essence cannot be caused to exist without also causing these further features. Aquinas uses the example of heat in water to illustrate the second way in which accidental features can be dependent. In this case, water’s heat depends on fire’s action. Heat is not a foundational feature of water, but rather one that is explained by the activity of an external cause. Not only did scholastic authors demarcate the difference between essential and accidental features in a non-modal way, but they also held that the division between the essential and the accidental did not align with the division between the necessary and the contingent. Following Aristotle, they maintained that there are some accidents that are necessarily possessed by their subjects. These necessary accidents, which are referred to as propria, are accidents which a subject has in virtue of its essential principles.4 Risibility, for example, is necessarily present whenever rationality is present. Thus, it is a necessary feature of its subject, though it is not an essential one since it is possessed in virtue of the prior feature of rationality. Though scholastic authors do not explicitly discuss contingent essential features, their theological beliefs suggest that they would have been open to such features. Christians believe that Jesus Christ, as God, has a divine nature and that he freely assumed a human nature as well. Thus, Jesus Christ is essentially human, but only contingently so, since he could have possibly not become human. Scholastic authors were in firm agreement that there was an essence which corresponded to each biological species. This is to say that there is a core set of characteristics or features

31

Gloria Frost

which makes each species the species which it is. There was less certainty and agreement about whether in addition to species essences there were also individual essences which corresponded to each member of a species. For instance, is there some set of characteristics or features which makes Socrates be Socrates? The question of whether there are individual essences is closely tied to the question of what it is that individuates different members of the same species. A common scholastic view held that the principle which individuates numerically distinct members of a species was matter. For example, Socrates and Plato each have the same human essence, yet they are different individuals since they are composed of different matter in which the human essence is instantiated. Many in the period thought that matter was in itself unintelligible and undefinable since it is merely a substratum in which essences inhere.5 Since matter was only the component that each member of the species had beyond its species essence and matter has no essential content of its own, it seems to follow from these suppositions that there were no individual essences. The core of what Socrates is is human and there are no other features outside of the essence of humanity which make him be what he is. Some interpreters attribute this line of reasoning against individual essences to Thomas Aquinas.6 Yet, the proper interpretation of Aquinas’s views is subject to debate.7 John Duns Scotus (1265/6–308) is an example of a scholastic figure who explicitly discusses the question of whether there are individual essences and argues in the affirmative.8 Scotus argues that whatever principle individualizes essences must combine with the essence to form an essential unity, i.e. an essence, or else the individual substance which results from the species essence and principle of individuation would be a mere accidental unity. This is to say that the individual would not be an essential unity unless it had its own essence. In Scotus’s view, the essence of the individual is constituted by its individuating principle and its species essence. Scotus rejects that matter or any combination of qualities can be the principle of individuation.9 In his view, the principle of individuation is haecceity, i.e. an individual difference, which is a reality that is individual of itself, which is to say that its individuation is brute and not dependent upon some further individuator. Accordingly, in Scotus’s view, an individual such as Socrates’s individual essence contains only his species essence, i.e. humanity, and his individual difference. Accidents, such as his particular size or color, are left out of his individual essence. Since a given individual difference and its species essence together form an essential unity, i.e. an individual essence, Scotus denies that a particular entity’s individual difference could combine with any other type of species essence. For example, Socrates’s individual difference cannot combine with the canine species essence to form a dog who is Socrates. Socrates’s individual difference is essentially united with the human species essence and thus, Socrates is necessarily human.10 Scotus did not think it was possible for human beings to cognize individual essences. But this was due only to our cognitive limitations. He thought that the essences of individuals were in principle intelligible, and in fact, known by God. Scholastic thinkers routinely question the human ability to fully cognize even the essences of species. A standard assumption in the period was that human knowledge originates in sense experience. Since the senses can only grasp sensible accidents, such as color, shape, texture and sounds, and essences are not objects of sense experience, the difficulty was posed of how we can have knowledge of the essences of material beings. Aquinas addresses this difficulty by appealing to the ontological relationship between a substance’s essence and its proper accidents. As we just saw above, scholastic thinkers maintain that substances possessed their proper accidents in virtue of their essences. Proper accidents are those accidents which naturally result from the essence. Because essences were explanatory of a substance’s proper accidents, Aquinas reasoned that through knowing the accidents of a being, some knowledge

32

Medieval

of its essence could be gained. Aquinas writes that “accidents reveal or give knowledge of essence just as proper effects give knowledge of a cause …”11 Just as we come to know something about the sun by knowing light, Aquinas thinks something of the essence of a substance can be known by knowing its proper accidents. Elsewhere he describes accidents as “signs of essential differences which are unknown to us”.12 Though Aquinas believes that proper accidents can give us some knowledge of essences, he believes that our knowledge of essences will always be incomplete. He famously writes: “[N]o philosopher is able to perfectly investigate even the nature of one fly …”13 There were other scholastic figures who were more pessimistic than Aquinas about the human ability to know essences because they rejected the view that essences could be known by knowing accidents. Medieval thinkers believed that some realities which exist are more perfect than others. Substances, for example, are more perfect than accidents because they are independent, while accidents are less perfect since they only exist in dependence on a substance. Francis of Marchia (c. 1290–c. 1344) argues that nothing can cause a distinct concept of a reality that is more perfect than itself.14 Thus, by knowing the accidents of a substance, we cannot gain a concept of the substance’s essence. Marchia thinks that there is an in-principle difficulty with trying to gain knowledge of a more perfect concept on the basis of a less perfect concept since the higher concept contains content which cannot be produced in the mind by grasping the lower concept. Thus, while Aquinas thinks that we can infer something about what a fly is in itself by experiencing the proper accidents that arise from its essence, Marchia contends that despite experiencing the fly through our senses we know nothing about what makes a fly be what it is.

2.2

Two Debates: Is Matter Part of the Essences of Material Substances? Are Essence and Existence Really Distinct?

Following Aristotle, medieval scholastic thinkers maintained that material substances were composites of two metaphysical principles, matter and substantial form.15 Matter is a receptive component which is of itself imperfect and incomplete, while forms are principles of perfection and completion. By receiving a substantial form, matter composes an actual material substance of a determinate type. For example, when matter receives a feline form, matter and that form together compose a particular cat. Forms are not only principles of actuality, but they are also principles of species. This is to say that it is in virtue of being actualized by a particular type of substantial form that matter composes a particular type of substance. For example, Felix the cat’s substantial form is what determines that his matter composes a cat rather than a dog. Given the important role that substantial form plays in determining the species of a substance, scholastic thinkers debated about whether the essences of material substances are to be identified with their forms or whether matter is part of the essence and definition of material substances. The significance of the question can be seen by considering an example taken from Aristotle and regularly referenced by the scholastics. Consider the two qualities of concavity and snubness. The essence and definition of concavity does not include the matter which is subject to concavity. To be concave is to have a surface that curves inward. Yet, nothing in the essence or definition of concavity specifies what type of material a concave object must be made of. For example, plastic, rubber and wooden objects can all be concave. By contrast, the essence and definition of snubness makes reference to a type of matter in which snubness is instantiated. Snubness is a concavity of the nose or flesh. Plastic, wooden or bronze objects cannot be snub since it is of the essence of snubness to be in a particular type of matter, namely a nose or flesh. What scholastic thinkers wondered about

33

Gloria Frost

is whether the essences of material substances, such as dogs, trees and humans, were more like concavity or more like snubness. Did such essences include reference to particular types of matter in which other more formal characteristics were instantiated? For example, is it part of the essence of Socrates to have the material parts of flesh and bones or does his essence only contain his formal characteristics such as rationality? The scholastic debate about this question was heavily shaped by texts in Aristotle and his Islamic commentators Avicenna (Ibn Sina) (c. 970–1037) and Averroes (Ibn Rushd) (1126–98) Albert.16 The first position in the debate, which is attributed to Averroes, maintains that the essence of a material substance is identical with its form, or put otherwise, matter does not enter into the essence or definition of a material substance. For example, flesh and bones are not part of the essence or definition of what a human is. In support of this view, it is argued that matter is not included in the essence of a material being because matter is that which receives essence and thus, must be itself outside of the essence. The second position, which is attributed to Avicenna, maintains that the essence and definition of any material substance must contain matter since material substances are essentially material. In this view, the essence of Socrates must include flesh and bones. It was argued in favor of this view that material substances are not purely form, rather they are composites of matter and form. Thus, their essences must include both their formal features and the matter out of which they are made. Another argument in favor of this position appeals to the difference between mathematical and physical objects. Mathematical forms such as circles or triangles can be in many different types of matter. For example, there can be a circle made out of bronze or gold or wood or ice. Physical objects, however, are not indifferent to the type of matter out of which they are composed. For example, to exist and be human, it seems that Socrates must be composed of flesh and bones. Unlike a circle, he cannot likewise be composed of wood or bronze. Some scholastics, such as Albert the Great (c. 1206–80), discussed this debate, but did not take a clear stand on which position was correct.17 Aquinas and Scotus sided with the Avicennian position that the essences of material substances contain matter.18 In clarifying this position, Aquinas introduces an important distinction between two types of matter: common matter and individual matter. Common matter is the type of matter that all individuals of a certain kind are made of. In the case of humans, common matter is flesh and bones. Individual matter is the determinate matter that a particular individual is made of. For example, Socrates is made up of flesh and bones that have a particular size and weight. Aquinas clarifies that it is common matter, rather than individual matter, that is contained in the essence and definition of a material substance. He writes: I call ‘designated matter’ that matter which is considered under determinate dimensions. This matter is not posited in the definition of man in so far as he is man, but would be posited in the definition of Socrates if Socrates were to have a definition. In the definition of man however non-designated matter is posited. For these bones and this flesh are not posited in the definition of man, but rather flesh and bones absolutely, which is the nondesignated matter of man.19 For example, it is of the essence of a human to be made of flesh and bones. Yet, no exact dimensions of flesh and bones belongs to the essence of a human being. If individuals were to have essences and definitions, Aquinas acknowledges that it would belong to the essence and definition of the individual to have matter under determinate dimensions. Scotus too adopted the distinction between common and individual matter and similarly claims that it is

34

Medieval

common, rather than individual matter that belongs to the essence of a material substance. He refers to individual matter as “accidental matter”.20 One of the most pressing questions about essence in thirteenth and fourteenth century Latin philosophy was the question of whether the essence of a created substance is distinct from its existence and if so, what is the nature of this distinction. Questions about the nature of the relationship between essence and existence are not central questions in contemporary philosophy, so some explanation is necessary. What medieval thinkers wondered is whether in addition to their essence, created substances are also constituted by a distinct actualizing principle which makes their essence constitute a real, existing substance. The question arose in scholastic philosophy in part because of remarks of authorities and in part because of a philosophical concern to differentiate the metaphysical structure of creatures from the metaphysical structure of God. In his De Hebdomadibus, Boethius (c. 470–524) claims that in composite created beings, a being’s existence and that which it is are different.21 While Boethius, should not regarded as a proponent of the real distinction defended by later scholastics, his texts nevertheless were often quoted in later discussions. Avicenna is another earlier source who spurred on scholastic discussions of the relationship of essence and existence. Scholastic figures regarded him as defender of an extreme version of the real distinction between essence and existence.22 In addition to the impetus of authorities, scholastic thinkers were motivated to discuss the relationship between essence and existence due to conceptual concerns. Medieval thinkers were committed to the view that there must be some intrinsic metaphysical difference between finite, created, contingent substances and God as a necessary, uncreated and infinite substance. It was standard to view God as metaphysically simple. This is to say that, in addition to lacking integral parts, in God there is no composition of substance and accident or matter and form or any other metaphysical constituents. It was commonly held that God’s absolute simplicity was required for God to be the primal, independent being since non-simple beings depend both on their constituents and on an efficient cause which unites their constituents. The fundamental difference in the metaphysical structure between God and creatures was thought to be some form of composition which is found in all creatures and absent in God. We have already seen above, that scholastic figures were committed to hylomorphism (i.e. matter-form composition) in material substances. In addition to material substances, scholastic figures held that there were created immaterial substances. Some medieval thinkers, such as Bonaventure (c. 1217–74) and other Franciscans, held that hylomorphic composition was universal. In addition to the physical matter which composes (together with substantial form) material substances, they posited that there was also spiritual matter which united with the substantial forms of created immaterial substances to compose them. For universal hylomorphists, matter-form composition was sufficient to differentiate created substances from the metaphysically simple God.23 Universal hylomorphism, however, was not universally accepted. Thomas Aquinas argued that “spiritual” matter was an oxymoron. Matter as such accounts for why a substance has three dimensions, extended in space and thus, it makes no sense to posit spiritual matter.24 Some thinkers who reject universal hylomorphism turn to essence and existence as a type composition in creatures which differentiates them metaphysically from God. Aquinas is a prominent scholastic figure who defends the essence-existence distinction in creatures. Aquinas views essence and existence as two complementary principles of created substances. Essence is the principle of a being in virtue of which a created substance is what it is and has quidditative content, while existence is the principle in virtue of which a created substance actually exists. In Aquinas’s view, essence and existence do not come together as two things, but rather as complementary principles ordered to each other as potentiality and

35

Gloria Frost

actuality. The relationship of essence to existence is similar to that of matter and form. When form actualizes matter and matter receives form, a material being is composed. Similarly, when existence actualizes an essence and that essence receives existence, an actual substance is composed. Aquinas sees existence as a principle that actuates and makes an essence compose a real existing thing rather than a mere possible being. Since existence is that which makes a form or nature real, Aquinas thinks of existence as a principle of perfection. He writes: “Existence is the actuality of all acts and on account of this it is the perfection of all perfections”.25 In Aquinas’s view, God’s existence is not received in a distinct essence principle. His essence principle and existence principle are identical.26 Thus, God is pure or unlimited existence and thus, God has unlimited or infinite perfection.27 Aquinas sometimes tries to establish the conclusion that essence and existence must be distinct principles in creatures by arguing that there can be just one being who is pure, unreceived existence.28 He reasons that the way that beings are multiplied is by adding something outside of their essence which individuates the essence. Thus, if there were multiple beings that were pure, unreceived existence, some other principle would be added to their existence to individuate them. The resulting composition would amount to an existenceessence composite since the individuator would be a formal-essential principle that is composed with existence. If there can be just one being who is unreceived, pure existence, then all other beings other than God must be composites of existence and essence.29 At other times, Aquinas argues for the conclusion that creatures are composites of essence and existence without presupposing that there is a God who is pure existence. In one of his early works, he writes: [E]very creature has finite existence. But existence not received in another is not finite, rather it is absolute. Therefore, every creature has existence received in something, and so it is necessary that it has at least two [principles], namely existence and that which receives existence.30 The argument above begins from the observation that no creature is infinite. It is clear that the existence of creatures is limited in various ways. Aquinas reasons that since the existence of creatures is limited, there must be some receptive principle which limits their principle of existence. As already mentioned, Aquinas sees existence as a principle of perfection. Furthermore, he believes that it was not self-limiting. Existence not received by any distinct nature or essence would be infinite.31 Since creatures clearly are limited, they must be composites of existence and another principle which is a source of imperfection, namely essence. William of Ockham (c. 1280–c. 1349) is a prominent medieval figure who rejects the real distinction between essence and existence. Ockham argues that if essence and existence are really distinct, then God must be able to preserve each separately. He assumes that if two principles are really distinct, then they are really separable.32 It seems absurd to posit an existence that is separable from essence since it would not be the existence of any essence. Likewise, it seems absurd to posit an essence that is separable from existence since it would be a real essence which is nevertheless not actualized. Thus, Ockham concludes that essence and existence cannot be really distinct.33 In response to the defender of the real distinction, Ockham claims that one cannot conclude from an essence’s previous non-existence that essence and existence are really distinct since when the essence did not exist, it was not an essence without existence. Rather, it was nothingness. Ockham’s reasoning can be applied to an example.34 A defender of the real distinction may be tempted to think that since there was

36

Medieval

a time when Socrates’s essence did not have actual existence, it follows that there must be some principle that is really distinct from the essence that accounts for the difference of the essence when it goes from being possible to actual. To this Ockham responds that it is not the case that something is received by an essence to transition it from possibility to actuality, because before the essence exists, it simply is not. In Ockham’s view, the terms “essence” and “existence” signify one and the same thing, an actually existing essence. Other medieval figures tried to adopt an intermediate position. Henry of Ghent (c. 1217–93) claims that existence does not add a distinct reality to essence, but rather it adds a distinct “intention” to essence. “Intentio” in Latin literally means “a stretching forth” or “reaching out toward”. So, in claiming that existence adds an intention to essence, he is suggesting that existence adds to an essence an ordering to something outside of itself. More specially, Henry claimed that existence adds to an essence a relation of dependence on God as efficient cause.35 For instance, before Socrates exists his essence does not depend on God as an efficient cause since God is not causing his essence to exist when he is not. When Socrates exists, his essence then has a relation of dependence on God as God is now efficiently causing his essence to be. John Duns Scotus is another medieval figure who maintained that the distinction between essence and existence was an intermediate type of distinction, rather a real or merely conceptual distinction. Scotus himself did not give much sustained attention in his works to discussing the precise nature of the essence-existence distinction. Accordingly, the proper interpretation of his views on this matter has been subject to scholarly debate.36

2.3 Conclusion In this chapter, we have seen that scholastic medieval thinkers developed views about essence which respond to familiar questions in contemporary philosophy and yet, they also raised questions about essence which are foreign to contemporary discussions. We saw that, like contemporary philosophers, scholastic figures thought about the difference between essential and accidental features. In contrast to the modal view of essence popular in contemporary philosophy, medieval Aristotelians distinguished essential from accidental features in terms of ontological fundamentality or independence. Accidental features were possessed in virtue of other prior features, while essential features did not depend on any prior feature. Like later philosophers, scholastic figures thought about whether there are only kind essences or also individual essences and what features might be contained in individual essences. There was not a clear consensus on these topics in the period. In addition to these more familiar questions about essence, we have seen that the shared Aristotelian commitments of scholastic philosophers, as well as their theological beliefs, raised questions about essence which are less familiar today. In light of their commitment to Aristotelian hylomorphism, medieval scholastic thinkers debated whether the essences of material substances were the same as their forms or whether such essences also included matter. We saw above that two of the most prominent medieval figures, Thomas Aquinas and John Duns Scotus, held that the essences of material substances must contain their material features as well as their formal ones. Finally, we saw that the medieval belief in a simple God who creates all other substances, together with the influence of earlier authoritative texts, raised the question of whether the essences of creatures were composed with an actualizing principle through which they are made real. Thomas Aquinas affirmed the real distinction between essence and existence, while many others challenged this view.

37

Gloria Frost

2.4

Related Topics

Modal Conceptions of Essence; Non-modal Conceptions of Essence; Essences of Individuals; Biological Species

Notes 1 For a contemporary attempt to defend such a view, see Gorman (2005) and (2014). 2 ST I q. 3, a. 4: “… quidquid est in aliquo quod est praeter essentiam eius, oportet esse causatum vel a principiis essentiae, sicut accidentia propria consequentia speciem, ut risibile consequitur hominem et causatur ex principiis essentialibus speciei; vel ab aliquo exteriori, sicut calor in aqua causatur ab igne”. Translations from Latin to English are my own. Suárez discusses this issue in depth in his Disputationes Metaphysicae 18. 3 See ST I q. 77, a. 6 ad 3. 4 Strictly speaking, a propria is any feature that arises through a substance’s essential principles and some propria are not necessarily possessed by their subjects. On the different types of propria, see Porphyry’s Isogogue ch. 4. For a discussion of Duns Scotus’s views on propria and their relationship to their subjects, see Cross (2017). 5 On scholastic debates about the intelligibility of matter see Pasnau (2011: 120–124). 6 See Veatch (1974). 7 Sandra Edwards has argued that although Aquinas denies that there were any positive qualities or characteristics, such as sizes or colors, that are essential to an individual material substance, he nevertheless thinks that a thing’s matter is essential to it. Accordingly, individual essences are constituted by an individual’s particular matter and its individuated species essence. See Edwards (1985). In support of her view, she cites passages such as ScG I, c. 65: “The essence of a singular is constituted by designated matter and the individuated form, just as the essence of Socrates is constituted from this body and this soul”. 8 See Scotus, Quaestiones Super Libros Metaphysicorum Aristotelis, question 7, article 7. For literature on his views, see King (2005). 9 On Scotus’s views on individuation see King (1992). 10 Here I am following King’s interpretation of Scotus’s views in King (2005). 11 De ver. q. 10, a. 1 ad 6: “ … accidentia designant vel notificant essentiam, ut proprii effectus notificant causam … ” 12 In VII Meta lec. 12, no. 1552: “ … signa quaedam differentiarum essentialium nobis ignotarum”. 13 Sybol. Apos., prologus: “ … cognitio nostra est adeo debilis quod nullus philosophus potuit unquam perfecte investigare naturam unius muscae … ” 14 For discussion of Marchia’s views, see Pasnau (2011: 126–128). 15 The matter referred to here is prime matter. Medieval thinkers were committed to the view of that prime matter is a real principle composing material substances. For different medieval views and debates about the nature of prime matter see Pasnau (2011: 35–52). 16 For discussion of the Aristotelian and Arabic background to Latin scholastic discussions of whether the essences of material substances contain matter, see Maurer (1951) and Galluzzo (2008). 17 On Albert’s views see Maurer (1951: 1700). 18 On Aquinas see Maurer, 195l. On Scotus, see Galluzzo (2008). 19 De ente, c. 2: “Et dico materiam signatam, quae sub determinatis dimensionibus consideratur. Haec autem materia in diffinitione hominis, in quantum est homo, non ponitur, sed poneretur in diffinitione Socratis, si Socrates diffinitionem haberet. In diffinitione autem hominis ponitur materia non signata; non enim in diffinitione hominis ponitur hoc os et haec caro, sed os et caro absolute, quae sunt materia hominis non signata”. See also In VII Meta., lec. 9, n. 1469. 20 See Galluzzo (2008). 21 Wippel (1982: 392). 22 Wippel (1982: 393). 23 On universal hylomorphism and its critique, see Wippel (1982: 408–410). 24 See De ente, ch. 4.

38

Medieval 25 De pot. q. 7, a. 2 ad 9: “[E]sse est actualitas omnium actuum, et propter hoc est perfectio omnium perfectionum”. 26 See, for example, ST I q. 3, a. 4. 27 See, for example, ST I q. 7, a. 1. 28 For a exposition and analysis of the various arguments Aquinas makes for the distinction between essence and existence, see Wippel (2000: 132–176). 29 For an example of Aquinas advancing an argument of this sort, see De ente, ch. 4. 30 In I Sent. d. 8, q. 5, a. 1 sed contra 2: “Praeterea, omnis creatura habet esse finitum. Sed esse non receptum in aliquo, non est finitum, immo absolutum. Ergo omnis creatura habet esse receptum in aliquo; et ita oportet quod habeat duo ad minus, scilicet esse, et id quod esse recipit”. 31 For more on this view in Aquinas, see Wippel (1998). 32 Aquinas does not share this assumption. For example, he thinks that prime matter and substantial form are really distinct and yet, prime matter cannot exist without a substantial form. 33 See Wippel (1982: 401–402). 34 See Wippel (1982: 401–402). 35 See Wippel (1982: 404). 36 See Wippel (1982: 405–407). For a recent article on Scotus’s views, see Cross (2013).

References

Primary Sources John Duns Scotus, (1997) Quaestiones Super Libros Metaphysicorum Aristotelis, in Andrews, R. et al. (eds.) Opera Philosophhica. St. Bonaventure, N.Y.: The Franciscan Institute. Francisco Suárez, Disputationes Metaphysicae, [online] Available at: https://homepage.ruhr-unibochum.de/michael.renemann/suarez/ [Accessed 11 Feb. 2022]. Thomas de Aquino, Opera omnia. [online] Available at: https://www.corpusthomisticum.org/iopera. html [Accessed 11 Feb. 2022]. I cite Aquinas’s works according to the following abbrevations: De ente et essentia (De ente) Expositio in Symbolum Apostulorum (Sym. Apos.) In duodecim libros Metaphysicorum expositio (In Meta.) In quatuor libros Sententiarum (In Sent.) Quaestiones disputatae de veritate (De ver.) Quaestiones disputatae de potentia (De pot.) Summa theologiae (ST) Porphyry, Isagogue, [online] Available at: http://www.logicmuseum.com/wiki/Authors/Porphyry/ isagoge/parallel [Accessed 11 Feb. 2022].

Secondary Sources Cross, Richard. (2013) ‘Duns Scotus on Essence and Existence,’ Oxford Studies in Medieval Philosophy, 1, pp. 174–202. Cross, Richard. (2017) ‘Duns Scotus on Disability: Teleology, Divine Willing, and Pure Nature’, Theological Studies, 78(1), pp. 72–95. Edwards, Sandra. (1985) ‘Aquinas on Individuals and Their Essences’ Philosophical Topics, 13:2, pp. 155–163. Galluzzo, Gabriele. (2008) ‘Scotus on the Essence and Definition of Sensible Substances,’ Franciscan Studies, 66, 213–232. Gorman, Michael. (2005) ‘The Essential and the Accidental,’ Ratio, 18, pp. 276–289. Gorman, Michael. (2014) ‘Essentiality as Foundationality,’ in Novotny, Daniel and Novak, Lukas (eds.) Neo-Aristotelian Perspectives in Metaphysics. London: Routledge, pp. 119–137. King, Peter. (1992) ‘Duns Scotus on the Common Nature and the Individual Differentia,’ Philosophical Topics, 20(2), pp. 51–76. King, Peter. (2005) ‘Duns Scotus on Singular Essences,’ Medioevo, 30, pp. 111–137.

39

Gloria Frost Maurer, Armand. (1951) ‘Form and Essence in the Philosophy of St. Thomas,’ Mediaeval Studies, 13(1), pp. 165–176. Pasnau, Robert. (2011) Metaphysical Themes: 1274–1671. Oxford: Oxford University Press. Veatch, Henry B. (1974) ‘Essentialism and the Problem of Individuation,’ Proceedings and Addresses of the American Philosophical Association, 48, pp. 64–73. Wippel, John. (1982) ‘Essence and Existence,’ in Kretzmann, N., Kenny, A., Pinborg, J. and Stump, E. (eds.) The Cambridge History of Later Medieval Philosophy: From the Rediscovery of Aristotle to the Disintegration of Scholasticism, 1100–1600. Cambridge: Cambridge University Press, pp. 385–410. Wippel, John. (1998) ‘Thomas Aquinas and the Axiom That Unreceived Act Is Unlimited,’ The Review of Metaphysics, 51(3), pp. 533–564. Wippel, John. (2000) The Metaphysical Thought of Thomas Aquinas. Washington, DC: The Catholic University of America Press.

40

3 MODERN Anat Schechtman

The early modern period in the history of philosophy—roughly, from the middle of the seventeenth century to the end of the eighteenth century—was a time of dramatic shifts in philosophical positions and traditions. At the beginning of this period, the Aristotelian paradigm that shaped most medieval philosophy was still dominant. But early modern thinkers increasingly subjected it to scrutiny, criticism, and creative reinterpretation. Naturally, discussions of essence by central figures in the period exemplify this dynamic. My aim here is to survey some of the most important developments, highlighting the ways in which early modern thinkers gradually leave the medieval Aristotelian tradition behind. A central theme is how differing conceptions of the scope of essence lead to differing conceptions of its theoretical roles.

3.1 The Traditional Notion of Essence According to the Aristotelian tradition that serves as a foil to much early modern philosophy, the nature or essence of x is what it is to be x. To use a standard example, what it is to be a human being, such as Hypatia, is to be a rational animal: being rational and being an animal are essential properties of Hypatia. Of course, Hypatia possesses other properties, such as having hair or being curious. These are not essential properties of Hypatia, but instead are “accidents,” for they are not part of what it is to be her.1 Philosophers working in this tradition generally agreed that essentiality bears an asymmetrical relation to necessity. To use a common example among Aristotelians, although it may be necessary that a human being is risible (capable of laughter), this property is not part of what it is to be a human being. So necessity does not entail essentiality. But essentiality does entail necessity: an entity cannot exist, or be the entity it is, without its essential properties.2 Consequently, such properties were taken to play several important theoretical roles, such as helping to explain persistence and unity. Returning to our example, that Hypatia is essentially a rational animal helps to explain why she is the same entity over time (she maintains those properties despite changing in various other respects) and a single individual (rather than a mere aggregate).3 Proponents of this traditional Aristotelian approach to essence held that essential properties play further theoretical roles. For instance, that Hypatia is a rational animal helps

DOI: 10.4324/9781003008750-5

41

Anat Schechtman

to explain why she is risible. By contrast, her risibility does not explain her status as a rational animal. Here we find an essential property explaining a non-essential property, but not vice versa.4 Proponents of the traditional approach similarly viewed essential properties as grounding kind membership. As I will sometimes say, essence was treated as “kinddetermining”, in the sense that an individual belongs to a general (i.e., not sui generis) kind or species—a group whose members are similar in theoretically important respects—in virtue of the complete set of its essential properties.5 For example, that an individual’s essence fully consists in the properties being rational and being an animal determines that the individual belongs to the kind human being (rather than, say, the kinds angel or bear). That the individual is risible is not essential to her, and so plays no such kind-determining role. Much of my discussion in what follows will focus on the traditional idea that essence is kind-determining. It should be clear from the above that the traditional view regards essence as selective, in the sense that only some of an entity’s properties are essential to it. The selectivity of essence is a precondition for the two theoretical roles described in the previous paragraph. An entity’s essential properties can explain its non-essential ones only if the former include some but not all of the entity’s properties. That essence is selective also enables it to be kind-determining, by making it possible for two distinct entities to share their essential properties, and so belong to the same kind, while still differing in their other properties. Having identified some of the core commitments of the traditional notion of essence, I can now state my main thesis: while some elements of the traditional notion remain a fixture of early modern thought, the view of essences as selective undergoes a significant change as the period unfolds. Whereas at the beginning of the period, the idea that essence is selective is very much still alive, various developments eventually call into doubt the utility of a notion of essence that includes this feature.

3.2 Early Modern Traditionalists: Descartes, Hobbes, Cavendish, and Conway René Descartes (1596–1650) was long considered “the father of modern philosophy” for supposedly almost single-handedly ushering in this philosophical era. This appellate is now largely—and rightly, I think—viewed as obsolete. Nevertheless, Descartes’ thought was highly influential in the early modern period, serving as both a point of departure and foil for many contemporaneous debates. Accordingly, his notion of essence is a good place for us to start. Many aspects of the traditional notion are on display in Principles of Philosophy (1644), where Descartes systematically presents the central tenets of his metaphysical system. He writes: [E]ach substance has one principal property which constitutes its nature and essence, and to which all its other properties are referred. Thus extension in length, breadth and depth constitutes the nature of corporeal substance; and thought constitutes the nature of thinking substance. Everything else which can be attributed to body presupposes extension, and is merely a mode of an extended thing; and similarly, whatever we find in the mind is simply one of the various modes of thinking. (CSM 1.210) This passage tells us that essential properties determine the kind to which a substance belongs—body or mind, respectively.6 And moreover, they help to explain a substance’s nonessential properties (i.e., they are that “to which all its other properties are referred”). The passage also expresses the selectivity of essence: while each substance has many properties, it has one essential property, extension or thought. Descartes goes on to write:

42

Modern

[S]hape is unintelligible except in an extended thing; and motion is unintelligible except as motion in an extended space; while imagination, sensation and will are intelligible only in a thinking thing. (ibid) Shape, motion, imagination, sensation, and will are accidental properties, which Descartes calls “modes”. This passage asserts that none of them is intelligible (i.e., liable to explanation) without reference to certain other properties, extension and thought, which are essential. Here, again, we see the selectivity of essence at work. Selectivity and its accompanying theoretical roles can also be discerned in Descartes’ bestknown work, the Meditations on First Philosophy (1641). In the Second Meditation, after realizing that “this proposition, I am, I exist, is necessarily true whenever it is put forward by me or conceived in my mind,” the meditator admits that he does “not yet have a sufficient understanding of what this ‘I’ is”. Subsequent reflection enables his mind to “perceive its own nature as distinctly as possible”: But what then am I? A thing that thinks. What is that? A thing that doubts, understands, affirms, denies, is willing, is unwilling, and also imagines and has sensory perceptions. (CSM 2.19) Turning from mind to body later in the Second Meditation, the meditator summarizes the results of his effort to identify the nature of a piece of wax as follows: [T]he wax was not after all the sweetness of the honey, or the fragrance of the flowers, or the whiteness, or the shape, or the sound, but was rather a body which presented itself to me in these various forms a little while ago, but which now exhibits different ones. But what exactly is it that I am now imagining? Let us concentrate, take away everything which does not belong to the wax, and see what is left: merely something extended, flexible and changeable. (CSM 2.20) Here the nature of the wax—what it is to be that body—is presented as selective. Properties such as taste, fragrance, and color are non-essential. That the wax is essentially extended partly explains these non-essential properties—the “various forms” in which the body presents itself to the senses—and determines the kind to which the wax belongs (viz., body). Commitment to the traditional view of essence is also evident in Descartes’ contemporaries, such as the materialist Thomas Hobbes (1588–1679). Although he rejects Descartes’ dualism about mind and body, his understanding of what a body is—its essence—is remarkably similar to Descartes’: [A] body is that, which having no dependence upon our thought, is coincident or coextended with some part of space.7 (De Corpore, 8.2) Margaret Cavendish (1623–1673) advances a version of materialism whose view of body highlights not space but “Place”: MATTER is that we name Body; which Matter cannot be less, or more, than Body … Also, Matter cannot be figureless, neither can Matter be without Parts. Likewise, there cannot be Matter without Place, nor Place without Matter; so that Matter, Figure, or

43

Anat Schechtman

Place, is but one thing: for, it is as impossible for One Body to have Two Places, as for One Place to have Two Bodies; neither can there be Place, without Body. (1668, I.1) Neither Hobbes nor Cavendish is keen on using the terms “nature” and “essence”, often speaking instead of the meaning of the word “body”.8 Nevertheless, insofar as both intend to usurp Cartesian dualism, the above passages seem very much designed to replace Descartes’ view that there are two distinct kinds of substance (mind and body), determined by two distinct essences (thought and extension), with a monistic view that recognizes just one kind (body), determined by just one essence (coextension with space, or Place). Moreover, the selective notion of essence undergirding their materialism is in my view the very same one that Descartes employs when articulating his dualism. It is also the one that Anne Conway (1631–79) relies on when developing her “vitalist,” three-species alternative to both dualism and materialism. She writes: [W]e must now determine how many species of things there are which are distinguished from each other in terms of their substance or essence. If we look closely into this, we will discover there are only three, which, as was said above, are God, Christ, and creatures … Since all phenomena in the entire universe can be reduced to these three aforementioned species as if into their original and peculiar causes, nothing compels us to recognize a further species. (1996, 30) The essential properties of God, Christ, and creatures, which determine the species to which they belong, are immutability, partial mutability, and complete and utter mutability, respectively: [B]ecause the three aforementioned species exhaust all the specific differences in substances which can possibly be conceived by our minds, then that vast infinity of possible things is fulfilled in these three species … . Certainly insofar as something can be called an entity, it is either altogether immutable like God, the supreme being, or altogether mutable, that is for good or bad, like a creature, which is the lowest order of being, or partly mutable in respect to good, like Christ, the son of God, the mediator between God and creatures. (ibid) Conway also suggests that essential properties explain non-essential properties, the latter being among the “phenomena” of which the former are the “original and peculiar causes”.9 To be clear, my claim is not that when it comes to the character of essence, Descartes, Hobbes, Cavendish, and Conway agree on every point. They do not.10 Nor is it that they concur with the traditional notion in every single respect.11 Rather, they converge on the view of essence, current in the Aristotelian tradition, as selective, kind-determining, and (according to Conway) explanatory.

3.3 First Countermovement: Spinoza and Leibniz As we move forward in the seventeenth century, an alternative to the traditional notion of essence begins to emerge. Most evident in the works of Benedict de Spinoza (1632–77) and G.W. Leibniz (1646–1716), this alternative assigns a great many essential properties to any particular entity. Whereas Descartes and other proponents of the traditional view deem a small number of an entity’s properties as essential, Spinoza and Leibniz opt for a highly

44

Modern

inclusive approach that in its extreme form maintains that all of an entity’s properties are essential to it. As we will see, this “superessentialist” view makes it difficult for essence to play the two theoretical roles highlighted above. Although superessentialism is often associated with Leibniz, it is his predecessor and arguably major influence, Spinoza, who laid its foundations, although he does not go as far as Leibniz in embracing it.12 Indeed, as I will now explain, the inclusivity of essence is a natural consequence of the conceptual dependence relations that power Spinoza’s metaphysical system. In his magnum opus, the Ethics (1677), Spinoza often claims that one thing can be correctly conceived only through another thing. For example, early in the Ethics, he maintains that an effect can only be cognized, or conceived, through its cause: The cognition of an effect depends on, and involves, cognition of its cause.13 (E1a4) This thesis is often called the “causal axiom”. Later in the Ethics, Spinoza provides the following definition of essence: I say that to the essence of any thing belongs that which, being given, the thing is necessarily posited and which, being taken away, the thing is necessarily taken away; or that without which the thing can neither be nor be conceived, and which can neither be nor be conceived without the thing. (E2def2) Combining what follows “or” in this definition of essence with the causal axiom, some scholars have interpreted Spinoza as holding that all of an entity’s causal properties (i.e., those properties that identify its cause) are essential to it.14 Further evidence for this interpretation is gleaned from Letter 60, where Spinoza claims that “the idea or definition of a thing should express its efficient cause,” and from E5a2, where he claims that “[t]he power of an effect is defined by the power of a cause, insofar as its essence is explained or defined by the essence of its cause”.15 Spinoza often equates an entity’s essence with its conatus, power, or its striving to remain in existence: The striving [conatus] by which each thing strives to persevere in its being is nothing but the actual essence of the thing. (E3p7) [T]he power of each thing, or the striving by which it (either alone or with others) does anything, or strives to do anything that is (by p6), the power, or striving, by which it strives to persevere in its being, is nothing but the given, or actual, essence of the thing itself. (E3p7d) According to Spinoza, the power of an entity to remain in existence, and other powers as well by which it “does anything”, are explained by its essence. As this indicates, Spinoza’s commitment to a fairly inclusive view of essence is compatible with the idea that essential properties explain non-essential properties.16 At the same time, such a view implies rejection of the traditional idea that essence is kinddetermining. Spinoza’s determinism entails that each thing has a unique causal history. Given this, if an entity’s essence includes its cause, then any two entities with distinct causal histories—say, Hypatia and Socrates—do not share an essence. But if they do not share an

45

Anat Schechtman

essence, then they do not share an essence that determines their membership in the same kind. Socrates’ essential properties do not place him in the same group as Hypatia, but render him sui generis. Likewise for Hypatia: since her essence is unique, it is too particular to determine membership in a general kind.17 To summarize, Spinoza adopts an inclusive view of essence that diverges from the traditional view with which the early modern period began, insofar as it rejects the role of essence as kind-determining. As we will now see, Leibniz follows the same trajectory as Spinoza, though to an even greater extent, arguing that all of an entity’s properties are essential to it. Like Spinoza, Leibniz moves from considerations about what it takes to conceive an entity to what is included in its essence. Regarding the former, Leibniz famously holds that every property of a given substance belongs to what he calls its “complete notion” or concept: The complete or perfect notion of an individual substance contains all of its predicates, past, present, and future. For certainly it is now true that a future predicate will be, and so it is contained in the notion of a thing. And thus everything that will happen to Peter or Judas, both necessary and free, is contained in the perfect individual notion of Peter and Judas … (“Primary Truths”, GA 32) We have said that the notion of an individual substance includes once and for all everything that can ever happen to it and that, by considering this notion, one can see there everything that can truly be said of it, just as we can see in the nature of a circle all the properties that can be deduced from it.18 (Discourse on Metaphysics §13) As indicated by the latter passage’s reference to “the nature of” the target entity, Leibniz often treats the complete notion of a thing as equivalent to its essence. Elsewhere he makes the equivalence fully explicit, speaking of “the essence or individual notion of a substance” (Discourse on Metaphysics §16). Consider also Leibniz’s remark that [A] thing can remain the same, though changed, if from its very nature it follows that the same thing must pass successively through different states. I myself am said, indeed, to be the same as I was before, because my substance involves all my past, present and future states. (Notationes Generales, A 6.4.556) If “my substance involves all my … states” or properties, then my complete notion does so as well. Since my complete notion is equivalent to my essence, it follows that my essence involves all my properties. Since the reasoning is perfectly general, it follows that the essence of every entity includes all of that entity’s properties.19 Leibniz embraces the implications of this maximally inclusive approach to essence, insisting that even the most arbitrary-seeming properties of a thing are essential to it, and that since they belong to the entities’ complete notion, they can be known a priori (i.e., independently of sense experience): God, seeing Alexander’s individual notion or haecceity, sees in it at the same time the basis and reason for all the predicates which can be said truly of him, for example, that he vanquished Darius and Porus; he even knows a priori (and not by experience) whether he died a natural death or whether he was poisoned, something we can know only through history. (Discourse on Metaphsysics §8)

46

Modern

It, therefore, also follows that he would not have been our Adam, but another Adam, had other events happened to him. (AG 72–73; cp. Discourse on Metaphsysics §30) Leibniz’s move to maximal inclusivity has significant repercussions for the theoretical roles of essence. First, on a fully inclusive view of essence, essential properties do not explain nonessential properties, for the simple fact that there are no properties of the latter sort to be explained.20 Second, like Spinoza, Leibniz is committed to denying that essence is kinddetermining. According to Leibniz, no two distinct individuals share an essence; therefore, no individual essence determines membership in a general kind.21 Not only being rational, but also vanquishing Darius and Porus, in the case of Alexander the Great, or eating from the forbidden fruit, in the case of Adam, belongs to their respective essences—as well as all their other properties. Once more, each substance is sui generis. What explains Spinoza’s and Leibniz’s commitment to an inclusive view of essence? A full answer, which will likely point to multiple factors, goes beyond the scope of this short chapter.22 But we have already seen one consideration that is important to both philosophers, namely, the connection between essence and conception. Both Spinoza and Leibniz hold that to fully conceive of something is to grasp what it is to be that thing—hence, its essential properties.23 Consequently, an entity’s essence encompasses whatever is required to fully conceive of it. Each philosopher endorses an expansive view of what is required to fully conceive of an entity: one must conceive its causal properties in Spinoza’s case, or all of its properties, in Leibniz’s case. Either way, the result is an inclusive view of essence.

3.4 Second Countermovement: Locke The final position I would like to examine consists in a different kind of rejection of the traditional view of essence than the one implied by superessentialism. In the Essay Concerning Human Understanding, John Locke (1632–1704) famously distinguishes two types of essence, the first of which he calls “real”: First, Essence may be taken for the very being of anything, whereby it is what it is. And thus the real internal, but generally (in substances) unknown constitution of things, whereon their discoverable qualities depend, may be called their essence. (Essay, III.iii.15) In contrast, what Locke calls “nominal” essences are “nothing else but, those abstract complex ideas, to which we have annexed distinct general names” (Essay, III.iii.17), which, in our ignorance of the real essences of things, we use to classify the latter as falling into kinds: But, it being evident that things are ranked under names into sorts or species, only as they agree to certain abstract ideas, to which we have annexed those names, the [nominal] essence of each genus, or sort, comes to be nothing but that abstract idea which the general, or sortal (if I may have leave so to call it from sort, as I do general from genus), name stands for. (Essay, III.iii.15) As a slogan, we may say that real essences are in the world, whereas nominal essences are in the mind. We may use one of Locke’s recurring examples, a piece of gold, to illustrate this distinction.24 The real essence of the piece consists in the insensible qualities that give rise to all its sensible

47

Anat Schechtman

qualities: the latter include being yellow, heavy, fusible, malleable, soluble in aqua regia, tasteless, odorless, and so on. Its nominal essence consists in a complex idea of the mind, made from ideas of a certain subset of these sensible qualities. Anything that manifests the latter qualities will be classified by us as belonging to the kind gold.25 In Locke’s words, supposing the nominal essence of gold to be a body of such a peculiar colour and weight, with malleability and fusibility, the real essence is that constitution of the parts of matter on which these qualities and their union depend … . (Essay III.vi.6) Now, it is tempting to hold that under the best circumstances, our classificatory scheme captures genuine divisions in nature, kinds to which entities belong independently of our choice of that scheme. Nominal essences would then correspond to real essences, at least in the ideal case. Yet Locke argues that this position is unfounded. There is no reason to think that the qualities flagged by any classificatory scheme correspond to a genuine division in nature, even in the best circumstances. Instead, what he claims to be the “more rational” position is that any classificatory scheme is grounded on similarities that perceivers such as ourselves observe among the sensible qualities of things. (Essay III.iii.17) These observations lead us to divide those things into kinds. Rather than approximating the real essences of things, such divisions simply reflect how we happen to perceive those things. It is often assumed that Locke’s position is a version of skepticism rooted in his empiricism, and specifically, in his position that all ideas originate in sense perception. On this reading, Locke endorses the traditional notion of essence: the real essences of things are selective, kinddetermining, and help to explain non-essential properties. Yet Locke’s empiricism compels him to hold that because these essences are insensible, they are unknowable to us. So we could have no reason, founded solely on the natures of things, to group various entities together in one way rather than another. Consequently, our efforts to sort things into kinds according to their essences are bound to be arbitrary. There is however an alternative interpretation of Locke’s position that takes him instead to reject the traditional notion of essence. According to this interpretation,26 an entity such as a piece of gold or a human being possesses a real essence, which is insensible (hence, given Locke’s empiricism, unknowable) and explains the entity’s sensible qualities. At the same time, these essences are not selective, and hence do not determine the kinds to which those entities belong. This is because essences are unique: each entity has its own essence, consisting in the insensible qualities possessed by it and it alone that give rise to all its sensible qualities.27 It follows that the work of sorting things into kinds must be shouldered by something else. This interpretation emphasizes that Locke is cognizant of the traditional notion of essence: [T]hose who, using the word essence for they know not what, suppose a certain number of those essences, according to which all natural things are made, and wherein they do exactly every one of them partake, and so become of this or that species. (Essay III.iii.17) Locke here acknowledges a view according to which essences explain non-essential, sensible properties, while also determining the kind or species to which the individual belongs (“and so become this or that species”). The problem, Locke seems to suggest, is that these two roles are in tension, and so one of them must go, and with it, the traditional notion of essence. On the one hand, Locke reasons, if essence is to be kind-determining, then it must be selective. For it must include only those properties by virtue of which the entity is classified as

48

Modern

belonging to a particular kind. If Hypatia belongs to the kind human, for example, then her essence must contain only those properties, such as being rational and being an animal, possession of which determines her membership in this kind. On the other hand, if essence is to explain all of an entity’s sensible properties, then it must be inclusive. For it must include sufficiently many non-sensible properties of sufficiently many different types to explain all of the entity’s sensible properties—including such things as the specific texture of Hypatia’s hair, the particular skin tone of a particular region of her nose, and the distinctive sound of her laughter. This means that Hypatia’s essence must include more than being rational and being an animal. In short, Locke suggests that essences are kind-determining only if they are selective, but they explain an entity’s sensible properties only if they are inclusive. But essences cannot be both selective and inclusive. So we cannot hold that they are both kind-determining and explanatory of sensible properties. We must choose. Locke himself opts for the latter, replacing the former with a view on which kinds are human constructs. What Locke tends to emphasize, however, is not the above tension per se, but rather what it implies for the utility of the traditional notion of essence. He writes: For I would ask anyone, what is sufficient to make an essential difference in nature, between any two particular beings, without any regard had to some abstract idea, which is looked upon as the essence and standard of a species? All such patterns and standards, being quite laid aside, particular beings, considered barely in themselves, will be found to have all their qualities equally essential, and everything, in each individual, will be essential to it, or, which is more true, nothing at all. (Essay III.vi.5; see also III.vi.4 and 39) We might summarize Locke’s reasoning in this passage as follows. Nominal essences reflect perception and convention-dependent classificatory schemes. When these are “laid aside,” we are left with real essences comprised of all of an entity’s non-sensible properties, including sufficiently many properties of sufficiently many different types to explain all of the entity’s sensible properties. They are all “equally essential” to it. But that is tantamount to saying that “nothing at all” is essential to it. Nothing useful is gained by labeling each and every one of the entity’s non-sensible properties as essential. What explains Locke’s commitment to the inclusive view of essence that leads him to this conclusion? Above, I suggested that an explanation of Spinoza’s and Leibniz’s commitment to this view will reference their endorsement of a link between conception and essence. Since Locke’s empiricism prohibits him from accepting any such link, we must look elsewhere to explain why he adopts the inclusive view. One plausible hypothesis is that it follows from his commitment to corpuscularianism, according to which the sensible qualities of bodies arise from properties of minute insensible parts or “corpuscles”. Such a view does not privilege a subset of corpuscularian properties. Without some further reason to think that there is such a privileged subset, it is a quick step to the conclusion that either all of them are essential to an entity, or none are. To conclude, we have seen that at the beginning of the early modern period, the traditional Aristotelian notion of essence as explanatory, kind-determining, and selective is very much still at play in the works of figures such as Descartes, Hobbes, Cavendish, and Conway. Later in the period we see movements away from this traditional view, towards superessentialism in the case of Spinoza and Leibniz, and towards a view of essence as theoretically useless in the case of Locke. What both movements have in common is their embrace of an inclusive view of essence that undermines its kind-determining role.28

49

Anat Schechtman

3.5

Related Topics

Ancient Medieval Modal Conceptions of Essence Non-Modal Conceptions of Essence Identity, Persistence, and Individuation Unity Conventionalism

Notes 1 Traditionally, an essence is, or is stated by, a real definition. For discussion of this connection, and of the Aristotelian tradition more broadly, see Chapters 1 and 2 in this volume. Throughout I use the terms “nature” and “essence” interchangeably. 2 For further discussion of the connection between essence and modality, see Chapters 7 and 8 in this volume. 3 On the theoretical roles of essence in contemporary philosophy, see Chapters 19 and 27 in this volume. 4 Essential properties were also taken to partly explain accidental properties; they did so together with other factors (e.g., environmental conditions). For example, Ockham writes: “it is clear to the senses that hot water, if left to its own nature, reverts to coldness” (Quodlibet III.6, cited in Pasnau 2011, §24.4). The cooler temperature, which is an accidental property of water, is explained by the nature of water, together with a feature of the environment, namely, the removal of a heat source. 5 This characterization applies to general kinds that do not admit of further division into subkinds, or to what Aristotle calls “lowest species” (see Chapter 1 in this volume). 6 Scholars tend to view Descartes as recognizing two and only two kinds, in contrast with the many kinds recognized by scholastic philosophers. Descartes however sometimes seems to allow for a greater plurality of kinds; see, e.g., the First Replies (CSM 2.84) and the Second Replies (CSM 2.106–7). For discussion, see Kaufman (2014) and Brown and Normore (2019, ch. 2). 7 See also Leviathan 34.2: a body “fills or occupies some certain room or imagined place, and depends not on the imagination, but is a real part of what we call the universe”. 8 Despite this and other similarities, Hobbes’s and Cavendish’s versions of materialism are significantly different. For a recent discussion, see Duncan (2022, chs. 2 and 4). 9 For an interesting discussion of Conway’s view of essences, see Grey (2020). 10 For example, they disagree about the ontological status of essences: like scholastic thinkers, Descartes, Cavendish, and Conway appear to be realists about essence, whereas Hobbes (1839–1845, 7: 221) adopts a deflationary approach, according to which essences are “words artificial belonging to the art of logic, and signify only the manner how we consider the substance itself”. See Pasnau (2011, §27.5) for a helpful discussion. 11 As was said above, in note 6, Descartes arguably accepts a far narrower range of essences than most Aristotlelians, and the same is true for the other early modern traditionalists we have discussed. 12 For Leibniz’s relation to Spinoza, see, e.g., Newlands (2018) and Laerke (2008). 13 For discussion of the causal axiom, see Lin (2020). 14 See, e.g., Della Rocca (2008, 93ff) and Newlands (2018, 115-16). 15 For a helpful overview and subsequent rejection of this line of argument, see Bender (forthcoming). 16 For further discussion of the explanatory role of essences in Spinoza, see Ward (2011), Viljanen (2008), Newlands (2018, ch. 5), and Bender (2023). 17 The thesis that Spinozistic essences are unique, and therefore not kind-determining, is defended by, e.g., Martin (2008), Ward (2011), and Hübner (2015). While Spinoza sometimes does speak of shared, kinddetermining essences, scholars have argued that on such occasions Spinoza is referring not to real essences, which are mind-independent, but rather, as Hübner (2015, 16ff) puts it, to “mind-dependent species-essences, constructed by finite minds but grounded in actual, recognized similarities”. 18 For further discussion of Leibniz’s doctrine of complete notions, including textual sources, see di Bella (2004, ch. 1 and 2018).

50

Modern 19 See, e.g., Mates (1972), Mondadori (1973), and Look (2013) for this very common line of interpretation. For challenges to it, see, e.g., Cover and Hawthorne (1999), Adams (1994, pp. 12ff), and Kim (manuscript). I am grateful to Jun Young Kim for a helpful discussion of this literature. 20 At the same time, as we have seen, Leibniz says of Alexander the Great that his individual notion—and hence his essence—is “the basis and reason” for all his properties. Perhaps the thought here is that among the essential properties, some are more fundamental, and thus explain other, less fundamental essential properties. 21 It is open to Leibniz to hold, in departure from the tradition, that only a subset of an entity’s essential properties determines the kind to which it belongs—e.g., in the case of Alexander the Great and Adam, their rationality and animality. For an interpretation of Leibniz as indeed holding this view, see Look (2009). 22 See Ward (2011) and Di Bella (2018) for discussion of some of the motivating factors behind, respectively, Spinoza’s and Leibniz’s inclusive view of essence. 23 This is arguably a version of the traditional link between essence and definition mentioned in note 1. 24 See, e.g., Essay III.iii.18. 25 Here I have in mind what Owen (1991, 107) calls “the real essence of an unsorted particular,” by contrast with “the real essence of a sorted particular”. While the former is a set of insensible qualities that give rise to all the entity’s sensible qualities, the latter is the subset of insensible qualities that give rise to the subset of sensible qualities comprising the entity’s nominal essence. 26 This interpretation is drawn from the discussion in Pasnau (2011, §27.7). Versions of it are also proposed in Owen (1991), Phemister (1990), and Look (2009). 27 The real essence of an individual is distinct from its substratum: the former consists of a collection of insensible qualities; the latter is not a collection of qualities but something that has or “supports” qualities. See Essay III.ii.23 for Locke’s discussion of substratum. 28 I am grateful to Sebastian Bender, Kathrin Koslicki, Michael Raven, Dominik Perler, and participants in the Theoretical Philosophy Colloquium at the Humboldt-Universität zu Berlin for their helpful comments. I am also grateful to John Bengson for extensive help with this article.

References Adams, R.M. (1994) Leibniz: Determinist, Theist, Idealist. Oxford: Oxford University Press. Bender, S. (2023). “Spinoza on the Essences of Singular Things,” Ergo 9.10: 1–24. Brown, D. and C. Normore (2019) Descartes and the Ontology of Everyday Life. Oxford: Oxford University Press. Cavendish, M. (1668) The Grounds of Natural Philosophy. London: A. Maxwell. [Cited by part and chapter]. Conway, A. (1996) The Principles of the Most Ancient and Modern Philosophy, A.P. Coudert and T. Corse (trans.). Cambridge: Cambridge University Press. Cover, J.A. and J. O′Leary-Hawthorne (1999) Substance and Individuation in Leibniz, Cambridge: Cambridge University Press. Della Rocca, M. (2008) Spinoza. New York: Routledge. Descartes, R. (1985–1992) The Philosophical Writings of Descartes (2 vols.), J. Cottingham, R. Stoothoff, and D. Murdoch (trans.). Cambridge: Cambridge University Press. [CSM; cited by volume and page]. Di Bella, S. (2004) The Science of the Individual: Leibniz’s Ontology of Individual Substance. Dordrecht: Springer. Di Bella, S. (2018) “The Complete Concept of an Individual Substance,” in M.R. Antognazza (ed.) The Oxford Handbook of Leibniz. Oxford: Oxford University Press. Duncan, S. (2022) Materialism from Hobbes to Locke. Oxford: Oxford University Press. Grey, J. (2020) “Species and the Good in Anne Conway’s Metaphysics,” in C. Marshall (ed.) Comparative Metaethics: Neglected Perspectives on the Foundations of Morality. New York: Routledge. Hobbes, T. (1839–45) The English Works of Thomas Hobbes of Malmesbury (11 vols.), W. Molesworth (ed.). London: J. Bohn. [cited by volume and page]. Hobbes, T. (1994) Leviathan, E. Curley (ed.). Indianapolis: Hackett. [cited by chapter and article]. Hübner, K. (2015) “Spinoza on Essences, Universals, and Beings of Reason,” Pacific Philosophical Quarterly 96.2: 58–88.

51

Anat Schechtman Kaufman, D. (2014) “Cartesian Substances, Individual Bodies, and Corruptibility,” Res Philosophica 91.1: 71–102. Kim, J. (manuscript) “Leibniz’s Tripartite Distinction of Properties”. Laerke, M. (2008) Leibniz lecteur de Spinoza. La genèse d’une opposition complexe. Paris: Honor Champion. Leibniz. G.W. (1923–1980) Staatliche Schriften und Briefe, series I-VIII. Berlin: Akademie-Verlag. [A; cited by series, volume, and page]. Leibniz. G.W. (1989) Philosophical Essays, R. Ariew and D. Garber (trans.), Indianapolis: Hackett. [AG; citations from the Discourse on Metaphysics are by DM and section number]. Lin, M. (2020). “The Many Faces of Spinoza’s Causal Axiom,” in D. Perler and S. Bender (eds.) Causation and Cognition in Early Modern Philosophy. New York: Routledge. Locke, J. (1975) An Essay Concerning Human Understanding, P. Nidditisch (ed.), Oxford: Oxford University Press. [Essay; cited by book, chapter, and section]. Look, B. (2009) “Leibniz and Locke on Natural Kinds,” in V. Alexandrescu (ed.) Branching Off: The Early Moderns in Quest for the Unity of Knowledge. Bucharest: Zeta Books. Look, B. (2013) “Leibniz’s Modal Metaphysics,” in E.N. Zalta (ed.) Stanford Encyclopedia of Philosophy, https://plato.stanford.edu/archives/spr2013/entries/leibniz-modal/ Martin, C. (2008). “The Framework of Essences in Spinoza’s Ethics,” British Journal for the History of Philosophy 16.3: 489–509. Mates, B. (1972) “Individuals and Modality in the Philosophy of Leibniz,” Studia Leibnitiana 4: 21–57. Mondadori, F. (1973) “Reference, Essentialism, and Modality in Leibniz’s Metaphysics,” Studia Leibnitiana 5: 74–101. Newlands, S. (2018). Reconceiving Spinoza. Oxford: Oxford University Press. Owen, D. (1991) “Locke on Real Essence,” History of Philosophy Quarterly 8.2: 105–118. Pasnau, R. (2011) Metaphysical Themes 1274–1671. Oxford: Oxford University Press. Phemister, P. (1990) “Real Essence in Particular,” Locke Studies 25: 27–55. de Spinoza, B. (1985) The Collected Works of Spinoza (2 vols), E. Curley (ed. and trans.). Princeton: Princeton University Press. [The Ethics [E] is cited by part, followed by the following abbreviations: a = axiom, d = demonstration, def = definition, p = proposition.] Viljanen, V. (2008) “Spinoza’s Essentialist Model of Causation,” Inquiry 51.4: 412–437. Ward, T. (2011) “Spinoza on the Essences of Modes,” British Journal for the History of Philosophy 19.1: 19–46.

52

4 PRAGMATISM Andrew Howat

Pragmatism is a rich philosophical tradition stretching from 1870s Cambridge, Massachusetts to today’s global scholarship. Scholars typically divide the tradition into three phases—Classical (Peirce, James, Dewey, Addams, Cooper, etc.), mid-century (e.g. C.I. Lewis, Quine, Ramsey, later Wittgenstein), and neo-pragmatist (e.g. Rorty, Putnam, Brandom, and Price). This chapter focuses solely on the Classical phase. It is a popular assumption that pragmatists are anti-essentialists. For those primarily familiar with neo-pragmatism, this assumption is understandable. Rorty claims pragmatism “is simply anti-essentialism applied to notions like ‘truth’, ‘knowledge’, ‘language’, ‘morality’, and similar objects of philosophical theorizing …” (1982: 162). Seigfried describes pragmatism as an “anti-essentialist, practice-centered perspective” (2004:13). In defining a “pragmatic view of concepts” held by Carnap, Dutilh Novaes writes that “he rejects the idea that there are objective, atemporal ‘essences’ that concepts ought to capture” (forthcoming: 13). Despite its popularity, the assumption that pragmatism is anti-essentialism is clearly unwarranted, since even the relatively precise term “Classical pragmatism” is not the name of any one thesis or set of theses. Such a strong claim would require presenting the whole range of characteristically pragmatist theses, clearly defining anti-essentialism, then demonstrating the appropriate entailments. Rorty, along with seemingly everyone else who shares his view, does none of these things.1 Instead, he implicitly relies upon the popular view that Peirce’s original argument for the “pragmatic maxim” from 1878 implies that definition can never be a source of philosophical insight (merely assuming, without argument, that this includes both nominal and real definitions). According to a standard narrative about pragmatism’s history, this claim then influenced James and Dewey, leading them to make various seemingly antiessentialist remarks, which indirectly influenced figures such as Quine, whose commitment to anti-essentialism is clear cut (Chapter 29). While this standard narrative gets many things basically right, it still does not entail that either that pragmatism is anti-essentialism, nor even that all the Classical pragmatists endorsed it, or so I shall argue. The upshot is that the relationship between Classical pragmatism and essentialism is complicated and contested in interesting and understudied ways.2 After some terminological preliminaries (§1), this chapter defends the following claims. First, there are at least two different interpretations available of C.S. Peirce’s views on essence and essentialism (§2). One of them suggests that Peirce may have endorsed his own novel,

DOI: 10.4324/9781003008750-6

53

Andrew Howat

pragmatist understandings of essence/essentialism. William James’s few remarks on the topic are somewhat ambivalent (§3), evincing a superficial anti-essentialism that seemingly anticipates Quine’s views, while remaining consistent with an anti-realist form of essentialism. Although John Dewey’s pragmatism (§4) is the most vividly anti-essentialist in spirit, once again there are prominent scholars who seemingly reject that interpretation, partly because Dewey’s attitude to metaphysics in general is difficult to establish. Some of his anti-essentialist remarks seem grounded in his own pragmatist, empirically naturalist metaphysics, while others suggest an outright metaphysical quietism that seemingly rules out “essence” and “essentialism” as meaningful terms.

4.1 Pragmatism and Traditional Philosophical Terminology Pragmatism is at least partly a metaphilosophy. As we shall see, its founding documents focus on methodological questions about philosophical inquiry and its traditional terminology (Talisse 2017; Howat 2023). For example, Peirce’s pragmatism emerges from his critique of Descartes’ conception of inquiry, while James took pragmatism to show that once any appropriate philosophical terminology is clarified in purely practical terms, many apparent disputes (especially metaphysical ones) dissolve and/or turn out to be merely verbal (James 1898, 1907). Many critics either ignore or misunderstand pragmatism’s metaphilosophical dimension. It has a reputation as an approach to philosophy that rejects metaphysics tout court as a meaningless endeavor (Atkin 2016: 222). Yet Peirce and Dewey made many metaphysical claims, often labelling them explicitly as such (Dewey 1929/1925: 47). Peirce even makes arguments for the indispensability of metaphysics (1.129, 1905; Giladi 2017; Legg and Giladi 2018). The pragmatists did, however, challenge the way we understand the nature, methods, and purpose of metaphysics. They strongly repudiate the idea that metaphysics is ‘first philosophy’, for example. Peirce regarded metaphysics as instead dependent upon what he called the ‘normative sciences’ (logic, ethics, and aesthetics, see e.g. EP 2.146, 1903). Dewey warned against the misguided ‘quest for certainty’ he thinks typically animates metaphysical theorizing. He may have rejected as incoherent the idea of a metaphysics of existence, instead arguing in favor of a metaphysics of experience (compare Bernstein 1961; Sleeper 2001; and Shook 2000). Both argued for a vision of metaphysics that is fallibilist, naturalistic, and a posteriori. Both also argued in their own ways against metaphysical realism and for what Robert Lane (2018: 64) calls a “basic idealism”. On Peirce’s view, the concept of eine Welt an sich—a world considered apart from how it could ever be experienced or cognized is empty. Peirce’s basic idealism implies that the world as we must cognize it—as our eventual investigation-fixed beliefs would eventually represent it to be—is reality, and it makes no sense to speak of something that resides behind or apart from our experiences, something that causes those experiences and subsequent thoughts but can never be represented by them Both are also arguably process philosophers, whose metaphysics focuses on problems contemporary metaphysicians tend to ignore (such as philosophical explanations for evolutionary processes and of emergence/self-organization, e.g. Seibt 2022; Reynolds 2002). Thus, in addition to the challenge of anachronism—an attempt to classify late 19th and early 20th century ideas using early 21st century terminology—there are also deep methodological difficulties involved in any attempt to state a clear relationship between pragmatism and essentialism, especially when the latter is often understood as a creature of contemporary non-naturalistic, realist metaphysical frameworks and a priori methods. Nevertheless, I shall

54

Pragmatism

rely on the following definitions in hopes of drawing at least some informative connections between the two: Essence: “A property P is an essential property of being an F iff anything is an F partly in virtue of having P. A property P is the essence of being an F iff anything is an F in virtue of having P. The essence of being F is the sum of its essential properties” (Devitt 2008: 345).3 Essentialism: the doctrine that (at least some) kinds have (at least some) essential properties.4 Realist Essentialism: if entities have essences (or essential properties), then they possess them “independently of how these entities are described, conceptualized or otherwise placed with respect to our specifically human interests, purposes or activities” (Koslicki, forthcoming). Anti-realist Essentialism: essences (or essential properties) are, or at least can be in some sense mind-dependent, socially constructed, or ‘conferred’ (see Chapters 31 and 32, for precedents see e.g. Ásta 2008 and Khalidi 2015). Anti-essentialism (active): the denial of essentialism based on some competing metaphysical claim about kinds. Anti-essentialism (passive): refusal to endorse essentialism based, not on any competing metaphysics, but as part of an effort to maintain a ‘strategic silence’ on any/all metaphysical questions given their unintelligibility (see Price 2011: 258).

4.2 Peirce C.S. Peirce invented pragmatism in the 1870s, in Cambridge, Massachusetts (though he did not use the term “pragmatism” until much later). The reason Rorty and others believe that pragmatism is anti-essentialism is the way they interpret Peirce’s “pragmatic maxim”, which I briefly explain below. They assign it skeptical or quietist implications about essence/ essentialism, and/or about metaphysics writ large, much like the logical positivists’ verification principle (Ayer 1968: 45). I will suggest that (a) this interpretation is probably mistaken, and (b) even if it weren’t, it does not entail Peirce was an anti-essentialist. I do so by briefly showing how the very same evidence is used by other scholars to argue that Peirce proposed a novel pragmatist understanding of essence/essentialism. The resulting positive view is underspecified but could, I suspect, be developed into a form of anti-realist essentialism. This generates something of a stalemate. I briefly show that Peirce’s few explicit remarks about essence do not resolve this stalemate, although they do suggest that Peirce was, at a minimum, open to the idea that both artificial and natural kinds have essences. This, I take it, is sufficient to show that Rorty’s blunt pronouncements about pragmatism and antiessentialism are unwarranted. However, Rorty famously dismisses Peirce’s role in founding pragmatism (1980: 720), so I also consider James and Dewey in the remaining sections. Most scholars trace pragmatism’s founding to Peirce’s 1878 article How to Make Our Ideas Clear (hereafter Ideas). There he proposes a methodological principle—the “pragmatic maxim”—for the clarification of abstract concepts such as reality. Pragmatism was initially the idea that only by applying this maxim could one achieve the highest possible level of clarity about concepts of philosophical interest. Peirce begins Ideas by reflecting upon then orthodox views about making ideas clear. In Descartes and Leibniz, for example, we find influential approaches to making ideas “clear and distinct”. Peirce understands these two traditional grades of clarity as follows. First grade

55

Andrew Howat

clarity (“clearness”) about a kind such as, say, lithium, means basic familiarity, or the ability to apply the concept in ordinary cases (e.g. recognizing samples of lithium based on its distinctive appearance). Second grade clarity (“distinctness”) means one can define lithium in exact, abstract terms, such as by reference to its atomic number. Notice that it is second grade clarity that most closely resembles the identification of an essence as we now understand that term—something like a constitutively necessary condition for a substance’s belonging to kind K. Peirce insists that while distinctness is part of our journey to understanding any idea and is “indispensable to exact reasoning”, there is nevertheless a higher form of clarity available. Scholars typically dub this third grade of clarity “pragmatic meaning”. Peirce proposes the pragmatic maxim as a rule for achieving it: Consider what effects, that might conceivably have practical bearings, we conceive the object of our conception to have. Then, our conception of these effects is the whole of our conception of the object. (5.402, 1878, my emphasis) The italicized phrase suggests that Peirce thinks pragmatic meaning effectively exhausts our conception of the relevant object or kind. For example, it seems to entail the idea that once one understands all of lithium’s many “effects with practical bearings” (e.g. if you place lithium in water, expect it to burn and possibly explode), there is simply nothing more to be said about its “nature”.5 Notice, however, that this method of conceptual clarification construes it as an ongoing project unlikely ever to end (hence Peirce’s fondness for the mathematical metaphor of an “ideal limit of inquiry” in his conception of truth, see Legg 2014). Thus, Peirce’s initial statement of the maxim makes definition seem largely redundant for philosophical purposes. He implies that its only real function lies in introducing a novel word to someone wholly unfamiliar with it. It is only by identifying an object’s potential “effects with practical bearings” that we learn something philosophically substantive about it. Peirce later labels claims that do this “precepts” and argues that they are “more serviceable than a definition” (my emphasis, 2.330, 1902): If you look into a textbook of chemistry for a definition of lithium, you may be told that it is that element whose atomic weight is 7 very nearly. But if the author has a more logical mind he will tell you that if you search among minerals that are vitreous, translucent, grey or white, very hard, brittle, and insoluble, for one which imparts a crimson tinge to an unluminous flame, this mineral being triturated with lime or witherite rats-bane, and then fused, can be partly dissolved in muriatic acid; and if this solution be evaporated, and the residue be extracted with sulphuric acid, and duly purified, it can be converted by ordinary methods into a chloride, which being obtained in the solid state, fused, and electrolyzed with half a dozen powerful cells, will yield a globule of a pinkish silvery metal that will float on gasolene; and the material of that is a specimen of lithium. The peculiarity of this definition—or rather this precept that is more serviceable than a definition—is that it tells you what the word lithium denotes by prescribing what you are to do in order to gain a perceptual acquaintance with the object of the word. Rorty takes a commitment to using the pragmatic maxim to be a rejection of and alternative to any approach to philosophical theorizing that treats definition as a route to genuine insight. He does this by appealing to Peirce’s famous subsequent remarks about truth, which suggest to many scholars a clear rejection of the correspondence theory (qua definition

56

Pragmatism

of truth) in favor of a third-grade clarification specifying truth’s practical consequences for inquiry and belief (1982: 162). This interpretation of the pragmatic maxim is understandable and popular. At one point Peirce bluntly states that “[n]othing new can ever be learned by analyzing definitions”. Nevertheless, Rorty’s interpretation neglects numerous inconvenient truths. Chief among these is the fact that Peirce seems to have verbal or nominal definition in mind here, not real definition (i.e. the type of definition thought by many contemporary metaphysicians to state essences).6 Rorty might, however, have read Peirce’s pragmatism as an argument for the claim that definition can never supply anything more than merely nominal clarity—that “real” definition of the sort touted by Fine 1994 for example, is impossible. It is easy to imagine Rorty suggesting that the contrary view amounts to what Peirce called “ontological gibberish”. This seems to be roughly the position Cheryl Misak takes (2004: 34–5): [Peirce] thinks that if one engages in the project of definition, the danger is that one word will be ‘defined by other words, and they by still others, without any real conception ever being reached’ (CP 5. 423, 1905, see also MS 329, 1904). This, he says, is the flaw of much of metaphysics. The pragmatic maxim, on the other hand, gets to the ‘real conception’ by articulating the consequences of hypotheses containing the term. Thus he thinks that the pragmatic project overshadows any other. If correct, this reply suggests that Peirce denies essentialism understood as the search for real definitions that state essences. However, it is not obvious that this is the only way to understand the term “essentialism” (it is not entailed by the definition of the term given above). Some Peirce scholars argue that he thinks the “effects with practical bearings” we identify using the maxim are the essences of kinds like lithium. Murphey may have been the first to propose this interpretation (1993 [1961]: 158, my emphasis): Instead of seeking a qualitative essence from which the behavior of the thing follows, Peirce now [in 1878] identifies the essence with behavior. The essence of a thing is the sum of the habits it involves. Accordingly, our objective in the investigation of a thing is to discover the laws governing its behavior--i.e. its habits--not the form which serves as the basis for natural classification, for there is no such form. Almeder extends this idea, writing (1980: 181, note 14): … one might argue that Peirce’s conception of the universal in terms of law is not a repudiation rather than a restatement of the scholastic theory of essence [sic]. Indeed, the scholastic maxim operari sequitur esse would seem to suggest that the essence of an object is to be specified in terms of its activities or operations. Hulswit argues that “Peircean essences are of the nature of habit; and habits are, at least in principle, subject to evolution” (1997: 739). This is important because many apparent pragmatist critiques of essence (especially Dewey’s) focus on the idea that they cannot be permanent or fixed. Absent further argument, such a reinterpretation of essence seems compatible with essence/essentialism as defined in §1, even if the resulting view would bear little resemblance to mainstream versions.

57

Andrew Howat

Whether it is Rorty or Murphey et al. who are correct depends, of course, on what Peirce himself thought about essence. Surprisingly, only Hulswit cites Peirce’s own definition of the term, which appears in an unpublished “Note on Metaphysics” from 1909 (6.337): Coming to Essence, this in its epistemological force is that intelligible character which truly defines what a general or indefinite, that is, what an indeterminate monadic predicate primarily asserts, so that all else that it asserts is the necessary consequence of this epistemological essence. It is easy to state what the essences of artificial objects are: The essence of a stove is that it is intended to diffuse warmth. But as to the essence of natural objects, if they have any, we are unable as yet to give them. We are only able to state the essence of our common names for such things. The metaphysical essence is the intelligible element of the possibility of its Being, or so much of that as is not a mere consequence of the rest. This passage does not easily settle the issue, but it does provide some interesting clues. First, it suggests that Peirce seemingly understood essence as tied to definition, not (pace Murphey et al.), to third grade clarification, behavior, etc. (though Peirce was arguably inconsistent in his use of “definition”, so this is not decisive). Second, Peirce thinks “artificial objects” have essences—namely their functions, which is surely an endorsement of (at least) psychological essentialism about conventional kinds (Neufeld 2022). Third, Peirce seems to distinguish between nominal and real essence in his remark about “the essence of our common names”. He also explicitly recognizes this distinction elsewhere (NEM 4:285) and even seems to suggest that, in retrospect, he should have incorporated it into the presentation of the grades of clarity (ILS: 193). Fourth, he seems open to the possibility that natural objects might well have real essences, but also claims we cannot yet give them. There are at least two possible explanations for Peirce’s reticence on this final point. If we assume that Peirce is acknowledging the possibility of real definitions, then he may have been hesitant simply because in the early 1900s, chemistry had not yet advanced beyond Thomson’s “plum pudding” model. Thus, our contemporary notion of individuating elements by appeal to atomic number, for example, was not fully fledged (readers may have noted the use of atomic weight in the above quotation from 2.330). Similarly, absent a satisfactory account of chemical bonding, it may have been less tempting to Peirce to think of truths like “water is H2O” as real definitions, or Peirce may have had other suspicions about their adequacy (of the sorts described in Chang 2012, for example). If, however, we assume that Peirce has third grade clarifications in mind when he speaks of essences (mistakenly writing “definition” instead of “precept”), his hesitancy might simply reflect the fact that inquiry is forever ongoing and given his basic idealism, it is only the ultimate conception that captures the real essences of kinds. Hulswit (1997: 739) seems to have this in mind when he writes: There is no metaphysical essence independent of our knowing processes; the metaphysical essence is independent of how you and I or anyone else at a specific moment characterizes the world. In the long run, however, the classifications of the scientific society will reflect the metaphysical essence of things. Much of the confusion arises from the fact that, though Peirce tries to break with traditional ways of conceiving the question of universals, essences, or kinds, he does not break with traditional language.

58

Pragmatism

Thus, in summary, there are those—like Rorty—who read Peirce as a strident anti-essentialist based solely on his critique of definition as a source of philosophical insight in 1878. This interpretation glosses over various important historical and textual nuances. There are also those who read Peirce as endorsing some alternative form of pragmatist essentialism, where the essences of things are the laws or habits that govern their behavior. This interpretation does not gel perfectly with the way Peirce defined “essence”. However, that definition by itself seems consistent, at a minimum, with psychological essentialism about conventional kinds, and potentially with metaphysical essentialism about natural kinds. No scholar has yet produced a definitive account of “Peircean essentialism”, but I suspect one could well be built out of the raw materials in Hulswit 1997 and 2002 (which connects essence to Peirce’s unique scientific account of final causation), and Short 2007 (which argues that Peirce anticipated the modal essentialism of Kripke and Putnam). These precedents are at least sufficient to show that Rorty’s case for equating pragmatism and anti-essentialism is overstated.

4.3 James Rorty might respond to the above by arguing that Peirce, despite his founding role, was not really a pragmatist. He might point out that Peirce himself even re-labelled his position “pragmaticism”, precisely to distinguish his views from those of James and Schiller (see e.g. Misak 2016). Thus, in this section I consider whether close examination of James’s views provides stronger support for describing pragmatism as anti-essentialism. I argue it comes closer but still falls short. James hugely admired Peirce’s maxim and set out to popularize, apply, and extend it in novel ways. However, in announcing his understanding of “the pragmatic method” to the world in 1898, he offered a much more strident interpretation of its deflationary and skeptical implications (James 1898: 290–1). This interpretation no doubt reflected James’s prior challenge to the traditional notion of essences as “absolute”: There is no property ABSOLUTELY essential to any one thing. The same property which figures as the essence of a thing on one occasion becomes a very inessential feature upon another … Whichever one of these aspects of its being I temporarily class it under, makes me unjust to the other aspects. But […] I always am classing it under one aspect or another, I am always unjust, always partial, always exclusive. My excuse is necessity -- the necessity which my finite and practical nature lays upon me. My thinking is first and last and always for the sake of my doing, and I can only do one thing at a time … . (James 1890/2018: 1121) What does it mean to reject the idea that essences are “absolute” and how does that connect with the definitions in §1? The contemporary reader is likely to understand “absolute” in terms of mind-independence. That is, on an absolutist interpretation, a kind having an essence must be a wholly mind-independent affair. For example, it must be a matter of lithium’s possessing the atomic number 3, a property it possesses entirely independently of human minds, concepts, etc. If James is denying that essences are mind-independent in this sense, that suggests he is an anti-essentialist. The problem with this is obvious from the definitions given in §1—such a claim is compatible with anti-realist essentialism. On such views, the essentiality of a property is grounded in our practical interest in that property, not the other way around. If “anti-realist essentialism” is a coherent position, then James could merely be endorsing a view along these lines.

59

Andrew Howat

Whatever view James is endorsing, he sees it as resulting from Peirce’s idea that concepts are purposive—that they are instruments for doing certain kinds of work or solving certain kinds of problems. This is presumably why James goes on (shortly after the above passage, Ibid: 1122) to write that: the only meaning of essence is teleological, and … classification and conception are purely teleological weapons of the mind. The essence of a thing is that one of its properties which is so important for my interests that in comparison with it I may neglect the rest … The properties which are important vary from man to man and from hour to hour. This view influenced Dewey, and Kitcher (2012: 136) argues that the result is as follows: Although James and Dewey are both adamant that there is an independent reality to which our thoughts and actions respond … they insist that this independent reality is not independently structured: it doesn’t come pre-divided into privileged objects and kinds of objects. Again, while this can reasonably be read as anti-essentialism, it can also clearly be read as anti-realist essentialism. For my part I can find nothing in James’s corpus that could help settle the question of which of these two is the better label. A better approach might be to examine how these ideas evolved in Dewey’s work, to see whether he understood or developed them in a decisively anti-essentialist direction.

4.4 Dewey Rorty’s favorite pragmatist is undoubtedly Dewey. Thus, if any Classical pragmatist ought to exhibit a clear commitment to anti-essentialism, it should be him. Rorty’s case for reading him this way has merit, but again is neither definitive nor uncontested. On the contrary, there is no shortage of Dewey scholars positively outraged by Rorty’s interpretation (e.g. Sleeper 1986). There are at least three ways to read Dewey on essence/essentialism. First, there are remarks that suggest a passive anti-essentialism, rooted in metaphysical quietism. Second, a focus on his later “naturalist” phase tends to suggest an active anti-essentialism, rooted in his own “empirical naturalist” metaphysics (thereby contradicting the first reading).7 Third, as with Peirce, there are those who reconstruct out of Dewey’s work a novel pragmatist approach to essence, one that amounts to a sort of “objective relativism” (Boisvert 1988; Mounce 2002:158). I will briefly summarize each interpretation, but the final lesson is already clear—none of the figures usually credited with founding pragmatism had a clear or comprehensive stance on essence/essentialism. Many of Dewey’s critical remarks about essence suggest he endorses what (in §1) I defined as a passive form of anti-essentialism, i.e. a rejection of essence/essentialism rooted in metaphysical quietism. Macarthur writes that “the aim of the quietist, in the region of philosophical thought to which it applies, is not to embrace philosophical doctrines or theories but to earn the right to live without them” (2008: 196). On this view, Dewey does not argue for the claim “essences do not exist”, but rather tries to persuade us that we can live without the concept altogether. Works like The Quest for Certainty and Reconstruction in Philosophy certainly seem to fit this quietist profile. In both, Dewey offers a debunking genealogy of essence. He argues that

60

Pragmatism

we are motivated to postulate them by the “precariousness” of our empirical existence, especially with respect to values: … today many persons find a peculiar consolation in the face of the unstable and dubious presence of values in actual experience by projecting a perfect form of good into a realm of essence, if not into a heaven beyond the earthly skies, wherein their authority, if not their existence, is wholly unshakeable. (Dewey 1960/1929: 34) Later in the work he generalizes from the case of values to essences of other kinds, e.g. physical (in Chapter 5), and mathematical (in Chapter 6). Dewey 1920: 61 also contrasts the worldview of the ancients, which embraced the idea of essences, with the worldview of the contemporary scientist, clearly favoring the latter: The laws in which the modern man of science is interested are laws of motion, of generation and consequence. He speaks of law where the ancients spoke of kind and essence, because he what he wants is a correlation of changes, an ability to detect one change occurring in correspondence with another. He does not try to define and delimit something remaining constant in change. He tries to describe a constant order of change. This reading of Dewey sees him presenting essences/essentialism as an otiose remnant of a bygone era, while urging us to abandon them and turn our focus to real (i.e. pragmatically legitimate) problems instead (Gale 2002; Mounce 2002; Seigfried 2004). In some of his later works, Dewey seems to many scholars to exhibit a change of heart. Richard Gale writes: “as Dewey matured as a philosopher, he came to realize that metaphysics need not be pernicious, since there is a legitimate empirical way to do it, which […] can make us better inquirers” (2002: 483, see also Bernstein 1961). Experience and Nature appears to fit this profile, though Rorty, Hook, Seigfried, and others attempt to explain this appearance away.8 Gale 2002: 487 summarizes the core argument of the book as follows: 1 2 3 4

Every existent is an experience. Every experience has the generic traits of existence. The generic traits of existence are inquiry-related traits. Every existent has inquiry-related traits.

The “generic traits of existence” claim is that every existent (a) exhibits some mixture of the precarious and the stable, the settled and unsettled, the determinate and the indeterminate; (b) possesses a unique qualitative individuality; and (c) is an event or at least processual (Gale 2002: 485). How do these claims amount to or entail anti-essentialism? On many definitions, an essential property is intrinsic, i.e. it is a property an object has solely “in virtue of the way that thing itself, and nothing else, is” (Lewis 1983: 197).9 Thus, while it might make sense to think that essential properties are things we can and do experience (e.g. the four sides of my rectangular desk), it is hard to see how it can make sense to think those properties are themselves experiences. This would mean that, for example, lithium did not have an essence until someone, somewhere experienced it (e.g. somehow detected its atomic number). However, three problems remain for those who consider this a definitive case for Dewey’s anti-essentialism. First, there are those who argue that a property’s being mind-dependent and

61

Andrew Howat

thus extrinsic/relational (e.g. qua socially constructed) is compatible with its being essential (see Chapter 31). Second, Gale points out that what Dewey means by the term “experience” is almost certainly not what we normally mean by it (2002: 502). For Dewey, the term includes both actual and merely possible experiences (in our usual sense of the term). It also includes both the object experienced (the “what”) and the way we experience it (the “how”). Such an “all-inclusive” conception of experience does not obviously conflict with essentialism in the way alleged above, though it may conflict in other, yet unexplored ways. Third, as Godfrey Smith 2016 argues, Dewey may have entirely rejected the intrinsic/extrinsic distinction (in merely one of his many assaults on traditional metaphysical terminology), making it even more difficult to state any clear relationship between his views and essentialism of whatever stripe. Finally, some who read Dewey as embracing a pragmatist metaphysics believe, ironically, that it produces a commitment to essentialism, albeit in a novel pragmatist form. This view differs from the Peircean one gestured at in §1 (Dewey appears far less comfortable with scholasticism in any form). Its primary inspiration seems instead to be James’s (1909a) radical empiricism, whose central claim is that “the relations between things, conjunctive as well as disjunctive, are just as much matters of direct particular experience, neither more so nor less so, than the things themselves” (James 1909b: 826). Author and advocate of “Dewey’s Metaphysics” Raymond Boisvert writes that “Dewey does not admit any essence hidden behind or within the accidental qualities of an entity”, but then immediately qualifies the claim in a rather surprising way (1988: 132): Dewey rejects the hypostatized interpretation of forms which would make of them an inner essence which is really “what” a thing is. In spite of this rejection, the what remains a significant part of his analysis. If my interpretation is correct, in elaborating on this “whatness” we can find Dewey’s alternative to the classical doctrine of form. Boisvert attempts to reconstruct this alternative from Dewey’s Art as Experience (1934), writing that “the remarks made there about forms can be generalized into a defensible ontological position”. This ontological position is widely thought to be “objective relativism”, which Lovejoy defined as follows: The existence and character of experienced data depends upon the occurrence of percipient events and therefore upon the nature and situation of the organism which has the experience; and it is only ‘in relation’ to a given organism that the object known possesses the character exhibited by the datum. Nevertheless, all perceptual content is stoutly declared to be ‘objective’.10

4.5

Summary and Conclusion

This chapter may leave the reader with more questions than answers about pragmatism’s relationship to essence and essentialism. If so, this reflects three important, but esoteric facts about the tradition. First, over the course of his career, the views of pragmatism’s founder C.S. Peirce on metaphysics evolved in ways that fundamentally altered his pragmatism (so much so he renamed it). Second, James and Dewey, like most subsequent pragmatists, often downplayed or ignored this evolution, while advocating for and popularizing various “pragmatist” views that Peirce rejected. Third, the anti-essentialist understanding of concepts that many now associate with pragmatism probably evolved gradually out of James and

62

Pragmatism

Dewey’s appropriation and modification of Peirce’s maxim, but was not solidified until much later (e.g. by Quine), especially once essence/essentialism began to be reframed in modal logical terms. Thus, while we can easily find various claims and arguments redolent of anti-essentialism in the Classical pragmatist tradition and it is reasonable to think these laid the groundwork for subsequent assaults on notions of essence/essentialism, it is likely that Peirce, James, Dewey and perhaps others simply took issue with a traditional metaphysically realist notion of essences/essentialism. The result of these objections might be something more like antirealism about essence/essentialism and arguments for developing a distinctively pragmatist version of that position. Subsequent pragmatists, such as Rorty, with the benefit of new logical insights and techniques, may then have decided that such an endeavor was folly. If so, then they were taking pragmatism in a new direction not dictated by its historical roots. Related Topics: Natural Kind Essentialism, Social Construction, Conferralism, Quinean Anti-Essentialism, Wittgenstein. Further reading: Peirce’s On Science and Natural Classes (1902) in Vol. 2 of The Essential Peirce, Short 2007 (especially Chapters 4–6), Hulswit 1997 and 2002.

Notes 1 Again, 1982: 162. Rorty frequently made claims of the form “Pragmatists believe x” or “We pragmatists claim that y”. Yet by any reasonable historical assessment, the label “pragmatism” is associated with numerous diverse claims, including (a) beliefs are at least partly constituted by habits of action; (b) the pragmatic maxim is the best method for clarifying intellectual concepts; (c) Peirce’s “method of science” is the best method of fixing belief; (d) truth is at least partly constituted by what would be believed at the end of inquiry; (e) the truth of p is at least partly constituted by the practical benefits of belief in p; (f) under certain conditions, non-evidential considerations can contribute to holding a belief’s being epistemically or morally justified (or both); (g) metaphysical realism is false, (h) one can be both anti-skeptical and a fallibilist, (i) there is no fundamental dichotomy between fact and value, and (j) in some philosophically important sense “meaning is use”, etc. Pragmatism is associated with all these claims, but different pragmatists endorse different ones for different reasons. 2 Many major texts on pragmatism contain few or even zero mentions of “essence” and “essentialism” (even texts with titles like “Peirce’s Scientific Metaphysics”, Reynolds 2002 and “The Metaphysics of William James and John Dewey”, Martland 1963). 3 I have intentionally avoided a modal characterization of essence/essentialism for two reasons. First, there is obviously much disagreement on the centrality of modal notions in our understanding of essence ( Chapter 8). Second, one thing that is abundantly clear in the historical record is that Peirce, James, and Dewey disagreed strongly on the nature and sources of necessity (see Hickman 1987: 447). 4 This is adapted from Robertson Ishii and Atkins 2020. I focus solely on kind essentialism in this chapter. I do so because when philosophers think of pragmatism and essentialism together, they generally think of a contrast between a metaphysically realist view of kinds (as “carving nature at the joints”) and an anti-realist one (where kinds are partly or wholly determined by our practical interests, linguistic frameworks, conceptual schemes, etc.), e.g. Carnap 1950, Putnam 1987, etc. 5 The attentive reader will have noticed an apparent equivocation here between “our conception of” the object and the object itself. Just because the “effects with practical bearings” exhausts the former, does not obviously entail that they should exhaust the latter. Thus, why should Peirce’s remarks here be taken to have anything to do with essences, rather than with merely verbal/nominal definitions? The important thing to remember here is Peirce’s basic idealism, defined by Lane 2018. This view allows that there can be a very large gap between our current conception of an object and the object itself (as we’d expect if inquiry concerning elements, etc. is ongoing). But it also entails there cannot be a gap between our ultimate conception (what we would believe about the object, were inquiry concerning it carried indefinitely far) and reality.

63

Andrew Howat 6 In 1910 Peirce explicitly also denied that he was discounting the significance of distinctness/ definition and instead argued for the “symmetrical development” of all three grades of clarity. I believe this has radically revisionary implications for our understanding of pragmatism’s conception of truth and of the maxim. As Lane 2018 shows, it entails that Peirce did not reject correspondence theory. Instead, he set out to clarify correspondence theory by specifying its practical content. I suspect these arguments bolster the case against Rorty’s anti-essentialist reading of Ideas, see Howat 2021. 7 David Macarthur writes: “Taken as a whole, Dewey’s career wavers uncomfortably between metaphysical quietism and a metaphysics of experience” ( Macarthur 2008, fn. 14: 207. See also Rorty in Cahn 1977). 8 Shea 1989:84 writes: “[Rorty and Hook] claim that Dewey made a mistake calling his magnum opus, Experience and Nature, a work in metaphysics. And they think he knew he made a mistake and repudiated it”. Gale 2002: 488 cites Rorty’s remark that “It is easier to think of the book [Experience and Nature] as an explanation of why nobody needs a metaphysics rather than as itself a metaphysical system”. See also Seigfried 2001:15. 9 This is not to say, as the editors helpfully pointed out to me, that it is correct simply to assume that essences are intrinsic, as doing so immediately rules out many claims that are seriously entertained by essentialists (e.g. Kripke’s essentiality of origins or that essences might be associated with social entities). 10 This is cited in Mounce 2002, with useful critical discussion: 158–165.

References If a passage occurs in Peirce’s Collected Papers, the citation is “n.m (year)” where n is the volume number, m the paragraph number, and the year is that of the quoted text. ILS refers to Peirce 2014. NEM refers to Peirce 1976. Almeder, Robert. The Philosophy of Charles S. Peirce. Totowa, NJ: Rowman and Littlefield, 1980. Ásta. “Essentiality Conferred.” Philosophical Studies 140, no. 1 (2008): 135–148. Atkin, Albert. Peirce. New York, NY: Routledge, 2016. Ayer, A.J. The Origins of Pragmatism: Studies in the Philosophy of Charles Sanders Peirce and William James. San Francisco: Cooper, Freeman, and Co, 1968. Bernstein, Richard J. “John Dewey’s Metaphysics of Experience.” The Journal of Philosophy 58, no. 1 (1961): 5–14. Boisvert, R.D. Dewey’s Metaphysics. Fordham American Philosophy. New York, NY: Fordham University Press, 1988. Cahn, S.M. New Studies in the Philosophy of John Dewey. Donald F. Koch American Philosophy Collection. Hanover, NH: University of Vermont, 1977. Carnap, Rudolf. “Empiricism, Semantics, and Ontology.” Revue Internationale de Philosophie (1950): 20–40. Chang, Hasok. Is Water H2O?: Evidence, Realism and Pluralism. Vol. 293. Springer Science & Business Media, 2012. Devitt, Michael. “Resurrecting Biological Essentialism.” Philosophy of Science 75, no. 3 (2008): 344–382. Dewey, John. Experience and Nature. Paul Carus Lectures. London, UK: George Allen & Unwin Ltd, 1929/1925. Dewey, John. Reconstruction in Philosophy. Beacon Paperback. Beacon Press, 1957/1920. Dewey, John. The Quest for Certainty: A Study of the Relation of Knowledge and Action. Gifford Lectures. New York, NY: Capricorn Books, 1960/1929. Dutilh Novaes, Catarina. “Carnap Meets Foucault: Conceptual Engineering and Genealogical Investigations.” Forthcoming in Inquiry. 10.1080/0020174X.2020.1860122. Fine, Kit. “Essence and Modality”. Philosophical Perspectives 8(Logic and Language) (1994): 1–16. Gale, Richard M. “The Metaphysics of John Dewey.” Transactions of the Charles S. Peirce Society 38, no. 4 (2002): 477–519. Giladi, Paul. “Battling for Metaphysics: The Case for Indispensability.” Metaphysica 18, no. 1 (2017): 127–150.

64

Pragmatism Godfrey-Smith, Peter. “Dewey and the Question of Realism.” Noûs 50, no. 1 (2016): 73–89. 10.1111/ nous.12059. Hickman, Larry. “R. W. Sleeper, ‘The Necessity of Pragmatism: John Dewey’s Conception of Philosophy’ (Book Review).” Transactions of the Charles S. Peirce Society. Buffalo, NY: The Society, 1987. Howat, Andrew. “Pragmatism and Correspondence.” Philosophia 49, no. 2 (2021): 685–704. Howat, Andrew. “Pragmatism and Philosophical Methods.” In Routledge Companion to Pragmatism, S. Aikin and R. Talisse (Eds.). New York, NY: Routledge, 2023. Hulswit, Menno. “Peirce’s Teleological “Approach to Natural Classes.” Transactions of the Charles S. Peirce Society 33, no. 3 (1997): 722–772. Hulswit, Menno. From Cause to Causation: A Peircean Perspective. Vol. 90. Springer Science & Business Media, 2002. James, W. Delphi Complete Works of William James (Illustrated). Delphi Series Nine. Hastings, UK: Delphi Classics, 1890/2018. James, William. “Philosophical Conceptions and Practical Results.” In The Pragmatism Reader: From Peirce Through the Present, Robert B. Talisse and Scott F. Aikin (Eds.), 66–78. Princeton University Press, 1898/2011. James, William. Pragmatism. Cambridge University Press, 1907/2014. James, William. William James: Writings 1902–1910: The Varieties of Religious Experience/ Pragmatism/A Pluralistic Universe/The Meaning of Truth/Some Problems of Philosophy/Essays, Vol. 2. New York, NY: Literary Classics of the United States, Inc, 1909b/1987. James, William. Essays in Radical Empiricism. Vol. 3. Cambridge, MA: Harvard University Press, 1976. Khalidi, Muhammad Ali. “Three Kinds of Social Kinds.” Philosophy and Phenomenological Research 90, no. 1: 96–112, 2015. Kitcher, Philip. Preludes to Pragmatism: Toward a Reconstruction of Philosophy. Oxford, UK: Oxford University Press, 2012. Koslicki, K. “Modality and Essence in Contemporary Metaphysics”, forthcoming in Modality: A Conceptual History (Oxford Philosophical Concept Series), Sam Newlands and Yitzhak Melamed (Eds.). Oxford, UK: Oxford University Press. Lane, Robert. Peirce on Realism and Idealism. Cambridge, UK: Cambridge University Press, 2018. Legg, Catherine. “Charles Peirce’s Limit Concept of Truth: Peirce’s Limit Concept of Truth.” Philosophy Compass 9, no. 3 (March 2014): 204–213. Legg, Catherine, and Paul Giladi. “Metaphysics—Low in Price, High in Value: A Critique of Global Expressivism.” Transaction of the Charles S. Peirce Society 54, no. 1 (2018): 64–83. Lewis, David. “Extrinsic Properties.” Philosophical Studies 44, no. 2 (1983): 197–200. Macarthur, David. “Pragmatism, Metaphysical Quietism, and the Problem of Normativity.” Philosophical Topics 36, no. 1 (2008): 193–209. Martland, T. R. “Dewey’s Rejection and Acceptance of a Metaphysic.” The Monist 48, no. 3 (1963): 382–391. https://doi.org/monist196448323. Misak, C. J. Truth and the End of Inquiry: A Peircean Account of Truth. Expanded pbk. ed. Oxford, New York: Clarendon Press; Oxford University Press, 2004. Misak, Cheryl. Cambridge Pragmatism: From Peirce and James to Ramsey and Wittgenstein. Oxford, UK: Oxford University Press, 2016. Mounce, Howard. The Two Pragmatisms: From Peirce to Rorty. New York, NY: Routledge, 2002. Murphey, Murray G. The Development of Peirce’s Philosophy. Indianapolis, IN: Hackett Publishing, 1993/1961. Neufeld, E. Psychological Essentialism and the Structure of Concepts. Philosophy Compass, 17, no. 5 (2022). Peirce, Charles S. Collected Papers of Charles Sanders Peirce, C. Hartshorne and P. Weiss (vols. i–vi), A. Burks (vols. vii and viii) (Eds.), Cambridge, MA: Belknap Press, 1931–58. Peirce, Charles S. The New Elements of Mathematics, Carolyn Eisele (Ed.). The Hague: Mouton, 1976. Peirce, Charles S. Illustrations of the Logic of Science, Cornelis de Waal (Ed.). Chicago: Open Court, 2014. Price, Huw. Naturalism without Mirrors. Oxford, UK: Oxford University Press, 2011. Putnam, Hilary. “Truth and Convention: On Davidson’s Refutation of Conceptual Relativism.” Dialectica 41, no. 1–2 (1987): 69–77.

65

Andrew Howat Reynolds, Andrew. Peirce’s Scientific Metaphysics: The Philosophy of Chance, Law, and Evolution. Nashville, TN: Vanderbilt University Press, 2002. Robertson Ishii, Teresa and Philip Atkins. “Essential vs. Accidental Properties”, The Stanford Encyclopedia of Philosophy (Winter 2020 Edition), Edward N. Zalta (Ed.). https://plato.stanford. edu/archives/win2020/entries/essential-accidental/ Rorty, R. “Pragmatism, relativism, and irrationalism.” In Proceedings and Addresses of the American Philosophical Association, 53 (1980): 717–738. Rorty, Richard. Consequences of Pragmatism: Essays, 1972–1980. University of Minnesota Press, 1982. Seibt, Johanna, “Process Philosophy”, The Stanford Encyclopedia of Philosophy (Spring 2022 Edition), Edward N. Zalta (Ed.). https://plato.stanford.edu/archives/spr2022/entries/process-philosophy/ Seigfried, Charlene Haddock. “Pragmatist Metaphysics? Why Terminology Matters.” Transactions of the Charles S. Peirce Society 37, no. 1 (2001): 13–21. Seigfried, Charlene Haddock. “Ghosts Walking Underground: Dewey’s Vanishing Metaphysics.” Transactions of the Charles S. Peirce Society 40, no. 1 (2004): 53–81. Shea, William M. “[Review of John Dewey’s Theory of Art, Experience and Nature: The Horizons of Feeling; Dewey’s Metaphysics, by T. M. Alexander & R. D. Boisvert].” American Journal of Education 98, no. 1 (1989): 83–88. Shook, John R. Dewey’s Empirical Theory of Knowledge and Reality. Vanderbilt University Press, 2000. Short, Thomas Lloyd. Peirce’s Theory of Signs. Cambridge, UK: Cambridge University Press, 2007. Sleeper, Ralph W. The Necessity of Pragmatism: John Dewey’s Conception of Philosophy. University of Illinois Press, 2001. Talisse, Robert. “Pragmatism and the Limits of Metaphilosophy”. The Cambridge Companion to Philosophical Methodology, Soren Overgaard and Giuseppina D’Oro (Eds.). Cambridge, UK: Cambridge University Press, 2017.

66

5 CONTEMPORARY (PHENOMENOLOGICAL TRADITION) Kevin Mulligan

5.1 Introduction “Essence-intoxicated” was the way one early phenomenologist, Scheler, was described.1 The epithet applies equally well to the other early phenomenologists. Husserl, Pfänder, Reinach, Stein, Geiger, Ingarden and Hering all believed in essence, that there are essences, that they are knowable, that they are essential to philosophy and terribly important. Belief in a transcendental ego or the transcendental reduction was not essential to being a phenomenologist. Belief in essence and its knowledge through intuition was. That there are “laws about essences (Wesenheiten)” is “one of the most important things in philosophy and—if one thinks it through to the end—in the world”.2 The essential facts discovered by phenomenology, Scheler claims, underlie “all other facts, the facts of the natural world-view and the facts of the scientific world-view, the connections between essential facts underlie all other connections”.3 The phenomenologists’ confidence in their ability to intuit essences and connections between them (essential connections) seems to have done much to discredit phenomenology and to have led to a lack of interest in their views about essence and its roles, in particular its relation to modality and grounding, not least amongst later so called phenomenologists. Meinong, unlike Husserl and his followers, did not go in for intuiting essences and was never philosophically intoxicated. But he, too, thought that essences and natures play a central role in philosophy. What, then, did the phenomenologists take essences and their connections to be? What rôles, philosophical and non-philosophical, did they assign to essence? What sort of contact, epistemic and non-epistemic, with essence did they take themselves to enjoy?4

5.2

Essences and Their Connections

What is an essence? Essences, it is often thought, are properties which, if exemplified, are exemplified essentially. This is the answer Husserl may well have in mind in his programmatic account in 1913: a temporal object “has its stock of essential predicables (Prädikabilien), which it “exemplifies” (zukommen).5 Here, then, are two familiar ideas: predicates combine with the adverb “essentially” to form more complex predicates and we may ascend from such predications to mention of

DOI: 10.4324/9781003008750-7

67

Kevin Mulligan

essential properties. This type of ascent, as we shall see in §5, corresponds, Husserl thinks, to a special type of epistemic ascent. Reinach, too, understands essence in this way. There is an apparently different way of understanding essence. On this view, essences are kinds, species, types or ideal objects which are or may be instantiated rather than exemplified. Thus Husserl gives as examples of what he calls “judgments about essence” A color as such differs from a tone as such. The essence (the “genus”) color is not the essence (“genus”) tone.6 In his Logical Investigations he often refers to species (Spezies), their instances or cases and instantiation (ist ein Einzelfall von, Vereinzelung) and seems to think of colors, tones as well as propositions, judgment, supposition, desire and feeling as such species. Unfortunately, he does not tell us what the relation is between red or color, on the one hand, and properties such as being red or being a color, on the other hand. The phenomenologists sometimes distinguish between essence understood as something general, a property or kind, and individual essences.7 Sometimes this difference is marked by reserving Wesenheit (essentiality), Eidos or Idee (idea) for the former and Wesen for the latter.8 Often, Wesen is used to mean sometimes one and sometimes the other and the reader has to struggle. Individual essences function in some respects like tropes, which play a central role in Husserl’s ontology (he calls them Momente). Individual essences, like tropes, are incomplete without and depend on substances and the dependence is a relation of specific or token dependence. But whereas every substance is incomplete without its individual essence, the substance does not depend on the tropes which depend on it. Husserl’s philosophy of tropes resembles in many respects the account given by D. C. Williams, who was familiar with Husserl’s views. In particular, Williams endorses Husserl’s view that tropes are “abstract parts” of substances. Williams also notes the similarities between tropes and individual essences.9 To be an essence, then, is to be something general, a multiply exemplifiable property or a multiply instantiable kind, or something particular, an individual essence. A quite distinct account of essence is given by Scheler in 1913 and later: “An essence (Wesen) or whatness is, in this sense as such neither something general nor something individual”.10 This is also the view of Stein—Wesenheiten are “neither general nor individual”11—who attributes the view to Duns Scotus. Versions of the view that essence is in itself indifferent or neutral with respect to generality and particularity go back to Avicenna and Aquinas. On Scheler’s view, then, essence is as such neither general nor individual but there are alsogeneral and individual essences. An essence is general in certain contexts and individual in other contexts. The claim resembles the neutral monism of Mach (versions of which are later endorsed by James and Russell): “elements” are not in themselves mental or physical or bodily but are only such in certain combinations with other elements. But Scheler’s view of essence concerns the two formal categories of generality and particularity whereas Mach’s elements are neutral with respect to material categories such as the mental and the psychological. One immediate precursor of Scheler’s view is James’ 1890 claim that the “conception of an abstract quality is, taken by itself, neither universal nor particular”.12 Scheler expounds his view as follows: The essence red, for example, is given in the general concept as well as in each perceivable nuance of this color. The differences between universal and particular meanings come about only in relation to the objects in which an essence comes to the fore. Thus, an essence becomes universal if it comes to the fore in a plurality of otherwise

68

Contemporary (Phenomenological Tradition)

different objects as an identical essence: in all and everything that “has” or “bears” this essence. The essence can, on the other hand, also constitute the nature of an individual thing without ceasing to be such an essence.13 Scheler and Stein are silent about how to talk about neutral essence. One plausible view is that infinitives are the expressions which have neutral essences as their semantic values. Then infinitival specifications of essence of the form “To F is to G”—for example, to be a person is to be free—may be distinguished from specifications of the form “To F is just to G”—to know that p is just to believe correctly that p and for good reason.14 These specifications are characterized by the way they may occur in constructions such as “To F is in part to G, in part to H”, “What it is to F”, “Part of what it is to F”. The phenomenologists never seem to consider the role of infinitives in their philosophies of essence although like so many philosophers they employ infinitives in advancing essentialist, philosophical claims and, like other heirs of Brentano, nominalized infinitives (das Urteilen). Husserl’s official account of the right way to talk about essence accords pride of place to generic uses of nominal expressions, as above. But when he advances philosophical claims he sometimes employs infinitives, sometimes combining them with nominal expressions (“The essence of consciousness is to be conscious ‘of something’”).15 A statement of essence (Wesensaussage), then, formulated with nominal expressions or infinitives, corresponds to a connection between essences (Wesenszusammenhang). (A related term is Wesensverhalt, which is then contrasted with Sachverhalt or state of affairs). What is the relation between essence and essential connections, on the one hand, and generality, in particular, law-likeness, modality and the apriori, on the other hand? How are expressions such as “wesensgesetzlich”, “wesensnotwendig”, “Wesensallgemeinheit”, so beloved of the phenomenologists, to be understood?

5.3

Essence, Generality, Modality and the Apriori

If a philosophical claim has the form: to be F (an N) is to be G (or: An N as such is G) then, say many phenomenologists, whatever is F is G and a fact of the first form grounds a fact of the second form. This sort of generality is frequently called “essential generality” (Wesensallgemeinheit) and is intimately linked to but distinct from necessary generality. In order to understand these claims it is important to be bear in mind the difference between two opposed views about the relation between laws and modality and two opposed views of facts. Should modality be understood in terms of laws? Is to be necessary that p nothing more than for there to be a law that p or laws from which it follows that p? Frege outlines a view of this type at the beginning of his Begriffsschrift. Russell thought that for it to be necessary that p is just for it to be always the case that p. A very different view is that necessity cannot be reduced to laws. This sort of view is defended by Meinong and the phenomenologists. The most fundamental type of necessity (Wesensnotwendigkeit) is grounded in law-like connections which are grounded in the essences of different types of objects. One view of facts is that they are, as Frege claimed, true truth-bearers. A radical alternative is that facts are obtaining states of affairs, the view of Meinong, Husserl and Reinach. If, like these philosophers, one thinks that both truth and necessity have bearers, how should instances of “It is true that p”, “It is necessary that p”, and “It is necessary that it is true that p” be understood? The bearers of truth and of necessity are very different, say Husserl and Reinach: the bearers of truth are propositions but the locus of necessity is states of affairs.16 Similarly, according to Meinong the bearers of necessity are objectives. One apparent consequence of this sort of view is that propositions (or other truth-bearers) enjoy modal

69

Kevin Mulligan

properties only derivatively. Another is that states of affairs which are general and obtain in virtue of essences and essential connections obtain necessarily and so are facts (but not matters of fact). “An essential connection”, Reinach says, “is what necessity stands on”.17 As far as I can see, no phenomenologist provides an account of the relation between general states of affairs, de re necessity and the necessity of states of affairs.18 Scheler points to the difference between essential connections and the generalities they ground by saying that essential connections are unilateral, bilateral, etc., but do not involve the notion of a condition, necessary or sufficient. “In an ‘essential connection’ nothing like a ‘condition’ is to be found … Only the application of essential connections leads to ‘conditions’”.19 Since his essential connections connect only neutral essences they do not connect objects and general properties. Unilateral essential connections abound in the writings of the phenomenologists. Equivalences which are rooted in bilateral essential connections (instances of “To F is just to G”) are much rarer.20 Husserl and other phenomenologists often distinguish between essential generality (essential necessity) and natural generality (empirical necessity), the generality of ideal laws and the generality of scientific or real laws.21 The proposition that all bodies are heavy does not “have the unconditional generality” of ideal laws; as a natural law it brings with it a positing of existence, of the existence of nature itself, of spatio-temporal actuality”. Ideal laws involve no such positing of existence and, unlike natural laws, are fully grounded in essence. Is the generality peculiar to essential generality just the type of generality to be found in natural generality? Scheler’s answer to this question modifies earlier, Austrian views. According to Brentano, all generality is negative: to judge that every F is G is just to deny that there are Fs which are not G. In an equally psychological mood, Mach, in a discussion of the sense and value of natural laws, had claimed that “natural science can be understood as a collection of instruments … for the greatest possible restriction (Einschränkung) of our expectation in future cases”. Scheler subsequently distinguishes between natural and essential generality by allowing that the former is negative and the latter, “insofar as it is a consequence of an essential connection”, positive. Scheler gives no argument for the claim that essential generality is positive but it perhaps rests on the view that the generality of a universal such as Red is more fundamental than that of predicative generality. Husserl thinks that “Red” in “Red is a color” is a proper name and, like Reinach, that the semantic value of “Red” is something general, a general quality or object.22 Scheler does not say what he takes the negative nature of natural generality to involve but the simplest version of his view is perhaps the claim that in the case of natural, inessential generality every F is G because no F is not G.23 Early phenomenology abounds in taxonomies of essence over and beyond the four-way distinction (between general properties, general kinds, individual and neutral essences) with which we began: essence formal vs material, proper vs improper, conditional vs nonconditional, exact vs vague; propositions about essence vs eidetic propositions; essences vs concepts, essence vs sense, the essence of essences vs the essence of non-essences. Similarly, modality and the apriori, with which essence and generality are so closely bound up are the objects of distinctions and theories. The subjective apriori is sharply distinguished from the objective apriori and it is argued or assumed that the latter is the more fundamental. Husserl added to the already mentioned distinction between essential and natural necessity what he called non-theoretical (axiological and practical) necessity and necessitation. Essential and natural necessity are the two forms of theoretical necessity but axiological and practical necessity are non-theoretical forms of necessity. Axiological and practical necessity are forms of essential necessity but differ from theoretical, essential necessity in virtue of the fact that the axiological principle of the Excluded Fourth (everything is either positively valuable,

70

Contemporary (Phenomenological Tradition)

negatively valuable or axiologically indifferent) has no counterpart in the world which is the object of theory.24 The theses of the early phenomenologists, in every area of philosophy, are typically formulated with the help of the above mentioned distinctions between essentiality, necessity and necessary possibilities and the claim that necessary truths are grounded in essences and their connections.25 Although the later Wittgenstein asserted that there was no such thing as phenomenology, the early Wittgenstein is an Austrian thinker who, in sharp contrast to Frege and Russell, is very much a friend of essence. He refers frequently in his Notebooks and in the Tractatus to just the essences which figure prominently in Husserl’s Logical Investigations, in particular to the essences of logic, language and the world. Like Husserl he distinguishes sharply between “essential generality” and “contingent generality” (Tractatus 6.1232). Where Husserl had claimed that “‘empirical laws’ … are non-genuine (unechte) laws”.26 Wittgenstein says that “all law-likeness” belongs to logic (Tractatus 6.3). He also refers to a number of necessary connections, just the connections which, he thinks, the uniquely meaningful language of science cannot express.27

5.4

Essence and Internal Relations

The distinction between internal (inner) and external (outer) relations was almost as important to Husserl and Meinong as to their English contemporaries.28 Unlike the latter, they did not think all relations were external or that they were all internal. Favorite examples of internal relations were the relations between colors, such as the similarity, difference or opposition between colors and the position of orange between red and yellow. In 1903, Meinong (Meinong 1969) sets out an account of the apriori judgments and states of affairs (“objectives”) such as that orange lies between red and yellow and red differs from green which is perhaps still the fullest account of internal relations. In 1906, he (Meinong 1973) makes quite clear how he thinks the modal features of such propositions and of their semantic values differ from their essential features. That red differs from green and “that it cannot be other than different, that this being different is necessary” rests on “the nature or make-up (Beschaffenheit, property) of what is judged”. Our proposition holds “because red is red, green is just green … difference is difference; it hangs on the nature of what is judged and has its ground in this”.29 The philosophies of essence and essential connections of Husserl and Reinach are clearly indebted to a very old philosophical tradition.30 This emerges, for example, in Husserl’s frequent references to, and reliance on, what he calls “genuine, Aristotelian kinds and species” and the trees they form. But the essentialist philosophy of internal relations, they thought, goes back to Locke and Hume.31 For in distinguishing between types of relations, Hume “posed a problem for the phenomenologist”: “Why are certain classes of relations and not others founded in the essence of the terms of the relations (Beziehungspunkte)?”.32 Hume’s account of relations of ideas, Husserl argued, is a forerunner of his own account of relations which are grounded in essence. Hume, argued Husserl and Reinach, had an inadequate account of ideas and relations. But he understood that there is a type of modality which is distinct from and “determined”—Hume’s word—by ideas, namely, possibility and impossibility, even if he did not accept the correct view, which is that essence directly and fully grounds necessities and necessary possibilities. And above all Hume had seen that there are relations of ideas which are non-formal, for example, ideas of colors. He therefore, they think, anticipates their views about material, non-formal (neither formal-logical nor formal-ontological) modality. The Hume of

71

Kevin Mulligan

Husserl and Reinach is not, they point out, that of Kant; nor, we may add on their behalf, is he the Hume of the logical positivists, the logical empiricists and of many subsequent anglophone philosophers, who followed in Kant’s footsteps.33 The philosophy of internal relations developed by Brentano’s heirs is by no means limited to the exploration of the spaces of sensory qualities, to what was sometimes called the geometry of colors and tones. It extends to other qualities, such as values, and to value-space and to the space of orientation (left-right, up-down). Intentionality, verification, involuntary expression, motivation and the reason to F relation were all often understood as internal relations. The phenomenologists sometimes call internal relations essential connections. But they do not think all essential connections are relations. Thus Reinach, in the course of denouncing (like many phenomenologists) “the impoverishment of the apriori”, the fact that even philosophers who accept “the fact of the apriori”, typically “reduce it to a small province of its actual domain”, says this of Hume. The Scottish philosopher “enumerates a few relations of ideas” which are “surely, a priori connections. But why he restricted such connections to relations, and then only to some few of them, is not clear”.34 Hume avoided the later mistake of reducing the apriori to the formal apriori and thus paved the way for the phenomenologists.35 Early and late, Wittgenstein appeals to the distinction between internal and external relations. His views about their relations to modality and essence change considerably. Like Brentano’s heirs, he aims to understand the relations between qualities in spaces, of which he distinguishes many varieties (e.g. in his Philosophical Remarks). And he, too, thinks of intentionality, expression, motivation and verification as internal relations.36

5.5 Rôles The two main theoretical or philosophical rôles of essence are to ground the most basic kind of necessity in the way already sketched and, as we shall see, to be the object of a kind of intuition or understanding. But essence plays many other non-theoretical or nonphilosophical rôles, according to our philosophers.

5.6

Essence Guides

Essence and essential connections guide us. More exactly, the woman in the street and the scientist, in their interactions with the contingent facts making up the objects of the natural world-view and of the scientific world-view, are, Scheler asserts, guided by essence. “In all non-phenomenological experience” essential connections “function … as ‘structures’ and ‘laws of form’ of experiencing in the sense that they are never ‘given’ in such experience but experiencing takes place in accordance (nach) with them”.37 They are, it might be said, transparent. It is only in “the experience of incorrectness, of deviation from a law, which we are not conscious of as a law, that we have a dawning awareness that some [essential – KM] insight was leading and guiding (führte) us.38 He gives two examples of the phenomenon he has in mind, one of which makes clear that the laws which guide us are rules: Whenever we for example infer according to (nach) a law of inference, without inferring “from” (aus) it, obey an aesthetic rule (like the productive artist), without in any way having this rule as a formulated proposition in mind, then essential insights come into play (in Funktion)—without thereby standing explicitly before the mind.39

72

Contemporary (Phenomenological Tradition)

5.7 The Functionalization of Knowledge of Essence The phenomenologist, unlike the woman in the street and the scientist, intuits in “phenomenological experience” what “lies hidden already as a form in natural and scientific experience”,40 a form sometimes called the subjective apriori which, as we have seen, is supposed to guide us. But where does this sort of form come from? It is, Scheler thinks, the result of a process: To see essences as essences is something other than to know contingent facts (to perceive, judge, etc.), in conformity with the guidance of previously intuited essences. In the latter action the essence does not come to separate consciousness. It is only a case of the functioning of essential knowledge—as a selection-process … —without itself being given. It renders supraliminal for the knowledge of contingent existence everything which accords with the intuited essence or is a possible application of essential connections and structures. In this way the originally objective apriori becomes the subjective apriori; …41 One of the “least understood” properties of knowledge of essence, a property Scheler does not really succeed in describing, is that it “becomes functional as a law governing the mere ‘employment’ of the intellect with regard to contingent facts; under its guidance, the intellect grasps, analyses, intuits and judges the contingent factual world as ‘determined’ in ‘accordance’ with essential connections”.42 Pace Kant, the subjective apriori is as varied as it is historically variable. Thanks to the “functionalization” of insights into essence—the fact that they take on the status of rules which are followed—there is “a sort of true growth of the human mind both in an individual life and in the course of history”, growth which is mediated by tradition.43 According to the later Wittgenstein, too, we are guided by rules. According to one interpretation, Wittgenstein came to think of the sentences which purport to express essential, philosophical truths as rules for the use of words: essence, he says, is specified by grammar. While there are many phenomena described by Wittgenstein which the phenomenologists would classify as examples of the subjective apriori, Wittgenstein roundly rejects what they called the objective apriori, essential, essentially necessary connections. “We have”, he famously declares, “a system of colors as we have a system of numbers. Do the systems lie in our nature or in the nature of numbers or colors?” His answer: “Not in the nature of numbers or colors”.44

5.8

Essence Intrudes

Another role of essence is to get in the way. Like a good logical positivist or empiricist in Vienna, Prague or Berlin, Scheler asserts in 1926 that science must eliminate all appeals to essence. Essence and essential questions have a bad habit of intruding into science and mathematics. “Positive science” increasingly “excludes metaphysical questions and above all questions of essence from its domain”. Questions about sense or meaning which cannot be settled by observation and measurement must be “radically eliminated” from “the domain of positive science”. This elimination should be insisted on much more rigorously than is currently the case “in mathematics and theoretical physics”.45 Questions “which cannot be decided by possible observation and measurement together with mathematical inference” do not belong to positive science.46 Why? The propositions made true by essential connections have no observable consequences. That is why pragmatists are essence-blind. It is

73

Kevin Mulligan

the specific task of all positive science to eliminate all questions of essence … in order to retain … all those questions for the settling of which an activity can be specified which decides … by letting the expected reaction by the world (to our action) decide.47 Unlike the pragmatists, the logical positivists and the logical empiricists but like Husserl and other early phenomenologists, Scheler believes firmly in the possibility of an ontology or “rational science” and metaphysics of nature. If we pay attention to the things a scientist finds evident or obvious, we shall find that these are often the essential connections belonging to such an ontology of nature.48 The phenomenologists, it should be noted, do not use the terms “ontology” and “metaphysics” in the ways which are now familiar. By “ontology” they refer to disciplines which contain only non-contingent truths, formal (for example, the formal theory of wholes and parts) and non-formal (for example, the theory of color space or valuespace or of social objects). Metaphysics they take to be the philosophical discipline which combines truths of ontology with contingent truths. Their sharp distinction between ontology and metaphysics, on the one hand, and positive science, on the other hand, is rejected by those who think of theories as containing claims which are more or less contingent. The fundamental reason why Husserl or Scheler cannot accept such a view is that, as we have seen, they take the locus of necessity to be states of affairs rather than truth-bearers.

5.9

Knowledge of Essence and Bird’s Eye Views

Are essences and their connections knowable? If so, how does such knowledge come about? What are the roles of such knowledge? What sort of knowledge is involved?49

5.10

The Intuition of General Essences (Ideas) and Conceptual Ladders

Before we look at the nature of the intuition of essence it will be useful to bear in mind one putative rôle of this type of intentionality. In an early criticism of Husserl’s Logical Investigations, Wundt reproached its author for often failing to say positively what a judgment, a wish, a meaning is. Instead, Husserl devotes himself to explaining at great length what a judgment, a wish, a meaning is not and arrives at an “empty tautology”, e.g. “a judgment is – just a judgment”. In a reply to Wundt, Scheler concedes that he has correctly described the course of many phenomenological discussions but objects that he has completely failed to understand their point. A phenomenological book is not a book which simply “wants to communicate observations and describe what has been observed or to prove something inductively or deductively”.50 The two types of books require very different attitudes in their readers. The sole aim of a phenomenological discussion is to “bring the reader … to the intuition of what, according to its essence, can only … be intuited”. In bringing the reader to this intuition, “all the propositions in the book, all entailments, all preliminary definitions, all chains of inference and proofs have, taken together, only the function of a pointer which indicates what is to be intuited”51. The phenomenologist aims to circle round her subject-matter until she arrives at the point at which a “tautology” indicates to the reader: “Look, then you will see it! This is the sense of what Wundt takes to be a mere ‘tautology’”.52 An example of the sort of procedure Scheler has in mind is the way Husserl circles in on the phenomenon of suspension of judgment (Urteilsenthaltung) in 1913. To suspend judgment, he says, is not to doubt; it is not to “merely think” that, for example, water-nymphs are performing a roundelay”, for in this case there is no elimination of an active belief; it is not to

74

Contemporary (Phenomenological Tradition)

think in the sense of assuming or supposing.53 Husserl here adds to his three negative claims only the theses that suspending judgment is subject to the will and that to suspend judgment is to put it out of action. The first thesis does not distinguish suspending judgment from assuming or supposing, which he also takes to be directly subject to the will. The second is perhaps only a metaphor.54 The function of the “numerous negations” in the writings of the phenomenologists, Scheler says, “is to circle in on a phenomenon from all sides through successive elimination of the variable complex wholes into which the phenomenon enters … until nothing remains except—the phenomenon itself”; it is the phenomenon’s “non-definability in all possible attempts to define it which displays it as a genuine phenomenon”.55 Even images, for example metaphors, by mutually checking each other, may help to bring something to intuition.56 In a book belonging to positive science, on the other hand, what the author has in mind is never itself directly given “but always only as bound through certain relations to other objects”.57 In phenomenological philosophy, “one “talks a little less, keeps silent more and sees more” than in other types of philosophy, one sees, too, “what about the world is perhaps the nomore talkable about”. This is, he says, the pathos peculiar to such philosophy.58 Intuition of essences and their connections, then, comes after and crowns a great deal of conceptual effort. But these connections also verify and falsify predications, predications of essence. Such predications, like predications of contingent matters of fact, must, the phenomenologists like to say, be such that they can be cashed out (a metaphor they seem to have taken over from Schopenhauer and James). The two basic roles for the intuition of essence, then, are to crown conceptual effort and to verify predications of essence. The rhetoric of non-visual seeing, looking and showing and their pathos in Scheler’s 1913–4 description of what goes on in a phenomenological book have one or two similarities with Wittgenstein’s description of what is going on in his book and its pathos: in each case, the function of the conceptual apparatus has the primary function of showing something, it is a ladder leading to a revelation.59 But Scheler, unlike Wittgenstein, takes the sentences making up his ladder to be meaningful.

5.11

The Essence of Intuition of Essence

Husserl was convinced that there is a deep similarity between the ordinary perception of temporal entities, such as things and tropes and the intuition of essences and their connections. This is no mere “external analogy but a radical community”.60 Indeed he thinks that visual perception of a redness or shape trope or visual imagining of these or inner perception or quasi-perception of a trope of judging allow him to ascend to intuition of their kinds or essences. Scheler, on the other hand, roundly denies that the intuition of essence (or of a connection between essences) is anything like sensory perception. One reason he gives for this claim is that sensory perception is of what resists or may resist but in the ideal sphere there is no resistance. Resistance and its possibility are what justify the assumption that what we perceive is mind-independent. But the assumption that essence continues to obtain when not the object of thought makes no sense at all and would not be justified even if it did make sense. “Husserl’s analogy” between essences and temporal items, which makes of essence “a rigid block of being” is “fundamentally wrong”. The analogy which has misled Husserl, he seems to think, is what Scheler elsewhere calls one of “the covert pictures (image, Bild)” which lead philosophers into error. Essence is indeed identifiable but the identifiability of something is no clue to its essence. Husserl is however right to say that essences are not, like mathematical

75

Kevin Mulligan

entities, any sort of ficta.61 Scheler, as we have noted, thinks essence is not at bottom either general or particular but neutral. Intuition in the basic sense, formless intuition (that is, presumably, intuition which is not of something as a particular or as general) is the sort of act of which neutral essence is the correlate. Scheler’s own final, positive account of essences and their intuition, according to which—contra Husserl—general essences are always exemplified and intuition is productive is bound up with his panentheistic metaphysics.

5.12

Stigmatic Intuition vs Bird’s Eye Views

Hartmann, a philosopher of essence who cannot be suspected of any sort of pan/en/theism, also criticizes Husserl’s account of intuition. In 1921 and later, he distinguishes between “stigmatic”, punctual intuition and “conspective intuition”.62 Husserl and other phenomenologists have favored the former and overlooked the latter and thus failed to grasp their complicated interrelations. In “conspective intuition”, in “surveying conspectively” (konspektiv übersehen) many different connections are grasped. It is an overview (Zusammenschau); it is to see through (durchschauen) and survey (überschauen). The grasp of a theorem belongs to stigmatic intuition, that of a proof to conspective intuition.63 Wittgenstein was to agree that “surveyability (Übersichtlichkeit) belongs to a proof”,64 but differs from Hartmann and the phenomenologists in thinking that a proof establishes rather than discovers connections. More generally, if Hartmann’s conspective intuition resembles in some respects Wittgenstein’s overview or bird’s eye view (übersichtliche Darstellung), which allows us to “see the connections”, Wittgenstein thinks that such seeing reveals an order and not, like the phenomenologists, the order of a domain.65 The roles of what both Hartmann and Wittgenstein call a survey (Überschau) are not the same. Husserl’s early account of the intuition of essence in his Logical Investigations developed in different directions. One is his already mentioned account of the suspension of judgment. Another is his method of variation.66 The accounts of intuition set out by Husserl, Scheler and Hartmann can be considered from the point of view of the distinction between knowledge and its sources (perception, memory, revelation, demonstration, etc.). Just as intuition, according to Scheler, is based on conceptual work, so, too, what Hartmann calls stigmatic intuition may be thought to be based on conspective intuition. Similarly, variation and the suspension of judgment, it might be thought, are the sources of what Husserl thinks of as the intuition of essential connections.

5.13

The Intuition of Individual Essences

The early phenomenologists have less to say about individual essences and our knowledge of these than they do about general essences and their intuition. An individual essence to which Scheler attaches a great deal of importance is the individual selfness (Ichheit) of a self. Like many heirs of Brentano (and unlike Husserl after his announcement of his discovery of his transcendental ego in 1913), Scheler thinks of selves or souls as unities, in particular as unities of psychological functions such as hearing, desiring and emoting, rather than as any sort of simple, underlying substratum (or psychological bare particular). And each such unity, he claims, contains an individual selfness, itself distinct from the general essence of which individual selfnesses are instances.67 Scheler’s argument for this claim seems to be that inner perception is essentially of what is psychological and an individual selfness is epistemically prior to psychological functions.68 Another individual essence, Scheler thinks, is that of a person, where a person is not any sort of self or soul but, like a self or soul, a unity, and not

76

Contemporary (Phenomenological Tradition)

any sort of simple substratum (or mental bare particular). A person is a unity of mental acts as a soul is a unity of psychological functions.69 The individual essence of a person, now, is an individual value essence (Wertwesen) and comprises the values determining what is intrinsically valuable for the person, for example, a person’s vocation. Personal identity, then, is not to be understood in value-free terms, for example, in terms of theoretical or intellectual mental acts, such as memory. The individual value essence of a person, it is claimed, is revealed through a particular type of understanding, the understanding of the lover, the understanding motivated by love.70 Scheler’s two examples of individual essences are examples of wholly material essences and contrast, in this respect, with the now popular individual essences ascribed by instances of “x is self-identical”. Indeed the phenomenologists sometimes claim that identity has a ground or explanation, namely essence.71 To this it might be objected that the individual essence of an object makes its self-identity necessary but does not make it self-identical.

5.14 Essence vs Family Resemblances Although some progress has been made in understanding and evaluating what Husserl had to say about essence and modality, knowledge of essence and related topics such as dependence, grounding and mereology,72 the same cannot be said of what other phenomenologists had to say about these topics. It also seems to be the case that the applications by the phenomenologists of essentialist views in different areas of philosophy have rarely been critically examined. Their philosophy of mind, their phenomenology of mental acts and their objects, invariably takes an essentialist form. It is somewhat surprising that the essentialist status of such claims has come in for so little attention since there are obvious and wellknown alternatives. One such is the view made popular by the later Wittgenstein’s claim that many mental predicates are family resemblance terms and the related claim that the processes, states and events of e.g. understanding have nothing substantial in common but are connected by numerous partial similarities.73 How might a phenomenologist reply to objections of this type? Almost certainly by appealing to a type of essential connection to which they were strongly attached, between different types of mental acts and their “correlates”—between judgment (belief) and facts (obtaining states of affairs) or truth, between conjecture and probability, between emotion and values, between certainty and actuality, between intentions and oughtness and between love (and hate) and value-height or individual values. All judgings, the reply might run, have in common their relation to facts or truth, and so on in each case. The relation of correlation is not to be understood in the same way in each case nor is there unanimity about what the act-correlate tie amounts to. (Do emotions reveal value (Husserl, Meinong, Marty) or are emotions a response to values which are the objects of some sort of prior affective grasp (Reinach, Scheler, von Hildebrand)?). Of course, many philosophers, including Wittgenstein, have thought that some or all of the correlates mentioned are ficta, mere shadows, like the entities often held by the phenomenologists and others to be the bearers of the correlates mentioned—abstract or ideal propositions or states of affairs. If that is the case, one may wonder whether an essentialist philosophy of mind has any future.74 During the 20th century many essence-lite philosophies or descriptions were to see the light of day. On these views, the truth of some truth-bearers is typically explained by or traced back to linguistic facts, normative or not. The truths or facts are those the phenomenologists distinguished as analytic and necessary or synthetic and necessary. Their truth is accounted for in terms of one or another so called linguistic theory of the apriori. Or these truth-bearers

77

Kevin Mulligan

are said to be true in virtue of meanings. Wittgenstein’s assertion that essence is specified by grammar is perhaps the most influential of all these essence-lite accounts.75 It is surprising that the different essence-lite views have still to be confronted with the non-lite views of the essence-intoxicated early phenomenologists.

5.15

Related Topics

Marko Malink’s chapter, Ancient (Chapter 1). Anat Schechtman’s chapter Modern (Chapter 3) Robert Michels’ chapter, Contemporary (Analytic Tradition) (Chapter 6). Alessandro Torza’s chapter, Modal Conceptions of Essence (Chapter 7) Fabrice Correia’s chapter, Non-Modal Conceptions of Essence (Chapter 8). Marco Marabello’s chapter, Essences of Individuals (Chapter 9) Tuomas Tahko’s chapter, Natural Kind Essentialism (Chapter 10) Arata Hamawaki’s chapter, Wittgenstein (Chapter 33)

Notes 1 The description is applied by Ortega y Gasset to his friend. It is a modification of a famous description of Spinoza as “God-intoxicated” ( Ortega y Gasset 1966). 2 Reinach 1989b I: 542–3. 3 Scheler 1957: 448. 4 The best introduction to the way the early phenomenologists understood essence is Reinach’s article “Über Phänomenologie” ( Reinach 1989b, English tr. Reinach 1969). The most detailed and lucid accounts of essence by phenomenologists are due to Husserl, Reinach, Scheler, Ingarden 1925, Hering 1921 and Spiegelberg 1930. On Husserl’s essentialism, cf. Segelberg 1999, Hochberg 1999, Smith 1989, Ingarden 1996, Mulligan 2004, Spinelli 2021. On Ingarden on essence, cf. Haefliger 1994. On essentialism in early phenomenology, cf Seifert 1996, Pöll 1936. Hartmann, a philosopher strongly influenced by the essentialisms of the early phenomenologists, is also the author of detailed accounts of essence. 5 Husserl 1950: §2 12–3. 6 Husserl 1950 §5. 7 See Tuomas Tahko’s chapter, Natural Kind Essentialism () and Marco Marabello’s chapter, Essences of Individuals () 8 Husserl 1950 §3. Ingarden (1996 §26) suggests that Husserl does not allow for individual essences. 9 Williams 1953. 10 Scheler 1966: 68, cf Scheler 1973: 48–9. 11 Stein 1986: 98. 12 James 1950, I, ch. XII, 473. 13 On the view that Aristotelian forms are neither particular nor universal, cf. Code 1984, Koslicki 2018. In his chapter on the traditional lack of distinction between what Frege called “Fallen-Unter” and “Unterordnung”, Angelelli describes ousia as “a sort of kaleidoscopic entity, neither universal nor singular, or which is both” ( Angelelli 1967: 119). 14 Mulligan (2018) argues that if the infinitives employed in philosophical claims such as “To be a person is to be free” have a semantic value, the best candidate for this role is neutral essence. D. P. Henry ( Henry 1984: 140–1,173, 172, 215, 234; 1986: 182) endorses a view along these lines and attributes it to medieval philosophers (without giving any textual evidence for this claim). Henry then argues that infinitives belong to the semantic category of predicates. But a predicate is a finite expression. In their formal account of the copula in infinitival specifications of essence, which they call generalized identity, Correia and Skiles also treat infinitives as predicates ( Correia & Skiles 2019). 15 Husserl 1950: 358. 16 Reinach 1989d; Mulligan 2004.

78

Contemporary (Phenomenological Tradition) 17 “So ist also der Wesenszusammenhang dasjenige, worauf die Notwendigkeit sich aufbaut” ( Reinach 1989c I: 70; Reinach 1976: 164–5). What Reinach here calls an essential connection is what is called below (§3) an internal relation. On the relation between essence and necessity, cf. Hildebrand 1960: 64–9. See Alessandro Torza’s chapter, Modal Conceptions of Essence ( Chapter 7) and Fabrice Correia’s chapter, Non-Modal Conceptions of Essence ( Chapter 8). 18 The view that the bearers of necessity are states of affairs was to be endorsed by a philosopher familiar with the writings of the phenomenologists, D. C. Williams: “necessity can qualify nothing short of the states of affairs which make statements or judgments necessarily true” ( Williams 1963: 602), cf. Forbes 1989: 130–1. Williams goes on to say that it is “plausible that the pure essences generally are the principal if not only vehicles of metaphysical necessity” ( Williams 1963: 625). On Reinach on the bearers of modality, see Künne 1987 185–7. On Meinong’s early objections to the reduction of necessity to laws and later endorsement of the view that the bearers of necessity are states of affairs (objectives), see Grossman 1974: 31–5. 19 Scheler 1966:375; cf Hartmann 1966: 344. 20 On essential or real definitions, cf Husserl 2012 §24. In his Logical Investigations, Husserl does not call all general states of affairs which are grounded in essence necessities. He reserves the terms “analytic necessities” and “synthetic necessities” for particularizations of laws which are so grounded (Husserl LI III §§12–3, Husserl 1988: 234). 21 Husserl 1975 §72; Husserl 1973. 22 Reinach 1989 II: 354. For the history and criticisms of this sort of claim, cf Barth 1974, Angelelli 1967. 23 Scheler 1995: 106; Mach 1980: 455. Another Austrian philosopher, Popper, argues somewhat later that scientific laws are prohibitions. 24 Conditional vs non-conditional essence ( Reinach 1989a I: 250–1, 186), exact vs vague or morphological essences ( Husserl 1975 LU III §9, 1984: 248–51; Becker 1923, Scheler 1960: 356–8; propositions about essence vs eidetic propositions ( Husserl 1950 §5); essences vs proper essences ( Husserl 1974 70, 455); essences vs concepts ( Husserl 1948: 240–1); essence vs sense ( Stein 1980: 180ff..), the essence of essences vs the essences of non-essences ( Husserl 2012: 34). For Husserl’s account of non-theoretical necessity, cf. Husserl 1988 §11; on this account and its relation to Fine’s three types of necessity, cf. Mulligan 2022. If necessity is rooted not in general essence but in neutral essence, then what is necessary need no longer be general. In the case of normative necessity, Scheler draws just this conclusion and allows for oughts which are essentially bound to particular individuals, times and places. In the case of theoretical necessity, cf Ingarden (1964: 87ff.) on the essentiality of origin. 25 For examples and discussion, see my contribution to The Routledge Handbook of Metaphysical Grounding ( Mulligan 2020). 26 Husserl 1975 §23 82. 27 28 Cf. Mulligan 2024. See Arata Hamawaki’s chapter on Wittgenstein. 28 See Robert Michels’ chapter, Contemporary (Analytic Tradition). 29 Meinong 1973: 9. 30 See Marko Malink’s chapter, Ancient. 31 See Anat Schechtman’s chapter Modern. 32 Husserl 1948: 471. 33 See Reinach 1989b, Husserl 2012: 155–67. 34 Reinach 1969: 215 (translation modified), 1989 I: 546. 35 Internal relations relate objects. If infinitival expressions are not names and their semantic values are not objects, essential connections are not relations. Internal relations come in many kinds—identity, difference, distance, motivation, etc. The copula in infinitival specifications of essence does not display this multiplicity. 36 On Wittgenstein’s uses of the distinction between internal and external relations, cf. Mácha 2015, Mulligan 1991, 2012. 37 Scheler 1966: 71. 38 Scheler 1954: 198. 39 Scheler 1954: 198. Scheler, Husserl, Meinong and other heirs of Brentano frequently appeal to the distinction between inferring according to and inferring from. They are more likely to have learned of it from Bolzano than from Charles Dodgson. 40 Scheler 1966:66.

79

Kevin Mulligan 41 42 43 44 45 46 47 48

49 50 51 52 53 54

55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72

Scheler 1954: 207–8. Scheler 1954: 198. Scheler 1954:198. On the subjective apriori, cf. Pap 1944 Wittgenstein Zettel §357. Scheler 1960: 234. Scheler 1960: 209. Scheler 1960: 220. Scheler 1960: 220. Cf. “The growth of objective knowledge consists [ … ] in a constant interplay between an a priori element—knowledge of essence—and an a posteriori element, the empirical testing of existential hypotheses whose possibility has already been anticipated a priori ( Lowe 2013: 156). See Antonella Mallozzi’s chapter, Epistemology of Essence ( Chapter 14). Scheler 1957: 391. Scheler 1957: 391. Scheler 1957: 391. Husserl 1950: 65. Husserl 1950: 65. Husserl takes his epochē or method of suspending judgment to have been anticipated by Descartes’ method of doubt. His first negation was perhaps suggested to him by Bolzano’s observation that Descartes was wrong to think that doubt is immediately subject to the will. His second negation seems to be a reference to what Broad called entertaining a proposition. Husserl says that the image or picture of bracketing is best applied to the content of a judging when this act is put out of action. Whatever suspending a judgment is supposed to be Husserl clearly thinks of it as being fairly easy. The diametrically opposed view that it is very difficult is put forward by Scheler ( Scheler 1995: 254ff..). For the view that suspending judgments, whether non-contingent as in Husserl and Scheler or contingent, is difficult, cf Meylan forthcoming. Scheler 1957: 392. Pfänder deliberately uses a large number of metaphors in his account of the nature of love ( Pfänder 1916). Later phenomenologists, such as Heidegger and Merleau-Ponty, sometimes seem to think that phenomenology must be wholly metaphorical or figurative. Hertz’s Preface to his Mechanics, a text admired by Scheler and Wittgenstein, employs just such a relational conception of essence in science, a conception taken up by Carnap in the Aufbau. On Hertz on relational essence, cf. Hüttemann 2009: 145–68. Scheler 1957: 393. In the course of endorsing and developing Reinach’s account of the essence of movement towards and through a point, Ajdukiewiz comments: “Being a phenomenologist [Reinach] does not define any of the kinds of contact but is satisfied with stirring our intuition ( Ajdukiewicz 1978: 199). Husserl 1950: 14. Scheler 1957: 256. On the differences between sensory perception and intuition, cf. Ingarden 1996 §26. Hartmann 1965 chs 65, 67–9, 72, 74; Hartmann discusses in detail the relations between understanding and conspective intuition ( Hartmann 1966 ch. 47). On the differences between understanding and intuition of essence, see Lowe 2013, 2014. Hartmann 1965: 521, 541. Wittgenstein Remarks on the Foundations of Mathematics, 1, §154. Wittgenstein Philosophical Investigations, §132. Husserl 2012, 1948. On Husserl’s method of variation, cf Spinelli 2021, Michels 2020. Scheler 1966: 378–9, 94–5, 373–4, 389, 413. Scheler 1966: 408. See Annina Loets chapter, Persons ( Chapter 22). Scheler 1966: 481. McTaggart’s account of love resembles in some respects that of his German contemporary. Thus McTaggart distinguishes between the value “in” a person or self and the value of a person or self (McTaggart The Nature of Existence, II §788). Pfänder 1921:334, cf. 178, 180, 234, 485–7; for the view that identity is not grounded in essence cf. Reinach 1989a II: 566. Analyses of Husserl’s essentialist accounts of mereology, dependence and foundation in his third Logical Investigation ( Husserl 1984, Husserl 1973) by Simons, Smith, Johansson, Fine, Correia and Casari should be mentioned first in this connection.

80

Contemporary (Phenomenological Tradition) 73 The relation between essentialism and the account of focal meaning by Aristotle and others (as well as, say, the theory of prototypes) has, as far as I can see, never been thoroughly tested with reference to mental predicates. On Wittgenstein on family resemblances, see Arata Hamawaki’s chapter (). 74 Did Wittgenstein take his descriptions of family resemblances to provide objections to essentialist claims? Cf. Specht 1969, Hallett 1991, Mulligan 2012, Pouivet 2014. Whatever Wittgenstein may have thought, the precise relations between essentialism and family resemblances await exploration, as do their relations to focal meanings. On essentialism and focal meanings, cf Finnis 2011 9ff.. On Husserl on prototypes, cf. Taieb 2021. 75 The first 20th century essence-lite view of non-contingent truths and connections is to be found in Brentano and in the writings of those of his heirs who tended towards nominalism. Unlike its better known successsors, it is entirely non-linguistic.

References Where appropriate, original dates of publication are given in brackets. Ajdukiewicz, K. 1978 “Change and Contradiction”, The Scientific Word Perspective, Dordrecht: Springer, 192–208. Angelelli, I. 1967 Studies on Gottlob Frege and Traditional Philosophy, Springer Netherlands. Barth, E. M. 1974 The Logic of the Articles. A Contribution to the Study of Conceptual Structures, Dordrecht & Boston: Reidel. Becker, O. 1923 “Beiträge zur phänomenologischen Begründung der Geometrie und ihre physikalischen Anwendungen”, Jahrbuch für Philosophie und phänomenologische Forschung, IV, 493–560. Correia, F. & Skiles, A. 2019 “Grounding, Essence, and Identity”, Philosophy and Phenomenological Research, 98(3), 642–670. Finnis, J. 2011 (1980) Natural Law and Natural Rights, Oxford: Clarendon Press. Forbes, G. 1989, Languages of Possibility, Oxford: Basil Blackwell. Grossman, R. 1974 Meinong, London, Routledge & Kegan Paul. Haefliger, G. 1994 Über Existenz. Die Ontologie Roman Ingardens, Dordrecht: Springer. Hallett, G. 1991 Essentialism: A Wittgensteinian Critique, New York: Suny Press. Hartmann, N. 1965 (1921) Grundzüge einer Metaphysik der Erkenntnis, Berlin: de Gruyter. Hartmann, N. 1966(1938)Möglichkeit und Wirklichkeit, Berlin: de Gruyter. Henry, D. P. 1984 That Most Subtle Question (Quaestio subtilissima). The Metaphysical Bearing of Medieval and Contemporary Linguistic Disciplines, Manchester: Manchester University Press. Henry, D. P. 1986 “Universals and Particulars”, History and Philosophy of Logic, 7(2), 177–183. Hering, J. 1921 “Bemerkungen über das Wesen, die Wesenheit und die Idee. Edmund Husserl zum 60. Geburtstag gewidmet”, Jahrbuch für Philosophie und phänomenologische Forschung, IV, 495–543. Hildebrand, D. v. 1960 (1948) What is Philosophy?, Milwaukee: The Bruce Publishing Company. Hochberg, H. 1999 Complexes and Consciousness, Stockholm: Thales. Husserl, E. 1948 Erfahrung und Urteil, Hamburg: Claassen Verlag. Husserl, E. 1950 Ideen zu einer reinen Phänomenologie und phänomenologischen Philosophie I (Husserliana III), ed. by W. Biemel, The Hague: Nijhoff (abbr. as Ideas). Husserl, E. 1973 Logical Investigations, translation of Husserl 1975, 1984, transl. by J. Findlay, London: Routledge and Kegan Paul, (“LI 1 (II, III”) refers to the first (second, third) Investigation) Husserl, E. 1975 Logische Untersuchungen (1900–01, 1913), ed. by E. Holstein, The Hague: Martinus Nijhoff. Husserl, E. 1974 (1929) Husserl, E Formale und transzendentale Logik: Versuch einer Kritik der logischen Vernunft, (Husserliana XVII), Berlin: de Gruyter. Husserl, E. 1984 Logische Untersuchungen. Zweiter Band – I. Teil, Zweiter Band - II. Teil, (Husserliana XIX), ed. by U. Panzer. Husserl, E. 1988 Vorlesungen über Ethik und Wertlehre 1908–1914, (Husserliana XXVIII), ed. by U. Melle, Dordrecht: Kluwer. Husserl, E. 2012 Zur Lehre vom Wesen und zur Methode der eidetischen Variation. Texte aus dem Nachlass (1891–1935), (Husserliana XXXXI), Berlin: de Gruyter. Hüttemann, A. 2009 “Pluralism and the Hypothetical in Heinrich Hertz’s Philosophy of Science”, The Significance of the Hypothetical in the Natural Sciences, ed. by Heidelberger & Schiemann, 145–168.

81

Kevin Mulligan Ingarden, R. 1925 “Essentiale Fragen. Ein Beitrag zu dem Wesensproblem”, Jahrbuch für Philosophie und phänomenologische Forschung VII, 125–304. Ingarden, R. 1964 Der Streit um die Existenz der Welt. Existentialontologie, Vol. 1, Tübingen: Max Niemeyer Verlag. Ingarden, R. 1996 Zur Grundlegung der Erkenntnistheorie, 1. Teil, Das Werk, Tübingen: Niemeyer. James, W. 1950 (1890) The Principles of Psychology, Vol. I, New York: Dover. Künne, W. 1987 “The Intentionality of Thinking: The Difference between State of Affairs and Propositional Matter”, Speech Act and Sachverhalt: Reinach and the Foundations of Realist Phenomenology, ed. K. Mulligan, Dordrecht: Nijhoff, 175–187. Lowe, E. J. 2013 Forms of Thought, Cambridge: Cambridge University Press. Lowe, E. J. 2014 “Grasp of Essences versus Intuitions: An Unequal Contest”, Intuitions, ed. T. Booth & D. Rowbottom, Oxford: Oxford University Press, 256–268. Mach, E. 1980 (1906) Erkenntnis und Irrtum. Skizzen zur Psychologie der Forschung, Darmstadt: Wissenschaftliche Buchgesellschaft. Mácha, J. 2015 Wittgenstein on Internal and External Relations: Tracing all the Connections, London: Bloomsbury. Meinong, A. v. 1969 (1903) “Bemerkungen über den Farbenkörper und das Mischungsgesetz”, Gesamtausgabe, Vol. I, Graz. Meinong, A. v. 1973 (1906) “Ueber die Erfahrungsgrundlagen unseres Wissens”, Gesamtausgabe, Vol. V, Graz. Meylan, A. “On Suspending Judgment”, forthcoming Michels, R. 2020 “Husserlian Eidetic Variation and Objectual Understanding as a Basis for an Epistemology of Essence”, Logos and Episteme II (3), 333–355. Mulligan, K. 1991 “Colours, Corners and Complexity: Meinong and Wittgenstein on Internal Relations”, Existence and Explanation: Essays in Honor of Karel Lambert, eds. B. C. van Fraassen, B. Skyrms, & W. Spohn, Dordrecht: Kluwer, 77–101. Mulligan, K. 2004 “Essence and Modality. The Quintessence of Husserl’s Theory”, Semantik und Ontologie. Beiträge zur philosophischen Forschung, eds. M. Siebel & M. Textor, Frankfurt: Ontos Verlag, 387–418. Mulligan, K. 2012 Wittgenstein et la philosophie austro-allemande, Paris: Vrin. Mulligan, K. 2018 “De l’infinitif, de l’essence neutre et de leurs rapports”, Sujet libre, (Festschrift de Libera), eds. J.-B. Brenet & L. Cesalli, Paris: Vrin, 227–232. Mulligan, K. 2020 “Austro-German Phenomenologists”, The Routledge Handbook of Metaphysical Grounding, ed. M. Raven, 90–101. Mulligan, K. 2022 “Logical Norms, Logical Truths & (Normative) Grounding”, Bolzano and Grounding, eds. B. Schnieder & St. Roski, Oxford: Oxford University Press, 244–275. Mulligan, K. 2024 “Essence, Modality & Generality – Wittgenstein & Husserl”, Wittgenstein’s PreTractatus Writings. Interpretation and Reappraisal, eds. M. Marion, & J. Plourde, Palgrave, forthcoming. Ortega y Gasset, J. 1966 “Max Scheler. Un embriagado de esencias (1874–1928)”, Obras Completas, Vol. IV (1929–1933), Madrid: Revista de Occidente, 507–511 Pap, A. 1944 “The Different Kinds of A Priori, Philosophical Review, (5), 465–484. Pfänder, A. 1916 “Zur Psychologie der Gesinnungen”, in Jahrbuch für Philosophie und phänomenologische Forschung, I & III. Halle: Max Niemeyer. Pfänder, A. 1921 “Logik”, Jahrbuch für Philosophie und phänomenologische Forschung, IV. Halle: Max Niemeyer. Pöll, W. 1936 Wesen und Wesenserkenntnis, Munich: Verlag von Ernst Reinhardt. Pouivet, R. 2014 “Wittgenstein’s Essentialism”. Liber Amicorum Pascal Engel, eds. J. Dutant, D. Fassio, & A. Meylan, Université de Genève, Faculté de Lettres, Département de Philosophie, 449–464. Reinach, A. 1969 “Concerning Phenomenology”, tr. D. Willard, The Personalist, 50, 194–221. Reinach, A. 1976 “Kant’s Interpretation of Hume’s Problem”, trans. of Reinach 1989c by J. N. Mohanty, The Southwestern Journal of Philosophy, 7. Reinach, A. 1989a Sämtliche Werke: Textkritische Ausgabe in 2 Bänden, eds. K. Schuhmann & B. Smith, Munich: Philosophia Verlag. Reinach, A. 1989b (1921) “Was ist Phänomenologie?”, Reinach 1989 1, 531–550 Reinach, A. 1989c (1911) “Kants Auffassung des Humeschen Problems”, Reinach 1989 I, 67–93. Reinach, A. 1989d (1911) “Zur Theorie des negativen Urteils”, Reinach 1989, I, 95–140.

82

Contemporary (Phenomenological Tradition) Scheler, M. 1954 Vom Ewigen im Menschen, Gesammelte Werke V, Bern: Francke Verlag. Scheler, M. 1957 (1933) “Lehre von den drei Tatsachen”, Schriften aus dem Nachlass, I, Zur Ethik und Erkenntnislehre, Gesammelte Werke, X, Bern: Francke Verlag, 431–502. English tr. 1973 “The Theory of the Three Facts”, Selected Philosophical Essays, tr. D. R. Lachterman Evanston: Northwestern Press, 202–287. Scheler, M. 1960 (1926) “Erkenntnis und Arbeit. Eine Studie über Wert und Grenze des pragmatischen Motivs in der Erkenntnis der Welt”, Die Wissensformen und die Gesellschaft, Gesammelte Werke, VIII, Bern: Francke, 191–382. Scheler, M. 1966 (1913–1916) Der Formalismus in der Ethik und die materiale Wertethik. Neuer Versuch der Grundlegung eines ethischen Personalismus, Gesammelte Werke, II, Berne, Francke. Scheler, M. 1973 Formalism in Ethics and Non-Formal Ethics of Values. A New Attempt toward the Foundation of an Ethical Personalism, English tr. of Scheler 1966, Evanston: Northwestern University Press. Scheler, M. 1995 Späte Schriften, Gesammelte Werke, IX, Bern: Francke. Segelberg, I. 1999 Three Essays in Phenomenology and Ontology, Stockholm: Thales. Seifert, J. 1996 Sein und Wesen, Heidelberg: Universitätsverlag C. Winter. Smith, B. 1989 “Logic and Formal Ontology”, Husserl’s Phenomenology: A Textbook, eds. J. N. Mohanty & W. McKenna, Lanham: University Press of America, 29–67. Specht, K. 1969 The Foundations of Wittgenstein’s Later Philosophy, Manchester: Manchester University Press. Spiegelberg, H. 1930 “Ueber das Wesen der Idee. Eine ontologische Untersuchung”, Jahrbuch für Philosophie und phänomenologische Forschung, 1–238. Spinelli, N. 2021 “Husserlian Essentialism”, Husserl Studies 37(211), 147–168. Stein, E. 1980 (1917) Zum Problem der Einfühlung, Munich: Kaffke. Stein, E. 1986 Endliches und ewiges Sein, Freiburg: Herder. Taieb, H. 2021 “The Structure and Extension of (Proto)Type Concepts: Husserl’s Correlationist Approach”, History and Philosophy of Logic 43(2), 129–142. Williams, D. C. 1953 “On the Elements of Being: I”, Review of Metaphysics 7(1), 3–18. Williams, D. C. 1963 “Necessary Facts”, Review of Metaphysics 16(4), 601–626. Thanks to Anthony Fisher, Richard Glauser, Marko Malink, Robert Michels, Tuomas Tahko, Julian Toader and the Editors for their help.

83

6 CONTEMPORARY (ANALYTIC TRADITION) Robert Michels

6.1 Introduction Due to the strong influence of logical empiricism, the notion of essence seemed like a relic of the past to many analytic philosophers working in the first half of the 20th century. Yet, at the beginning of the 21st century, analytic philosophers considered the notion worthy of serious discussion and even relied on it in philosophical explanations and theories. This chapter gives a roughly chronological overview of the history of essence in 20th century philosophy in the analytic tradition, focusing on a number of important developments leading from the logical positivists’ opposition to essence to its current renaissance in analytic metaphysics and beyond. Not every use of the word “essence”, or sometimes also “nature” or “identity”,1 evokes the metaphysical notion of essence. Likewise, not every essentialist view is described as such. It is indeed difficult, perhaps impossible to pinpoint a single notion of essence present in all discussions of essence and essentialism in 20th century analytic philosophy. What one can do instead is to distinguish a number of distinct conceptions of essence which, to use the later Wittgenstein’s term, are connected by relations of family resemblance. According to the first, Lockean, conception, the essence of a thing is its internal constitution which determines its discoverable qualities but is, like Kant’s objects in themselves, itself beyond the grasp of our senses. This conception appears in antiessentialist remarks by the early Moore and Carnap, serving as a quickly dismissed foil, and later reappears in the philosophical and psychological literature on psychological essentialism. The second conception is Aristotle’s. It is discussed in some detail by Stebbing and in contrast to the first conception emphasizes a direct connection between essence and definition. The third, which is found in the works of Barcan Marcus and Quine, conceives of essence in terms of the notion of necessity captured by the necessity operator of first-order modal logic and characterizes particular essentialist claims in terms of particular theorems or semantic posits about this logic. The fourth, which is, with good reason, retrospectively attributed to Moore by Kit Fine, defines essence in terms of necessity. The fifth is Fine’s own neo-Aristotelian conception, according to which essence is a primitive notion which can be used to define necessity.

84

DOI: 10.4324/9781003008750-8

Contemporary (Analytic Tradition)

6.2

Essence and the Birth of Analytic Philosophy 6.2.1 Russell

Russell and Moore’s revolt against the idealist metaphysics dominant at Cambridge around the turn of the 20th century is an important part of the founding myth of analytic philosophy.2 Their conversion to analytic philosophy impacted both Russell’s and Moore’s attitude towards essence, but as we will see, in different ways. In his first publications, Russell was still suffering from, as he much later put it, being “indoctrinated with the philosophies of Kant and Hegel” (Russell 1959: 11) at Cambridge. While these works were not focused on essence, Russell freely used essentialist terms in them. Discussing geometry, Russell for example writes that “points are wholly constituted by relations, and have no intrinsic nature of their own” (Russell 1896: 15). Similarly, he explicitly refers to the essence of quantity and of space in his discussion of quantity and number in Russell (1897b: 331), and to the essential properties of space in Russell (1897a, §61 and §80). Russell’s break with idealism meant that he also took a critical stance towards certain essentialist views. He was deeply impressed by the new logic developed by Frege and others and recognized that it could serve as the backbone of a new, scientific philosophy which fundamentally questioned many traditional philosophical ideas and theories (see Russell 1914: 191ff). Like Moore, he in particular opposed the Cambridge Idealist’s “axiom of internal relations”, the claim that “Every relation is grounded in the natures of the related terms” (Russell 1907: 37). Russell explicitly linked the notion of nature to that of a “scholastic essence” (Russell 1907: 44), but it is not evident from his arguments against the axiom and the metaphysical holism linked to it (Russell 1907; Russell 1912: ch. 14), whether he took it to have the modal dimension characteristic of contemporary essentialist views.3 Russell much later took a more clear-cut anti-essentialist stance in his History of Western Philosophy, where he objects to Aristotle’s conception of essential properties as properties which a thing “cannot change without losing its identity”, since he considered this conception “a transference to metaphysics of what is only a linguistic convenience” (Russell 1946: 193). Russell thought that the only way to make sense of essence is to treat “essence” as a synonym of “nominal definition”: “In fact, the question of ‘essence’ is one as to the use of words[;] a word may have an essence, but a thing cannot” (ibid.).

6.2.2

Moore

In his second dissertation The Metaphysical Basis of Ethics (Baldwin and Preti 2011: 117ff; submitted in 1989), G. E. Moore lauded Kant for having “pointed out that there is nothing absolutely ‘inner’ in the objects of experience” and for giving “the final blow to the doctrine of ‘essences’ and ‘faculties’, as principles of explanation” (Baldwin and Preti 2011: 223). However, while Moore rejected essences as explanatory entities, he accepted necessary connections between consecutive states of the world, i.e. between distinct entities. He thereby opposed Hume, implicitly embraced something resembling a traditional essentialist view and even approvingly linked these necessary connections to the Aristotelian notion of formal causation.4 Moore’s main discussion of essence is contained in two papers focused on the notion of intrinsicality, or the equivalent notion for relations, internality. The first, “The Conception of Intrinsic Value” (G. E. Moore 1993c), was written between 1914–7 and first published in 19225 as a companion piece to his Principia Ethica (G. E. Moore 1993b).6 A crucial claim in the paper is that value-predicates depend solely on the intrinsic nature of the things which have them.

85

Robert Michels

The notion “nature” on which Moore relies in this and his second paper on intrinsicality, is amodal. He defines the nature of a thing as “all its qualities as distinguished from its relational properties” (G. E. Moore 1993a: 103). However, his main focus is on the modal notion of intrinsicality, which he uses to define intrinsic value: A kind of value is intrinsic if and only if, […] when anything possess it, that same thing or anything exactly like it would necessarily or must always, under all circumstances, possess it in exactly the same degree. (G. E. Moore 1993c: 290) Moore spends a significant part of the paper trying to get at the notion of necessity involved. This notion is unconditional and neither identical to causal, nor to logical necessity (see G. E. Moore 1993c: 294, 291–2, 295). His example of a necessity of this sort concerns color: “if a given patch of colour be yellow, then any patch which were exactly like the first would be yellow too” (G. E. Moore 1993c: 295). His discussion ends with the remark that “what precisely is meant by this unconditional ‘must’, I must confess I don’t know” (G. E. Moore 1993c: 295). From a contemporary perspective, there seem to be two ways to answer Moore’s implied question: He might have been looking for what we now call metaphysical necessity (see Baldwin 1993: xxiii–xxiv), or instead for a sui generis notion of normative necessity (Fine 2002: 259). Moore’s second paper of interest is “External and Internal Relations” (G. E. Moore 1993a). It was first published in 1922 and directly engages with the axiom of internal relations. In it, Moore argues that the fact that some relations are internal does not imply that all of them are, pace what defenders of the axiom thought. The connection to essence is apparent in Moore’s definition of intrinsicality: A relational property is internal to an object A if, and only if, (x = A) entails x. Moore paraphrases the definition as saying that “A could not have existed in any possible world without having ” (G. E. Moore 1993a: 92), and later in the paper writes x, where that the definiens is equivalent to the claim that it is a necessary truth that (x = A) “ ” stands for material implication and, adopting what is sometimes called a “weak” notion of necessity, to the claim that it is a necessary truth that A (see G. E. Moore 1993a: 99). As Fine (1994) points out, Moore’s definition of internality is in fact a version of the modal view of essence which defines essentiality in terms of necessity. Still, it is important to point out a subtle difference to the contemporary version of the view. In his earlier paper, Moore struggled to identify the kind of modality involved in his definition of intrinsic value. His discussion of the notion of entailment reveals a similar struggle. Opposing Russell, Moore clearly distinguishes entailment from material implication (see Baldwin and Preti 2011: 100) and characterizes it as a relation between properties, such as red and colored (Baldwin and Preti 2011: 90). On the other hand, Moore also writes that entailment is the relation “which holds between the premisses and conclusion of a syllogism in Barbara” (Baldwin and Preti 2011: 101) and approvingly cites the anti-psychologists about logic of his time. Moore’s work on intrinsicality hence highlights what would become a recurring motif in early analytic philosophy, namely the struggle to explain the necessity of certain truths which the early phenomenologists called material necessities or laws of essence (see Chapter 5) from a perspective critical of traditional metaphysical theorizing.

6.3

Logical Empiricism and Post-Logical Empiricism 6.3.1

Wittgenstein

In the 1920s and 30s, the most important group working in what has become known as analytic philosophy was the Vienna Circle (see Uebel 2021). While not part of the Vienna

86

Contemporary (Analytic Tradition)

Circle himself, Wittgenstein heavily inspired its members. In his Tractatus Logico-Philosophicus (Wittgenstein 1922), Wittgenstein made frequent use of essentialist language,7 but it is important to note that he was committed to the doctrine that “[t]he only necessity that exists is logical necessity” (Tractatus §6.37). For Wittgenstein, essence was closely connected to logical form, which in the Tractatus not only pertained to language, but also to the world. Wittgenstein too struggled to fit certain necessary truths into his philosophical system. The problem manifested itself in the Tractatus in form of the color exclusion problem. This problem arises due to two conflicting commitments, namely to the logical independence of atomic states of affairs (§2.061) and to the logical necessity of exclusion relations between colors (§6.3751). The only paper he published in an academic journal, “Some Remarks on Logical Form” (Wittgenstein 1929), was an attempt to solve this problem. In the paper, Wittgenstein argued that we sometimes discover logical form only “a posteriori, and not by conjecturing about a priori possibilities” (Wittgenstein 1929: 163). He suggested that the logical form of color sentences involves numbers representing coordinates in a color space, so that “[o]ne shade of colour cannot simultaneously have two different degrees of brightness or redness” (Wittgenstein 1929: 167). Echoing Moore, Wittgenstein assumed “that the relation of difference of degree [e.g. of brightness or redness] is an internal relation” (Wittgenstein 1929: 168). His attempt to solve the problem by modifying the notion of logical form however stretches this notion beyond its limits, as it leads him to assert that the “definite rules of syntax” of a “perfect notation” would prescribe that “there is no logical product [i.e. conjunction]” (Wittgenstein 1929: 171) of incompatible elementary color statements, de facto putting them outside the scope of logic. Given the heavy cost of the solution, one could say that Wittgenstein failed to find a place for what others considered essential truths in his early philosophy.8

6.3.2 Carnap Before becoming one of the core members of the Vienna Circle, Carnap set out to realize the vision of scientific philosophy aired in the programmatic remarks at the end of Russell’s Our Knowledge of the External World.9 In his book, Der logische Aufbau der Welt (Carnap 1928; English transl.: Carnap 1967), Carnap presented a framework for what he took to be the only scientifically adequate methodology for doing philosophy, constructional theory. The task of constructional theory was to construct constructional systems, hierarchical systems of definitions, built up level by level in order to rationally reconstruct the whole of scientific knowledge, using only logical vocabulary and a small number of basic, non-logical concepts. Carnap thought these systems to be by design unfit to address “problems of essence” (see part V, ch. A), i.e. traditional metaphysical problems like the mind-body problem, since he took these problems to be outside the realm of science (see § 182). The “metaphysical” notion of essence likewise had no place within constructional theory. Carnap did admit talk of “constructional essence” by courtesy, but for him, this was just a somewhat misleading way to refer to the logical definition of the relevant concept within a constructional system (see § 59, § 161). Carnap strengthened his critique in his 1932 paper “The Elimination of Metaphysics Through Logical Analysis of Language” (Carnap 1959). There we find “essence” on a list of “specifically metaphysical terms [which] are devoid of meaning” (Carnap 1959: 67) which are used in “[t]he alleged statements of metaphysics which […] have no sense, assert nothing, are mere pseudo-statements” (Ibid.). Carnap’s early anti-essentialism was at least partly motivated by his commitment to extensional logic. He would later abandon the strictly extensional standpoint and contribute

87

Robert Michels

substantially to the development of modal logic in Meaning and Necessity (Carnap 1947), a book which paved the way for developments in this area which proved to be of crucial importance to the discussion of essence.

6.3.3 Schlick Schlick, the founder of the Vienna Circle, was also no friend of essence. In his Allgemeine Erkenntnislehre published in 1918, Schlick opposed the phenomenologists’ claim that we can “see” essences. He cited Bergson and Husserl as philosophers who wrongfully distinguish the empirically founded conceptual knowledge of the natural sciences from a genuinely philosophical kind of knowledge pertaining to essence which can be acquired through intuition (see Schlick 2009: 291). In his 1932 paper “Gibt es ein materiales Apriori?” (Schlick 2008) Schlick argued against the existence of the material a priori, the category to which the essentialist judgments which are supposed to result from so-called essential seeing belong. Schlick’s arguments are based on the logical empiricist thesis that there are only two kinds of sentences, those which express judgments which are synthetic and a posteriori, and those which express tautologies and are both analytic and a priori (see Schlick 2008: 462). Pace the phenomenologists, he insisted that sentences like “The same surface cannot be red and green all over” are a priori, but not material (see Schlick 2008: 461). Drawing on Wittgenstein’s Tractatus, Schlick argued that such sentences are “purely conceptual in nature, their validity is a logical one, they have tautological, formal character” (Schlick 2008: 466; my translation) and drawing on (Wittgenstein 1929) he argued that the “logical grammar” of color words guarantees that the same property cannot be designated by two distinct color terms like “red” and “green” (see Schlick 2008: 469).

6.3.4

Stebbing

Stebbing helped popularize the views of the members of the Vienna Circle in the UK, but was not uncritical of them. She did however share their negative attitude towards essence. Her A Modern Introduction to Logic (Stebbing 1942), contains a critical discussion of Aristotle’s theory of definition, which is more detailed and shows more historical awareness than Carnap’s or Schlick’s brief discussions of essence.10 Besides discussing the connection between essence and definition, Stebbing states that according to Aristotle, “[w]hat is essential is necessarily related to the subject, what is non-essential is accidentally related to it” (Stebbing 1942: 429), that essential characteristics make something the thing it is, and that they are indispensable to it. She also points out that Aristotle’s notion is an undefinable, fundamental, theoretical primitive (p. 430), and discusses the medieval distinction between essence and propria which Fine later linked to his distinction between constitutive and consequential essence (see Stebbing 1942: 430; Fine 1995b). Her discussion also mentions the generic nature11 of the Aristotelian notion of essence: the subject of Aristotelian definitions are species, or concepts, not names (see Stebbing 1942: 430, 432). This is noteworthy, since many later discussions, including Fine’s most influential paper, focused primarily on the essences of particulars (see also Chapter 9). Even though Stebbing’s discussion is explicitly historical, it contains a number of (partly enthymematic) systematic criticisms. She in particular argued that Aristotle’s distinction between essence and propria could not be generalized beyond mathematical subject matters and that even concerning them, the distinction cannot be understood in terms of

88

Contemporary (Analytic Tradition)

demonstration, as intended (see Stebbing 1942: 431). She also points out that Aristotle’s assumption that “essence is fixed and unalterable” (Stebbing 1942: 431–2) clashes with the fact that contemporary mathematics allows multiple definitions of the same mathematical entity. Stebbing’s book furthermore contains the objection to objectual definition which, as mentioned previously, Russell later also made in his (Russell 1946): “Do we define expressions or what the expressions stand for? Many logicians hold that it is the latter that is defined. This, however, is a mistake. Mill stated the correct view when he said “All definitions are of names, and names only” (Stebbing 1942: 426). Finally, Stebbing also drew on contemporary biology to argue against Aristotle’s idea of an exclusive and exhaustive division of kinds of things into definable species, i.e. of the pervasiveness of essences in nature. She approvingly cites Goodrich, who writes about species that “it is usually scarcely possible to find any character at all sufficiently conspicuous and constant to distinguish them from each other”.12 Stebbing’s clearly anti-essentialist conclusion is that “[m]odern theories of organic evolution have combined with modern theories of mathematics to destroy the basis of the Aristotelian conception of essence, and hence to throw doubt upon the traditional theory of definition” (Stebbing 1942: 433). Still, it is worth noting that just like Moore, Wittgenstein, and Schlick, Stebbing saw the need to account for truths stating certain intrinsic relations which phenomenologists would have taken to express laws of essence. She in particular accepted Moore’s explanation of the relation between “This is red” and “This is colored” in terms of his notion of entailment (see Stebbing 1942: 223).

6.3.5

Quine

Quine’s essay “On What There Is” (Quine 1948) played a crucial role in establishing metaphysics as a sub-discipline of analytic philosophy by giving it a logical underpinning. Quine showed that someone fully committed to the logical empiricists’ logic-based methodology could still answer metaphysical questions without wandering off into the murky territory of the “intuitive metaphysics” (Carnap, 1928: §182) they dismissed. While Quine paved the way for analytic metaphysics, his main contribution to the philosophy of essence was negative. He provided a number of highly influential arguments against what he called Aristotelian essentialism, “the doctrine that some of the attributes of a thing (quite independently of the language in which the thing is referred to, if at all) may be essential to the thing, and others accidental” (W. V. O. Quine 1953b: 173–4; see Chapter 29 for a more detailed discussion of the arguments). The arguments focus on problems regarding the semantic interpretation of modal operators, especially as they occur in formulas like x Fx , in which a quantifier outside the scope of a modal operator binds a variable inside that operator’s scope. Such formulas express de re modal claims, modal claims about particular objects in the domain of quantification. Quine considered de dicto modal claims, claims in which a modal operator applies to a closed formula involving no free variables, less problematic (see e.g. Quine 1948: 24), but thought that by accepting de re modal claims, one was led directly into the “metaphysical jungle” (W. V. O. Quine 1953b: 174) of Aristotelian essentialism, a place friends of “desert landscapes” (Quine 1948: 23) like himself abhorred. Pace Aristotelian essentialism, Quine insisted that “necessity does not properly apply to the fulfillment of conditions by objects […], apart from special ways of specifying them” (W. V. O. Quine 1953a: 151). It is clear that Quine identified essentiality and necessity13 and that his arguments left an important mark on the discussion of essentialism (see e.g. Cartwright 1968). However, unlike

89

Robert Michels

contemporary participants in this discussion, who usually take this identification to be a thesis about metaphysical necessity, Quine clearly stated that his arguments target “strict necessity”, which following C. I. Lewis and Carnap in Carnap (1947) was just another word for analyticity (W. V. O. Quine 1953a: 153). Quine briefly discussed generalizations of his arguments to physical modality and counterfactuals (W. V. O. Quine 1953a: 158), but he did not entertain the idea of a specifically metaphysical notion of necessity.14

6.4

The Way Towards the Modal View of Essence

6.4.1 Propositional Modal Logic: C. I. Lewis Crucial advances in modal logic from the 1950s to the 70s helped address Quine’s arguments against Aristotelian essentialism and Carnap’s claim that essence is a meaningless notion. C. I. Lewis made an important early contribution to the development of contemporary modal logic by axiomatizing the logic of strict implication (see Lewis 1918: ch. V) and later the standard systems of propositional modal logic S1-S5 (Lewis and Langford 1959: appendix II). Lewis did not provide a semantic interpretation for his logics,15 but paved the way for work on the semantics of quantified modal logic which crucially shaped the later discussion of essence.

6.4.2

Quantified Modal Logic: Barcan Marcus

The development of C. I. Lewis’s modal logic in Lewis and Langford’s Symbolic Logic contains some discussions of quantifiers in modal logic (see e.g. Lewis and Langford 1959: 270, 323; for propositional quantifiers: 179), but not a fully general quantified modal logic. Such a logic was introduced by Ruth Barcan Marcus in Barcan (1946) and extended to a quantified modal logic with a notion of identity in Barcan (1947). Unlike Carnap’s take on quantified modal logic in Carnap (1947), hers directly connected to the different standard systems of modal logic established by Lewis (see Fitting 2020) and did not assimilate necessity to analyticity. Barcan Marcus’s approach in these early papers was purely proof-theoretical, i.e. purely syntactical, which meant that it sidestepped Quine’s semantic arguments against quantified modal logic (see Janssen-Lauret 2022: section 6.2). An important feature of Barcan Marcus’s logic which would leave a lasting mark both on the development of quantified modal logic, as well as on its application in metaphysics are the Barcan x Fx and x Fx x Fx .16 A logic which and Reverse-Barcan formulas: x Fx verifies all instances of the two formulas allows one to transition between modal claims involving quantification into a modal context (de re modality) and within a modal context (de dicto modality), providing a way to respond to Quine’s skepticism about de re modality. Even more importantly in the current context, Barcan Marcus later defended “minimal essentialism” in her 1971 paper “Essential Attribution”, a view characterized by the formula x Fx x ¬ Fx (“there is something which is necessarily F, but there is something which is not necessarily F”); (Barcan Marcus 1993: 62; for an in-depth discussion, also relating the view to Fine’s arguments against the modal view of essence, see Leech 2023). Any quantified modal logic which has this thesis as a theorem allows non-trivial necessary predications, i.e. predications which necessarily characterize some things as opposed to others. Barcan Marcus can be taken to show that this and two stronger versions of Aristotelian essentialism (or extensions of these views) are not, as Quine called them, “indefensible” (Quine 1960: 183). Discussing a particular application of the view, she concluded that “[i]nstead of leading into a metaphysical jungle, formulation of our analysis

90

Contemporary (Analytic Tradition)

[of Aristotelian essentialism] within a causal interpretation of QML [Quantified Modal Logic] is suggestive and illuminating” (Barcan Marcus 1993: 69–70).

6.4.3 Quantified Modal Logic: Kripke With axiomatic modal logic in both its propositional and quantified varieties put on a prooftheoretic basis, what was still missing was a workable semantics. Progress was made by several philosophers, including Carnap, Kanger, Montague, and Hintikka, but the development of the now standard relational, model-theoretic, or Kripke semantics for modal logic found its culmination in Kripke’s logical papers.17 This semantics was first introduced for the quantified version of the modal logic S5 with identity and proven complete in Kripke (1959). Kripke’s semantics was based on the same idea as the one developed in Carnap (1947), namely that modal notions like necessity and possibility can be interpreted semantically in terms of truth in possible worlds. Two crucial features of the new semantics were that it treated possible worlds as simple points of evaluation instead of as Carnapian statedescriptions or as extensional models and that it interpreted modal operators semantically using a binary accessibility relations between possible worlds. This new approach allowed Kripke to establish the correspondence between the normal modal systems introduced by C. I. Lewis and certain formal properties of the accessibility relation in a semantic structure called a frame (see S. A. Kripke 1963a). This result was crucial for the further development of modal logic. Kripke’s work also shone further light on the metaphysical significance of the Barcan formulas. These formulas are true in constant-domain Kripke models, models in which the quantifiers and variables are interpreted in the same domain of objects with respect to all possible worlds. S. A. Kripke (1963b) introduced variable domain models and showed that the Barcan and reverse Barcan formulas are falsified in some of them. This shows that the inter-translatability between de re and de dicto modal claims ensured by the formulas is tied to a particular constraint on the semantic interpretation of the modal operators.

6.4.4

Direct Reference, Rigid Designation, Metaphysical Modality, and the Modal View

The availability of modern modal logic at least partly addressed Carnap’s charge of meaninglessness and Quine’s semantic objections against quantification into modal contexts. It also helped bring a metaphysical view, which was arguably already tacitly present in some attacks as well as cautious defenses of essence, into the lime-light: the modal view of essence, according to which essentiality either coincides with, or can be analyzed in terms of necessity. Quantified modal logic hence also promised to provide a fruitful heuristic for metaphysical speculation about essence: Assuming that the essential properties of a thing are those properties which it has in all possible worlds, the claim that an object essentially has a property can be justified by arguing that the object has the property in all, or rejected by arguing that it lacks the property in some worlds. Modal logic alone licensed neither this sort of reasoning, nor substantial metaphysical claims,18 but a number of interrelated developments in the philosophies of language, of science and in metaphysics helped legitimize both. Leading proponents of essentialist ideas, including in particular Barcan-Marcus (Barcan Marcus 1961: 310), adopted the Millian (see Mill 1974: book I, ch. II) view that proper names are directly referential, i.e. refer directly to their bearers without the mediate step through a definite description. Kripke defended the closely connected view that proper names are rigid designators, i.e. that they refer to the same object with respect to all possible worlds (see Kripke 1980: 48ff). Putnam generalized these theses to natural kind and magnitude terms

91

Robert Michels

(see Putnam 1970; Putnam 1973). These semantic views explicitly contradicted the thendominant descriptivist view, according to which the meaning of singular terms, including proper names, consists in definite descriptions of their referents (see e.g. Russell 1905; Searle 1958). They hence avoided difficult questions regarding the reference of proper names in different possible worlds, which, given descriptivism, would threaten to hinder reasoning about the contingent and necessary properties of objects. They also offered a bold counterpoint to Quine’s claim that we cannot make sense of the idea that things necessarily (or essentially) have properties independently of how we refer to them. Even though these views did arguably not by themselves entail substantial essentialist claims (see Salmon 2005), they certainly made it easier to entertain and defend them. Putnam for example argued that the use of natural kind terms like “water” or “tiger”, indicate certain characteristic features of objects which he took to be likely explainable in terms of an underlying essential nature which those objects share with others of their kind (Putnam 1970: 188). Similarly, Kripke argued that theoretical identifications involving natural kind terms, such as that of gold being the element with atomic number 79, of heat being molecular motion, or of cats being animals, are necessary, if true (Kripke 1980: 138). More controversial are Kripke’s defense of the essentiality of origin (Kripke 1980: 113; see also Chapter 11) or his argument that there could not have been unicorns (Kripke 1980: 24, 156–7). Another closely related development was a shift from a semantic to a more metaphysical perspective. Carnap and Quine pursued a purely semantic approach to modality and focused on analyzing the meaning of modal notions. Semantic considerations about the meaning of proper names still played a crucial role in Kripke’s Naming and Necessity, but Kripke did not shy away from metaphysical considerations which, so to say, put the objects first. He for example argued that when reasoning about what might or might not be the case concerning a particular object, we need not first worry about finding a criterion of trans-world identity, a description listing certain of its characteristic properties which we can then use to identify the object in different possible worlds. Instead “we begin with the objects, which we have, and can identify, in the actual world. We can then ask whether certain things might have been true of the objects” (Kripke 1980: 53). A question which was lingering at least since Moore’s work on intrinsicality was the question of the kind of necessity involved in essentialist claims. Both Moore’s work, as well as Wittgenstein’s problematic attempt to account for necessary exclusion-relations between colors, suggest that essential truths are, pace Schlick or Ayer (Ayer 1952: 148 f) neither analytic, nor conceptually necessary or tautologous, but could rather be taken to be necessary in a different, so far elusive, sense. An answer which has now found widespread acceptance among analytic metaphysicians is that the necessity inherent to essential truth is metaphysical necessity. In her pioneering 1938 paper “Logical and Metaphysical Necessity” (Kneale 1938) Martha Kneale applied the notion of metaphysical necessity to propositions “concerned with the most general characteristics of reality and the necessary relations between them” (Kneale 1938: 264). Her main example were propositions about the relation between tenses, which are not analytic since their negations are not logical contradictions. The notion however only became a staple in analytic philosophy about 30 years later in a climate more friendly to both metaphysical and modal reasoning, after Kripke advocated for it in Naming and Necessity. An important step towards establishing this climate was Kripke’s insistence on a distinction which was already stressed by Kneale, namely that between apriority and necessity.19 Kripke argued that apriority should be considered an epistemological notion and that as such, it should be clearly distinguished from necessity, which he considered to be a metaphysical notion. This controverted the idea that necessity and

92

Contemporary (Analytic Tradition)

apriority coincide, an idea which can even be tracked back to the B-introduction to Kant’s Critique of Pure Reason. It also paved the way for the recognition that there are truths which are a priori, but contingent on the one, and a posteriori and necessary on the other hand. One of Kripke’s by now well-worn examples of a truth of the latter kind is “Water is H2O”. He argued that this claim expresses a necessary truth about the substance water, but since this truth had to be discovered empirically, our knowledge of it is not a priori (Kripke 1980: 127). Like the move to a directly referential theory of natural kind terms, the dissociation of necessity from apriority effectively contributed to broadening the subject matters about which essentialist claims could be made. However, it also raised substantial epistemological questions about the possibility of knowledge of essences (see Chapter 14). The advances in modal logic, the new approach to meaning, reference, and possible worlds, and the recognition of a sui generis notion of metaphysical necessity all helped establish the modal view as the de facto standard approach to essence. An explicit formulation of it is arguably provided by Forbes, who writes that “if P is an essential property of x, then for all possible worlds w, if x exists in w then x has P in w” (Forbes 1985, p. 94; see also Chapter 7). This, or equivalent views of essentiality were widely assumed in discussions of essentialist views in the 1970s and 80s, including for example in Chisholm’s defense of mereological essentialism (Chisholm 1973), or Plantinga’s book The Nature of Necessity (Plantinga 1974).20

6.5

Fine

By the 1980s, the modal view of essence was well-established, but not wholly uncontroversial. Wiggins for example argued against framing essentialist claims in terms of a sentential necessity-operator (Wiggins 1976), Almog rejected the identification of essence with necessity and argued for a non-modal notion of essence tied to “what questions” (Almog 1991), and David Lewis saw no substantial role to play for essence in his metaphysics, but thought that if one so wishes, one could use his counterpart theory to make sense of a flexible, contextdependent notion of essence (Lewis 1986: 252). Despite such reservations, the modal view of essence arguably remained the standard view. This changed after Kit Fine’s paper “Essence and Modality” (Fine 1994). In it, Fine raised four objections to the modal view, proposed a Neo-Aristotelian view which links essence to real definition, and argued that instead of explaining essence in modal terms, we can use essence to account for different kinds of necessity, including in particular metaphysical necessity.21 Both the objections, as well as Fine’s positive proposal from “Essence and Modality” have been immensely influential. While this paper continues to receive most attention, important aspects of Fine’s theory of essence are presented in other papers, including in particular a discussion of a range of different notions of essence (Fine 1995b), an essentialist account of ontological dependence (Fine 1995a), and the logic of essence (Fine 1995c). In the latter paper, Fine develops a logic for a first-order language involving an essentialist sentence operator (the formal counterpart to Fine’s primitive notion “true in virtue of the nature of”) and in particular proves that what many take to be the standard logic of metaphysical necessity, the system S5 of first-order modal logic with constant domains, reduces to Fine’s system E5+ in the sense that any theorem of the former is also a theorem of the latter system (Fine 1995c: p. 267, Theorem 4). In his later paper (Fine 2000), Fine presents a semantics for this logic, showing that semantically, as well as syntactically, the logic of essence can be seen as an extension of what is often thought to be the standard system of quantified modal logic (see Chapter 16).

93

Robert Michels

The importance of Fine’s work on essence can hardly be overstated. His papers have not only re-established the notion of essence as a theoretical primitive, they have also paved the way for what has been called “a hyperintensional revolution” (Nolan 2014: 149) in metaphysics, a movement to rely on notions which are more fine-grained than the modal notions which were abundantly used by analytic metaphysics in the last third of the 20th century.

6.6

The Post-Finean Landscape in Metaphysics

Fine’s work on essence was a watershed moment in the metaphysical discussion of essence, so it is useful to briefly survey recent developments in this area in relation to his work. Many metaphysicians like Lowe (Lowe 2008; Lowe 2016) are in agreement with Fine’s approach to essence or rely on it in their works. Others like Correia and Zylstra have elaborated particular aspects of his approach, so as e.g. to include a generic notion of essence besides the objectual notion, to elaborate the reduction of different kinds of necessity to essence (Correia 2006; Correia 2012), or to expand on the idea that pluralities have essences (Zylstra 2019). There are also a number of metaphysicians who agree with Fine in rejecting the modal view, but propose alternatives to his approach to essence. Fine himself has developed a new account of essence which relates it and grounding to necessary and sufficient conditions respectively (Fine 2015). In a similar spirit (Correia and Skiles 2019) assimilate essence to generalized identity. Hylomorphic accounts of essence have been discussed and defended in Oderberg (2011) and Koslicki (2018). Kment (2014) and Hale (2013) both propose theories of modality in which a primitive notion of essential truth plays a crucial role. Some critics of Fine focus on particular aspects of his view, including its compatibility with metaphysical contingentism (Teitel 2017; for discussion see Werner 2020), assumptions built into the logic of essence (Ditter 2020), or its potential to provide a reductive explanation of modality (Casullo 2020; Leech 2021; Mackie 2020; Noonan 2018; Romero 2019; for discussion see Wallner and Vaidya 2020). Others acknowledge Fine’s counterexamples to the modal view, but argue that the view can be saved by relying on additional, non-modal22 devices in addition to necessity, either in the form of further theoretical notions or special logical frameworks (Wildman 2013 – sparseness, Cowling 2013; Melo 2019 – naturalness, Rizzo 2022 – grounding, Denby 2014; Bovey 2021 – intrinsicality, Correia 2007 – Priorean strict implication, Zalta 2006 – theory of abstract objects). Fine’s influence notwithstanding, there are some authors who have developed essentialist theories which keep the core identification of essentiality and de re necessity of the modal view intact. Two examples are Paul’s modal account of essence which draws on David Lewis’s view of essence as variable and context-dependent (Paul 2004; Paul 2006), and Mackie’s minimal essentialism (Mackie 2006). Outside of the specialized discussion in metaphysics, essentialist views are more often than not understood in terms of either the modal view, or even weaker views which link essence to extensional definition. This is especially noticeable in areas of philosophy or specific discussions closely connected to the sciences or to other humanities. An example of a relatively popular theory which (in most of its variants) relies on a modal view of essence is dispositional essentialism in the metaphysics of science (see Chapters 12 and 13). Views like essentialism about species in the philosophy of biology (see Chapter 18), psychological essentialism (see Chapter 15, and e.g. Gelman 2003; Leslie 2013), and a number of essentialist views in the philosophies of psychiatry (see Chapter 23), race (see Chapter 24), and sex and gender (see Chapter 25) are in contrast highly controversial, specifically because of pervasive

94

Contemporary (Analytic Tradition)

skeptical views about the unique (extensional) definability of relevant natural kind or species terms in these discussions.

6.7

Related Topics

Chapters 29 (Quinean Anti-Essentialism), 7 and 8 (Modal and Non-Modal Conceptions of Essence) elaborate topics which are briefly touched on in this chapter. The discussions of Aristotle’s, Locke’s, and Wittgenstein’s conceptions of essence in Chapters 1, 3, and 33 provide particularly relevant historical background. Systematic and historical views from roughly the same period which are not covered in this chapter are discussed in Chapters 4 (Pragmatism), 5 (essence in 20th century phenomenology), 30 (Conventionalism), and 32 (Conferralism). Connections to further chapters are pointed out in the chapter.

Acknowledgments I would like to thank audiences at the Handbook of Essence workshop, sessions of the Mind and Metaphysics colloquium at the University of Bern, and the eidos seminar at the University of Geneva and in particular Philip Blum, Fabrice Correia, Kit Fine, Vera Hoffmann-Kolss, Rebekka Hufendiek, Frederique Janssen-Lauret, David Kovacs, Karol Lenart, Jon Litland, Valentina Luporini, Kevin Mulligan, Paolo Natali, Gonzalo Rodriguez-Pereyra, David Schroeren, Alan Sidelle, Jonas Werner and the editors of the handbook, Kathrin Koslicki, and Mike Raven. Work on this paper was supported by a grant of the Swiss National Science Foundation (grant number 185435, PI: Kevin Mulligan).

Notes 1 Fine (1994) and Lowe (2008) use them interchangeably, Almog (2010) distinguishes “essence” and “nature”. 2 See for example ( Preston, n.d.: section 6.1). 3 Cf. ( Russell 1912: 84) and (G. E. Moore 1993a: 103). The modal character of the axiom was brought out more clearly by Moore and later also by Ayer (1952: 146ff). 4 See Baldwin and Preti (2011: 52). 5 See G. E. Moore (1993c: p. 280, editor’s note). 6 See Hurka (2015: section 6.2). 7 See in particular §2.011, §3.1431, §3.3421, §4.03, §5.471, §5.4711. 8 See Chapter 33 for a more detailed discussion of essence in Wittgenstein. 9 See Schilpp (1963: 12). 10 I am grateful to Frederique Janssen-Lauret for pointing me to Stebbing’s book and this section in particular. 11 See Chapter 1 for discussion. 12 Cited in Stebbing (1942), p. 439. Original source: ( Goodrich 1924), pp. 150–1. For more on essence in biology, see Chapter 18. 13 See the alternative statement of the doctrine of Aristotelian essentialism in W. V. O. Quine (1953a: 155). 14 He would much later write that ‘[m]etaphysical necessity has no place in my naturalistic view of things’. ( Quine 1991: 270) 15 See Mares (2016) for discussion. 16 These canonical formulations of the principles are not identical to, but implied by formulas involving a strict conditional instead of the necessity-operator in her paper. See Barcan (1946), axiom schema 11 and theorem 37. 17 See ( Goldblatt 2005: sec. 4; Ballarin 2021: sec. 3.2).

95

Robert Michels 18 The proof of the necessity of identity could be taken as an exception, but see Burgess (2013) for discussion. See also Williamson (2013). 19 For a discussion of Kneale’s view of metaphysical necessity including a detailed comparison to Kripke’s view, see Leech (2019). 20 Whether authors like Forbes or Plantinga can indeed be said to advocate an analysis of the notion of essence might not be as clear-cut as it is usually assumed to be. See Chapter 8, footnote 1. 21 See Chapter 8 and Michels (2019) for more detailed discussions of Fine’s objections and his positive view. 22 See Torza (2015).

Bibliography Almog, J. (1991) “The What and the How,” Journal of Philosophy 88(5):225. Almog, J. (2010) “Nature Without Essence,” Journal of Philosophy 107(7):360–383. Ayer, A. J. (1952) Language, Truth and Logic, 2nd ed., New York: Dover Publications. Baldwin, T. (1993) “Editor’s Introduction,” in G. E. Moore (ed.) Principia Ethica. Revised Edition. With the Preface to the Second Edition and Other Papers, Cambridge: Cambridge University Press. Baldwin, T. and Preti, C. (eds.) (2011) G. E. Moore: Early Philosophical Writings. Cambridge: Cambridge University Press. Ballarin, R. (2021) “Modern Origins of Modal Logic,” in E. N. Zalta (ed.) The Stanford Encyclopedia of Philosophy, Fall 2021. https://plato.stanford.edu/archives/fall2021/entries/logic-modal-origins/; Metaphysics Research Lab, Stanford University. Barcan, R. C. (1946) “A Functional Calculus of First Order Based on Strict Implication,” Journal of Symbolic Logic 11(1):1–16. Barcan, R. C. (1947) “The Identity of Individuals in a Strict Functional Calculus of Second Order,” Journal of Symbolic Logic 12(1):12–15. Barcan Marcus, R. (1961) “Modalities and Intensional Languages,” Synthese 13 (4): 303–322. Barcan Marcus, R. (1993) “Essential Attribution,” in R. Barcan Marcus (ed.) Modalities. Philosophical Essays, 53–70. Oxford: Oxford University Press. Bovey, G. (2021) “Essence, Modality, and Intrinsicality,” Synthese 198(8):7715–7737. Burgess, J. P. (2013) “On a Derivation of the Necessity of Identity,” Synthese 191(7):1–19. Carnap, R. (1947) Meaning and Necessity. A Study in Semantics and Modal Logic. Chicago: The University of Chicago Press. Carnap, R. (1959) “The Elimination of Metaphysics Through Logical Analysis of Language,” in A. J. Ayer (ed.) Logical Positivism, New York: The Free Press. Carnap, R. (1928) Der Logische Aufbau Der Welt, 3rd ed., Hambrug: Felix Meiner Verlag. Carnap, R. (1967) The Logical Structure of the World and Pseudoproblems in Philosophy, Chicago: Open Court. Cartwright, R. L. (1968) “Some Remarks on Essentialism,” Journal of Philosophy 65(20):615–626. Casullo, A. (2020) “Essence and Explanation,” Metaphysics 2(1):88–96. Chisholm, R. M. (1973) “Parts as Essential to Their Wholes,” The Review of Metaphysics 26(4):581–603. Correia, F. (2006) “Generic Essence, Objectual Essence, and Modality,” Noûs 40(4):753–767. Correia, F. (2007) “(Finean) Essence and (Priorean) Modality,” Dialectica 61(1):63–84. Correia, F. (2012) “On the Reduction of Necessity to Essence,” Philosophy and Phenomenological Research 84(3):639–653. Correia, F. and Skiles, A. (2019) “Grounding, Essence, and Identity,” Philosophy and Phenomenological Research 98(3):642–670. Cowling, S. (2013) “The Modal View of Essence,” Canadian Journal of Philosophy 43(2):248–266. Denby, D. A. (2014) “Essence and Intrinsicality,” in R. Francescotti (ed.) Companion to Intrinsic Properties, Berlin & Boston: De Gruyter. Ditter, A. (2020) “The Reduction of Necessity to Essence,” Mind 129(514):351–380. Fine, K. (1994) “Essence and Modality,” Philosophical Perspectives 8, Logic and Language: 1–16. Fine, K. (1995a) “Ontological Dependence,” Proceedings of the Aristotelian Society 95:269–290. Fine, K. (1995b) “Senses of Essence,” in W. Sinnott-Armstrong (ed.) Modality, Morality, and Belief. Essays in Honor of Ruth Barcan Marcus, Cambridge: Cambridge University Press.

96

Contemporary (Analytic Tradition) Fine, K. (1995c) “The Logic of Essence,” Journal of Philosophical Logic 24(3):241–273. Fine, K. (2000) “Semantics for the Logic of Essence,” Journal of Philosophical Logic 29(6):543–584. Fine, K. (2002) “The Varieties of Necessity,” in K. Fine (ed.) Modality and Tense. Philosophical Papers, Oxford: Oxford University Press. Fine, K. (2015) “Unified Foundations for Essence and Ground,” Journal of the American Philosophical Association 1(2):296–311. Fitting, M. (2020) “Intensional Logic,” in E. N. Zalta (ed.) The Stanford Encyclopedia of Philosophy, Spring 2020. https://plato.stanford.edu/archives/spr2020/entries/logic-intensional/; Metaphysics Research Lab, Stanford University. Forbes, G. (1985) The Metaphysics of Modality, Oxford: Clarendon Press. Gelman, S. A. (2003) The Essential Child: Origins of Essentialism in Everyday Thought, Oxford: Oxford University Press. Goldblatt, R. (2005) “Mathematical Modal Logic: A View of Its Evolution,” in D. M. Gabbay (ed.) Handbook of the History of Logic Vol. 6, Amsterdam: Elsevier. Goodrich, E. S. (1924) Living Organisms. An Account of Their Origin & Evolution, Oxford: Oxford University Press. Hale, B. (2013) Necessary Beings: An Essay on Ontology, Modality, and the Relations Between Them, Oxford: Oxford University Press. Hurka, T. (2015) “Moore’s Moral Philosophy,” in E. N. Zalta (ed.) The Stanford Encyclopedia of Philosophy, Fall 2015. https://plato.stanford.edu/archives/fall2015/entries/moore-moral/; Metaphysics Research Lab, Stanford University. Janssen-Lauret, F. (2022) “Ruth Barcan Marcus and Quantified Modal Logic,” British Journal for the History of Philosophy 30(2):353–383, doi: 10.1080/09608788.2021.1984872. Kment, B. (2014) Modality and Explanatory Reasoning, Oxford: Oxford University Press. Kneale, M. (1938) “Logical and Metaphysical Necessity,” Proceedings of the Aristotelian Society 38(1):253–268. Koslicki, K. (2018) Form, Matter, Substance, Oxford: Oxford University Press. Kripke, S. A. (1959) “A Completeness Theorem in Modal Logic,” Journal of Symbolic Logic 24(1):1–14. Kripke, S. A. (1963a) “Semantical Analysis of Modal Logic I. Normal Propositional Calculi,” Zeitschrift für Mathematische Logik Und Grundlagen Der Mathematik 9(5-6):67–96. Kripke, S. A. (1963b) “Semantical Considerations on Modal Logic,” Acta Philosophica Fennica 16:83–94. Kripke, S. A. (1980) Naming and Necessity, Cambridge: Harvard University Press. Leech, J. (2019) “Martha Kneale on Why Metaphysical Necessities Are Not a Priori,” Journal of the American Philosophical Association 5(4):389–409. Leech, J. (2021) “From Essence to Necessity via Identity,” Mind 130(519):887–908. Leech, J. (2023) “Ruth Barcan Marcus and Minimal Essentialism,” Ratio 36(4):289–305. Leslie, S.-J. (2013) “Essence and Natural Kinds: When Science Meets Preschooler Intuition,” Oxford Studies in Epistemology 4: 108–166. Lewis, C. I. and Langford, C. H. (1959) Symbolic Logic, 2nd ed., New York: Dover Publications. Lewis, C. I. (1918) A Survey of Symbolic Logic, 1st ed., Berkeley: University of California Press. Lewis, D. (1986) On the Plurality of Worlds, Oxford: Blackwell Publishing. Lowe, E. J. (2016) “Metaphysics as the Science of Essence,” Analysis 19(7):1–30. Lowe, E. J. (2008) “Two Notions of Being: Entity and Essence,” Royal Institute of Philosophy Supplements 83(62):23–48. Mackie, P. (2006) How Things Might Have Been, Oxford: Clarendon Press. Mackie, P. (2020) “Can Metaphysical Modality Be Based on Essence?,” in M. Dumitru (ed.) Metaphysics, Meaning, and Modality. Themes from Kit Fine, Oxford: Oxford University Press. Mares, E. (2016) “The Development of C. I. Lewis’s Philosophy of Modal Logic,” in M. Cresswell, E. Mares and A. Rini (eds.) Logical Modalities from Aristotle to Carnap: The Story of Necessity, Cambridge: Cambridge University Press. Melo, T. X. de (2019) “Essence and Naturalness,” Philosophical Quarterly 69(276):534–554. Michels, R. (2019) “On How (Not) to Define Modality in Terms of Essence,” Philosophical Studies 176(4):1015–1033. Mill, J. S. (1974) A System of Logic, Ratiocinative and Inductive, Collected Works of John Stuart Mill Vol. VII, Toronto: University of Toronto Press, London: Routledge and Kegan Paul.

97

Robert Michels Moore, G. E. (1993a) “External and Internal Relations,” in G. E. Moore (ed.) Selected Writings, London: Routledge. Moore, G. E. (1993b) Principia Ethica. Revised Edition. With the Preface to the Second Edition and Other Papers, Cambridge: Cambridge University Press. Moore, G. E. (1993c) “The Concept of Intrinsic Value,” in Principia Ethica. Revised Edition. With the Preface to the Second Edition and Other Papers, Cambridge: Cambridge University Press. Nolan, D. (2014) “Hyperintensional Metaphysics,” Philosophical Studies 171(1):149–160. Noonan, H. W. (2018) “The New Aristotelian Essentialists,” Metaphysica 19(1):87–93. Oderberg, D. S. (2011) “Essence and Properties,” Erkenntnis 75(1):85–111. Paul, L. A. (2004) “The Context of Essence,” Australasian Journal of Philosophy 82(1):170–184. Paul, L. A. (2006) “In Defense of Essentialism,” Philosophical Perspectives 20(1):333–372. Plantinga, A. (1974) The Nature of Necessity, Oxford: Oxford University Press. Preston, A. (no date) “Analytic Philosophy,” in The Internet Encyclopedia of Philosophy. https://www. iep.utm.edu/analytic/. Putnam, H. (1970) “Is Semantics Possible?,” Metaphilosophy 1(3):187–201. Putnam, H. (1973) “Explanation and Reference,” in G. Pearce and P. Maynard (eds.) Conceptual Change, Dordrecht: D. Reidel. Quine, W. V. O. (1948) “On What There Is,” in W. V. O. Quine (ed.) From a Logical Point of View, 2nd ed., Cambridge: Harvard University Press. Quine, W. V. O. (1953a) “Reference and Modality,” in W. V. O. Quine (ed.) From a Logical Point of View, 2nd ed., Cambridge: Harvard University Press. Quine, W. V. O. (1953b) “Three Grades of Modal Involvement,” in W. V. O. Quine (ed.) The Ways of Paradox and Other Essays, 1st ed., London: Random House. Quine, W. V. O. (1960) Word and Object, Cambridge: The MIT Press. Quine, W. V. O. (1991) “Two Dogmas in Retrospect,” Canadian Journal of Philosophy 21(3):265–274. Rizzo, J. De (2022) “A Ground-Theoretical Modal Definition of Essence,” Analysis 82(1):32–41. Romero, C. (2019) “Modality Is Not Explainable by Essence,” Philosophical Quarterly 69(274):121–141. Russell, B. (1896) “The Logic of Geometry,” Mind 5(17):1–23. Russell, B. (1897a) An Essay in the Foundations of Geometry, Cambridge: Cambridge University Press. Russell, B. (1897b) “On the Relations of Number and Quantity,” Mind 6(3):326–341. Russell, B. (1905) “On Denoting,” Mind 14(56):479–493. Russell, B. (1907) “On the Nature of Truth,” Proceedings of the Aristotelian Society 7(1):28–49. Russell, B. (1912) The Problems of Philosophy, Oxford: Oxford University Press. Russell, B. (1914) Our Knowledge of the External World: As a Field for Scientific Method in Philosophy, London: Routledge. Russell, B. (1946) History of Western Philosophy, London: Routledge. Russell, B. (1959) My Philosophical Development, New York: Simon and Schuster. Salmon, N. U. (2005) Reference and Essence, 2nd ed., Amherst: Prometheus Books. Schilpp, P. A. (1963) The Philosophy of Rudolf Carnap, The Library of Living Philosophers Vol. 11, La Salle: Open Court. Schlick, M. (2008) “Gibt Es Ein Materiales Apriori?,” in J. Friedl and H. Rutte (eds.) Die Wiener Zeit. Aufsätze, Beiträge, Rezensionen 1926–1936, Moritz Schlick Gesamtausgabe Band I/6, Wien, New York: Springer. Schlick, M. (2009) Allgemeine Erkenntnislehre. Moritz Schlick Gesamtausgabe Band I/1, Wien, New York: Springer. Searle, J. R. (1958) “Proper Names,” Mind 67(266):166–173. Stebbing, S. (1942) A Modern Introduction to Logic, 3rd ed, London: Methuen & Co. Ltd. Teitel, T. (2017) “Contingent Existence and the Reduction of Modality to Essence,” Mind 128(509):39–68. Torza, A. (2015) “Speaking of Essence,” Philosophical Quarterly 65(261):754–771. Uebel, T. (2021) “Vienna Circle,” in The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Fall 2021. https://plato.stanford.edu/archives/fall2021/entries/vienna-circle/; Metaphysics Research Lab, Stanford University. Wallner, M. and Vaidya, A. (2020) “Essence, Explanation, and Modality,” Philosophy 95(4):419–445. Werner, J. (2020) “Contingent Objects, Contingent Propositions, and Essentialism,” Mind 130(520): 1283–1294.

98

Contemporary (Analytic Tradition) Wiggins, D. (1976) “The de Re ‘Must’: A Note on the Logical Form of Essentialist Claims,” in G. Evans and J. McDowell (eds.) Truth and Meaning, Oxford: Oxford University Press. Wildman, N. (2013) “Modality, Sparsity, and Essence,” Philosophical Quarterly 63(243):760–782. Williamson, T. (2013) Modal Logic as Metaphysics, Oxford: Oxford University Press. Wittgenstein, L. (1922) Tractatus Logico-Philosophicus, London: Routledge & Kegan Paul. Wittgenstein, L. (1929) “Some Remarks on Logical Form,” Aristotelian Society Supplementary Volume 9(1):162–171. Zalta, E. N. (2006) “Essence and Modality,” Mind 115(459):659–693. Zylstra, J. (2019) “Collective Essence and Monotonicity,” Erkenntnis 84(5):1087–1101.

99

PART 2

Essence and Essentialisms Themes and Variations Kathrin Koslicki and Michael J. Raven

Introduction Once a theorist is willing to entertain the notion of essence, they are immediately faced with a grab-bag of questions, distinctions, and choice points about how to understand it and how to formulate essentialist doctrines. Part II concerns some of the most prominent of these. In recent decades, it has become common to distinguish between modal and non-modal conceptions of essence (Chapters 7 and 8). A key choice point, then, is which conception better captures the notion of essence. Another key distinction is between the essences of individuals and the essences of kinds (Chapter 9). Given the notion of essence, various essentialist doctrines may then be formulated in terms of it. Natural kind essentialism holds that there are natural kinds that have essences (Chapter 10). Origin essentialism, prominent especially in Kripke’s work, holds that an individual’s origins (and, specifically, often their material origins) are essential to them (Chapter 11). Scientific essentialism holds that entities recognized by our best science have essences (Chapter 12). Dispositional essentialism holds that at least some properties have a noncategorical, dispositional nature (Chapter 13). The notion of essence, as it figures in these various essentialist doctrines, gives rise to important epistemic, linguistic, and logical questions. The epistemology of essence concerns how essences are knowable and, if so, whether there is a distinctive route to gaining such knowledge (Chapter 14). The language of essence concerns whether, and how, our linguistic or mental representations impose essentialist constraints on what we talk, or think, about (Chapter 15). Finally, the logic of essence concerns how formal (proof-theoretic and semantic) tools may be used to regiment and study the notion of essence (Chapter 16). Chapter 7, Alessandro Torza’s “Modal Conceptions of Essence” focuses on approaches to essence that conceive of it as a fundamentally modal notion. On the dominant version of this approach, an item has a property essentially if and only if it has that property of necessity. In effect, essence is de re modality. Although the modal conception of essence can be found throughout the history of philosophy, it was especially prominent in the early 20th century among both proponents and opponents of essence alike. The status quo changed in the late 20th century with Kit Fine’s influential counterexamples to the modal conception. Torza’s

DOI: 10.4324/9781003008750-9

Kathrin Koslicki and Michael J. Raven

chapter elaborates on the modal conception of essence, Fine’s counterexamples, and some recent attempts to formulate a modal conception immune to them. Chapter 8, Fabrice Correia’s “Non-modal Conceptions of Essence” focuses on approaches to essence that do not conceive of it as a fundamentally modal notion. Such approaches owe much to Kit Fine’s influential work. Fine’s counterexamples to the modal conception of essence were not merely intended to refute it. Their force, according to Fine, derived from our independent grasp of a non-modal conception of essence. Indeed, Fine even suggested a program of reducing metaphysical modality to this non-modal conception. Correia’s chapters elaborates on this non-modal conception of essence as well as the idea that modality has its “source” in essence. Chapter 9, Marco Marabello’s “Essences of Individuals” concerns the key distinction between the essences of individuals and the essences of kinds, focusing especially on the former. To illustrate, we may distinguish between an individual tiger, say Tony, and tigers in general. Given this distinction, we may ask: what is essential to Tony the tiger, and what is essential to tigers? The questions may receive different answers. It may be that whatever is essential to tigers must also be essential to each individual tiger, including Tony. But it needn’t follow from this that whatever is essential to Tony must also be essential to tigers in general. Marabello’s chapter surveys prominent approaches to the essences of individuals. One such approach posits individual essences, sometimes called “thisnesses” or “haecceities”, that account for the identities of individuals but without regard to their qualities or what sort of thing they are. On this view, although Tony is a tiger, Tony is not essentially a tiger. Another approach, sometimes called “sortal essentialism”, holds that an individual essentially is the sort of thing that it is. On this view, Tony is essentially a tiger. Chapter 10, Tuomas Tahko’s “Natural Kind Essentialism” focuses on the doctrine that there are real and objective categories in nature. Essentialism about natural kinds may come in three varieties: that it is essential to an individual that it belongs to a kind; that each instance of kind shares an essence; and that a kind has an essence. Tahko’s chapter focuses largely on the second variety: that instances of a kind are unified by a shared essence. Much recent interest in this form of natural kind essentialism derives from Kripke’s and Putnam’s work. They argued, for example, that water has a chemical nature and so, in particular each water sample essentially has the molecular structure of H2O. This is also an example of a view on which natural kind essences are intrinsic to their instances. But some have also considered views on which natural kind essences may be extrinsic or relational. Tahko surveys these and other approaches to natural kind essences. Chapter 11, Teresa Robertson Ishii’s “Origin Essentialism” concerns the idea that an individual’s origin is essential to it in the sense that it could not have had an entirely different origin. This idea, in various forms, rose to prominence from Kripke’s influential Naming and Necessity. Kripke’s idea was that there are no genuinely metaphysically possible scenarios in which an individual has an origin that does not overlap its actual origin. For example, Queen Elizabeth, who was a child of Elizabeth Bowes-Lyon and George VI, could not have been a child of Harry and Bess Truman instead. And a table that was in fact originally made from wood could not have been originally made from ice. Robertson Ishii’s chapter discusses origin essentialism as well as a famous paradox that arises for it. Chapter 12, Travis Dumsday’s “Scientific Essentialism” focuses on a particular essentialist system for scientific kinds. This is the system developed in Brian Ellis’s work. It combines elements of both natural kind essentialism and dispositional essentialism, but with a few twists. Ellis’s system may be contrasted with a broadly Humean view on which, roughly speaking: laws of nature are mere contingent regularities, causality is non-logical, dispositions

102

Essence and Essentialisms

depend on the laws of nature, and there are no irreducibly modal properties in the world. Scientific essentialism, at least of Ellis’s variety, rejects all components of the Humean view. Dumsday’s chapter elaborates on just what this scientific essentialist view is and considers some of its respective costs and benefits. Chapter 13, Ka Ho Lam’s “Dispositional Essentialism” concerns the doctrine of dispositional essentialism. The doctrine relies on a distinction between dispositional properties that possess a propensity or power to affect things and categorical properties that do not. Dispositional essentialism holds that some properties are essentially dispositional. After elaborating on this doctrine, Lam considers two challenges to it. The first challenge is that all properties are fundamentally categorical. This has inspired some to posit a hybrid view on which a property may be both essentially dispositional and categorial. Lam discusses some problems facing this hybrid view. The second challenge is that developments in quantum mechanics may require rethinking our very conception of dispositions. Chapter 14, Antonella Mallozzi’s “Epistemology of Essence” surveys several prominent approaches to the epistemology of essence. Addressing its epistemology is not only of intrinsic interest, but important as a rejoinder to anti-essentialists. This is because some of the most prominent skeptical challenges to essence derive from doubts about their knowability. Those who adopt a modal conception of essence may wish to subsume its epistemology under the epistemology of modality. But those who adopt a non-modal conception of essence would, it seems, have to pursue another approach. Mallozzi considers several prominent approaches in the literature, focusing especially on work by Kripke, Lowe, Hale, Oderberg, Elder, and Kment. Chapter 15, Katherine Ritchie’s “Language of Essence” considers how linguistic and mental representations may sometimes suggest, impose, or reveal, essentialist constraints on their subject matter. Ritchie’s chapter focuses on four prominent cases. First, generic generalizations, such as ‘Birds fly’, often have essentialist connotations; in this case, that birds essentially fly. Second, natural kind terms, such as ‘water’ and ‘tiger’, are, post-Kripke and Putnam, increasingly thought to have essentialist implications. Third, slurs and derogatory terms are often taken to insinuate the inevitability or essentiality of certain negative stereotypes. Fourth, and more generally, some have thought that nouns almost always have essentialist implications for the categories they express. Chapter 16, Jon Litland’s “Logic of Essence” surveys some of the prominent formal approaches to essence. Most of these approaches were developed in pioneering work by Kit Fine in the 1990s. Some of this work, however, is rather technically demanding. Perhaps because of this, it has remained unfamiliar even to some of those working on essence. Litland’s chapter provides a general and accessible overview of a formal language for essence, as well as a proof theory and semantics for it.

103

7 MODAL CONCEPTIONS OF ESSENCE Alessandro Torza

7.1 Introduction Philosophers traditionally distinguish between having a property essentially rather than accidentally—or being in a certain way essentially rather than accidentally, if property talk is to be avoided. Socrates is essentially human and accidentally Xanthippe’s husband. Water is essentially the substance with chemical structure H2O and accidentally the substance that humans drink to survive. The number 2 is essentially prime and accidentally the number of kings of Sparta at any time during the 5th century BC. The singleton set {Steve Jobs} essentially contains Steve Jobs and accidentally contains a co-founder of Apple Computer. Somewhat more controversially, humans have their actual parents essentially and their actual jobs accidentally. The aforementioned examples suggest that the distinction between a thing’s essential and accidental properties is modal in character. Because Socrates and Xanthippe might never have met, he could have failed to be her husband, but not to be human. Because there might have been no humans on Earth, water could have failed to be the substance that we drink to survive, but not to have chemical structure H2O. And so on and so forth. The present chapter takes the suggestion at face value by considering a number of modal characterizations of the essential/accidental distinction that have been articulated and discussed since the early 20th century, as well as some of the challenges that they face.

7.2

Modality

Modal talk involves expressions such as “necessary”, “possible” and “contingent”, as well as their derivatives. For the purpose of regimenting modal statements, we will employ the language of first-order logic with the addition of two monadic sentential operators: N (it is necessary that) and P (it is possible that). The standard interpretation of the resulting first-order modal language is based on Kripke (1963), which draws on the Leibnizian idea that necessary truth amounts to truth at all worlds, and possible truth amounts to truth at some world. Kripke’s theory has two built-in assumptions. One is ontological: individuals can exist across worlds. The other is semantic: individual constants (proper names) are rigid, i.e., they refer to the same individual across worlds.

DOI: 10.4324/9781003008750-10

105

Alessandro Torza

Truth-at-a-world is defined as expected, except for sentences governed by quantifiers or modal operators: [K ] x is true at world u iff is true of every individual existing at u. [K ] x is true at world u iff is true of some individual existing at u. [KN] N is true at world u iff is true at every world accessible to u. [KP] P is true at world u iff is true at some world accessible to u. Accessibility is intended as a relation of relative possibility: for v to be accessible to u is for v to be possible from u’s “viewpoint”. Unless stated otherwise, it will be assumed that the relation is universal (i.e., that every world is accessible to every world), so that N captures absolute necessity. Under this assumption, the accessibility relation may be omitted.1 Once truth simpliciter is identified with truth at the actual world, Kripke semantics will output biconditionals such as the following:

Hx)) is true iff, at every world, all a “Necessarily, all Greeks are human” (i.e., N x(Gx Greeks are human. b “Aristotle could have been Spartan” (i.e., PSa) is true iff, at some world, Aristotle is Spartan. c “Some possible cat is necessarily a mammal” (i.e., P x (Cx NMx)) is true iff, at some world, there is a cat such that, at every world, it is a mammal. A crucial distinction is the one between de dicto statements, which say of a proposition (dictum) that it is necessary/possible, and de re statements, which say of an individual (res) that it has some property necessarily/possibly. The distinction may be characterized syntactically in the following manner: a statement is de re if either it contains an individual constant within the scope of a modal operator, or it quantifies into the scope of a modal operator; otherwise it is de dicto. Thus, “Necessarily, all Greeks are human” is de dicto insofar as it ascribes necessity to the proposition that all Greeks are human. On the other hand, “Aristotle could have been Spartan” is de re since it ascribes the property of possibly being Spartan to Aristotle. “Some possible cat is necessarily a mammal” is also de re in that it ascribes the property of necessarily being a mammal to some possible cat. Let us say that a predicate “P” is de re if the application of “P” to an individual constant produces a de re statement. Modal conceptions of essence share the core idea that something is essentially so and so just in case it satisfies a suitable de re predicate. The way such a predicate is specified is what distinguishes the various modal conceptions of essence.

7.3

Classical Modalism

The modal conception of essence with the longest pedigree, classical modalism, is also the conceptually most parsimonious: it characterizes the meaning of the predicate “is essentially P” purely in terms of “P” and the vocabulary of standard quantified modal logic. Here are four such characterizations, of increasing logical strength, of what it is for something t to be essentially P: CONDITIONAL UNCONDITIONAL WEAK BARCAN STRONG BARCAN

N( x (x = t) Pt).2 NPt . NPt x~NPx. NPt x (Px ~NPx).

106

Modal Conceptions of Essence

The first two characterizations are far and away the most popular in literature from the last 100 years (Forbes 1985; Kripke 1980; Lewis 1968, 1986; Moore 1920; Paul 2006; Plantinga 1974; Quine 1953), although they can be found either directly or indirectly throughout the history of philosophy arguably as far back as Aristotle’s Topics (102b6–7) and Metaphysics (1019a1–4), cf. the chapters Origin Essentialism and Essence: Ancient for discussion. When interpreted in terms of Kripke semantics, CONDITIONAL says that Socrates is essentially a human iff he is a human at every world where he exists, whereas UNCONDITIONAL says that Socrates is essentially a human iff he is a human at every world. Choosing CONDITIONAL over UNCONDITIONAL comes down to whether one accepts (Hanson 2018; Plantinga 1985, Stephanou 2007) or rejects (Fine 1985; Kripke 1963; Pollock 1985) the thesis of serious actualism: that something can have a property only if it exists. For UNCONDITIONAL and serious actualism jointly entail that Socrates is essentially human only if he necessarily exists, which runs against the deep-seated intuition that humans are contingent beings. Thus, the essentialist who adopts UNCONDITIONAL is going to either reject serious actualism, or abandon intuition and embrace necessitism, the radical view that necessarily everything necessarily exists (Linsky and Zalta 1994; Williamson 2013). Now, call essentialism the thesis that something is essentially P, for some predicate “P”. There is a worry that neither CONDITIONAL nor UNCONDITIONAL is able to capture essentialism as the substantive metaphysical thesis motivated by such examples as the ones from sec. 1. In order to see this, consider the following predicates (read “x y” as “x is a part of y”, i.e., “either x is a proper part of y or x is identical with y”):3

P1 z =def

x (x = z).

P2 z =def z = z. P3 z =def z = t . P4 z =def z

t.

x (x = t)), Because the result of substituting “P1” for “P ” in CONDITIONAL is N( x (x = t) which is a truth of modal logic, CONDITIONAL-essentialism turns out to be trivially true. Likewise, substituting “P2” for “P ” in UNCONDITIONAL yields the logical truth N(t = t), hence UNCONDITIONALessentialism is also trivially true. To avoid this result, some have looked for stronger characterizations of essentialism (Marcus 1967; Parsons 1967). For example, WEAK BARCAN is not made automatically true by either “P1” or “P2”, since it requires of an essential property that it not be shared by all things. Nevertheless, WEAK BARCAN-essentialism is nearly trivialized insofar as x ~ N (x = t), which is true just in case t is not the only existent. “P3” yields N (t = t) On the other hand, STRONG BARCAN appears to be immune to (near) trivialization, since it is not satisfied by any of “P1”, “P2” or “P3”. A candidate satisfier of STRONG BARCAN-essentialism is the predicate “P4”. For on the common-sense assumption that objects may gain and lose parts, if t is x (x t ~N (x t)) is not a mereological simple then the substitution instance N (t t) true: the first conjunct follows from N(t = t), whereas the second conjunct is true insofar as t has at least some of its parts contingently. However, “P4” does not (nearly) automatically satisfy STRONG BARCAN, since it is not a truth of modal logic that parthood is a contingent relation—a thesis that can be, and has been rejected for at least some classes of objects (Chisholm 1973). The issue with STRONG BARCAN is that it is arguably too strong. For the essentialist wants to be able to assert that 2 is essentially prime. Given STRONG BARCAN, we get that 2 is necessarily prime and something is contingently prime. Yet, we would be hard pressed to find a prime

107

Alessandro Torza

number that could be the product of two smaller natural numbers. The underlying issue is that STRONG BARCAN flies in the face of the intuition that the property of being prime is necessary to anything that has it. A modal characterization that aims to be neither trivial nor too strong is the Discrimination-Based Account of essence, cf. (De 2020).4 On this view, t is essentially P iff (i) t is CONDITIONAL-essentially P, and (ii) there is no property X that is CONDITIONAL-essential to everything, and such that t’s being P is identical with t’s being X. If quantification into predicate position is allowed, and ‘“ ” expresses higher-order identity, the condition can be expressed thus: DBA

N ( x (x = t)

Pt)

~

X ( y N ( x (x = y)

Xy)

(Pt

Xt)).

This account is not trivialized by the aforementioned predicates, and it does not run into the problem affecting STRONG BARCAN. Nevertheless, DBA still appears to be too strong insofar as it is in tension with the conjunction of two generally accepted theses: 1 that having a property essentially is a noncontingent matter; 2 that necessity is noncontingent, as captured by the axiom schema 4, i.e. N

NN .5

For suppose that something t is DBA-essentially P in actuality, and let w be a possible world Pt) and so, since nothing whose only denizen is t. By 2, it is true at w that N( x (x = t) Py). It follows that it is true at w that exists at w other than t, that y N( x (x = y) Xy) (Pt Xt)). y N( x (x = y) Py) (Pt Pt), and so that X ( y N ( x (x = y) Hence, t is not DBA-essentially P at w, against 1.6 A related issue is that the right conjunct of DBA makes essence extrinsic. For example, a sample of water t has structure H2O DBA-essentially in actuality, yet it will not have it DBAessentially at any world where t is the only object. Yet some essential properties are intrinsic, including water’s microphysical structure (see Correia 2007; Denby 2014), as well as the section Hybrid Modalism in this chapter. Similarly, but without appealing to the axiom schema 4, it can be shown that WEAK BARCAN and STRONG BARCAN are also in tension with the necessity of essentiality. Because WEAK BARCAN, STRONG BARCAN and DBA do not fill their intended theoretical role, in the remainder of the chapter the discussion of classical modal views will be restricted to CONDITIONAL and UNCONDITIONAL, unless stated otherwise. It remains unclear whether there exists a suitable characterization of essentialism that includes the metaphysically substantive cases while excluding properties had as a matter of logic.7

7.4

Identity Across Worlds

The very intelligibility of de re predication, and so of a modal distinction between essential and accidental properties was subject to an influential skeptical challenge due to Quine (1953), although the technical and philosophical work of Ruth Barcan Marcus and Saul Kripke among others went a long way toward rehabilitating de re modality. (See the chapter Quine on Essence in this volume.) In particular, Kripke (1980) claims that “The question of essential properties so-called is […] equivalent to the question of ‘identity across possible worlds’,” (p. 42) by which he means that UNCONDITIONAL-essentialism is intelligible just in case something exists at multiple worlds. The reasoning behind the claim may be reconstructed thus: if a statement of the form NPt is truth-apt then, in virtue of its truth condition, we are committed to crossworld identities;

108

Modal Conceptions of Essence

conversely, if an individual t is one and the same across worlds, then it makes sense to ask about any predicate whether t satisfies it at each world, that is, necessarily. Insofar as identity—the relation that everything has with itself and nothing else—is “unproblematic” (p. 49), it follows that the intelligibility of essentialism and de re predication is likewise unproblematic, cf. (Plantinga 1973). This line of argument has been resisted on multiple fronts, although less by modal skeptics than by philosophers who think that Kripke’s metaphysical picture is either incomplete or incorrect. Some have indeed argued that the intelligibility of essentialism does not entail the existence of crossworld individuals. In particular, Lewis (1986) has defended genuine modal realism, the view that alternative worlds are just as concrete as the actual one, and pairwise disjoint. Because the individuals of this pluriverse are worldbound, Lewis specifies the truth conditions of modal statements in terms of a counterpart relation, rather than identity: the counterparts of something t at world w are the denizens of w that most resemble t.8 According to counterpart theory (Lewis 1968), “Socrates is necessarily a mammal” is true iff, for every world w and every w-counterpart Socrates* of Socrates, Socrates* is a mammal. Due to its greater generality compared to Kripke semantics, counterpart theory has been defended in some form or another also by philosophers who accept identity across worlds and reject genuine modal realism (Fara 2008; Forbes 1985; Hazen 1979; Stalnaker 2003; Varzi 2020; Wang 2015). The possibility of interpreting first-order modal language in counterpart-theoretic terms shows that the intelligibility of de re predication in general, and of a modal conception of essence in particular, can be decoupled from an ontology of crossworld individuals. Kripke (1980, p. 45 fn. 13) dismissed counterpart theory on the grounds that it incorrectly entails that “if we say ‘Humphrey might have won the election’ … we are not talking about something that might have happened to Humphrey but to someone else, a ‘counterpart’”. As Lewis (1986, p. 198) points out, however, the condition “x has a counterpart that won the election” is counterpart-theoretically equivalent with “x might have won the election”, which is ascribed to Humphrey himself, not to his counterpart. Furthermore, under certain assumptions the counterpart theorist is even able to simulate an ontology of crossworld individuals in terms of counterfactual identity. For every world w and w-counterpart Socrates* of Socrates, Socrates is counterfactually identical with Socrates* iff, had w been actual, Socrates would have been Socrates* (Torza 2012). The counterpart theorist can counter Kripke’s animadversions by (i) casting the intuitions about identity across worlds in terms of counterfactual identity, and (ii) arguing that counterfactual identity validates the same modal-logical theses (Leibniz’s law, necessity of identity/distinctness, etc.) as strict identity. An alternative avenue of resistance to Kripke’s argument for the intelligibility of essentialism involves the idea that crossworld identity is not unproblematic insofar as it requires criteria of identification. These may be either epistemic criteria allowing us to know which object is which across worlds (Hintikka 1970), or qualitative individual essences grounding the identity of objects across worlds (Chisholm 1967). Kripke (1980, p. 44) famously rejected criteria of identification across the board, arguing that worlds with built-in identities can be introduced by stipulation prior to any epistemic or metaphysical considerations. (See the chapters Identity, Persistence, and Individuation, and The Epistemology of Essence.)

7.5 De Dicto Essentialism We have so far investigated the question of how to best characterize expressions of the form “t is essentially P”, and it turned out that de re modality appears to provide a natural way of

109

Alessandro Torza

doing so. The literature however treats as essentialist claims that do not have that form. Consider the following (Kripke 1980; Putnam 1973, 1975): 1 Necessarily, water is H2O. 2 Necessarily, cats are mammals. Both are regarded as essentialist claims: 1 says something about the essence of water, namely that its microphysical structure is H2O, whereas 2 says something about the essence of cats, namely that they belong to the class Mammalia (cf. the chapter Natural Kind Essentialism.) On the face of it, 1 is an instance of UNCONDITIONAL-essentialism, since it may be regimented as the de re statement (7.1)

N(w = h),

provided that “w” and “h” are singular terms for the water kind and the H2O kind, respectively. However, the intended reading of 1 is typically about not the water kind, but water samples, that is

N

x (Wx

(7.2)

Hx),

where the predicates “W” and “H” pick out the water property and the H2O property, respectively. Being de dicto, 7.2 does not ascribe a modal property to any individual, and so it is not an essentialist statement in the sense discussed so far. The point is even more straightforward in the case of 2 given its standard first-order regimentation

N

x (Cx

(7.3)

Mx),

where the predicates “C” and “M” pick out the cat property and the mammal property, respectively. One might try to turn a statement like 7.2 into a de re statement by quantifying into predicate position, so as to express the idea that water and H2O are properties such that having the former necessarily implies having the latter:

X

Y (X

W

Y

H

N

x (Xx

Yx)).

(7.4)

However, all 7.4 says is that the actual cats are necessarily a subset of the actual mammals, which is not the intended meaning of 1, nor is it equivalent with 7.2. In order to comply with usage, Mackie (2006, p. 13) suggests instead that a statement be regarded as essentialist just in case it is either de re essentialist (viz., UNCONDITIONAL/ CONDITIONAL-essentialist) or it expresses an a posteriori de dicto necessity. Indeed, statements like 7.2 and 7.3 are quintessential cases of a posteriori truths, that is, truths that can only be known on the basis of empirical evidence. One problem with Mackie’s characterization is that it turns essentialism into a disjunctive phenomenon, in sharp contrast to what transpires from the literature, which by and large regards essence as a unified feature of reality. A non-disjunctive characterization is forthcoming if one pays attention to the way a posteriori de dicto necessities are justified. As argued in (Salmon 1981, pp. 166–7), our belief that, necessarily, water is H2O (i.e., eq. 7.2) may be justified as follows, where “Sxy” stands for “x has the same microphysical structure as y”:

110

Modal Conceptions of Essence

a t is a sample of H2O (i.e., Ht). Sxt)). b Necessarily, water has the same microphysical structure as t (i.e., N x (Wx c If something is a sample of H2O then, necessarily, what has the same microphysical N y (Syx Hy))). structure as that is also a sample of H2O (i.e., x (Hx Hx)). d Therefore, necessarily, water is H2O (i.e., N x (Wx The conclusion (d) is validly inferred from a bit of empirical data (a) plus two de re modal statements (b, c). In particular, b states that it is UNCONDITIONAL-essential to t that it have the same microphysical structure as any water sample. Thus, the justification of the original de dicto necessity (7.2) involves essentialism in the standard de re sense. This observation prompts the sketch of a characterization covering both de dicto and de re essentialism: a statement is essentialist iff it is a logical consequence of the conjunction of de re essentialist statements and a possibly empty set of non-de re statements.

7.6 Fine’s Objection to Classical Modalism Aside from early skepticism about de re modality due to Quine, classical modalism established itself as the standard conception of essence for nearly three decades, until Fine (1994) formulated an objection that single-handedly upended the status quo.9 In order to streamline the discussion, let us focus on a particular characterization, namely CONDITIONAL. As will soon become clear, Fine’s point carries over to a broad class of modal characterizations. The main claim is that CONDITIONAL fails to provide a sufficient condition for the ascription of essence: something can be necessarily P if existent, without being essentially P. (Fine does not take issue with the necessity side of CONDITIONAL, see the chapter Non-Modal Conceptions of Essence). The structure of the argument is remarkably simple. The main premise is that essence plays a definitional role: to define the nature of an object (equivalently: to define what an object is) amounts to providing a complete list of that object’s essential properties. Not only has this idea a pedigree spanning the history of philosophy, from Aristotle (Metaphysics, 1031a1) to Kripke (1980, pp. 124, 138)—it is also supported by the examples we have employed so far: it is in the nature of Socrates to be human; it is in the nature of water to have microphysical structure H2O; it is in the nature of 2 to be prime; it is in the nature of {Steve Jobs} to have Steve Jobs as a member. This premise is captured by the condition [F1] If t is essentially P, then it belongs to the nature of t that it is P.10 Now, consider the following three predicates: P4 z: = def z is such that there are infinitely many prime numbers; P5 z = def z is distinct from the Eiffel Tower; P6 z = def z is a member of {Socrates}. The second premise of the argument is that none of the properties expressed by the above three predicates belongs to the nature of Socrates. The claim is intuitively compelling: were we to spell out what it is for something to be Socrates, we might invoke facts about his DNA, or about the identity of his parents, but we would not invoke facts about prime numbers, or the Eiffel Tower, or set membership. By F1, it follows that [F2] none of the properties expressed by P4–P6 are essential to Socrates.

111

Alessandro Torza

The third premise is that the properties expressed by P4–P6 are necessarily had by Socrates if he exists. The claim is hardly controversial. Because it is necessarily the case that there are infinitely many prime numbers, Socrates is such that there are infinitely many prime numbers at any world where he is something.11 Because distinctness is a necessary relation, Socrates is not identical with the Eiffel Tower at any world where he is something.12 And because set membership is a necessary relation, Socrates is a member of his own singleton at any world where he is something.13 By CONDITIONAL, it follows that [F3] each of the properties expressed by P4–P6 is essential to Socrates, which contradicts F2. Fine concludes that, if we wish to restore consistency, CONDITIONAL ought to be abandoned: in order for something t to have a property essentially, it does not suffice that t has that property necessarily if t exists. CONDITIONAL’s failure to capture essence has a precise diagnosis. The fact that, say, Socrates is a member of {Socrates} at all worlds is necessary because of the nature of set membership, not because of the nature of Socrates. To put it slightly differently, it is essential to {Socrates} that Socrates is a member of {Socrates} and, because of that, it is necessarily the case that Socrates is a member of {Socrates}if Socrates exists. But it is not essential to Socrates that Socrates is a member of {Socrates}. Thus, whereas the essentiality of membership is asymmetric, the necessity of membership is not, cf. (Dunn 1990, pp. 89–91). Necessity, and modality in general, is not sufficiently fine-grained to represent facts about essence. This diagnosis leads to a generalization of Fine’s objection that undercuts not just CONDITIONAL, but classical modalism across the board. As shown in Torza (2015), there is no way to define the meaning of “t is essentially P” purely in terms of “P” and the vocabulary of standard quantified modal logic, whether first-order or higher-order, compatibly with the following conditions: i Socrates is not essentially a member of {Socrates}; ii {Socrates} essentially has Socrates as a member. Some have reacted by casting doubt on the intuitions leading to F2, and so to the asymmetry of membership essentiality (i–ii). Livingstone-Banks (2017) sets up a dilemma: either such intuitions are pre-philosophical, in which case they lack epistemic weight14; or they are philosophical, and so presuppose some alternative, nonmodal conception of essence, which makes them question-begging in an argument against classical modalism. He argues that the value of such intuitions is not in debunking classical modalism, but in motivating a definitional conception of essence; and that choosing this over the modal conception should be the result of systematic considerations about their respective theoretical virtues, rather than considerations about particular cases. Cowling (2013) has pushed back against F1, arguing that the essence-nature link is negotiable. In support of that claim, he advances an alternative view that identifies the nature of an object with its sparse properties, i.e., the ones that characterize the world “completely and without redundancy” (Lewis 1986, p. 60). Sparseness plays a number of key metaphysical roles such as making for qualitative similarity, and determining facts about laws and causation. On the Lewisian picture, sparse properties are the maximally specific properties invoked by fundamental physics: having a determinate value of mass, and having a determinate value of spin along a given direction are sparse; being green, and being the sum of two primes are not.

112

Modal Conceptions of Essence

Sparseness and essentiality are orthogonal: Socrates’s quantity of mass is sparse, though accidental to him, whereas humanity is essential to Socrates, though not sparse. By identifying a thing’s nature with the sparse properties that thing instantiates, claims about Socrates’s nature will fail to translate to claims about his essence, thus blocking Fine’s reductio.15 Pockets of resistance notwithstanding, there is now near-universal consensus that classical modalism is materially inadequate.16 The remainder of the chapter is devoted to modal conceptions of essence that aim to meet the Finean challenge. (Alternative strategies are discussed in the chapter Non-Modal Conceptions of Essence.)

7.7 Sophisticated Modalism Efforts have been made to show that a purely modal characterization of essence is able to accommodate Fine’s cases, as long as the modality involved is nonstandard. We will refer to such a strategy as sophisticated modalism. Although the distinction between standard and nonstandard modal notions is somewhat fuzzy, the two most prominent post-Finean purely modal conceptions of essence (to be discussed shortly) employ notions whose intended interpretation appeals to impossible worlds, which are not part of the standard toolbox of the semantics for modal logic. The least that can be said about impossible worlds is that they are ways the world cannot be. This fact leaves open which impossible worlds there are. On the strictest view, there are none. On the most liberal view, each way the world cannot be is an impossible world. As will soon become clear, in order for impossibilia to make room for a reply to Fine’s objection, one must go for an intermediate view.17 One sophisticated modal conception involves counterfactuals, that is, conditionals of the form if it was the case that then it would be the case that , in symbols:

> . The standard semantics for counterfactuals yields the following truth condition (Lewis 1973; Stalnaker 1968): [SL>] > is true at world u iff the possible -worlds that are closest to u are -worlds.18 Closeness is a binary relation on worlds which, on its intended interpretation, is defined by qualitative similarity: closer worlds are more similar. (The relevant criteria of similarity are specified by the context of utterance.) Thus, “if kangaroos had no tail, they would topple over” is true in actuality just in case the possible worlds more (relevantly) similar to ours where kangaroos have no tail are worlds where kangaroos topple over. Brogaard and Salerno (2013) favor the following nonstandard truth condition instead: [BS>] > is true at world u iff the possible or impossible -worlds that are closest to u are -worlds, provided that closeness is a binary relation on the class of all worlds, whether possible or impossible. The main rationale they offer for adopting BS> concerns counterpossibles, i.e., counterfactuals with impossible antecedents. SL> makes all counterpossibles, and so both of the following statements, vacuously true: i If arithmetic was complete and consistent, it would prove its own consistency; ii If arithmetic was complete and consistent, the moon would be made of cheese.

113

Alessandro Torza

However, one may object that, on a fairly natural way of reasoning about impossible scenarios, (i) is true and (ii) false. BS> offers a solution. If there are impossible worlds where arithmetic is both complete and consistent, the relevant closeness relation is such that the closest worlds where arithmetic is complete and consistent are worlds where arithmetic proves its own consistency, but the moon is not made of cheese. If that much is granted, BS> will make (i) true and (ii) false as desired. Brogaard and Salerno argue that BS> can help define essence in a way that accommodates Fine’s cases, given suitable assumptions about the closeness relation. On their characterization, t is essentially P iff t is CONDITIONAL-essentially P, and if nothing were P then t would not exist—that is: COUNTERFACTUAL

N( x (x = t)

Pt)

(~

xPx > ~

x (x = t)).

Now consider again the predicate P4 z =def z is such that there are infinitely many prime numbers and assume that there are impossible worlds at which there are at most finitely many primes. Because the closest such worlds need not be worlds where Socrates fails to exist, ~ xP4 x > ~ x (x = Socrates) is false. Thus, Socrates is not COUNTERFACTUAL-essentially such that there are infinitely many primes. Let us turn to the predicate P6 z =def z is a member of {Socrates}. Assume that (A1) there are impossible worlds where Socrates does and {Socrates} does not exist; and that (A2) at every possible or impossible world, {Socrates} exists iff Socrates is a member of {Socrates}. Pick a closeness relation such that (A3) some impossible world where Socrates does and {Socrates} does not exist is at least as close as any possible world where Socrates (and so {Socrates}) does not exist. A1-3 guarantee that among the closest worlds where nothing is a member of {Socrates}, and so Socrates is not a member of {Socrates}, is some impossible world where Socrates exists. Hence, ~ x (x {Socrates}) > ~ x (x = Socrates) is false, and Socrates is not COUNTERFACTUAL-essentially a member of {Socrates}. By the same lights, {Socrates} COUNTERFACTUAL-essentially has Socrates as a member. Socrates {Socrates}) is true. Moreover, For, as we already know, N( x (x = {Socrates}) the closet worlds where nothing has Socrates as a member, and so Socrates is not a member of {Socrates} are, by A2, worlds where {Socrates} does not exist. That is, ~ x (Socrates x) > ~ x (x = {Socrates}) is true. It can be concluded that, given A1-3, COUNTERFACTUAL entails the asymmetry of membership essentiality. The Brogaard-Salerno route ultimately appears to be inadequate, however. First, it is not clear that a closeness relation satisfying A3 is the relevant one, since it flouts the thesis that every possible world is closer than any impossible world (Mares 1997), as well as the thesis that similarity in matters of (e.g., set-theoretic) laws has more weight than similarity in matters of particular fact (Lewis 1979; Steward 2015). Second, consider the predicate P5 z =def z is distinct from the Eiffel Tower. In order for Socrates not to be COUNTERFACTUAL-essentially distinct from the Eiffel Tower, ~ x~(x = Eiffel Tower) > ~ x (x = Socrates) needs to be false, which in turn requires that some of the closest worlds at which the Eiffel Tower is all there is are worlds where Socrates exists. However, at such worlds the Eiffel Tower is identical with Socrates, a scenario that is

114

Modal Conceptions of Essence

arguably less, not more similar to actuality than one where the Eiffel Tower does and Socrates does not exist (Steward 2015). Third, and most importantly, COUNTERFACTUAL fails with respect to a close variant of Fine’s Socrates example, namely when modalized properties are involved (Torza 2015). If Socrates is not essentially a member of {Socrates} because his essence does not involve sets, a fortiori he is not essentially such that necessarily he is a member of {Socrates} if he exists. However, iii N ( x (x = Socrates)

N ( x (x = Socrates)

Socrates

{Socrates}))

is logically equivalent with N( x (x = Socrates) be true. Moreover, the counterfactual

Socrates

{Socrates}), which we know to

iv ~

x N ( y (y = x)

x

{Socrates}) > ~

x (x = Socrates)

is also true. For, at every possible or impossible world, something satisfies the predicate N( y (y = x) x {Socrates}) iff it is Socrates. So, (iv) is equivalent with v ~

x (x = Socrates) > ~

x (x = Socrates)

which is trivially true. Therefore, Socrates is COUNTERFACTUAL-essentially such that necessarily he is a member of {Socrates} if he exists, which flies in the face of the Finean wisdom. An alternative sophisticated modal conception of essence, drawing on (Prior 1957), is articulated and defended in Correia (2007). The proposal rests on a nonstandard semantics for modal discourse with three key ingredients: Aboutness. Sentence at u.19

is truth-evaluable at world u iff the objects that

is about exist

A sentence is about the objects referred to by individual terms occurring in it. Accordingly, “Plato is a student of Aristotle” is about Plato and Aristotle, whereas “There are prime numbers” is about nothing. Also notice that “Socrates exists” and “Socrates is Socrates” are not truth-evaluable at a world where Socrates does not exist. Priorean conditional. at every world where

Prior-implies ( is truth-evaluable.

) iff the material conditional

is true

The Priorean conditional behaves nonclassically. Suppose that Phaenarete is Socrates’ mother at all worlds where Socrates exists, and that Phaenarete exists at some world where Socrates does not. Then, “Socrates exists” Prior-implies “Phaenarete is a mother”. However, “Phaenarete is not a mother” does not Prior-imply “Socrates does not exist”, since there are worlds where Phaenarete exists but the material conditional “if Phaenarete is not a mother then Socrates does not exist” is not truth-evaluable, and so not true. Worlds. The intended interpretation of the present semantics involves both possible and impossible worlds. In particular, it is assumed that there are worlds where Socrates does but {Socrates} does not exist, as well as worlds where there are at most finitely many primes.

115

Alessandro Torza

With that being said, Correia puts forward the following characterization: t is essentially P iff PRIOR

x (x = t)

Pt .

So, t is PRIOR-essentially P just in case t is P at every possible or impossible world where t exists. PRIOR-essentialism is what one obtains from CONDITIONAL-essentialism by replacing the )) with the Priorean conditional. strict conditional (i.e., N( Given suitable assumptions about the nature of (im)possible worlds, PRIOR is able to accommodate Fine’s three examples against classical modalism. Assuming that there are impossible worlds such that Socrates exists but there are at most finitely many primes, Socrates is not PRIOR-essentially such that there are infinitely many primes. Since there are possible worlds where Socrates does and the Eiffel Tower does not exist, Socrates is not PRIORessentially distinct from the Eiffel Tower. Assuming that there are impossible worlds where Socrates does and {Socrates} does not exist, Socrates is not PRIOR-essentially a member of {Socrates}; and assuming that every possible or impossible world where {Socrates} exists is such that Socrates is a member of {Socrates}, {Socrates} has PRIOR-essentially Socrates as a member. By not involving a relation of closeness between worlds, PRIOR avoids the first two difficulties that were leveled against COUNTERFACTUAL. What PRIOR is unable to avoid is the third difficulty involving modal properties. Indeed, every possible or impossible world u where Socrates exists is such that “if Socrates exists then necessarily there are infinitely many prime numbers” is truth-evaluable, and indeed true at u. It follows that Socrates is PRIORessentially such that necessarily there are infinitely many prime numbers.20 But if the existence of numbers is irrelevant to Socrates’s essence, so is the necessary existence of numbers. Hence, Correia’s brand of sophisticated modalism fails to meet Fine’s challenge.

7.8

Hybrid Modalism

An alternative approach to characterizing essence compatibly with Fine’s remarks employs a combination of both modal and nonmodal resources, which we may therefore call hybrid modalism. Strategies of this sort differ by which nonmodal notions are brought to bear on the issue. Sparse modalism (Wildman 2013) characterizes essence by means of standard modal logic augmented with the notion of sparseness. Like Cowling (2013), Wildman identifies the nature of an object with the sparse properties it instantiates; unlike Cowling, he accepts the link between nature and essence (F1). Moreover, Wildman agrees with the consensus view that something is essentially some way only if cannot exist without being that way. The upshot is that something t is essentially P iff SPARSE

t is

CONDITIONAL-essentially

P, and P is sparse.

The resulting view immediately appears to be overly restrictive. Water is essentially H2O, yet the property of being H2O is not invoked by fundamental physics, hence it is not sparse on the standard, Lewisian sense. To overcome the objection, the intended reading of SPARSE employs a looser notion of sparseness, articulated in Schaffer (2004), such that for a property to be sparse is for it to be invoked by some scientific theory, whether fundamental or not. Thus, such properties as being H2O, and being vertebrate are Schaffer-sparse, though not Lewis-sparse.

116

Modal Conceptions of Essence

Sparse modalism so understood easily deals with some of Fine’s cases. The property of being such that there are infinitely many primes, and the property of being distinct from the Eiffel Tower are not sparse, hence not SPARSE-essential to Socrates. On the other hand, the view runs into trouble with the asymmetry of membership essentiality. If set theory makes reference to any sparse property, the only plausible candidate is the one picked out by the theory’s only primitive, namely the membership relation. So, both the property of being a member of {Socrates} and the property of having Socrates as a member are not sparse. It is therefore neither SPARSE-essential to Socrates that he is a member of {Socrates}, nor SPARSEessential to {Socrates} that it has Socrates as a member. Wildman thinks that the case of relational properties is best handled by a special-purpose clause such that s bears relation R essentially to t iff SPARSEREL

s is

CONDITIONAL-essentially

such that it bears R to t, and R is sparse.

(The condition generalizes to relations of higher adicity in the obvious fashion.) But SPARSEREL is no improvement over SPARSE. For if membership is indeed a sparse relation, and assuming the standard thesis that membership facts are necessary conditionally upon the members’ existence, then it is SPARSEREL-essential to Socrates that he is a member of {Socrates}, as well as SPARSEREL-essential to {Socrates} that it has Socrates as a member. Wildman ultimately bites the bullet and defends a metaphysics of sets that makes all membership facts essential, thus rejecting the asymmetry of membership essentiality. Such a move is problematic on two counts. First, sparse modalism is explicitly motivated as a way of meeting the Finean challenge. Insofar as it fails to accommodate what is arguably the hardest part of that challenge, the view is dialectically unstable. Second, the asymmetry of essentiality can also be detected in relations other than membership, which renders Wildman’s views about the metaphysics of sets beside the point. Consider the prima facie plausible view that in order to define natural numbers it suffices to state the axioms of (second-order) Peano arithmetic, whereas the negative integers are defined out of the natural numbers by means of further axioms, e.g., that every n is such that n + ( n) = 0. It is then essential to 1, but not to 1, that 1 + ( 1) = 0. Consider now the dyadic relation of adding up to zero. If that relation is sparse, and given the necessity of mathematical facts, then it is both SPARSEREL-essential to 1 that 1 + ( 1) = 0, and SPARSERELessential to 1 that 1 + ( 1) = 0; if the relation is not sparse, then it is neither SPARSERELessential to 1 that 1 + ( 1) = 0, nor SPARSEREL-essential to 1 that 1 + ( 1) = 0. Either way, sparse modalism fails to predict the observed asymmetry of essentiality. In order to capture such asymmetries, de Melo (2019) introduces the notion of slot-relative sparseness.21 The fact that Socrates is a member of {Socrates} tells us everything about the way {Socrates} is, but nothing about the way Socrates is. The fact that Anna loves Bob tells us something significant about the way Anna is, but very little about the way Bob is. These examples suggest that membership is sparse relative to the second but not the first slot, and that the loving relation is sparser relative to the first than the second slot. Construing sparseness as a relation between a property and a slot makes room for a more fine-grained characterization of essence in the case of relational properties. According to locally sparse modalism, s bears relation R essentially to t iff L-SPARSEREL

s is CONDITIONAL-essentially such that it bears R to t, and R is sparse relative to the slot occupied by s.

117

Alessandro Torza

When suitable assumptions are in place, it follows that {Socrates} is L-SPARSEREL-essentially such that it has Socrates as a member, whereas Socrates is not L-SPARSEREL-essentially a member of {Socrates}. So far, so good. However, locally sparse modalism is bound to break down whenever applied to relations that, unlike membership, are not asymmetric. For if 1 is L-SPARSERELessentially such that 1 + ( 1) = 0, the relation of adding up to zero must be sparse relative to its second slot. From this and the necessary fact that 1 + 1 = 0 it follows that 1 is L-SPARSERELessentially such that 1 + 1 = 0, against the aforementioned remark that the essence of natural numbers is exhausted by the axioms of Peano arithmetic, and so does not involve negative integers. The reason why locally sparse modalism flounders is that essentiality is a function of the relata, not of their position in a relation. As a consequence, the asymmetry of essentiality can be observed in asymmetric, as well as not asymmetric relations. In order to save the spirit of de Melo’s proposal, sparseness should therefore be construed as a relation between a property and an object, not between a property and a slot. Whether such a view is tenable, and whether it can be motivated on independent grounds are yet unexplored questions.22 Alternatively, instead of engineering a new notion of sparseness, one might look for some off-the-shelf relation between properties and objects that will play the suitable role. This route has been undertaken by Bovey (2021), who puts forward a characterization of essence in terms of standard modal logic plus a relational notion of intrinsicality.23 If something t is P, then t is intrinsically P if it is P in virtue of the way it is on its own; and it is extrinsically P if it is not intrinsically P, cf. (Bader 2013).24 For example, every massy object is intrinsically massy; every unaccompanied object is extrinsically unaccompanied; and every massy object is intrinsically such that something is massy, whereas every nonmassy object is extrinsically such that something is massy. Intrinsic modalism is the view that t is essentially P iff INTRINSIC

t is both

CONDITIONAL-essentially

P and intrinsically P.

The rationale behind the proposal is that part of what makes a property essential to a thing t is that the property is had by t no matter what everything else is like. The idea happens to track a good deal of our intuitions concerning what is essential rather than accidental: having chemical structure H2O is had by a water sample independently of what surrounds it, whereas being married to Xanthippe is had by Socrates in virtue of what further entities are like, including Xanthippe. The same intuition appears to underlie Fine’s examples. Socrates is extrinsically, and so not INTRINSIC-essentially such that there are infinitely many prime numbers. Likewise, Socrates is extrinsically, and so not INTRINSIC-essentially distinct from the Eiffel Tower. The situation is (once again) trickier when it comes to capturing the asymmetry of membership essentiality. Socrates is a member of {Socrates} in virtue of the way both Socrates and {Socrates} are. Because {Socrates} is not a part of Socrates, Socrates is not a member of {Socrates} in virtue of the way Socrates is on his own. Therefore, Socrates is extrinsically, and so not INTRINSICessentially a member of {Socrates}. Whether Socrates is a member of {Socrates} in virtue of the way {Socrates} is on its own depends on whether Socrates is a part of {Socrates}. On the standard, mereological notion of parthood, the parts of a set are its subsets, hence Socrates is not a part of {Socrates}. {Socrates} will then be extrinsically, and so not INTRINSIC-essentially such that it has Socrates as a member—against the desideratum. On a more liberal notion of parthood, however, the

118

Modal Conceptions of Essence

members of a set too are parts of the set (Fine 2010), which guarantees that the property of having Socrates as a member is had intrinsically by {Socrates}, and so that {Socrates} is INTRINSIC-essentially such that it has Socrates as a member. Thus, with a modicum of fine-tuning, intrinsic modalism is able to accommodate Fine’s examples. Whether the result generalizes is unclear, however. For we would like the proposal, jointly with suitable assumptions, to entail that 1 is, whereas 1 is not INTRINSIC-essentially such that 1 + ( 1) = 0. This in turn requires that 1 + ( 1) = 0 hold in virtue of the way 1 is on its own, and it does not hold in virtue of the way 1 is on its own. As to the former, one might argue as follows. Suppose that (*) if f (m) = n , then m is a part of n. Then, 0 is a part of 1, since 1 is the result of applying the successor operation to 0, and 1 is a part of 1, since 1 is the result of applying the inverse operation to 1. Assuming that the relevant parthood relation is transitive, it follows that the property of being such that 1 + ( 1) = 0 is intrinsic to 1, as desired. However, since 1 = ( 1), (*) implies that 1 is a part of 1. Because 0 is also a part of 1, it follows that 1 + ( 1) = 0 holds in virtue of the way 1 is on its own, and so that 1 is INTRINSICessentially such that 1 + ( 1) = 0, contrary to our desideratum. The same kind of challenge will present itself mutatis mutandis with other asymmetric instances of essentiality. Moreover, intrinsic modalism is straightforwardly incompatible with the essentiality of extrinsic properties. If Kripke (1980) and Salmon (1981) are right, everyone is essentially the offspring of their actual parents, yet the property of being Phaenarete’s son is extrinsic to Socrates (cf. the chapter Origin Essentialism). If Sider (2001) is right, we ought to regard the property of being human as extrinsic to everything that has it: something is human iff it is a maximal human-like object—on pain of falling prey to the problem of the many (Unger 1980). However, the consensus view is that Socrates is essentially human. The issue carries over to other sortal properties (cf. the chapters Identity, Persistence, and Individuation, and Biological Species). The aforementioned examples suggest that intrinsic modalism, while providing a good approximation of the target notion, is overly restrictive. Now, INTRINSIC tells us that, in order for t to be essentially P, t’s being P concerns t on its own. Here is a way of relaxing that condition: in order for t to be essentially P, t’s being P concerns t, or any further things involved in the explanation of t’s existence. A view along these lines is articulated in Rizzo (2022) by construing explanation in terms of the relation of metaphysical ground.25 The resulting view, call it GROUND, accommodates the Finean cases, is compatible with the existence of extrinsic essences, and fares at least as well as intrinsic essentialism with respect to the other examples discussed in this section. Notably, it appears to capture the asymmetry involving the relation of adding up to zero. For, arguably, facts involving 1 ground the existence of 1, and facts involving 0 ground the existence of 1. Ground being transitive, it 1 is GROUND-essentially such that 1 + ( 1) = 0. On the other hand, facts follows that involving 1 do not ground the existence of 1, hence 1 is not GROUND-essentially such that 1 + ( 1) = 0.

7.9 Conclusions While the dominant view throughout the 20th century, classical modalism has been found wanting and then quickly set aside as a consequence of Fine’s critique. Although essence is now widely regarded as more fundamental than necessity, and modality at large, two alternative strategies have been developed in order to meet the Finean challenge: sophisticated modalism, and hybrid modalism.

119

Alessandro Torza

The jury is still out on whether either is able to draw the line between essential and accidental properties as required. If some variant or other of modalism were to prove successful, we could draw a twofold moral. On the one hand, we would have evidence against the currently mainstream view that essence is prior to modality. On the other hand, one should not conclude that Fine’s point is unwarranted. For the gist of it is that the apparatus of standard modal logic is incapable of analyzing essence. So, any attempt at carrying out that analysis by either interpreting modality in a nonstandard fashion, or augmenting it with nonmodal machinery is going to be more grist to Fine’s mill.

7.10

Related Topics

For a discussion of relevant topics the reader may consult the following chapters from this volume: Chapter 1 Essence: Ancient Chapter 8 Non-Modal Conceptions of Essence Chapter 10 Natural Kind Essentialism Chapter 11 Origin Essentialism Chapter 14 Epistemology of Essence Chapter 18 Biological Species Chapter 19 Identity, Persistence, and Individuation Chapter 29 Quine on Essence

Notes 1 See Salmon (1989) for reasons to decouple absolute necessity from a universal accessibility relation. Pt). 2 A further modal definition of essence is sometimes mentioned in the literature, namely N(t = t This is materially equivalent with UNCONDITIONAL if self-identity is necessary, and with CONDITIONAL if self-identity is necessarily equivalent with existence. 3 The following remarks concerning trivialization make the routine assumption that modal logic has the property of normality, and that it preserves all truths of first-order logic. 4 De traces the Discrimination-Based Account of essence back to Della Rocca (1996) and Marcus (1967). 5 See Salmon (1981, pp. 229–52) for an influential line of argument against the modal schema 4; cf. ( Chandler 1976). See Roca-Royes (2011) for discussion. Note that the schema also fails on the counterpart theory of Lewis (1968). 6 The objection from the necessity of essentiality carries over to the account of essence favored by De (2020), which is DBA plus the condition that the property expressed by “P” be qualitative. 7 For an attempt at characterizing nontrivial essence by a combination of modal and nonmodal notions see Coates (2022). 8 Pace Lewis, it is possible to interpret de re predication in terms of identity even if worlds are disjoint ( Varzi 2006), and it is possible to be a modal realist even if worlds are not disjoint ( McDaniel 2004). 9 Fine’s line of argument against classical modalism is anticipated in Almog (1991, p. 232). 10 The converse of F1, although subscribed by Fine, is not required by the argument. 11 This step is warranted on the assumption that (i) for every condition there exists the predicate “x is such that ”, and that (ii) each predicate expresses a property. 12 The necessity of distinctness follows from the conjunction of Leibniz’s law and the axiom schema B N N (i.e., ), which is routinely associated with metaphysical modality. See Dummett (1993) for an argument against B, and Walters (2014) for a rebuttal; cf. ( Salmon 1989) on the logical status of B. 13 Notice that the thesis that set membership is a necessary relation does not follow from standard set theory (viz., ZFU), which is an extensional theory, nor does it follow from the necessitation of

120

Modal Conceptions of Essence

14 15 16 17 18 19 20 21

22 23 24

25

standard set theory, which is a de dicto intensional theory. Rather, it is a substantive de re assumption about the nature of sets. For discussion see ch. 5 of Forbes (1985). But see Kripke (1980, p. 42) on the value of intuition in modal matters. It is unclear whether the objection rests on substantive disagreement, or whether it boils down to an alternative convention on how to use the term “nature” ( Torza 2015). Gorman (2005) argues that essence cannot be captured neither modally nor definitionally. On the logical and philosophical relevance of impossible worlds see Berto and Jago (2013). Some differences between Stalnaker’s and Lewis’s semantics for counterfactuals are being glossed over which, while significant, do not affect the present discussion. Like Prior (1957) and unlike Correia (2007), it is being assumed that to exist at u is to be something at u. Nothing critical of what is said here hangs on that choice. Notice that Correia defines the Priorean conditional only for pairs of unmodalized sentences. In order to discuss predicates like “is necessarily such that there are infinitely many prime numbers”, that limitation has been lifted. According to de Melo, naturalness should also be relativized to kinds of objects, a modification he puts to work in addressing a number of objections to Wildman’s view that are not discussed here, including the ones in Skiles (2015). In the interest of simplicity, relativization to kinds is being set aside. Koslicki (forthcoming) expresses skepticism concerning the very idea of relativizing naturalness. An attempt at defining essence in terms of a nonrelational notion of intrinsicality, carried out in Denby (2014), runs into similar issues as sparse modalism, cf. ( Zylstra 2019). Notice that it is possible for something t to be P both in virtue of the way it is on its own, and in virtue of the way something else is—namely, when t’s being P is metaphysically overdetermined. Consequently, t’s being intrinsically P is compatible with t’s being P in virtue of the way something else is. On metaphysical ground see Fine (2012).

References Almog, J. (1991) “The What and the How,” The Journal of Philosophy, 88 (5):225–244. Bader, R. M. (2013) “Towards a Hyperintensional Theory of Intrinsicality,” The Journal of Philosophy, 110 (10):525–563. Berto, F. and Jago, M. (2013) Impossible Worlds, Oxford: Oxford University Press. Bovey, G. (2021) “Essence, Modality, and Intrinsicality,” Synthese, 198 (8):7715–7737. Brogaard, B. and Salerno, J. (2013) “Remarks on Counterpossibles,” Synthese, 190 (4):639–660. Chandler, H. S. (1976) “Plantinga and the Contingently Possible,” Analysis, 36 (2):106–109. Chisholm, R. M. (1967) “Identity through Possible Worlds: Some Questions,” Noûs, 1:1–8. Chisholm, R. M. (1973) “Parts as Essential to Their Wholes,” Review of Metaphysics, 26:581–603. Coates, A. (2022) “Essence, Triviality and Fundamentality,” Canadian Journal of Philosophy, 52 (5):502–516. Correia, F. (2007) “(Finean) Essence and (Priorean) Modality,” Dialectica, 61 (1):63–84. Cowling, S. (2013) “The Modal View of Essence,” Canadian Journal of Philosophy, 43 (2):248–266. De, M. (2020) “A Modal Account of Essence,” Metaphysics, 3 (1):17–32. Della Rocca, M. (1996) “Essentialism: Part 1,” Philosophical Books, 37 (1):1–13. De Melo, T. X. (2019) “Essence and Naturalness,” Philosophical Quarterly, 69 (276):534–554. Denby, D. (2014) “Essence and Intrinsicality,” In R. M. Francescotti (ed) Companion to Intrinsic Properties, Berlin: De Gruyter. De Rizzo, J. (2022) “A Ground-theoretical Modal Definition of Essence,” Analysis, 82 (1):32–41. Dummett, M. (1993) “Could There Be Unicorns?” in The Seas of Language, Oxford: Clarendon Press. Dunn, J. M. (1990) “Relevant Predication 3: Essential Properties,” in J. Dunn and A. Gupta (ed) Truth or Consequences, Dordrecht: Kluwer Academic Publishers. Fara, D. G. (2008) “Relative-sameness Counterpart Theory,” The Review of Symbolic Logic, 1 (2):167–189. Fine, K. (1985) “Plantinga on the Reduction of Possibilist Discourse,” in J. Tomberlin and P. van Inwagen (eds), Alvin Plantinga, Dordrecht: Springer. Fine, K. (1994) “Essence and Modality,” Philosophical Perspectives, 8:1–16.

121

Alessandro Torza Fine, K. (2010) “Towards a Theory of Part,” Journal of Philosophy, 107 (11):559–589. Fine, K. (2012) “Guide to Ground” in F. Correia and B. Schnieder (eds) Metaphysical Grounding, Cambridge: Cambridge University Press. Forbes, G. (1985) The Metaphysics of Modality, Oxford: Clarendon Press. Gorman, M. (2005) “The Essential and the Accidental,” Ratio, 18 (3):276–289. Hanson, W. H. (2018) “Actualism, Serious Actualism, and Quantified Modal Logic,” Notre Dame Journal of Formal Logic, 59 (2):233–284. Hazen, A. (1979) “Counterpart-theoretic Semantics for Modal Logic,” The Journal of Philosophy, 76 (6):319–338. Hintikka, J. (1970) “The Semantics of Modal Notions and the Indeterminacy of Ontology,” Synthese, 21 (3–4):408–424. Koslicki, K. (2024) “Modality and Essence in Contemporary Metaphysics,“ in S. Newlands and Y. Melamed (eds) Modality: A History (Oxford Philosophical Concepts), Oxford: Oxford University Press. Kripke, S. A. (1963) “Semantical Considerations on Modal Logic,” Acta Philosophica Fennica, 16: 83–94. Kripke, S. A. (1980) Naming and Necessity, Cambridge, MA: Harvard University Press. Lewis, D. K. (1968) “Counterpart Theory and Quantified Modal Logic,” The Journal of Philosophy, 65 (5):113–126. Lewis, D. K. (1973) Counterfactuals, Malden, Mass.: Blackwell. Lewis, D. K. (1979) “Counterfactual Dependence and Time’s Arrow,” Noûs, 13 (4):455–476. Lewis, D. K. (1986) On the Plurality of Worlds, Malden, MA: Wiley-Blackwell. Linsky, B. and Zalta, E. N. (1994). “In Defense of the Simplest Quantified Modal Logic,” Philosophical Perspectives, 8:431–458. Livingstone-Banks, J. (2017) “In Defence of Modal Essentialism,” Inquiry, 60 (8):816–838. Mackie, P. (2006) How Things Might Have Been: Individuals, Kinds, and Essential Properties, New York: Oxford University Press. Marcus, R. B. (1967) “Essentialism in Modal Logic,” Noûs, 1 (1):91–96. Mares, E. (1997) “Who’s Afraid of Impossible Worlds?,” Notre Dame Journal of Formal Logic, 38 (4): 516–526. McDaniel, K. (2004) “Modal Realism with Overlap,” Australasian Journal of Philosophy, 82 (1):137–152. Moore, G. E. (1920) “External and Internal Relations,” Proceedings of the Aristotelian Society, 20 (1):40–62. Parsons, T. (1967) “Grades of Essentialism in Quantified Modal Logic,” Noûs, 1 (2):181–191. Paul, L. A. (2006) “In Defense of Essentialism,” Philosophical Perspectives, 20 (1):333–372. Plantinga, A. (1973) “Transworld Identity or Worldbound Individuals?” in M. Munitz (ed) Logic and Ontology, New York, NY: New York University Press. Plantinga, A. (1974) The Nature of Necessity, Oxford: Oxford University Press. Plantinga, A. (1985) “Replies to My Colleagues,” in J. Tomberlin and P. van Inwagen (eds) Alvin Plantinga, Dordrecht: Springer. Pollock, J. L. (1985) “Plantinga on Possible Worlds,” in J. Tomberlin and P. van Inwagen (eds) Alvin Plantinga, Dordrecht: Springer. Prior, A. (1957) Time and Modality, Oxford: Oxford University Press. Putnam, H. (1973) “Meaning and Reference,” The Journal of Philosophy, 70:699–711. Putnam, H. (1975) “The Meaning of “Meaning”,” in Mind, Language, and Reality: Philosophical Papers, vol. 2. Cambridge: Cambridge University Press. Quine, W. V. O. (1953) “Three Grades of Modal Involvement,” Proceedings of the XIth International Congress of Philosophy, 14:65–81. Roca-Royes, S. (2011) “Essentialism vis-à-vis Possibilia, Modal Logic, and Necessitism,” Philosophy Compass, 6 (1):54–64. Salmon, N. U. (1981) Reference and Essence, Princeton, NJ: Princeton University Press. Salmon, N. U. (1989) “The Logic of What Might Have Been,” Philosophical Review, 98 (1):3–34. Schaffer, J. (2004) “Two Conceptions of Sparse Properties,” Pacific Philosophical Quarterly, 85 (1):92–102. Sider, T. (2001) “Maximality and Intrinsic Properties,” Philosophy and Phenomenological Research, 63 (2):357–364. Skiles, A. (2015) “Essence in Abundance,” Canadian Journal of Philosophy, 45 (1):100–112.

122

Modal Conceptions of Essence Stalnaker, R. (1968) “A Theory of Conditionals,” In N. Rescher (ed) Studies in Logical Theory, Oxford: Blackwell. Stalnaker, R. C. (2003) Ways a World Might Be: Metaphysical and Anti-Metaphysical Essays, Oxford: Oxford University Press. Stephanou, Y. (2007) “Serious Actualism,” Philosophical Review, 116 (2):219–250. Steward, S. (2015) “Ya Shouldn’ta Couldn’ta Wouldn’ta,” Synthese, 192 (6):1909–1921. Torza, A. (2012) “‘Identity’ without Identity,” Mind, 121 (481):67–95. Torza, A. (2015) “Speaking of Essence,” Philosophical Quarterly, 65 (261):754–771. Unger, P. (1980) “The Problem of the Many,” Midwest Studies in Philosophy, 5 (1):411–468. Varzi, A. C. (2006) “Strict Identity with No Overlap,” Studia Logica, 82 (3):371–378. Varzi, A. C. (2020) “Counterpart Theories for Everyone,” Synthese, 197 (11):4691–4715. Walters, L. (2014) “The Possibility of Unicorns and Modal Logic,” Analytic Philosophy, 55 (2):295–305. Wang, J. (2015) “Actualist Counterpart Theory,” The Journal of Philosophy, 112 (8):417–441. Wildman, N. (2013) “Modality, Sparsity, and Essence,” Philosophical Quarterly, 63 (253):760–782. Williamson, T. (2013) Modal Logic as Metaphysics, Oxford: Oxford University Press. Zylstra, J. (2019) “Essence, Necessity, and Definition,” Philosophical Studies, 176 (2):339–350.

123

8 NON-MODAL CONCEPTIONS OF ESSENCE Fabrice Correia

8.1 Introduction The story is well-known. Around the 1970s, in great part thanks to spectacular progress in modal logic, most analytic philosophers became convinced that the metaphysical modalities were respectable notions. Many of these philosophers subsequently embraced the conception of essence as de re necessity. Then came Kit Fine’s “Essence and Modality” (1994), which convinced many of those who embraced this modal conception of essence that the conception was flawed. Almost 30 years later, and despite sustained efforts to save the modal conception, there seems to be a near consensus that it cannot be upheld. How much of this story is true? The answer depends of course on how “most”, “many” and “near” are precisified. But irrespective of how the vagueness of the story is to be resolved to make it true, it is fair to say that “Essence and Modality” has had a deep impact on the way philosophers think about the connections between essence and modality. And it is also fair to say that a great part of this impact shows up in the impressive popularity that non-modal conceptions of essence have gained in the past 15 years or so.1 My main task in this chapter is to give the reader a good picture of the non-modal views about essence that have been developed since 1994, including the one that Fine himself advocated in “Essence and Modality”. This is no easy task: the directly relevant 1994 and post-1994 literature that I could find up to the present day (16 January 2022), excluding the relevant chapters of this volume, comprises no less than 87 items, articles or books. (About 70 of them appear in the bibliography of this chapter.) And this would presumably have been a much more difficult task a few years hence, since, as Figure 8.1 shows, the relevant literature seems to grow exponentially. My plan is as follows. In section 2, I start by spelling out the modal conception of essence that is the target of “Essence and Modality”, I then go through some important objections against this conception—some from Fine’s paper itself and some others—and I finally briefly review some replies to these objections. In section 3, I introduce a number of different nonmodal conceptions of essence. Section 4 is devoted to discussing objections to these conceptions. In sections 5 and 6, I deal with another influential view defended by Fine in “Essence and Modality”, namely the view that metaphysical necessity is to be understood in terms of essence. Section 5 spells out Fine’s own variant of the view and spells out other

124

DOI: 10.4324/9781003008750-11

Non-Modal Conceptions of Essence

Figure 8.1

“Number of relevant 1994 and post‐1994 publications about essence per year.”

variants that have been defended in the literature; section 6 discusses objections to the view. Finally, section 7 very briefly mentions further important topics and developments. I should flag that I have deliberately left some topics aside, as they are dealt with in other chapters of this volume—for instance the chapters Logic of Essence, Conferralism and Conventionalism. Let me also mention that the chapters Contemporary (Analytic Tradition) and Modal Conceptions of Essence are directly connected to the present chapter.

8.2

The Modal Conception and Its Problems

The modal conception of essence has it that to essentially have a property is to have it as a matter of necessity. This conception can naturally be made more precise in two slightly different ways, which correspond to two modal accounts of essence: the categorical account and the conditional account. According to the former, an object is essentially F just in case it is necessarily F; according to the latter, an object is essentially F just in case it is necessarily F if it exists.2 The locution “just in case” in the formulation of the accounts may be understood in various ways (see section 5 for some candidates), but for present purposes it will not be important to be specific on how exactly it is to be understood. What will be important is that the categorical account and the conditional account support the biconditionals (1) and (2), respectively:

125

Fabrice Correia

(1) An object x is essentially F iff necessarily, x is F (2) An object x is essentially F iff necessarily, x is F if x exists The specific targets of Fine’s (1994) objections to the modal conception of essence are precisely the categorical and the conditional accounts as I have formulated them.3 Fine’s objections against these accounts boil down to establishing that no account of essence should support the right-to-left direction of either (1) or (2). Plausibly, Socrates necessarily has the following properties if he exists: • • • •

Existing Being such that there are infinitely many primes Being numerically distinct from the Eiffel Tower Being a member of the singleton set {Socrates}

Taking this for granted, the right-to-left direction of (2) allows one to infer that Socrates has all these properties essentially. But intuitively, none of these properties is essential to him. Hence, the conditional account should be rejected.4 This objection can be turned into an objection against the categorical account if we make the last three properties of the list conditional upon the existence of Socrates (the first property cannot be invoked in this case). Alternatively, one may invoke a necessary existent instead of Socrates. The details are obvious. Such is the core argument that Fine puts forward against the two modal accounts of essence. The argument involves substantial modal assumptions (that Socrates is necessarily so and so) as well as substantial essentialist assumptions (that Socrates is not essentially so and so), but some of what Fine says (p. 5) suggests a modified argument that rests on a significantly weaker basis, and which is accordingly significantly stronger. The part of the argument that deals with the conditional account goes as follows: for at least one of the properties listed above, it should be possible to consistently maintain both that Socrates fails to have that property essentially and that he necessarily has the property if he exists; yet given the right-to-left direction of (2), this is impossible; therefore, the conditional account must be rejected. The substantial assumptions of the previous argument have thus been replaced by a much weaker assumption, viz. the statement that I wrote in italics. The modifications to make in order to get a corresponding argument against the categorical account are straightforward. Along the same line of thought, it is important to note that the pool of examples that Fine invokes can be broadened to a significant extent, thereby making the argument even more compelling. Variations on the singleton Socrates example can be used to illustrate the point. Appeal to singletons is not compulsory: instead of {Socrates}, one could invoke any set {Socrates, X} where X’s existence is necessitated by Socrates’s existence—one may take for X any necessary existent, or Socrates’s brain, say. Appeal to sets is not compulsory either: instead of {Socrates}, one could invoke any mereological fusion Socrates + X where X is any necessary existent that has nothing to do with the nature of Socrates—the empty set, or an arbitrary angel, say. The left-to-right directions of (1) and (2) are left untouched by the previous objections. Fine does not question these directions; he is actually explicit that he takes an essential property to be a property that the object necessarily has, or necessarily has if it exists (p. 4). Now an impressive number of philosophers have recently rejected, or at least questioned, the view that essential properties are (all) necessary, be it categorically or conditionally upon the existence of their bearers—see Gorman (2014), Leech (2018, 2021), Noonan (2018),

126

Non-Modal Conceptions of Essence

Romero (2019), Casullo (2020) and Mackie (2020).5 The common thought here is the following: if an essential property of an object is nothing but a property that pertains to what the object is, then there is, on the face of it, nothing modal in essentiality; why then should we believe that an essential property of an object is a property that the object must have (or must have if it exists)? We have here a further objection to the two modal accounts of essence. Yet another objection stems from distinctions introduced in Fine (1995a), namely the distinction between constitutive and consequential essence and the distinction between immediate and mediate essence, which Fine spells out as follows: An essential property of an object is a constitutive part of the essence of that object if it is not had in virtue of being a consequence of some more basic essential properties of the object; and otherwise it is a consequential part of the essence.6 (p. 57) In general, the mediate nature of an object will incorporate the nature of all of the objects upon which it depends; the nature will in this sense be hereditary. But the immediate nature will only include that which has a direct bearing on the nature of the object. (p. 61) The objection, which Fine does not himself put forward, is simply that the modal accounts are blind to these distinctions. It is not the place here to explain in detail how friends of the modal accounts of essence have reacted, or can react, to these objections (the right place for this is the chapter Modal Conceptions of Essence in this volume) but let me very briefly run through some main options. Against the last objection, one may question the meaningfulness of the relevant distinctions—as in Correia (2007: 83). Against the previous objection, one may rely on sheer “intuition” or positively argue that the relevant connection between essence and necessity is conceptual—as in Jago (2021). In reaction to Fine’s (1994) objections, four strategies (that can to a certain extent be mixed) have been pursued to save the modal conception of essence, or something reasonably close to it: (i) deny that some Finean counterexamples really are counterexamples—see Cowling (2013) and Wildman (2013); (ii) formulate a modal account of essence where the modality is non-standard and argue that Fine’s counterexamples, or at least some of them, are harmless against the account—see Correia (2007) and Brogaard & Salerno (2007); (iii) modify the modal accounts by simply adding to the modal condition on the property the requirement that the property should have a certain additional feature, e.g., that it should be sparse—as in Cowling (2013), Wildman (2013) and De Melo (2019)—or intrinsic—as in Denby (2014) and Bovey (2021); and finally (iv) put forward an account in terms of modality and some other heavy-duty metaphysical notion as in (iii) but in a less straightforward way—as for example in De Rizzo (2022), who invokes metaphysical grounding in addition to metaphysical necessity.7

8.3

Non-Modal Accounts

Let primitivism be the view that essence cannot be understood in other terms. In the wake of Fine’s objections to the modal accounts of essence, primitivism has been widely seen with a positive eye. In “Essence and Modality”, Fine does not address the question of whether primitivism holds, but in the contemporary “Senses of Essence” he is explicit that he takes the view to be plausible. Explicit endorsement of primitivism, or endorsement of the view as a working hypothesis, is present for instance in Correia (2006, 2012, 2020), Oderberg (2007), Lowe (2008, 2012, 2016), Koslicki (2012), Hale (2013, 2018a, 2018b), Kment (2006, 2014), Zylstra (2019a), Godman, Mallozzi & Papineau (2020) and Mallozzi (2021). The list of

127

Fabrice Correia

philosophers who endorse primitivism without having said so in print is certainly much longer. Fine’s (2015) more recent view of the essential features as those which are “constitutively necessary” to the relevant items is explicitly primitivist. However, some non-primitivist, non-modal accounts of essence have been put forward. In this section I simply spell out these accounts, reserving critical assessment for §4. Gorman (2005, 2014) proposes the following account (my formulation is faithful to the spirit, not the letter, of Gorman’s formulations): (3) x is essentially F just in case the following holds: (i) being F really characterizes x and (ii) there is no G such that being F really characterizes x because being G really characterizes x “Real” characterization is opposed to “dummy” or “fake” characterization; being (roughly) spherical really characterizes the football I have at home; being red or not red does not, even though the ball does have this property. The distinction is intuitive enough (which does not mean it is completely clear). “Because” is intended to express metaphysical grounding—on which the reader can consult, e.g., Correia & Schnieder (2012)—or something similar. Zalta (2006) develops a theory of essentiality based on his distinction between abstract objects (among which he counts sets, and therefore singletons) and non-abstract objects and on the approach to quantified modal logic advocated in Linsky & Zalta (1994, 1996). The details would take us too far, so let me give a brief sketch of the theory. An important claim of the theory is that whereas all objects exemplify (that is, have) properties, abstract objects, and only them, in addition encode properties. Another claim is that necessarily, everything necessarily exists, and that the role that most philosophers attribute to the concept of existence in modal contexts must rather be played by the concept of concreteness. That said, Zalta defines four notions of essentiality, three modal and one non-modal: • • • •

F F F F

is is is is

essential1 essential2 essential3 essential4

to to to to

x x x x

iffdf iffdf iffdf iffdf

necessarily, x exemplifies F necessarily, x exemplifies F if x is concrete F is essential2 but not essential1 to x x encodes F

The first one mirrors the categorical modal account of essence. The second one mirrors the conditional account, with the difference that concreteness rather than existence is invoked. The fourth notion applies only to abstract objects, whereas the other ones apply to all objects. As we will see in §4, Zalta puts the non-modal notion to work to deal with Fine’s singleton example. Rayo (2013, 2015) and Correia & Skiles (2019) advocate an account of essence in terms of “generalized identity”. Generalized identity is a concept of identity expressed by statements of type “For something to be F is for it to be G” and “For it to be the case that p is for it to be the case that q”, among other statements of similar types (see also Dorr (2016) on this concept). Generalized identity allows one to characterize several other notions of essence in addition to the usual notion of “objectual” essence that we have been concerned with so far, but for the sake of simplicity I will here leave these other notions aside (but see §7). The proposed account goes as follows: (4) x is essentially F just in case to be F is part of what it is to be x, i.e., just in case for something to be both x and F is for it to be x where “to be x” is short for “to be identical to x”.8

128

Non-Modal Conceptions of Essence

8.4

Assessment of the Non-Modal Accounts

An important argument in favor of the modal accounts of essence is that they are accounts of a prima facie rather obscure notion in terms of notions that are better understood—metaphysical necessity in the case of the categorical account, metaphysical necessity, existence and the material conditional in the case of the conditional account. Anyone impressed by this argument will consider primitivism problematic, on the grounds that obscure notions should not be taken as primitive. But the other non-modal accounts discussed in the previous section are also subject to this sort of objection from obscurity: the notions they invoke (real characterization, grounding, encoding, generalized identity), some will argue, are too obscure to be taken as basic in a respectable metaphysical theory. The best reply from friends of these accounts is to try to put forward illuminating theories of the notions they take as primitive. How do the non-modal accounts listed in the previous section fare with respect to the objections against the modal accounts discussed in §2? Primitivism obviously fares very well. The Finean counterexamples have no grip on the primitivists: they are free to hold, as Fine (1994) indeed does, that Socrates necessarily has the properties that Fine invokes in his objections and that he does not have them essentially; Fine (1994) believes that essential properties are necessary, but this belief is not forced upon him by his primitivism: a primitivist is free to hold that some essential properties are contingent; finally, there appears to be nothing to prevent a primitivist from recognizing, as Fine (1995a) indeed does, the distinction between constitutive and consequential essence and that between immediate and mediate essence. (Of course, once these latter distinctions are accepted, it will be plausible to hold that some of the essentialist notions involved in the distinctions should be defined in terms of other essentialist notions that are also involved—e.g., that consequential essence should be defined in terms of constitutive essence, or the other way around. Yet such definitions are consistent with primitivism, since they would define essentialist notions in terms of other essentialist notions.) On Gorman’s account, an essential property of an object must really characterize the object. Gorman claims that although there is some vagueness in the distinction between features that really characterize and features that do not, he holds that the features that Fine invokes in his counterexamples are “hardly borderline ones” (2005: 279). Gorman’s thought is that these features do not really characterize Socrates, and therefore that his own account escapes Fine’s objections. But is it so clear that none of the features invoked by Fine really characterizes Socrates? We may easily grant that being such that there are infinitely many primes fails to really characterize Socrates. But what about the other three features? Many have held the Kantian view that “existence is not a property”; following this thought it is perhaps fine to claim that existing does not really characterize Socrates. But in order to do so with full justification, we need to be clearer about what real characterization amounts to. What about being distinct from the Eiffel Tower and being a member of {Socrates}? They are both extrinsic features of Socrates. Is this what makes these features not really characterizing him? If so, then the account of essence will fall prey to the objection that some essential features are extrinsic (think for instance about the Kripkean view that Socrates’s biological origins are essential to him; on this view, see the chapter Origin Essentialism in this volume). If not, on what grounds should we deny that these two features really characterize Socrates? Once again, we need to know more about the notion of real characterization. As Gorman (2014: 131) explicitly acknowledges, his account seems to leave room for the possibility of contingent essential properties; if it does, then it escapes the second objection raised against the modal accounts. Presumably, though, the account could also be supplemented

129

Fabrice Correia

with principles that rule out this possibility and hence satisfy those who believe that essential properties are not contingent. Gorman’s concept of essence seems to be what Fine would call a constitutive-immediate concept. If it is such a concept, then Gorman’s account can be expanded into an account that also countenances concepts of consequential and mediate essence, and it therefore escapes the third objection. Let me here describe one way (among others) of defining such concepts in terms of a given constitutive-immediate concept, which is neutral between different underlying accounts of the constitutive-immediate concept. Let me adopt the following definitions: • • • •

E(x) is the set of all properties that are constitutively-immediately essential to object x x immediately depends on y iff y is a constituent of a property in E(x) x depends on y iff there is a chain of links of immediate dependence from x to y CI(x) is the set of all properties F of x such that for some properties F1, F2, … in E(x), x’s having F is a consequence of x’s having F1, F2, … • CM(x) is the set of all properties of x of the form [is such that Fy] where x depends on y and F ∈ CI(x) Being in CI(x) and being in CM(x) are two ways of being consequentially essential to x, and being in CM(x) is a way of being mediately essential to x. Assessing Zalta’s account of essence is another story. It is not completely clear what job his four notions of essence are supposed to do. Fine has one notion of essence in mind—the traditional, Aristotelian notion (see footnote 1). Zalta nowhere claims that one of his notions matches Fine’s. More generally, he does not seem to be concerned with that notion. Rather, the core of his paper, at any rate as far as the connection with Fine (1994) is concerned, consists in arguing that some of the distinctions that Fine identifies, and which Fine argues the standard modal accounts cannot capture, can be captured in his framework in terms of some of the notions of essence that he introduces. Hence, Zalta’s account is not in the same ballpark as the other ones and therefore we cannot assess it using the same criteria. Let me simply comment on Zalta’s treatment of the essentialist asymmetry that Fine takes there to be between Socrates and {Socrates}, which is where Zalta’s non-modal notion of essence, the fourth notion defined in the list, comes into play (see pp. 688–90). Fine holds that we can consistently maintain that Socrates is not essentially a member of {Socrates} while {Socrates} essentially contains Socrates as a member, which the modal accounts of essence do not allow, at least given independently plausible principles of modal set theory. Zalta argues as follows (I set aside many details for the sake of brevity): Singletons are abstract. {Socrates} encodes the property of having Socrates as a member, because in the relevant theory M that defines the sets we are interested in, it is true that {Socrates} has the property of containing Socrates (the “because” is justified by Zalta’s Equivalence Theorem). Therefore, having Socrates as a member is essential4 to {Socrates}. Now being a member of {Socrates} is not essential4 to Socrates: indeed, no property is essential4 to Socrates, because Socrates is not abstract and only abstract objects encode properties. Hence, essentiality4 allows one to capture the asymmetry between Socrates and {Socrates} that Fine has in mind. The argument does establish that there is a relevant essentialist asymmetry between Socrates and {Socrates}, at least if we grant that essentiality4 is a form of essentiality. Yet the argument succeeds only by accident, so to speak: the asymmetry vanishes indeed if the

130

Non-Modal Conceptions of Essence

example is appropriately changed. Take for instance the pair {Socrates} / {{Socrates}} instead of Socrates / {Socrates}. In theory M, {{Socrates}} has the property of having {Socrates} as a member and {Socrates} has the property of being a member of {{Socrates}}. By the Equivalence Theorem, it follows both that {{Socrates}} encodes the property of having {Socrates} as a member and that {Socrates} encodes the property of being a member of {{Socrates}}. It follows both that {{Socrates}} essentially4 has the property of having {Socrates} as a member and that {Socrates} essentially4 has the property of being a member of {{Socrates}}. As announced, there is no asymmetry in this case. Now surely, the essentialist asymmetry that Fine has in mind in the case of the original pair should also be present in the case of the new pair. Hence, Zalta has not succeeded in capturing this asymmetry. Let me finally turn to the Rayo/Correia & Skiles account. Rayo turns out to have a deflationary account of the connection between generalized identity and modality. On his view, “For something to be F is for it to be G” is equivalent to “Necessarily, for all x, Fx iff Gx” (2013: 71). Taking Rayo’s view for granted, the account under (4) above is equivalent to the conditional modal account and is therefore vulnerable to Fine’s (1994) objections. In Correia & Skiles (2019), we do not endorse Rayo’s deflationary account, and in what follows I will assume that the latter is incorrect. I take it that once generalized identity is properly construed, it is plausible to hold that although Socrates necessarily has all the properties invoked in Fine’s objections to the modal accounts of essence, they are not part of what it is to be Socrates—where “part of what it is to be Socrates” is understood along the lines expressed in (4). Consider, for instance, the singleton objection. Surely, it is not the case that for something to be Socrates and a member of {Socrates} is for it to be Socrates: being Socrates has nothing to do with facts about sets. And the view that this is not the case is certainly compatible with the view that Socrates is necessarily a member of {Socrates} if he exists. Note that the essentialist asymmetry between Socrates and {Socrates} is nicely captured on the proposed account, or so I take it: the view that being {Socrates} and a settheoretic container of Socrates is the same thing as being {Socrates} strikes me as compelling. The proposed account of essence allows one to give a powerful reply to the second objection raised against the modal accounts. Why should we believe that essential properties are necessary? Generalized identity, just like standard identity, obeys Leibniz’s Law (suitably qualified—there are opaque contexts for generalized identity, just like there are opaque contexts for standard identity). Using the Law and very weak principles of quantified modal logic, one can derive that an object x must be F if it exists (line 4 in the proof below; following a widespread view, “∃y(y = x)” is taken to state that x exists) from the assumption that x is essentially F, as this is understood according to the proposed account (line 1): 1 2 3 4

To be x and F is to be x Necessarily, ∀y((y = x & Fy) ⊃ Fx) Necessarily, ∀y(y = x ⊃ Fx) Necessarily, ∃y(y = x) ⊃ Fx

Assumption Quantified modal logic By 1 and 2 and Leibniz’s Law By 3 and quantified modal logic

Hence, the proposed account gives reason to believe that essential properties are necessary. See Correia & Skiles (2022) for a thorough discussion of this line of thought, and Leech (2021) for the critical background of this discussion. The concept of essence characterized by (4) in Correia & Skiles (2019) is arguably constitutive-immediate. It can therefore be expanded into an account that also countenances concepts of consequential and mediate essence (see the discussion of Gorman’s account), and it accordingly escapes the third objection against the modal accounts of essence.

131

Fabrice Correia

8.5 Accounting for Metaphysical Necessity in Terms of Essence “Essence and Modality” is well-known not only for its objections to the modal accounts of essence but also for its defense of a “reverse” account, that is, of an account of metaphysical modality in terms of essence. Fine’s essentialist account of metaphysical modality is primarily an account of metaphysical necessity—metaphysical possibility and the other alethic modalities in the family are to be defined in its terms in the usual ways. Fine’s account can be formulated as follows:9 (5) It is metaphysically necessary that p just in case for some object or objects X, it is true in virtue of the nature (or essence) of X that p—in symbols: □p just in case for some X, □X p Importantly, the account involves the idea that some truths are true in virtue of the nature of several objects taken together without being true in virtue of the nature of individual objects. How exactly the essentialist operator “It is true in virtue of the nature of … that ---” is to be understood in Fine’s account is not obvious. Can it be taken as a primitive or must it be analyzed somehow? Fine takes it that whatever the reply might be, the operator is irreducibly essentialist. See, e.g., Correia (2012, 2020), Michels (2019) and Fine (2020) for some relevant discussions. Fine’s (1994) own formulation of the account is not exactly (5). He holds that the metaphysically necessary truths are the propositions that are true in virtue of the nature of some objects (p. 9).10 This suggests that he at least accepts (5) on what I will call its weakest construal—namely (5) with “just in case” understood as expressing material equivalence (for which I will use the symbol ≡ below). He certainly endorses a stronger claim, but it is not clear exactly which one. Here are some interesting precisifications of (5) that result from various interpretations of the “just in case” locution: Strict equivalence Generalized identity Reduction Grounding

□(□p ≡ for some X, □X p) For it to be the case that □p is for it to be the case that for some X, □X p Its being the case that □p reduces to / consists in its being the case that for some X, □X p If □p, then the fact that □p is grounded in the fact that for some X, □X p

In Correia (2012) and Correia (2020), I describe Fine’s (1994) view about metaphysical necessity as reductive, but there I did not distinguish between claims of generalized identity and claims of reduction. Ditter (2020, 2022) also talks about reduction. Fine (2020) follows me in qualifying his view as reductive. Two other philosophers have endorsed (5) on its weakest construal, Lowe (2012: 939) and Hale (2018b: 124–6). Both, like Fine, accept something stronger. Lowe claims that “all metaphysical necessity […] is grounded in essence” (p. 939, his emphasis), and Hale holds that when a proposition is metaphysically necessary, this is because the proposition is true in virtue of the nature of some objects (p. 125). An interesting exploration of (5) not already cited above is Hale (1996). (5) on its weakest construal implies that every proposition that is metaphysically necessary is true in virtue of the nature of some objects, which can be picturesquely reformulated thus: every metaphysical necessity has its source in essence. Some philosophers have explicitly

132

Non-Modal Conceptions of Essence

endorsed only the weaker claim that only some metaphysical necessities have their source in essence—i.e., the claim that some, but not all, metaphysically necessary propositions are true in virtue of the nature of some objects. Hale (2013), for instance, holds that (at least some) logical necessities have their source in the nature of logical entities (§5.5), while essentialist truths are necessary in a “brute” way (pp. 158–9). Godman, Mallozzi & Papineau (2020) hold that the fact that it is necessary that gold has the atomic constitution that it actually does has its source in the nature of gold, but they doubt—in stark contrast with Hale (2013)—that the logical necessities have an essentialist source (p. 332). Note that these views of Hale (2013) and Godman, Mallozzi & Papineau (2020) do not constitute accounts of metaphysical necessity in terms of essence. Let me finally briefly mention two further accounts of metaphysical necessity in terms of essence. One is due to Kment (2006, 2014) and the other one to Correia & Skiles (2022). Kment’s account goes as follows (2014: 188): (6) □p just in case the proposition that p is true throughout the sphere around actuality that contains the worlds that match actuality with respect to the metaphysical laws Kment thinks of his account as a real definition of metaphysical necessity (see pp. 197–8) and therefore “just in case” in (6) should be understood accordingly. The details of the account are not completely straightforward, but there is no need to elaborate on them here. What needs to be stressed is that Kment includes among the metaphysical laws all the essentialist truths. This is what makes his account of metaphysical necessity an essentialist account. Correia & Skiles’s (2022) account is, just like the account of essence defended in Correia & Skiles (2019), formulated in terms of generalized identity. It goes as follows, where “just in case” is intended to express generalized identity:11 (7) □p just in case the proposition that p is a logical consequence of the true generalized identities Given (4) above, there is a mapping from the essentialist propositions to the generalized identities, that maps each essentialist proposition P which state that an object is essentially so and so to a proposition Q such that for P to be the case just is for Q to be the case. The fact that the true generalized identities comprise propositions that are outputted by this mapping already makes (7) deserve to be called an essentialist account. But there is a stronger reason why the account is, on our view, essentialist: we take there to be a mapping of the sort described above which outputs all the generalized identities. See Correia & Skiles (2022: 1295).12

8.6 Objections Against Accounts of Necessity in Terms of Essence In this section I review some objections against the kind of accounts of metaphysical necessity discussed in the previous section. For reasons of space but also in great part because of the available literature on the topic, I focus on objections that have been raised against the Finean account.13 Some of these objections may be turned, with more or less effort, into objections against the other two accounts discussed in the previous section, but I leave the elaboration of this point aside. Hale (2018b). Plausibly, essentialist truths are necessary: for all X and p, if □X p, then □□X p. The principle is not uncontroversial, see e.g., Robertson (2022: 160–1). But it is validated by Fine’s (1995b, 2000) logics of essence, so Fine could not block the objection to be raised

133

Fabrice Correia

below by simply denying it. Hale (2018b: 127–8) points out that given this principle and Fine’s essentialist account of necessity, every essentialist truth □X p is accompanied by infinitely many other such truths □X*□X p, □X**□X*□X p, … . He calls such series “troublesome-looking” regresses, but he argues that there actually is no trouble. His argument seems to be the following: (a) if for some X and p, there is a series of truths □X p, □X*□X p, □X**□X*□X p, … , then the situation is unproblematic if the series □X p, □X□X p, □X□X□X p, … (where the subjects of essentialist attribution are all identical) is also a series of truths, and it is problematic otherwise; fortunately, (b) for all X and p, if □X p, then □X□X p. Hale’s argument for (b) is hard to grasp, but Fine’s (1995b, 2000) logics of essence validate (b), and so he would not complain. The reason why Hale thinks Fine—or anyone else, for that matter—should accept (a) remains unclear. Van Cleve (2018). Van Cleve (2018: 16–7) argues, like Hale, that given that essentialist truths are necessary, the Finean is committed to there being series of essentialist truths of type □X p, □X□X p, □X□X□X p, … . He then argues that given a further principle put forward by Rosen (2010: 119), namely the principle that for all X and p, if □X p, then p because □X p, trouble follows. In Rosen’s own principle, “because” expresses metaphysical grounding, but for the purpose of Van Cleve’s argument we may take it to express another sort of explanatory connection.14 Van Cleve’s objection goes as follows. Take a given series of truths □X p, □X□X p, □X□X□X p, … . By Rosen’s principle, we then have the following series of explanatory truths: p because □X p □X p because □X□X p □X□X p because □X□X□X p ⁝ Van Cleve offers two objections against such series. The literature on metaphysical grounding features further material to determine whether these series, understood as involving groundtheoretic truths, are problematic—see, e.g., Dixon (2016) and Rabin & Rabern (2016) and the references therein. Teitel (2019) and Hale (2018b). Despite the efforts of Williamson (2013) and a few others, Contingentism—the view that possibly something possibly fails to be identical to something—is still orthodoxy. Teitel (2019) argues that contingentists are likely to accept what he calls Standard Contingentism, the view that what is essential to something may fail to be essential to something—in symbols and more perspicuously: ∃p(∃X□X p ∧ ⋄¬∃X□X p). He then argues that Standard Contingentism coupled with Fine’s account of necessity is inconsistent with the view that the correct modal propositional logic for metaphysical necessity is a normal modal logic that results from the minimal system K by adding at least the characteristic axiom of system S4, namely □A ⊃ □□A. Teitel’s argument can be formulated in the following, straightforward way: 1 From Standard Contingentism, principles validated by system K and weak principles for the sentential quantifier, we can infer ∃p¬□(∃X□X p ⊃ □∃X□X p). 2 Given Fine’s account, we can replace the two occurrences of ∃X□X p by □p salva veritate, so we can infer ∃p¬□(□p ⊃ □□p). 3 But this flies in the face of any normal modal logic that results from K by adding at least □A ⊃ □□A, since any such logic has □(□A ⊃ □□A) as a theorem.

134

Non-Modal Conceptions of Essence

Teitel officially does not take a stand on what must be given up in order to get rid of the inconsistency. Yet the argument clearly puts a lot of pressure on friends of Fine’s account of necessity, since the view that system S5 is the correct modal propositional logic for metaphysical necessity is, like Contingentism, orthodoxy, and S5 is a normal modal system that has □A ⊃ □□A as a theorem.15 Werner (2021) suggests an interesting reply to Teitel’s argument that presupposes HigherOrder Contingentism—roughly, the view that possibly some proposition possibly fails to be something. Werner argues that contingentists have good reasons to be higher-order contingentists, and that higher-order contingentists should reject the second step of the argument. Interestingly, Hale (2018b: 129–31) produced essentially the same argument. Hale (2018b) endorses the Finean account of metaphysical necessity, and his version of the argument shows that he also endorses Standard Contingentism. What he rejects is, in effect, also the second step of the argument, but in a way that is more radical than the way in which higher-order contingentists, according to Werner, should reject it. Details aside, he suggests that the principle that from □(A ≡ B) and □A one can infer □B—a principle that is unrestrictedly validated by system K—should be restricted to formulas A and B that do not contain essentialist operators. Wildman (2021). Wildman (2021) puts forward two arguments against Fine’s account of metaphysical necessity. The first one (§2) involves two events that cannot jointly occur, like the event T of a given coin’s landing tails at a time t and the event H of that same coin’s landing heads at t. (Wildman then gives another argument involving tropes instead of events, but it is not significantly different.) It is plausible to hold, Wildman argues, that it is part of the nature of T that the coin lands tails at t and likewise that it is part of the nature of H that the coin lands heads at t. Taking this for granted, Wildman goes on, by Fine’s account it follows that the coin lands both tails and heads at t, which of course is not the case. The precise target of Wildman’s objection is the following particular consequence of Fine’s account: for all p and X, if it is true in virtue of the nature of X that p, then p. This consequence of the account has no modal content at all, it simply states that the essentialist operator at work in Fine’s account is factive. The objection is thus best viewed, not as an objection against Fine’s account of metaphysical necessity, but as an argument to the effect that this essentialist operator is not factive. Now how could it be denied that the operator is factive? If it is true in virtue of the nature of X that p, then, surely, it is true that p; and if it is true that p, then p. The natural reply to Wildman’s first argument is to reject its essentialist premises. What is essential to T is conditional in character: it is not that the coin lands tails tout court, it is that the coin lands tails if T occurs. And likewise, what is essential to H is only the conditional proposition that the coin lands heads if H occurs. Using the conditional essentialist propositions as premises instead of the non-conditional essentialist propositions that Wildman invokes, one cannot derive problematic conclusions (pace what Wildman claims in footnote 15 on p. 1461). The second objection (§3) is more complex and for reasons of space I cannot adequately run through it. It crucially presupposes that Fine’s account is reductive, or that it involves the concept of grounding in more or less the way I suggested in the previous section. One possible line of reply, though not the only one, is that the account should not be understood in either of these ways. If it is understood as consisting in a statement of generalized identity (to wit: for it to be metaphysically necessary that p is for there to be some X such that it is part of the nature of X that p—see again the previous section), for instance, Wildman’s objection may not get off the ground.

135

Fabrice Correia

Bovey (2022). Bovey (2022) offers a complex argument against Fine’s account of metaphysical necessity which I cannot summarize in a satisfactory way without entering into too many details. The argument relies on three main principles; let me simply briefly comment upon them. One of the three principles is the principle that essentialist truths are necessary—which Bovey dubs NE. As I stressed at the very beginning of this section, the principle is plausible, and Fine cannot reject it if he sticks to his views on the logic of essence as expressed in Fine (1995b, 2000). The other two principles are Source and NT. Source is roughly my Grounding (see §5)—I say “roughly”, because Bovey invokes the generic notion of “metaphysical explanation” rather than grounding. NT states that if the fact that □q is metaphysically explained by the fact that p, so is also the fact that q. I have just stressed (see the discussion of Wildman’s second objection) that Grounding is not forced upon the Finean. I think the same is true of Source. Be that as it may, NT looks like the weakest element of the triad. In defense of the principle, Bovey simply claims that he finds it plausible. Here is an argument to the effect that NT needs substantial support in order to be accepted. Consider an account of metaphysical necessity in terms of possible worlds which invokes metaphysical explanation as follows: (8) For all q, if □q, then the fact that □q is metaphysically explained by the fact that for all possible worlds w, it is true at w that q From (8) and NT we get: for all q such that □q, the fact that q is metaphysically explained by the fact that for all possible worlds w, it is true at w that q. This consequence is highly implausible. But regardless of whether it is true or not, it surely can consistently be rejected by a proponent of (8). This shows that NT is not a conceptual truth. We need substantial arguments in order to accept it.

8.7

Further Topics, Further Developments

In this last section I wish to briefly run through two important topics that I have left aside in the previous sections, and to briefly mention further material on issues that I have only touched upon. (A) The notion of essence that most people who use it or work on it have in mind is certainly objectual: it is the notion of one or several objects (things, entities, …) being essentially so and so. In Correia (2006), I stressed that alongside this notion stands another, at least equally traditional notion of essentiality, the generic notion, and I argued that it cannot be reduced to the objectual notion.16 The contrast is illustrated by the following two claims: • It is part of what redness is that whatever is red is colored • It is part of what it is to be red that whatever is red is colored The first claim is objectual, it says of a given object, redness, that it is essentially such that whatever is red is colored. The second claim, a generic claim, is not. Fine’s (1994) objections against the modal accounts of objectual essence can be turned into objections against corresponding accounts of generic essence, and the Finean account of metaphysical necessity in terms of essence, if it is to be accurate, must take the generic dimension in addition to the objectual dimension into account. The non-modal conceptions of essence discussed in §3 can all accommodate the generic dimension, in a more or less

136

Non-Modal Conceptions of Essence

straightforward way—the account in Correia & Skiles (2019) actually accommodates it by default. Likewise, all the accounts of metaphysical necessity discussed in §5 either can be modified to take generic essence on board or already take it on board—the account in Correia & Skiles (2022) is of this latter sort. If we formulate essentialist claims in higher-order languages with the right linguistic resources, then many more kinds of essentialist statements will be countenanced. Thus, for instance, if we follow Fine’s idea of expressing essence using the operator □ with indices that correspond to the subjects of essentialist attributions, we may allow these indices to be of arbitrary grammatical types—the type of adverbs or the type of monadic sentential operators, for example. Ditter (2022) develops a theory of essence along these lines and defends a reduction of necessity to essence of a broadly Finean sort. Many details of the theory of essence and necessity developed in Correia & Skiles (2019, 2022) have been left aside in these papers, in particular the exact type of language in which the theory should be formulate, but it is clear that a formal implementation of the theory would require a rich higher-order language. (B) Aristotle famously held that the definition of something is a statement that expresses the essence of that thing (see, e.g., Metaphysics Z 5, 1031a11–12). Fine (1994) follows him on this connection between essence and what is now often called “real” definition. More recently, Fine and others have advocated a conception of real definition along the same broad lines but which is significantly more refined. Fine (2015) and Rosen (2015) both hold that a real definition involves not only an essentialist attribution but also a ground-theoretic attribution. To illustrate with a generic rather than an objectual example, both hold (details put aside) that a real definition of what it is for something to be F should imply that for some G, being G grounds being F. In Correia (2017), I also defend an account in the same family, and I argue that it is equivalent to another account that invokes not grounding but the notion of “carving reality at its joints”. On the connection between essence and real definition, see also Zylstra (2019b) and Hale (2021). (C) Here are two questions of intrinsic philosophical interest: 1 What do essentialist truths ground or otherwise metaphysically explain? 2 What grounds or otherwise metaphysically explain essentialist truths? Understood modally, these questions—especially the second one—have been at the center of philosophical investigations for decades. But the questions have also occupied friends of nonmodal conceptions of essence. I already mentioned Glazier (2017) on the first question (footnote14): according to him, if □X p, then the fact that □X p metaphysically explains the fact that p, where metaphysical explanation is not to be understood as ground-theoretic explanation. See Zylstra (2019c) and Vogt (forthcoming) for further discussion. The views labeled Reduction and Grounding in §5 are of course also relevant to this question. See also Romero (2019) and Wallner & Vaidya (2020). Regarding the second question, Dasgupta (2016) argues that essentialist facts are “autonomous”, i.e., “not apt for being grounded”. Raven (2020) explores a view, inspired by Almog (1991) and later developments, according to which they are grounded.

8.8

Related Topics

The chapters of this handbook that are most relevantly connected to this one are: Ancient, Contemporary (Analytic Tradition), Modal Conceptions of Essence, Origin Essentialism, Logic of Essence, Conventionalism, Conferralism. All have been mentioned at relevant places.

137

Fabrice Correia

Acknowledgments I have greatly benefitted from feedback on presentations linked to this chapter that I gave at three workshops: Atelier du Groupe d’Études en Métaphysique, Collège de France, Paris (online), June 2021, Routledge Handbook of Essence Workshop, Neuchâtel (online), July 2021, and Neo-Aristotelian Perspectives on Persistence and Modality, Mainz (online), October 2021. I am also grateful to Martin Glazier, Robert Michels and the editors of this Handbook for helpful comments on a draft of this chapter. This work was supported by the Swiss National Science Foundation’s project 100012_197172.

Notes 1 The notion of essence that Fine has in mind in the paper is the traditional notion, the one that Aristotle pointed to using the expressions “τὸ τί ἐστι” (the what it is) and “τὸ τί ἦν εἶναι” (the what it was to be) and for which philosophers subsequently used the Latin “essentia” and other words with the same root in more recent languages. (See the chapter Ancient in this volume.) The above story is meant to be a story about how philosophers have thought about this notion over the past 50 years or so, and I take this chapter to be about non-modal conceptions of this notion. During the workshop dedicated to this volume organized by Kathrin Koslicki and Marco Marabello (9-12 July 2021) and in personal conversation thereafter, Teresa Robertson questioned the view that the most prominent philosophers who theorized about essence in the 1970s and the 1980s had the traditional notion in mind: she suggested that they simply used “essential” / “essentially” as synonymous with “necessary” / “necessarily”, or at least that many of them did. This surprised me. Why on earth, I then thought, would philosophers suddenly decide to give new meanings to well-known important philosophical words that have been used for centuries in basically the same way? But I had a look at some relevant papers and books, and I indeed found some cases that fit Robertson’s description. Forbes (1985: 94) explicitly stipulates that he uses “essential property” for a property that is had in every possible world where the bearer of the property exists. Plantinga (1978) is a less clear example, but only slightly so: he makes massive use of the disjunctive phrases “essentially or necessarily” and “necessarily or essentially”, which strongly suggests that he takes “essentially” and “necessarily” to be semantically equivalent. Yet other cases do not fit Robertson’s description, or at least not so well. Barcan Marcus (1971) is a clear case: she uses “essential” in the Aristotelian sense and heavily stresses that essentiality, understood in the Aristotelian way, is not the same as necessity (even though, she also stresses, the two are connected in specific ways). Kripke (1980) looks like a mixed case. Some textual evidence suggests that he understands “essential” in the Aristotelian sense and that he takes it that being essential implies, but is not the same as, being necessary (see his claims about heat and molecular motion on p. 133), whereas some other parts suggest that he takes “essential” and “necessary” as synonymous (as for instance at the end of p. 135 where he discusses a hypothetical baptism of the substance gold). See Fine (2022) for further remarks on the concept of essence in Naming and Necessity. See also footnote 3 of the chapter Origin Essentialism in this volume for relevant considerations. It would be interesting to investigate in more detail what the various relevant philosophers from that period had in mind when they used talk of essence and essential properties. Whatever the upshot, however, the last two sentences of the paragraph to which this footnote is appended are certainly true. 2 Fine mentions a third precisification: an object is essentially F just in case it is necessarily F if it is identical to that very object. As Fine stresses, though, this third account arguably boils down to one of the other two accounts. 3 Fine (1994: 8) is skeptical that a purely modal account of essence, whatever its shape, could be acceptable. See Torza (2015) for an attempt to establish that no such account is viable. It is of course important here to be clear on what a purely modal account is. Torza takes a stand: for him, a purely modal account is one that can be formulated in a certain kind of language for quantified modal logic. 4 The example involving {Socrates} is the one that has proved to be the most difficult to handle by friends of the modal conception of essence. Interestingly, Dunn (1990: 90–1) was very close to discovering it. He indeed notes that if a set S contains only necessary existents, then the conditional account of essence predicts that every member of S is essentially a member of S—which he takes to go against a view on the nature of sets to which he wants to do justice. It is somewhat surprising that it

138

Non-Modal Conceptions of Essence

5

6 7 8

9

10 11 12

13 14 15

16

did not occur to him that we get the very same problem with any set S provided that the set contains only one member. See also Almog (1991). Few people are aware, I think, that Fine argues elsewhere that there is a conception of essence, distinct from the one that is under focus in “Essence and Modality” (and, I would add, quite exotic), according to which essential properties are not all necessary. See Fine (1995a: 66–8). There is an interesting issue as to which notion of consequence is adequate in this context, which I leave completely aside throughout this chapter. See Fine (1995a: 58–61) for a discussion. Dunn (1990), who questioned the standard modal accounts of essence a few years before the publication of “Essence and Modality”, discusses two accounts that belong to this category. They invoke relevant implication instead of grounding. I labeled the proposed account “non-primitivist”, but there is actually room for disagreement here. As stressed in Correia & Skiles (2019: 654), the view that generalized identity is itself an essentialist notion—a view that I find plausible—cannot be lightly dismissed. If that view is taken for granted, then the proposed account is primitivist. But in special way: it takes an unusual essentialist notion as primitive and characterizes all the other essentialist notions in terms of it. Fine also provides accounts of other kinds of necessity (e.g., conceptual and logical necessity) of the same form by restricting the quantifier “for some X” to relevant classes of objects (concepts in the case of conceptual necessity and logical concepts in the case of logical necessity). I will leave these accounts aside. Fine actually says “all objects” rather than “some objects” but given that his operator is monotonic—what is true in virtue of the nature of X is true in virtue of the nature of Y provided that every member of X is a member of Y—the two formulations are equivalent. We actually discuss three distinct accounts, but these accounts are equivalent given plausible assumptions. I skip the details. Rayo (2013: ch. 5) proposes an account of the truth-conditions for statements of a first-order modal language in possible worlds terms, where the possible worlds model is selected on the basis of a set of true generalized identities. Rayo’s account has some affinities with the account in Correia & Skiles (2022). Dorr (2016: 69) also discusses an account of metaphysical necessity in terms of generalized identity, which crucially presupposes a deflationary conception of the latter notion. See Correia & Skiles (2022: 1296–7) for a few remarks on Rayo (2013) and Dorr (2016) on metaphysical modality. The account proposed in Correia & Skiles (2022) has not yet been criticized. For a discussion of Kment’s (2006, 2014) account, see Lange (2015) and the reply in Kment (2015). This is important because one may reject Rosen’s principle if understood as involving metaphysical grounding while accepting it if understood as involving another sort of metaphysical explanation. This is what Glazier (2017) does. Teitel subsequently also argues that the combination of Standard Contingentism with Fine’s account is in tension with the characteristic axiom of system B, namely A ⊃ ◻⋄A, which is also a theorem of S5, but the argument invokes an extra principle that is far from being obvious and is for this reason less convincing than the first argument. The reply to the first argument suggested by Werner (see below) is also available for the second argument. Generic essence is also central to Rayo (2013) and Fine (2015). In Correia (2013), I exploit the sister notion of alethic essence, which can be seen as a special case of generic essence.

Bibliography Almog, J. (1991) “The What and the How,” The Journal of Philosophy, 88(5), 225–244. Barcan Marcus, R. (1971) “Essentialist Attribution,” The Journal of Philosophy, 68(7), 187–202. Barnes, J. (ed.) (1991) The Complete Works of Aristotle, Princeton: Princeton University Press. Bovey, G. (2021) “Essence, Modality, and Intrinsicality,” Synthese, 198(8), 7715–7737. Bovey, G. (2022) “On the Necessity of Essence,” Philosophical Studies, 179, 2167–2185. Brogaard, B. & Salerno, J. (2007) “A Counterfactual Account of Essence,” The Reasoner, 1, 4–5. Casullo, A. (2020) “Essence and Explanation,” Metaphysics, 2(1), 88–96. Correia, F. (2006) “Generic Essence, Objectual Essence, and Modality,” Noûs, 2006, 40(4), 753–767.

139

Fabrice Correia Correia, F. (2007) “(Finean) Essence and (Priorean) Modality,” Dialectica, 61(1), 63–84. Correia, F. (2012) “On the Reduction of Necessity to Essence,” Philosophy and Phenomenological Research, 84(3), 639–653. Correia, F. (2013) “Metaphysical Grounds and Essence,” in M. Hoeltje, B. Schnieder, & A. Steinberg (eds.), Varieties of Dependence, Munich: Philosophia Verlag, 271–291. Correia, F. (2017) “Real Definitions,” Philosophical Issues, 27(1), 52–73. Correia, F. (2020) “More on the Reduction of Necessity to Essence,” in M. Dumitru (ed.), Metaphysics, Modality, and Meaning. Themes from Kit Fine, Oxford: Oxford University Press, 265–282. Correia, F. & Schnieder, B. (2012) “Grounding: An Opinionated Introduction,” in F. Correia & B. Schnieder (eds.), Metaphysical Grounding: Understanding the Structure of Reality, Cambridge: Cambridge University Press, 1–36. Correia F. & Skiles, A. (2019) “Grounding, Essence, and Identity,” Philosophy and Phenomenological Research, 98(3), 642–670. Correia, F. & Skiles, A. (2022) “Essence, Modality, and Identity,” Mind, 131(524), 1279–1302. Cowling, S. (2013) “The Modal View of Essence,” Canadian Journal of Philosophy, 43(2), 248–266. Dasgupta, S. (2016) “Metaphysical Rationalism,” Noûs, 50(2), 379–418. Denby, D. (2014) “Essence and Intrinsicality,” in R. Francescotti (ed.), Companion to Intrinsic Properties, Berlin: De Gruyter, 87–109. De Melo, T. X. (2019) “Essence and Naturalness,” The Philosophical Quarterly, 69(276). 534–554. De Rizzo, J. (2022) “A Ground-Theoretical Modal Definition of Essence,” Analysis, 82(1), 32–41. Ditter, A. (2020) “The Reduction of Necessity to Essence,” Mind, 129(514), 351–380. Ditter, A. (2022) “Essence and Necessity,” Journal of Philosophical Logic, 51, 653–690. Dixon, T. S. (2016) “What is the Well-Foundedness of Grounding?,” Mind, 125(498), 439–468. Dorr, C. (2016) “To Be F Is To Be G,” Philosophical Perspectives, 30, 39–134. Dunn, J. M. (1990) “Relevant Predication 3: Essential Properties,” in J. M. Dunn & A. Gupta (eds.), Truth or Consequences, Dordrecht: Kluwer, 77–95. Fine, K. (1994) “Essence and Modality,” Philosophical Perspectives, 8, 1–16. Fine, K. (1995a) “Senses of Essence,” inW. Sinnott-Armstrong (ed.), Modality, Morality, and Belief. Essays in Honor of Ruth Barcan Marcus, Cambridge: Cambridge University Press, 53–73. Fine, K. (1995b) “The Logic of Essence,” Journal of Philosophical Logic, 24, 241–273. Fine, K. (2000) “Semantics for the Logic of Essence,” Journal of Philosophical Logic, 29, 543–584. Fine, K. (2015) “Unified Foundations for Essence and Ground,” Journal of the American Philosophical Association, 1, 296–311. Fine, K. (2020) “Comments on Fabrice Correia’s “More on the Reduction of Necessity to Essence”,” in M. Dumitru (ed.), Metaphysics, Meaning, and Modality. Themes from Kit Fine, Oxford: Oxford University Press, 466–470. Fine, K. (2022) “Some Remarks on the Role of Essence in Kripke’s Naming and Necessity,” Theoria, 88(2), 403–405. Forbes, G. (1985) The Metaphysics of Modality, Oxford: Clarendon Press. Glazier, M. (2017) “Essentialist Explanation,” Philosophical Studies, 174(11), 2871–2889. Godman, M., Mallozzi, A. & Papineau, D. (2020) “Essential Properties are Super-Explanatory: Taming Metaphysical Modality,” Journal of the American Philosophical Association, 3, 1–19. Gorman, M. (2005) “The Essential and the Accidental,” Ratio, 18, 276–289. Gorman, M. (2014) “Essentiality as Foundationality,” in D.D. Novotný & L. Novák (eds.), NeoAristotelian Perspectives in Metaphysics, New York & London: Routledge, 119–137. Hale, R. (1996) “Absolute Necessities,” Philosophical Perspectives, 10, 93–117. Hale, R. (2013) Necessary Beings. An Essay on Ontology, Modality, and the Relations Between Them, Oxford: Oxford University Press. Hale, R. (2018a) “Essence and Existence,” Revista de Filosofia de la Universidad de Costa Rica, 57, 137–155. Hale, R. (2018b) “The Basis of Necessity and Possibility,” Royal Institute of Philosophy Supplement, 82, 109–138. Hale, R. (2021) “Essence and Definition by Abstraction,” Synthese, 198, 2001–2017. Jago, M. (2021) “Knowing How Things Might Have Been,” Synthese, 198, 1981–1999. Kment, B. (2006) “Counterfactuals and the Analysis of Necessity,” Philosophical Perspectives, 20, 237–302.

140

Non-Modal Conceptions of Essence Kment, B. (2014) Modality and Explanatory Reasoning, Oxford: Oxford University Press. Kment, B. (2015) “Replies to Sullivan and Lange,” Philosophy and Phenomenological Research, 91(2), 516–539. Koslicki, K. (2012) “Essence, Necessity, and Explanation,” in T. Tahko (ed.), Contemporary Aristotelian Metaphysics, Cambridge: Cambridge University Press, 187–206. Kripke, S. (1980) Naming and Necessity, Harvard: Harvard University Press. Lange, M. (2015) “Comments on Kment’s Modality and Explanatory Reasoning,” Philosophy and Phenomenological Research, 91(2), 508–515. Leech, J. (2018) “Essence and Mere Necessity,” Royal Institute of Philosophy Supplement, 82, 309–332. Leech, J. (2021) “From Essence to Necessity via Identity,” Mind, 130(519), 887–908. Linsky, B. & Zalta, E. N. (1994) “In Defense of the Simplest Quantified Modal Logic,” Philosophical Perspectives, 8, 431–458. Linsky, B. & Zalta, E. N. (1996) “In Defense of the Contingently Non-Concrete,” Philosophical Studies, 84, 283–294. Lowe, E. J. (2008) “Two Notions of Being: Entity and Essence,” Royal Institute of Philosophy Supplement, 62, 23–48. Lowe, E. J. (2012) “What Is the Source of Our Knowledge of Modal Truths?,” Mind, 121(484), 919–950. Lowe, E. J. (2016) “Metaphysics as the Science of Essence,” in A. Carruth, S. Gibb & J. Heil (eds.), Ontology, Modality, and Mind. Themes from the Metaphysics of E. J. Lowe. Oxford, Oxford: Oxford University Press, 14–34. Mackie, P. (2020) “Can Metaphysical Modality be Based on Essence?,” in M. Dumitru (ed.), Metaphysics, Meaning, and Modality. Themes from Kit Fine, Oxford: Oxford University Press, 247–264. Mallozzi, A. (2021) “Putting Modal Metaphysics First,” Synthese, 198, 1937–1956. Michels, R. (2019) “On How (Not) to Define Modality in Terms of Essence,” Philosophical Studies, 176, 1015–1033. Noonan, H. W. (2018) “The New Aristotelian Essentialists,” Metaphysica, 19(1), 87–93. Oderberg, D. (2007) Real Essentialism, New York: Routledge. Plantinga, A. (1978) The Nature of Necessity, Oxford: Clarendon Press. Rabin, G. O. & Rabern, B. (2016) “Well Founding Grounding Grounding,” Journal of Philosophical Logic, 45(4), 349–379. Raven, M. (2020) “Explaining Essences,” Philosophical Studies, 178(4), 1043–1064. Rayo, A. (2013) The Construction of Logical Space, Oxford: Oxford University Press. Rayo, A. (2015) “Essence Without Fundamentality,” Theoria, 30(3), 349–363. Robertson, T. (2022) “Everything but the Kitchen Sink: How (Not) To Give a Plenitudinarian Solution to the Paradox of Flexible Origin Essentialism,” Philosophical Studies, 179, 133–161. Romero, C. (2019) “Modality is not Explainable by Essence,” The Philosophical Quarterly, 69 (274), 121–141. Rosen, G. (2010) “Metaphysical Dependence: Grounding and Reduction,” in R. Hale & A. Hoffman (eds.), Modality: Metaphysics, Logic, and Epistemology, Oxford: Oxford University Press, 109–136. Rosen, G. (2015) “Real Definition,” Analytic Philosophy, 56, 189–209. Teitel, T. (2019) “Contingent Existence and the Reduction of Modality to Essence,” Mind, 128(509), 39–68. Torza, A. (2015) “Speaking of Essence,” The Philosophical Quarterly, 65(261), 754–771. Van Cleve, J. (2018) “Brute Necessity,” Philosophy Compass, 13 (9), e12516. Vogt, L. (forthcoming) “Two Problems for Zylstra’s Truthmaker Semantics for Essence,” Inquiry: An Interdisciplinary Journal of Philosophy. Wallner, M. & Vaidya, A. (2020) “Essence, Explanation, and Modality,” Philosophy, 95(4), 419–445. Werner, J. (2021) “Contingent Objects, Contingent Propositions, and Essentialism,” Mind, 130(520), 1283–1294. Wildman, N. (2013) “Modality, Sparsity, and Essence,” The Philosophical Quarterly, 63(243), 760–782. Wildman, N. (2021) “Against the Reduction of Modality to Essence,” Synthese, 198(Suppl 6), S1455–S1471.

141

Fabrice Correia Williamson, T. (2013) Modal Logic as Metaphysics, Oxford: Oxford University Press. Zalta, E. N. (2006) “Essence and Modality,” Mind, 115(459), 659–694. Zylstra, J. (2019a) “Collective Essence and Monotonicity,” Erkenntnis, 84, 1087–1101. Zylstra, J. (2019b) “Constitutive and Consequentialist Essence,” Thought: A Journal of Philosophy, 8, 190–199. Zylstra, J. (2019c) “Making Semantics for Essence,” Inquiry: An Interdisciplinary Journal of Philosophy, 62(8), 859–876.

142

9 ESSENCES OF INDIVIDUALS 1 Marco Marabello

9.1

Individuals, Kinds, and Essentialism

A common distinction is the one drawn between individuals and kinds. On the one hand, individuals are entities such as the chair where I am now sitting, my cat Aristotle, the particles that compose the chair and my cat, and the 2023 Rugby World Cup, that is, particular objects or events (cf. Adams 1979). On the other hand, kinds are entities such as chairs, cats, and world-cup finals,2 that is, groupings of particular objects or events. Essentialism is usually characterized as the thesis that, for at least some objects, the properties those objects have can be divided between essential and accidental ones, between properties that an object couldn’t lack and properties that it could lack or, alternatively, properties that are definitive of the object’s nature and properties that aren’t. Furthermore, it is customary to use the term “essence” to denote the property or properties essential to a certain entity. This is a mild version of essentialism, and although there are others (cf. Robertson Ishii and Atkins 2022), here I will just assume this characterization as it is, by and large, the most common. For the present purposes, I also won’t discuss the question of what an essential property is. Whether the concept must be analyzed by means of modality or otherwise taken as primitive is an interesting question in its own but one that I will not tackle here (for more on this see Correia, this volume, and Torza, this volume). When this is necessary, I will refer to these two conceptions of essence as “modal conception” and “non-modal conception”, respectively. I will talk, though, of possible worlds, but this is meant just as a useful heuristic rather than an endorsement of the modal conception on my part. Now, granting the distinction between individuals and kinds, and that at least some objects have at least some essential properties, one faces two distinct types of questions: (1) what is essential to an individual? (2) what is essential to a kind? That is, what makes a certain individual, be that my cat, Socrates, or what have you, the individual it is? And what makes a certain kind, be that the kind chairs, the kind cats, or what have you, the kind it is? This chapter will survey various answers to the first question (cf. Plantinga 1974: ch. IV §10), while setting aside the second (see Tahko, this volume; Brigandt, this volume, for some answers to the second question). However, before plunging into these answers, a few remarks—on the individual/kind distinction—are in order.

DOI: 10.4324/9781003008750-12

143

Marco Marabello

First, as examples of an individual, I mentioned only concrete entities; however, an individual may also be abstract, e.g., sets (cf. Forbes 1985: 99) and propositions. In this chapter, I won’t be concerned with what is essential to abstract individuals, although some of the answers to be discussed could very well apply to them. Second, I used the term “particular” to gloss what an individual is. It is important to notice that one could use the two terms “particular” and “individual” to refer to different things. Lowe (2009: ch 1), for instance, uses the term “particular” to refer to a class of entities which includes tropes or property instances. According to him, these latter entities are examples of particulars which are not individuals. Third, the distinction between individuals and kinds has nothing to say about which one, if any, is more fundamental than the other. One could reasonably hold that individuals are not fundamental (Dasgupta 2009), but even in this case, one can ask what is essential to an individual, for the question of how a certain individual might have been would (arguably) still arise. Finally, the existence of two distinct questions, (1) and (2) above, depends on what one takes a kind to be. For one could take a kind to be just an individual. For instance, Hawley and Bird (2011: 206) maintain that there are three possibilities about the metaphysical status of kinds, either they are universals, or particulars, or sui generis entities (neither universals nor particulars). Under the second of these three possibilities falls the thesis according to which kinds are just individuals (see Brigandt, this volume, about species). More precisely, under this conception, kinds are spatio-temporal extended entities whose parts are the kind’s members. Clearly, if this is the correct metaphysics of kinds, there would be no reason to maintain that (2) is a distinct question from (1). Be that as it may, I will now turn to possible answers to question (1). In this chapter, I will just consider two answers: that individual essences (Section 9.2) and sortals (Section 9.3) are essential to individuals. Another answer, namely, that an individual’s origins are essential to it, is discussed in a separate chapter (see Robertson Ishii, this volume), and thus it won’t be considered here.

9.2

Individual Essences

A first answer to the question of what is essential to an individual is that its individual essence is. Proponents of this answer hold a thesis that one might call Individual Essentialism (IE), viz., that there are individual essences. But what is an individual essence? Roughly, an individual essence of an object is a property (or properties) that is necessarily unique to that object. Importantly, the uniqueness condition must hold across possible worlds, that is, no other individual y, actual or possible, could instantiate the individual essence of a distinct individual x. More precisely, an individual essence of an entity is, necessarily, a property (or properties) which is both necessary and sufficient for being that entity (cf. Plantinga 1974: ch. 5; Forbes 1985: 96; Mackie 2006: 19; Roca Royes 2011: 72). Some friends of possible worlds hold, moreover, that individual essences provide sufficient and necessary conditions for the trans-world identification of individuals. Now, granted that an individual essence is an essence of an individual, one faces two questions: first, what could play the role of an individual essence? Second, what, if any, are the reasons to hold that (IE) is true? I will consider these questions in turn. To begin with, Sean Connery’s property of being identical to Sean Connery seems a perfect candidate for being an individual essence of Sean Connery. For, necessarily, Sean Connery has this property if he exists, and necessarily, if some individual has this property, then he is Sean Connery. But properties of this kind have been deemed trivial by many (Della Rocca 1996; Forbes 1985: 96; Mackie 2006: 20), and thus, not apt to play the role that individual essences

144

Essences of Individuals

are supposed to play. Examples of properties that may play the role of non-trivial individual essences are Plantinga’s (1974) world-indexed properties, primitive haecceities or thisnesses (Rosenkrantz 1993), and individual forms (Koslicki 2018; 2020). First, world-indexed properties are properties with the following structure: having P in w, where P is a property and w a world. Roughly, they ascribe to an individual a feature P that the individual has in a certain world w (see Plantinga 1974: 63 for a precise characterization). For instance, Socrates’ property of being a citizen of Athens in @. World-indexed properties clearly are essential properties, for at any world in which Socrates exists, necessarily he has the property of being a citizen of Athens in @. But world-indexed properties can also play the role of individual essences when the feature that they ascribe to an individual in a world is unique to that individual in that world. Take the world-indexed property of being the protagonist of North by Northwest in @. This is an individual essence of Cary Grant, for necessarily, if Cary Grant exists, then he has this property, and necessarily, anyone who has that property will be identical to Cary Grant (see Koslicki 2020 for a critique of world-indexed properties as individual essences). Second, primitive haecceities or thisnesses could play the role of individual essences. The notion of a haecceity has a venerable pedigree in the history of philosophy and it is customary, but not uncontroversial (e.g., Plantinga 1976: 149), to credit John Duns Scotus as the first to introduce it in the philosophical debate (See Cross 2022 for more on the history of the notion). Although different authors understand different things by “thisness” or “haecceity”, if one understands a haecceity as an entity h that a certain individual a instantiates and that individuates that individual, then this entity h could play the role of an individual essence of a. For instance, Sean Connery, and only him, instantiates his haecceity, namely, being Sean Connery in every world in which he exists. However plausible, positing haecceities, at least if understood as necessarily existing entities à la Plantinga (1976), comes with a high ontological cost. Since it is usually assumed that all properties can be multiply instantiated and ground qualitative resemblances, haecceities cannot be grouped with other properties and thus they require accepting an additional ontological commitment (See Diekemper 2015 and Cowling 2022 for more on this). For these reasons, many are prone to reject them or to adopt a more lightweight characterization of haecceities (cf. Adams 1981). But besides their ontological cost, there is a real question of whether these haecceities could be essential, let alone individual essences. Certainly, according to some modal accounts of essence, this is not so. For instance, Denby (2014: 90) argues that for a property to be essential to an object, it must be a “core” and “central” property of that object, where a property is a core one when it is related to the “interesting features of things, such as their qualitative natures, their powers, and their causal relations to other things” (ibid.). But according to him, thisnesses aren’t “core” or “central” properties of objects in this sense, therefore they are not adequate candidates for essentiality. But then haecceities couldn’t play the role of individual essences either. Finally, for those with hylomorphist sympathies, individual forms may play the role of individual essences for those individuals that are compounds of matter and form (Koslicki 2018, ch. 3.4; 2020: §3.5). If an individual x in a certain world w has a certain individual form f, then, at any world, having f will be both necessary and sufficient for being x. Since the numerical identity of individuals across worlds is determined by the identity of their individual forms, proponents of this strategy face a question: what grounds the identity of individual forms? Here, the friends of individual forms have two possibilities (Koslicki 2020: 132), namely, either (i) grounding facts about the identity of individual forms in facts about the identity of other entities or (ii) taking the identity of individual forms as primitive. Koslicki

145

Marco Marabello

(2020) takes the latter route for the purposes of her discussion, however, she leaves open the possibility of taking the former (see Fine 2020 for a critique of Koslicki’s position). But why think that individuals have individual essences? That is, why think that (IE) is true? Although arguments in support of individual essences are put forward, amongst others, by Forbes (1985: chs. 5 and 6), Mackie (1987, 2006: ch. 2), and Plantinga (1974), for the present purposes I will just tackle Mackie (2006: ch. 2, §5)’s Indiscernibility Argument. The argument proceeds from three assumptions: A1. There cannot be bare identities and non-identities. A2. Identities and non-identities cannot be extrinsically grounded. A3. The correct way to interpret de re modal claims is by means of transworld identity. As they play a central role in the overall argument, let me comment on these three assumptions before proceeding with the rest of the argument. First, (A1), or as Mackie (1987: 180–1) calls it “The No Bare Identities Principle”, holds that facts about identity and non-identity must always be grounded. In other words, given a fact about the identity (or non-identity) of two objects x and y, this fact must be grounded in some features of x and y themselves. Second, (A2) restricts the facts that can ground identity (or non-identity) claims to facts about intrinsic features of the entities whose identity (or non-identity) is under discussion. Third, (A3) says that claims about de re necessity and possibility are rightly understood as claims about the transworld identity of those entities which figure in these claims, rather than as claims about their Lewisian counterparts. In other words, that Obama could have lost the 2008 election is rightly understood as a claim about an individual numerically identical to the actual Obama, rather than his counterpart, who in some world loses the 2008 election. Keeping these three assumptions in mind, we can now proceed to put forward Mackie’s Indiscernibility Argument whose primary goal is to lend plausibility to (IE). I will start by spelling out the argument in premise/conclusion form and then comment on each step. P1. At least some individuals do not have an individual essence. P2. For any two individuals x and y which do not have an individual essence, it is possible that x and y have the same essential properties but different accidental ones. P3. Let c and d be individuals which do not have an individual essence (from P1). C1. It is possible that c and d have the same essential properties but different accidental ones (from P2 and P3). P4. Possibly, any difference between c and d is due entirely to their accidental properties (from C1). P5. If it is possible that any difference between two individuals x and y is due entirely to their accidental properties, then it is possible that any way x could be is a way y could be. C2. Any way c could be is a way d could be and any way d could be is a way c could be (from P4 and P5). P6. If any way c could be is a way d could be and any way d could be is a way c could be, then it is possible that two worlds w1 and w2 are exactly alike in all respects expect for the identities of c and d. C3. It is possible that two worlds w1 and w2 are exactly alike in all respects except for the identities of c and d. The argument is indeed complex and deserves careful consideration. To start with, it is important to notice that the argument constitutes a reductio ad absurdum. Since the argument’s aim, as already hinted above, is to lend plausibility to (IE), Mackie supposes that some objects do not have individual essences (P1) and goes on to see what this entails.

146

Essences of Individuals

If two individuals x and y do not have individual essences, then it is possible that those individuals x and y have exactly the same essential properties but different accidental properties (P2). This follows from the definition of an individual essence. Recall that an individual essence is an essential property (or properties) which is unique to a certain individual. If two individuals have no individual essence then, nothing forbids them from having the same essential properties. Now, let’s assume that, for instance, Cary Grant and Sean Connery are among those individuals which do not have an individual essence (P3). It is then possible that Cary Grant and Sean Connery have exactly the same essence, viz., the same essential properties (C1). Since they share all their essential properties, any actual difference between Cary Grant and Sean Connery will then be entirely due to their accidental properties (P4). For instance, taking the property of being a human being as their only essential property, all differences between Cary Grant and Sean Connery are due to their accidental properties, for instance, the accidental property of having played James Bond that the second has but the first doesn’t. But if this is the case, then there are no ways Cary Grant could have been that Sean Connery could not have been (P5). In other words, and somewhat metaphorically, the possibilities open to one are the same as the possibilities open to the other: any way Cary Grant could be is a way Sean Connery could be. Less abstractly, if being born in the United States is a way Cary Grant could have been, then it is also a way Sean Connery could have been. This means that any sequence of events that could make up Cary Grant’s life is a sequence of events that could make up Sean Connery’s life (C2).3 But then, consider Sean Connery’s possible life in which he is a philosopher instead of an actor. Since any way Sean Connery’s life could be is a way Cary Grant’s life could be, the very same life is a possible life of Cary Grant (P6). Thus, consider two worlds w1 and w2. These two worlds are exactly alike, except that in w1 Sean Connery is living his philosopher life while in w2 Cary Grant is (C3). Since w1 and w2 differ only with respect to who is living the philosopher life, the two worlds differ only with respect to the identity of the individual who leads that life. There is no matter of fact as to whether the individual in w1 is Sean Connery or Cary Grant. And there is no matter of fact as to whether the individual in w2 is Cary Grant or Sean Connery. After all they share all their essential properties. This scenario delivers a case of a “bare difference” (Mackie 2006: 26) in the identities of Cary Grant and Sean Connery since the facts about their identity do not obtain in virtue of any other fact about their intrinsic features. Therefore, by assuming (P1), we have derived a conclusion which is a case of bare nonidentity, and this case contradicts our assumption (A1), that is, that there cannot be cases of bare identities and non-identities. Mackie pushes the argument further by pointing out that since, by assumption, Cary Grant and Sean Connery share all their essential properties, then any way Cary Grant could have been is a way Sean Connery could have been. But given that whatever is actual is possible (note that this is the dual of axiom T of modal propositional logic, viz., that whatever is necessary is actually the case), it follows that how Cary Grant actually is is a way Cary Grant could have been and thus, a way Sean Connery could have been. Then, one can imagine a world w1 where there is an individual, call him q-Grant, who lives a life indistinguishable from that of Cary Grant but who is nevertheless identical to Sean Connery. Moreover, one can also imagine a world w2 where there is an individual, call him qConnery, who lives a life indistinguishable from that of Sean Connery but who is nevertheless identical to Cary Grant. Surely though, one has no reason to think that these two worlds couldn’t be one and the same. Therefore, we arrive at a world ws completely indistinguishable from the actual one, except for the fact that Cary Grant and Sean Connery have switched

147

Marco Marabello

their roles. Here again, we have a case of bare difference since there is no fact about the two individuals’ intrinsic features in virtue of which Cary Grant in @ is identical to q-Connery in ws and Sean Connery in @ is identical to q-Grant in ws (Mackie 2006: 27–8; cf. Chisholm 1967 and Adams 1979). In this case too, the assumption that some individuals do not have an individual essence entails that there are cases of bare identities and non-identities. But as we’ve seen, this contradicts the “No Bare Identities Principle” (A1). There are four possible ways to react to the argument just given. To begin with, one may reject (P1), that is, that at least some individuals do not have individual essences. But since the argument proceeded from the three assumptions (A1), (A2), and (A3), one could also reject any of these assumptions to avoid the contradiction. In other words, one may argue that bare identities (and non-identities) aren’t problematic, that identities (and nonidentities) can be extrinsically grounded, or that transworld identity does not provide the correct interpretation of de re modal claims. Each of these possibilities seems to be a trade-off. Let’s look at them in turn. First, one can resist the argument by holding that all individuals have individual essences. But those sympathetic to this strategy should provide plausible candidates to play the role of (non-trivial) individual essences. Forbes (1985), for instance, argues that this is not possible for certain categories of objects and thus, he adopts a piecemeal solution to the puzzle. According to Forbes, for some categories of objects, namely biological organism, one can find plausible candidates to play the role of individual essences, while for other categories this is not possible. For the latter categories, one should adopt a counterpart model of de re modal claims instead of a transworld identity model. Mackie (2006: ch. 3) argues at length against Forbes’ attempt to find (non-trivial) individual essences for such individuals, but also generalizes her argument to draw the conclusion that “the attempt to find such individual essences is doomed to failure” (Mackie 2006: 69). Here two points can be made respectively against Mackie’s argument and Forbes’ strategy. First, it is not clear why one must find plausible candidates for individual essences. After all, it may be sufficient to show, as Mackie’s indiscernibility argument does, that rejecting individual essences has unpalatable consequences, while leaving it open what these individual essences are. Second, against Forbes’ strategy, one could argue that there is no reason to endorse a piecemeal solution. After all, it is not clear why one should not adopt a counterpart-theoretic approach across the board once it is accepted for certain entities. A second possibility to resist the argument is to deny that bare identities and non-identities are problematic. This is indeed Mackie’s preferred strategy to avoid the contradiction engendered by the argument. Mackie’s choice follows from the exclusion of all other relevant possibilities rather than from a positive argument in favor of bare identities. Nevertheless, it is not clear what really supports (A1). As Forbes (1985: 125) writes: “It is difficult to find an irresistible argument for the principle that facts about identities and differences must be grounded in some way”. Forbes (ibid.), though, holds that “it is part of our concept of identity, whether transworld or transtemporal, that there are no ungrounded facts about such identities”, and supports this claim by presenting some cases which allegedly show the unintelligibility of bare identities. For instance, he considers the case of a world exactly like ours but where the famous steel tower in Paris is different from the actual one. But to make sense of this situation one cannot imagine, e.g., that the material from which the tower was made is plastic rather than steel, but only that the scenario differs in the identity of the tower. Forbes (1985: 125) concludes that the seeming unintelligibility of such scenario is a “measure of the plausibility of the view that transworld differences must be grounded”.

148

Essences of Individuals

As a third possible way out, one could reject the thesis that identities cannot be extrinsically grounded (cf. Forbes 2002 for more on the notion of identities that are extrinsically grounded). Indeed, that facts about identity can be extrinsically grounded is something accepted when discussing cases of identity over time. For instance, Parfit (1984) holds that whether an individual x at t1 is identical to an individual y at t2 depends on the absence of other candidates for the identity with x. Clearly, cases about identity over time are different from cases about identity across possible worlds; however, Mackie (1989; 2006: ch. 4) argues against this way of resisting the argument because, according to her, to avoid conflict with the logic of identity, the friends of extrinsically grounded facts about identity must resort to bare identities. Finally, as already pointed out, one could resist the argument by abandoning the transworld identity interpretation of de re modal claims and adopting a counterpart theoretic interpretation. According to this latter interpretation, individuals are world-bound, i.e., they exist only at one world. For example, that Socrates could have been a doctor, according to this approach, should be understood as claiming that at least one of his counterparts is a doctor. This interpretation of de re modal statements naturally goes together with a Lewisian picture of possible worlds, where possible worlds are worlds as concrete as ours, since it is otherwise difficult to imagine how one and the same individual could inhabit these many concrete possible worlds. However, as Mackie (2006: ch. 5) points out, for those without sympathy for Lewis’ concrete modal realism, the solution seems unpalatable. In sum, one can resist the Indiscernibility Argument in multiple ways, however, all these present tradeoffs. Mackie argues that accepting bare identities is the best possible solution, even if bare identities sound paradoxical since commitment to them forces us to accept “that the total history of the world, up to the time that a particular individual (you, for example) came into existence at a certain place was insufficient for the individual that then and there came into existence to be that individual, rather than something else” (Mackie 2006: 92).

9.3 Sortal Essentialism The biblical story of Lot’s wife tells that, while fleeing Sodom, Lot’s wife turned to look at the city and became a statue of salt. In Die Verwandlung, Kafka narrates that Gregor Samsa woke up one day to find himself turned into a vermin. J.K. Rowling’s character Professor McGonagall is able to turn into a cat at will. If these cases defy your imagination, then you probably have the intuition that Lot’s wife couldn’t have turned into a statue of salt, Gregor Samsa couldn’t have turned into vermin, and McGonagall couldn’t turn into a cat at will. This is the position adopted by sortal essentialists.4 Sortal Essentialism (SE) is yet another answer to the question of what is essential to an individual. It is the thesis that some sortal to which the individual actually belongs is essential to it. For the sake of clarity, some remarks over the use of the term “sortal” are in order. An important distinction to draw is that between sortal terms, sortal concepts, and sortals. First, sortal terms are terms such as “shark”, “river”, “carrot”, etc … Second, sortal concepts are the concepts “expressed or conveyed” (Lowe 2007: 515) by sortal terms. Thus, the concept of a shark is conveyed by the sortal term “shark”. Finally, sortals (also known as kinds) are the entities to which the sortal terms and concepts refer. Although determining what kind of entities sortals are is beyond the scope of this chapter, I will refer to sortals as properties to which individuals belong (note that this need not be the case, cf. Grandy and Freund 2021). So, assuming that (SE) is true, if Lot’s wife is a human being, then she couldn’t have failed to be a human being. That is, necessarily, at any world in which Lot’s wife exists, she is a human being. Or take the sword Excalibur, if it exists, then it is necessarily a sword (cf. Baker 2007:

149

Marco Marabello

ch. 3; Passinsky, this volume for essentialism about artifacts). Similarly, one could hold that if a certain event belongs to a certain sort, then it is necessarily of that sort. For instance, assuming that being a killing is a sort, Kennedy’s killing is necessarily a killing (cf. Forbes 1985: ch. 8; Bennett 1988: ch. 4 for essentialism about events). By and large, (SE) is regarded as a plausible thesis (but see Mackie 1994; 2006: chs. 7–8 for arguments against the view). But any friend of (SE) must answer three distinct, although related, questions. First, what is a sortal? Second, assuming that an individual may fall under different sortals, to which sortal does the individual essentially belong and to which sortals does it belong only accidentally? And third, why think at all that some sortals are essential properties of individuals? I will tackle these three questions in turn (cf. Mackie 2006: 119). The notion of a sortal may be familiar to most; familiar as it may be, however, there is no characterization which is widely agreed upon. Nevertheless, how one characterizes the notion of a sortal is crucial: different characterizations will yield different results about which sortal, if any, is essential to a certain individual, at least if (SE) is not properly restricted. I will consider some ways of characterizing sortalhood, which however are not exhaustive (see also Feldman 1973 for a thorough discussion of some characterizations). One way of characterizing sortal properties is to hold that sortal properties are those that provide an answer to What is it? questions. This characterization goes back at least to Aristotle’s Categories (see Malink, this volume for more on this). Wiggins (2001: 8) seems to have something like this characterization in mind (cf. Grandy and Freund, 2021; Feldman 1973: 269). But this criterion has at least one problem. Answers to What is it? questions may be provided by properties such as kitten, or child. But clearly one cannot hold that being a kitten is essential to a certain cat, since this would amount to holding that when the kitten grows up, it will be a different individual. A second possible characterization is the mereological one (Feldman 1973: 273 and ff.). Roughly, a property is a sortal if and only if the property is not instantiated by proper parts of an object x, if the property is instantiated by x. This criterion should strike anyone as inadequate since there are clear counterexamples. Consider a subatomic particle which has other subatomic particles as proper parts. For instance, a proton which is composed of three quarks. If the mereological characterization is correct, then being a subatomic particle wouldn’t count as a sortal. But this seems implausible. Furthermore, coupled with (SE), this characterization seems even more implausible. For one can reasonably hold that a particular subatomic particle, say a neutron, is not essentially a neutron. As a matter of fact, the phenomenon of beta minus decay consists in the transformation of a neutron into a proton together with an electron and an antineutrino (Khalidi 2013: 24, fn. 14). Since a neutron actually can turn into a proton, a fortiori a neutron could have been a proton. However, it seems plausible to hold that, while a neutron could have been a different kind of subatomic particle, it couldn’t have been something different from a subatomic particle. That is, while a neutron is not essentially a neutron, a neutron is essentially a subatomic particle. But the mereological characterization of sortalhood does not allow this result. A third way of characterizing sortalhood is by means of the notion of counting (Strawson 1959: 168; Wallace 1965: 9; Wiggins 1967: 1). According to this characterization, a property is a sortal just in case it provides a “criterion for counting” the entities that fall under that sortal (Wallace 1965: 9). The gist of the idea is that asking how many Fs there are makes sense if and only if “F” stands for a sortal property (see Feldman 1973 for different ways of cashing out this idea). Thus, since as you enter the seminar room on your first day of class, it makes sense to ask how many students there are, then being a student counts as a sortal property. Clearly though, if one holds (SE) together with this extremely liberal characterization of a

150

Essences of Individuals

sortal property, then one should provide a way to differentiate those sortals that are essential from those that aren’t, for properties like being a student cannot seriously be taken as essential. The latter characterization is intertwined with the idea that sortals provide principles of individuation or criteria of identity (Mackie 2006: 120). Roughly, a criterion of identity is one that allows us to say if a certain individual x who belongs to sortal F is numerically identical to an individual y belonging to the same F (Lowe 2007: 515). Although the view that sortals are necessary to provide criteria of identity has been challenged (Ayers 1974), it seems plausible to hold that if sortals are supposed to give criteria for counting, they must provide criteria of identity. After all, how could you say how many cats are in the house if you cannot say whether the cat that was in the bedroom is the same as the one that is now in the kitchen? As Mackie (2006: 121) notes characterizing sortals as those properties that provide principles of individuation yields the result that, say, being a red thing is not a sortal while being a human being is (see Geach 1980: 63–4). Still, it seems that one could fall under different sortals if these provide the same criteria of identity (Lowe 2007). For instance, a particular cat falls under the sortal being a cat and being a feline. But then, if (SE) is true, which sortal—among those under which it actually falls—is essential to an individual? It seems evident that no matter which characterization of sortalhood one chooses, problems arise when (SE) comes into the picture. The friend of (SE) here has two moves, which do not exclude each other, at her disposal: either to find a characterization of sortal property that does not yield implausible results when coupled with (SE), or to restrict (SE) in a plausible way. The latter move seems to be the most palatable. After all, there seems to be no reason to hold an unrestricted version of (SE), that is, a thesis that holds that any sortal to which an individual belongs is essential to it. But which are the sortals that are essential to a certain individual? A first try consists in distinguishing between substance and phase sortals. Phase sortals, viz., sortals to which an individual belongs only for a certain time of their existence, are sortals such as being a kitten or being a child. Substance sortals, on the other hand, are sortals to which an individual belongs for the whole duration of its existence. Using Mackie’s (2006: 121) slogan, for any substance sortal F: “once an F, always an F”. Still, it should be clear that, although necessary, being a substance sortal is not sufficient for being an essential sortal. For there is no implication between the fact that an individual x was, is, and always will be an F and the fact that necessarily, if x exists, x falls under F. For instance, even if being a human being is a substance sortal of Socrates, this would rule out only Kafkian scenarios where Socrates is a human in 401 BCE and a gadfly (not in the metaphorical use of Plato) in 400 BCE, but not a scenario in which Socrates, at some other possible world, has been for the whole of his life a gadfly (Mackie 2006: 122). So, an essential sortal must meet some further condition (or conditions) that, together with being a substance sortal, jointly suffices for being an essential sortal. David Wiggins argues that being an ultimate (or fundamental) sortal provides this further condition. According to Wiggins, ultimate or fundamental sortals are the most general sortals that correspond to a criterion of identity (cf. Mackie 2006: 132) who uses “principle of individuation” and “criterion of identity” interchangeably). For instance, being a human being is, for Wiggins, an instance of an ultimate sortal (Roca Royes 2011: 70), and thus, Mr. Samsa is essentially a human being. Wiggins’ restriction of (SE) to ultimate sortals allows one to solve puzzling cases such as the one presented above of a cat that falls under the sortal being a cat and being a feline. Although both these sortals provide the same criterion of identity, only one can wear the badge of ultimate sortal, namely being feline.

151

Marco Marabello

One question remains to be answered: why should one hold that certain sortals are essential to individuals? Both Brody (1980) and Wiggins (1980; 2001) provide some reason to hold (SE). Roughly, Brody’s particular theory of counterfactual possibilities allows him to infer that substance sortals are essential sortals. Brody’s proposal, in short, is to understand counterfactual possibilities in terms of possible futures and possible pasts. Possible futures and pasts, in Brody’s model, overlap (at some time) the actual life of an object. Since an individual who falls under a substance sortal at some time of its life, falls under it at all times of its life, it follows from the fact that possible futures and pasts overlap the actual life of the individual, that the individual will fall under that sortal in every possible future or past. Therefore, according to Brody, substance sortals are essential but this is in virtue of the model of counterfactual possibilities that he endorses (Brody 1980: 123). In the case of Wiggins, the heavy lifting is done by the notion of a criterion of identity. Wiggins (1980: 122)5 explicitly holds that an individual could not have a different principle of individuation (criterion of identity) from its actual one. Mackie (2006: 131) calls this principle “Essentiality of a Principle of Individuation”. Since Socrates’ ultimate sortal is being human, being human determines his actual criterion of identity. At any other world w, where Socrates exists, Socrates has this same criterion of identity. But then Socrates couldn’t fall under a different sortal since, e.g., the sortal being a gadfly provides a different criterion of identity that is incompatible with the one provided by being human (cf. Lowe 2007 who holds a similar view). As already mentioned, there seems to be a widely shared endorsement of (SE), at least among those philosophers with essentialist sympathies. Indeed, few have argued against (SE), besides Mackie (2006: chs. 7–8), who was cited above. Examples of arguments against (SE) can be found in Wetzel (2000) and more recently in Grandjean (2022). But despite this wide consensus, it is not clear what would be lost if one were to reject (SE). For one, Forbes (2017: 886) argues that a rejection of (SE) entails a commitment to bare particulars, viz. propertyless substrata which can instantiate any property whatsoever. More precisely, Forbes points out that if (SE) were false, then that Socrates could be a gadfly (in a non-metaphorical sense) would be true. But this, according to him, does not represent a genuine possibility for Socrates. For Socrates to be a gadfly one should be able to conceive two different states of affairs in which one and the same thing, i.e., Socrates, is a human being and a gadfly, respectively (Forbes: ibid.). But to make sense of this scenario, one has to commit to the bare particulars picture of objects. Clearly, determining whether bare particulars are themselves something to be avoided is beyond the scope of this chapter, but those who are not sympathetic to (SE) should be aware of the possible consequences of rejecting it. Arguably, another advantage of (SE) is that it provides solutions to puzzles of material constitution such as the puzzle of Dion and Theon. To illustrate this well-known puzzle (see Scarpati, this volume for more on this), suppose that a man, Dion, has a proper part called “Theon” which coincides with Dion’s body minus his left leg. At a certain moment, Dion’s left leg is amputated. Apparently, now we have two distinct objects which are spatio-temporally coincident and constituted of the same matter. But this seems impossible. So, which one went out of existence? Burke (1994) argues that (SE) provides a straightforward and appealing answer: it is Theon who goes out of existence when Dion’s leg is amputated. By (SE), if something is a person, it is essentially so. But this means that if something is not a person, it is essentially not a person too. Before the amputation of Dion’s leg, Theon wasn’t a person and essentially so. After the amputation, Dion’s body minus the left leg is not numerically distinct from Theon even though it is qualitatively identical and spatiotemporally continuous with

152

Essences of Individuals

Theon, since Dion’s body minus his left leg is now a person (Burke 1994: 134). Thus, rejecting (SE) could deprive one of some alleged solutions to material constitution’s puzzles.

9.4 Conclusion In this chapter, I have discussed various takes on what may be essential to individuals. First, I discussed the thesis that individual essences are essential to individuals and I presented Mackie’s Indiscernibility Argument which makes a serious case for individual essences. Mackie herself rejects individual essences and holds that to resist the contradiction engendered by the argument, one has to commit to bare identities. However, despite Mackie’s claim that attempts to find plausible candidates for individual essences are destined to fail, further work may potentially show that there are indeed plausible candidates to play this role. Second, I discussed the thesis known as Sortal Essentialism (SE). As already mentioned, (SE) is, by and large, regarded as a plausible thesis and few people have attacked it. Nevertheless, further work is needed to support the thesis, as its intuitive plausibility does not constitute a good enough reason to endorse it.

9.5 Further Readings Forbes’ The Metaphysics of Modality is an excellent starting point for reading about the issues discussed in this chapter. Although complex, the book is accessible as it provides the reader with all the philosophical tools needed. The printed copy is unfortunately out of press, but Professor Forbes kindly shares a free pdf of the book on his website (https://spot.colorado. edu/~forbesg/metmodpage.html). It goes without saying that anyone interested in the topic treated in this chapter will find plenty of ideas in Penelope Mackie’s works. Her excellent, clear, and accessible, book How Things Might Have Been expands on many of her previous works and is highly recommended. The “Transworld Identity” entry of the Stanford Encyclopedia of Philosophy, which she co-wrote with Mark Jago, contains many useful references.

9.6

Related Topics

Correia, Fabrice, “Non-Modal Conceptions of Essence” Brigandt, Ingo, “Biological Species” Malink, Marko, “Essence: Ancient” Passinsky, Asya, “Artifacts, Artworks, and Social Objects” Robertson Ishii, Teresa, “Origin Essentialism” Scarpati, Maria, “Identity, Persistence, and Individuation” Tahko, Tuomas, “Natural Kind Essentialism” Torza, Alessandro, “Modal Conceptions of Essence”

Notes 1 This chapter is dedicated to Penelope Mackie who sadly passed away in December 2022. Although I never met Professor Mackie in person, I am deeply indebted to her. Her contributions, especially those on free will and the metaphysics of modality, have had a huge influence on my own research and interests. I will continue to be inspired by the clarity and depth of Professor Mackie’s writings, which I take as paradigms of excellent philosophy. I would also like to acknowledge the help I received from

153

Marco Marabello

2 3 4 5

the editors of this volume, Kathrin Koslicki, and Mike Raven, and from Vincent Grandjean, Camille Neidhardt, and Maria Scarpati. Following Lowe (2009), if a certain object is, say, a cat, I use the plural of “cat” in italics to refer to the kind to which that object belongs. Mackie (2006: 26) states that the two lives must be specified in an “owner independent” way. That is, there cannot be any specification of one of the two lives that implies who is the “owner” of that life. For more on this see Mackie (2006: Ch. 2, fn. 15). Sometimes this thesis is labeled “Kind Essentialism”. Using the term “sortal” instead of “kind” allows a sharp contrast with the distinct thesis of Natural Kind Essentialism. Interestingly, in his Sameness and Substance Renewed there is no mention of such a principle.

References Adams, R. M. (1979). Primitive Thisness and Primitive Identity. The Journal of Philosophy, 76(1), 5. 10.2307/2025812 Adams, R. M. (1981). Actualism and Thisness. Synthese, 49(1), 3–41. 10.1007/BF01063914 Ayers, M. R. (1974). Individuals Without Sortals. Canadian Journal of Philosophy, 4(1), 113–148. 10.1 080/00455091.1974.10716925 Baker, L. R. (2007). The Metaphysics of Everyday Life: An Essay in Practical Realism. Cambridge University Press. Bennett, J. (1988). Events and Their Names (Repr). Clarendon Pr. Brody, B. A. (1980). Identity and Essence. Princeton University Press. Burke, M. B. (1994). Dion and Theon: An Essentialist Solution to an Ancient Puzzle. Journal of Philosophy, 91(3), 129–139. Chisholm, R. M. (1967). Identity through Possible Worlds: Some Questions. Noûs, 1(1), 1. 10.2307/ 2214708 Cowling, S. (2022). Haecceitism. Stanford Encyclopedia of Philosophy. Cross, R. (2022). Medieval Theories of Haecceity. Stanford Encyclopedia of Philosophy. Dasgupta, S. (2009). Individuals: An Essay in Revisionary Metaphysics. Philosophical Studies, 145(1), 35–67. 10.1007/s11098-009-9390-x Della Rocca, M. (1996). Essentialism: Part 1. Philosophical Books, 37(1), 1–13. 10.1111/j.1468-0149.1 996.tb02508.x Denby, D. (2014). Essence and Intrinsicality. In R. M. Francescotti (Ed.), Companion to Intrinsic Properties. DE GRUYTER. 10.1515/9783110292596.87 Diekemper, J. (2015). The Ontology of Thisness. Philosophy and Phenomenological Research, 90(1), 49–71. 10.1111/phpr.12010 Feldman, F. (1973). Sortal Predicates. Noûs, 7(3), 268. 10.2307/2214351 Fine, K. (2020). Comments on Kathrin Koslicki’s “Essence and Identity.” In K. Fine (Ed.), Metaphysics, Meaning, and Modality (pp. 429–434). Oxford University Press. 10.1093/oso/9780199652624.003. 0026 Forbes, G. (1985). The Metaphysics of Modality. Clarendon Press. Forbes, G. (2002). Origins and Identities. In A. Bottani, M. Carrara, & P. Giaretta (Eds.), Individuals, Essence, and Identity: Themes of Analytic Metaphysics (pp. 319–340). Kluwer Academic. Forbes, G. (2017). Essentialism. In B. Hale, C. Wright, & A. Miller (Eds.), A Companion to the Philosophy of Language (pp. 881–901). Wiley. Geach, P. T. (1980). Reference and Generality (3rd ed). Cornell University Press. Grandjean, V. (2022). The Bare Past. Philosophia, 50(5), 2523–2550. Grandy, R. E., & Freund, M. A. (2021). Sortals. The Stanford Encyclopedia of Philosophy. https://plato. stanford.edu/archives/sum2021/entries/sortals/ Hawley, K., & Bird, A. (2011). What are Natural Kinds? Philosophical Perspectives, 25(1), 205–221. 10.1111/j.1520-8583.2011.00212.x Khalidi, M. A. (2013). Natural Categories and Human Kinds Classification in the Natural and Social Sciences. Cambridge University Press. Koslicki, K. (2018). Form, Matter, Substance. Oxford University Press. Koslicki, K. (2020). Essence and Identity. In K. Koslicki (Ed.), Metaphysics, Meaning, and Modality (pp. 113–140). Oxford University Press. 10.1093/oso/9780199652624.003.0007

154

Essences of Individuals Lowe, E. J. (2007). Sortals and the Individuation of Objects. Mind & Language, 22(5), 514–533. 10.1111/j.1468-0017.2007.00318.x Lowe, E. J. (2009). More Kinds of Being: A Further Study of Individuation, Identity, and the Logic of Sortal Terms (2nd ed). Wiley-Blackwell. Mackie, P. (1987). Essence, Origin and Bare Identity. Mind, XCVI(382), 173–201. 10.1093/mind/ XCVI.382.173 Mackie, P. (1989). Identity and Extrinsicness: Reply to Garrett. Mind, XCVIII(389), 105–117. 10.1093/ mind/XCVIII.389.105 Mackie, P. (1994). Sortal Concepts and Essential Properties. The Philosophical Quarterly, 44(176), 311. 10.2307/2219612 Mackie, P. (2006). How Things Might Have Been: Individuals, Kinds, and Essential Properties. Clarendon Press; Published in the United States by Oxford University Press. Mackie, P., & Jago, M. (2022). Transworld Identity. In The Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/entries/identity-transworld/ Parfit, D. (1984). Reasons and Persons (Issue 2, pp. 311–327). Oxford University Press. Plantinga, A. (1974). The Nature of Necessity. Clarendon Press. Plantinga, A. (1976). Actualism and Possible Worlds. Theoria, 42(1–3), 139–160. 10.1111/j.1755-25 67.1976.tb00681.x Robertson Ishii, T., & Atkins, P. (2022). Essential vs. Accidental Properties. The Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/archives/win2020/entries/essential-accidental Roca-Royes, S. (2011). Essential Properties and Individual Essences. Philosophy Compass, 6(1), 65–77. 10.1111/j.1747-9991.2010.00364.x Rosenkrantz, G. S. (1993). Haecceity: An Ontological Essay. Springer. Strawson, P. F. (1959). Individuals (Reprinted, transferred to digital printing). Routledge. Wallace, J. R. (1965). Sortal Predicates and Quantification. The Journal of Philosophy, 62(1), 8. 10.23 07/2023313 Wetzel, L. (2000). Is Socrates Essentially a Man? Philosophical Studies, 98(2), 203–220. Wiggins, D. (1967). Identity and Spatio-Temporal Continuity. Blackwell. Wiggins, D. (1980). Sameness and Substance. Harvard University Press. Wiggins, D. (2001). Sameness and Substance Renewed. Cambridge University Press.

155

10 NATURAL KIND ESSENTIALISM Tuomas E. Tahko

10.1

Introduction: What Are Natural Kinds?

Natural kind essentialism (NKE) may be understood as a view about the essences of natural kinds themselves, the essences of the members of natural kinds, or both. At a general level, NKE is usually understood as the idea that there is something in common among the members of a given natural kind and that this commonality means that the members of the kind share a natural kind essence. On some views, this commonality is also explained by the shared essence. This rough idea leaves much to be specified, but before we get started, some initial observations should be made. First, the distinction between individual vs. general essences is of crucial importance to NKE. Is it essential to an individual to be of kind K? I will set aside this question of essentialism regarding individuals.1 Instead, I will understand NKE in terms of general essences: each member of a given kind K shares the same general essence of K (cf. Lowe 2008: 35). Every entity will have a general essence in the sense that it is a member of a given ontological category or kind, but it may or may not be essential to the individual that it is a member of that kind. This suggests that natural kind essences are not sufficient for individuating the distinct members of a natural kind.2 The general essence of K may involve one or more properties that are essential to K’s members. But notice that if kinds are understood as something over and above their members (say, as sui generis universals, genuine entities in their own right), then we may also ask what is essential to the kind K (the entity, such as a universal) itself, rather than the individuals that instantiate the kind. So, strictly speaking we can distinguish between three different but closely related views (cf. Bird and Tobin 2022 who distinguish two of these): 1 It is essential to individual X that it belongs to kind K (individual essentialism). 2 Each individual member of a given kind K has a general or natural kind essence, which may consist of one or more properties that are essential to all members of K (NKE). 3 The kind K (a sui generis entity) has an essence, which may also include the fact that each of its members has certain shared properties (sui generis kind essentialism).

156

DOI: 10.4324/9781003008750-13

Natural Kind Essentialism

(1)–(3) are logically independent, although some combinations may be more plausible than others and some views about what type of entities natural kinds are may constrain one’s choices with regard to (1)–(3). There is also some intentional vagueness in the definitions of (2) and (3), since there are a variety of ways in which we may understand the relationship between natural kind essences and the essential properties that are definitive of the kind’s members. Let me note one particularly interesting issue: the possibility of kind change (also noted, e.g., by Keinänen and Hakkarainen 2017, and Bird and Tobin 2022). For instance, is it possible for an individual cat to become a dog while retaining its individual essence, i.e., remaining the same entity? This type of change of species membership would violate (1) but not (2) (assuming that the shared essential properties of K’s members do not include the membership in K). Another potential example, from physics, is β--decay, where weak interaction converts an atomic nucleus into a different atomic nucleus (atomic number increases by one). Markku Keinänen and Jani Hakkarainen (2017) mention one such case, namely, the possibility of the radioactive 14C atom decaying into 14N. Again, this would violate (1) but not (2). Should we accept this type of kind change or remain committed to the essentiality of kind membership? This question is crucial for (1). One possibility would be to deny that there are individual essences like those of atomic nuclei or biological species, given that kind change may be possible in those cases. Alternatively, one might deny that there are any individual essences at all, but in this case (1) becomes redundant. However, I will focus on (2), which is the clearest sense of NKE. In contrast, if one assumes that natural kinds are sui generis universals, then (3) would concern the question of what is essential to universals of this type. This is certainly an interesting question, but one that concerns the metaphysics of ontological categories rather than NKE (see Keinänen and Tahko 2019, and Hommen 2021 for further discussion). We should distinguish NKE from: Dispositional Essentialism (DE), which is a view about properties, stating that at least some of them have dispositional essences, e.g., the tendency for negatively charged particles to attract positively charged particles (e.g., Bird 2007).3 Scientific Essentialism (SE), which is best understood as a collection of views about natural kinds and related topics, such as laws of nature (e.g., Ellis 2001).4 Many proponents of NKE also accept DE or SE. However, DE does not directly concern kinds, but rather properties. The case of negatively charged particles attracting positively charged particles could possibly be explained just with reference to the property charge, or its causal powers, which give rise to the relevant laws of nature. On the other hand, SE is a broader view, which may be considered to partially consist of a version of NKE. I shall assume realism about natural kinds: we cannot reduce natural kinds to some other type of entity. While I will be focusing on an essentialist characterization of natural kinds, it may be useful to start with a more general account of what natural kinds are. Let me quote a classic passage from John Stuart Mill: In so far as a natural classification is grounded on real Kinds, its groups are certainly not conventional; it is perfectly true that they do not depend upon an arbitrary choice of the naturalist (Mill 1843: 720). Setting aside the details of Mill’s own view, which was not essentialist, the key thought is that natural kinds track or reflect some natural or “objective” divisions in mind-independent

157

Tuomas E. Tahko

reality (they “carve at the joints”) (cf. Psillos 2002, Devitt 2005, Lowe 2006, Chakravartty 2007, Tahko 2015a, Bird and Tobin 2022).5 I should immediately note that the mindindependence criterion has recently received increasing scrutiny (see, e.g., Franklin-Hall 2015, Khalidi 2016, Ereshefsky 2018, and Tahko 2022). Nevertheless, realism about natural kinds must be associated with some sense of “objectivity”, in contrast to various forms of conventionalism and conferralism about natural kinds.6 Moreover, my focus will be on the metaphysics of natural kinds rather than on the semantics of natural kind terms, although in some cases these two issues are very closely intertwined (see Koslicki 2008 on this distinction, and Häggqvist and Wikforss 2018 for a contemporary take). As Nathan Salmon (2005) famously argues, we cannot derive metaphysical essentialism from semantic arguments. Kinds may be conceived as simple or complex universals, sui generis entities, homeostatic property clusters, bundles of (natural) properties, fundamental ontological categories (substantial universals), Aristotelian (instantiated) vs. Platonic (uninstantiated) universals, and so on (for examples, see Boyd 1999, Ellis 2001, Lowe 2006, Hawley and Bird 2011, and Hommen 2021). I will be concentrating on classical examples of natural kinds, drawn primarily from the fields of physics, chemistry, and biology, i.e., the “natural” sciences—I will not be discussing psychological or social kinds. Traditionally, discussion of such “higher-level” kinds has had anti-realist connotations, although contemporary literature on natural kinds is less constrained in this regard, and there is no reason in principle why an account of natural kinds could not be extended to cover higher-level kinds as well (see, e.g., Mallon 2016 for a realist account of “human kinds”, i.e., kinds such as mental kinds, which have properties that can be altered by human activity).7 Finally, how do we understand essence or essentialism more generally? This is a topic that comes up throughout this volume and in its introduction, so I will not pursue it here. Many of the versions of NKE that I will discuss assume a modal conception of essence, while some assume a non-modal conception.8 I will not attempt to determine the implications of assuming one or the other view of essence, but it is worth noting that the historical background, from Kripke and Putnam, is intertwined with the modal conception, whereas some contemporary authors (e.g., Kit Fine, E. J. Lowe, Kathrin Koslicki, and Tuomas Tahko) prefer the non-modal conception.

10.2 The Kripke–Putnam Framework NKE was popularized by Saul Kripke (1980) and Hilary Putnam (1975). Their views are often discussed in parallel, but in fact there are important differences, which Putnam (1990) later specified. Putnam expressed doubts about extending his famous Twin Earth scenario across metaphysically possible worlds. The Twin Earth scenario asks us to imagine a planet identical with Earth in every way except for one detail: Instead of water, Twin Earth has a substance that has every superficial characteristic of water, but a radically different molecular constitution, XYZ. So, XYZ is superficially like water, but has a different intrinsic structure. In this scenario, the chemical properties of water are reproduced by some molecular structure other than H2O, namely, XYZ, which is a placeholder for some potentially complex chemical structure.9 Should we consider this substance to be water? The expected reaction is to deny that XYZ is water. On the usual reading of the Twin Earth scenario, the metaphysical possibility of XYZ reproducing the chemical properties of water in some world with alternative laws of physics is assumed, but the conclusion that we are expected to draw from the scenario is that water has its microstructure essentially, where “microstructure” is a placeholder for whatever

158

Natural Kind Essentialism

determines the chemical properties of water. The interpretation of the Twin Earth scenario in the context of NKE is, however, controversial (as discussed in Tahko 2015a). Putnam later suggested that the question about the possible variation of the laws of physics with regard to water “makes no sense”, mainly because of his general suspicion of metaphysical modality (Putnam 1990, p. 70). The suggestion is that we should conceive of the Twin Earth scenario as concerning a remote location in our own universe instead. In contrast, Putnam reads Kripke to be concerned with the stronger reading, involving metaphysical necessity. It is important to note both the exegetical issue as well as its implications for the strength of the underlying claim regarding NKE: do natural kinds have their microstructure by metaphysical necessity (i.e., essentially, given that Kripke and Putnam operate with the modal notion essence) or merely by nomological necessity? This is a question that is closely related to the debate about laws of nature and the discussion regarding Dispositional Essentialism (DE) and Scientific Essentialism (SE). Many proponents of DE, such as Alexander Bird (e.g., Bird 2007, 2018), and the main proponent of SE, Brian Ellis (e.g., 2001), favor a view according to which laws of nature are metaphysically necessary (where the relevant properties/kinds exist). This suggests that natural kinds like water may have their microstructure by metaphysical necessity, as there are scientific reasons to think that a variation in microstructure while retaining the chemical properties of water is not physically possible (see Tahko 2015a). However, some favor a view whereby at least some laws may be metaphysically contingent (e.g., Lowe 2006), which leaves this issue open (albeit Lowe himself thinks that natural kinds and metaphysically necessary laws are closely linked; see Tahko 2015b for discussion). Setting aside these complications, the standard interpretation of what may be called Kripke–Putnam Essentialism about kinds (without necessarily associating this with the actual, considered view that either Kripke or Putnam may have held) may understood as follows: Kripke-Putnam Essentialism (KPE): if every member or sample of the kind K possesses property p, in every possible world, then possession of property p is part of the essence of the kind K, i.e., p is an essential property of K. KPE is often associated with the view that the essential properties of kinds are intrinsic and microstructural, although as we will see, both of these assumptions can and should be questioned. In Kripke’s work, an important part of the background is that theoretical identity statements such as “Water is H2O” should be understood as identities involving rigid designators, making them metaphysically necessary. Rigid designators in general, such as proper names, pick out the same individual in all possible worlds where that individual exists. Since water was discovered to be H2O with the help of empirical research, this makes “Water is H2O” an a posteriori metaphysical necessity. Another classic example, also from Kripke, is the association between a chemical element and its atomic number, e.g., “gold is the element with atomic number 79”. However, it is controversial whether natural kind terms are rigid designators (for discussion, see, e.g., LaPorte 2004, 2012, Soames 2005, Linsky 2006, and Schwartz 2021). For the purposes of outlining the traditional KPE view, let us nevertheless assume that terms like “water” and “gold” are rigid, which entails that in every possible world in which gold exists, its atomic number is 79. Let me now summarize the reasoning behind KPE in a more systematic fashion. It’s a key part of the view that we start from some empirical information. Next, we include the assumption that natural kind terms like “water” are rigid designators and pick out the same microstructure in all possible worlds—this is in line with Kripke, but Putnam’s Twin Earth scenario can be used

159

Tuomas E. Tahko

to further support the underlying intuition, controversial as it may be. If we additionally grant that essential truths are simply truths that hold in all possible worlds (the modal view of essence), then it follows that water has its actual microstructure essentially, and thus “Water is H2O” is a metaphysically necessary a posteriori essentialist truth. This is something that is traditionally thought to be supported by the necessity of identity: if x and y are the same object then it is necessary that they are the same object (see Kripke 1971). The necessity of identity is often considered to be knowable a priori. In summary, the following steps lead us toward KPE (although these are not presented as a logical argument here): 1 An empirical discovery leading to an a posteriori theoretical identity statement S, which associates property p (and only p) with every sample/member of kind K in the actual world. 2 A general a priori principle about the necessity of identity and rigid designation: if the terms “a” and “b” are rigid designators and the objects that they designate are identical, then they are necessarily identical. 3 The statement S involves rigid designators on both sides of the identity and hence expresses a metaphysically necessary, a posteriori identity. 4 Since S states a metaphysically necessary theoretical identity, K possesses property p in every possible world and hence the possession of property p is part of the essence of K, i.e., p is an essential property of K. Each of these steps can be questioned (perhaps with the exception of the first, empirical generalization), and they all involve significant background assumptions. Besides the already mentioned issue regarding intrinsicness (to be discussed in section 3) and the assumption that natural kind terms act as rigid designators, these steps can be questioned on the basis of the underlying assumptions about the connection between microstructural properties and the chemical (superficial) properties of chemical elements like water (to be discussed in section 4), as well as the role of a priori vs. a posteriori content. On the last issue, further scrutiny should be given to (2), i.e., the claim that the only a priori content that is needed to establish supposed metaphysically necessary identities like “Water is H2O” concerns the necessity of identity. It is arguable that further a priori content about the essence of natural kinds would be required, i.e., that there is a necessary connection between the chemical properties of a chemical substance and the microstructure of that substance (see the addenda to Kripke 1980, and Tahko 2009, 2015a, as well as section 4 for further discussion). Let us now move on to a more detailed discussion about the role of intrinsic vs. extrinsic essences in the context of NKE.

10.3

Intrinsic vs. Extrinsic Essences

Traditional examples of NKE tend to mention cases like chemical elements and compounds, which are commonly defined in terms of their intrinsic properties, such as their nuclear charge and molecular composition. The notion of an intrinsic property is familiar from David Lewis, who suggests that they are properties “which things have in virtue of the way they themselves are” (Lewis 1986: 61). Extrinsic properties are properties which things have in virtue of their relations to other things. There are well-known issues concerning the precise understanding of intrinsicness (e.g., Sider 1996), but it will become clear that some natural kind essences would seem to have non-intrinsic elements. So, something must be said about the possibility of extrinsic or relational essences. The need for extrinsic or relational essences has also been recognized in contexts such as the essences of artworks and other artifacts (see Baker 2000).10

160

Natural Kind Essentialism

The view that natural kind essences must be intrinsic is commonly associated with Kripke and Putnam (see, e.g., Okasha 2002 and Needham 2011), but it may not be fully accurate (see LaPorte 2004 and Williams 2011). Neither Kripke nor Putnam formulated intrinsic NKE explicitly, but it is clear from their examples that some internal structure was assumed to be the defining or most important characteristic of many natural kinds, and this is likely to have led to the conception that all natural kind essences are intrinsic. Consider the following passage from Okasha: [B]oth Putnam and Kripke appear to believe that essential properties of species can be found if we penetrate beyond the “superficial characteristics” of organisms into their “hidden structure”. Thus Putnam (1975) claims that the true criterion for being a lemon is having the “genetic code” of a lemon—this, rather than any observable traits, is the essence of lemonhood, he claims (p. 240). Similarly, Kripke (1980) maintains that having the right “internal structure” is the true criterion for being a tiger—this shared “internal structure” is the essence of tigerhood, he thinks (p. 121). (Okasha 2002: 198) It is not difficult to find individual examples or passages from Kripke and Putnam that corroborate the interpretation that they consider natural kind essences to be intrinsic. Consider Putnam’s treatment of lemons and acids, where the “essential nature” is stated to be “chromosome structure, in the case of lemons, and being a proton-donor, in the case of acids” (Putnam 1970: 188). The idea that chromosome structure or similar intrinsic features would conclusively capture the essence of biological species is of course a very controversial claim, and one of the main reasons why intrinsic NKE has been subjected to criticism, as seen above.11 Similar criticisms can be raised for chemical kinds as well (van Brakel 1986, VandeWall 2007, and Needham 2011). The case of acids is worth a specific mention. Strictly speaking, being a proton-donor does not look like a purely intrinsic or even a microstructural property, as it suggests that there is something that the proton is being donated to. Indeed, the definition of acids (and bases) has gone through a variety of changes involving both microand macroscopic characteristics (Stanford and Kitcher 2000: 115 ff.; see also Tahko 2020: 810). Neil Williams has pointed out that Putnam seems to have accepted diseases as natural kinds, and further, that “natural kinds of disease could turn out to have (or do in fact have) a cause as their essence” (Williams 2011: 155). Williams continues: As these causes are partly, if not entirely, extrinsic to the disease instances, the causal relation can be translated into a (non-intrinsic) relational property of the disease instances, from which it follows that the essences of certain natural kinds are relational (and non-intrinsic). (Williams 2011: 155.) If Williams is right, then it would seem that Kripke–Putnam Essentialism (KPE)’s association with intrinsic (and microstructural) essences may have been overstated. Or perhaps more accurately: KPE does not necessarily reflect the commitments of Putnam (or Kripke). So, NKE does not need to be as closely tied to intrinsic essences as it is sometimes perceived to be. This point has been forcefully made in the literature (e.g., Boyd 1999, Griffiths 1999, Millikan 1999, Okasha 2002, LaPorte 2004, and Brigandt 2009). For instance, Samir Okasha (2002) has argued that NKE can be maintained even in the biological context if we accept relationally defined kinds (and hence extrinsic or relational essences). In this context, the causal history or etiology associated with evolutionary process is thought to

161

Tuomas E. Tahko

be in a key role. We may generally talk of etiological kinds to capture those kinds whose members share a common causal history or origin (see Khalidi 2021). Besides biological species (and other biological cases), potential examples include, e.g., cosmological and geological kinds, which are commonly defined at least in part in terms of their causal history (see also Cleland, Hazen, and Morrison 2021). Extrinsic or relational features may however manifest in other ways as well, not necessarily just in terms of etiology or causal history. Recent work on chemical and biochemical kinds has made this quite clear (e.g., Hendry 2006, Tobin 2010, Needham 2011, Tahko 2015a, Bartol 2016, Havstad 2018, and Tahko 2020). Consider this passage from Emma Tobin on “moonlighting” proteins: Some proteins “moonlight” and thus, perform a secondary function in different parts of the organism. There are two types of moonlighting protein: extrinsically structured moonlighting (ESM) and intrinsically unstructured moonlighting (ISM). Extrinsic cases of moonlighting are cases where extrinsic contextual factors affect the functional role of the protein. An example is the presence of globular crystallin proteins, which have two functional roles. These play a structural role in the lens of the eye, and also act as a catalytic enzyme elsewhere in the organism. (Tobin 2010: 42) The functional role of these proteins does not appear to be purely intrinsic, given that it relates to the protein’s behavior in a certain context, such as the lens of the eye in Tobin’s example. Moreover, biological functions themselves are commonly analyzed in terms of their etiology, and indeed some have argued that “history” is an unavoidable element of theories of function (Garson 2019, see also Bartol 2016). But insofar as the biological function of, say, proteins, is part of their essence, we need to account for this aspect as well. Even if intrinsicality cannot be easily maintained, the case of (ESM) described above does not necessarily undermine microstructural essentialism. As Tobin notes, the microstructuralist could perhaps accommodate these cases “since alternative functional roles are determined by changes in the molecular environment (such as localization, ligand binding and so on)” (Tobin 2010: 42). Yet, Tobin does think that (ISM) is a more serious problem for microstructuralism. In this regard, the debate continues (see Goodwin 2011, Havstad 2018, and Tahko 2020). Finally, there have recently been some attempts to revitalize (at least partially) intrinsic biological essentialism (e.g., Devitt 2008, Dumsday 2012, and Austin 2018). Interesting as these attempts are, I will have to omit a more detailed discussion here.12

10.4 Microstructural Essences Even though intrinsic essences and microstructural essences seem to be closely linked, there may be cases where these two come apart. Interestingly, traditional Kripke–Putnam essentialism focuses on superficial or phenomenological, i.e., macroscopic properties, but takes the (potentially intrinsic) microstructure to be definitive of the kind. It is then a further question whether these two can come apart, that is, whether or not there is a 1:1 correlation between the microstructure and the macroscopic properties. Microstructural essentialism is the view that prioritizes microstructure. In fact, the commitment to microstructural essentialism has been so central that, for some, the failure of finding determinate microstructural essences has led to the rejection of higher-level natural kinds, such as biological kinds (Ellis 2001). But as we saw in the previous section, not all apparent threats to microstructuralism are as serious as they may first seem. The core of the problem is that

162

Natural Kind Essentialism

microstructuralism is often understood in a simplified fashion, e.g., not taking into account the 3D-shape of molecules (cf. Tobin 2010, Havstad 2018). Moreover, a sophisticated version of microstructuralism also ought to consider the fact that the same compositional formula of molecules like H2O can be divided into different isomers (i.e., distinct substances with the same compositional formula). One possibility would be to understand the “sameness” of microstructure in a more fine-grained sense, taking into account possible stereoisomers, structural isomers, and so on (see Weisberg 2005 and Tahko 2015a for discussion). Microstructuralism is often also associated with intrinsic essences, but it does not necessarily entail that all microstructural properties must be intrinsic, as we’ve just seen in the case of biological functions in section 3. Recent work on natural kinds has taken a strong stand against microstructuralism. One of the more sophisticated versions of microstructuralism has been put forward by Robin Hendry (2006: 872), who suggests that “water is the substance formed by bringing together H2O molecules and allowing them to interact spontaneously”—the thought here is that in order to account for the properties of water, we need to consider the interactions between the molecules in a body of water. In his forceful critique of microstructuralism, Paul Needham (2011: 16) cites this very passage, but remains unsatisfied of the need for microstructuralism, stating that even if microstructure is indeed present, it doesn’t follow that “the details of the microstructure of any particular substance are reasonably well known, and certainly not that they are independent of macroscopic constraints or somehow determine the macroscopic features of substances or that substances are in some clear sense ‘nothing but’ their microconstituents”. Needham is right to point out the presence of macroscopic constraints as well as the difficulty of specifying what the microstructure of a substance involves in any given case—even if there are some relatively easy cases as well. We saw that in the classic version of NKE, which is often associated (right or wrongly) with Kripke and Putnam, it is considered to be knowable a priori that substances such as water have their actual microstructure essentially. But how exactly should we understand the requirements of microstructural essentialism? In particular, does microstructural essentialism require that microstructure determines the macroscopic features of substances in the sense that Needham seems to assume? It may be that something weaker is sufficient. Consider the following two claims (cf. Tahko 2015a: 804): 1 Necessarily, there is a 1:1 correlation between (all of) the chemical properties of a chemical substance and the microstructure of that substance. 2 Necessarily, a sample of chemical substance A is of the same chemical substance as B if and only if A and B have the same microstructural composition. Notice that neither (1) nor (2) explicitly say anything about the type of connection—say, dependence or determination—that holds between microstructure and macroscopic features. Thinking back to Putnam’s Twin Earth scenario, it seems to focus on (2) rather than (1), since the scenario concerns a case where we encounter a substance, XYZ, which replicates the chemical properties of water. Interestingly, for the Twin Earth scenario to be possible at all, we would have to assume the falsity of (1), since we would then have two microstructures that are associated with (all of) the same chemical or macroscopic properties—a many:1 relationship. While this may be of relatively little consequence regarding the upshot for the semantics of natural kind terms that Putnam was mainly interested in, it does make a difference for the metaphysical reading of the scenario that Kripke and others may have had in

163

Tuomas E. Tahko

mind. The most plausible way around this complication is to read the Twin Earth scenario as concerning merely the epistemic possibility of encountering a substance like water with a different microstructure, and to read (2) in terms of physical necessity, as Putnam (1990: 69) seems to have intended. What about (1)? Importantly, even if a correlation between microstructure and chemical properties does exist, it might not always be possible to specify the microstructure without relying on macroscopic features, like Needham has argued. We have good reasons to think that there is no general formula according to which microstructure determines macroscopic features. So, if microstructural essentialism holds any promise, we must understand it in a more restricted sense than it has been traditionally understood—this seems to be the consensus in recent work (e.g., Tobin 2010, Tahko 2015a, Bartol 2016, Havstad 2018, and Tahko 2020).

10.5

Refined Natural Kind Essentialism

Let us finish by considering more refined or restricted versions of NKE. Consider a recent proposal from Godman, Mallozzi, and Papineau (2020), which is Kripkean in spirit. They suggest that to be an essential property is to be a super-explanatory property. On this view, natural kinds (or their properties) are typically “unified by certain super-explanatory core properties” (Godman, Mallozzi, and Papineu 2020: 2). Such a core property, they propose, is some single property which causally explains the occurrence of the multiple properties shared by instances of a kind. One example they provide is the property of atomic constitution, say, of gold, which explains why all samples of solid gold have the same chemical and physical properties. This proposal has clear connections to Richard Boyd’s homeostatic property cluster (HPC) theory, which has been previously developed in a more essentialist framework by Kathrine Hawley and Alexander Bird (2011). According to the HPC theory, the clustering of properties in members of natural kinds is explained by a shared causal structure among the members of a natural kind; each member of a kind is a property cluster kept in homeostasis by a (often lower-level) causal mechanism. There have been many different versions of this type of causal account of natural kinds (e.g., Khalidi 2013). However, Godman, Mallozzi and Papineu specify that where they differ from the usual causal accounts is that, on their view, “the great preponderance of natural kinds owe their clustering of properties, not just to some causal structure or other, but to one single underlying property that serves as the common cause of all the other clustered properties” (2020: 6). Importantly, the relevant super-explanatory properties may in some cases be intrinsic and microstructural, but Godman, Mallozzi, and Papineu also accept the possibility of shared causal history serving the super-explanatory role. Godman, Mallozzi, and Papineu conclude by saying that their aim is to reduce kind essences to “a specific kind of nomological structure”, namely, their super-explanatory role (2020: 14). The resulting view is interesting, but highly controversial: [I]t is a consequence of our view that modal Kind essences would stay fixed even in possible worlds where the relevant laws of nature were different. Even if variation in laws of nature meant that H2O were no longer odorless, colorless, tasteless, and so on, it would still be water. This result of the view is in clear tension with those accounts of NKE which argue that there is a metaphysically necessary connection between the core properties such as molecular composition and the macroscopic properties of the kind. Compare this with David

164

Natural Kind Essentialism

Oderberg’s neo-Aristotelian essentialist account, whereby the properties that “flow” from the essence of a particular object belonging to a given kind are caused by and originate in the form of that kind (Oderberg 2011: 99–103).13 We should be careful here though. For what reason do we have to think that there are any causal connections between the essential properties of natural kinds such as, say, electron—or indeed between the kind itself and the properties of its members? In other words, why should we think that either the kind itself or its super-explanatory property (or core properties, if we accept more than one) causally explain why the properties of a natural kind cluster together? There is no scientific evidence that would suggest such a connection in cases like electron or other supposedly fundamental kinds, and indeed some, like Anjan Chakravartty (2007: 171), have suggested that fundamental natural kinds, such as electrons, have their core properties (mass, charge, spin) as a matter of brute fact. A generalized version of the HPC theory or something in the lines of Godman, Mallozzi, and Papineau’s proposal would seem to go toward providing this type of explanation: each kind is a property cluster kept in homeostasis by a causal mechanism, which could be anything from nucleosynthesis to DNA synthesis. But while the HPC theory has been very popular, it is also clear, and well-known, that the account is unlikely to apply to all natural kinds (see, e.g., Khalidi 2013: sec. 2.6). For instance, it is difficult to see what kind of causal mechanism could be postulated to be responsible for the resemblance amongst chemical elements: a shared atomic number and nuclear charge are not mechanisms in HPC’s sense, but rather just individual properties. The case of supposedly fundamental kinds like electron constitutes even stronger evidence against this type of account because of the lack of causal links between the kind and its properties or indeed among any of the properties. This latter objection seems equally challenging for Godman, Mallozzi, and Papineau’s account. One possible reaction is that some form of pluralism about natural kinds is required, i.e., that different epistemic or theoretical interests entail different accounts of kinds (for discussion of pluralism in the chemical context, see Hendry 2012, and in a biochemical context, Bartol 2016). Either way, I believe that the issue regarding the connection between a natural kind and the properties of its members, or the “unity problem” as Oderberg (2011: 90) calls it, is still at the heart of NKE: “Why, in the case of a K with putative essential properties F, G, and H, are those properties always and only found together in the Ks, assuming that the essential properties specify what a K is such as to distinguish Ks from every other kind of thing?” (ibid.).14 For example, if it is essential to the kind electron that its members have unit negative charge, then what exactly is the relationship between the property of unit negative charge and the essence of the kind electron? Is the essence of the kind something over and above the essential properties of the kind’s members, and if so, what? In asking this question, we have returned to some of our initial distinctions, specifically the distinction between NKE as I defined it and sui generis kind essentialism, according to which the kind K (a sui generis entity) has an essence. We need a clearer sense of these issues especially when faced with the problems that the causal accounts have encountered. Unfortunately, a deeper analysis of the prospects for giving such an account will have to take place elsewhere.15

10.6

Related Topics

• Alessandro Torza: “Modal Conceptions of Essence” • Fabrice Correia: “Non-modal Conceptions of Essence” • Marco Marabello: “Essences of Individuals”

165

Tuomas E. Tahko

• • • • •

Teresa Robertson Ishii: “Origin Essentialism” Travis Dumsday: “Scientific Essentialism” Ka Ho Lam: “Dispositional Essentialism” Asya Passinsky: “Artifacts, Artworks, and Social Objects” Ingo Brigandt: “Biological Species”

Acknowledgments Thanks to the editors, Kathrin Koslicki and Mike Raven, as well as Will Morgan, Sam Kimpton-Nye, and Milenko Lasnibat for their helpful feedback on an earlier version of this chapter.

Notes 1 2 3 4 5 6 7 8 9

10 11 12 13 14 15

See Marco Marabello’s “Essences of Individuals” and Teresa Robertson Ishii’s “Origin Essentialism”. See Maria Scarpati’s “Identity, Persistence, and Individuation”. See Ka Ho Lam’s “Dispositional Essentialism”. See Travis Dumsday’s “Scientific Essentialism”. The phrase “carves at the joints” originates in Plato’s Phaedrus 265e. See Alan Sidelle’s and Jonathan Livingstone-Banks’ “Conventionalism”, and Anand Vaidya and Michael Wallner’s “Conferralism”. See Asya Passinky’s “Artworks, Artifacts, and Social Objects”, Danielle Brown’s “Psychiatric Kinds”, Ron Mallon’s “Race”, and Esther Rosario’s “Sex and Gender”. See Alessandro Torza’s “Modal Conceptions of Essence” and Fabrice Correia’s “Non-modal Conceptions of Essence”. What is a “chemical property”? Putnam was originally interested in the macroscopic, phenomenological properties such as boiling point or solubility. Whether these reduce to microscopic properties such as electron configuration is debatable ( Needham 2011). I define a chemical property as a property of a chemical substance in virtue of which the substance can undergo chemical reactions, but plausibly these properties also determine the phenomenological properties ( Tahko 2015a: 802). See Asya Passinsky’s “Artworks, Artifacts, and Social Objects”. See Ingo Brigandt’s “Biological Species”. See Ingo Brigandt’s “Biological Species”. See also Kimpton-Nye (2022) for a different route to a similar result, via modal necessitarianism, the view according to which all worlds are nomologically identical. See also Dumsday (2010) on “complex essences” and Hommen (2021). See Tahko (2015a, 2020, 2021, 2022) and Keinänen and Tahko (2019).

References Austin, C. J. (2018) Essence in the Age of Evolution: A New Theory of Natural Kinds, New York and London: Routledge. Baker, L. R. (2000) Persons and Bodies: A Constitution View, Cambridge: Cambridge University Press. Bartol, J. (2016) “Biochemical Kinds,” British Journal for the Philosophy of Science 67: 531–551. Bird A. (2007) Nature’s Metaphysics: Laws and Properties, Oxford: Clarendon Press. Bird, A. (2018) “The Metaphysics of Natural Kinds,” Synthese 195: 1397–1426. Bird, A. and Tobin, E. (2022) “Natural Kinds,” in E. N. Zalta (ed.) Stanford Encyclopedia of Philosophy, (Spring 2022 Edition), https://plato.stanford.edu/archives/spr2022/entries/natural-kinds/. Boyd, R. (1999) “Homeostasis, Species, and Higher Taxa,” in R. A. Wilson (ed.) Species: New Interdisciplinary Essays, Cambridge, MA: MIT Press, pp. 141–185. Brigandt, I. (2009) “Natural Kinds in Evolution and Systematics: Metaphysical and Epistemological Considerations,” Acta Biotheoretica 57 (1–2): 77–97.

166

Natural Kind Essentialism Chakravartty, A. (2007) A Metaphysics for Scientific Realism: Knowing the Unobservable, Cambridge: Cambridge University Press. Cleland, C. E., Haze, R. M., and Morrison, S. M. (2021) “Historical Natural Kinds and Mineralogy: Systematizing Contingency in the Context of Necessity,” PNAS 118 (1): 1–9. Devitt, M. (2005) “Scientific Realism,” in F. Jackson and M. A. Smith (eds.) Oxford Handbook Of Contemporary Philosophy, Oxford: Oxford University Press, pp. 767–791. Devitt, M. (2008) “Resurrecting Biological Essentialism,” Philosophy of Science 75: 344–382. Dumsday, T. (2010) “Natural Kinds and the Problem of Complex Essences,” Australasian Journal of Philosophy 88 (4): 619–634. Dumsday, T. (2012) “A New Argument for Intrinsic Biological Essentialism,” Philosophical Quarterly 62 (248): 486–504. Ellis, B. (2001) Scientific Essentialism, Cambridge: Cambridge University Press. Ereshefsky, M. (2018) “Natural Kinds, Mind Independence, and Defeasibility,” Philosophy of Science 85: 845–856. Franklin-Hall, L. (2015) “Natural Kinds as Categorical Bottlenecks,” Philosophical Studies 172: 925–948. Garson, J. (2019) “There Are No Ahistorical Theories of Function,” Philosophy of Science 86: 1146–1156. Godman, M., Mallozzi, A., and Papineau, D. (2020) “Essential Properties Are Super-Explanatory: Taming Metaphysical Modality,” Journal of the American Philosophical Association 6: 316–334. Goodwin, W. (2011) “Structure, Function, and Protein Taxonomy,” Biology and Philosophy 26: 533–545. Griffiths, P. E. (1999) “Squaring the Circle: Natural Kinds with Historical Essences,” in R. A. Wilson (ed.) Species: New Interdisciplinary Essays, Cambridge, MA: MIT Press, pp. 208–228. Havstad, J. C. (2018) “Messy Chemical Kinds,” British Journal for the Philosophy of Science 69: 719–743. Hawley, K. and Bird, A. (2011) “What Are Natural Kinds?” Philosophical Perspectives 25: 205–221. Hendry, R. F. (2006) “Elements, Compounds, and Other Chemical Kinds,” Philosophy of Science 73: 864–875. Hendry, R. F. (2012) “Chemical Substances and the Limits of Pluralism,” Foundations of Chemistry 14: 55–68. Hommen, D. (2021) “Kinds as Universals: A Neo‑Aristotelian Approach,” Erkenntnis 86: 295–323. Häggqvist, S. and Wikforss, Å. (2018) “Natural Kinds and Natural Kind Terms: Myth and Reality,” British Journal for the Philosophy of Science 69 (4): 911–933. Keinänen, M. and Hakkarainen, J. (2017) “Kind Instantiation and Kind Change - A Problem for FourCategory Ontology,” Studia Neoaristotelica 14 (2): 139–165. Keinänen, M. and Tahko, T. E. (2019) “Bundle Theory with Kinds,” Philosophical Quarterly 69 (277): 838–857. Khalidi, M. A. (2013) Natural Categories and Human Kinds: Classification in the Natural and Social Sciences, Cambridge: Cambridge University Press. Khalidi, M. A. (2016) “Mind-Dependent Kinds,” Journal of Social Ontology 2: 223–246. Khalidi, M. A. (2021) “Etiological Kinds,” Philosophy of Science 88 (1): 1–21. Kimpton-Nye, S. (2022) “Laws of Nature: Necessary and Contingent,” Philosophical Quarterly 72 (4): 875–895. Koslicki, K. (2008) “Natural Kinds and Natural Kind Terms,” Philosophy Compass 3/4: 789–802. Kripke, S. (1971) “Identity and Necessity,” in M. K. Munitz (ed.) Identity and Individuation, New York, NY: New York University Press, pp. 135–164. Kripke, S. (1980) Naming and Necessity, Cambridge, MA: Harvard University Press. LaPorte, J. (2004) Natural Kinds and Conceptual Change, Cambridge: Cambridge University Press. LaPorte, J. (2012) Rigid Designation and Theoretical Identities, Oxford: Oxford University Press. Lewis, D. (1986) On the Plurality of Worlds, Blackwell, New York. Linsky, B. (2006) “General Terms as Rigid Designators,” Philosophical Studies 128 (3): 655–667. Lowe, E. J. (2006) The Four-Category Ontology: A Metaphysical Foundation for Natural Science, Oxford: Oxford University Press. Lowe, E. J. (2008) “Two Notions of Being: Entity and Essence,” Royal Institute of Philosophy Supplements 83 (62): 23–48. Mallon, R. (2016) The Construction of Human Kinds, Oxford: Oxford University Press. Mill, J. S. (1843) “A System of Logic, Ratiocinative and Inductive,” in J. M. Robson (ed.) Collected

167

Tuomas E. Tahko Works of John Stuart Mill, Toronto: University of Toronto Press and London: Routledge and Kegan Paul, 1974. Millikan, R. G. (1999) “Historical Kinds and the ‘Special Sciences’,” Philosophical Studies 95: 45–65. Needham, P. (2011) “Microessentialism: What is the Argument?,” Noûs 45: 1–21. Oderberg, D. (2011) “Essence and Properties,” Erkenntnis 75: 85–111. Okasha, S. (2002) “Darwinian Metaphysics: Species and the Question of Essentialism,” Synthese 131: 191–213. Psillos, S. (2002) Causation and Explanation, Montreal: McGill-Queen University Press. Putnam, H. (1970) “Is Semantics Possible?,” Metaphilosophy 1(3): 187–201. Putnam, H. (1975) “The Meaning of “Meaning”,” in Putnam 1979, Mind, Language and Reality: Philosophical Papers, Vol. 2, Cambridge University Press, pp. 215–271. Putnam, H. (1990) “Is Water Necessarily H2O?” in J. Conant (ed.) Realism with a Human Face, Cambridge, MA: Harvard University Press, pp. 54–79. Salmon, N. U. (2005) Reference and Essence, second edition, New York: Prometheus Books. Schwartz, S. P. (2021) “Against Rigidity for Natural Kind Terms,” Synthese 198: 2957–2971. Sider, T. (1996) “Intrinsic Properties,” Philosophical Studies 83: 1–27. Soames, S. (2005) Reference and Description: The Case Against Two-Dimensionalism, Princeton: Princeton University Press. Stanford, P. K. and Philip, K. (2000) “Refining the Causal Theory of Reference for Natural Kind Terms,” Philosophical Studies 97: 99–129. Tahko, T. E. (2009) “On the Modal Content of A Posteriori Necessities,” Theoria: A Swedish Journal of Philosophy 75 (2): 344–357. Tahko, T. E. (2015a) “Natural Kind Essentialism Revisited,” Mind 124: 795–822. Tahko, T. E. (2015b) “The Modal Status of Laws: In Defence of a Hybrid View,” Philosophical Quarterly 65 (260): 509–528. Tahko, T. E. (2020) “Where Do You Get Your Protein? Or: Biochemical Realization,” British Journal for the Philosophy of Science 71 (3): 799–825. Tahko, T. E. (2021) Unity of Science, Elements in Philosophy of Science, Cambridge: Cambridge University Press. Tahko, T. E. (2022) “Natural Kinds, Mind-Independence, and Unification Principles,” Synthese 200, 144. Tobin, E. (2010) “Microstructuralism and Macromolecules: The Case of Moonlighting Proteins,” Foundations of Chemistry 12: 41–54. van Brakel, J. (1986) “The Chemistry of Substances and the Philosophy of Mass Terms,” Synthese 69: 291–324. VandeWall, H. (2007) “Why Water Is Not H2O, and Other Critiques of Essentialist Ontology from the Philosophy of Chemistry,” Philosophy of Science 74: 906–919. Weisberg, M. (2005) “Water is Not H2O,” in D. Baird, E. Scerri, and L. McIntyre (eds.) Philosophy of Chemistry: Synthesis of a New Discipline, New York: Springer, pp. 337–345. Williams, N. E. (2011) “Putnam’s Traditional Neo-Essentialism,” Philosophical Quarterly 61: 151–170.

168

11 ORIGIN ESSENTIALISM Teresa Robertson Ishii

11.1

Introduction

Interest in origin essentialism was spurred by the landmark Naming and Necessity lectures delivered by Saul Kripke at Princeton University in January 1970. (The relevant portions of the lectures appear in (Kripke [1972] 1980: 110–5). Unless otherwise indicated, this chapter’s quotations of and references to Kripke’s work are from these pages.) Kripke endorsed three origin essentialist theses. • Queen Elizabeth, who was in fact the child of Elizabeth Bowes-Lyon and George VI, could not have been the child of Harry and Bess Truman instead. • Queen Elizabeth could not have originated from “a totally different sperm and egg” from those from which she actually originated. • The particular table at which Kripke was pointing could not have been made from “a completely different block of wood” or other non-wooden material (emphasis deleted). The first two claims should of course be taken to generalize to other people and similar organisms and the third to generalize to other tables and similar material artifacts. But little more than this can safely be attributed to Kripke. Whether or not his origin essentialist theses are true, they rightly had a huge impact on philosophy. For one thing, each claim attributes a modal property (a property having to do with metaphysical necessity or possibility, with how things metaphysically must or could be) to a particular individual independently of how that individual is designated. For example, the first claim says of the Queen—of the woman herself—that she has the property of necessarily not being the child of Harry and Bess Truman. The mere intelligibility of Kripke’s theses discredited a formerly influential view of Quine (1960: 195–200) according to which saying of an individual that she has a modal property makes about as much sense as saying of an individual, independently of how he is designated, that he was so-called because of his size. (See this volume Chapter 29 “Quine on Essence”.) For another thing, Kripke’s origin essentialist theses provided plausible examples of the necessary a posteriori. To be clear, what is claimed to be necessary and knowable only a posteriori is, to take the first claim again, the non-modal fact that Queen Elizabeth is not the child of Harry and Bess Truman. Kripke’s

DOI: 10.4324/9781003008750-14

169

Teresa Robertson Ishii

origin essentialist claims thus provided a vivid illustration of a valuable philosophical lesson: the distinction between necessity and contingency is different from the distinction between a priority and a posteriority. These significant contributions aside, the claims were also simply captivating, so that origin essentialism became a topic in its own right. It is worth emphasizing that Kripke’s claims do not concern the issue of what pertains to the nature of a particular object. The word “essential” (and its cognates) is used in multiple ways in contemporary analytic philosophy. Sometimes “essential” is used to mean modally essential, where an object x has a property P modally essentially iff it is metaphysically necessary that x has P (if x exists).1 Sometimes “essence” is used to mean nature or what has sometimes been called whatness.2 This usage has become dominant in the wake of “Essence and Modality” (Fine 1994). But throughout this chapter, unless otherwise indicated, I will use “essential” in its modal sense. It is important in today’s philosophical environment to realize that using the word “essential” in this way carries no commitment to the so-called modal account of essence (whatness). (See this volume Chapter 7 “Modal Conceptions of Essence”.) Consideration of S and S′ will clarify. S: A property P is essential to x iff it is metaphysically necessary that x has P. S is ambiguous (polysemous). If one reads “essential” in its modal sense, S is straightforwardly analytic. If one reads “essential” in its whatness sense, then S expresses the modal account of whatness. S′: x is a bank iff x is the land along the side of a river. S′ is ambiguous. If one reads “bank” in its fluvial sense, S′ is straightforwardly analytic. If one reads “bank” in its financial sense, it expresses the altogether silly fluvial account of financial banks. Just as one who uses “bank” in its fluvial sense does not thereby endorse a ridiculous account of financial banks, one who uses “essential” in its modal sense does not thereby endorse the modal account of whatness essence.3 §2 takes a closer look at Kripke’s origin essentialist claims. §3 presents expansion (or addition) arguments for such claims. §4 discusses the paradox of flexible origin essentialism and provides brief concluding remarks about the continuing significance of origin essentialism to contemporary philosophy.

11.2 A Closer Look at Origin Essentialism The second of Kripke’s claims about the Queen, the one involving the egg and sperm, was probably thought by Kripke to be more basic than the first, since he indicated that the type of parent he had in mind was what is more explicitly called a gametic parent. The gametic parents are “the people whose body tissues are sources of the biological sperm and egg” from which a child arises. If one assumes, as I suspect Kripke did, that a gamete must have been produced by the very organism that in fact produced it—and such an assumption would itself be another origin essentialist claim—one can legitimately move from the claim about the Queen’s gametic origin to the claim about her parental origin. (Cf. McGinn 1976.) It is worth noticing that Kripke does not claim that the Queen must have had the parents that she actually did. Nor does he claim that she had to have arisen from the egg and sperm from which she actually arose. The claim that she could not have had Harry and Bess Truman

170

Origin Essentialism

as parents, though it must of course be taken to generalize to other pairs of people, may be taken to leave open the possibility that she could have had Harry Truman and Elizabeth Bowes-Lyon as parents. This understanding is a way of making sense of Kripke’s claim that the Queen could not have come from a “totally different” egg and sperm. In a similar vein, Kripke says that the table could not have come from a “completely different” block of wood. The phrase “completely different” is odd, given the compelling view that numerical identity and distinctness do not admit of degrees. It is natural therefore to take Kripke to mean that the table, which in fact came from a particular block of wood consisting of certain matter, could not have come from a different block of wood consisting of entirely different matter—that is, consisting of matter that does not overlap (have matter in common with) the matter that made up the block of wood when the table came into existence. Kripke’s origin essentialist claims are thus compatible with the possibility of something’s having a somewhat different origin.4 Kripke’s claims are sometimes confused with the claim that a person could not have had a genome that differs from the one they actually have. (Cf. Cooper 2015; Janssen-Lauret 2021; Pust 2001.) Even if Kripke had thought not merely that the Queen could not have originated from “a totally different sperm and egg” but that she must have originated from the egg and sperm from which she actually did, there is no implication concerning the Queen’s genomic characteristics. In fact, Kripke says, “I might have been deformed if the fertilized egg from which I originated had been damaged in certain ways, even though I presumably did not yet exist at that time”. It is natural to think that the kind of damage Kripke has in mind is genomic damage. It is also natural to think that what goes for deleterious change goes for beneficial change and that what goes for the fertilized egg goes for the unfertilized egg and also for the sperm. Evidently, there are many routes to Kripke’s possibly having a genome that differs from the one he actually has that are consistent with Kripkean origin essentialism. Many find origin essentialist claims intuitively compelling. But a word of caution about counterfactual conditionals is in order as one tries to assess one’s own intuitions. It is common to think that if one’s parents (or their gametes) had not gotten together, then one would not exist. And it may be tempting to take this counterfactual conditional to support a strong form of origin essentialism—stronger than that evidently advocated by Kripke—according to which it is impossible for one to exist without one’s parents (or their gametes) having gotten together. But the temptation should be resisted, since the inference is faulty. It is like moving from the claim that if Mike had not joined the Coast Guard, then he would not have received Veteran’s Administration benefits to the claim that it is impossible for Mike to have received Veteran’s Administration benefits without having joined the Coast Guard. Although many find Kripke’s claims intuitively compelling, some do not. So argumentation is especially important. It is also important for another reason—it says something about the plausible extensions of Kripke’s claims. If the argument is successful, can it also be successfully applied mutatis mutandis to yield that a human being could not have originated from entirely different matter? Or to yield some version of genetic origin essentialism? Does a similar argument support that at least some natural formations like mountains, lakes, or streams have their original matter essentially? Or that events have their causes essentially? Or that the #BlackLivesMatter movement could not have been sparked by the murder of George Floyd instead of having been sparked by the acquittal of George Zimmerman? If one adopts a hylomorphic view of entities like the Supreme Court, where the idea is roughly that the “matter” of the Court is its justices while the “form” of the Court is its organizational structure, should one conclude that the Court’s original “matter” is essential to it? That is, should one conclude that the Supreme Court must have originated with a group of justices that overlap the group of

171

Teresa Robertson Ishii

justices it actually originated with? Does a similar argument support that a biological species could not have evolved “from a different branching in the tree of life”? Moving further afield—outside the realm of origins—does a similar argument support that chemical elements and compounds have their structures essentially? (On lakes, see Hughes 1994. On events, see van Inwagen 1983: 169–70 and Hughes 1994. On biological species, see Pedroso 2014 (p60 quoted) and this volume Chapter 18 “Biological Species”. On chemical elements and compounds, see Salmón 1981.) Questions like these can be answered only by looking carefully at arguments. §3 turns to this task. The reader who is interested in a potential extension of Kripke’s claims is encouraged to consider how (or whether) the premises of the arguments to be discussed can be modified for the purpose. Such a reader is also encouraged to consider the plausibility of each of the resulting premises.

11.3

Expansion Arguments for Origin Essentialism

Kripke did not merely advocate origin essentialism, he famously—perhaps infamously—provided “something like proof” in footnote 56 of Naming and Necessity. To be more careful, he provided a sketch of an argument for his claim about the table. Although the sketch is in some places murky, this much is clear: the argument involves first considering the actual world in which our table is originally made from a certain hunk of matter (or, by extension, in which the Queen arises from a certain pair of gametes), and then considering a possible world that expands or adds to the actual world in the sense that it includes our actual table’s being originally made from the hunk of matter from which it was actually originally made and in addition includes a table’s being originally made from a hunk of matter that is “completely different” from that hunk of matter (or, by extension, includes the Queen’s arising from the pair of gametes from which she actually arose and in addition includes a person’s arising from a pair of gametes that are “totally different” from that pair). (I suppress further reference to the Queen, leaving it to the reader to consider whether the analogs of the premises in the argument concerning the table are plausible in the case of the Queen.) Kripke’s phrase “hunk of matter” will be understood to have roughly the same sense as “portion of matter” or “bit of matter” or simply “matter”. A hunk of matter (in this sense) may be scattered—a bit here, a bit there. The phrase “x is originally made from matter y” or simply “x is made from y” is to be understood to mean that x is originally made entirely from all of y, which is to say that no matter other than y is used to make x and that there is none of y that is not used to make x. It will help to keep things manageable to assume that exactly the same matter exists in all the worlds under consideration and that if matter y and z overlap in one world, they overlap in all. (This simplification is not entirely without import (Ballarin 2013). It limits the discussion—and the argument’s premises and conclusions—to hunks of matter that co-exist in a given world. The expository purposes it serves more than compensate.) Below are two salient candidate compossibility assumptions that may be involved in “generating” the expanded world. UNISPECIFIC COMPOSSIBILITY:

If x is a table originally made from matter y and it is possible for some table or other to be originally made from matter z, which does not overlap matter y, then it is also possible for x to be a table originally made from matter y and in addition some table or other to be originally made from matter z. (Cf. Salmón 1981: 203, (IV))

172

Origin Essentialism BISPECIFIC COMPOSSIBILITY:

If x is a table originally made from matter y and it is possible for x′ to be a table originally made from matter z, which does not overlap matter y, then it is also possible for x to be a table originally made from matter y and in addition x′ to be a table originally made from matter z. (Cf. Salmón 1981: 201, (III))

UNISPECIFIC COMPOSSIBILITY is intuitively plausible. Consider any particular table x, the matter y from which it was originally made, and any matter z that does not overlap y. (This will be our starting point for exploring both UNISPECIFIC COMPOSSIBILITY and BISPECIFIC COMPOSSIBILITY, so keep your finger on it.) UNISPECIFIC COMPOSSIBILITY yields that if z could be made into a table, then it could be that x is a table originally made from y and in addition z is made into a table. The expanded possible world (the world of the consequent, so to speak) is specific as to the “first” table (that is, to x) but unspecific as to the “second”. It is exceedingly plausible that such an expanded world is indeed possible. And this plausibility is completely independent of the plausibility of origin essentialism itself. As such UNISPECIFIC COMPOSSIBILITY is appropriate to use in an argument for origin essentialism. But, as we shall see, BISPECIFIC COMPOSSIBILITY is another story. It yields that if z could be made into a table x′, then it could be that x is a table originally made from y and in addition x′ is a table originally made from z. The expanded possible world in this case is specific as to both the “first” table (that is, to x) and the “second” (that is, to x′). When x and x′ are distinct, this is quite plausible. And its plausibility is independent of origin essentialism. But what if x is x′? In this case, BISPECIFIC COMPOSSIBILITY yields that if there is a possible world in which x is a table originally made from matter z (matter that does not overlap x’s actual original matter y), then there is a possible world in which x is a table with two distinct origins, y and z. The consequent violates the following nearly incontrovertible principle. (But Spencer and Tillman 2019 deny a generalization—to all material objects—of a variant of this principle.) ORIGIN UNIQUENESS:

It is impossible that a single table x is originally made from matter y and in addition is originally made from distinct matter z. (Cf. Salmón 1981: 200, (I))

So the conditional together with ORIGIN UNIQUENESS would yield that there is no possible world in which x is a table originally made from matter z (matter that does not overlap x’s actual original matter y). But why would anyone who did not already accept origin essentialism accept the conditional? Without independent support, BISPECIFIC COMPOSSIBILITY should be powerless to move one who is genuinely on the fence about origin essentialism. (Cf. Salmón 1981: 212–6; Robertson and Forbes 2006: 369; Rohrbaugh and deRosset 2006: 376.) Whereas BISPECIFIC COMPOSSIBILITY is objectionably strong, UNISPECIFIC COMPOSSIBILITY is not. But that means that resources in addition to ORIGIN UNIQUENESS are needed. A sufficiency principle fills the bill. SUFFICIENCY1:

If it is possible that x is a table originally made from matter z, then it is necessary that any table originally made from matter z is the very table x and no other. (Cf. Salmón 1981: 206, (V))

173

Teresa Robertson Ishii

Consider a table, a, which was originally made from matter m. Consider too some matter m* that has no matter in common with m, but which could itself be made into a table. This satisfies the antecedent of (an instance of) UNISPECIFIC COMPOSSIBILITY. (I will henceforth suppress some of the fussy details, like the one indicated by the occurrence of “(an instance of)” in the previous sentence.) What UNISPECIFIC COMPOSSIBILITY then says is that it is possible for a to be originally made from m while in addition m* is made into a table. ORIGIN UNIQUENESS then says that in any such possible world, these tables are distinct from one another. Now consider some particular possible world in which all this happens, and let b be the table there that is originally made from m*. SUFFICIENCY1 says then that in all possible worlds any table that is originally made from m* is b and not some other table. Since we have already established that a is distinct from b, this means that a could not have been made from m*.5 Since there is nothing special about a, b, m, and m*, the following conclusion is warranted on the basis of this reasoning, provided that SUFFICIENCY1 is true. ORIGIN ESSENTIALISM:

If a given table is originally made from certain matter, then it is necessary that the given table is not a table originally made from any non-overlapping matter.

(This argument, together with the first two modifications below, follows Salmón 1981. Cf. Salmón 1979. Forbes 1986 gives an argument concerning biological organisms that is roughly in the same vein, especially insofar as it appeals to a sufficiency principle. Cf. Forbes 1985; Forbes 1980.) But SUFFICIENCY1 is worrisome. Suppose that a is a dining table. Intuitively, m could just as well have been used to make a coffee table instead. Consider any possible world in which this happens. SUFFICIENCY1 dictates that in any such world the coffee table made from m is a. But it is intuitively dubious that a could have been a coffee table. This worry may be addressed by mentioning the “plan” according to which the table is made. (A lot may be packed into this notion of a plan—perhaps, for example, being made by a particular woodworker should count as part of the plan.) SUFFICIENCY2:

If it is possible that x is a table originally made from matter z according to plan u, then it is necessary that any table originally made from matter z according to plan u is the very table x and no other. (Cf. Salmón 1981: 210–1, (V′))

But SUFFICIENCY2 is also problematic. Suppose there is a table, c, that is originally made from matter n according to some plan p. As time goes by, c undergoes various repairs until it comes to be constituted by matter that does not overlap n at all. At this point, while c still exists, n is made into a table, d, according to p. Clearly tables c and d are distinct from one another, but SUFFICIENCY2 identifies them. SUFFICIENCY3 avoids this. SUFFICIENCY3:

If it is possible that x is the only table originally made from matter z according to plan u, then it is necessary that any table that is the only table originally made from matter z according to plan u is the very table x and no other. (Cf. Salmón 1981: 229, (V′′))

But SUFFICIENCY3 faces a related worry (McKay 1986). Consider the same “recycling” situation just described. At least some have the intuition that each of c and d could be made

174

Origin Essentialism

without the other—that is, that there is a possible world in which c is the only table originally made from n according to p and another possible world in which d is the only table made this way. If so, SUFFICIENCY3 incorrectly identifies c and d. Others however have predecessor essentialist intuitions, according to which c could have existed without d, but d could not have existed without c (Forbes 1994: §3; Salmón 2005: Ap. VI, p. 373). A slight modification provides a more forceful complaint (Robertson 1998). Consider another recycling situation. Suppose that c is originally made from n according to p and that over time all but one molecule of c’s original matter is replaced. Let n′ be the hunk of matter consisting of the matter that was replaced together with one additional molecule (of the same type as the sole original molecule remaining in c). Suppose further that while c still exists, n′ is made into a table, d′, according to p. Intuitively, since n′ differs from n by only a single molecule, there is a possible world in which c is the only table made from n′ according to p. Moreover, there is a possible world—namely, the actual world—in which d′ is the only table made in this way. Sufficiency3 incorrectly identifies c and d′. This prompts SUFFICIENCY4. SUFFICIENCY4:

If it is possible that x is originally made from matter z according to plan u and is the only table originally made partly from any of matter z, then it is necessary that any table that is originally made from matter z according to u and is the only table originally made partly from any of matter z is the very table x and no other. (Cf. Salmón 2005: Ap. VI, 374; Hawthorne and Gendler 2000: 290) This (along with a companion modification of

UNISPECIFIC COMPOSSIBILITY)

yields

MODIFIED ORIGIN ESSENTIALISM:

If a given table is originally made from certain matter, then it is necessary that the given table is not a table originally made from non-overlapping matter while being the only table originally made partly from any of that matter. (Cf. Salmón 2005: Ap. VI, 375)

This conclusion is significantly different from the claims with which we began. It evidently leaves open possibilities—such as a table’s being made from matter z that does not overlap the matter from which it was actually originally made, provided that some other table is also originally made partly from some of z—that are at odds with the origin essentialist claims with which we began. Moreover SUFFICIENCY4 will not be acceptable to those who have the intuition that each of c and d′ could be originally made from n′ according to p while being the only table originally made partly from any of matter n′. SUFFICIENCY4 would incorrectly identify c and d′. Insofar as predecessor essentialism—a view accordingly to which d′ could not exist without c—enjoys less intuitive support than origin essentialism did in the first place, this version of the argument (like the one using SUFFICIENCY3) leaves something to be desired. While none of this is devastating, these considerations provide some motivation for developing an alternative argument. Explicitly hoping to avoid problems faced by sufficiency principles, some argued for origin essentialism by trying to provide appropriate independent support for BISPECIFIC COMPOSSIBILITY (Rohrbaugh and deRosset 2004). Simplifying a fair bit, some critics of this attempt showed that a premise of the resulting argument for BISPECIFIC COMPOSSIBILITY, the so-called locality of prevention, in fact entailed a sufficiency principle of the sort that the argument sought to avoid (Robertson and Forbes 2006). Others argued roughly both that the locality of prevention was too weak to support BISPECIFIC COMPOSSIBILITY and that sufficient

175

Teresa Robertson Ishii

supplementation would result in a question-begging argument for origin essentialism (Cameron and Roca 2006). The argument was modified in response (Rohrbaugh and deRosset 2006). Subsequent critics endeavored to show that the argument from UNISPECIFIC COMPOSSIBILITY and the argument from BISPECIFIC COMPOSSIBILITY are flawed in a common way. One argued roughly that insofar as the desired conclusion encompasses not merely hunks of matter that co-exist in a given world, but also any possible (according to the given world) hunks of matter, both styles of argument presuppose a recombination thesis—one according to which any two non-overlapping possible hunks of matter are compossible—that does not fit naturally into a unified theoretical framework with the argument’s other assumptions (Ballarin 2013). Another argued that any argument for origin essentialism in the broad style of Kripke’s footnote 56 would inevitably appeal to some metaphysical principle that leads to an implausible essentialist thesis, such as the claim that a table’s place of origin is essential to it (Damnjanovic 2010). Moreover, this critic urged that often the relevant deviant essentialist thesis would undermine an assumption involved in the corresponding argument for origin essentialism. Such an argument would thus be self-undermining. Neither the arguments for origin essentialism nor the criticisms of those arguments are decisive. This is typically the case for philosophical theses. It is uncommon however for the mere intelligibility, the mere plausibility, or the mere consistency of a philosophical view to be of much importance. §1 mentioned the significance of the mere intelligibility and the mere plausibility of Kripke’s origin essentialist claims. §4 will touch on the significance of the mere consistency of flexible origin essentialism.

11.4 The Paradox of Flexible Origin Essentialism As we have seen, Kripke’s origin essentialist claims are modest: although they exclude the possibility of something’s having a completely different origin, they allow the possibility of that thing’s having a somewhat different origin. Indeed, Kripke evidently affirms the possibility of a slightly different origin for the table (Kripke [1972] 1980: 50–1). An interesting paradox arises from this combination of views together with some additional claims. For ease of exposition, let us make some simplifying assumptions. First, swap tables for simple tripods that are made from kits consisting solely of three qualitatively identical interlocking legs (Salmón 2021). Next, assume both FLEXIBILITY and ESSENTIALISM. FLEXIBILITY:

If t is a tripod originally made from kit k, then t could have been a tripod originally made from any kit that overlaps k by at least two legs.

ESSENTIALISM:

If t is a tripod originally made from kit k, then t could not have been a tripod made from any kit that does not overlap k by at least two legs.

Let Trip be a tripod made from a kit consisting of legs L1, L2, and L3. FLEXIBILITY yields that Trip could have been a tripod originally made from L1+L2+L13. (Note that on my naming convention for legs, each distinct name consisting of “L” and a numerical subscript corresponds to a distinct leg.) In other words, there is a possible world, w1, in which Trip is a tripod originally made from L1+L2+L13. More fully, there is a world, w1, possible according to w0 (to give the actual world a name), in which Trip is a tripod originally made from L1+L2+L13. Intuitively, FLEXIBILITY is not the sort of claim that can be true merely

176

Origin Essentialism

contingently. If it is true, then it is necessarily true. In other words, if FLEXIBILITY is true, then it is true in all possible worlds. If so, then FLEXIBILITY is true in w1. And thus, there is a world, w2, possible according to w1, which is itself possible according to w0, in which Trip is a tripod made from L1+L12+L13. For short, there is a possibly possible world, w2, in which Trip is a tripod made from L1+L12+L13. In other words, it is possibly possible that Trip is a tripod made from L1+L12+L13. If whatever is possibly possible is possible, then it is possible that Trip is a tripod made from L1+L12+L13. But this violates ESSENTIALISM. We have derived a contradiction from plausible assumptions. Some would solve this paradox by rejecting FLEXIBILITY (Chisholm 1976: 145–58). Others would solve it by rejecting ESSENTIALISM (Mackie 2006). Many would prefer a solution that allows retaining both. There are two options. The first is to block the derivation of the contradiction by saying that although it is possibly possible that Trip is made from L1+L12+L13, it is not possible (Chandler 1976; Salmón 1981; Salmón 1986). This solution is unorthodox, since it violates the widely accepted logic for metaphysical modality, S5. (Prominent advocates of S5 include Fine (1977: 119), Forbes (1985; and especially relevantly, 1984), Plantinga (1974), and Williamson (2016). Williamson (2013: 44) states, “most metaphysicians accept S5”.) According to S5, it is logically true (that is, true as a matter of logic, not merely true) that the relation between worlds u and v of v’s being possible according to u is reflexive, symmetric, and (most relevantly) transitive. (On truths vs. logical truth: It is true but not logically true that Biden is president. Those who deny this may be wrong, but they do not thereby exhibit logical confusion.) The second is to block the derivation of the contradiction by saying that there is no world possible according to w1, in which Trip is made from L1+L12+L13. That is, there is no world such as w2. On this solution, FLEXIBILITY is true, but only contingently so (Hawthorne 2006: 241n8). This solution requires a radical departure from the standard philosophical practice of treating general philosophical claims, like FLEXIBILITY, as necessary if true. (Think how odd it would be for utilitarianism to be true, but only contingently so. Cameron (2007) offers opposing considerations in regard to a different philosophical thesis.) This solution is motivated by the desire to retain S5, in spite of S5’s delivering strange results in the case at hand. According to S5, what is (im)possible does not vary from possible world to possible world. FLEXIBILITY and ESSENTIALISM dictate that in w0, Trip’s origin (im)possibilities are tethered to L1+L2+L3. Since w0 and w1 are both possible worlds, S5 dictates that in w1 Trip’s origin (im)possibilities are also tethered to L1+L2+L3, in spite of the fact that in w1 Trip is not originally made from L1+L2+L3. Thus according to w1, Trip could be made from some kits, such as L1+L2+L3, that have two legs in common with L1+L2+L13 (the kit from which Trip originates in w1) and could not be made from other kits, such as L1+L12+L13, that have two legs in common with L1+L2+L13 (again, the kit from which Trip originates in w1). This is highly counterintuitive. How can it be that in other possible worlds tripods are so different from the way they are in the actual world? How can the way things happen to be be so special in this way? Some philosophers are sanguine about the prospects of rendering this counterintuitive result palatable (Hawthorne 2006: 241n6; cf. Leslie 2011; Kment 2018.) Others are dubious (Salmón 1981; Robertson Ishii 2022). By contrast, rejecting the claim that whatever is possibly possible is possible in response to a prima facie counterexample is in keeping with standard philosophical practice. It is akin to rejecting the (formerly orthodox) view that if all S are P then some S are P in response to the prima facie counterexample provided by the plausible view that all humans are mortal and moreover there are no non-mortals. (If all humans are mortal, then all non-mortals are non-

177

Teresa Robertson Ishii

human. So, if the formerly orthodox view were correct, then some non-mortals would be nonhuman, and thus there would be non-mortals.6) Prima facie, if we accept both FLEXIBILITY and ESSENTIALISM, the lesson of the paradox is that not everything that is possibly possible is possible. This lesson is significant. If a thing’s modally essential properties vary from possible world to possible world, then insofar as whatness essences are not supposed to vary in this way, the modal essence of a thing is not determined by the whatness essences of all things, pace Fine 1994 (Robertson Ishii 2022). But what if we do not accept both FLEXIBILITY and ESSENTIALISM? Significantly, if we find flexible origin essentialism merely consistent, then insofar—and it is pretty far—as it is prima facie implausible that flexible origin essentialism is consistent, but inconsistent with FLEXIBILITY’s being necessary if true, then we thereby have reason to reject that logic dictates that whatever is possibly possible is possible. And that would mean that S5 is not the correct logic for metaphysical modality.7 Fifty years ago, the mere intelligibility of Kripke’s origin essentialist claims challenged—then overturned—then current philosophical orthodoxy. Similarly, the mere consistency of flexible origin essentialism challenges current philosophical orthodoxy. It remains to be seen whether the orthodox view that S5 is the correct logic for metaphysical modality will also be overturned.

11.5

Related Topics

Modal Conceptions of Essence [Chapter 7] Biological Species [Chapter 18] Quine on Essence [Chapter 29]

Notes 1 A word about the parenthetical “if x exists”: if modal essentiality is explicated in the existenceconditioned way, it is plausible that Socrates is modally essentially human whereas if it is explicated in the categorical way, this is not so, since being human requires existing and it is possible for Socrates not to exist. Instead what is plausibly essential to Socrates on the categorical explication is the property of being human if existent. The fact that the concept of a modally essential property admits of two non-equivalent explications should not be particularly worrisome. Compare: the concept of a prime number may be explicated in a way that includes the number one (as in “a number whose only factors are one and itself”) or in a way that does not (as in “a number with exactly two factors”). Perhaps the two explications of modal essentiality arise from the fact that being necessary like being relevant is relational. Relevant to what? Necessary for what? If it is for existing, then that yields the existence-conditioned explication. If it is for being, then that yields the categorical explication. The distinction between existing and being need not be spooky: Bertrand Russell does not exist, but he has being enough to render it intelligible to attribute to him properties like having written “On Denoting”. 2 Recognizing “the dominance of the modal reading of “essence” in current discussions,” Almog (1991: 230n2 quoted) introduced the term “whatness” for the sense of “essence” that interested him. 3 By saying that “it is only in the last twenty years or so that the modal approach to essentialist metaphysics has really come into its own,” Fine (1994: 3) intimated that modal metaphysicians of the 1970s and 1980s held “the modal account of essence”. But since Fine cited no one of that period by name, it was uncertain to whom he attributed the account. Nonetheless a high-profile group of Fine’s readers evidently accepted his suggestion that modal metaphysicians of the 1970s and 1980s held the view in question. The view is described, without citation, as “widespread” ( Correia 2007: 63), “once the dominant account” ( Wildman 2013: 760), “once-dominant” ( Skiles 2015: 100), and “traditional” ( Leech 2018: 311). Both Wildman (2021: 1456n2) and Zylstra (2019: 339) cite exactly two still-living philosophers of that period as advocating the view: Kripke ([1972] 1980) and Plantinga (1974). Cowling (2013) names others of the usual suspects (though perhaps attributing to them a

178

Origin Essentialism

4

5 6 7

different, but related, view), including two prominent advocates of origin essentialism: Forbes and Salmón. In the draft of his contribution to the present volume, Fine writes, “Quine, like all others who worked within the framework of quantified modal logic, adopted a modal view of essence [whatness]” (draft: 8). In personal communication, each of Forbes, Kripke, Plantinga, and Salmón has confirmed that he never held the modal account (view, conception, etc.) of whatness. All used “essential” in its modal sense. In conversation, Kripke said that given that humans originate from just two (and not, for example, from a hundred) gametes, he is inclined to think that the Queen could not have existed without originating from both of the gametes from which she actually originated. The thought is that although in principle a thing’s origin could be slightly different but not too different, for creatures originating in just two things, any difference is too much. Here the argument uses the necessity of the NECESSITY OF DISTINCTNESS, the claim that if x and y are distinct then that is necessarily so. The NECESSITY OF DISTINCTNESS is generally taken to be a logical truth. I here take it as such. This obviates the need always to make its use explicit. In a domain restricted to living things, the point is obvious. In an unrestricted domain, one may take “mortals” to be things that either live and die or never live. To be more careful: my talk of “consistency” and “inconsistency” here is to be understood against the background of uncontroversial assumptions (such as that Trip is originally made from L1+L2+L3); in other words, when I say “consistent”, I mean consistent with the uncontroversial background assumptions.

References Almog, J. (1991) “The What the How [I],” Journal of Philosophy 88.5: 225–244. Ballarin, R. (2013) “The Necessity of Origin: A Long and Winding Route,” Erkenntnis 78.2: 353–370. Cameron, R. (2007) “The Contingency of Composition,” Philosophical Studies 136.1: 99–121. Cameron, R. and Roca, S. (2006) “Rohrbaugh and deRosset on the Necessity of Origin,” Mind 115.458: 361–366. Chandler, H. (1976) “Plantinga and the Contingently Impossible,” Analysis 36: 106–109. Chisholm, R. (1976) Person and Object: A Metaphysical Study, La Salle, IL: Open Court. Correia, F. (2007) “(Finean) Essence and (Priorian) Modality,” Dialectica 61: 63–84. Cooper, R. (2015) “How Might I Have Been?,” Metaphilosophy 46.4-5: 495–514. Cowling, S. (2013) “The Modal View of Essence,” Canadian Journal of Philosophy 43.2: 248–266. Damnjanovic, N. (2010) “No Route to Material Origin Essentialism?,” Erkenntnis 72: 93–110. Fine, K. (1977) “Prior on the Construction of Possible Worlds and Instants,” Postscript to Prior and Fine 1977: 116–161. Fine, K. (1994) “Essence and Modality,” Philosophical Perspectives 8: 1–16. Forbes, G. (1980) “Origin and Identity,” Philosophical Studies 37: 353–363. Forbes, G. (1984) “Two Solutions to Chisholm’s Paradox,” Philosophical Studies 46.2: 171–187. Forbes, G. (1985) The Metaphysics of Modality, Oxford: Clarendon Press. Forbes, G. (1986) “In Defense of Absolute Essentialism,” in French, A., Uehling, T., Wettstein, H. (eds.), Midwest Studies in Philosophy XI: Studies in Essentialism (3–31), Minneapolis, MN: University of Minnesota Press. Forbes, G. (1994) “A New Riddle of Existence,” Philosophical Perspectives 8: 415–430. Hawthorne, J. (2006) “Determinism De Re,” in his Metaphysical Essays (239–243), New York: Oxford University Press. Hawthorne, J. and Gendler, T. (2000) “Origin Essentialism: The Arguments Reconsidered,” Mind 109.434: 285–298. Hughes, C. (1994) “The Essentiality of Origin and the Individuation of Events,” Philosophical Quarterly 44.174: 26–44. Janssen-Lauret, F. (2021) “Anti-Essentialism, Modal Relativity, and Alternative Material-Origin Counterfactuals,” Synthese 199: 8379–8398, 10.1007/s11229-021-03167-8. Kment, B. (2018) “Chance and the Structure of Modal Space,” Mind 127.507: 633–665. Kripke, S. ([1972] 1980) Naming and Necessity, Cambridge: Harvard University Press.

179

Teresa Robertson Ishii Leech, J. (2018) “Essence and Mere Necessity,” Royal Institute of Philosophy Supplement 82: 309–332. Leslie, S. (2011) “Essence, Plenitude, and Paradox,” Philosophical Perspectives 25: 277–295. Mackie, P. (2006) How Things Might Have Been, Oxford: Clarendon. McGinn, C. (1976) “On the Necessity of Origin,” Journal of Philosophy 73.5: 127–135. McKay, T. (1986) “Against Constitutional Sufficiency Principles,” in French, A., Uehling, T., Wettstein, H. (eds.), Midwest Studies in Philosophy XI: Studies in Essentialism (295–304), Minneapolis, MN: University of Minnesota Press. Pedroso, M. (2014) “Origin Essentialism in Biology,” Philosophical Quarterly 64.254: 60–81. Plantinga, A. (1974) The Nature of Necessity, Oxford: Clarendon. Prior, A. and Fine, K. (1977) Worlds, Times, and Selves, Amherst: University of Massachusetts Press. Pust, J. (2001) “Natural Selection Explanation and Origin Essentialism,” Canadian Journal of Philosophy 31.2: 201–220. Quine, W. (1960) Word and Object, Cambridge: MIT Press. Robertson, T. (= Robertson Ishii, T.) (1998) “Possibilities and the Arguments for Origin Essentialism,” Mind 107.428: 729–749. Robertson, T. (= Robertson Ishii, T.) and Forbes, G. (2006) “Does the New Route Reach its Destination?,” Mind 115.458: 367–373. Robertson Ishii, T. (2022) “Everything but the Kitchen Sink: How (Not) to Give a Plenitudinarian Solution to the Paradox of Flexible Origin Essentialism,” Philosophical Studies 179: 133–161, 10.1 007/s11098-021-01654-9. Rohrbaugh, G. and deRosset, L. (2004) “A New Route to the Necessity of Origin,” Mind 113.458: 705–725. Rohrbaugh, G. and deRosset, L. (2006) “Prevention, Independence, and Origin,” Mind 115.458: 375–385. Salmón, N. (1979) “How Not to Derive Essentialism from the Theory of Reference,” Journal of Philosophy 76: 703–725. Salmón, N. (1981) Reference and Essence, Princeton: Princeton University Press. Reprinted with original pagination in Salmón 2005. Salmón, N. (1986) “Modal Paradox: Parts and Counterparts, Points and Counterpoints,” in French, A., Uehling, T., Wettstein, H. (eds.), Midwest Studies in Philosophy XI: Studies in Essentialism (75–120), Minneapolis, MN: University of Minnesota Press. Reprinted in Salmón 2005: 273-244. Salmón, N. (2005) Reference and Essence (expanded second edition with new appendices), Amherst, NY: Prometheus Books. Salmón, N. (2021) “Modal Paradox [II]: Essence and Coherence,” Philosophical Studies 178: 3237–3250, 10.1007/s11098-020-01599-5. Skiles, A. (2015) “Essence in Abundance,” Canadian Journal of Philosophy 45.1: 100–112. Spencer, J. and Tillman, C. (2019) “Necessity of Origins and Multi-Origin Art,” Inquiry 62.7: 741–754. van Inwagen, P. (1983) An Essay on Free Will, Oxford: Clarendon Press. Wildman, N. (2013) “Modality, Sparsity, and Essence,” Philosophical Quarterly 63.253: 760–782. Wildman, N. (2021) “Against the Reduction of Modality to Essence,” Synthese 198: 1455–1471, 10.1 007/s111229-017-1667-6. Williamson, T. (2013) Modal Logic as Metaphysics, Oxford: Oxford University Press. Williamson, T. (2016) “Modal Science,” Canadian Journal of Philosophy 46.4-5: 453–492. Zylstra, J. (2019) “Essence, Necessity, and Definition,” Philosophical Studies 176: 339–350.

180

12 SCIENTIFIC ESSENTIALISM Travis Dumsday

12.1

Introduction

As the term is chiefly used here, “scientific essentialism” refers to a philosophical system founded by the influential Australian Brian Ellis. Focusing heavily as it does on the metaphysics of natural kinds and natural-kind essentialism (NKE), it would be helpful to begin this chapter with some quick introductory remarks about the latter (while also pointing the reader to the more detailed chapter entry on NKE, by Tuomas Tahko). To that end: while there are various ways of defining “NKE”, for present purposes let’s suppose it includes the following claim: our physical universe (the natural realm) contains individual objects, some or all of which belong to kinds, kinds whose essences function as the criteria for kindmembership and are either constituted by or ground the characteristics of those kinds.1 For example, our physical universe contains objects belonging to the kind electron, whose essence functions as the criterion for membership in that kind and is either constituted by or grounds the characteristics of that kind (such as negative charge, half-integral spin, and a precise rest mass of 9.109 × 10−28 grams). That is a deliberately inclusive, sunny formulation of a necessary condition for NKE, rigged with an eye to neutrality concerning well-known points of controversy amongst self-identified advocates of the view, such as: (i) whether every object must belong to a kind, or whether instead a truly bare particular is metaphysically possible (or even actual, with spacetime points occasionally plugged as candidates); (ii) whether a single object can belong to multiple real kinds simultaneously, or only one at a time; (iii) whether an object belongs to its kind(s) necessarily, or whether instead it can switch its kind-membership(s) while still remaining in existence; (iv) whether a kind-essence is reducible to its characteristics, or whether instead it is in some sense distinct from those characteristics and serves to ground them; (v) whether kinds are universals, such that NKE entails realism versus nominalism, or whether NKE can instead remain neutral here; (vi) what ontological category “object” refers to, and whether NKE entails realism about substances or instead remains compatible with other options (like substance-eliminativist bundle theories or versions of radical ontic structural realism); (vii) what sorts of characteristics constitute or are grounded in kind-essences (whether dispositional properties only or categorical properties only or both), etc. So there remains plenty of room for friendly in-house debates among NKE folk. It’s a big tent.

DOI: 10.4324/9781003008750-15

181

Travis Dumsday

What unites those sheltering under that tent? The most effective unifier of all: common enemies. Proponents of NKE have faced an imposing horde of both analytic and continental opponents, from Humeanism to constructivism to positivism to Kantian idealism to postmodernism and beyond. Indeed, during the heyday of positivism within anglophone philosophy, explicit development and defence of NKE was largely restricted to holdouts operating outside the analytic mainstream (e.g., Scholastics, Peircean Pragmatists, funky Neoplatonists). With the demise of positivism throughout the 1950s and 1960s and the attendant revival of metaphysics within analytic philosophy, discussions of essentialism returned and achieved considerable prominence. Instrumental in this revival were the works of Saul Kripke and Hilary Putnam, but others made valuable contributions in the 1970s and 1980s—e.g., Roy Bhaskar (1975), Baruch Brody (1980), Nancy Cartwright (1989), Evan Fales (1979, 1986), Milton Fisk (1973), Alfred Freddoso (1986), Rom Harré & E.H. Madden (1975), Michael Loux (1974; 1978), E.J. Lowe (1989), and Martin Tweedale (1984). Though sometimes deploying empirically based thought experiments and illustrations, they and others working on NKE were for the most part interested primarily in its semantics and metaphysics. With some notable exceptions (like Bhaskar and Cartwright), on the whole they were less interested in the details of how NKE linked up (or not) with either science or the philosophy of science. (That is not to say that they were ill-informed about the science, but simply that the projects they were pursuing tended not to involve detailed engagement with it.) The single most important figure in establishing a sturdier bridge between NKE and philosophy of science was Brian Ellis. In doing so, he also played a significant role in solidifying the metaphysics of science as a respectable sub-discipline. He did so by way of a now relatively uncommon scholarly path: the construction of a philosophical system. He christened it “scientific essentialism” (SE). That label had appeared earlier in the literature, sometimes to refer to the views of Kripke and Putnam, but Ellis’ work gave it both a new lease on life and a different import. Ellis’ SE is a relatively comprehensive system that integrates NKE with a broader set of metaphysical and epistemological commitments, usually with an eye to implications for major debates in philosophy of science (such as that between scientific realism versus anti-realism). By way of quickly clarifying how SE relates to NKE and dispositional essentialism (DE), each of which has a separate entry in this volume: SE entails a commitment to both NKE and DE, insofar as both are core components of Ellis’ system. However, the reverse entailment certainly does not hold; one can be a proponent of either NKE or DE, or both, without adopting SE. Ellis’ development and defence of SE over the course of a series of papers and books from the early 1990s up through the early 2010s sparked a good deal of discussion; that discussion involved not only SE conceived as his own distinctive approach, but also analogous approaches tying NKE together with wider issues in the philosophy of science. Today the label “scientific essentialism” is periodically used to reference such analogous approaches, but for the most part it remains associated with Ellis’ system. My aim here is to provide a concise overview of SE as developed by Ellis. The remainder is divided as follows: in the next section I summarize the main commitments of SE as laid out in key works, most importantly his 2001 monograph Scientific Essentialism. I pay special attention to the roles played by NKE within SE, including Ellis’ distinctive views on issues (i)–(vii). In section three I conclude with some brief evaluative comments. 2

182

Scientific Essentialism

12.2

Ellis’ System: Scientific Essentialism

SE is a fairly wide-ranging system, and when discussing any such system it can be difficult to know quite where to begin, or which thread of the tapestry to start tugging on. Still, having alleged that advocates of NKE are unified most of all by a shared set of opponents, perhaps it would be appropriate to begin at the spot where Ellis himself sometimes begins when laying out the basics of his system: its contrast with Humeanism. Ellis writes (2000: 329–30): By ‘Humeanism’, I mean a group of related theses: (a) that matter is essentially passive, (b) that the laws of nature are behavioural regularities of some kind, (c) that the laws of nature are contingent, (d) that causal relations hold between logically independent events or states of affairs, (e) that the identities of objects are independent of the laws of nature, (f) that the dispositional properties of things are not intrinsic, but depend on what the laws of nature happen to be, and (g) that modal properties supervene logically on non-modal properties, so that we could not have any difference in modal properties if there were no differences in non-modal properties. I do not say that every Humean holds all of these theses, or even that they all originate with Hume. But they do constitute what most philosophers would consider to be a Humean package. The Humean package does not derive only from Hume, but has roots in the whole mechanistic philosophical movement which led up to him. Ellis understands Humeanism to be the conjunction of those seven claims, which he sees as interconnected. SE differs from Humeanism on each point. It maintains that material objects are not inert, but rather are intrinsically possessed of a range of active and passive causal powers (i.e., capacities/potencies/dispositions). What we typically think of as “laws of nature” supervene on and flow from those powers and their regular expression. Laws are thus neither extrinsic governing principles dictating the behavior of otherwise static lumps of matter, nor are they mere descriptions of regularities devoid (bizarrely) of any ontological grounding. Because laws depend on powers, laws are contingent only in the sense that it remains a contingent fact which powers happen to be instantiated in our world. But once a certain set of powers is in place, the laws are entailed thereby (with absolute necessity). The powers are in turn intimately related to the identities of the objects to which they belong, and in that manner the identities of objects are not independent of the laws of nature. Returning to our earlier example, negative charge is a causal power enabling its possessor to repel other negatively charged particles and to attract the positively charged (with a certain degree of force, along a certain vector, subject to a host of ceteris paribus clauses …). An electron is inherently negatively charged, such that as a material object it is far from inert; consequently, its identity as that sort of particle cannot be independent from the laws of nature, in this case the laws of electromagnetism. Given that it is a power, negative charge also counts as a modal property, insofar as it helps to delimit the range of things that an electron can and cannot do—its range of real-world possibilities, in other words. Ellis puts it as follows (2001: 48): “If a and b are electrons, then it is necessarily true that they are negatively charged. And, necessarily, if a and b are negatively charged, they are intrinsically disposed to repel each other. Therefore, if a and b are electrons, it is necessarily true that they are intrinsically disposed to repel each other. It is therefore a necessary truth, not a contingent one, that electrons are intrinsically disposed to repel each other”. Again, the necessities here are absolute; in every possible world in which there is an electron, that electron is negatively charged, etc. It is utterly impossible that there be an electron lacking negative charge. These

183

Travis Dumsday

are metaphysical necessities about members of a kind: in every possible world in which the kind is instantiated by individual objects, certain facts follow about those objects. For Ellis such necessities carry the same modal weight as pure logical necessities (like the principle of non-contradiction), but he will often distinguish verbally between logical and metaphysical necessities by way of emphasizing that the latter are not purely formal—they obtain of real entities in our actual world. This is another contrast with the Humean, who is happy to admit logical necessities (necessary formal relations between abstract concepts) but for whom such modal force never obtains in relationships between real entities in the physical world. For Ellis, by contrast, what happens between an electron and a proton approaching each other in physical space (leaving aside built-in ceteris paribus clauses) is as strongly necessary as any logical theorem.3 Those are not the only points of contrast between SE and Humeanism—others include SE’s commitment to the reality of universals (in contrast to Humean nominalism) and its rejection of strict empiricism—but they are among the points of contrast that are especially significant when it comes to their respective implications for thinking about the scientific enterprise and its relationship to metaphysics. Ellis’ rejection of strict empiricism contributes to his particular brand of scientific realism, namely, causal process scientific realism (a label he employs in his (2001: 157)). On this view, science properly aims at discovering objective truths about the physical world (including the unobservable realm), and scientists rightly accept as real any entities playing an ineliminable role in causal explanations. This version of scientific realism thus boasts a more robust requirement for realism than do some others, since in order to count as having a real referent it is not enough that some term or concept merely show up in our most successful scientific theories (a standard that would likely entail adopting mathematical Platonism (among other things), which Ellis rejects). Rather, what gets posited as having real referents are those terms or concepts needed for explaining the behavior of things. And when there are multiple competing claims about what the relevant truthmakers are for the propositions of some true scientific theory, and science itself cannot adjudicate between those claims, then at times it becomes the job of metaphysics to discover which truthmakers are the best candidates and which (if any) can be decisively ruled out. For example, physics may not be able to decide by itself whether the laws of electromagnetism are made true by the causal powers of the kinds involved in electromagnetic interactions or instead are made true by transcendent governing abstracta (as on Maudlin’s (2007) primitivist ontology of laws) or divine decree (as on Foster’s (2004) theistic ontology of laws) or by nothing whatever (as on Humeanism). But the metaphysician can attempt to develop arguments showing that the first alternative is rationally preferable in various ways (e.g., in terms of simplicity and explanatory scope and fit with background knowledge) to its competitors. Ellis sums up part of this general perspective (2009: 7–8): Science is not limited, as the early positivists believed it to be, to just describing what can be observed, and constructing theories that will enable scientists to make sound predictions concerning future observations. Scientists do construct realistic theories about things, and some of them should certainly be taken seriously as descriptive of what there is. But not all theories are like this, and what many of them tell us about reality is unclear. There are also good reasons to think that there are essential limitations to scientific knowledge of the world, and that there are important questions that arise out of science about the nature of reality that cannot be answered by further scientific inquiry. Answering these questions is what metaphysics is all about … . It is reasonable,

184

Scientific Essentialism

therefore, to identify the task of metaphysics as being to identify the kinds of truthmakers there are for the various kinds of known truths about the world, and to construct an overall picture of reality that is adequate to accommodate them. The connection between essentialism and causal process realism also highlights one of the ways in which SE differs from other versions of the doctrine, such as traditional Aristotelian essentialism and Kripkean essentialism, which Ellis thinks were much less invested in essences having an explanatory role.4 In terms of historical precursors he points to Locke’s essentialism as closer in spirit to his own, positing as it does a role for hidden (i.e., empirically unobservable) essences as explanatory factors in the observable properties and behaviors of objects. Yet Locke himself devoted little attention to defending the reality of such hidden essences.5 And of course we today have a remarkable advantage unforeseen by Locke, in that the essences of at least some natural kinds are no longer hidden, but are in fact well-known to science (2001: 55): “The basic sciences of physics, chemistry, and microbiology have now revealed many of the intrinsic properties and structures of the things that exist in nature, and we are now justifiably able to say what makes them the kinds of things they are and why they are disposed to behave and display the properties they do”. Later on he adds (2001: 247) that “the essential properties of things are not occult, but discoverable and measurable by the ordinary methods of science. There is nothing especially hidden or obscure about them. They can be found just by observing how the instances of the various natural kinds act on, and interact with, each other”. We have already seen some components of the metaphysics Ellis is constructing, by way of supplying those genuine truthmakers for scientific truths and a larger ontological framework in which to situate them. Let’s turn now to examine more of its details, beginning with his account of natural kinds. For Ellis’ SE, there are six fundamental ontological categories: substances (both substanceuniversals like electron and particular concrete substances such as individual electrons); properties (both property-universals like negative charge and particular concrete chargetropes); and processes (both process-universals like electromagnetic attraction and their concrete dynamic instantiations). He observes (2005b: 462) that this is rather like E.J. Lowe’s well-known 4-category ontology, except with the addition of processes.6 Ellis recognizes that although it is common to use “natural kinds” as synonymous with “natural kinds of substance”, and although he often uses the label in this way himself, strictly speaking it applies not only to substance-universals but also to property-universals and process-universals. Each category of universal (or each kind of kind, to speak more confusingly) comes in a hierarchy, with the bottom level of the hierarchy consisting of infimic (i.e., lowest-level/most specific) species and the higher levels consisting of generic/determinable universals.7 As an example of some levels within the substance-universal category, Ellis (2001: 74) lists as a very general kind element, as a less general species of that kind inert gas, and as an infimic species of that same kind helium atom. All levels of the hierarchy are objectively real (in fact infimic species depend for their existence on the higher-level kinds they necessarily fall under),8 though the higher levels of the hierarchy are only instantiated via the instantiation of an infimic species (i.e., no concrete object is just an element without also being some more specific sort of element, or just a shape without also being a square versus a circle). No universals can exist uninstantiated—Ellis persistently rejects Platonic realism in favor of moderate realism, writing for instance that (2009: 97) “universals, as we normally understand them, are Aristotelian, that is, they exist if and only if they are instantiated”. Ellis openly acknowledges that this creates complications for how to think about some properties (especially the uninstantiated but seemingly legitimate

185

Travis Dumsday

determinate values of instantiated determinable universals) and also how to think about necessary mathematical truths, and at times he seems conflicted on this issue; however, in the end he maintains (2009: 116–9) that these properties and truths can be accommodated within SE without having to adopt outright Platonism. Natural kinds of substance have kind-essences consisting in a set of properties (in the case of fundamental substances like electrons that have no proper parts) or properties + structure (in the case of compound substances like molecules). He writes (2001: 22): “Therefore, the sets of properties or structures that define these natures and thus distinguish the natural kinds from each other, are distinctive of the natural kinds. Traditionally, such properties have been called the essential properties of the kinds, and together they constitute what is known as the real essence of the kind”. All real natural kinds have essences, such that the lack of an essence marks a putative kind as non-natural, or at least less-than-wholly natural. (Observe too that while Ellis occasionally contrasts structures with properties (as he does in the preceding quote), he does in fact categorize structures as properties, specifically as categorical rather than dispositional properties. He considers relations likewise to be properties (e.g., for him a distance relation between two particles counts as a categorical property). This fits in with his aforementioned six-category ontology, in which structures and relations don’t get categories of their own.) For Ellis, one implication of the preceding point is that true natural kinds are only found in physics, chemistry, and maybe some areas of microbiology. What we think of as typical biological species—cats and dogs and cacti—are not natural kinds, in part because their high degree of individual genetic variability means that they lack a shared essence. (E.g., two cats are not intrinsically type-identical in the way that two electrons are intrinsically typeidentical.) Biological taxonomy involves identifying lineally related organisms sharing clusters of various traits, but that does not require that they be viewed as genuine natural kinds—in fact we are better off referring to them by a different label altogether, that of “cluster kinds” (2001: 31–2 and 168–70; 2002: 28–32 and 154–7; 2009: 58–60).9 As biological organisms, we human beings also fail to count as members of a genuine natural kind. Consequently, none of the social sciences studies a genuine natural kind. And since all laws of nature depend on genuine natural kinds and their essential properties, no putative social science laws (like the “laws” of economics) actually count as laws, strictly speaking (2001, ch. 5; 2002: 157–65). When it comes to the ontological makeup of individual substances, Ellis affirms what has become known as “primitive substance theory”. PST has a number of defenders in the recent literature.10 The idea is that an individual substance just is the instantiation of a substanceuniversal; in other words, a substance-universal is instantiated by an individual substance, not in an individual substance. Particular objects are not compounds of more basic ontological ingredients, as posited by competitor accounts like substratum theory (wherein a substance is a compound of substratum + attributes) and some versions of hylomorphism (wherein a substance is a compound of prime matter + substantial form). Rather, objects are truly basic ingredients in ontology. A particular object may have spatially distinct parts, but it is not composed out of radically diverse ontological principles in the way suggested by those competitors. This was also part of Ellis’ original reason for resisting the reduction of substance-universals to property-universals—i.e., the risk that attempting the reduction might have the unintended consequence of pushing some toward substratum theory, should they presume (contra bundle theory) that an instantiated property-universal would have to be instantiated in something. Ellis writes (2001: 91–2):

186

Scientific Essentialism

For I do not believe, and never have, in the existence of bare particulars, or featureless substances, in which the supposed properties of being an electron or being a proton might be instantiated … . The idea that there is a property of being an electron which something or other, presumably an underlying substance, might have does not belong to the conceptual framework of modern science. It belongs, rather, to the conceptual scheme of the Council of Trent, which attempted to make sense of the Catholic doctrine of transubstantiation. If you can believe in a metaphysics which allows that the ‘species essence’ of electronhood can inhere in a substance, the identity of which is independent of the properties that constitute this essence, then I shall acknowledge your entitlement to believe that there are things that might, or might not, have the property of electronhood. But this seems to me to be just medieval mumbo-jumbo of a kind that has no place in a modern philosophy of science. Leaving aside Ellis’ readily understandable misunderstanding of the complex Roman Catholic doctrine of transubstantiation,11 this does underscore part of his justification for adopting PST in preference to major alternatives. Regarding the property category, we have already encountered one of Ellis’ core commitments, namely, his dispositionalism—i.e., his claim that causal powers are among the irreducible properties in nature. He presents multiple arguments in favor of this view; for instance, he notes that the laws of nature need some sort of ontological ground (it is implausible to maintain with the Humeans that exceptionless regularities are just brute facts), and that categorical properties cannot do that explanatory work by themselves (2001: 115): “There are no regularities of behaviour that are specific to things of a given shape or size, for example, or to the members of the extensions of any other categorical property, as there are laws governing the interactions of charged or massive particles”. Categorical properties seem to entail no specific behaviors (knowing something’s shape or size tells us nothing about what sorts of things it can do or have done to it), so it makes sense to think that there is some other sort of property that does carry such entailments: powers. As we have seen already, Ellis sees powers as providing the ground for laws of nature—or, more specifically, the causal laws of nature. And while the various versions of nomological necessitarianism (especially the Dretske-Tooley-Armstrong theory or “DTA theory”) are superior to Humean and conventionalist accounts of laws, they suffer from serious difficulties of their own (2001: 215–7; 2002: 97–100), such that dispositionalism comes out as the unambiguous leader of the pack.12 It is worth observing too that while he tends to focus on causal laws in his writings on the nomic, Ellis (2000: 331) does affirm that there are other sorts of laws having other sorts of grounds: “(a) the global laws, which describe the intrinsic nature of the events and processes which are metaphysically possible in our world, (b) the general structural principles, such as those of General Relativity, and Pauli’s Exclusion Principle, which define the spacetime or causal structure of all worlds of the same natural kind as ours, and (c) laws of essential nature, which describe the real essences of the various natural kinds of things, such as fundamental particles, chemical substances, and fields, existing in our world”. Dispositionalism has become a fairly popular view in the current literature (thanks in part to Ellis), and there have emerged several competing ideas concerning the relationship between dispositional and categorical properties. Some believe that both are equally real and fundamental, a position often known as “mixed view dispositionalism”. This is Ellis’ consistent position (2001: 127; 2002: 68–70 and 173–5; 2009, ch. 5), and it certainly has other defenders.13 Opposed to it are “pan-dispositionalism” (on which all fundamental properties are dispositional),14 “identity theory dispositionalism” (on which all fundamental properties

187

Travis Dumsday

have both dispositional and categorical identity conditions),15 and “neutral monism” (on which the dispositional/categorical distinction is held to be merely conceptual, such that in reality the fundamental properties are of themselves neither dispositional nor categorical).16 Ellis takes a stance on several other in-house debates amongst dispositionalists, for instance arguing for the reality of both single-track and multi-track dispositions17; for the idea that all natural necessities (as opposed to logical necessities and analytic truths) are rooted in kinds and their essential powers18; for the interdefinable nature of dispositions and the causal processes they ground, even though the latter belong to an irreducibly distinct ontological category19; and for the claim that the range of dispositions instantiated in our world is circumscribed not by any higher-level laws or other ontological principles, but rather by the kind-essence of our world itself, which belongs to its own variety of natural kind (thereby also providing a solution to the much-discussed “global laws problem” facing dispositionalism).20 This has obviously been quite a cursory overview of SE. Much more could be said about the commitments laid out above and Ellis’ supporting arguments for them. And there are interesting areas of the system I haven’t even touched on here (e.g., its distinction between incidental versus accidental properties (2001: 76–8), its interpretation of quantum mechanics (2009, ch. 4), its philosophy of time (2009: 119–28), its account of free-will (2009: 129–39)). But before concluding with a few brief evaluative remarks, it might be useful to recap his take on issues (i)–(vii), as laid out in the Introduction. So: i For Ellis, every object must belong to at least one kind. No bare particulars are allowed, whether inherently bare yet always instantiating some contentful property or properties (the mainstream view among advocates of bare particularism) or actually bare. ii A single object can belong to multiple real kinds simultaneously, provided that this be understood in the sense of its occupying a place in an extensive hierarchy of kinds (E.g., an individual electron belongs to the infimic species electron and only to that infimic species, but it also belongs to the higher-level generic kind lepton, the still-higher-level generic kind fundamental particle). Ellis does not, however, countenance a single thing belonging simultaneously to multiple disjoint kinds that aren’t placed within a single hierarchy. iii An object cannot switch out its most basic kind-membership (its infimic species) for another, since nothing would persist through the change. (Again, no bare particulars allowed.) iv A kind-essence is reducible to the properties and structures that constitute it. (I.e., there is no essence over and above those properties, ontologically prior to them and grounding them.) v Kinds are universals. vi Ellis originally maintained that substances are real and irreducible (both individual substances and substance-universals) and that primitive substance theory is the correct substance ontology (as opposed to substratum theory or bundle theory or hylomorphism). However, in his later work he changed this aspect of his system and argued that substances are reducible to properties + processes. vii The characteristics constituting a kind-essence can include both dispositional and categorical properties.

12.3

Assessing Scientific Essentialism

There can be no doubt about the originality of Ellis’ system, nor about its widespread and well-deserved influence. Many prominent philosophers of science have engaged with Ellis’ work over the years,21 and he continues to be cited regularly.

188

Scientific Essentialism

Still and all, a philosophy with which one can battle on at least some fronts is more interesting than one which is wholly agreeable, so I would like to end by pointing to some areas of Ellis’ system that could perhaps use some further development and defence (or modification). First and perhaps most importantly, Ellis may have been too quick to accept the idea that biological species must be cluster kinds rather than natural kinds, and by extension too dismissive of the prospects for a workable form of intrinsic biological essentialism. While still very much a minority position within philosophy of biology, versions of that view have attracted a number of recent defenders.22 (This is on top of much interesting work done on relational and historical essentialisms in biology.) One upshot of this work is that the genetic variability displayed amongst individual organisms of the same species may not be an insuperable barrier to recognizing them as belonging to real biological kinds with real kindessences. It may be time for SE to open its tent a little wider by way of welcoming in the biological sciences. That could in turn have the effect of opening up the social sciences to analysis by SE, given that Ellis’ central reason for excluding them was his contention that human beings are not members of a single real natural kind. (For further discussion of biological essentialism, see again the chapter by Brigandt in the present volume.) Second, some have questioned Ellis’ view that kind-essences must be reducible to their associated defining properties and structures, rather than being in some sense distinct from (and ontologically prior to) those properties and structures. Oderberg (2007; 2011) for instance argues that the latter picture of kind-essences is preferable because it plays a necessary role in solving the so-called “unity problem”—i.e., the question of what holds together the prima facie separable essential properties of a kind. Moreover, Dumsday (2013; 2019) argues that recognizing a priority of kind-essence over essential properties plays a key role in the defence of dispositionalism. Third, Ellis’ (2009) switch from his previous longstanding realism about substance to a bundle theory/process ontology is questionable. While his earlier preferred substance ontology—primitive substance theory—does face significant objections (e.g., surrounding its ability to provide a workable account of individuation), in my opinion it remains preferable to any thoroughgoing reductionism about the substance category, for reasons familiar from the substance ontology battlefield (e.g., Martin’s (1980) well-known allegation that bundle theorists are driven into conceptual confusions regarding the parts versus the traits of a thing). Fourth and finally, there seems a tension in Ellis’ system between his moderate realism about universals and his acceptance not merely of the reality of determinable/generic universals but even their ontological priority over infimic species. Historically, the latter claim especially has been more associated with Platonic than Aristotelian accounts of universals, and for good reason. Similarly (though more controversially) there is good reason to think that Platonism is a better fit with primitive substance theory than is Aristotelianism, so at least if one is working with the earlier substance-realist version of SE then that seems a further motive for considering a switch to a Platonic ontology of universals. Such suggestions for altering SE and/or shoring up its defences amount to relatively minor quibbles. It remains a highly significant contribution to the philosophy of science and will surely continue to generate profitable discussions and new developments in the field.23

12.4

Related Topics

Natural-Kind Essentialism, by Tuomas Tahko Dispositional Essentialism, by Ka Ho Lam Biological Essentialism, by Ingo Brigandt

189

Travis Dumsday

Notes 1 I take this formulation of NKE to be compatible with (though not strictly equivalent to) the more concise formulation helpfully provided by Tahko in his contribution to the present volume: “Each individual member of a given kind K shares some property or properties (general essence) that are essential to members of K”. A few additional provisos/qualifications worth mentioning: (A) My formulation of NKE is stated in terms of individual objects in the physical realm, but this is not intended to exclude the possibility of individual objects existing non-physically and belonging to non-physical kinds. (B) Similarly, my formulation is not intended to exclude the possibility of there being members of natural kinds that are not themselves individual objects. Some ontologies posit entities that exist independently and are irreducibly real but that cannot plausibly be considered individual objects (e.g., the “substances” of one-category universalist bundle theory, according to which what we think of as individual objects are actually bundles of compresent universals; those ontologies in which there is a fundamental distinction between objects and stuff, and only the latter is ultimately real; radical versions of ontic structural realism, in which what we think of as individual objects are actually irreducible structures or matrices of relations). Such ontologies may be compatible with NKE, if the entities they posit can plausibly be characterized as belonging to kinds the essences of which function as criteria for kind-membership. However, I take no stance on this here. (C) In speaking of NKE involving kinds having criteria for membership, one might be led to question whether the positing of homeostatic property cluster kinds (HPC-kinds) counts as affirming a version of NKE. There is room for dispute about this, but my own inclination is to think that HPCkinds may fit within the rubric of NKE provided that one is willing to countenance kinds having disjunctive or otherwise fuzzy membership criteria. However, as we shall see later in this piece, Brian Ellis thought that HPC “kinds” did not properly count as natural kinds in the fully robust sense he advocated in his version of NKE. 2 A point of clarification at the outset: an overview of SE does not amount to an overview of Brian Ellis’ philosophical work—far from it. Today an Emeritus Professor at La Trobe University, he has been publishing in the field since the 1950s. A proper accounting of his impressive corpus would be a valuable (and much larger) project, but for present purposes our focus must be narrower. 3 For further discussion of debates surrounding the connections between essence and modality, see the entries by Wildman and Correia in the present volume. 4 A case could certainly be made that Ellis has not taken into sufficient account the ways in which essences play an explanatory role in Aristotle’s essentialism. For more on the latter, see the entry in this volume by Malink. 5 Elsewhere he writes (2000: 330): “One has to go all the way back to Aristotle to find a truly notable defender of essentialism”. Ellis has nothing good to say about the Scholastic tradition. Fan that I am of John Duns Scotus et al., and given that the Scholastics actually developed some interesting and novel defences of NKE, this seems to me a minor blindspot in his scholarship. 6 Later on, Ellis (2009: 66–9) drops the substance category, suggesting that it is reducible to properties + processes. (Thus all substance-universals would be analyzable without remainder into propertyuniversals and/or process-universals, and by parity all individual substances would likewise be reducible to individual properties and/or processes.) However he develops this model less thoroughly than his earlier six-category ontology, so I will focus on that earlier account here. 7 Ellis never says whether he thinks there is a highest-level generic kind or kinds, prudently dodging Neoplatonic knots over the reality of Being. Regarding the now somewhat archaic terminology of “infimic” species, this was a common term in older literature on kinds, and refers simply to the lowest-level member of the relevant kind-hierarchy. For instance, members of the alleged biological kind “cat” belong to a great many higher-level kinds (mammal, living thing, physical substance, substance, etc.), but to no real lower-level kinds. “Cat” would in this sense be an infimic species, its members belonging also to higher-level kinds but not to anything lower down on the Porphyrian tree (or its modern analogues). 8 For instance, he writes ( 2005a: 78): “Within these hierarchies of kinds, the more general ones are ontologically more fundamental than the more specific. They are more fundamental (a) because they cannot be constituted by their species: the generic property of mass, for example, cannot be thought of as a disjunctive property that includes all specific masses as disjuncts, because not every specific mass is instantiated, and (b) because the generic kinds must exist if any of their species exist, but the more specific kinds need not exist if the generic kind does. The hierarchies of kinds are thus

190

Scientific Essentialism

9 10 11

12 13 14 15 16 17

18

19

20 21 22 23

hierarchies of ontological dependence, with those kinds at the lower levels being ontologically dependent on those above them”. For more on the debates surrounding biological essentialism, see the entry in this volume by Brigandt. See Broackes (2006), Fales (1990), Hoffman (2012), Loux (1974; 1978; 2002), Lowe (1994; 1998; 2000; 2006; 2012; 2013), Macdonald (2005), and Wiggins (2001). While some mediaeval theologians (like Giles of Rome) did employ inherently propertyless prime matter as part of their model for how transubstantiation works, in fact most of those accounts (including the problematic Thomist model that eventually became normative within Catholicism) make no use of it, and could in fact be formulated in terms compatible with PST. Having whetted readers’ appetites for a deep dive into the metaphysics of transubstantiation (right??), I would recommend Adams (1992) as a fine starting point, and my own (forthcoming) for the full gospel truth. For more on dispositional essentialism, see the entry in the present volume by Lam. See for instance Cross (2005), Molnar (2003), and Oderberg (2007). See Bauer (2013), Bird (2007), Bostock (2008), Coleman (2010), Mumford & Anjum (2011), and Shoemaker (1980). See Heil (2003; 2005b; 2012), Martin and Heil (1999), Ingthorsson (2013), Jacobs (2011), and Strawson (2008). See Bartels (2013) and Mumford (1998, ch. 8). One is tempted to cite Lowe (2001; 2006) as another neutral monist, but his version of dispositionalism is so distinctive that to classify it as falling within any of these four groups would be contentious. See Dumsday (2016) for more on Lowe’s account. Ellis & Lierse (1994: 29) write: “In the simplest cases, dispositions are said to be single-track. That is, they can manifest themselves in only one kind of circumstance, and then only in one kind of way. Such dispositions are characterised by a single subjunctive conditional. More often than not, however, dispositions are multi-track. That is, they can reveal their presence in a range of antecedent circumstances, yielding to a range of different consequent events”. One implication of which is that imaginability is not a proper test of metaphysical possibility, contra the Humean. Ellis writes ( 2001: 232–3): “The imaginability test of logical possibility derives from the assumption that what a thing can do or become depends only on its manifest image. For this is what the imagination has to work with. It starts with the manifest image and transforms it. But why should we suppose that all imaginable transformations are possible? … .What is really possible for a given thing to do or become does not depend only on the transformability of its manifest image. It depends on what kind of a thing it is, and how and of what it is constituted … . There might conceivably be a creature in some possible world that looks like a horse, which can transform into something that looks like a cow. But it could not possibly be a horse since horses are incapable of any such transformations. Nor could the result of the transformation be a cow because cows cannot be produced in this way”. Ellis (2000: 332) writes: “The causal processes which are the displays of a given kind of causal power must be essentially similar. For the same kind of power must always have the same law of action, whether that law be deterministic or probabilistic. Therefore, the processes which arise from the exercise of a given kind of causal power must all be processes of the same natural kind. Essentialism therefore leads us to think of the world as containing, not only natural kinds of objects or substances, but also natural kinds of processes. The processes we call ‘causal processes’ are thus species of natural kinds of processes”. Later he adds ( 2000: 343): “Causal powers, and the natural kinds of processes which are their displays, thus go hand in hand, and are interdefinable”. See especially Bigelow, Ellis & Lierse (1992). I will not attempt to provide a rundown of the many discussions of Ellis’ system present in the literature, but here are a few representative examples: Armstrong (1999); Hacking (2007); Heil (2005a); Khalidi (2009); and Lange (2004; 2005). See for instance Austin (2017; 2019), Devitt (2008; 2011; 2018a; 2018b; 2023), Dumsday (2012), Oderberg (2007), and Walsh (2006). I would like to extend my sincere thanks to Kathrin Koslicki and Mike Raven for the kind invitation to contribute to the volume, and for their helpful and extensive input on several early drafts. My thanks also to my many fellow contributors, who provided useful questions and suggestions during our online workshop in the summer of 2021. Finally, tremendous gratitude is of course owed to Brian Ellis for his remarkable contributions to analytic metaphysics and philosophy of science.

191

Travis Dumsday

References Adams, M. (1992) “Aristotle and the Sacrament of the Altar: A Crisis in Medieval Aristotelianism”, in R. Bosley and M. Tweedale (eds.) Aristotle and His Medieval Interpreters, Calgary, AB: University of Calgary Press, 195–249. Armstrong, D. (1999) “The Causal Theory of Properties: Properties According to Shoemaker, llis, and Others,” Philosophical Topics 26, 25–37. Austin, C. (2017) “Aristotelian Essentialism: Essence in the Age of Evolution,” Synthese 194, 2539–2556. Austin, C. (2019) Essence in the Age of Evolution: A New Theory of Natural Kinds, New York, NY: Routledge. Bartels, A. (2013) “Why Metrical Properties are not Powers,” Synthese 190, 2001–2013. Bauer, W. (2013) “Dispositional Essentialism and the Nature of Powerful Properties,” Disputatio 5, 1–19. Bhaskar, R. (1975) A Realist Theory of Science, Leeds: Leeds Books. Bigelow, J.. Ellis, B., and Lierse, C. (1992) “The World as One of a Kind: Natural Necessity and Laws of Nature,” British Journal for the Philosophy of Science 43, 371–388. Bird, A. (2007) Nature’s Metaphysics: Laws and Properties, Oxford: Oxford University Press. Bostock, S. (2008) “A Defence of Pan-Dispositionalism,” Metaphysica 9, 139–157. Broackes, J. (2006) “Substance,” Proceedings of the Aristotelian Society 106, 131–166. Brody, B. (1980) Identity and Essence, Princeton, NJ: Princeton University Press. Cartwright, N. (1989) Nature’s Capacities and Their Measurement, Oxford: Oxford University Press. Coleman, M. (2010) “Could there Be a Power World?” American Philosophical Quarterly 47, 161–170. Cross, T. (2005) “What is a Disposition?” Synthese 144, 321–341. Devitt, M. (2008) “Resurrecting Biological Essentialism,” Philosophy of Science 75, 344–382. Devitt, M. (2011) “Natural Kinds and Biological Realisms”, in J. K. Campbell, M. O′Rourke, and M. Slater (eds.) Carving Nature at its Joints: Natural Kinds in Metaphysics and Science, Cambridge, MA: MIT Press, 155–173. Devitt, M. (2018a) “Historical Biological Essentialism,” Studies in History and Philosophy of Biological and Biomedical Sciences 71, 1–7. Devitt, M. (2018b) “Individual Essentialism in Biology,” Biology & Philosophy 33, 1–22. Devitt, M. (2023) Biological Essentialism, Oxford: Oxford University Press. Dumsday, T. (2012) “A New Argument for Intrinsic Biological Essentialism,” Philosophical Quarterly 62, 486–504. Dumsday, T. (2013) “Using Natural-Kind Essentialism to Defend Dispositionalism,” Erkenntnis 78, 869–880. Dumsday, T. (2016) “Lowe’s Unorthodox Dispositionalism,” Res Philosophica 93, 79–101. Dumsday, T. (2019) Dispositionalism and the Metaphysics of Science, Cambridge: Cambridge University Press. Dumsday, T. (Forthcoming) “Transubstantiation Through the Lens of Spacetime Substantivalism,” Theology & Science. Ellis, B. (2000) “Causal Laws and Singular Causation,” Philosophy & Phenomenological Research 61, 329–351. Ellis, B. (2001) Scientific Essentialism, Cambridge: Cambridge University Press. Ellis, B. (2002) The Philosophy of Nature: A Guide to the New Essentialism, Montreal & Kingston: McGill-Queen’s University Press. Ellis, B. (2005a) “Marc Lange on Essentialism,” Australasian Journal of Philosophy 83, 75–79. Ellis, B. (2005b) “Universals, the Essential Problem and Categorical Properties,” Ratio 18, 462–472. Ellis, B. (2009) The Metaphysics of Scientific Realism, Montreal & Kingston: McGill-Queen’s University Press. Ellis, B., and Lierse, C. (1994) “Dispositional Essentialism,” Australasian Journal of Philosophy 72, 27–45. Fales, E. (1979) “Relative Essentialism,” British Journal for the Philosophy of Science 30, 349–370. Fales, E. (1986) “Essentialism and the Elementary Constituents of Matter,” Midwest Studies in Philosophy, 11, 391–402. Fales, E. (1990) Causation and Universals, New York, NY: Routledge.

192

Scientific Essentialism Fisk, M. (1973) Nature and Necessity: An Essay in Physical Ontology, Bloomington, IN: Indiana University Press. Foster, J. (2004) The Divine Lawmaker: Lectures on Induction, Laws of Nature, and the Existence of God, Oxford: Oxford University Press. Freddoso, A. (1986) “The Necessity of Nature,” Midwest Studies in Philosophy 11, 215–242. Hacking, I. (2007) “Natural Kinds: Rosy Dawn, Scholastic Twilight,” Philosophy: Journal of the Royal Institute of Philosophy 82 (Supplement 61), 203–239. Harré, R., and Madden, E. H. (1975) Causal Powers: A Theory of Natural Necessity, Oxford: Blackwell. Heil, J. (2003) From an Ontological Point of View, Oxford: Oxford University Press. Heil, J. (2005a) “Kinds and Essences,” Ratio 18, 405–419. Heil, J. (2005b) “Dispositions,” Synthese 144, 343–356. Heil, J. (2012) The Universe as We Find It, Oxford: Oxford University Press. Hoffman, J. (2012) “Neo-Aristotelianism and Substance,” in T. Tahko (ed.) Contemporary Aristotelian Metaphysics, Cambridge: Cambridge University Press, 140–155. Ingthorsson, R. D. (2013) “Properties: Qualities, Powers, or Both?” Dialectica 67, 55–80. Jacobs, J. (2011) “Powerful Qualities, Not Pure Powers,” Monist 94, 81–102. Khalidi, M. (2009) “How Scientific is Scientific Essentialism?” Journal for the General Philosophy of Science 40, 85–101. Lange, M. (2004) “A Note on Scientific Essentialism, Laws of Nature, and Counterfactual Conditionals,” Australasian Journal of Philosophy 82, 227–241. Lange, M. (2005) “Reply to Ellis and Handfield on Essentialism, Laws, and Counterfactuals,” Australasian Journal of Philosophy 83, 581–588. Loux, M. (1974) “Kinds and the Dilemma of Individuation,” Review of Metaphysics 27, 773–784. Loux, M. (1978) Substance and Attribute: A Study in Ontology, Dordrecht: Reidel. Loux, M. (2002) Metaphysics: A Contemporary Introduction, 2nd ed., New York, NY: Routledge. Lowe, E. J. (1989) Kinds of Being: A Study of Individuation, Identity and the Logic of Sortal Terms, Oxford: Blackwell. Lowe, E. J. (1994) “Primitive Substances,” Philosophy and Phenomenological Research 54, 531–552. Lowe, E. J. (1998) “Form Without Matter,” Ratio 11, 214–234. Lowe, E. J. (2000) “Locke, Martin, and Substance,” Philosophical Quarterly 50, 499–514. Lowe, E. J. (2001) “Dispositions and Laws,” Metaphysica 2, 5–23. Lowe, E. J. (2006) The Four-Category Ontology: A Metaphysical Foundation for Natural Science, Oxford: Oxford University Press. Lowe, E. J. (2012) “A Neo-Aristotelian Substance Ontology: Neither Relational Nor Constituent”, in T. Tahko (ed.) Contemporary Aristotelian Metaphysics, Cambridge: Cambridge University Press, 229–248. Lowe, E. J. (2013) “Neo-Aristotelian Metaphysics: A Brief Exposition and Defense,” in E. Feser (ed.) Aristotle on Method and Metaphysics, New York, NY: Palgrave Macmillan, 196–205. Macdonald, C. (2005) Varieties of Things: Foundations of Contemporary Metaphysics, Oxford: Blackwell. Martin, C. B. (1980) “Substance Substantiated,” Australasian Journal of Philosophy 58, 3–10. Martin, C. B., and Heil, J. (1999) “The Ontological Turn,” Midwest Studies in Philosophy 23, 34–60. Maudlin, T. (2007) The Metaphysics Within Physics, Oxford: Oxford University Press. Molnar, G. (2003) Powers: A Study in Metaphysics, Oxford: Oxford University Press. Mumford, S. (1998) Dispositions, Oxford: Oxford University Press. Mumford, S., and Anjum, R. L. (2011) Getting Causes from Powers, Oxford: Oxford University Press. Oderberg, D. (2007) Real Essentialism, New York, NY: Routledge. Oderberg, D. (2011) “Essence and Properties,” Erkenntnis 75, 85–111. Shoemaker, S. (1980) “Causality and Properties,” in P. van Inwagen (ed.) Time and Cause, Dordrecht: Reidel, reprinted in J. Kim and E. Sosa (1999) Metaphysics: An Anthology, Oxford: Blackwell, 253–68. Strawson, G. (2008) “The Identity of the Categorical and the Dispositional,” Analysis 68, 271–282. Tweedale, M. (1984) “Armstrong on Determinable and Substantival Universals,” in R. J. Bogdan (ed.) D.M. Armstrong, Dordrecht: Reidel, 171–189. Walsh, D. (2006) “Evolutionary Essentialism,” British Journal for the Philosophy of Science 57, 425–448. Wiggins, D. (2001) Sameness and Substance Renewed, Cambridge: Cambridge University Press.

193

13 DISPOSITIONAL ESSENTIALISM Ka Ho Lam

13.1

Introduction

Dispositional essentialism is the view that at least some properties have dispositional essences, i.e., these properties essentially endow their bearers with certain dispositions.1 Roughly speaking, a disposition is a tendency to manifest various characteristic effects in the appropriate circumstances.2 For example, a vase, being fragile, has the disposition to shatter when struck. Just as a vase, no matter how fragile it is, may never shatter, an object may possess dispositions that never manifest. In this regard, dispositions are modal in nature. Rather than standing for any actual effect, they point toward what could or would obtain (Jacobs 2011: 83; Vetter 2015: 33). Dispositional essentialists are particularly interested in fundamental properties, such as charge, mass, and spin, for these properties are supposed to participate in the laws of nature. According to them, the fact that these properties are defined in dispositional terms should be taken as prima facie evidence of dispositional essentialism (Bird 2007: 7; Molnar 1999: 13; Mumford 2006: 476).3 For instance, it seems that the property charge is exhausted by its dispositional characters, such as repelling like charge and accelerating in a certain way when placed in a forced field.4 Apparently, not all properties in the laws of nature are genuinely dispositional, e.g., spatiotemporal relations, orientations, and locations. It is hard to see what powers these properties essentially endow their bearers with. If these properties are not essentially dispositional, then they cannot be defined by what they dispose their bearers to do, for any dispositional character that happens to accompany them can only be accidental.5 Thus, unlike dispositional properties, these categorical properties are not, in and of themselves, modal. As Barbara Vetter indicates, “[b]eing modal sets dispositions apart from categorical properties (if there are any)” (Vetter 2015: 33).6 And categoricalism, endorsed by philosophers such as David Lewis and David Armstrong, denies dispositional essentialism by maintaining that all (fundamental) properties are purely categorical, i.e., they do not possess any dispositional character essentially. There are different versions of dispositional essentialism. The most radical one, standing in direct opposition to categoricalism, is dispositional monism. Advocated by Alexander Bird (2007) and Stephen Mumford (2004), it claims that all fundamental properties are purely dispositional. Like many debates in philosophy, these two polar positions have their own pros

194

DOI: 10.4324/9781003008750-16

Dispositional Essentialism

and cons, and there is no consensus as to which is correct. A way to break the stalemate between them is to formulate an intermediate position that retains their strengths and avoids their weaknesses. Here we have two options. On the one hand, as Brian Ellis (2012) and George Molnar (2003) suggest, one should adopt a dualist position that posits both categorical properties and dispositional properties on the fundamental level. On the other hand, one can opt for a hybrid position, according to which fundamental properties are powerful qualities, i.e., they are comprised of two ontologically distinct parts, namely a dispositional (or powerful) part and a categorical (or qualitative) part. In other words, the essence of a fundamental property is not exhausted by its dispositional characters—it also consists of categorical characters. However, this proposal, which has attracted increasing attention in recent literature, appears to be incoherent, given that the dispositional is supposed to be modal and the categorical is supposed to be non-modal. Therefore, as I am going to show, a crucial issue of the hybrid position is to revise our conceptions of the categorical and the dispositional (Engelhard 2010; Giannotti 2019; Giannotti 2021; Ingthorsson 2013; Jacobs 2011; Taylor 2013; Taylor 2018; Williams 2019).7 Clearly, in assuming that (some) fundamental properties are essentially dispositional, both the dualist position and the hybrid position commit to dispositional essentialism. In the following, I am going to illustrate how challenges from categoricalism as well as modern physics motivate dispositional essentialists to revise the assumption that the dispositional is modal whereas the categorical is non-modal. I will begin by surveying two central debates between categoricalism and dispositional monism. The first one is concerned with the identities of properties (Section 13.2), and the second one is concerned with the laws of nature (Section 13.3). Next, I will examine how the dualist position may resolve the problems faced by dispositional monism and categoricalism in these two debates, as well as indicate the difficulties it may run into. Then, I will explain why, like the dualist position, the hybrid position also encounters a problem in attempting to explicate how the categorical and the dispositional are connected. I will look at how this problem requires hybrid theorists to reconceive the distinction between the categorical and the dispositional (Section 13.4). Finally, I will move on to some of the challenges from modern physics that dispositional essentialists should pay attention to in revising the categorical-dispositional distinction.

13.2 The Identities of Properties A central complaint about categoricalism is that it makes properties unknowable. As I have already mentioned, dispositional essentialists believe that properties possessing dispositional essences are identified by what they dispose their bearers to do. On the contrary, categoricalists contend that since properties do not possess any dispositional character essentially, their identities cannot be so determined. Consider the property charge again. If categoricalism is true, it means that none of its dispositional characteristics, be it repelling like charges or accelerating in a forced field, is essential to it. Stripping it of all its dispositional characteristics, what remains for its identity is simply its “thisness” and “not-thatness” (Williams 2019: 58). Hence, the properties of a thing, according to categoricalists, “tell us nothing whatever about how it will behave” (Ellis & Lierse 1994: 28). Many philosophers therefore argue that categoricalism entails quidditism, the view that properties possess primitive, non-qualitative identities. This renders categoricalism an unattractive position. Critics believe that, in postulating that properties are individuated by their non-qualitative, intrinsic aspects, namely quiddities, categoricalism is “guilty of supposing

195

Ka Ho Lam

that ‘there are more things in heaven and earth than physics has dreamt of’” (Vetter 2015: 8; see also Kimpton-Nye 2021: 3422), as quiddities do not seem to play any explanatory role in physics. Quidditism has undesirable epistemological repercussions. Given that, qua quiddities, categorical properties are by nature inert and cannot interact with us in any way, they are cut off of our cognitive reach (Marmodoro 2020: 58). So, assuming that all properties are categorical, it would be possible for them to be swapped around in some systematic way in another possible world, without objects in that world behaving any differently from the way they do in our world (Armstrong 2012: 28–9; Bird 2007: 74–5). As the causal powers of different categorical properties are not distinctive of them, we basically have no way to distinguish between different categorical properties (Ellis 2012: 19). Hence, in another possible world, the property mass could perform the role of the property charge in our world, and vice versa, without our knowing it.8 This seems to suggest that we can never know for certain the identity of any property, which is by no means good news for many philosophers.9 In contrast, dispositional monism avoids this problem by maintaining the properties, qua powers, have their identity assigned by their causal roles. Given that a property’s identity goes hand in hand with the way it disposes its bearer to various manifestations, it cannot dispose its bearer differently without ceasing to be the property it is (Bird 2007; Shoemaker 1980). However, a closer look suggests that dispositional monism may not fare any better than categoricalism in fixing the identity of property. The promise that we can come to know what a property is by knowing what manifestation that property orients its bearer toward may turn into a vicious regress (Bird 2007: 101). The regress arises because we recognize dispositions by their manifestations, or more accurately, by the properties these manifestations involve. Yet, since these properties, according to dispositional monism, are also dispositions, they can only be recognized by their own manifestations, which are also dispositions, and so on. If, as dispositional monists maintain, dispositional properties were the only properties there are, then at no stage of the regress would we encounter any property of which recognition does not point toward a further manifestation. In short, while there is no effect knowing which will reliably inform us about a property’s identity in the case of categoricalism, there are infinitely many effects to know in the case of dispositional monism. In fact, the regress is not only epistemological, but also ontological. As we have seen, the nature of a disposition is fixed relationally rather than intrinsically, i.e., by the manifestations it is a disposition for (Tugby 2012: 725). Thus, the identity of a dispositional property is constituted by its relations to other dispositional properties, i.e., by the effects it diposes its bearer to manifest in appropriate circumstances. Yet, the effects in question are supposed to be dispositional as well, hence their identities are no less constituted by their relations to other dispositional properties. The universe of dispositional monism is thus “an infinite series of properties that have no nature on their own except for a connection to something else that is equally lacking in any independent nature” (Ingthorsson 2013: 61; see also Jacobs 2011: 87; Lowe 2006: 138).10 Of course, one may attempt to disprove this unending regress by asserting that it is unlikely for there to be infinitely many fundamental properties. Be that as it may, it only means that there is a vicious circle somewhere (Bird 2007: 132; Lowe 2006: 138). Dispositional monists believe that the regress, be it epistemological or ontological, is not fatal. They claim that the identity of each property can be settled holistically by its relations to other properties within the whole structure of properties (Bird 2007; Mumford 2004). According to Bird, if this structure of properties fulfills certain graph-theoretic requirements, we can uniquely determine the position, hence the identity, of each property in the structure.11 Consequently, for dispositional monists, the identity of a property is constituted by its

196

Dispositional Essentialism

relations to other properties (Bird 2007; Bird 2012; Mumford 2004; Tugby 2013; Tugby 2017; Williams 2019).

13.3

The Laws of Nature

Another major disagreement between dispositional monism and categoricalism revolves around how we should comprehend the lawhood of the laws of nature. Roughly speaking, laws of nature state the causal relations between properties. For instance, Coulomb’s law states the relation between various properties, namely point charges, the displacement between these charges, and the electrostatic force generated. In other words, properties participate in causal laws. Given that dispositional monists and categoricalists conceive the nature of properties differently, they therefore understand laws of nature differently. As we have seen, according to dispositional monists, given that fundamental properties are dispositional in nature, the relations they stand in to each other constitute their essences. They go on to assert that causal laws fall out as consequences of these relations between fundamental properties (Bird 2012: 36). So it is a law that like-charged particles repel each other because the property charge essentially disposes its bearers to this effect. Laws are distinguished from accidental regularities as only the former, but not the latter, spring from the dispositional essences of fundamental properties. Hence, properties that possess dispositional essences do not just participate in laws—they actually ground those laws (Vetter 2012: 201). And causal laws so understood are metaphysically necessary. Since an essentially dispositional property cannot be the property it is without engaging in the same relations to other properties, the same set of laws will hold in another possible world so long as the properties that ground these laws exist in that world. How do categoricalists account for the laws of nature, if they deny that properties have dispositional essences? Here we find two radically different models, namely the Humean regularity view (Lewis 1973; Lewis 1986; Lewis 1994, also Beebee 2000; Earman & Roberts 2005a; Earman & Roberts 2005b) and nomic necessitarianism (Armstrong 1983; Dretske 1977; Tooley 1977). The former model suggests that causal laws are just regularities or patterns of things possessing properties. Laws do not govern how these properties, which are categorical in nature, interact. Instead, they merely supervene on the successions of events or facts. In contrast, the latter model suggests that laws of nature are second-order relations of properties. Unlike the regularity view, here laws of nature have a governing function. They impose orders on the interactions of properties. Nevertheless, laws of nature in both models are contingent in the sense that it is possible for an alternative set of laws to obtain in another world where exactly the same set of properties exists. This is because, in principle, properties themselves, devoid of any dispositional essence, do not prevent different patterns of things or different second-order relations from being implemented in a different world. A common criticism of the regularity view is that the Humean laws it depicts “fail to perform the explanatory task that is central to lawhood” (Vetter 2012: 204; also Armstrong 1983: 40; Bird 2007: 86). First of all, laws are supposed to explain their instances. For example, we expect to explain why some particular white substance dissolves in some colorless, odorless liquid by reference to the law that sodium chloride dissolves in water. Yet, understood as nothing but generalizations about the “vast mosaic of local matter of particular facts” (Lewis 1986: ix), Humean laws fail to meet this expectation. Since laws so conceived “are what they are in virtue of the Mosaic rather than vice versa” (Maudlin 2007: 72), we cannot appeal to them to explain the particular instances of the mosaic or why the relevant generalizations are true (Lange 2013: 257–8).

197

Ka Ho Lam

Of course, for Lewis, not all true generalizations are laws. As suggested by his “best systems account”, contingent generalizations are laws if and only if they are expressed as theorems or axioms in every true deductive system that achieves the best combination of simplicity and strength (Lewis 1973). Nevertheless, generalizations satisfying this “metaphysically innocuous” criterion do not thereby become something “over and above” the regularities they represent. So, if it is a law that all Fs are Gs, then the corresponding Humean law can at best be the universal generalization ∀x(Fx→Gx). Yet, it states nothing more than the regularity that all Fs are G. What is missing is an “element of necessitation” that explains the regularity in question, namely that all Fs have to be G (Vetter 2012: 206). What about nomic necessitarianism? As Armstrong indicates, what categoricalists need is “a sort of intermediate necessity for laws between Humeanism and the necessary connection favored by power theorists” (Armstrong 2012: 31). In order to avoid the shortcoming of the regularity view, such a sui generis notion of necessity must be robust enough for laws to explain their instances. Consequently, laws of nature are more than just universal generalizations such as ∀x(Fx→Gx). Moreover, this nomic necessity should not be as strong as the metaphysical necessity that characterizes laws of nature under dispositional monism. Given that categorical properties lack dispositional essences, not only the relations between them will not hold in all possible worlds, but also, the modal force of these relations must come from a source external to these properties. According to Armstrong, laws should thus be understood as second-order relations between properties, namely N (F, G). Here N stands for a law-making, second-order universal that governs the relation between the properties F and G, i.e., two first-order universals. Bird objects that this intermediate notion of necessity is untenable as it leads to a vicious regress. Presumably, the second-order universal N is responsible for the modal force of the regularity (R) between the properties F and G, i.e., ∀x(Fx→Gx). This is because F and G, being categorical, are not modal themselves. So what about the modal force of the connection between N and R? Clearly N does not just supervene on R, for this is exactly the regularity view that nomic necessitarianism finds unsatisfactory. Yet, N does not entail R either. For if it does, it means that in any world where N, F, and G exist, the law N (F, G) will hold. Yet, being a categorical property, N is not supposed to support any metaphysically necessary law, as it lacks the intrinsic modal force that necessarily disposes itself toward R. So, to ensure that N will sustain R, what we need is the nomic necessity that N is supposed to embody. Again, such nomic necessity has to come from a higher order universal linking N and R, namely N’. What results is a third-order relation N’ (N, R). We will then face the same question concerning the connection between N’ and the relation R’ (i.e., the relation between N and R). So we run into a vicious regress.12 The above considerations suggest that dispositional monism has two advantages over categoricalism. First, in maintaining that laws of nature spring from the dispositional essences of fundamental properties, it accounts for their lawhood. Second, it secures a notion of necessity that allows the laws of nature to fulfill their explanatory role (Bird 2007: 205, Ellis 2001: 219).13 Nevertheless, dispositional monism faces the challenge of how to accommodate properties that do not appear to be dispositional in the laws of nature, such as spatial displacement, orientations, location and spatiotemporal relations. For instance, Coulomb’s law states the relation between electrostatic force, point charges and the displacement between them. But displacement does not appear to be dispositional: for what effects would a given spatial displacement between two entities dispose them to exhibit? There are two options for dispositional monists. They can either dismiss these properties as non-fundamental or explicate them as dispositional. In his discussion of temporal relation and

198

Dispositional Essentialism

spatial displacement, Bird opts for the latter option. As he indicates, according to general relativity, the mass of each object is understood as its power to change the curvature of spacetime. Thus, by the action-reaction principle, according to which a potential effect should also be a potential cause (Brown & Lehmkuhl 2017), space and time should be treated as causes of change that possess dispositional essences, rather than unchanging backgrounds that are not involved in any causal interaction, as assumed by Newtonian mechanics (Bird 2007: 166; Bird 2012: 39). We will take a closer look at whether dispositional monism is compatible with general relativity in Section 13.6. Now let us first turn to the two different intermediate positions between dispositional monism and categoricalism.

13.4

Property Dualism

As I have shown in the last two sections, while categoricalism is guilty of quidditism, dispositional monism has difficulty in fixing the identities of properties as it is likely to generate an infinite regress or vicious circle. Likewise, although dispositional monism surpasses categoricalism in explicating the laws of nature, it does not readily accommodate properties that do not appear to be dispositional. Property dualism, in acknowledging both categorical properties and dispositional properties on the fundamental level, can be viewed as an attempt to combine the benefits of both types of property in order to eschew the different problems faced respectively by dispositional monism and categoricalism. At first glance, property dualism is able to halt the regress dispositional monism runs into and fix the identities of dispositional properties by appealing to categorical properties. Given that the identities of categorical properties do not depend on their relations to other properties, the buck will stop with them (Bird 2007: 138). Yet, a possible drawback is that these categorical properties are equally culpable of quidditism. As Bird indicates, quidditism is an issue not just for categoricalism, but for categorical properties in general: “[i]f one takes quidditism to be problematic, then that infects the conception of any property as categorical” (Bird 2012: 38, emphasis in the original). So it seems that if we reject categoricalism because of quidditism, then we should reject property dualism too, given that it also postulates pure categorical properties in its ontology. Recall that quidditism is problematic because quiddities are basically unknowable to us. As a rejoinder to Bird’s criticism, both Ellis and Molnar argue that, although categorical properties do not confer powers upon their bearers like dispositional properties do, they are “factors” or “dimensions” that are sensitive to powers, i.e., they are respects in which things can change in causal interactions. While powers act to change their values, they also delimit the sorts of manifestations produced by powers (Molnar 2003: 163; Ellis 2001: 9–10).14 We can therefore learn about the identities of these categorical properties indirectly through the abilities and actions of the relevant causal powers they are associated with. So, on the one hand, in reserving an active causal role for dispositional properties, property dualism is able to account for the lawhood of the laws of nature in terms of dispositional essences.15 On the other hand, in assigning a passive causal role to categorical properties, property dualism manages to accommodate non-dispositional properties in the laws of nature without succumbing to quidditism. However, this attempt to fend off the charge of quidditism may lead to a dilemma. As Neil Williams argues, unless they have certain causal roles to play under property dualism, otherwise purely categorical properties would simply be “ghostly” causal danglers (Williams 2019: 106). Nevertheless, if causal roles are the only roles they play, then property dualism may collapse into dispositional monism. Recall that according to dispositional monism,

199

Ka Ho Lam

relations between fundamental properties are necessary due to their dispositional essences. As Williams indicates, if categorical properties have only causal roles to play and are also connected to dispositional properties necessarily, then they will be indistinguishable from dispositional properties. In other words, categorical properties will become no less modal than dispositional properties. But if their connections to dispositional properties are accidental, then we have to admit two types of laws, one necessary and one contingent. Williams believes that this is not parsimonious (Williams 2019: 110).16 In view of such dilemma, Williams contends that we should give up property dualism and adopt the hybrid position.

13.5

The Hybrid Position

Rather than positing two different types of properties, the hybrid position holds that fundamental properties are both categorical and dispositional. While admitting that fundamental properties possess dispositional essences, the hybrid position maintains at the same time that the dispositional essence of a fundamental property does not exhaust its essence—the full essence of a fundamental property also consists of a categorical part. At first glance, this proposal looks dubious. For if the dispositional is modal and the categorical is non-modal, this means that the essence of a property is composed of two radically different, if not incoherent, parts. Indeed, as we will see, proponents of the hybrid position seldom adhere to the traditional understanding that the categorical is outrightly non-modal, in opposition to the dispositional. Before we take a closer look at how its proponents construe the categorical, let us first examine the motivations for the hybrid position. Understanding what challenges the hybrid position attempts to tackle will offer us a better view of why a revision of the notion of categoricality is imminent under its agenda. One of the arguments in support of the hybrid position is that pure powers fail to account for mental qualia or the phenomenological character of our experience. As Jonathan Jacobs argues, mental qualia are not purely relational, for if they are, it means that everything is “zombie-like, disposed to act in certain ways but empty on the inside, empty of all qualitative nature” (Jacobs 2011: 87). Thus, unlike the dualist position, the central goal for the hybrid position to introduce a categorical aspect into the world of powers is neither to solve the regress problem related to the identities of dispositional properties nor to accommodate the non-dispositional factors in the laws of nature. As we have seen, according to Williams, the main problem of property dualism stems from the fact that the categorical properties must be assigned a causal role lest they become “ghostly” danglers. On the contrary, in the hybrid account, the categorical part of a property’s essence is supposed to account for the qualitative or phenomenal characters of the world rather than causal interactions (Jacobs 2011: 87; Williams 2019: 100).17 As Williams maintains, a property is categorical in the sense that it is not powerful. Nonetheless, it is also not merely a quiddity (Williams 2019: 102). So the categorical qualities, being causally irrelevant, would not become part of the causal/dispositional structure of the world. And although the categorical qualities are not causally operative, the properties they are part of will not become “ghostly” danglers. This is because these properties are not purely categorical. Given that they are dispositional at the same time, they remain causally powerful.18 Again, the crucial question is: are the categorical part and the dispositional part of a property’s essence connected contingently, or necessarily? As Armstrong alleges, the connection cannot be contingent. For if it is, then it means that the categorical part could have different powers or even no power at all (Armstrong 1997: 84). Molnar and Mumford

200

Dispositional Essentialism

raise a similar concern: how is it possible for the two parts of one and the same property to vary independently of each other? For if this could be the case, then the two different parts can actually exist separately. And this sounds like a return to property dualism (Molnar, 2003: 150; Mumford 2007: 85). Indeed, as Williams admits, the categorical part of a property’s essence is connected to its dispositional part necessarily in a hybrid account. So, the identity of a property is determined “twice over” given that each property type is defined both holistically by its unique position in the power constellation or structure with other powers and individually by its qualitative character that does not involve in casual interactions (Williams 2019: 118). So no two properties that have the same causal powers would differ in their qualitative characters. If they differ in their dispositional characters, then, necessarily, they would not have the same categorical character (Tugby 2012: 178). As we have seen in the case of property dualism, the connection between the dispositional properties and the categorical properties cannot be necessary (nor can it be contingent). This is because, when the categorical properties only have causal roles to play, they will be assimilated by the dispositional structure and become indistinguishable from the dispositional properties. The hybrid position avoids this problem because the categorical aspect of a property is not causally relevant. However, in assuming that the dispositional part and the categorical part of a property are connected necessarily, the hybrid position will conflict with the initial assumption that the categorical, unlike the dispositional, be non-modal. Here is how the conflict arises. According to the hybrid position, the dispositional characters of the fundamental properties form a structure that covers all causal relations in the world. Nonetheless, their categorical characters, despite being causally irrelevant, form a separate isomorphic structure at the same time, since they are connected to the dispositional characters in a one-to-one manner. And the relations between the categorical characters in this categorical structure are necessary. This is because, on the one hand, the categorical characters are connected to the corresponding dispositional characters necessarily, and on the other hand, the dispositional characters in the dispositional structure are connected to one another necessarily. Thus, in a world where the same set of properties exist, the same dispositional structure and the same categorical structure will obtain, and the two structures are isomorphic. Moreover, the relations in both structures are metaphysically necessary. As a result, the categorical becomes no less modal than the dispositional. This contradicts the common assumption that the categorical, unlike the dispositional, has “no essential or other non-trivial modal character” (Bird 2007: 67).

13.6 The Challenges from Modern Physics One way to defend the hybrid position against the problem I have just presented is to revise our conception of the categorical. Traditionally, the categorical is usually understood negatively as non-dispositional (Bird 2007: 66), for it seems difficult to come up with an informative and coherent notion of it (Ingthorsson 2013: 57). And as we have seen, many philosophers in the debate, whether they are dispositional monists, categoricalists, or property dualists, presuppose that the dispositional and the categorical are mutually exclusive, in the sense that the former is modal in nature while the latter is not (Armstrong 1997; Armstrong 2005; Bird 2007; Ellis & Lierse 1994; Molnar, 2003; Mumford 1998). In contrast, as I have mentioned earlier, hybrid theorists often do not share this assumption. Recently, many of them have attempted to provide a more substantial characterization of the categorical. We have already seen that the categorical characters are identified as the

201

Ka Ho Lam

intrinsically qualitative features rather than just quiddities in Williams’ account (also Jacobs 2011; Smith 2016; Taylor 2018). Many dispositional essentialists have also questioned the traditional, modal reading of the categorical-dispositional distinction.19 For instance, by comparing how the categorical is differentiated from the dispositional in dispositional monism, categoricalism and the hybrid position respectively, Lorenzo Azzano (2021) argues that having a dispositional essence does not necessarily make a property dispositional. Likewise, according to Joaquim Giannotti (2019; 2021), the fact that a property is qualitative does not forbid it from being necessarily dispositional. And by appealing to the notions of metaphysical grounding and essential dependence, Matthew Tugby (2020) contends that properties are qualitative quiddities, and their qualitative natures account for the dispositional characters they possess. Given the scope of this chapter, I am not able to examine these proposals here. Yet, what is at stake is not only the conception of the categorical, but also the conception of the dispositional. To conclude this chapter, let me point out some of the challenges from modern physics dispositional essentialists face in relation to the latter. As I have mentioned at the very beginning, fundamental physics provides us with prima facie evidence for endorsing dispositional essentialism. We have also seen how Bird appeals to general relativity in trying to explain away the impression that space and time are categorical. Ellis and Lierse also argue that scientific evidence appears to be decisive against categoricalism, given that “with few exceptions, the most fundamental properties that we know about are all dispositional” (Ellis & Lierse 1994: 32). Nonetheless, there are complaints alleging that dispositional essentialism actually fails to keep up with the developments in science. Dispositional essentialists should pay attention to these complaints when attempting to modify the conception of the dispositional. One such complaint comes from Steven French. As he indicates, dispositional essentialism, in emphasizing the causal role of dispositional properties, is likely to run into trouble when extending to the quantum realm, “where attributions of causation are famously problematic” (French 2020: 193). More precisely, French argues that the nature of the fundamental physical magnitudes is not exhausted by their dispositional characteristics, which are understood exclusively in terms of their causal roles. A case in point is the conservation of charge. While it is a law that charge is conserved, it is not correct to say that charge bestows the disposition to conserve itself (French 2020: 194; Yates 2013: 119). Such conservation of properties appears to be related to certain fundamental symmetry via Noether’s Theorem, which are regarded as constraints, or meta-laws in quantum physics (French 2020: 199). If, as it is argued, much of the dynamical behavior associated with charge can be attributed to this symmetry in the quantum context, then it means that there are aspects of this property—in particular, the ways it features in the relevant laws—that a dispositional essentialist account fails to capture (French 2020: 194). Vassilios Livanios (2018) raises a similar concern regarding whether dispositional essentialism can accommodate the kind of metaphysical contingency presupposed by Hamilton’s principle of least action, as noticed by Joel Katzav (2004; 2005).20 Based on these scientific findings, French argues that we should do away with dispositional properties and construe the fundamental level of reality as being governed by laws and structures, from which powers emerge, rather than the other way around (French 2014: 292). In contrast, as a preliminary proposal, Bird suggests that we should regard principles such as conservation and symmetry as “pseudo-laws” that will eventually be disposed of as science advances. For bird, these principles appear to be features of our theoretical background structures that should be eliminated in our physical theories rather than features of the world

202

Dispositional Essentialism

that needed to be accommodated within our metaphysics. (Bird 2007: 213–4). More specifically, in replying to the challenge of Noether’s theorem, David Yates (2013) suggests that, by replacing the traditional modal conception of essence with Kit Fine’s primitivist conception of essence, we can formulate a more liberal notion of causal role for properties that is not confined to their dispositional characters.21 And this will allow us to derive the principle of charge conservation. Likewise, philosophers like Vetter (2015) and Kistler (2020) attempt to provide a more encompassing account of dispositional essentialism by replacing the traditional understanding of dispositions with a more flexible notion of powers or potencies. To conclude, many of the debates between dispositional essentialists and categoricalists hinge on the assumption that the dispositional is modal and the categorical is non-modal. Of particular interest are the debates revolving around the identities of properties and the laws of nature. The hybrid position, in postulating that fundamental properties consist of both a dispositional aspect and a categorical aspect, promises a way out of the stalemate between dispositional monism and categoricalism in these debates. However, as I have shown, the hybrid position is not consistent with the assumption that the categorical is non-modal. Furthermore, new findings in quantum mechanics also pose challenges to the framework of dispositional essentialism. Thus, in order to serve as a viable intermediate position between dispositional monism and categoricalism, the hybrid position, in addition to adopting a more viable reading of the categorical, must also take into consideration these new scientific findings and revise the conception of the dispositional accordingly.

13.7

Related Topics

“Scientific Essentialism” “Natural Kind Essentialism” “Laws and Explanation”

Notes 1 According to common usage, the term “dispositional essentialism” denotes specifically the position that certain properties have dispositional characters essentially. I will therefore put aside the broader essentialist view that certain objects possess dispositions essentially. As the latter will constitute another debate. 2 I will employ the terms “dispositions”, “powers” and “potentialities” interchangeably. Some theorists assign them different meanings. For example, Vetter (2015) takes dispositions to be forming a subset of potentialities, which include also abilities and powers. She thinks that powers or potentials are weaker than dispositions, given that having a certain disposition entails having a certain power or potential, yet the reverse is not always the case. In contrast, Kistler (2020) thinks that dispositions belong to a semantic category, while powers, which are supposed to feature in the truthmakers of dispositional attributions, belong to an ontological category. 3 Whether the terminological usage dictates the ontological fact is of course, a controversial issue. Those who hold a negative view include: Barker (2013: 608), French (2014: 242) and Williams (2019: 14). 4 While this chapter focuses on the examples of fundamental physics, it should be noted that debates about dispositional essentialism are also motivated by questions in other fields of science and areas of philosophy. For instance, Hüttemann & Kaiser (2018) explore what biological dispositions are and how the notion of disposition sheds light on the explanatory practice in biology (see also Austin 2017). Likewise, within the dispositional essentialist framework she develops, Vetter attempts to elucidate human abilities in terms of dispositions, in order to answer various questions concerning human agency ( Vetter 2019, see also Vetter & Jaster 2017). As Dumsday (2019) points out,

203

Ka Ho Lam

5 6

7

8

9

10 11

12

13

dispositional essentialists are engaged in two different projects. On the one hand, the project of defending their positions against different rival theories; on the other hand, the project of investigating how the internal development of dispositional essentialism is connected to other topics in metaphysics, such as material composition and persistence. The current chapter focuses on the former project. The discussion concerning whether or not these properties are dispositional can be traced back to the exchange between Mellor (1974; 1982) and Prior (1982). As it will become clear, the assumption that categorical properties are non-modal is debatable. For the sake of the argument, I will confine this assumption to the fundamental properties of concrete objects. This is because the properties of abstract objects may constitute an immediate counterexample to this assumption. Suppose one is a structuralist about mathematical objects and maintains that natural numbers are defined by their positions in the structure of the number line. Consequently, being a natural number entails that it stands in a necessary relation with other natural numbers. For example, being the number one is understood to stand in a necessary relation (i.e., being the successor of) with being the number zero. However, the properties being the number zero and being the number one appear to be categorical. I am indebted to Robert Michels for this counter-example. The powerful qualities view should be carefully distinguished from another hybrid view defended by Heil (2003), Kistler (2020), Martin (1993; 1997; 2008), Martin & Heil (1999), Mumford (1998), Strawson (2008), Schroer (2010; 2013). According to these philosophers, the distinction between the categorical and the dispositional is merely conceptual, not ontological. And whether a property is categorical or dispositional depends on our theoretical perspectives and instrumental concerns. As Azzano points out, supporters of such view “would have no problem admitting that agreements and disagreements on matters of dispositionality and categoricity are merely verbal, or, at least, they do not pick up any genuine difference in one’s background metaphysics” ( Azzano 2021: 2954). However, if the distinction is only conceptual, it is unclear to what extent this hybrid view is realist about dispositionality or commits to dispositional essentialism. Apart from permutation, two similar worries about categoricalism are the possibility of replacement and the possibility of duplication. The former suggests that categoricalism allows a property in the actual world, e.g., charge, to be replaced by an alien property, e.g., schmarge, in another possible world. The latter implies that categoricalism permits two properties in another possible world to play the role of charge in the actual world ( Yates 2013: 96). As we will see when we move on to the debate about the laws of nature, whether one finds the permutation of powers counter-intuitive largely depends on how one conceives the causal roles of properties. Indeed, many categoricalists are willing to bite the bullet here. For example, Lewis (2009) contends that the permutation argument exactly supports his view of “Ramseyan Humility”, i.e., that we are irredeemably ignorant of the identities of the fundamental properties. Likewise, Armstrong (2012: 29) argues that the permutation argument presents merely a possibility that should not be taken too seriously by metaphysicians, for our actual world is the only world there is. For example, Chakravartty believes that “[r]egresses of this kind are commonly short-lived, since causal chains originating with the property instances one attributes are connected, in cases where one justifiably claims knowledge of them, to one’s sensory modalities” ( Chakravartty 2007: 136). Bird argues that this requires there to be an appropriate asymmetry in the total set of relations in the dispositional structure. For a detailed illustration of the technicalities involved, see Bird (2007: 138–46). For an amended version of the structuralist solution to the regress problem, see Bigaj (2010). Barker & Smart (2012) contend that Bird’s own dispositional essentialist account is susceptible to the same charge of vicious regress he himself levels against nomic necessitarianism. Yet, according to Coates (2021: 916–7), Barker and Smart’s argument fails. This is because Bird’s original argument is concerned with the question of validation, namely whether the Armstrongian laws can explain regularities, but Barker and Smart’s argument is concerned with the question of explanation, namely whether the connection between the Armstrongian laws and the regularities can be explained. Kimpton-Nye (2021: 3432) suggests that dispositional essentialism may be guilty of explanatory circularity in its attempt to provide a grounding explanation for the laws of nature by appealing to dispositional essences. This is because, according to dispositional monism, while the relations a dispositional property stands in with other properties constitute its essence (the explanans), these relations are understood as the laws of nature (the explanandum) at the same time. For example, as Kimpton-Nye (2021: 3430) points out, dispositional essentialists, in response to the question “why

204

Dispositional Essentialism

14

15

16

17

18

19

20 21

do charged objects interact in accordance with Coulomb’s law?” will probably answer something like, “because Coulomb’s law is part of the essence of the property charge”. And such an answer does not appear to be more satisfying than explaining opium’s ability to make one sleeps in terms of its dormitive power. As an illustration, Ellis suggests that living a long distance from Sydney is a factor in determining whether one can walk there, though the distance is neither a physical causal power nor anything that has such power ( Ellis 2012: 20). Hence, being “dimensions” or “factors” of causal powers, categorical properties determine what manifestations are certain, possible, or impossible for a given disposition. It should be noted that the position of property dualism I sketch here is slightly different from Ellis’ own “new essentialism”, despite the fact that Ellis is one of the strongest advocates of the dualist position. According to Ellis’ new essentialism, laws are ontologically dependent on natural kinds (and not vice versa). Therefore, the laws of nature do not derive from the dispositional essences of powerful properties, but directly from the essences of natural kinds. For Ellis’ new essentialism, see Chapter 12, “Scientific Essentialism” (this volume). For the relation between natural kind essentialism and laws of nature, see Chapter 10, “Natural Kind Essentialism” (this volume) and Chapter 20, “Laws and Explanation” (this volume). It is unclear to what extent this constitutes a fatal problem to the dualist position. As Tuomas Tahko (2015) contends, physics provides us with genuine examples of both necessary and contingent laws (strictly speaking, Tahko’s view is not a mixture of dispositional monism and categoricalism as he assumes that metaphysically necessary laws emerge from the essences of fundamental natural kinds, rather than from the essences of dispositional properties). Michele Paoletti (2020) also lists different sources for the contingent status of laws in a purely dispositional world. Indeed, intuitions differ as to whether laws of nature are necessary or contingent. While Bird thinks that it is “an appealing and widely held” view that laws are in some way necessary ( Bird 2012: 37), it is argued that laws could have been different from what they are ( Beebee, 2000; Kistler 2020; Lewis 1986; Sidelle 2002). Thus an account that admits both necessary and contingent laws at least has the advantage of accommodating both intuitions. For an account that posits both metaphysically necessary and contingent laws based on the kinematical/dynamical distinction used in physical theorizing, see Hirèche et al. (2021). Another motivation for postulating the categorical aspect in addition to the dispositional aspect is to tackle what Marmodoro calls the “ghostliness” of power ( Marmodoro 2020: 60). It is often criticized that an ontology of nothing but pure powers lacks enough actuality. As Armstrong contends, in such an ontology, “particulars would seem to be always re-packing their bags as they change their properties, yet never taking a journey from potency to act” ( Armstrong 1997: 80). Another way to put the question is: what do pure dispositions do when not manifesting (Psillos 2006)? According to many hybrid theorists, e.g., Tugby, if properties are powerful qualities, then their categorical (or qualitative) aspects, which are not just potentialities, can serve as the ontological grounds for their continued existence when their dispositional aspects are not manifesting. Yet, Marmodoro suggests that dispositional monists, without positing the categorical aspect, can answer this challenge by distinguishing between occurrent activity and potential activity, since the former is sufficient for actuality ( Marmodoro 2020: 60). Of course, the question remains as to how the hybrid position accommodates factors in scientific laws that do not appear to be dispositional, for example, locations and distances. Williams argues that we should regard them as “parts of states of affairs, and have no peculiar being in their own right”, rather than as properties ( Williams 2019, 109). Apart from reconceiving the categorical/dispositional distinction in non-modal terms, proponents of the hybrid position may also revise the notion of modality in question. As Michael Raven points out to me, philosophers who construe the categorical/dispositional distinction in modal terms seem to be operating with a criterion of something’s being modal as: F is modal if for some G, necessarily if F then G. Yet, a viable criterion for F’s being modal may need to be stated purely in terms of F’s relations to some G. For a comprehensive reply to Katzav’s argument, see Smart & Thébault (2015). For the distinction between modal and non-modal conceptions of essence, see Chapter 7, “Modal Conceptions of Essence” (this volume) and Chapter 8, “Non-modal Conceptions of Essence” (this volume).

205

Ka Ho Lam

Further Readings Bird, A. (2007) Nature’s Metaphysics: Laws and Properties. Oxford: Clarendon Press. (A comprehensive overview of the debates between dispositional monism, categoricalism and the two intermediate positions.) Williams, N. E. (2019) The Powers Metaphysic. Oxford: Oxford University Press. (A detailed account of the hybrid position.) French, S. (2014) The Structure of the World: Metaphysics and Representation. Oxford: Oxford University Press. (A defense of the view that takes structure rather than properties as fundamental.) Vetter, B. (2015) Potentiality: From Dispositions to Modality. Oxford: Oxford University Press. (An extension of dispositional essentialism to metaphysical modality.) Yates, D. (2013) “The Essence of Dispositional Essentialism,” Philosophy and Phenomenological Research, vol. LXXXVII, no. 1, pp. 93–128 (An attempt to revise the notion of dispositional essence in order to accommodate findings in physics.)

References Armstrong, D. M. (1983) What is a Law of Nature? Cambridge: Cambridge University Press. Armstrong, D. M. (1997) A World of States of Affairs. Cambridge: Cambridge University Press. Armstrong, D. M. (2005) “Four Disputes about Properties,” Synthese, vol. 144, no. 3, pp. 309–320. Armstrong, D. M. (2012) “Defending Categoricalism” in A. Bird et al. (eds.), Properties, Powers and Structures: Issues in the Metaphysics of Realism. New York: Routledge, pp. 27–41. Austin, C. J. (2017) “Evo-devo: A Science of Dispositions,” European Journal for Philosophy of Science, vol. 7, no. 2, pp. 373–389. Azzano, L. (2021) “Dispositionality, Categoricality, and Where to Find them,” Synthese, pp. 2949–2976. Barker, S. (2013) “The Emperor’s New Metaphysics of Power,” Mind, vol. 122, no. 487, pp. 605–653. Barker, S. & Smart, B. (2012) “The Ultimate Argument against Dispositional Monist Accounts of Laws,” Analysis, vol. 72, no. 4, pp. 714–722. Beebee, H. (2000) “The Non-governing Conception of Laws of Nature,” Philosophy and Phenomenological Research, vol. 61, no. 3, pp. 571–594. Bigaj, T. (2010) “Dispositional Monism and the Circularity Objection,” Metaphysica, vol. 11, no. 1, pp. 39–47. Bird, A. (2007) Nature’s Metaphysics: Laws and Properties. Oxford: Clarendon Press. Bird, A. (2012) “Monistic Dispositional Essentialism” in A. Bird et al. (eds.), Properties, Powers and Structures: Issues in the Metaphysics of Realism. New York: Routledge, pp. 35–41. Brown, H. & Lehmkuhl, D. (2017) “Einstein, The Reality of Space and the Action-Reaction Principle” in P. Ghose (ed.) Einstein, Tagore and the Nature of Reality. London: Routledge, pp. 9–36. Chakravartty, A. (2007) A Metaphysics for Scientific Realism: Knowing the Unobservable. Cambridge: Cambridge University Press. Coates, A. (2021) “Essence and the Inference Problem,” Synthese, vol. 198, pp. 915–931. Dretske, F. (1977) “Laws of Nature,” Philosophy of Science, vol. 44, no. 2, pp. 248–268. Ellis, B. (2001) Scientific Essentialism. Cambridge: Cambridge University Press. Ellis, B. (2012) “The Categorical Dimensions of the Causal Powers” in A. Bird et al. (eds.), Properties, Powers and Structures: Issues in the Metaphysics of Realism. New York: Routledge, pp. 11–26. Ellis, B. & Lierse, C. (1994) “Dispositional Essentialism,” Australasian Journal of Philosophy, vol. 72, no. 1, pp. 27–45. Earman, J. & Roberts, J. T. (2005a) “Contact with the Nomic: A challenge for Deniers of Humean Supervenience about Laws of Nature. Part I: Humean Supervenience,” Philosophy and Phenomenological Research, vol. 71, no. 1, pp. 1–22. Earman, J. & Roberts, J. T. (2005b) “Contact with the Nomic: A challenge for Deniers of Humean Supervenience about Laws of Nature. Part I: Humean Supervenience,” Philosophy and Phenomenological Research, vol. 71, no 2, pp. 253–286. Engelhard, K. (2010) “Categories and the Ontology of Powers: A Vindication of the Identity Theory of Properties” in A. Marmodoro (ed.), The Metaphysics of Powers: Their Grounding and their Manifestations. London: Routledge, pp. 41–57.

206

Dispositional Essentialism French, S. (2014) The Structure of the World: Metaphysics and Representation. Oxford: Oxford University Press. French, S. (2020) “Doing away with Dispositions: Power in the Context of Modern Physics” in A. S. Meincke (ed.), Dispositionalism: Perspectives from Metaphysics and the Philosophy of Science. Cham: Springer, pp. 189–212. Giannotti, J. (2019) “The Identity Theory of Powers Revisited,” Erkenntnis, vol. 86, vol. 3, pp. 603–621. Giannotti, J. (2021) “Pure Powers are not Powerful Qualities,” European Journal of Analytic Philosophy, vol. 17, no. 1, pp. 5–29. Heil, J. (2003) From an Ontological Point of View. Oxford: Oxford University Press. Hirèche, S., Linnemann, N., Michels, R., & Vogt, L. (2021) “The Strong Arm of the Law: A Unified Account of Necessary and Contingent Laws of Nature,” Synthese, vol. 199, pp. 10211–10252. Hüttemann, A. & Kaiser, M. I. (2018) “Potentiality in Biology” in K. Engelhard & M. Quante (eds.) Handbook of Potentiality. Dordrecht: Springer, pp. 401–428. Ingthorsson, R. D. (2013) “Properties: Qualities, Powers, or Both?,” Dialectica, vol. 67, no.1, pp. 55–80. Jacobs, J. D. (2011) “Powerful Qualities, Not Pure Powers,” The Monist, vol. 94, no. 1, pp. 81–102. Katzav, J. (2004) “Dispositions and the Principle of Least Action,” Analysis, vol. 64, pp. 206–214. Katzav, J. (2005) “Ellis on the Limitations of Dispositionalism,” Analysis, vol. 65, pp. 92–94. Kimpton-Nye, S. (2021) “Reconsidering the Dispositional Essentialist Canon,” Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition, vol. 178, pp. 3421–3441. Kistler, M. (2020) “Powers, Dispositions and Laws of Nature” in A. S. Meincke (ed.), Dispositionalism: Perspectives from Metaphysics and the Philosophy of Science. Cham: Springer, pp. 89–114. Lange, M. (2013) “Grounding, Scientific Explanation, and Humean Laws,” Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition, vol. 164, no. 1, pp. 255–261. Lewis, D. K. (1973) Counterfactuals. Cambridge, MA: Harvard University Press. Lewis, D. K. (1986) “Causal Explanation” in D. Lewis (ed.) Philosophical Papers, Volume II. Oxford: Oxford University Press, pp. 214–240. Lewis, D. K. (1994) “Humean Supervenience Debugged,” Mind, vol. 412, pp. 473–490. Lewis, D. K. (2009) “Ramseyan Humility” in D. Braddon-Mitchell & R. Nola (eds.) Conceptual Analysis and Philosophical Naturalism. Cambridge, MA: MIT Press, pp. 203–222. Livanios, V. (2018) “Hamilton’s Principle and Dispositional Essentialism: Friends or Foes?” Journal for General Philosophy of Science, vol. 49, pp. 59–71. Lowe, E. J. (2006) The Four-Category Ontology: A Metaphysical Foundation for Natural Science. Oxford: Oxford University Press. Marmodoro, A. (2020) “Power, Activity and Interaction” in A. S. Meincke (ed.), Dispositionalism: Perspectives from Metaphysics and the Philosophy of Science. Cham: Springer. pp. 55–66. Martin, C. B. (1993) “The Need for Ontology: Some Choices,” Philosophy, vol. 68, no. 266, pp. 505–522. Martin, C. B. (1997) “On the Need for Properties: The Road to Pythagoreanism and Back,” Synthese, vol. 112, no. 2, pp. 193–231. Martin, C. B. (2008) The Mind in Nature. Oxford: Oxford University Press. Martin, C. B. & Heil, J. (1999) “The Ontological Turn,” Midwest Studies in Philosophy, vol. 23, no. 1, pp. 34–60. Maudlin, Tim. (2007) The Metaphysics within Physics. Oxford: Oxford University Press. Meincke, A. S. (ed.) (2020) Dispositionalism: Perspectives from Metaphysics and the Philosophy of Science. Cham: Springer. Molnar, G. (1999) “Are Dispositions Reducible?,”The Philosophical Quarterly (1950‐), vol. 49, no. 194, pp. 1–17. Molnar, G. (2003) Powers: A Study in Metaphysics. Oxford: Oxford University Press. Mumford, S. (1998) Dispositions. Oxford: Clarendon Press. Mumford, S. (2004) Laws in Nature. London: Routledge. Mumford, S. (2006) “The Ungrounded Argument,” Synthese, vol. 149, no. 3, pp. 471–489. Mumford, S. (2007) David Armstrong. Stocksfield: Acumen. Paoletti, M. P. (2020) “Five Sources of Contingency for Dispositionalism,” Metaphysica, vol. 21, no. 1, pp. 9–30.

207

Ka Ho Lam Schroer, R. (2010) “Is There More Than One Categorical Property? Philosophical Quarterly, vol. 60, no. 241, pp. 831–850. Schroer, R. (2013) “Can a Single Property be both Dispositional and Categorical?” Metaphysica, vol. 14, no. 2, pp. 63–77. Shoemaker, Sydney. (1980) “Causality and Properties” in Paul van Inwagen (ed.) Time and Cause. Dordrecht: Reidel, pp. 109–135. Smart, B. & Thé bault, K. (2015) “Dispositions and the Principle of Least Action Revisited,” Analysis, vol. 75, no. 3, pp. 386–395. Smith, D. C. (2016) “Quid Quidditism Est?” Erkenntnis, vol. 81, pp. 237–257. Strawson, G. (2008) “The Identity of the Categorical and the Dispositional,” Analysis, vol. 68, no. 300, pp. 271–282. Tahko, T. E. (2015) “The Modal Status of Laws: In Defense of a Hybrid View,” The Philosophical Quarterly, vol. 65, no. 260, pp. 509–528. Taylor, J. H. (2013) “In Defense of Powerful Qualities,” Metaphysica, vol. 14, no. 1, pp. 93–107. Taylor, J. H. (2018) “Powerful Qualities and Pure Powers,” Philosophical Studies, vol. 175, pp. 1423–1440. Tooley, M. (1977) “The Nature of Laws,” Canadian Journal of Philosophy, vol. 7, no. 4, 667–698. Tugby, M. (2012) “Rescuing Dispositionalism from the Ultimate Problem: Reply to Barker and Smart,” Analysis, vol. 72, no. 4, pp. 723–731. Tugby, M. (2013) “Graph-Theoretic Models of Dispositional Structures,” International Studies in the Philosophy of Science, vol. 27, no. 1, pp. 23–39. Tugby, M. (2017) “Power Worlds and the Problem of Individuation,” American Philosophical Quarterly, vol. 54, no. 3, pp. 269–281. Tugby, M. (2020) “Grounding Theories of Powers,” Synthese, vol 198, no. 12, pp. 11187–11216. Vetter, B. (2012) “Dispositional Essentialism and the Laws of Nature” in Birds et al. (eds.), Properties, Powers and Structures: Issues in the Metaphysics of Realism. New York: Routledge, pp. 201–215. Vetter, B. (2015) Potentiality: From Dispositions to Modality. Oxford: Oxford University Press. Vetter, B. (2019) “Are Abilities Dispositions?” Synthese, vol. 196, no. 1, pp. 201–220. Vetter, B. & Jaster, R. (2017) “Dispositional Accounts of Abilities,” Philosophy Compass, vol. 12, no. 8, pp. 1–11. Williams, N. E. (2019) The Powers Metaphysic. Oxford: Oxford University Press. Yates, D. (2013) “The Essence of Dispositional Essentialism,” Philosophy and Phenomenological Research, vol. LXXXVII, no. 1, p. 93–128.

208

14 THE EPISTEMOLOGY OF ESSENCE Antonella Mallozzi

How do we know what’s essential to a given entity, say a particular individual or a natural kind? Is that something one can discover empirically, or by reasoning alone? Can we know the essence of a certain entity, or only some of its essential properties? Is investigating the essence of say, chemical elements or artifacts different from investigating the essence of numbers and moral qualities? These are some of the key questions that the epistemology of essence aims to address. This entry is a survey of the main answers that have been proposed within recent debates. Essentialist notions are usually regarded as modal notions, even though many think that essence is not definable in terms of de re necessity—especially following Fine’s (1994) influential paper (compare the entries on Modal Conceptions of Essence, Torza, Chapter 7, this volume; and Non-Modal Conceptions of Essence, Correia, Chapter 8, this volume). By means of counterexamples, Fine criticizes so-called modalism, which defines essence in terms of an object’s necessary properties. A prominent version is conditional modalism: x is essentially F iff (df) Necessarily, if x exists, then x is F. Fine proposes that the analysis should go the other way around. Drawing from Aristotle’s metaphysics, the Finean conception defines de re necessity in terms of an entity’s essence or nature, which is thought to be captured by a real definition.1 While being a modalist excludes endorsing the Finean conception, the reverse doesn’t hold. Indeed, anyone who accepts a notion of de re necessity may accept a modal view of essence. As Fine himself put it,2 the real question is whether in addition to the modal conception of essence there is another conception that further specifies what essence is. The question is whether modalism is sufficient for accounting for essence.3 There are important epistemological consequences depending on whether one chooses modalism or the Finean conception of essence. If you go for modalism, the epistemology of essence will coincide with knowledge of metaphysical necessity. That might be a significant advantage of the view since, if successful, an account of our knowledge of necessity will yield an account of knowledge of essence as a byproduct. On the other hand, if you favor the Finean conception (according to which not all the necessary properties of an entity are part of its essence), you might need two separate explanations of our knowledge of essence and necessity. A widespread strategy starts by giving an account of our knowledge of essence; then it investigates how that yields in turn

DOI: 10.4324/9781003008750-17

209

Antonella Mallozzi

knowledge of metaphysical necessity. This strategy assumes a direction of epistemic priority that matches the Finean conception of essence. Particularly, the strategy assumes that our epistemic procedures match the following bridge-principle connecting essence to necessity: If it is essential to x that it is F, then it is necessary that x is F. The idea is that since metaphysical necessity depends on essence, our knowledge of modality should proceed from knowledge of essence. An account of the former thereby presupposes an account of the latter. Note, however, that the Finean conception need not entail the epistemic priority of essence over modality. There might be cases where one comes to know a metaphysically necessary truth “directly”, so to say, and learn only later about the essence facts that underlie it. Perhaps, one might not get to know certain essence facts at all.4 Some authors find it problematic that the process of coming to know metaphysical necessity via knowledge of essence involves, at least conceptually, an “extra step” compared to the modalist conception (Casullo 2020; Horvath 2014; Tahko 2017). This needs to be explained, no matter how obvious or “automatic” it might seem. The crucial question is how we can come to know the bridge-principle connecting essence to necessity, since a real definition (e.g. gold is essentially the element with the atomic number 79) doesn’t make any explicit claim about metaphysical possibility or necessity (Horvath 2014). Following Kripke’s original suggestion, some philosophers conjecture that we know that transition a priori—whether via intuition, or reflection on the concepts involved (Hale 2013, Mallozzi 2021, Casullo 2020, Kment 2021, Jago 2021). Vaidya & Wallner (2020) propose that the bridge-principle follows from the essence of essence, in virtue of the fact that essences themselves have modal bearing. Presumably, we thereby know it a priori. Tahko (2023) maintains that epistemic subjects need not know or otherwise grasp the principle in order to gain modal knowledge. Its proper role is “rather at the level of our theory of modal epistemology” (3592 ). The idea seems to be that we may rather learn about the principle and its crucial role for modal knowledge from epistemologists working on the issue. But internalists might complain that, in general, an epistemology that relies on principles or mechanisms that might not be available to the knower does not provide an adequate explanation of that knowledge.5 While (on a Finean conception of essence) accounting for knowledge of essence may be crucial for explaining knowledge of necessity, I will set aside the issue of how we may know the bridge-principle and the connection to the epistemology of modality. Instead, we shall focus on knowledge of essence itself, i.e. on how a subject can gain (non-testimonial) knowledge of essence, by looking at the main available accounts from recent literature.

14.1

Preliminaries

Given the central role that the notion of essence has traditionally had in metaphysics, it’s surprising that only a few authors have explicitly addressed the question of how we know essence and essential properties and tried to give a systematic account of it. (I should signal, however, that recently the issue of knowledge of essence has been gaining prominence within the epistemology of modality due to a renewed interest in essentialist knowledge and its candidate role for knowledge of necessity. See Mallozzi Vaidya & Wallner 2021 for discussion). Several views within the epistemology of the a priori as well as the epistemology of metaphysics could be applicable to knowledge of essence, as well; perhaps some of them have that goal implicitly (examples include accounts of intuition and/or the understanding as an epistemic source, such as Bealer 1987, 1996; Bengson 2015; BonJour 1998; Chudnoff 2013; Sosa 2007). Still, in the present chapter we shall focus on those few recent theories that offer a dedicated, systematic account of knowledge of essence—an account that aims to explain specifically how one can

210

The Epistemology of Essence

come to know essence facts such as water is essentially H2O, Socrates is essentially human, or a triangle is essentially a plane figure with three sides and three angles. We shall assess how each theory scores with respect to the goal of explaining, or offer an illuminating characterization of, our knowledge of paradigm essence facts. Here are a few highlights from the discussion to follow. Preliminarily, all the accounts we’ll look at share robust realist and non-skeptical assumptions: essence facts exist objectively and mind-independently, and we are in a position to know them. Thus, I won’t discuss forms of skepticism and deflationism about essence and essentialist knowledge. (See the entries on Conventionalism, Sidelle and Livingstone-Banks, Chapter 30, this volume; Conferralism, Vaidya and Wallner, Chapter 32, this volume; and Quinean Anti-Essentialism, Fine, Chapter 29, this volume). Furthermore, all the accounts we discuss understand essence in terms of the nature of an entity x or the real definition of x, namely according to what above has been referred to as the “Finean conception of essence”. (Kripke, who is the first author we review, is often regarded as a modalist. The point is controversial; but defenders of the modalist interpretation might want to consider Kripke’s view as an exception to the list.)6 The broad Finean characterization is often spelled out in different ways: as we will see, accounts differ as to what essence or real definition exactly is. Indeed, each theory we’ll examine comes as an integral part of a broader philosophical system that includes a metaphysics of essence. Providing an epistemology of essence that complements the corresponding metaphysics seems a highly desirable feature for a candidate theory. If correct, such a theory would thereby meet, within the study of essence, what Peacocke (1999) has called the “Integration Challenge”: i.e. the general challenge of providing an account of how we can know truths in a given area, which fits a corresponding metaphysical account of that area. Furthermore, the theories we discuss seem to grant the metaphysics of essence a certain priority over the associated epistemology. They proceed (implicitly or explicitly) based on the methodological tenet that accounting for knowledge of essence requires at least some preliminary account of what essence is. This is in effect a “metaphysics-first” approach to the epistemology of essence (Mallozzi 2021). A further main issue is whether knowledge of essence is a priori or a posteriori. The answer may depend on one’s broader philosophical and methodological views, besides one’s conception of essence. Accounts that favor rationalist methods tend to treat knowledge of essence as a priori. According to those, we may gain essentialist knowledge via understanding, conceptual competence, imaginative exercises, and/or intuition. Whereas, more naturalistically oriented accounts see knowledge of essence as distinctively empirical, often as a product of scientific investigation. Of course, it’s not unusual that both a priori and empirical methods jointly contribute to knowledge of essence. Vice-versa, a priori and a posteriori knowledge of essence might require separate accounts (Hale 2013). Complications might arise if the a priori-a posteriori distinction doesn’t actually capture a genuine epistemological difference, and so isn’t philosophically useful after all (see e.g. Williamson 2013. For discussion: Boghossian & Williamson 2020). Additionally, essentialism has traditionally emphasized the connection between essence and explanation. The broad idea is that the essence of a given entity is what explains that entity’s certain other properties and behaviors. Aristotle is thought to have held that the essence of a kind is that feature that explains why kind-members have certain other (nonessential) features (Hauser 2019). Similarly, Locke held that the essence of a material substance consists in its internal constitution (its atomic or molecular structure) because this explains, causally, the observable, macroscopic features of the substance. Contemporary debates have witnessed a resurgence of this idea, starting with the work of Putnam (1975) and

211

Antonella Mallozzi

Kripke (1980). More recently, Gorman (2005), Kment (2014), and Godman Mallozzi and Papineau (2020) among others have offered accounts of essence that hinge on explanation, often pairing it with an empiricist epistemology. There’s a familiar worry that explanation is merely subjective and/or context-dependent and so it won’t help identify genuinely essential properties on a realist view. However, the notion at play in those accounts is either causal explanation (especially in the Locke-PutnamKripke strand, as well as the contemporary accounts), or so called “metaphysical explanation”. (The latter is sometimes identified with grounding: see the entry on Essence, Grounding, and Explanation, Kovacs, Chapter 20 of this volume.7 Aristotle’s notion of formal causation is indeed a sui generis kind of causation, which may broadly fit the category of metaphysical explanation). Both sorts of explanations capture real relations obtaining “out there” in the world, namely the cause of a given fact, or the reason why it obtains respectively. These are fully objective and mind-independent. Thus, they shouldn’t be treated as epistemic notions representing subjective, merely interest-dependent relations.8 Finally, although this entry focuses on contemporary contributions from analytic debates, I should mention Husserl’s (1973) account as an early prominent proposal from the phenomenological tradition (for discussion, see Mulligan, Chapter 5, this volume). Husserl held that we may gain knowledge of essence via an imaginative process called “eidetic variation”, whose outcome is justified by intuition. This is a process of abstraction where a subject voluntarily imagines different versions of a certain object, while having it retain some core similarities to the original. As the subject encounters cognitive resistance regarding specific variations, an invariant common structure is uncovered, i.e. the eidos or essence of the object. The subject has in effect an intuition or insight into the essence of the entity. Key questions for Husserl’s eidetic method concern the arbitrariness of the process, its reliability, and a worry of vicious circularity—for the success of the method might require some prior grasp of essence (for discussion: Mohanty 1991; Kasmier 2010; Wallner 2023).

14.2

Kripke

Kripke revived the study of modal metaphysics and essentialist notions within analytic philosophy in the 1970s. Although he didn’t lay out a systematic account of knowledge of essence, he might be thought of as “author zero” in these debates. His groundbreaking thesis that we can know certain necessities a posteriori (1980) has a crucial role for the investigation of our knowledge of essence. Kripke’s view is that we may come to know a posteriori necessities by inferring according to a model which involves a conditional whose antecedent is a non-modal proposition and whose consequent is a modal one. The conditional is If P, then necessarily P. According to Kripke, we may establish, via empirical means, that the antecedent of this conditional, P, is in fact true. Then, by applying modus ponens the consequent is inferred, so we can conclude necessarily P. The inference crucially relies on the conditional premise, If P, then necessarily P, which for Kripke we know “a priori via philosophical analysis”. Still, the conclusion of the inference is a posteriori, since one of the premises (namely P), was established empirically. What kinds of propositions does P stand for in the Kripkean model? Kripke’s examples include informative identity-claims featuring rigid designators, “A is identical to B” (e.g. “Hesperus=Phosphorus”). But also truths concerning substance composition (e.g. “water is H2O”); fundamental kind membership (e.g. “cats are animals”); or individual biological origins (e.g. “the Queen has actual parents X and Y”). Crucially, Kripke appears to suggest that these are all essential truths.9 It’s essential to water that it is composed of H2O. Being an

212

The Epistemology of Essence

animal is part of the essence of cats. And it’s essential to the Queen’s being that very person that she’s originated from that particular egg and sperm. Thus, the key idea for the epistemology of essence that we draw from Kripke’s work is that essential truths are often knowable a posteriori, in most cases via scientific investigation.

14.3

Lowe

Casullo has deemed Lowe’s account “the most sustained attempt in the literature to develop an epistemology of essence” (2020: 593). Lowe holds that knowledge of essence (which for him wholly grounds knowledge of metaphysical modality) consists in understanding what a given entity x is or would be (Lowe 2012). This means understanding or grasping the real definition of x. The distinction between nominal definition vs. real definition historically goes back to Locke, who himself developed it from Aristotle. While nominal definition defines the word that picks out an entity in the world, real definition defines the thing (res) itself. Nominal definitions are what we find in a good English dictionary; real definitions capture the essence or true nature of things. A real definition for Lowe is in many cases a generative principle, namely a principle stating what it is for something to come into existence. (Note that the entities in question are “in a suitable sense capable of generation”; so, not God, nor universals. 2012: 935–6). Lowe’s examples often involve abstracta from geometry: e.g. “A circle is the locus of a point moving continuously in a plane at a fixed distance from a given point” (2012: 935). One may come to know the essence of a circle or what it takes for there to come into being a circle simply by understanding its real definition. More generally, Lowe stresses that knowledge of essence is “a product simply of understanding—not of empirical observation” (2008: 39). However, as Lowe’s account purports to be general and range over any entity, canonical Kripkean cases of the necessary a posteriori, e.g. “Necessarily, water is H2O”, seem to pose an immediate challenge. For how are we to access the true nature of water simply via understanding? In response: first, according to Lowe, establishing standard cases of a posteriori necessities requires knowing certain general criteria of identity (e.g. that two material objects cannot exist in the same place at the same time, 2008: 26). For him, those are a priori premises that we may only grasp via understanding. Thus, Lowe contends that coming to know standard cases of a posteriori necessities requires further a priori metaphysical knowledge, which supposedly undermines their a posteriori status (2008: 44). This is somewhat surprising, however. For even granting the need for such underlying knowledge, adding extra a priori premises to a piece of inferential reasoning that contains empirical premises doesn’t make it any less empirical, let alone a priori. Second, a closer look at Lowe’s metaphysics of essence might help us better understand his epistemological claims. We saw that Lowe identifies essences with real definitions. Importantly, he also maintains that we shouldn’t think of those as further entities or reify them (on pain of infinite regress: 2008: 39). The essence or real definition of something is just what the entity is. Thus, it would be a mistake for example to identify essence with molecular structure, or an individual’s DNA. At the epistemological level, while the discovery of the latter is usually thought to require empirical (scientific) investigation, understanding what an entity is doesn’t involve any such requirement. But then, one might wonder, what is the essence of water, say, if not its microstructure? When directly addressing this question, Lowe distinguishes between different conceptions of essence, apparently based on pragmatic and epistemic factors. On the one hand, he

213

Antonella Mallozzi

concedes that “if we are using the term ‘water’ to talk about a certain chemical compound whose nature is understood by theoretical chemists, then indeed we should say that it is part of the essence of this compound that it consists of H2O molecules” (2008: 33). On the other hand, “the existence of this compound is a relatively recent discovery” that wouldn’t have been possible without modern chemical knowledge (2008: 44). Hence, Lowe concludes that our everyday use of the term “water”, as well as our forebears’, doesn’t pick out “a chemical compound whose nature is now understood by theoretical chemists” and it’s wrong to “assume that it is part of its essence that it is composed of H2O molecules” (2008: 44). That’s also somewhat surprising. One might contend that it’s part of the commitment to essentialism that the way we use a term should in no way affect the properties of the thing picked out by the term. (Similarly for what we know about the thing). A better way to understand those divergences is by acknowledging that our linguistic competence and everyday use with a given term might be lacking in certain cases, and so it might not successfully capture the essence of the corresponding entity. How we think and talk about x vs. the nature of x are completely independent issues. Additionally, following Locke, Lowe emphasizes that “we do, and should” classify kinds of substances based on their macroscopic appearance and behavior, rather than “their supposed ‘real essences’” (Lowe 2011: 17). But this general criterion also raises questions. How are we to tell which of those observable features are essential to an entity? How do we select the “right” set of properties? Perhaps Lowe would insist that this just is the process of understanding what an entity is: identifying somehow the right set of properties amounts to grasping the real definition. Even so, the criterion seems to conflict with Lowe’s rejection of the role of sensory perception for knowledge of essence. For how are we to learn about things’ observable characteristics, if not via observation? (Similarly in cases involving alleged “ostensive” not linguistic real definitions, e.g. colors: 2012: 942). Still, Lowe’s treatment of canonical cases of a posteriori knowledge of essence might not be as problematic as it seems. An important claim is that, even where a thing is clearly definable and its real definition is a generative principle, that doesn’t entail that the thing actually exists. This is a point Lowe also conveys with a slogan, “Essence precedes existence” (Lowe 2008: 45). The priority of essence to existence is for him not only ontological but also epistemic: we need to know what the essence of an entity is before establishing whether it exists (2008: 40. For discussion: Tahko 2017, 2018; Casullo 2020). This means that empirical investigation would only be needed after we have grasped the essence of an entity, which is again achieved purely via understanding. Indeed, Lowe seems to be defending a transcendental thesis. We must know at least to some extent what an x is, that is, have at least partial knowledge of its essence, in order to think and talk about x at all (2012: 945. For discussion: Sgaravatti 2016; Tahko 2017, 2018, 2023; Vaidya & Wallner 2021). The point generalizes: knowledge of essence is for Lowe a prerequisite to have any knowledge, including empirical knowledge (2008: 33). Additionally, Lowe clarified more recently that “there is really no such thing as ‘purely’ a priori knowledge, nor any such thing as ‘purely’ a posteriori knowledge” (Lowe 2014: 268). The a priori is supposedly cyclical or “bootstrapping” with a posteriori knowledge; so perhaps he would ultimately allow that some empirical information plays a role for knowledge of essence. (For discussion: Tahko 2017). However, one wonders how we obtain knowledge of essence via understanding, exactly. On the one hand, Lowe rejects the evidential role of intuitions as well as conceivability for knowing metaphysical claims in general (2014). On the other hand, he excludes that grasping the essence of an entity is a form of linguistic or conceptual mastery: “knowing the nature or essence of a (possible) kind of being or entity cannot be reduced to knowing the meanings of

214

The Epistemology of Essence

words or understanding concepts and knowing logical relations between them” (2008: 33; 2014). But then how is understanding supposed to operate? Ruling out linguistic/conceptual mastery seems especially problematic. In the epistemology of the a priori understanding is usually identified with some grasp of linguistic/conceptual items. Knowledge of analyticity in particular is knowledge one may gain a priori in virtue of understanding a certain sentence/ proposition; where the relevant a priori beliefs/inferences are partly constitutive of understanding, or otherwise based on such an understanding (Boghossian 2020). Once conceptual mastery is off the table, it isn’t clear how understanding is supposed to operate, and so ultimately how one may arrive at knowledge of essence. In his (2008) Lowe appealed to a “grasp of, or rational insight into, certain necessary relationships … [an] insight into their natures or essence” (33). That this may yield a priori knowledge of necessary truths is an idea that several authors have defended (Bealer 1992; BonJour 1998; Chudnoff 2013). But elsewhere Lowe explicitly rejects intuition as a source of essence knowledge, as mentioned (2014: 256). The notion of understanding and its role for knowledge of essence should thus be clarified. (Similarly: Bengson Cuneo and Shafer-Landau 2022: §4. For further discussion: Casullo 2020; Mallozzi (ms.); Wallner 2023).

14.4

Hale

Hale (2013) gives two separate accounts of a priori vs. a posteriori knowledge of essence, where the underlying assumption is that to know the essence or nature of something is to know its real definition.10 For Hale, a priori knowledge of essence is grounded in knowledge of meaning. It shouldn’t strike one as problematic that we may know nominal definitions a priori, especially if we think of that as the result of linguistic stipulation. But how is the definition of a word supposed to give us access to the nature of the thing picked out by that word? Hale holds that cases where we may know real definitions a priori are cases where an entity’s real definition and nominal definition coincide: “it often happens that … the correct definition of a thing and the correct definition of a word for the thing can be stated using the very same words” (254). Straightforward examples, for Hale, are those where we can give an explicit definition of a word, namely, one that states analytically necessary and sufficient conditions for its application. Take the case of square. The explicit definition of “square” can be stated as a plane figure that has four straight sides of equal length, meeting at right-angles. But that’s just what being a square is, namely its real definition. As he puts it, “precisely because such a definition gives necessary and sufficient conditions for the word ‘square’ to apply, there is no mystery how we can know that there is no more (and no less) to being square than satisfying those conditions” (255). More generally, Hale says that in such cases the essence is transparent, namely a priori accessible. Further examples include simple analytic truths such as spinsters are unmarried women or cobs are male swans. Additionally, for Hale we can gain a priori knowledge of essence via knowledge of meaning when definitions are implicit. Take the logical constants. According to a popular view, we know implicitly what the logical constants mean a priori, by being disposed to assent to basic logical truths or inferences involving them. For example, in the case of “and”, one needs to be disposed to assent to statements of the form: If A and B, then A. But mastering “and” in this way is just knowing what conjunction is, namely its essence. For Hale, the implicit nominal definition of “and” coincides with its real definition, thus we can access the latter a priori simply by knowing the former. Similarly for the case of numbers. On the basis of definitions by abstraction, one can gain knowledge of the nature of cardinal numbers a

215

Antonella Mallozzi

priori (256–7. Also Hale 2021). Thus, in both implicit and explicit cases, knowledge of meaning allegedly suffices for a priori knowledge of essence. By contrast, cases of a posteriori knowledge of essence have the key feature that the nominal definition is different from the real definition. For example, the nominal definition of “water” is (roughly) the drinkable liquid that is found in rivers and lakes on Earth. Whereas, the real definition of water is H2O. Thus, we need empirical investigation to find out what a thing is or its real definition. (Incidentally, for Hale our knowledge of metaphysical possibility, which is itself grounded in knowledge of essence, is “always and only a posteriori”. 2013: 7). Hale maintains that a posteriori knowledge of essence respects the Kripkean model for knowledge of a posteriori necessities. Appropriate inferences proceed according to specific Kripke conditionals that instantiate general principles such as the necessity of identity or the necessity of origins (e.g. If Hesperus is Phosphorus, then necessarily Hesperus is Phosphorus; If this table is not made of ice, it is necessarily not made of ice). But how do we know such principles? As we saw, for Kripke that’s a matter of a priori philosophical analysis. An a priori proof can be offered for the necessity of identity (Kripke 1971), while the necessity of origins is “susceptible of something like a proof” (1980: 114 fn.56. Kripke’s argument is notoriously controversial. See e.g. Salmon 1979; Cameron 2005. Also, we might note that the availability of a proof doesn’t entail that a subject actually grasps it, which might undermine the justification of the resulting modal knowledge). While Hale largely agrees with Kripke, he also interprets him as a modalist. Instead, for Hale, essence grounds necessity, which suggests in turn that knowledge of essence is prior to knowledge of necessity. Thus, Kripke’s account needs opportune modifications to explain a posteriori knowledge of essence, not (merely) knowledge of necessity. Hale’s key point is that the particular Kripke conditionals can be inferred from certain general principles of essence: “principles asserting, schematically, that such-and-such a property is essential to its instances” (269). Hale’s proposed principles concern substance, kind membership and kind inclusion, besides the necessity of origins and of identity themselves, as well as canonical theoretical identifications. For example, it is a general principle of essence that any object is essentially an object of a certain general kind (essentiality of kind-membership); or that each living thing essentially has the particular origin it has (essentiality of origins). Once applied to specific instances, these principles cover many standard cases of a posteriori knowledge of essence. Of course, the crucial question is how we know those principles of essence. Hale offers extensive discussion of kind-membership. There’s a broad abductive argument for it, as the principle might be taken to underwrite a “simple and straightforward” answer to the question What is it to be a given object?, as well as to help draw the distinction between accidental vs. essential properties (276). But Hale’s main argument for kind-membership hinges on pure sortals and their connection to the identity of objects (270). While sortal predicates in general represent kinds, pure sortals in particular for Hale capture that restricted class of properties that are not merely semantic restrictions of other kinds (e.g. man working for the Home Office is a merely semantic restriction of man; whereas human being isn’t a merely semantic restriction of mammal). Also, pure sortals apply to objects throughout the whole of their existence. Thereby, Hale points out that kind-membership indirectly “asserts, in effect, that each object falls under some pure sortal concept” (274). Thus, one might provide support for kind-membership by showing that the identity conditions of kind-members are captured by pure sortals. Hale’s twofold epistemology of essence raises several questions. The account of a priori knowledge of essence seems to presuppose that treating a priori knowledge of essence in terms

216

The Epistemology of Essence

of knowledge of meaning simplifies one’s explanatory task. This is a familiar move in the epistemology of the a priori: reducing the a priori to the analytic is alleged to make it less mysterious, especially compared to accounts that appeal instead to special mental faculties like rational intuition. However, this strategy is controversial. BonJour (1998) has offered extensive criticism of the attempt of reducing the a priori to the analytic, for several different conceptions of analyticity. Main issues for those attempts include implicit reliance on one’s antecedent grasp of the truths of logic, as well as tacit appeal to rational insight. Indeed, for BonJour, appeal to rational insight is inevitable to explain how a priori knowledge and justification might result from understanding of meaning. Moreover, even somebody like Boghossian, who has devised what’s probably the most promising strategy for explaining cases of the a priori in terms of analyticity (more precisely, epistemic analyticity, Boghossian 1996), is cautious in claiming that knowledge or understanding of meaning is sufficient for a priori justification. Even in cases where understanding of meaning seems constitutive of a priori justification, Boghossian stresses that a proper account will need to rely on the existence of true bridge-principles connecting understanding and justification in the appropriate way. And “It’s a non-trivial question in the theory of the a priori whether there are such bridge principles” (2020: 188). Indeed, Boghossian has also recently conceded that “one cannot escape appealing to intuitions in the theory of the a priori” (2020: 186). Substantive normative principles are main examples, as they can’t be merely encoded in the ingredient canonical normative concepts like in the case of analytic truths. If we can know a priori the true nature or essence of right and wrong at all, that must be via intuition not knowledge of meaning (Boghossian 2021). With this in mind, one wonders how exactly a priori knowledge of essence is supposed to result solely from knowledge of meaning, in Hale’s account. First, granted that in such cases nominal and real definition coincide, how do we tell which cases have this feature? One should somehow trust that all there is to being a certain entity is captured by the nominal definition; but that might require some independent grasp or access to the real definition. Second, is knowledge of meaning really sufficient for knowledge of essence? Some examples of analytic truths seem problematic, e.g. all cobs are male swans; or, all vixens are female foxes. These truths involve biological kinds, whose essence we may only discover empirically.11 You haven’t clarified what being a cob is by simply defining it as a male swan. Hence those analytic truths appear at best to provide a nominal definition not a real definition. The case of spinster, or bachelor, may seem more promising as those are social kinds that are importantly partly a result of our stipulations. But one might argue, á la BonJour, that our knowledge of such truths actually depends on our grasp of the underlying logical truth that All FGHs are F, so knowledge of meaning isn’t strictly sufficient for the resulting knowledge. Even most basic cases involving the logical constants don’t seem immune from these concerns. We saw that knowing what “and” means requires being disposed to infer in appropriate ways. But one might contend that that involves more than mere knowledge of meaning; for one needs to “see” that those inferences are correct, i.e. have some independent grasp of the logical law itself. In sum, knowledge of meaning might only give us a partial account of a priori knowledge of essence. Some appeal to intuition or rational insight into the nature of things might be what’s missing in Hale’s account (for further discussion: Mallozzi ms). Regarding Hale’s account of a posteriori knowledge of essence, a crucial question concerns the justification of the general principles of essence. To be fair, Hale is well aware of this (2013: 269). In the case of kind-membership, we saw that the principle largely relies on pure sortals. Wallner (2023) argues that the capacity of identifying and applying pure sortals requires itself essentialist information, so that knowledge of essence seems tacitly presupposed

217

Antonella Mallozzi

by the principle, rather than being achieved through it. A more general issue we might raise is whether one can identify some common ground for the principles, which may help us establish them and systematically guide us to knowledge of essence. An obvious candidate for a Neo-Aristotelian conception is explanation (more below), so it is somewhat surprising that Hale doesn’t explore this route. Wallner (2023) has argued that both Lowe’s and Hale’s epistemologies of essence are structurally incomplete, namely incomplete with respect to the account of epistemic justification they propose. For Wallner, it might prove fruitful to turn to Husserl’s account of knowledge of essence by incorporating his method of eidetic variation as well as his commitment to intuitive awareness of universals. Incidentally, within the philosophy of science, Brown (1994) has also defended the use of intuition and thought-experiments for knowledge of universals, which for him include truths about the physical world. Postulating essential properties is itself an a priori assumption for Brown, though confirmed by the success of scientific theories. Within moral epistemology, Bengson, Cuneo, and Shafer-Landau (2022) have stressed the importance of intuition-based approaches to knowledge of essence for the normative domains, as detailed in particular by eighteenth-century moral philosopher Richard Price, and possibly for other domains, as well. Finally, Jago (2021) shares important aspects with Hale’s theory in that he holds a crucial connection between kind-concepts (compare Hale’s pure sortals) and knowledge of essence. Though Jago argues for a “tendency” linking our knowledge that something is essentially F with our capacity for conceptualizing things under a kind F. For him, that’s just part of how we refer to objects in thought and language more generally. Jago’s externalist framework allegedly ensures that the tendency is reliable and safe from error.

14.5 Oderberg Oderberg’s (2007; 2011) account draws extensively from Aristotle’s. Oderberg distinguishes essences from essential properties (or necessary accidents) as well as, of course, mere accidents. Essential properties flow from the essence, namely are caused by or originate in the essence. (The notion of causation at stake is Aristotelian formal causation, a unique notion which Oderberg stresses is “different from any other kind of causation” 2011:102). For example, having a capacity for humor is an essential property of humans, whereas being a rational animal is the essence of being human. The capacity for humor (along with many other essential features) is caused by or originates in the essence of being human; whereas being a rational animal constitutes being human or is the real definition of being human (which Oderberg understands in terms of genus and differentia). But what are essences, exactly? For Oderberg, essence “consists of the parts or elements that constitute the thing as the kind of thing it is” (2011: 98). Those parts or elements are often improperly called “properties”; but, like Lowe, Oderberg contends that essence itself is not a bundle of properties: “What constitutes the essence are not properties at all” (Ibid.). An essence is instead a principle of unity or formal cause, which “keeps together” and unifies all the essential properties of a given entity. As Oderberg says with a slogan, “essence explains unity” (2007: 47). Without essences qua unifiers, we wouldn’t be able to answer the “unity problem” of how to account for the “unified repertoire of behaviors, operations, and functions indicative of a single, integral entity”, especially as it persists through change (45). Importantly, for Oderberg, essence cannot be a bundle of “privileged” properties, like it is suggested in the Locke-Kripke-Putnam tradition, for we would still have to address the unity problem of what holds those bundles together. Nor could essence be distinct from form,

218

The Epistemology of Essence

otherwise it would be totally extrinsic to the thing and dangerously similar to a bare substratum—namely, something that we don’t know what is but somehow plays the role of unifying or standing behind a thing’s qualities. (Compare Witt’s chapter on Unity, this volume). Thus, given this metaphysical conception of essence, the relevant question for us becomes how do we know the form or principle of unity an entity, as well as its essential properties? Oderberg follows Locke and the subsequent tradition in criticizing apriorism about knowledge of essence, at least for actual essences and essences that themselves aren’t the object of a priori investigation (e.g. mathematics). Consider essential properties first. For Oderberg, we know them based on empirical observation. For example, in the case of natural kinds, although we must distinguish between essential properties vs. properties universally possessed by kind-members, the latter are “nearly always essential” (2007: 50). Thus, it’s methodologically fair to assume they are essential based on empirical observation “in the absence of further argument or demonstration to the contrary” (Ibid.). Still, empirical investigation alone won’t suffice for knowledge of essence. Against Elder, Oderberg denies that there is an “empirical test for essence” (52 ff. More below). Instead, Oderberg appeals to what seem to be “armchair” methods. Certain “intellectual judgments that are metaphysical in nature” are needed to establish the essence of a thing as well as the essential properties that flow from it. We use “reason and common experience” to determine when a given quality is part of the essence (51). This is a familiar procedure: one supposes counterfactually in imagination that the entity in question lacks a given quality and asks whether it would be coherent to conclude that the entity would continue to display its characteristic properties and behavior. If that’s the case, then the quality in question is not part of the essence. But if removing the quality would cause “a general disturbance or radical change” in the functions of the thing (Ibid.), then it is part of the essence (or flows from it). Also importantly, like Lowe, Oderberg stresses that we may not know the complete essence of a thing; or even not know it at all, while still being justified in ascribing it to the thing. (See Tahko 2018 for a comparison between Oderberg and Lowe’s accounts). What about knowledge of essence itself qua form or principle of unity? Although “essence explains unity”, an abductive route seems out of the question for Oderberg. He remarks that this explanatory role is only derivative and the appeal to explanation must be made with caution (2007: 47). Instead, he invokes a priori metaphysical reasoning for identifying and classifying entities (based on genera and species), namely a fundamental grasp of ontological categories. “It is a metaphysical judgment that certain properties indicate that an object has a certain essence, i.e. that it has a substantial form that puts it into one category rather than another” (2007: 162). At bottom, for Oderberg, that there must be a metaphysical principle of unity is something that “we can only deduce by a priori metaphysical reasoning” (2011: 97). In sum, a combination of a priori and empirical methods is needed to gain essentialist knowledge. Oderberg’s account raises several questions. Lowe (2010) expressed concerns regarding, inter alia, his notion of real definition in terms of genus and differentia. Besides, we might expect naturalistically oriented metaphysicians to look with suspicion at the notion of essence qua substantial form or principle of unity, as well as at the notion of formal causation (admittedly, those are issues that Oderberg inherits from Aristotle). Focusing on the epistemology, a main issue seems to be how to tell the difference between parts or elements of the essence vs. essential properties that flow from it. Furthermore, Oderberg should address traditional concerns regarding the epistemic powers of imagination. How and to what extent is imagination and counterfactual reasoning a reliable guide to knowledge of essence and

219

Antonella Mallozzi

modal knowledge more broadly? (For an overview: Mallozzi Vaidya & Wallner 2021). Finally note that, following Mumford, Oderberg claims that in counterfactual reasoning we focus on a given object “as it is in this world” and on “how it would behave in this world were such and such a feature removed from it” (51). But one wonders how that should yield knowledge of essence and necessary accidents, which by definition belong to an entity at all possible worlds.

14.6

Elder

Elder (2004) defends what he calls a “commonsense” ontology against those “austere” views according to which only the most fundamental entities posited by microphysics really exist. For Elder, ordinary objects like trees, tables, and human beings all exist “in ontological strictness” (ix). Furthermore, they have essential properties by nature, which we can learn about empirically. Essences for Elder come in clusters of properties. Some properties of an entity are ensured or required by the presence of other properties; vice-versa, if a certain property that is part of an essential nature is taken away, other properties will have to be absent, as well. Elder maintains that there is an empirical test for essences, the “test of flanking uniformities”, which we regularly use “without quite realizing it” (23). In the case of natural kinds, we can gain evidence that a given property f is essential to a kind K by looking at kinds that display some contrary properties, while being otherwise similar to K. One should “see whether, among the members of (what seem to be) natural kinds roughly similar to Ks, differing from Ks by possessing some one property or another contrary to f, there are uniformly found other properties contrasting with other properties uniformly possessed by Ks” (37). For example, we are “warranted to judge” that atomic number 79 is an essential property of gold because other physical elements, which have atomic numbers contrary to 79, also display other contrary properties such as melting point, specific gravity, etc. (Ibid). Importantly, the test is strictly empirical, as it “doesn’t require that we know—via a priori insight, or via armchair expression of our conventions of individuation—“template” truths about the kinds of kinds (physical elements, mineral formations, chemical compounds, etc.) into which nature’s specific kinds fall” (38). Like Oderberg, Elder is also cautious regarding the connection between essence and explanation. Against a certain line of thought that traces back to Kripke, Elder doubts that “explanatory richness” is either a necessary or a sufficient condition for essentiality (4–5). In particular, Elder rejects that causal explanation might cast light on essence. While he maintains that something must “hold together” the clusters of essential properties, he excludes that they have a single underlying cause (26). Rather, for Elder, what plays this unifying role are the laws of nature (26–7). Here are some issues for Elder’s proposal. Oderberg (2007) wonders what exactly is meant by “rough similarity” among kinds, and whether that might lead to obvious errors in performing the test. Besides, he finds it problematic that the test requires that one observe “not merely that certain contrasting properties are absent when others are, but that they must be” (53). Rea (2002) argues that the test might single out properties that are only necessary conditions for membership in a particular natural kind, not truly essential properties. Elder considers Rea’s worry but replies that it rests on the mistaken assumption that “the objects that populate the world can lose membership in a given natural kind without ceasing to exist” (75–6). Kind-membership is instead “a life-and-death issue”, thus the test successfully identifies essential properties. But one might insist that the test might overachieve. For

220

The Epistemology of Essence

example, even though silver has a melting point that’s contrary to gold’s melting point, arguably silver might have had a different melting point while still being silver, say under very different environmental conditions. Perhaps melting point isn’t an essential feature of chemical elements, but the result of a combination of factors. Finally, one might wonder whether the test might even erroneously detect merely accidental properties for kindmembership. After all, nothing seems to preclude that those clusters of properties co-occur uniformly in kind-members merely due to random coincidence, not because they’re genuinely essential. Based on Elder’s test, it isn’t clear that we have a criterion to exclude such errors. Note that the idea—dismissed by Elder—that the essence of an entity is what is responsible for many other features of the entity has been recently developed by Godman Mallozzi and Papineau (2020). For them, essences are special core properties that have distinctive explanatory powers in accounting for how things are—essences are superexplanatory. More precisely, the essence of an entity is what causes and explains many of that entity’s other (non-essential) properties. In the case of natural kinds, the atomic constitution of a chemical element, say silver, explains why all samples of silver consistently share a whole host of properties and behaviors, such as density, electrical and thermal conductivity, disposition to combine chemically, and so on. Atomic number is thus the essence or “nature” of silver because of its unique role in explaining all those features that all samples of silver exhibit. Mallozzi (2021) further argues that in such cases knowledge of essence relies on scientific investigation aimed at identifying what plays such a superexplanatory role for a kind.

14.7

Kment

Kment’s account of knowledge of essence (2021) also hinges on explanation. The account is an integral part of his theory of modal knowledge and is based on his own modal metaphysics (2014). For Kment, modal facts are partially grounded in what he calls “metaphysical laws”, which include essential truths. These truths state conditions for being a certain entity or for instantiating a certain property or relation and play a distinctive explanatory role. For example, it is an essential truth for Kment that all gold atoms have atomic number 79. This truth explains why all and only atoms with atomic number 79 are gold atoms, as well as why a particular atom having atomic number 79 is an atom of gold. Accordingly, for Kment (2021), we may acquire knowledge of these essential truths thanks to the distinctive role they play in both metaphysical and causal explanation. How do we do so? There are mainly two methods. First, abductively, via inference to the best explanation. Second, a priori, via conceptual or linguistic competence. As to the first method, we may gain knowledge of essence via abductive inferences from facts that we know perceptually or a priori. Kment holds that identifying what causes or grounds some non-fundamental fact is an abductive process that crucially involves making assumptions about metaphysical laws, particularly essential truths. That’s because such truths or real definitions serve as covering laws in particular cases. For example, according to our best explanation, the fact that my cup of coffee is hotter than your glass of iced tea is grounded in the fact that the mean molecular kinetic energy of the former is higher than that of the latter. The covering law that is assumed to hold here is the real definition of beinghotter-than: one object is hotter than another if the mean molecular kinetic energy of the former exceeds that of the latter. Thus, we may establish essential truths via inference to the best explanation, given the supporting role they play for grounding and causal explanation. However, one might worry that abductive inference is a shaky basis for knowledge of essence. It is often remarked that abduction is an ampliative type of reasoning, and the best

221

Antonella Mallozzi

explanation might not necessarily be a mark of truth. Abduction only warrants an inference to the probable, or approximate truth. Additionally, several hypotheses might exemplify comparable theoretical virtues, which may prevent one from selecting one as the best explanation. Within Kment’s account, that results in turn in being undecided between as many candidate essential truths. As for the method of conceptual competence, Kment holds that competence with a term often requires at least implicit knowledge of (part or all) the real definition of the thing picked out by the term. For example, being competent with “vixen” requires knowing that all and only vixens instantiate the property of being female foxes, which is the real definition of vixen. Or in the case of modality, being competent with the notions of metaphysical necessity and possibility requires that one knows, at least tacitly, that their real definitions themselves essentially involve the metaphysical laws (similarly, Peacocke 1999). Thus, for Kment competent speakers are “in a position to know” essential truths just in virtue of such competence. Conceptual knowledge is a further, a priori source of knowledge of essence. However, similar issues to those raised for Hale’s account affect Kment’s proposal, as well. For it is not clear how speakers may extract the relevant knowledge that’s encoded in the ingredient terms.12

14.8 • • • • • • • • • •

Related Topics

Modal Conceptions of Essence Non-Modal Conceptions of Essence Natural Kind Essentialism Essences of Individuals Quinean Anti-Essentialism Conventionalism Conferralism Contemporary (Analytic) Contemporary (Phenomenological) Laws and Explanation

Notes 1 Many of the questions raised in this chapter are discussed in Aristotle’s Posterior Analytics. Versions of the Aristotelian conception developed by Fine are also widespread throughout antiquity, the medieval period, and the early modern era. See the corresponding entries by Malink, Frost, and Schechtman in this volume. 2 In conversation. 3 Several philosophers have objected to Fine’s views. Some arguments target Fine’s counterexamples. Others question the notion of real definition. Why are the features listed in a real definition necessary features of the object? Mackie (2020) remarks that “it looks as if the account of essence in terms of real definition is intended to deliver a modal rabbit out of a non-modal hat. And I don’t see how this can be done”. Also, how can we identify real definitions? For Gorman (2005), Fine’s account presupposes that we already know what the real definitions of things are. (See the chapter on Modal Conceptions of Essence, Torza this volume). 4 More precisely, we can distinguish between a truth that simply registers the essence of some entity without mentioning that that’s the essence of the entity—call it an “essential truth” (e.g. ‘water is H2O’) vs. a truth that explicitly reports an essence fact—call it an “essence truth” (e.g. ‘water is

222

The Epistemology of Essence

5 6 7 8 9 10

11 12

essentially H2O’). I’m following Bengson et al. (2023)’s terminology here. Raven (2020) has coined a similar “status/report” distinction. Thanks to John Bengson for raising this issue. Obviously we can’t enter the Kripkean exegesis here, but note that many disagree with the modalist interpretation, including Fine himself (in conversation) and, for what is worth, myself. See also Fine (2022) and Robertson Ishii’s chapter, this volume. See also Glazier (2017) for an alternative proposal. For epistemic readings of metaphysical explanation see e.g. Thompson (2016) and Maurin (2019). Note that for the modalist the consequent of the conditional is an essential truth (that registers the essence of the thing without explicitly saying so); but not so for the non-modalist. However, the antecedent may be regarded as an essential truth on both views. The dichotomy between a priori vs. a posteriori knowledge of essence might depend to an extent on the types of entities at stake. Do different kinds of objects have different types of essence? Compare e.g. the essence of water, or zebras, vs. the essence of a circle. If so, that might ground corresponding differences in epistemic methods involved with discovering such essences. See Roca-Royes (2017, 2018) for two different epistemologies for abstracta vs. concreta. An alternative account of essence truths involving sortal classifications is given in Fine (2005). I’m grateful to John Bengson, Kathrin Koslicki, Mike Raven, and Michael Wallner for helpful comments on an earlier version of this chapter.

References Bealer, G. (1987) The Philosophical Limits of Scientific Essentialism, Philosophical Perspectives 1: 289–365. Bealer, G. (1992). The Incoherence of Empiricism. Proceedings of the Aristotelian Society, Supplementary Volumes 66: 99–143. Bealer, G. (1996). The Possibility of Philosophical Knowledge, Philosophical Perspectives 10: 143–150. Bengson, J. (2015). The Intellectual Given. Mind 124 (495): 707–760. Bengson, J., Cuneo, T., and Shafer-Landau, R. (2022). Pricean Reflection. British Journal for the History of Philosophy. DOI: 10.1080/09608788.2021.2007846 Bengson, J., Cuneo, T., and Shafer-Landau, R. (2023). The Source of Normativity. Mind 132 (527): 706–729. Boghossian P. (1996). Analyticity Reconsidered. Noûs 30(3): 360–391. Boghossian, P. (2021). Normative Principles are Synthetic A Priori. Episteme, 18, 367–383. Boghossian, P. and Williamson, T. (2020). Debating the A Priori. Oxford: Oxford University Press. BonJour, L. (1998). In Defense of Pure Reason. Cambridge: Cambridge University Pres. Brown, J. R. (1994). Smoke and Mirrors: How Science Reflects Reality. Routledge. Casullo, A. (2020). Is Knowledge of Essence the Basis of Modal Knowledge? Res Philosophica 97 (4): 593–609. Cameron, R. (2005). A Note on Kripke’s Footnote 56 Argument for the Essentiality of Origin. Ratio XVIII 3: 262–275. Chudnoff, E. (2013). Intuition. Oxford: Oxford University Press. Elder, C. (2004). Real Natures and Familiar Objects. Cambridge, MA: MIT Press/Bradford Books. Fine, K. (1994). Essence and Modality. Philosophical Perspectives 8: 1–16. Fine, K. (2005). Necessity and Non-Existence. In Id., Modality and Tense. Philosophical Papers (pp. 321–356). Oxford: Oxford University Press. Fine, K. (2022). Some Remarks on the Role of Essence in Kripke’s “Naming and Necessity. Theoria 88 (2): 403–405. Glazier, M. (2017). Essentialist explanation. Philosophical Studies 174: 2871–2889. Godman, M., Mallozzi, A., and Papineau, D. (2020). Essential Properties Are Super-Explanatory: Taming Metaphysical Modality. Journal of the American Philosophical Association 6(3): 316–334. Gorman, M. (2005). The Essential and The Accidental. Ratio XVIII 3: 276–289. Hale, B. (2013). Necessary Beings. Oxford: Oxford University Press. Hale, B. (2021). Essence and Definition by Abstraction. Synthese 198: 2001–2017. Hauser, C. (2019). Aristotle’s Explanationist Epistemology of Essence. Metaphysics 2(1): 26–39. Horvath, J. (2014). Lowe on Modal Knowledge. Thought 3(3): 208–217.

223

Antonella Mallozzi Husserl, E. (1973). Experience and Judgment. Investigations in a Genealogy of Logic. Revised and ed. by L. Landgrebe. Transl. by J. S. Churchill and K. Ameriks. London: Routledge & Kegan Paul. Jago, M. (2021). Knowing How Things Might Have Been. Synthese 198: 1981–1999. Kasmier, D. (2010). A Defense of Husserl’s Method of Free Variation. In S. Luft and P. Vandevelde (Eds.), Epistemology, Archaeology, Ethics: Current Investigations of Husserl’s Corpus (pp. 21–40). New York: Continuum Kment, B. (2021). Essence and Modal Knowledge. Synthese 198: 1957−1979. Kment, B. (2014). Modality and Explanatory Reasoning. Oxford: Oxford University Press, 2014). Kripke, S. (1980). Naming and Necessity. Cambridge, MA: Harvard University Press Kripke, S. (1971). Identity and Necessity. In M. K. Munitz (Ed.), Identity and Individuation (pp. 135–164). New York University Press. Lowe, E. J. (2014). Essence vs. Intuition: An Unequal Contest. In A. R. Booth and D. P. Rowbottom (Eds.), Intuitions (pp. 256–268). Oxford University Press. Lowe, E. J. (2010). Real Essentialism - David S. Oderberg. The Philosophical Quarterly 60(240): 648–652. Lowe, E. J. (2012). What is the Source of our Knowledge of Modal Truths. Mind 121(484): 919–950. Lowe, E. J. (2011). Locke on Real Essence and Water as a Natural Kind: A Qualified Defence. Proceedings of the Aristotelian Society, Supplementary Volumes 85: 1–19. Lowe, E. J. (2008). Two Notions of Being: Entity and Essence. Royal Institute of Philosophy Supplement 62: 23–48. Mackie, P. (2020). Can Metaphysical Modality be Based on Essence? In M. Dumitru (Ed.), Metaphysics, Meaning, and Modality. Themes from Kit Fine (pp. 247–264). New York and Oxford: Oxford University Press. Mallozzi, A. (2021) Putting Modal Metaphysics First. Synthese 198: 1937–1956. Mallozzi, A. (ms.) Knowing the Nature of Things A Priori. Mallozzi, A., Vaidya, A., and Wallner, M. (2021). The Epistemology of Modality. In The Stanford Encyclopedia of Philosophy (Fall 2021 Edition) https://plato.stanford.edu/archives/fall2021/entries/ modality-epistemology/ Maurin, A. (2019). Grounding and Metaphysical Explanation: It’s Complicated. Philosophical Studies 176: 1573–1594. Mohanty, J. N. (1991). Imaginative Variation in Phenomenology. In T. Horowitz & G. J. Massey (Eds.), Thought Experiments in Science and Philosophy (pp. 261–272). Lanham: Rowman & Littlefield. Oderberg, D. (2011). Essence and Properties. Erkenntnis 75: 85–111. Oderberg, D. (2007). Real Essentialism. London: Routledge. Peacocke, C. (1999). Being Known. Oxford: Oxford University Press. Putnam, H. (1975). The Meaning of ‘Meaning’. In K. Gunderson (Ed.), Minnesota Studies in the Philosophy of Science VII: Language, Mind, and Knowledge (pp. 131–193). Minneapolis: University of Minnesota Press. Raven, M. J. (2020). Is Logic Out of This World? Journal of Philosophy 117 (10): 557–577. Rea, M. (2002). World Without Design. Oxford: Oxford University Press. Roca-Royes, S. (2017). Similarity and Possibility: An Epistemology of De Re Modality for Concrete Entities. In B. Fischer and F. Leon (Eds.), Modal Epistemology After Rationalism (pp. 221–246). Dordrecht: Springer. Roca-Royes, S. (2018). Rethinking the Epistemology of Modality for Abstracta. In I. Fred-Rivera and J. Leech (Eds.), Being Necessary: Themes of Ontology and Modality from the Work of Bob Hale (pp. 245–265). Oxford: Oxford University Press. Salmon, N. (1979). How Not To Derive Essentialism from the Theory of Reference. Journal of Philosophy 76 (12): 703–725. Sgaravatti, D. (2016). Is Knowledge of Essence Required for Thinking about Something? Dialectica 70 (2): 217–228. Sosa, E. (2007). A Virtue Epistemology: Apt Belief and Reflective Knowledge. Oxford: Oxford University Press. Tahko, T. (2017). Empirically-Informed Modal Rationalism, in B. Fischer and F. Leon (Eds.), Modal Epistemology After Rationalism (pp. 29–45). Dordrecht: Springer. Tahko, T. (2018). The Epistemology of Essence. In A. Carruth, S. C. Gibb, and J. Heil (Eds.), Ontology, Modality, Mind: Themes from the Metaphysics of E.J. Lowe (pp. 93–110) Oxford: Oxford University Press.

224

The Epistemology of Essence Tahko, T. (2023). Possibility Precedes Actuality. Erkenntnis 88: 3583–3603. Thompson, N. (2016). Grounding and Metaphysical Explanation. Proceedings of the Aristotelian Society 116 (3): 395–402. Vaidya, A., and Wallner, M. (2021). The Epistemology of Modality and the Problem of Modal Epistemic Friction. Synthese 198: 1909–1935. Vaidya, A., and Wallner, M. (2020). Essence, Explanation, and Modality. Philosophy 95 (4): 419–445. Wallner, M. (2023). In Search for a Structurally Complete Epistemology of Essence. In D. Prelevic and A. Vaidya (Eds.), The Epistemology of Modality and Philosophical Methodology (pp. 150–175). Routledge. Williamson, T. (2013). How Deep Is the Distinction between A Priori and A Posteriori Knowledge? In A. Casullo and J. Thurow (Eds.), The A Priori in Philosophy (pp. 291–313). Oxford: Oxford University Press.

225

15 LANGUAGE OF ESSENCE Katherine Ritchie

How we structure our utterances and what words we use can affect what is conveyed, what comes to mind, and the sorts of judgements we are apt to make. For example, saying that someone is “an addict” and saying that someone “struggles with addiction” might lead to different judgments about the degree to which addiction is a defining feature of the person, whether this feature is contingent, and how much it explains about the person’s behavior. Similarly, “Raccoons eat avocados” and “Some raccoon has eaten an avocado” convey propositions of very different strengths. The former is a generic generalization which might be taken to be saying something (perhaps false) about a central feature of the entire kind raccoons. The latter, in contrast, is merely describing what some raccoon has done on an occasion. Expressions like “gold” have been taken to be semantically distinct from descriptions like “a soft yellow metallic substance” in that that the former, but not the latter, has its extension determined by causal relations between a naming event and a substance with an underlying essence (Kripke 1980). Philosophers and psychologists have considered the ways using locutions like “an F”, kind terms, and generics might, in one way or another, be getting at what people take to be essential (or not) to a person or kind. In this chapter I will consider the way that language affects our judgments about essence. The discussion will center on how we represent the world psychologically and linguistically, but I will also consider ways these representations might place constraints or reveal ways the world really is. For instance, it could be that certain linguistic constructions elicit robust judgments about some feature (e.g., eating avocados, having stripes, being athletic) being central to someone’s or some kind’s identity. If so, then it might be that we would judge that the person or kind would not persist through changes to this feature. Or, it might be that we would then judge that the true nature of the person or the kind involves F-ness. Or perhaps that a kind member that failed to express this feature was in some way defective or an anomaly. Our language might also impose this as a restriction on anything that a term picks out. For instance, something might need to have the chemical structure H2O in order to fall in the extension of “water”. That a term requires anything in its extension to have essence E does not show that there really is any thing which has E; the term could be empty, picking out nothing at all. Language itself might mislead us in systematic ways, so that we very often judge that something is essential when in fact it is not. While much of my focus here will be on our representations rather than reality, in the final section, I draw out some broader

226

DOI: 10.4324/9781003008750-18

Language of Essence

ramifications that essentialist language might have for other philosophical projects, including metaphysics. The chapter is structured as follows. First, I consider the ways “essence” is usually understood in debates about essentialist thought and language (§15.1). These do not always track how philosophers interested in essence use the term. Then I consider four components of natural language that have been argued to involve essentialism or to elicit essentialist thinking—generics, slurs and derogatory terms, natural kind terms, and nouns (§15.2). I offer examples of each and consider arguments that purport to show that these linguistic constructions or expression types essentialize. Finally, I consider what the foregoing discussion might reveal about philosophical methodology, metaphysics, and social and political projects (§15.3).

15.1

Representational Essentialism

While many of the other chapters in this handbook are focused on the metaphysics of essence (e.g., Wildman, this volume; Correia, Chapter 8, this volume; Mackie, Chapter 9, this volume; Tahko, Chapter 10 this volume; Robertson, Chapter 11, this volume), this chapter will focus primarily on representational or psychological essentialism. Understood in this way, essentialism is about how we humans reason, explain, and make judgments about things and stuff in the world. The psychologist Susan Gelman takes representational essentialism to involve two components. First, a kind component “that people treat certain categories as richly structured ‘kinds’ with clusters of correlated properties” (2004: 408). And, second, an essence component “that people believe a category has an underlying property (essence) that cannot be observed directly but that causes the observable qualities that category members share” (ibid.). Understood in this way, psychological essentialism can apply to our representations of kinds—like sharks and gold and men and women. The essence component of psychological essentialism has also been found to hold of representations of individuals (Christy, Schlegel & Cimpian 2019; De Freitas, Tobia, Newman & Knobe 2017). For instance, one might think that there is some deeper essence to being you, that determines your identity across time. It isn’t your appearance, but this deeper thing (perhaps your DNA, perhaps some essential character traits) that makes you you. To unify psychological essentialism about kinds and individuals, one could adopt an idea from Newman and Knobe (2019). They state that at its core psychological essentialism involves a “tendency to try to explain observable features in terms of a further unifying principle” (Newman & Knobe 2019: 2). This way of framing psychological essentialism leaves open two questions. First, how are essences represented? Second, what sorts of explanation relations might hold between observable features and underlying essences? Psychologists tend to hold that we often fail to have specific and well elaborated representations of essences and instead represent kinds with a placeholder essence that could be further explicated and about which we might defer to experts (Medin & Ortony 1989). For example, someone might think that there is some underlying essence that makes polar bears polar bears, without having a representation of exactly what that is. Much of the psychological research on essentialism has focused on investigating representations of animal kinds and human categories like race and gender. Researchers have found that underlying essences of these categories are usually taken to be, in some way or other, biological in nature. For instance, people might represent women as sharing DNA and take that to be the essence of the kind women. This, however, is not to say that this representation is correct as biological essentialism about gender is widely discredited (e.g., de Beauvoir 1972 [1949];

227

Katherine Ritchie

Haslanger 2003; Stoljar, Chapter 26, this volume). Rather, the claim is that people often represent social categories like race and gender as having biological essences. More recently, some have argued that essences could also involve underlying values/ideals or a telos. For instance, membership in the category scientist plausibly does not depend on biology, but it might still involve positing an essence—in this case perhaps something about seeking out truths about the nature of the world via systematic observations and testing (Newman & Knobe 2019; Tobia, Newman, Knobe, 2020). Some kinds might have a teleological essence relating to, e.g., their role in an ecosystem (Rose & Nichols 2020; cf. Neufeld 2021). Until very recently, the explanation relation was uniformly taken to be causal in nature. For instance, as we see stated in the Gelman quote above, essences are often taken to be underlying features that cause observable features. More recently, some have argued that a more general realization relation might be required to capture the broader range of essentialized representations. Newman and Knobe (2019) argue that the underlying value taken to be the essence of scientists might realize, but not cause, certain behaviors like carrying out meticulous observational studies. Psychologists have developed a number of experimental methods to see whether children and adults hold essentialist views about various categories and individuals. The main features that have been tested for are the following (see Gelman 2003: Ch. 1; Rhodes & Moty 2020 for further discussion): • Hidden and Not Directly Observable: if a category is not determined by observable features (e.g., things that are blue), but in a way that relies on something unobservable and underlying, then it is taken to be represented as having an essence (Markman 1989; Keil 1989; Gelman 2003; Bloom 2004). • Inductive Potential and Homogeneity: if a category is seen as having high inductive potential, this is argued to show that it is represented as a kind with an underlying essence. Category members in essentialized categories are also taken to be fairly homogenous. (Markman 1989; Gelman 2003, 2004; cf. Noyes & Keil 2019; Noyes, Dunham, Keil, & Ritchie 2021) • Heritability and Mutability: if membership in a category is taken to be immutable (or at least quite stable across situations) or inheritable this is taken to show that the category is represented as a kind with an underlying essence (Keil 1989; Gelman & Wellman 1991) • Explanation: if a category figures in explanations like the following, a is F because a is a member of category C, this is taken to support the view that C is represented as a kind with an underlying essence (Prasada & Dillingham 2006, 2009) • Discrete Category Boundaries: essentialized categories have discrete boundaries. Something either is or is not a member of the category, rather than sort of being a member of the kind. A number of different procedures have been designed to test for these features. To test whether category membership is determined by observable features or something underlying and unobservable, Gelman and Markman (1986; 1987) designed an appearance foil triad task. In this task participants are presented with three images. For example, one set of images included pictures of a black beetle, a green leaf insect, and a green leaf. While the leaf insect and leaf were superficially very similar, if the images were labeled as “bug” and “leaf” Gelman and Markman found that children expected the two bugs to be more similar than the leaf and the leaf insect. This shows that inductive potential aligns with category labels and does not just rely on commonalities in observable features like color and shape.

228

Language of Essence

To test for heritability, Gelman and Wellman (1991) developed the switched-at-birth task (Gelman & Wellman,1991). This paradigm involves asking participants whether some creature that is born to parents of kind K1 but raised by parents of kind K2 will have features characteristic of K1s or K2s. For instance, will a skunk baby that is raised by dogs bark (like dogs) or spray an unpleasant smelling substance (like skunks) when it is scared? If participants tend to think that it will have the skunk-property, this is taken to show that kind membership is not easily mutable and so, that the kind is represented as having an essence. It has been found that many biological and social kinds like race and gender are judged in ways that fit with psychological essentialism (Rothbart & Taylor 1992; Haslam, Rothchild, & Ernst 2002). Essentialist thinking plays a role in how we categorize, the inductive inferences we are apt to draw, what explanations we take to be licensed, and so on. These can be advantageous. For instance, essentialism might play a role in explaining why a drug that helps horses is not effective as a treatment in humans—the two kinds have different biological essences. Essentialism has also been argued to lead to significant harms, including dehumanization and increased stereotyping (Bastian & Haslam 2006, 2007; Haslam et al. 2002; Tirrell 2012; Livingstone Smith 2020; Neufeld 2020; but cf. Haslam & Levy 2006). Better understanding how language can elicit essentialism may be an important component in determining how to successfully mitigate prejudice and forge positive social change. While language is not the only driving force behind psychological essentialism (it might be an innate cognitive bias, there could be cultural effects, etc.), the language we use affects our judgments and our tendencies to essentialize. In the next four subsections, I consider four construction or expression types that have been argued to elicit psychological essentialism.

15.2

Linguistic Constructions and Expression Types 15.2.1

Generic Generalizations

Various construction types and expressions have been argued to essentialize. Generics, or generic generalizations, have been the primary focus of psychologists working on language and essentialism, and have gained prominence in recent work in philosophy largely due to the work of Sarah-Jane Leslie (2008, 2013, 2017). Generics are constructions in which a property is generalized across a category or kind without overt quantifiers like some, all, or many. Generics have been of particular interest for a number of reasons including that their semantics has proved difficult to correctly formalize, their connections with normativity (as in 1), and their connections with stereotypes about groups of people as in 2–3. In English generics can be expressed using bare plurals (1–4), indefinite singulars (5), or definite singulars (6). 1. 2. 3. 4. 5. 6.

Fans don’t let fans drive drunk/high.1 Girls are bad at math. Black people are good at basketball. Birds fly. A true friend always has your back. The whale is a mammal.

Bare plural constructions (BPs) are the most permissive form with which to express generic generalizations in English. They are also the most widely studied. I’ll focus my discussion here primarily on BPs, but consider arguments that the truth of indefinite singular generics (ISs) require that the property predicated be essential to the kind at the end of this section.

229

Katherine Ritchie

To begin let’s consider two important features of generics. First, they can require more than just an accidental universal generalization.2 For instance, suppose that as it turns out, everyone who has ever been a member of the US Supreme Court has an even social security number. Nevertheless, 7 does not strike most as true (Dahl 1975). 7. Supreme Court Justices have even SSNs. This shows that generics do, or can, require something stronger than actual universal generalization. They seem to have some sort of modal force. Call this feature Non-Accidental Generalization. A non-accidental generalization suggests that there must be some explanation for why the generalization holds. That the kind or category has an underlying essence would serve as such an explanation. For instance, birds fly because they have an underlying essence that leads them to share this type of locomotion. The second important feature of generics is that they can be true even if there are (perhaps many) exceptions. For instance, someone who learns of Ada Lovelace’s mathematical accomplishments, can still retain their belief that 2 is true. Similarly, even though penguins do not fly, most of us accept that 4 is a true generalization. Further, there isn’t a precise proportion of category members that need to have a property to be true. While most birds fly, less than half of birds lay eggs. Yet, people accept that it is true that birds lay eggs. Leslie (2008) pointed out that even when a minuscule number of a kind’s members have a property, a generic might be accepted as in 8. 8. Mosquitoes carry West Nile virus. Given this feature, which I’ll call Tolerate Exceptions, generics are hard to falsify. A number of theorists have argued that generics elicit essentialist thinking about kinds (e.g., Gelman 2003; Haslanger 2011; Rhodes, Leslie, & Tworek 2012; Langton, Haslanger, & Anderson 2012; Cimpian & Markman 2011; Wodak & Leslie 2017). For instance Wodak and Leslie claim that we should understand a generic like “Ks are F” “as by default communicating that members of the kind [K] share some distinctive, non-obvious and persistent property or underlying nature that casually grounds their common properties and dispositions, and that the property [F] is characteristic of that kind (i.e., widely posed by individuals who are [K] in virtue of their shared intrinsic nature)” (2017: 279). There is a significant body of empirical research suggesting that generics essentialize. Rhodes, Leslie, and Tworek (2012) found that generic language facilitates the formation and transmission of essentialist beliefs about novel social categories. They argued for their conclusion by showing that heritability, inductive potential, and explanation ratings were higher when children heard generics about novel kinds (e.g., “Zarpies climb trees”) than other sorts of construction (“this Zarpie/this one climbs trees”). So they conclude that hearing generic language leads to the formation of essentialist beliefs about a category. FosterHanson, Leslie, and Rhodes (2016) found that even negated generics like 9 increase judgments that there is an essence that kind members share. 9. Girls don’t hate math. While this research provides evidence that the use of generics can augment essentialist thinking, there are other bodies of work suggesting an alternative explanation. Some argue that generics are connected to kinds, but not necessarily kinds that are taken to have essences

230

Language of Essence

(Noyes & Keil 2019; Ritchie & Knobe 2020, Hoicka et al. 2021). They argue that some of the features taken to reveal a commitment to psychological essentialism are better understood in terms of representing a category as a kind or in terms of a psychological tendency to generalize, neither of which are necessarily connected to essentialist thinking. Hoicka et al. (2021) argue that both generics and high-proportion quantifiers for novel kinds, like “Daxes hate ice cream” or “Most Daxes hate ice cream”, lead adults and to some degree children to generalize properties across category members, but not to essentialize. In other work generics have been argued to connect to structural reasoning (Vasilyeva, Gopnik, & Lombrozo 2018; Ritchie 2019; Vasilyeva & Lombrozo 2020). For example Vasilyeva and Lombrozo (2020) found that people accepted generics when there were structural explanations for the connection between a category and property, like explanations involving governmental policies or access to health care. The truth of a generalization and a kind having inductive potential could rely on positioning within a broader external structure, without there being an innate unobservable biological essence (see also, Noyes, Dunham, Keil, & Ritchie 2021). Notice that structural explanations of generics and inductive potential also offer an explanation for Non-Accidental Generalization. In this case, rather than an inherent essence, it is shared social structural positioning that elicits the acceptability of the generic, judgments of strong inductive potential, and so on. Given that generics are licensed and true even without an intrinsic essentialist explanation, it is too strong to conclude that generics always essentialize. It has been argued, however, that generics of a particular form—namely indefinite singular generics, like 5 above—are always connected to essence (Lawler 1973; Burton-Roberts 1977). The idea is that the differences in distribution patterns of generic interpretations of BPs (10a, 10b) and ISs (11a, 11b) reveals that ISs only have generic interpretations when the predicated property is essential to the kind. 10. a. Bachelors are tall. b. Bachelors are unmarried. 11. a. A bachelor is tall. b. A bachelor is unmarried. Notice that while both 10a and 10b can have generic interpretations, the pair in 11 is different. It appears as if 11a has only an existential interpretation, while 11b has both a generic and existential reading. Since being tall is plausibly not essential to being a bachelor, but being unmarried is, the restriction to essential properties for IS generics could explain this difference in readings. However, it cannot adequately capture the felicitous generic interpretations of sentences like 12–13. 12. A toddler is a handful. 13. An elephant never forgets. These properties are not plausibly essential to the kinds toddler or elephant. So, it is too strong to require that properties predicated in IS generics be essential (see Cohen (2001) for an alternative account and further discussion). While evidence is mixed, there is evidence that generics increase essentialism and that IS generics are more felicitous with essential or definitional properties than other properties. There are differences across domains that show that these patterns are not completely general

231

Katherine Ritchie

(e.g., children and adults tend to essentialize animal categories more than artifactual categories; Gelman 2003; Keil 1989), nevertheless, the connection between generics and essentialism is fairly closely tied to certain linguistic construction types. It is generic interpretations of constructions of the form “As are F” or “An A is F”, across various choices of “A” and “F”, that elicit increased essentialism. In contrast, there are several threads of research focusing on subsets of nouns and arguing that these expressions essentialize. I consider classic arguments about natural kind terms and essence next, then turn to arguments that slurs essentialize in the following section.

15.2.2 Natural Kind Terms In analytic philosophy, the most famous discussions of the connections between language and essence are Putnam-Kripke style arguments about natural kind terms (like “water”, “tiger”, and “gold”) and names (like “Paderewski”). I’ll focus on the arguments applied to natural kind terms (perhaps correctly thought of as names of species) here. Putnam’s versions of the arguments squarely target internalism about meaning, that the meaning of a term is what individual language-users psychologically represent and which thereby determines the extension of the term. Kripke’s arguments target descriptivism about meaning, that the meaning of a term is a cluster of descriptions and reference is fixed through satisfying descriptions. Both hold views that connect to essentialism (Kripke 1980, Putnam 1975; see Koslicki 2008 for an overview). Putnam (1975) appeals to a thought experiment involving Twin Earth to argue for an externalist view of natural kind terms. In it we are asked to imagine a world superficially similar to ours with a substance that looks, smells, feels, and behaves like water, but which has a different chemical composition abbreviated as XYZ. We are asked to also suppose that on Twin Earth there are Twin English speakers who use a language that is phonologically and orthographically identical to English. The Putnam has us imagine an Earthian, Oscar, and his Twin Earthian duplicate, Twin Oscar. Both have thoughts about liquids, being thirsty, and so on. Putnam argues that while Oscar and Twin Oscar are duplicates, intuitively, Oscar thinks and talks about water (H2O) while Twin Oscar thinks and talks about twin water (XYZ). So, he concludes, meanings just aren’t in the head! External features including microstructural properties that language users may not represent at all are relevant to the extension of natural kind terms. Kripke argues that a natural kind term like “tiger” does not have a descriptive meaning like the ferocious carnivorous four-legged feline with tawny fur, black stripes, and a white belly. We might find, for example, a tiger without a white belly or with only three-legs and it would still fall in the extension of “tiger”. So meeting these descriptions is not necessary. Moreover, we might find out that no tigers are really orange (we were all suffering from an optical illusion). So, Kripke argues, the extension of “tiger” is not determined by meeting this description. Rather, Kripke holds that we take the term “in advance” to “designate a species” with an underlying internal structure (= essence) we perhaps have yet to discover and then hold “that anything not of this species, even though it looks like a tiger, is not in fact a tiger” (Kripke 1980: 121). The Putnam-Kripke arguments fit closely with the idea outlined in §15.1 that categories with essences are not determined by observable features but by an underlying essence. Given this view of the semantics of natural kind terms, some purported kind terms are necessarily empty (i.e., that they necessarily have nothing in their extension). For example, Kripke argues that unicorns are necessarily non-existent and that the actual word “unicorn” as used by us is necessarily empty (Kripke 2013). The argument relies on the view that unicorns are supposed to be a mythological species and the view that to be a species is to have an

232

Language of Essence

essence, not merely to be superficially similar (e.g., a horse-like creature with a single horn on its forehead). So, while there are possible horse-like creatures with a single horn, these possible creatures could have various different underlying biological natures, evolutionary histories, and so on. If “unicorn” had an extension it would need to pick out one of these possible species. But, there is nothing to single out possible species 1, from possible species 2, from … in order to make it the referent of “unicorn”. So, Kripke concludes, we cannot say that unicorns are a possible species and the word “unicorn” as used by us is necessarily empty. These arguments focus on a particular class of expressions—natural kind terms. Putnam, Kripke, and those who accept their conclusions do not hold that all kind terms have nondescriptive meanings. For instance, “bachelor” is plausibly a term for a social kind, but its extension very well may be fixed by description. We are not nearly as likely to defer to experts about the meaning of “bachelor” as we are for the meaning of “pyrite”. And the meaning of “bachelor” was not plausibly fixed by ostending to a sample bachelor. The Putnam-Kripke arguments are meant to apply to terms for animals, plants, and chemical substances. Natural kind terms might be taken to be semantically special, for instance their semantics might be externalist and name-like, while other kind terms are not name-like and do not have externalist meanings. However, one might also question whether they are really special in this sense. For instance, Wikforss argues that if one combines the view that natural kind terms are semantically special with the view that which kind terms are natural kind terms is only discoverable a posteriori one is committed to view that “there is a category of terms that is special from a semantic point of view, even though identifying this category depends on the development of sophisticated empirical theories, such as contemporary chemistry or evolutionary theory” (2010: 68). She takes it that a view of semantic specialness that requires detailed empirical research is implausible. So, perhaps we should not think that there is something semantically special about natural kind terms, even if there are some kind terms that pick out kinds which are metaphysically special. Another issue related to natural kinds terms and essences is whether Putnam-Kripke style arguments deliver metaphysical conclusions about kinds having essences. I’ll turn to arguments about the metaphysical import of these arguments in §15.3.

15.2.3

Slurs and Derogatory Terms

Another class of terms that has been argued to essentialize are slurs and other derogatory terms. Tirrell (2012) analyzes derogatory language used leading up to and during the Rwandan genocide. She argues that to be a deeply derogatory term, an expression must meet a condition that the negative content is “presumed to convey an essential aspect of the target … and in so doing, must create and enforce hierarchy” (2012: 191). Deeply derogatory terms express something about a target’s essence, make differences between groups seem inevitable, and are morally laden. Moreover, if differences are immutable and some are naturally worse than others, this could serve to justify differential treatment and making any supposedly required rehabilitation or change impossible. While Tirrell argues that many forces beyond language are involved in atrocities like genocide, she takes language to play a role. This role is partially played given the way deeply derogatory terms essentialize. Neufeld argues for an essentialist semantic analysis of slurring expressions. She argues that slurs “are kind terms encoding an “essence” of a social group, which is taken to explain a number of negative features attributed to the group” (2019: 2). On her view someone falls into the extension of a slur, S, for a social group G just in case they have the essence of G and this causes them to have stereotypical negative features of Gs. Since no one has a racial,

233

Katherine Ritchie

gender, ethnic, or other essence that causes them to have negative stereotypical features, on Neufeld’s account slurs are empty. They are “a species of failed kind terms; they are terms which, although introduced with the intention of designating kinds, fail to do so” (2019: 2). Notice the similarities with her argument and Kripke’s argument about unicorns. Both Neufeld and Kripke argue that certain terms are empty and fail to designate kinds. The difference is that while Neufeld takes slurs to be intended to designate kinds, fictional kind terms like “unicorn” were likely not introduced with such intentions (as least supposing they were introduced as terms for fictitious kinds).

15.2.4

Nouns and Labels

In the previous two subsections, we saw arguments that certain classes of nouns—slurs and natural kind terms—involve positing essences. We now turn to the broader view that nouns as a lexical category are poised to elicit essentialist thinking. Research has found that children and adults take nouns rather than adjectives or verbs to convey more information about a category (Gelman & Markman 1987; Markman 1989; Gelman & Coley 1990; Gelman & Heyman 1999; Waxman 1990; Jaswal & Markman 2002; Walton & Banaji 2004; Carnaghi et al. 2008). In experimental work Markman and Smith (reported in Markman 1989) found that in pairs with an adjectival and a nominal version of an expression, like in 14a and 14b, people took the nominal construction (i.e., 14b) to be stronger and more informative. 14. a. Sam is liberal. b. Sam is a liberal. This has led several psychologists and philosophers to argue that the lexical category noun is connected to being a kind and to essentialism (Wierzbicka 1986; Markman 1989; Gelman 2003; Ritchie 2021b). For example, Markman argues that using a noun for a category expresses something more central to the entity’s identity, has stronger inductive potential, and provides more essential information “stating what [the entity] is” (1989: 134). Leslie (2017) focuses much of her discussion of essentialism and language on generics, but she suggests that labels more generally might essentialize. Ritchie and Knobe (2020) use linguistic evidence to argue that lexicalized nominal expressions, like “women”, are represented as kinds which have members, opening the door for essentialism, while failing to invariantly lead to essentialist thinking. In other work, Ritchie (2021a, 2021b) argues that the semantics for predicate nominals like in 14b involves categorizing the subject as a member of the kind and trigger a presupposition that underpins our essentialist biases, whereas predicate adjectives like 14a merely involve an individual being assigned a property. This view too is not specific to particular categories of nouns (e.g., natural kind terms), but is meant to hold of predicate nominals of the form “A is an F” in general. If nouns are connected to essentialism, then this suggests that while it is not false that generics, natural kind terms, or slurs essentialize, it is something about nouns rather than, say, slurs that explains the connection between a particular expression and its propensity to essentialize.

15.3 Broader Connections: Metaphysics, Methodology, and Social Political Projects In §15.2 we saw arguments connecting linguistic constructions, expression types, and lexical categories with essentialism. To conclude, we’ll consider broader connections

234

Language of Essence

between language and essentialism and issues in metaphysics, methodology, and social political projects. Essentialist language might be thought to have important metaphysical upshots. For example, one interpretation of Putnam-Kripke arguments about essence and natural kinds terms is that a particular semantic theory delivers answers to metaphysical questions about the essence of persons (e.g., origin essentialism) or natural kinds (e.g., that chemical kinds have microstructural essences). However, Salmon (1979) argues that Kripke’s theory of reference does not deliver conclusions about kinds or individuals having essences, but rather assumes essentialist theses from the start. Kripke (1980) notes he did not intend to show essentialism followed from a theory of reference. Research on how humans posit underlying essences has been argued to have import for philosophical methodology. Leslie (2013) argues that cognitive biases related to essence—like the way use of generics and nouns elicits judgments that categories have innate essences, are homogenous, and so on—put pressure on the idea that intuitions are metaphysically illuminating. For example, consider again Putnam’s Twin Earth thought experiment. Even if one shares Putnam’s intuition that Twin Oscar’s use of “water” picks out twin water (XYZ) not water (H2O), Leslie asks us to question the force of the argument. This sort of essentialist judgment is ingrained in human cognition from a very early age. Why should one think, she argues, that it tracks the truth about the nature of kinds? Finally, essentializing language has connections to social-political philosophy and social justice projects. A number of philosophers have argued for linguistic prescriptions based on connections between language and essentialism. These have focused largely on the argument that generics should be avoided given their connections to essentialist thinking and its potential for pernicious effects (Haslanger 2011; Langton, Haslanger, & Anderson 2012; Leslie 2017; Wodak & Leslie 2017; Wodak, Leslie, & Rhodes 2015). If generics increase the propensity for people to take social kinds to have biological essences (e.g., to think there are racial essences) and these lead to dehumanization and prejudice, it is a short step to arguing that social generics should not be used. Moreover, as we saw with the Tolerate Exceptions condition, generics are also hard to falsify as they tolerate, sometimes many, exceptions. This is particularly worrisome when considering that generics can be used to express stereotypes. Since generics tolerate exceptions, pointing to exceptions to racial, gender, and other stereotypes might do little to combat them. For reasons like this, we might think social generics ought to be avoided. Some argue for broader prescriptions against essentializing language. For instance, Leslie suggests that “reducing the use of labels” (i.e., lexicalized nouns) “for racial, ethnic, and religious groups may reduce the extent to which children grow up essentializing these groups” so this strategy should be considered (2017, 420). Dembroff and Wodak (2018) argue we should not use pronouns that have gender features in part as a strategy to avoid gender essentialism. Other philosophers argue that generics and labels can be part of important social justice projects, so a general prohibition against social generics is too strong (Saul 2017; Ritchie 2019; Sterken 2020). For example, Saul (2017) argues that generic claims like “boys like pink too” are examples of “generic claims that campaigners for social justice might well want to make, as part of a social critique” (2017, p. 12). Ritchie (2019) argues that generics can be used to accurately describe the systematic nature of structural forms of oppression given the Non-Accidental Generalization feature. Ritchie (2021a) argues that the propensity for nouns like “woman” to elicit essentialist thinking reveals a limitation for anti-essentialist ameliorative projects that aim to change the meaning of terms in order to promote social change (e.g.,

235

Katherine Ritchie

Haslanger, 2000). Just changing a meaning won’t be sufficient to avoid essentialism if nouns as a lexical category are intimately connected to psychological essentialism. However, she does not conclude that this requires eliminating nouns for social categories, but suggests ameliorators must attend to our propensity to essentialize and to other features that affect how pernicious essentialism is. Language can affect our judgments about essence, making us take categorization to be immutable, inductively potent, explanatory, and so on. We have considered how these judgments might be tied to construction type (i.e., generic generalizations), expressions (i.e., natural kind terms, slurs), and lexical category (i.e., nouns). We concluded by considering some connections philosophers have drawn between essentializing language and metaphysics, methodology, and social-political philosophy. Future research ought to explore these issues further.

Notes 1 From a freeway sign in Los Angeles before the 2022 Super Bowl. 2 I say “can” as some of the constructions called generics do not. For instance, some generics might be true just in virtue of a majority of instances having the feature as in Leslie’s example “Barns are red” ( Leslie 2008).

Works Cited Bastian, B., & Haslam, N. (2006) “Psychological Essentialism and Stereotype Endorsement” Journal of Experimental Social Psychology 42(2), 228–235. Bastian, B., & Haslam, N. (2007) “Psychological Essentialism and Attention Allocation: Preferences for Stereotype-Consistent versus Stereotype-Inconsistent Information” The Journal of Social Psychology 147(5), 531–541. de Beauvoir, S., (1972) [1949] The Second Sex. Harmondsworth: Penguin. Bloom, P. (2004) Descartes’ Baby: How the Science of Child Development Explains what Makes Us Human. New York: Basic Books. Burton-Roberts, N. (1977) “Generic Sentences and Analyticity” Studies in Language 1, 155–196. Carnaghi, A., Maass, A., Gresta, S., Bianchi, M., Cadinu, M., & Arcuri, L. (2008) “Nomina Sunt Omina: On the Inductive Potential of Nouns and Adjectives in Person Perception” Journal of Personality and Social Psychology 94, 839–859. Christy, A. G., Schlegel, R. J., & Cimpian, A. (2019) “Why Do People Believe in a “True Self”? The Role of Essentialist Reasoning about Personal Identity and the Self” Journal of Personality and Social Psychology 117(2), 386–416. Cimpian, A., & Markman, E. M. (2011) “The Generic/Non-Generic Distinction Influences How Children Interpret New Information about Social Others” Child Development 82(2), 471–492. Cohen, A. (2001) “On the Generic Use of Indefinite Singulars” Journal of Semantics 18 (3), 183–209. Correia, F. (this volume) “Non-modal Conceptions of Essence,” in K. Koslicki & M. Raven (Eds.) Routledge Handbook of Essence. Routledge. Dahl, O. (1975) “On Generics,” in E. L. Keenan. (Ed.) Formal Semantics of Natural Language. Cambridge: Cambridge University Press. De Freitas, J., Tobia, K. P., Newman, G. E., & Knobe, J. (2017) “Normative Judgments and Individual Essence” Cognitive Science 41, 382–402. Dembroff, R. & Wodak, D. (2018). “He/She/They/Ze” Ergo: An Open Access Journal of Philosophy 5. Foster-Hanson, E., Leslie, S. J., & Rhodes, M. (2016) “How Does Generic Language Elicit Essentialist Beliefs?,” in A. Papafragou, D. Grodner, D. Mirman, & J. C. Trueswell (Eds.) Proceedings of the 38th Annual Conference of the Cognitive Science Society. Philadelphia, PA: Cognitive Science Society. Gelman, S. A. (2003). The Essential Child: Origins of Essentialism in Everyday Thought. USA: Oxford University Press.

236

Language of Essence Gelman, S. A. (2004) “Psychological Essentialism in Children” Trends in Cognitive Sciences 8 (9), 404–409. Gelman, S. A. & Coley, J. D. (1990) “The Importance of Knowing a Dodo is a Bird: Categories and Inferences in 2-year-old children” Developmental Psychology 26, 796–804. Gelman, S. A. & Heyman, G. D. (1999) “Carrot-Eaters and Creature-Believers: The Effects of Lexicalization on Children’s Inferences About Social Categories” Psychological Science 10(6), 489–493. Gelman, S. A., & Markman, E. M. (1986) “Categories and Induction in Young Children” Cognition 23, 183–209. Gelman, S. A., & Markman, E. M. (1987). “Young Children’s Inductions from Natural Kinds: The Role of Categories and Appearances” Child Development 58, 1532–1541. Gelman, S. A., & Wellman, H. M. (1991) “Insides and Essences: Early Understandings of the NonObvious” Cognition 38 (3), 213–244. Haslam, N. & Levy, S. R. (2006) “Essentialist Beliefs about Homosexuality: Structure and Implications for Prejudice” Personality and Social Psychology Bulletin 32, 471–485. Haslam, N., Rothschild, L. & Ernst, D. (2002) “Are Essentialist Beliefs Associated with Prejudice?” British Journal of Social Psychology 41, 87–100. Haslanger, S. (2000) “Gender and Race: (What) are They? (What) do We Want Them to Be?” Noûs 34 (1):31–55. Haslanger, S. (2003). “Social Construction: The “Debunking” Project,” in F. Schmitt (ed.) Socializing Metaphysics. Landham, MD: Rowman & Littlefield. Haslanger, S. (2011) “Ideology, Generics, and Common Ground,” in C. Witt (ed.) Feminist Metaphysics: Explorations in the Ontology of Sex, Gender and the Self. Dordrecht: Springer Netherlands. Hoicka, E., Saul, J., Prouten, E. Whitehead, L., & Sterken, R. (2021) “Language Signaling High Proportions and Generics Lead to Generalizing, but Not Essentializing, for Novel Social Kinds” Cognitive Science 45(11), e13051. Jaswal, V. K. & Markman, E. M. (2002). “Children’s acceptance and use of unexpected category labels to draw non-obvious inferences,” in W. Gray & C. Schunn, C. (Eds) Proceedings of the 24th Annual Conference of the Cognitive Science Society, Erlbaum. Keil, F. C. (1989) Concepts, Kinds, and Cognitive Development. MIT Press. Koslicki, K. (2008) “Natural Kinds and Natural Kind Terms” Philosophy Compass 3 (4), 789–802. Kripke, S. (1980) Naming and Necessity. Cambridge, MA, USA: Harvard University Press. Kripke, S. (2013) Reference and ExistenceL The John Locke Lectures. Oxford University Press. Langton, R. Haslanger, S. & Anderson, L. (2012) “Language and Race,” in G. Russell & D. G. Fara (eds.), The Routledge Companion to Philosophy of Language. Routledge. Lawler, J. (1973) “Studies in English Generics” University of Michigan Papers in Linguistics 1, 1. Leslie, S.-J. (2008) “Generics: Cognition and Acquisition” Philosophical Review 117 (1), 1–47. Leslie, S.-J. (2013) “Essence and Natural Kinds: When Science Meets Preschooler Intuition” Oxford Studies in Epistemology 4, 108–165. Leslie, S.-J. (2017) “The Original Sin of Cognition: Fear Prejudice, and Generalization” Journal of Philosophy 114 (8), 393–421. Mackie, P. (this volume) “Essences of Individuals,” in K. Koslicki & M. Raven (Eds.) Routledge Handbook of Essence. Routledge. Markman, E. (1989) Categorization and Naming in Children: Problems of Induction. Cambridge, MA: MIT Press. Medin, D. L., & Ortony, A. (1989) “Psychological Essentialism,” in S. Vosniadou & A. Ortony (Eds.) Similarity and Analogical Reasoning. New York: Cambridge University Press. Neufeld, E. (2019) “An Essentialist Theory of the Meaning of Slurs” Philosophers’ Imprint 19(35), 1–29. Neufeld, E. (2020) “Pornography and Dehumanization: The Essentialist Dimension” Australasian Journal of Philosophy 98(4), 703–717. Neufeld, E. (2021) “Against Teleological Essentialism” Cognitive Science 45(4), e12961. Newman, G. & Knobe, J. (2019) “The Essence of Essentialism” Mind and Language 34 (5), 585–605. Noyes, A., & Keil, F. C. (2019) “Generics Designate Kinds but Not Always Essences” Proceedings of the National Academy of Sciences 116(41), 20354–20359.

237

Katherine Ritchie Noyes, A., Dunham, Y., Keil, F. C., & Ritchie, K. (2021) “Evidence for Multiple Sources of Inductive Potential: Occupations and their Relations to Social Institutions” Cognitive Psychology 130, 101422. Prasada, S. & Dillingham, E. M. (2006) “Principled and Statistical Connections in Common Sense Conception” Cognition 99, 73–112. Prasada, S. & Dillingham, E. M. (2009) “Representation of Principled Connections: A Window onto the Formal Aspect of Common Sense Conception” Cognitive Science 33, 401–448. Putnam, H. (1975) “The Meaning of ‘Meaning’“ Minnesota Studies in the Philosophy of Science 7, 131–193. Rhodes, M., Leslie, S.-J., & Tworek, C. M. (2012) “Cultural Transmission of Social Essentialism. PNAS 109, 13526–13531. Rhodes, M. & Moty, K. (2020) “What is Social Essentialism and How Does it Develop?,” in M. Rhodes (ed.) Advances in Child Development and Behavior—The Development of Social Essentialism Vol. 59, Elsevier. Ritchie, K. (2019) “Should We Use Racial and Gender Generics?” Thought: A Journal of Philosophy 8 (1), 33–41. Ritchie, K. (2021a) “Essentializing Language and the Prospects for Ameliorative Projects” Ethics 131 (3), 460–488. Ritchie, K. (2021b) “Essentializing Inferences” Mind & Language 36(4), 570–591. Ritchie, K. & Knobe, J. (2020) “Kindhood and Essentialism: Evidence from Language,” in M. Rhodes (ed.) Advances in Child Development and Behavior—The Development of Social Essentialism Vol. 59, Elsevier. Robertson, T. (this volume) “Origin Essentialism,” in K. Koslicki & M. Raven (Eds.) Routledge Handbook of Essence. Routledge. Rose, D. & Nichols, S. (2020) “Teleological Essentialism: Generalized” Cognitive Science 44 (3). Rothbart, M. & Taylor, M. (1992) “Category Labels and Social Reality: Do We View Social Categories as Natural Kinds?,” in G. R. Semin & K. Fiedler (Eds.) Language, Interaction and Social Cognition. London: Sage. Salmon, N. U. (1979) “How Not to Derive Essentialism from the Theory of Reference” Journal of Philosophy 76 (12), 703–725. Saul, J. (2017) “Are Generics Especially Pernicious?” Inquiry, 1–18. Smith, D. L. (2020) On Inhumanity: Dehumanization and How to Resist It. New York, NY: Oxford University Press. Sterken, R. K. (2020) “Linguistic Interventions and Transformative Communicative Disruption” in H. Cappelen, D. Plunkett & A. Burgess (eds.) Conceptual Engineering and Conceptual Ethics. Oxford: Oxford University Press. Stoljar, N. (this volume) “Social Justice and Essence,” in K. Koslicki & M. Raven (Eds.) Routledge Handbook of Essence. Routledge. Tahko, T. (this volume) “Natural Kind Essentialsm,” in K. Koslicki & M. Raven (Eds.) Routledge Handbook of Essence. Routledge. Tirrell, L. (2012) “Genocidal Language Games,” in I. Maitra & M. K. McGowan (eds.), Speech and Harm: Controversies Over Free Speech. Oxford University Press. Tobia, K. P., Newman, G. E., & Knobe, J. (2020) “Water is and is not H2O” Mind & Language 35(2), 183–208. Vasilyeva, N., Gopnik, A., & Lombrozo, T. (2018) “The Development of Structural Thinking about Social Categories” Developmental Psychology 54(9), 1735–1744. Vasilyeva, N. & Lombrozo, T. (2020) “Structural Thinking about Social Categories: Evidence from Formal Explanations, Generics, and Generalization” Cognition 204, 1–14. Walton, G. M., & Banaji, M. R. (2004) “Being What You Say: The Effect of Essentialist Linguistic Labels on Preferences” Social Cognition 22(2), 193–213. Waxman, S. R. (1990) “Linguistic Bias and the Establishment of Conceptual Hierarchies: Evidence from Preschool Children” Cognitive Development 5, 123–150. Wierzbicka, A. (1986) “What’s in a Noun? (Or: How do Nouns Differ in Meaning from Adjectives?)” Studies in Language 10, 353–389. Wildman, N. (this volume) “Modal Conceptions of Essence,” in K. Koslicki & M. Raven (Eds.) Routledge Handbook of Essence. Routledge. Wikforss, Å. (2010) “Are Natural Kind Terms Special?,” in H. Beebee & N. Sabbarton-Leary (eds.), The Semantics and Metaphysics of Natural Kinds. Routledge.

238

Language of Essence Wodak, D. & Leslie, S.-J. (2017) “The Mark of the Plural: Generic Generalizations and Race,” in P. C. Taylor, L. M. Alcoff & L. Anderson (eds.) The Routledge Companion to the Philosophy of Race. Routledge. Wodak, D., Leslie, S. J. & Rhodes, M. (2015) “What a Loaded Generalization: Generics and Social Cognition” Philosophy Compass 10(9), 625–635.

Related Topics Griffith, A. (this volume) “Social Construction (Chapter 31),” in K. Koslicki & M. Raven (Eds.) Routledge Handbook of Essence, Routledge. Stoljar, N. (this volume) “Social Justice and Essence (Chapter 26),” in K. Koslicki & M. Raven (Eds.) Routledge Handbook of Essence, Routledge. Tahko, T. (this volume) “Natural Kind Essentialsm (Chapter 10),” in K. Koslicki & M. Raven (Eds.) Routledge Handbook of Essence, Routledge. Robertson Ishii, T. (this volume) “Origin Essentialism (Chapter 11),” in K. Koslicki & M. Raven (Eds.) Routledge Handbook of Essence, Routledge.

239

16 LOGIC OF ESSENCE Jon Erling Litland

Unlike in the case of modal logic there is as of yet no unified framework in which the various proposed logics of essence can be situated and compared; however, the most developed and familiar logic of essence is undoubtedly that of (Fine 2000; 1995b). This chapter accordingly adopts the approach of presenting (slightly simplified versions of) the language, proof theory, and semantics of Fine’s logic. Along the way we note various choice points, where different formal choices correspond to different conceptions of essence, and in this connection we discuss other approaches in the literature.

16.1

The Language of Essence

Let us consider four different essentialist claims 1 2 3 4

Socrates is essentially human i , i are essentially additively inverse square roots of 1 It is part of what it is to be human that humans are rational It is essential to disjunction that for any propositions p, q one may infer p and q

q from both p

In Fine’s system, essentialist claims are expressed using an indexed sentential operator . This notation for essentialist claims has become quite common in the metaphysics literature, where most authors take the operator to be indexed by (pluralities of) objects. Thus a is taken to be the operator “it true in virtue of the nature of a that …”. However, for the purposes of the logic of essence a slightly more general format is desirable. Following Fine, we take the essentialist operator to be indexed by predicates. If F is a predicate then F is the sentential operator: “It is true in virtue of the natures of the objects that are F that …”. More formally, the language of the logic of essence (henceforth LE) has: • • • •

Infinitely many first order variables x0, x1, …. For each n infinitely many pure n-place predicate letters F0, F1, … Infinitely many 1-place rigid predicate letters R0 , R1, … The existence predicate E, the identity predicate =, and the dependence predicate .

240

DOI: 10.4324/9781003008750-19

Logic of Essence

• The abstraction operation . • The essentialist operator . • The logical operations , ¬ ,

. (We use standard abbreviations for ,

, .)

The distinction between rigid and pure predicates is crucial for understanding the logic. The pure predicates can be thought of as standing for properties that do not depend on any objects, or as purely qualitative properties. The best way of thinking of rigid predicates is as follows. For any objects x0, x1, … consider the property being one of them. The rigid predicates stand for properties like that. The formation rules for the logic of essence are as follows: • • • • •

If x1, …, xn are variables and F is an n-place predicate then Fx1 … xn is a formula; Standard rules for ∀, ¬, ∨ If A is a formula and F a 1-place predicate then F A is a formula Every n-place predicate letter is an n-place predicate If A is a formula and x is a variable occurring in A then x. A is 1-place predicate.

(Note that allows us to form complex predicates; unlike in more familiar first-order logics the predicates extend beyond the predicate letters.) Let us distinguish between constitutive and consequential essence. (For more on this distinction see Correia’s Chapter 8, “Non-Modal Conceptions of Essence”.) The constitutive essence of an object only contains propositions that are “directly definitive of that object” (Fine 1994, 57) whereas the consequential essence of an object also contains the consequences of what is directly definitive of the object. LE is a logic of consequential essence; however, for the logic to be interesting we cannot let in all logical consequences, but must work with a “constrained” notion of consequence (Fine 1994, 59–60). The importance of working with a constrained notion of consequence comes up when we consider the essentialist conception of dependence. According to this conception a depends on b iff it is essential to a that some proposition involving b is the case (Fine 1995a). (For critical discussion of this essentialist notion of dependence see (Wilson 2020; Koslicki 2012).) For this notion of dependence to be non-trivial essence cannot be closed under full logical consequence. For suppose otherwise and suppose a p. Now let b be any object and H (b) any proposition involving b. Given that p H (b) is a logical consequence of p, a will depend on b. But since b was arbitrary this shows that a depends on every object. Let us return to how the four essentialist claims above can be expressed. The first two cases are easily accommodated. To express that Socrates is essentially human we first form the predicate x. x = s—being such that one is identical to Socrates. We then have: 1

x . x = s Hs—or,

it is true in virtue of the things that are identical to Socrates, that Socrates is

human As is common, we introduce the following shorthand and write a for the sentential operator x . x = a ; extending this notation in the obvious way we write x . x = a x = b …. We a, b, c … for may read this as “it is true in virtue of the natures of a, b , c , … that”. Our second essentialist claim above can then be expressed as follows: •

2 i, i (i)

=

1

( i)2 =

1

i+i=0

241

Jon Erling Litland 2 1 ( i)2 = 1 i + i = 0) (For the record, the official form is x . x = i x = i (i) = The third essentialist claim above raises a problem: what is the object in virtue of which every human is rational? Perhaps the property of being human? This yields the claim (using “H ” for the predicate “being human” and “R” for the predicate “being rational”):



property of being human

x (Hx

Rx)

As pointed out already by (Correia 2006) this is unlikely to work. The problem is that properties are of no use unless one knows what it is for them to be instantiated. And the obvious account is this. For any open sentence (x) the property of being is instantiated by an object a if and only if (a). But this immediately yields Russell’s paradox. For let R be the property that is instantiated by anything that does not instantiate itself. This property instantiates itself if and only if it does not instantiate itself. There are many possible responses to this, but perhaps the most natural is simply to deny that there is such a property as the Russell property. One way of doing that is by disallowing use of the instantiation relation in defining a property.1 This may solve the problem posed by an inconsistent theory of properties, but it creates grave problems for expressing essentialist claims. For even though there may be no property of non-self-instantiation it seems that one can make sense of essentialist claims like: • It is essential to being non-self-instantiating that one does not instantiate oneself • To be non-self-instantiating is to not instantiate oneself • For something to non-self-instantiate just is for that thing not to instantiate itself In these claims “being non-self-instantiating” functions as a predicate and not as a singular term denoting a property. It seems that such claims of “generic essence” cannot be expressed in Fine’s framework. The solution that stays closest to Fine’s framework involves going “higher-order”. Instead of working in a first-order language we work in a higher-order language where we can quantify into syntactic positions other than those of singular terms. In particular, one can quantify into sentence and predicate positions. Moreover, one has higher-order identity predicates that can be flanked by predicates and sentences, and one allows the -operator to bind variables of higher-type. Thus, where “H ” is the predicate “being human,” we can form to be the higher-order predicate X . X = H . We then allow the essentialist operator indexed by such higher-order predicates. The claim that it is essential to being human that humans are rational can then be expressed as follows: •

X.H=X

x (Hx

Rx)

(Ditter 2022) has shown how to develop such a higher-order logic of essence in great detail. A different way of dealing with the problem of generic essence is this. Several authors—(Rayo 2013; Dorr 2016; Correia and Skiles 2019)—have drawn our attention to “just-is,” or “to be … is to be … ” statements. Following (Linnebo 2014) we will call these “generalized identities”. Consider, e.g., the claim that to be a vixen just is to be a female fox. We can formalize this claim as follows: • Vx

x

Fem(x)

Fox(x)

242

Logic of Essence

Here is the generalized identity operation. What is distinctive about this operation is that it also binds and generalizes the variable x. The generalized identity claim is thus about what it is to be a vixen, in general, not just what it is to be this or that vixen. (Correia and Skiles 2019, 649–50) suggest that such generalized identity claims give rise to essence claims as follows: • Being F is what it is to be G in full iff: Gx x Fx • Being F is what it is to be G in part iff: there is some H such that Gx

x

Fx

Hx

(Fine 1994) discusses some alternative ways of expressing essence.2 Let us consider them briefly here. Perhaps the most natural way of expressing essentialist claims is by means of predicate modification. (Cf. (Wiggins 1976) on the form of de re necessity claims.) For any predicate P there is the predicate P , which we may pronounce “being essentially P ”. Using to construct complex predicates the first two essentialist claims above can then be expressed as follows: • ( H) s • ( ( xy. x + y = 0

x 2 = ( 1)

y 2 = ( 1))) i ( i)

If we allow ourselves higher-order logic the third essentialist claim can be expressed as follows: • ( ( X . y (Xy

Ry)) H

(We may attempt the following pronunciation: “to be human is to be such that it is essential to one’s being so, that if one is so, then one is rational”) None of the above ways of expressing essentialist claim reify essences; the final, “algebraic”, way of expressing essentialist claims does. On this algebraic conception an entity is simply related to a collection of entities that constitute its essence.3 There are two reasons for being interested in the algebraic conception. Consider again the fourth essentialist claim: “It is essential to disjunction that for any propositions p, q one can infer p q from both p and q”. In a higher-order framework one could write down something like: •

p q ((p

p

q)

(q

p

q))

But this is problematic. First, given the connection between essence and dependence it makes disjunction depend on , , and . Secondly, it fails to respect the idea that what is essential to is a rule. If we avail ourselves of an algebraic conception of essence could simply be assigned as its essence the rules governing . A reason for taking the essence of the logical operations to be rules is discussed by (Fine 1994, 58–9). Suppose that Socrates’s essence is “logically inert” in that it does not contain any logical operations; perhaps Socrates is essentially rational (Rs) and essentially an animal (As) but that it is not part of Socrates’s essence that he is rational and he is animal (Rs As). It should nevertheless follow from the essence of Socrates together with the essence of conjunction that Socrates is rational and an animal. Suppose, however, that the essence of (q p q)). By taking the essence of conjunction was a proposition—perhaps, p q (p Socrates together with the essence of conjunction, we get the following joint essence:

243

Jon Erling Litland

[R(s), A(s), p q(p (q p q))]. But unless we can apply the rules of Universal Instantiation and Modus Ponens we cannot get from this to R (s) A (s). By allowing the essence of to be the rule, from “p, q one can infer p q” the problem is solved. Such a rulebased conception of essence is worked out in detail in (Correia 2020; 2012). This concludes our discussion of different ways of formalizing essentialist claims. We now turn our attention to the proof-theory of the logic of essence, working for the remainder of this chapter with Fine’s sentential operator formulation of essentialist claims.

16.2

The Proof Theory of Essence

While LE is a logic of consequential essence we still want to insist that if a proposition is in the essence of an object a then that proposition only contains objects pertaining to the nature of a. It is therefore important that we can make sense of being a constituent of a proposition. Let E be a formula or a predicate and let x1, …, xm be the free variables in E, and let P1, …, Pn be the rigid predicate symbols in E. We then use y E as an abbreviation for y = x1 … y = xm P1 (y) … Pn (y). We use E for y. y E. (Intuitively, “being a Gx). constituent of E”.) Finally, we write F G as a shorthand for x(Fx We can now introduce the axioms of LE.4 We begin with some rules that have close modal analogues. Recall that the language of quantified modal logic extends the language of classical first-order logic by adding a sentential operator (for “necessarily, …”). A normal modal logic L is any extension of classical first-order logic that contains all instances of the Kripke B) ( A B) and is closed under the rule of necessitation: if A is a axiom (A theorem of L, then so is A. The essentialist analogues of these principles are: • Some proof system for classical first-order logic. B) ( FA • K F (A F B) • RN If A is a theorem of LE, then A A is a theorem. The K-axiom is the obvious analogue of the similar modal principle. Stated in English it amounts to this. Suppose that it lies in the natures of the objects that F that if A then B; suppose, further, that it lies in the nature of the objects that F that A; then it lies in the nature of the objects that F that B. This is clearly correct. Clearly, every theorem of the logic of essence should be true in virtue of the natures of something—but which things? RN ensures that if A is a theorem of LE then A is true in virtue of the constituents of A; one might say that if A is logically true, then A is true in virtue of the natures of its own constituents. It is RN together with K that makes LE a logic of consequential essence. Let 0 be the predicate x. x x and suppose F is a classical validity without constants and rigid predicates then it follows from RN that 0 F ; for instance, 0 x (x = x). (More generally, if it is essential to x that A and B is a logical consequence of A, then B is essential to x provided that all the rigid predicates and constants occurring in B also occur in A.) However, the requirement that F be without constants and rigid predicates is required. The proposition that Socrates is identical to Socrates—s = s—is not true in virtue of no objects whatsoever. Rather, what RN gives us is just that it is essential to Socrates that Socrates is identical to Socrates— s s = s. More generally, we get that x x x = x. A higher-order logic like Ditter’s allows us to draw finer distinctions. First, in Ditter’s system x (x = x) is not true in virtue of no objects; rather, x (x = x) is true in virtue of the nature of identity and universality.

244

Logic of Essence

Secondly, while it is essential to Socrates together with identity that he is self-identical, it need not be essential to Socrates by himself that he is self-identical. Thirdly, in LE essences are closed under conjunction in the sense that if it is essential to x that A and it is essential x that B then it is essential to x that A B. This means that LE cannot accommodate the possibility of logically inert essences noted above. A higher-order logic of essence can do this. This is also an advantage the higher-order approach has over the generalized identity approach of (Correia and Skiles 2019). For them something gets to be part of the essence of an item only by being part of a conjunction that is in the essence of that item; this, too, rules out logically inert essences. Even if one wants to work with a consequential conception of essence, a potential problem with RN comes from its interaction with classical logic. The following—known as Pierce’s Law—is a classical validity

((p

q)

p)

p)

Importantly, this principle is not intuitionistically valid. Somebody might hold the view that Pierce’s Law, while valid, is not true in virtue of the nature of the conditional and universality, but that in addition the natures of negation and disjunction are needed. (Pierce’s Law is derivable from Excluded Middle, and one might hold that Excluded Middle is true in virtue of the natures of disjunction and negation.) By working with a rule of proof like RN one cannot capture this distinction. Here is a possible solution. Instead of having a rule of proof that only checks whether A is a theorem, one can have a rule that is sensitive to the derivation of the theorem. The rule might take the form: If D is a derivation of A from axioms of our logic of essence, let R0 , R1, …, Rm be the rigid predicates in D and let c1, …, c n be the constants and variables occurring in D . Then let G be the predicate x.(R0 x … Rm x x = c1 … cn = x). The rule then says that one can continue the derivation D by concluding G A. Another consequence of working with classical logic and RN is that the following principles come out correct. (Here 1 is the universal predicate x. x = x) • •

x

x

1

x

yx = y 1 yx = y

Or in words: • For each thing, it lies in its nature that it is something • It lies in the nature of all the things that for each thing it lies in the nature of all things that that thing is something It is natural to gloss these principles as ruling out contingentism, the view that some things contingently exist. (For more about contingentism and necessitism, see (Williamson 2013).) However, this would be too quick. Fine is explicit that he understands the quantifiers possibilistically, as ranging over “all possible objects and not just the actual objects” (Fine 1995b, 244). Understood possibilistically, x x yx = y does not mean that every object exists necessarily. (Fine proposes that existence be expressed by an existence predicate rather than by the quantifiers.) It is of course a good question whether an actualist can give an adequate account of possibilistic quantification;5 and it would be interesting to develop a logic of essence with actualist quantification. We refer the reader to (Teitel 2019) and (Werner 2021) for

245

Jon Erling Litland

extended discussion of this issue; see also Correia’s Chapter 8, “Non-Modal Conceptions of Essence”.) We next come to the two rigidity axioms: • Rigidity Px • N-Rigidity ¬Px

P Px

P, x

¬ Px

Once we think of rigid predicates as standing for properties of the form being one of x0, x1, … then the Rigidity axiom is forced on us; it simply captures that for some objects and any object amongst them it is essential to those objects that the object is amongst them. When philosophers reason modally about plurals they typically also accept the principle that if an object is not one amongst some objects it is necessary that the object is not amongst them (Linnebo 2016). The essentialist analogue is N-rigidity, which can be derived from Rigidity using the 5-principle to be introduced below. But even if one rejects the 5-principle it seems reasonable to accept N-rigidity as a separate principle. Next, let us consider three axioms without modal counterparts. • Monotonicity F

G

(

FA

G A)

As we noted above LE allows us to reason about collective essences, about what is true in virtue of the natures of some objects. Monotonicity tells us that if it is true in virtue of the natures of the F s that A, and, moreover, all Fs are G then it is true in virtue of the natures of the Gs that A. Some care has to be taken with interpreting this. Monotonicity does not claim that if it lies in the nature of being F that A and the Fs are all Gs then it lies in the nature of being G that A. That claim is false (and on its face it is also a higher-order claim that cannot be expressed in LE). Suppose that (sadly) all humans are smokers and that (less sadly) it is part of the nature of being human that one is rational. Clearly, it does not follow that it is part of the nature of being a smoker that one is rational! However, Monotonicity does not entail this. Monotonicity only commits us to there being something in the natures of the entities who are in fact smokers in virtue of which they are rational. But there is: their being human. With that clarification aside, one should view Monotonicity as true by stipulation; accepting Monotonicity amounts to treating F as meaning “true in virtue of the natures of some of the objects that are F ”. On that reading Monotonicity is incontrovertible. This is not to say that there are not other notions of essence for which Monotonicity fails. For instance, it is not clear—though see (Zylstra 2019a)—that Monotonicity holds for the operator “it lies in the constitutive essence of the F s taken together that”. Or consider an example involving definition: one might, e.g., hold that i , i are fully defined by being additively inverse roots of 1. But it would be absurd to hold that i, i together with Socrates are fully defined by being such that i , i are additively inverse roots of 1. This shows that the operator “the objects that F are fully defined by being such that” does not satisfy Monotonicity. To explain the principles of Chaining and Localization we need some notation involving dependence. We write a b for b a. We write a F for x (Fx a x). (Intuitively: “some F depends on a.”) We write cF for ( y. y F). (Intuitively: “being something on which the F s depend”.) • Chaining

cF A

FA

246

Logic of Essence

Chaining says that what is essential to the objects on which the F s depend is essential to the F s. This principle must be interpreted carefully. The following looks like a counterexample. Smiles ontologically depend on mouths; and it is essential to mouths that they are parts of faces. However, it is not essential to smiles that they are part of faces. (Smiles aren’t even parts of faces, let alone essentially so.) This, however, is not a counterexample: it does not follow from Chaining that it is essential to smiles that they are parts of faces; what follows is just that it is essential to smiles that mouths are part of faces. Like with Monotonicity the best attitude towards Chaining treats it as true by stipulation. Accepting Chaining amounts to working with a mediate notion of essence. It is quite plausible that the metaphysically basic notion of essence is something like constitutive immediate essence (for discussion see (Miller 2022; Nutting, Caplan, and Tillman 2018). But, however that may be, it is no objection to LE that it is a logic of mediate consequential essence, and not of a different notion of essence. Finally, we come to the crucial principle of Localization. • Localization

FA

x A

x

F

Localization says that if A is true in virtue of the nature of the Fs then each constituent of A is something on which the F s depend. In the special case where there is only one F , Localization x y schematically captures the essentialist account of dependence. What y A x A says is that if it is essential to y that A and x is a constituent of A then y depends on x. However, in the case where there are many F s Localization appears questionable. The problem is that dependence is given an “individualistic spin”. In LE, x F —the claim that the F s depend on x—is defined as the claim that there is a particular F that depends on x. There are natural metaphysical views on which this individualistic localization claim fails. Suppose we hold that points in space are derived entities. We might, for instance, hold that points are to be understood as the points of intersection between lines. On such a view, a given point can, of course, be understood as the point of intersection of any number of lines, but there is no one line such that the point has to be understood as being on it. Take now the plurality of points that are on a given line l. Then it is plausible to hold that it is essential to these points taken together that they are on l. But it is not essential to any one of the points that they are on l. The points thus depend collectively on the line, but none of them individually depends on the line. What to make of this alleged counterexample to Localization? The problem is generated by the first-order nature of LE; this forces the dependence relation to be flanked by singular terms on either side. In a higher-order framework, one can allow relations between what rigid predicates stand for and objects. In such a framework there is no problem allowing collective dependence-relations. In fact, already in his account of essence in terms of Priorian modality (Correia 2007, 79) allows such collective dependence. Fine himself is aware of counterexamples like this (Fine 1995b, 243, 249–50). He responds that we should take Localization to amount to a stipulation about which notion of essence we have in mind. The alleged counterexamples to Localization above turn on cases where some objects are simultaneously defined, or understood in terms of each other. In such cases, should the individual essences of each object contain propositions about the others? Or should we take a more austere line and ban such “reciprocal essences,” and rather say that these objects fundamentally stand in certain relations? Accepting Localization means that we work with the first notion of essence. (See further (Fine 1994, 65–6).)

247

Jon Erling Litland

We next come to some further analogues of modal principles. • T. F A • 4. F A • 5. ¬ F A

A F

F A, A ,F

¬

where F is a quasi-rigid predicate. 6 F A , where F is a quasi-rigid predicate.

The essentialist T-axiom says that whatever is essentially so, is so. We will treat this as a truism (though see Correia’s Chapter 8, “Non-Modal Conceptions of Essence”.) Before considering the merits of 4 and 5 the restrictions in them require some explanation. First, a quasi-rigid predicate is defined as follows: every rigid predicate is quasi-rigid; any predicate of the form x. x = y is quasi-rigid; and if G1, …, Gn are quasi-rigid so is x.(G1 x … Gn x).7 Without the restriction to quasi-rigid predicates 4 would not be correct. Let F be the predicate “set most commonly used as an example in recent philosophy”. Then we have {Socrates}. However, it is not essential to {Socrates} that it is the set most F Socrates commonly used as an example in recent philosophy—that could easily have been {Plato} or some interesting pure set. Similar reasoning shows that the restriction to quasi-rigid F is required also in the 5-axiom. In the case of the 5-axiom we also have to add the constituents of A to the outer operator. It is not essential to Socrates that Plato is human; call the latter proposition Hp. With F being the predicate x. x = s we thus have ¬ F Hp. However, since the proposition ¬ F partly pertains to Plato, it is not essential to just Socrates that this proposition should be true. However, it seems reasonable to hold that it lies in the essence of Socrates and Plato together that it does not lie in the essence of Socrates that Plato is human. Both 4 and 5 are principles of “iterated essence”. (Dasgupta 2014) and (Glazier 2017) reject these as principles for constitutive essence. Here is how Dasgupta’s explains this: “To state the constitutive essence of Socrates might require stating that he is human. But it is odd to think that his essential core also includes the fact that it is part of his essential core that he is human. This latter, iterated claim of essence is something that follows from (or is grounded in) his essential core and not part of the essential core itself”. (Dasgupta 2014, 591) This might be correct as far as constitutive essence is concerned, but, familiarly, the response is that if this is so then this is not the notion of essence LE is intended to cover.8 The situation is clarified in Ditter’s higher-order logic of essence where the 4 axiom takes the following form: F A , F F A. This axiom schematically expresses the thought that the following is essential to the essentialist operator : for any quasi-rigid F and any p, if it is essential to things that are F that p, then it is essential to and F that it is essential to F that p. This axiom only commits us to it being part of the constitutive essences of Socrates and essence itself that Socrates is essentially human; but it leaves it open that it is not constitutively essential to Socrates that he is essentially human. (For more on iterated essence, see (Ditter, Forthcoming) Assuming, then, that iterated essentialist claims make sense, let us turn to the merits of 4 and 5. What 4 ensures is that objects do not lose essential features; what 5 ensures is that objects cannot gain essential features. It turns out that if an object cannot gain essential features, then it cannot lose them either—and so 5 entails 4 (but not vice versa). (We will not give the proof here.) Why should one think that objects cannot lose their essences? Considerations of origin essentialism put pressure on the view. Many have been attracted to the thesis that the origins of a material object are essential to it. For instance, one might hold that it is essential to a given table of wood that it be made of (roughly) the piece of wood it is in fact made of. (We

248

Logic of Essence

are prescinding from considerations of vagueness.) However, most hold that the “roughly” matters: the table could have been made from a slightly different piece of wood. This gives rise to a familiar modal problem. One is tempted to accept that necessarily, if the table is made of wood W , then the table could have been made of a slightly different piece of wood W1. Let us M (T , W1). However, one is also tempted to hold that the table symbolize this M (T , W ) could not have been made from a radically different piece of wood Wn ; thus ¬ M (T , Wn). However, given that for all i it is true that if the table of made of Wi , then it is possible that the n

table is made of Wi+1 one is led to conclude that … M (T , Wn). Considerations like these led (Salmon 1989) to reject the modal 4-principle (or equivalently: ). What do such cases mean for the logic of essence? One possibility is simply to give up the essentiality of origin, or to find a formulation that is immune to examples like the above. Another possibility is to take this to be a counterexample to the 4-axiom for essence. Alternatively, one might take these examples to show that one should distinguish between several distinct notions of essence (Salmon 2021). A more radical proposal has recently been suggested by (Yablo, n.d.). Yablo considers rejecting that what is essentially so is necessarily so. He can thus say that it is essential to the table that is made of exactly wood W ; that it is essential to the table that it is essential to the table that it is made of W ; and so on. But these claims about the iterated essence of the table do not preclude that the table could have been of W1 instead; it is just that when the table is made of W1 it has a different essence.9 (For more on the essentiality of origin, see Robertson Ishii’s Chapter 11 “Origin Essentialism”.) Somewhat simplified, here is how (Fine 1995b, 248–9) argues in favor of 5. Suppose x is the only F . And suppose that ¬ x A. The interesting case is when x depends on all the constituents of A. (Otherwise, Localization will take care of things for us.) Consider all the propositions p1, p1, … true in virtue of the nature of x. If it is true in virtue of the nature of x that p1, p1, … are all the propositions true in virtue of x then since A only has constituents “within the ken” of x it will follow that A is not one of these propositions. It thus lies in the nature of x that it is not essential to x that A. To reject 5 one has to reject that it lies in the nature of x that p1, p2 , … are all the propositions true in virtue of the nature of x. Given that we have accepted 4, the only issue is whether some proposition q that is not essential to x could become essential to x. Intuitionistic mathematics provides a possible case. According to the theory of free choice sequences there are infinite sequences of rational numbers that are generated by, for each entry, freely choosing what that entry should be. Such sequences are merely potentially infinite in the sense that there is no stage when the entire sequence is completed. (See (Linnebo and Shapiro 2019) for a detailed discussion of potential infinity.) We thus have the following situation. At any given stage n in the generation of a sequence s it is compatible with its nature that at entries m > n s can take distinct values k and l . (Given that the values for entry m have not been chosen yet, the choice of value is still free.) However, it is not essential to s that it can take distinct values at stage m: once stage m has been reached the value at m is frozen. Part of the interest in essence is the proposed reduction of necessity to essence. (See further Torza’s Chapter 7, “Modal Conceptions of Essence”, Correia’s Chapter 8, “Non-Modal Conceptions of Essence”, and Koslicki’s Chapter 21 “The ’Reduction’ of Necessity to NonModal Essence”.) Minimally, the reduction of necessity to essence requires that it is necessary that A iff A is true in virtue of the nature of all objects. If the reduction of necessity to essence holds, the logic of the operator 1 thus determines the logic of metaphysical necessity. What logic is that?

249

Jon Erling Litland

It is quite easy to see that thus A • 1A B) ( 1A • 1 (A • 1A 1 1A x 1A • 1 xA • If A is a theorem, so is

1

satisfies the axioms of S4 with the converse Barcan formula,

1B)

1A

However, as shown by (Ditter 2020) one does not get S5, in particular, one does not get the B-axiom, A→□1⋄1A. To obtain the B axiom Fine added the Domain Axiom. This is the principle: •

xRx

R

xRx, where R is a rigid predicate.

What the Domain Axiom says is that it is essential to all the objects there are that they are all the objects there are. The Domain Axiom is hardly trivial. Those who are drawn to potentialist views in the philosophy of mathematics may well think that it is false: for take any objects whatsoever, one might think that it is compatible with their natures that there is a further object. In particular, such philosophers might hold that no matter which objects we have it is compatible with their natures (taken together) that there be a set of exactly those objects (Linnebo 2010; Parsons 1983). Moreover, as shown by (Ditter 2020) if RN can be applied to the Domain Axiom serious trouble ensues. Say that an object is essentially independent if it is compatible with its essence that the only objects that exist are itself and the objects on which it depends. (This is a weak condition. An object can be essentially independent without its being possible for it to exist without other objects.) The Domain Axiom together with RN entails that there are no independent objects. (We prove this semantically below.) How should one respond to this conflict? One possibility, defended at some length in (Ditter 2022), is to reject the Domain Axiom and settle for S4 as the correct modal logic. (Ditter 2020) explores another approach: restrict the application of RN to the theorems of LE without the Domain Axiom. We briefly consider how this might be done below.

16.3

Semantics

In this brief section on the semantics of the logic of essence our focus will be on the semantics given in (Fine 2000). But some other approaches should be mentioned. (Correia 2000) develops a semantics for a propositional logic of essence. (Zylstra 2019b) has attempted to use the framework of truthmaker semantics (see e.g., Fine 2017a; 2017b) to develop a logic of essence. Since truthmaker semantics affords us a good notion of subject matter (Yablo 2018), such an approach is promising, but as argued by (Vogt 2021) Zylstra’s approach has serious problems. An advantage of analyzing essence in terms of generalized identity is that any semantics for the theory of generalized identity will eo ipso be a semantics for a theory of essence. Different views of propositional granularity will then give rise to more or less finegrained accounts of essence.10

250

Logic of Essence

Returning now to Fine’s semantics, a model M is a quadruple W , I ,

,

where:

i W is a non-empty set of worlds; ii I is a function taking each w ∈ W into a non-empty set Iw , the individuals contained in w; iii is a reflexive and transitive relation on w Iw such that if a Iw and a b then b Iw . iv is a function that takes each constant a to an element (a) of some Iw ; each rigid predicate R into a subset (R) of some Iw , and each pure n-place predicate F and world w into a set (F , w) of n-tuples of Iw v We assume that for each a w Iw there is a constant c such that (c) = a How should we understand the notion of an object a being contained in a world? (Fine 2000, 543) writes that “the presence of an object in a world is not taken to guarantee its existence but merely its possibility. Thus, each world will be taken to embody its own ‘view’ of which objects are possible and which are not”. To understand what is meant by a world containing the possibility of an object an analogy with modal logic may help. In modal logic we employ accessibility relations. is true at the world w iff is true at every accessible world w . It would, however, be a mistake to try to understand necessity in terms of truth-at-a-world and the accessibility relation; rather, the accessibility relation should be understood in terms of necessity. Thus: for w to be accessible from w is for any proposition that is necessarily true at w to be true at w . Similarly, we should understand what it is for a world w to contain an object a in essentialist terms. For the world w to contain the object a is for every proposition true in virtue of the nature of a to be true in w. If S is a set of objects in w Iw we define c (S)—the closure of S under —in the obvious way. The requirement that the individuals contained in a world are closed under the dependence relation is what ensures that the Chaining and Localization axioms are valid. If M is a model and E a sentence or closed predicate we define [E]M —the objectual content—of E as

{ (a1), …, (an)}

(P1)



(Pm)

Here a1, …, an are all the constants in E and P1, …, Pm are all the rigid predicate symbols in E. We say that E is defined at a world w iff [E]M Iw , that is, if the objectual content of E is contained in the objects of w. We now go on to define for sentences A and predicates H the notions w A—the truth of A at w—and Hw the extension of the predicate H at w. However, we only define these notions for those worlds w containing the objectual contents of A and H. This, in effect, gives us a three-valued semantics: if [A] is not contained in w then neither A nor ¬A will be true at w. The following way of thinking might be helpful. Think of a proposition as having both objectual content and intensional content. The objectual content is the set of objects the proposition is about; the intensional content is the set of worlds at which the proposition is true. We can then think of sentences as standing for propositions, and we can hold that propositions are only true or false at worlds that contain their objectual content. • • • • •

w w w w w

H (a1, …, an) iff (a1), …, (an) a = b iff (a) = (b) a b iff (a) (b) ¬ A iff not w A A B iff w A or w B

Hw , for any non-rigid predicate H

251

Jon Erling Litland

• w • w

F

xA (x) iff w A iff

• [A] c (Fw) and • v A, whenever Fv

(a) for each a

Iw

Iw

• Fw = (w, F) • Pw = (P), for rigid P • ( x . A (x))w = {a Iw : w A (a)} A sentence A is valid iff for every model M and every world w at which A is defined, w A. It will be instructive to consider how the semantics validates the principles of Necessitation, Localization, Chaining, and 5. Let us begin with Necessitation, which highlights the importance of restricting the definition of truth to the worlds at which the sentences are defined. A w. Suppose that A is valid. Let w be any world at which A is defined. We thus have [A] A w . A is defined there, and thus v A. Thus, w Take any world v such that Iv A A. Notice that if we had not restricted the definition of truth of A to worlds where A is defined, Necessitation would be unacceptable. For suppose A is true at every world, but undefined at some—A could for instance be s = s—and let w be a world where A is not defined. Then we do not have [A] Iw , and thus we do not have w A A. The proof of the Chaining principle goes as follows. Suppose w cF A. We first show that A is defined at every world containing c (Fw). From the assumption we know that A is defined at every world containing c (cF)w . But we can show c (cFw) = cc (Fw) = c (Fw). So let v be such that Iw c (Fw). By the above Iw c (cFw), and so v A. But then w F A. The Localization axiom falls out as follows. Suppose w a A. Then we know that FA [A] c (Fw) and (a) [A ]. By the definition of closure, there is some b Fw such that b a. x (Fx x a). What makes this proof go through is that the dependence And thus w relation is a binary relation on w Iw : the only way some objects can depend on an object, then, is for one of the former to depend on the latter. To model the kind of collective dependence discussed above, one would have to let the dependence relation be a relation between sets of objects of w Iw . Finally, let us consider how the 5-axiom is validated. Suppose that w ¬ F A where F is a rigid predicate. There are two cases: either [A] c (Fw) or not. If not, consider any world v with Iv Fw . Given that F is rigid Fw = Fv and thus [A] c (Fv). But then v F A and thus u [ A ] c ( F ) I F w w . Find a with u w such that u A. Let F , A ¬ F A. So let us suppose that v be any world such that Iv Fw A w . Since F is rigid we have Fw = Fu = Fv and thus Iv Fu. But then v ¬ F A, and since v was arbitrary we can conclude that w F , A ¬ F A. It is worth noting that it is in this case that we assume that essences cannot “grow”. To model failures of 5 we add a (reflexive, transitive) accessibility relation E to our models. (Intuitively, when wEv and Iv Fw not only does v contain all the objects that are F in w, anything which is essential to the Fs in w is true of those objects in v.) We retain the first clause for the essentialist operator but change the second clause to • v A, for all v such that wEv and Iv

Fw

A failure of 5 is given by three worlds w, u, v as in the above proof such that wEu , wEv , but not vEu. This corresponds to the world u’s respecting the essences the Fs have at w, but not respecting the essences the Fs have v.

252

Logic of Essence

Finally, let us show how RN and the Domain Axiom rules out independent objects. If RN can be applied to the Domain axiom, then our models have to satisfy the following condition: • For all w, v if Iw

Iv , then Iw = Iv

(The models are equivalence models.) Suppose now that a is an independent object; for simplicity, assume that a only depends on itself. Suppose further that a is not the only object y (y a). Since a is independent we must that exists. At the actual world w we then have w have w a ¬ y (y = a). Let u witness that w a ¬ y (y = a). Then Iu Iw . But then, since we y (y a). are looking at equivalence models, Iu = Iw contradicts that w Above we suggested that one might restrict RN to proofs that do not use the Domain Axiom. (Ditter 2020, 369–75) shows how to modify the semantics to accommodate this. Models are now quintuples W , I , , w, where w is a designated “actual” world of the model and where if Iw Iv then Iw = Iv . A sentence is true in a model if it is true in the actual world of the model; and a sentence is valid if it is true in all models. These models validate the Domain Axiom and the restricted version of RN.

16.4

Summary and Further Work

This chapter has only touched the surface of the logic of essence, the treatment of the semantics for the logic of essence having been particularly cursory, and we have said nothing about the intricate details of the completeness proof in (Fine 2000). It should, however, be clear that this is an area where important work remains to be done both philosophically and technically. Let us end the chapter listing some topics for further research: • Developing algebraic and/or predicational logics of essence, perhaps by extending Correia’s work on a rule-based conception of essence. • The existing proof-theories of the logic of essence are all presented as Hilbert systems—i.e., axiomatically. It would be of great interest to develop natural deduction and sequent calculi for logics of essence. • Notwithstanding the problems with Zylstra’s approach, truthmaker semantics remains a promising avenue for the logic of essence. • It remains an open question how the logic of essence should be developed if we insist on actualist quantification. • It would be interesting to explore the use of the logic of essence to develop set-theoretic potentialism. • In that connection it would be interesting to develop logics of essence in weaker background logics, such as intuitionistic logic.

16.5 • • • • • •

Related Topics

Torza’s Chapter 7, “Modal Conceptions of Essence” Correia’s Chapter 8, “Non-modal Conceptions of Essence” Robertson Ishii’s Chapter 11, “Origin Essentialism” Ritchie’s Chapter 15, “Language of Essence” Kovacs’s Chapter 20 “Essence, Grounding, and Explanation” Koslicki’s Chapter 21, “The ‘Reduction’ of Necessity to Non-Modal Essence”

253

Jon Erling Litland

Notes 1 A more radical option would be to accept the Russell property but weaken the logic. 2 For a yet further approach to expressing essentialist claims see ( Fine 2015; Fine 2016b). 3 The difference between the algebraic and the other approaches is over whether the fundamental way of expressing essentialist claims reify essence. If one opts for the sentential operator formulation one can introduce the essence of an item x as the collection of propositions that are true in virtue of the nature of x; and on the predicational approach one can take the essence of an item to be the collection of properties the item has essentially. 4 We are leaving out some axioms and rules that, while clearly valid on the intended interpretation, are of a more technical character and are there to allow the completeness theorem to be proved. 5 This question is especially pressing in a context where one wants to reduce necessity to essence; in that case one cannot simply take the notions of possibility and necessity for granted. It is thus unclear whether one of the standard accounts of possibilistic first-order quantification—to quantify over the possible F s is to possibly quantify over the things which are actually F (see for instance Fine 1985)—is available. In recent years the discussion has focused largely on giving an account of higherorder possibilist quantifiers ( Fritz and Goodman 2016; Fine 2016c; Williamson 2013). It is unclear how this dispute changes when it is framed within a logic of essence. 6 It is with iterated claims like this that taking the operator to be indexed by predicates pays off. Otherwise, we would have to help ourselves to plural quantification. 7 In Fine’s axiomatization, F is required to be rigid in the 4 and 5 axioms; the formulations with quasirigidity are derivable from Fine’s axiomatization. 8 ( Glazier 2017, 2885) does not just reject the logical principles 4 and 5. He argues for the stronger principle of the Inessentiality of Essence: • There do not exist s, t and A (where s and t are not necessarily distinct) such that

s

tA

9 ( Leech 2020) and ( Mackie 2020) have also questioned that what is essential is necessary. Leech’s criticisms are particularly interesting because they purport to show that even if one thinks of essence as a form of generalized identity, one cannot derive the necessity of essence. For a response see. ( Correia and Skiles 2021) 10 Theories like those of ( Dorr 2016; Goodman 2018) yield fine-grained accounts of essence; theories like those of ( Correia and Skiles 2019; Fine 2016a) yield a medium-grained account of essence.

References Correia, Fabrice. 2000. “Propositional Logic of Essence.” Journal of Philosophical Logic 29 (3): 295–313. Correia, Fabrice. 2006. “Generic Essence, Objectual Essence, and Modality.” Noûs 40 (4): 753–767. Correia, Fabrice. 2007. “(Finean) Essence and (Priorean) Modality.” Dialectica 61 (1): 63–84. 10.1111/ j.1746-8361.2006.01079.x. Correia, Fabrice. 2012. “On the Reduction of Necessity to Essence.” Philosophy and Phenomenological Research 84 (3): 639–653. Correia, Fabrice. 2020. “More on the Reduction of Necessity to Essence.” In Metaphysics, Meaning, and Modality: Themes from Kit Fine, edited by Mircea Dumitru, 265–282. Oxford University Press. Correia, Fabrice, and Alexander Skiles. 2019. “Grounding, Essence, and Identity.” Philosophy and Phenomenological Research 98 (3): 642–670. 10.1111/phpr.12468. Correia, Fabrice, and Alexander Skiles. 2021. “Essence, Modality, and Identity.” Mind 131 (524): 1279–1302. Dasgupta, Shamik. 2014. “The Possibility of Physicalism.” Journal of Philosophy 111 (9/10): 557–592. Ditter, Andreas. 2020. “The Reduction of Necessity to Essence.” Mind 129 (514): 351–380. Ditter, Andreas. 2022. “Essence and Necessity.” Journal of Philosophical Logic 51: 653–690. Ditter, Andreas. Forthcoming. “Are There Iterated Essentialist Truths?” Analysis. Dorr, Cian. 2016. “To Be F Is To Be G.” Edited by John Hawthorne and Jason Turner. Philosophical Perspectives 30: Metaphysics 30: 39–134. Fine, Kit. 1985. “Plantinga on the Reduction of Possibilist Discourse.” In Alvin Plantinga, 145–186. Springer. Fine, Kit. 1994. “Senses of Essence.” In, 53–73.

254

Logic of Essence Fine, Kit. 1995a. “Ontological Dependence.” Proceedings of the Aristotelian Society, New Series 95: 269–290. Fine, Kit. 1995b. “The Logic of Essence.” Journal of Philosophical Logic 24 (3): 241–273. Fine, Kit. 2000. “Semantics for the Logic of Essence.” Journal of Philosophical Logic 29 (6): 543–584. Fine, Kit. 2015. “Unified Foundations for Essence and Ground.” Journal of the American Philosophical Association 1 (2): 296–311. 10.1017/apa.2014.26. Fine, Kit. 2016a. “Angellic Content.” Journal of Philosophical Logic 45 (2): 199–226. Fine, Kit. 2016b. “Identity Criteria and Ground.” Philosophical Studies 173 (1): 1–19. Fine, Kit. 2016c. “Williamson on Fine on Prior on the Reduction of Possibilist Discourse.” Canadian Journal of Philosophy 46 (4–5): 548–570. Fine, Kit. 2017a. “A Theory of Truthmaker Content I: Conjunction, Disjunction and Negation.” Journal of Philosophical Logic 46 (6): 626–674. Fine, Kit. 2017b. “A Theory of Truthmaker Content II: Subject-Matter, Common Content, Remainder and Ground.” Journal of Philosophical Logic 46 (6): 675–702. Fritz, Peter, and Jeremy Goodman. 2016. “Higher-Order Contingentism, Part 1: Closure and Generation.” Journal of Philosophical Logic 45 (6): 645–695. Glazier, Martin. 2017. “Essentialist Explanation.” Philosophical Studies 174 (11): 2871–2889. Goodman, Jeremy. 2018. “Agglomerative Algebras.” Journal of Philosophical Logic 48 (4): 631–648. Koslicki, Kathrin. 2012. “Varieties of Ontological Dependence.” In Metaphysical Grounding: Understanding the Structure of Reality, edited by Fabrice Correia and Benjamin Schnieder, 186. Cambridge University Press. Leech, Jessica. 2020. “From Essence to Necessity Via Identity.” Mind 130 (519): 8870908. 10.1093/ mind/fzaa012. Linnebo, Øystein. 2010. “Pluralities and Sets.” Journal of Philosophy 107 (3). Linnebo, Øystein. 2014. “‘Just Is’-Statements as Generalized Identities.” Inquiry: An Interdisciplinary Journal of Philosophy 57 (4): 466–482. 10.1080/0020174x.2014.905037. Linnebo, Øystein. 2016. “Plurals and Modals.” Canadian Journal of Philosophy 46 (4–5): 654–676. Linnebo, Øystein, and Stewart Shapiro. 2019. “Actual and Potential Infinity.” Noûs 53 (1): 160–191. 10.1111/nous.12208. Mackie, Penelope. 2020. “Can Metaphysical Modality Be Based on Essence.” In Metaphysics, Meaning, and Modality: Themes from Kit Fine, edited by Mircea Dumitru, 247–264. Oxford University Press. Miller, Taylor-Grey. 2022. “On the Reduction of Constitutive to Consequential Essence.” Ergo: An Open Access Journal of Philosophy 9 (55). Nutting, Eileen S., Ben Caplan, and Chris Tillman. 2018. “Constitutive Essence and Partial Grounding.” Inquiry: An Interdisciplinary Journal of Philosophy 61 (2): 137–161. 10.1080/0020174X.2017.1392895. Parsons, Charles. 1983. “Sets and Modality.” In Mathematics in Philosophy, 298–341. Cornell University Press. Rayo, Agustin. 2013. The Construction of Logical Space. OUP Oxford. Salmon, Nathan. 1989. “The Logic of What Might Have Been.” Philosophical Review 98 (1): 3–34. Salmon, Nathan. 2021. “Modal Paradox II: Essence and Coherence.” Philosophical Studies 178 (10): 3237–3250. Teitel, Trevor. 2019. “Contingent Existence and the Reduction of Modality to Essence.” Mind 128 (509): 39–68. Vogt, Lisa. 2021. “Two Problems for Zylstra’s Truthmaker Semantics for Essence.” Inquiry: An Interdisciplinary Journal of Philosophy. Werner, Jonas. 2021. “Contingent Objects, Contingent Propositions, and Essentialism.” Mind 130 (520): 1283–1294. 10.1093/mind/fzaa080. Wiggins, David. 1976. “The De Re Must: A Note on the Logical Form of Essentialist Claims.” In, 285–313. Williamson, Timothy. 2013. Modal Logic as Metaphysics. Oxford University Press. Wilson, Jessica. 2020. “Essence and Dependence.” In Metaphysics, Meaning, and Modality: Themes from Kit Fine, edited by Mircea Dumitru. Oxford University Press. Yablo, Stephen. 2018. “Reply to Fine on Aboutness.” Philosophical Studies 175 (6): 1495–1512. Yablo, Stephen. n.d. “Everything Exists.” Zylstra, Justin. 2019a. “Collective Essence and Monotonicity.” Erkenntnis 84 (5): 1087–1101. Zylstra, Justin. 2019b. “Making Semantics for Essence.” Inquiry: An Interdisciplinary Journal of Philosophy 62 (8): 859–876. 10.1080/0020174x.2019.1570865.

255

PART 3

Applications Kathrin Koslicki and Michael J. Raven

Introduction Having approached philosophical discussions of essence from a historical point of view in Part I and set out prominent approaches and distinctions concerning essence in Part II, we are now ready in Part III of this Handbook to turn to applications of the notion of essence to specific questions, domains, and phenomena. Among the chapters included in Part III, some focus on particular aspects of the theoretical roles that are often associated with essence. In this vein, Chapter 19 asks how essence is related to the notions of identity, persistence, and individuation; Chapter 20 discusses the connections between essence, explanation, and grounding; Chapter 21 examines purported explanations of modality in terms of essence; while Chapter 27 considers the merits of appealing to essences as principles of unity. Other chapters included in Part III confront specific cases that have been thought, in one way or another, to spell trouble for those committed to essences. Thus, artifacts, artworks, and social kinds, discussed in Chapter 17, have been thought to present difficulties for essentialists due to their apparent mind-dependence and relativity to the interests of human agents. Chapter 18 confronts the question of how and whether essentialism about biological species may or may not be compatible with modern science, particularly evolutionary biology. Further challenges and controversies, which arise from attempts to apply essentialism to mental disorders, race, and the sex/gender distinction, are discussed in Chapters 23–25, respectively. The remaining chapters examine links between the notion of essence and other central philosophical concepts, in particular those of persons (Chapter 22), social justice (Chapter 26), and ethical value (Chapter 28). Chapter 17 Asya Passinsky’s “Artifacts, Artworks, and Social Objects” focuses on two main questions: first, whether, given their apparent mind-dependence and relativity to human interests, it is even reasonable to attribute essences to artifacts, artworks, and social objects (e.g., borders); and, secondly, assuming that an affirmative answer is given to the first question, how such essences might be characterized. The chapter begins by setting out some distinctions which are useful in homing in on the notion of essence that is relevant to the remainder of the discussion. Passinsky then turns to a critical examination of prominent responses that have been given to the two main philosophical questions concerning the essences of artifacts, artworks, and social objects addressed in the chapter. In particular,

DOI: 10.4324/9781003008750-20

Kathrin Koslicki and Michael J. Raven

proponents of a functionalist-intentionalist approach hold that the essences of artifacts or social objects can be functionally characterized by reference to the (individual or collective) intentions of agents who are responsible for their creation. While artworks, as Passinsky notes, are generally assumed to be a kind of artifact, the relationship between the categories of artifact and social object is less straightforward and is open to different construals depending on the approach at issue. Chapter 18 Ingo Brigandt’s “Biological Species” begins with a primer on biological systematics, before turning to the question of whether an individual is essentially a member of its species, and whether a species essentially belongs to its higher taxa. The position that biological species are associated with kind-essences came to be challenged starting in the 1970s, giving rise to the still dominant view that species are individuals. Alternative conceptions of species, which construe species as homeostatic property clusters, appear to make room for a revised notion of species-essence, as do recent attempts to rehabilitate neoAristotelian approaches to biological species. The prevailing sentiment among biologists, philosophers of biology, and philosophers of science more broadly, however, remains antiessentialist and debates concerning biological species and natural kinds are often carried out without invoking a notion of essence. Chapter 19 Maria Scarpati’s “Identity, Persistence, and Individuation” focuses on the question of how puzzles concerning a thing’s identity, persistence, and individuation are impacted by different sorts of essentialist commitments. Among the puzzles discussed by Scarpati are, first, the question of whether a single region of spacetime can be occupied by numerically distinct but spatiotemporally coincident objects (e.g., a statue and the matter constituting it); and, secondly, the question of what, if anything, grounds the modal differences between such apparently numerically distinct but spatiotemporally coincident objects. The chapter closes by exploring the distinction between essential features and individual essences, particularly as it relates to questions concerning individuation. Chapter 20 David Kovacs’ “Essence, Grounding, and Explanation” sets out four different ways in which essence might be taken to relate to the notion of grounding or metaphysical explanation, i.e., the type of connection that is often expressed by means of non-causal “in virtue of” or “because”-claims: (i) Unity: essence and grounding belong to a unified set of explanatory concepts; (ii) Supplementation: essence and grounding both contribute in their own way to a distinctive type of explanation; (iii) Independence: essence is a sui generis notion with no straightforward conceptual links to grounding; and (iv) Irrelevance: essence is not an explanatory notion at all and therefore should not be viewed as connected to explanatory notions like grounding. As Kovacs brings out in the chapter, a wide variety of attitudes are represented in the literature even among philosophers who are sympathetic to the recent “hyperintensional” turn in metaphysics when it comes to the question of how the notions of essence and grounding interact with one another. Chapter 21 Kathrin Koslicki’s “The ‘Reduction’ of Necessity to Non-Modal Essence” examines the non-modalist’s program of explaining metaphysical modality in terms of essence. The chapter proceeds by considering a range of different explanatory connections and strategies which raise the question of how the non-modalist’s program of explaining metaphysical modality in terms of essence is best carried out. In each case, in order to derive a metaphysically necessary truth from a statement of essence, it turns out to be necessary to bring in additional facts concerning the essences of related entities beyond those that are explicitly appealed to in the initial essentialist premise-set. Since no logical “one-size-fits-all” strategy appears to be available, a successful execution of the non-modalist’s program

258

Applications

requires a non-logical engagement with specific cases to make good on the promise that metaphysically necessary truths can, in all cases, be derived from statements of essence. Chapter 22 Annina Loets’ “Persons” highlights three Lockian ideas concerning the question of what persons are: that persons are the sorts of beings (i) that have certain distinctive cognitive capacities: (ii) that are, at least in principle, accountable for their actions; and (iii) that are identical from one time to another when certain broadly psychological conditions hold. As the chapter brings out, the literature on persons presents us with a complex and nuanced array of choice points when it comes to the question of how each Lockian thesis should be understood in isolation or in combination with one another. In the end, it remains an open question whether an appeal to personhood is in fact required in order to debate philosophically significant questions concerning someone’s moral standing or the ability to survive death. Chapter 23 Danielle Brown’s “Psychiatric Kinds” addresses the question of whether natural kind essences can be ascribed to psychiatric kinds. Homeostatic property cluster approaches to psychiatric kinds appear to run into trouble when no single underlying causal mechanism can be found which unifies clusters of symptoms that are associated with a given mental disorder. Psychiatric kinds furthermore seem to give rise to the sorts of feedback mechanisms characteristic of interactive kinds. Finally, what is counted as a mental disorder can vary from one time to another depending on the social or moral values that are prevalent in a given context. More recent approaches propose to carve out a space for psychiatric kinds by explicitly acknowledging and offering means to adjudicate the role of human values in scientific classifications. Chapter 24 Ron Mallon’s “Race” discusses a particular version of racial essentialism according to which human races exist and all and only members of a race share simple, intrinsic, and natural properties which play the roles associated with essences. Mallon explains why this account is widely regarded as mistaken and objectionable on other grounds. As Mallon notes, however, the mere rejection of this particular account is not by itself sufficient to block the tendency to generalize in pernicious ways about members of racial groups. The chapter closes by pointing to other strategies which might contribute towards achieving the normative aims motivating those who are opposed to racial essentialism. Chapter 25 Esther Rosario’s “Sex and Gender” surveys essentialist and anti-essentialist theories of sex and gender. In particular, the chapter considers externalist, internalist, and contextual approaches to sex and gender and examines their connections to key debates concerning the sex/gender distinction. Rosario isolates three problems to which extant theories of sex and gender are susceptible: the Inclusion Problem, the Definition Problem, and the Exclusion Problem. Lastly, the chapter highlights ways in which the distinction between essentialist and anti-essentialist accounts of sex and gender can give rise to unclarities. Chapter 26 Natalie Stoljar’s “Social Justice” considers five areas of overlap between issues of social justice and philosophical conceptions of essence. First, do the social groups assumed in the definition of oppression (genders, races, etc.) have essences? Second, for the purposes of feminism, anti-racism or other political action, is it necessary to posit essences that unify the respective social groups? Third, could gender, race, sexual orientation, etc. be essential properties of the individual members of social groups? Fourth, how do false beliefs in essences sustain oppression and undermine social justice? And fifth, can the nature of properties like genders or races be explicated using the definitional model of essence? As the chapter brings out, opinions are divided as to whether essences are helpful or needed in order to answer questions of importance for social justice.

259

Kathrin Koslicki and Michael J. Raven

Chapter 27 Charlotte Witt’s “Unity” develops the idea that essences unify composite individuals beginning with Aristotle’s theory and tracing its re-emergence in contemporary metaphysics. According to this idea, the job of essences—how they make a living—is that they unify the parts of an organized whole, an individual that is not identical to the sum of its parts. The unifying role of essence is, arguably, prior to other roles ascribed to essence in grounding an individual’s identity, persistence, or kind-membership, since the latter presuppose the existence of an individual whose parts have already been unified. Recently the idea of essence or form as unifying has also been applied to human social kinds, like races and genders. In this vein, Witt’s own “uniessentialism” proposes that gender serves as a principle of normative unity which ties together the norms to which social individuals are responsive. Chapter 28 Julie Tannenbaum and Stavroula Glezakos’ “Ethical Value” provides an overview of some important ethical positions that rest on claims about essence. As the chapter brings out, the nature of essence of human beings plays a central role in Aristotelian virtue theory as well as in debates about the morality of enhancement. By contrast, Buddhist ethics adopts an anti-essentialist perspective, not only when it comes to human beings but with regard to individuals more broadly. Philosophical debates about euthanasia, abortion, and moral responsibility appeal to the essence of killing (as opposed to letting die) as well as the essence of a person over time. Some discussions of sexual ethics involve claims about the nature of sexual activity. Lastly, claims about the essence of disability and gender have been used both to oppress people, and to motivate arguments against oppression and marginalization.

260

17 ARTIFACTS, ARTWORKS, AND SOCIAL OBJECTS Asya Passinsky

17.1

Introduction

Artifacts, artworks, and social objects are familiar to us from everyday life. Artifacts include practical items such as tables, chairs, screwdrivers, can openers, and laptops, as well as artworks such as paintings, sculptures, novels, and musical works.1 Social objects include social and institutional things such as dollars, borders, states, corporations, and universities. Although we are all familiar with such entities, it is far from clear what their nature or essence consists in and whether they even have a real nature or essence. The aim of this chapter is to survey and critically examine various positions on these two central philosophical issues concerning essence and artifacts, artworks, and social objects. Since there are many different notions of essence in the philosophical literature, it is important to clarify at the outset the notion that I will be invoking in this chapter. This is the contemporary neo-Aristotelian one, well known from the work of philosophers such as Kit Fine (1994, 1995a, 1995b), E.J. Lowe (2008, 2018), and Kathrin Koslicki (2012, 2018). NeoAristotelians hold that where x is an object such as Socrates or the Eiffel Tower, a specification of x’s essence provides an answer to the distinctively metaphysical question “What is x?”. Similarly, where F is a way of being such as knowing or being conscious, a specification of F’s essence provides an answer to the distinctively metaphysical question “What is it to F?”. Furthermore, neo-Aristotelians hold that this notion of essence cannot be analyzed in terms of the notion of metaphysical modality,2 and they instead take the notion to be primitive (at least for the purposes of present theorizing). Finally, neo-Aristotelians maintain that the notion of essence is intimately tied to the idea of real definition. A real definition is a definition of a worldly item (e.g., the kind water) rather than a definition of a linguistic item (e.g., the word “water”). It is thought that in specifying the essence of a given item, we are thereby providing a real definition of that item. It is important to also mention several claims that neo-Aristotelians are not committed to. First, neo-Aristotelians are not committed to substantive claims about the sorts of properties that can figure in an item’s essence. Thus, they do not maintain that these properties must be intrinsic, microphysical, simple, or discoverable by science. In this regard, neo-Aristotelian essentialism differs markedly from some other brands of essentialism, including essentialism about natural kinds in the philosophy of science literature and essentialism about social

DOI: 10.4324/9781003008750-21

261

Asya Passinsky

categories in the social theory literature.3 This is pertinent because it is generally agreed that artifacts, artworks, and social objects do not have essences which are intrinsic, microphysical, simple, or discoverable by science. Second, while some neo-Aristotelians maintain that essentialist truths obtain independently of us and our practices, it is not part of the neoAristotelian conception of essence that this be so. The conception itself is compatible with a constructivist view according to which essentialist truths are constructed by human interests, intentions, practices, or the like.4 This is relevant because some prominent views of artifacts and social objects have it that the natures of these entities are in some sense constructed by us. There are two distinctions pertaining to essence which bear on our discussion. The first is the distinction between individual and kind essence.5 Individual essence concerns the essence of individuals or particulars, such as the table in my apartment, Michelangelo’s David, or the dollar bill in my wallet. Kind essence concerns the essence of kinds, such as table, statue, or dollar bill. The second distinction is between objectual and predicational (or generic) essence.6 A statement of objectual essence aims to answer a question of the form “What is x?”, such as “What is this table?”. A statement of predicational (or generic) essence aims to answer a question of the form “What is it to F?”, such as “What is it to be a table?”. While these two distinctions are closely related, they are not the same. For arguably, kinds have both an objectual and a predicational essence. For example, it may be thought that part of the objectual essence of the kind table is that this kind is essentially an artifactual kind; and it may be thought that part of the predicational essence of the kind table is that being a table essentially involves having been made for a certain purpose. While there are many interesting questions concerning the individual essence of particular artifacts and social objects—for example, questions about the essentiality of their origins—my focus in this chapter will be on questions concerning the essence of the kinds artifact and social object as well as questions concerning the essence of kinds of artifacts and social objects. Moreover, I will specifically focus on the predicational essence of these kinds. In the next section, §2, I take up the question of whether the kinds artifact and social object—as well as kinds of artifacts and social objects—have essences. In the subsequent two sections, I assume an affirmative answer to this question and inquire into what these essences may be: §3 addresses the questions “What is it to be an artifact?” and “What is it to be an artifact of kind K?”, while §4 addresses the questions “What is it to be a social object?” and “What is it to be a social object of kind K?”. It is worth noting that some of the authors whose views I will be discussing in these two sections do not formulate their own views in terms of a neo-Aristotelian notion of essence—or in essentialist terms at all. However, since the commitments of neo-Aristotelian essentialism are so minimal, I take it that my essentialist re-construal of their views is sufficiently faithful to the original versions. I conclude, in §5, by briefly considering the relation between artifacts and social objects.

17.2 Essentialism Essentialism about the kind artifact, the kind social object, and kinds of artifacts and social objects maintains that these kinds have essences. Since my focus in this chapter is on the predicational essence of such kinds, I will be specifically concerned with the view that these kinds have non-trivial predicational essences, i.e., that for any such kind K, there is at least one non-trivial essentialist truth of the form “It is essential to x’s being a K that p”.7 Thus, our target essentialist doctrine about artifacts and social objects maintains that (i) there is at least one non-trivial truth of the form “It is essential to x’s being an artifact that p”; (ii) there is at least one non-trivial truth of the form “It is essential to x’s being a social object that p”; and

262

Artifacts, Artworks, and Social Objects

(iii) for any artifactual or social object kind K, there is at least one non-trivial truth of the form “It is essential to x’s being a K that p”. The most important challenge to this essentialist doctrine stems from the mind-dependent character of artifacts and social objects. As we shall see, prevailing theories of artifacts and social objects agree that these kinds are mind-dependent, at least in the sense that their existence or the existence of their instances depends upon the existence of minds. While different authors understand the relevant dependence in different ways, it is widely held that the dependence is constitutive as opposed to causal,8 and necessary as opposed to contingent. If being suitably mind-independent is a criterion for being real,9 then this strong form of mind-dependence arguably threatens the reality of the kinds in question. And if only real or natural kinds have essences because only such kinds are associated with the sorts of properties that can constitute an essence,10 then the compromised reality of these kinds in turn compromises essentialism.11 Muhammad Ali Khalidi has challenged the first step of this argument against essentialism (Khalidi 2015: 109–10).12 He argues that mind-independence is not a plausible criterion for realism because there are kinds which are mind-dependent in the relevant sense but are nevertheless real. For example, psychological states such as beliefs, desires, and pains necessarily depend upon the existence of minds in a constitutive way, but they are nevertheless real. However, it may be countered that artifacts and social objects exhibit a stronger form of mind-dependence than psychological states, at least according to some prominent views. Thus, some authors maintain that an object x belongs to an artifactual kind K at least partly in virtue of the fact that x’s maker intended it to be a K.13 And some authors maintain that an object x belongs to a social object kind K at least partly in virtue of the fact that the relevant community takes it to be a K.14 In contrast, it is not the case that a given psychological state x is a belief partly in virtue of being taken to be a belief or partly in virtue of having been intended to be a belief. Whether this stronger form of minddependence impugns the reality of artifacts and social objects is an open question. On the one hand, it may be thought that this form of mind-dependence impugns the objectivity of artifacts and social objects,15 and that being objective is a necessary condition for being real or fully real.16 On the other hand, it may be argued that psychological states are themselves real, and that an entity’s reality cannot be compromised by dependence upon something which is itself real.17 The second step of the argument against essentialism may also be challenged on two grounds. First, recall that neo-Aristotelians do not impose any substantive constraints on the sorts of properties that can figure in an item’s essence. Thus, the fact that a given kind is not associated with certain special sorts of properties—for example, intrinsic, microphysical, simple properties which are discoverable by science—does not preclude it from having an essence. Second, there seem to be non-trivial predicational essentialist truths even about unnatural or unreal kinds. For example, consider the kind witch. This is an unreal kind if anything is. Yet arguably, it is essential to x’s being a witch that x has evil magical powers. The existence of this non-trivial predicational essentialist truth about witch ensures that this kind has an essence in the relevant sense, given our earlier characterization of essentialism.18 A similar argument could be advanced for other putatively unnatural or unreal kinds, such as phlogiston, wizard, unicorn, and so on. Of course, these examples do not establish that every unnatural or unreal kind has an essence. However, it would be oddly arbitrary if some unnatural or unreal kinds had essences whereas others did not. So, the examples do lend some support to the general claim that every unnatural or unreal kind has an essence.19

263

Asya Passinsky

17.3 The Essence of Artifacts If the kind artifact and artifactual kinds such as table have essences, then what are these essences? Risto Hilpinen nicely summarizes what I take to be the naïve or ordinary view of artifacts, writing that “an artifact may be defined as an object that has been intentionally made or produced for a certain purpose” (Hilpinen 2011: para. 1).20 Let us call this the “functional-intentionalist view of artifacts”. Since even philosophers who ultimately reject this view oftentimes take it as their starting point, it will be useful to consider the view in some detail. Given our neo-Aristotelian conception of essence, the functional-intentionalist view may be construed as providing an account of the full or complete essence of being an artifact: FUNCTIONAL-INTENTIONALIST VIEW OF ARTIFACTS:

It is essential to x’s being an artifact that it is an object which has been intentionally made for a certain purpose.

There are three central components to this view: creationism, intentionalism, and functionalism. Let me elaborate upon each of these in turn. Creationism says that artifacts are essentially made or created, as opposed to found or discovered in nature. At a minimum, making or creating involves bringing into existence a new object. For example, to make a screwdriver is to bring into existence a new object—viz., a screwdriver—which did not exist before. In typical cases of creation of ordinary material artifacts, the making process involves physically modifying or rearranging some pre-existing material. Thus, the process of making a screwdriver may involve attaching the blade to the handle, and the process of making a sculpture may involve chiseling a block of marble. More generally, the making process may be taken to involve some sort of “work” on preexisting material. This work may consist of physical modification or arrangement, nonphysical arrangement, indication or selection, appropriation, or the like.21 One advantage of this more general conception of the making process is that it can accommodate “readymades” and “found objects”. Thus, for example, Hilpinen argues that Marcel Duchamp made his famous Fountain by selecting and preparing a urinal for presentation in an art gallery (Hilpinen 1993: §6), and Lynne Baker argues that a wine rack can be made out of a conveniently shaped piece of driftwood by being brushed off and then being used as a wine rack (Baker 2007: ch. 3, n. 8).22 It is important to appreciate that according to such views, readymades and found objects are not literally ready-made or found; rather, these objects are created through minimal means. Another advantage of the more general conception of the making process is that it can accommodate abstract artifacts, such as musical works and fictional characters. Thus, for example, Jerrold Levinson argues that a composer makes a musical work by indicating a sound/performing means structure (Levinson 1980: §4), and Simon Evnine maintains that an author makes a fictional character by indicating a set of properties (Evnine 2016: §4.4). Intentionalism says that artifacts are essentially tied to the intentions of their makers. In particular, artifacts are essentially the products of intentional activity, as opposed to products of unintentional activity.23 This distinguishes artifacts from naturally occurring objects which are in some sense “made” by natural processes, such as a cliff formed through weathering and erosion. Moreover, artifacts are essentially the intended products of such intentional activity, as opposed to unintended byproducts of intentional activity.24 This distinguishes artifacts from unintended byproducts, such as pencil shards that are produced when sharpening a pencil or debris that is produced at a construction site.

264

Artifacts, Artworks, and Social Objects

Functionalism says that artifacts are essentially tied to intentions with a specific content, namely intentions to produce something which will serve a certain purpose or function. For example, screwdrivers are intentionally produced to turn screws, chairs are intentionally produced to be sat on, and clocks are intentionally produced to tell time. It is natural to suppose that any given artifact essentially has an “intended” or “proper” function in virtue of having been intentionally made to serve this purpose or function.25 Thus, a screwdriver has the proper function of turning screws, a chair has the proper function of being sat on, and a clock has the proper function of telling time. A natural corollary of intentionalism and functionalism is that the essence of artifactual kinds lies in their associated proper functions.26 For example, the essence of screwdrivers lies in the function of turning screws, the essence of chairs lies in the function of being sat on, and the essence of clocks lies in the function of telling time. Let us call this corollary the “functional-intentionalist view of artifactual kinds”. The functional-intentionalist view of artifactual kinds may be construed as providing an account of the full or complete essence of being an artifact of kind K: FUNCTIONAL-INTENTIONALIST VIEW OF ARTIFACTUAL KINDS:

It is essential to x’s being a member of artifactual kind K that it has proper function F in virtue of having been intentionally made to serve function F.

A well-known virtue of this view is that it correctly classifies malfunctioning artifacts. For example, consider a misshapen screwdriver that cannot turn screws. Intuitively, it is still a screwdriver despite being unable to perform the characteristic function of screwdrivers. The functional-intentionalist view can account for this because the misshapen screwdriver still has the proper function of turning screws in virtue of having been intentionally made for this purpose. Let me briefly comment on two further aspects of the functional-intentionalist view of artifacts and artifactual kinds. First, the view does not assume that artifacts can only be made by human beings. Thus, for example, if beavers intentionally construct dams for the purpose of creating ponds which protect them from predators, then beaver dams are artifacts according to the functional-intentionalist view.27 As Beth Preston notes, this aspect of the view fits well with the evidence we have on sophisticated animal cognition (Preston 2018: §1). Second, the functional-intentionalist view restricts artifacts to objects, or what metaphysicians sometimes call “things”. Yet it may be thought that there are artifacts which belong to other ontological categories. For example, Evnine argues that there are artifactual events, such as theatrical performances (Evnine 2016: ch. 7). If that’s right, then the functional-intentionalist view is unduly restrictive. However, this is not so much an objection to the view as a challenge to show how the view can be extended to entities of other ontological categories. Having presented the functional-intentionalist view of artifacts and artifactual kinds, let me now turn to consider some of the main challenges and alternatives to this view. I will begin with functionalism. One important challenge to functionalism is raised by Amie Thomasson, who argues that some artifacts are not intended to serve any function at all (Thomasson 2014: §4.2). Her examples include doodles, idly produced paper clip sculptures, and works of art that are intentionally created, but not with any particular purpose in mind. Insofar as these are genuine artifacts, they are counterexamples to the functional-intentionalist view of artifacts. Thomasson’s solution to this problem is to replace the requirement that artifacts have an intended function with the more general requirement that artifacts have “intended features”, where these features may be functional, structural, perceptible, or receptive and normative (Thomasson 2014: §§4.2–4.3). Thomasson’s view avoids the counterexamples

265

Asya Passinsky

because the artifacts in question are plausibly taken to have some intended features which are not functional. For example, an idly produced paper clip structure may have the intended feature of being composed of paper clips, and a work of art may have the intended feature of being regarded or treated as a work of art. There are also important objections to the functionalist component of the functionalintentionalist view of artifactual kinds. One objection, pressed by Paul Bloom, is that something can be a member of artifactual kind K without having been intentionally made to serve the characteristic function associated with K (Bloom 1996: 5–6). Bloom gives the example of a chair that is made only “for show”. This chair was not made with the intention that it be sat on, and yet it is still a chair. Another objection, pressed by Thomasson, is that some artifactual kinds—in particular, kinds of artworks—are not associated with any one characteristic function (Thomasson 2014: 48). For example, paintings can be intentionally made to serve myriad purposes, including decoration, documentation, self-expression, and political persuasion. But arguably, there is no single purpose which paintings are characteristically made to serve. Thomasson proposes an alternative intentionalist account of artifactual kinds which is meant to avoid these objections. Here is Thomasson’s formulation of the account: Necessarily, for all x and all artifactual kinds K, x is a K only if x is the product of a largely successful intention that (Kx), where one intends (Kx) only if one has a substantive concept of the nature of Ks that largely matches that of some group of prior makers of Ks (if there are any) and intends to realize that concept by imposing K-relevant features on the object. (Thomasson 2003b: 600) Construed in essentialist terms, the account holds that it is essential to x’s being a member of artifactual kind K that it is the product of a largely successful intention to make a K. In cases where there are prior makers of Ks, intending to make a K requires (i) having a substantive concept of K which largely matches that of some prior makers, and (ii) intending to make something which has K-relevant features (i.e., criterial features associated with the concept of K). In “prototype” cases where there are no prior makers of Ks, the first condition only requires that the maker have a substantive concept of K. In such cases, the maker determines the success conditions for her own act of creation. Thomasson’s view avoids both of the objections raised above. By allowing K-relevant features to be non-functional, the view can accommodate artifactual kinds that are not associated with any one characteristic function, such as paintings. And by allowing K-relevant features to constitute a cluster rather than a strict set of necessary and sufficient conditions, it can accommodate instances of artifactual kinds that were not intended to serve the characteristic function associated with the kind, such as a chair made “for show”. While Thomasson’s view of artifacts and artifactual kinds dispenses with functionalism, it upholds intentionalism insofar as it ties the essence of artifacts and artifactual kinds to the intentions of their makers.28 Kathrin Koslicki develops an important challenge to such intentionalist views of artifactual kinds (Koslicki 2018: §8.4.1, 2023).29 The challenge stems from the observation that the intentions of users can diverge from—and arguably override—the intentions of makers. Koslicki illustrates this possibility with the following case.30 Suppose that Alexander Graham Bell, the inventor of the telephone, intended his device to be an aid for the hearing-impaired. However, later users of the telephone regard it as a long-distance communication device and intend to use it accordingly. Arguably, the intentions of these later telephone users override the intentions of the inventor of the telephone, so that a telephone is essentially a device for long-distance communication rather

266

Artifacts, Artworks, and Social Objects

than an aid for the hearing-impaired. Yet accounts such as Thomasson’s, which hold that the original maker of an artifactual kind K stipulatively determines what the K-relevant features are, seem committed to saying that a telephone is essentially an aid for the hearing-impaired. Evnine proposes a plausible response to this challenge, namely that users can also be makers (Evnine 2022: 5–6). Specifically, by intentionally “counter-using” instances of an old artifactual kind K in a new way, users can take on the role of makers and create a new artifactual kind K’. Thus, in the case of the telephone, the later users intentionally “counter-use” instances of the original artifactual kind, thereby creating a new artifactual kind which we now call “telephone”. However, it’s unclear whether this line of response can adequately address a version of Koslicki’s challenge which pertains to individual artifacts rather than artifactual kinds. Consider the following case. A building is originally constructed to serve as a house. But over time, members of the community come to regard and use the building as a church. Arguably, after enough time has passed, the building comes to be a church. Yet intentionalist accounts such as Thomasson’s, which hold that it is essential to x’s being an artifact of kind K that its maker intended it to be a K, seem committed to denying that the building comes to be a church. Applying Evnine’s idea of users as makers, we may say that the users of the building create a new individual artifact—viz., a church—through their “counter-use” of the house. However, assuming that the house is not thereby destroyed, we are left with the counterintuitive result that there is both a church and a house standing in the same place at the same time.31 The natural alternative to an intentionalist view of artifactual kinds is a use-based view which ties the essence of artifactual kinds to the intentions or practices of users rather than makers.32 But such a use-based view faces challenges of its own. Perhaps the most serious problem concerns prototype production. Consider, for example, the production of the very first screwdriver. Since at the time of production there are not yet any users of screwdrivers, it is hard to see how a use-based view can account for the apparent fact that this prototype screwdriver is a screwdriver.33 Another problem, raised by Koslicki, concerns cases in which the users of an artifact are in some sense mistaken in their practice (Koslicki 2023: 226–33). Koslicki gives the example of so-called “amulets”. The users of these items regard them as having the function of warding off evil spirits, and they use them accordingly. A use-based view seems to be committed to saying that amulets are essentially for warding off evil spirits and the items in question are amulets because their users ascribe this function to them and use them for this purpose. Yet arguably, the users are mistaken in their classificatory practice and the items in question are more aptly classified as jewelry. A use-based view does not have the resources to secure this alternative classification. Despite their differences, both use-based views and intentionalist views uphold creationism insofar as they maintain that artifacts are genuinely created. This idea has also been challenged in the literature. One significant challenge comes from the literature on the ontology of music. A prominent view in this field is Platonism, which holds that musical works are abstract sound structures that exist eternally.34 Platonism entails that musical works are not literally created, since they are not brought into existence at any particular point in time. Given that musical works are artifacts, this constitutes a challenge to the idea that artifacts are essentially created as opposed to discovered. However, there is a forceful objection to Platonism, namely that it denies our deeply held belief that art is genuinely creative. Here is Levinson’s eloquent formulation of the objection: The whole tradition of art assumes art is creative in the strict sense, that it is a godlike activity in which the artist brings into being what did not exist beforehand—much as a demiurge forms a world out of inchoate matter. The notion that artists truly add to the

267

Asya Passinsky

world, in company with cake-bakers, house-builders, law-makers, and theoryconstructers, is surely a deep-rooted idea that merits preservation if at all possible. The suggestion that some artists, composers in particular, instead merely discover or select for attention entities they have no hand in creating is so contrary to this basic intuition regarding artists and their works that we have a strong prima facie reason to reject it if we can. (Levinson 1980: 8, emphasis in original) Platonists have attempted to reply to this objection, for instance by emphasizing the creative nature of discovery.35 But as Julian Dodd notes, many philosophers still treat this consequence of Platonism as a reductio of the view that musical works are abstract sound structures (Dodd 2008: 1119). Another challenge to creationism comes from the literature on material constitution. Consider a statue that is molded from a pre-existing lump of clay. Monists say that the statue is numerically identical to the lump of clay, whereas pluralists maintain that the statue is numerically distinct from the lump of clay. Monists may argue that when the statue is molded from the clay, there is no new object—viz., a statue—that comes into existence. Rather, an old object acquires a new property, viz., being a statue. Thus, the statue is not literally created from the lump of clay. And likewise for other ordinary material artifacts which are “made” from preexisting material. But there is a well-known problem with monism, namely the problem from Leibniz’s Law. According to Leibniz’s Law, a and b are identical only if they share all properties in common. But the statue and the lump of clay do not appear to share all properties. For instance, the lump of clay could continue to exist if rolled into the shape of a ball, whereas the statue could not exist in such circumstances; and the statue may be well-made while the lump of clay is not well-made. Monists have developed various strategies for dealing with this objection, and pluralists have criticized these strategies.36 Here it will suffice to note that one’s position on the monism-pluralism debate may lead one to reject the idea that artifacts are genuinely created.

17.4 The Essence of Social Objects Let us now turn to social objects. If the kind social object and social object kinds such as border have essences, then what are these essences? A prominent and prima facie plausible idea, which can be traced back to the work of John Searle (1995, 2010),37 is that social objects are a special kind of artifact—namely, artifacts which are created and maintained not through mere individual intentionality but through collective intentionality, specifically the collective acceptance of rules or principles.38 Let us call this the “collective acceptance view of social objects”. Construed in essentialist terms, this view provides the following account of the full or complete essence of being a social object: COLLECTIVE ACCEPTANCE VIEW OF SOCIAL OBJECTS:

It is essential to x’s being a social object that it is an object which has been intentionally made and maintained for a certain purpose, through the collective acceptance of rules or principles.

The correlative view of the full or complete essence of being a social object of kind K is as follows: COLLECTIVE ACCEPTANCE VIEW OF SOCIAL OBJECT KINDS:

It is essential to x’s being a member of social object kind K that it has proper function F in virtue of being collectively taken to have function F.

268

Artifacts, Artworks, and Social Objects

There are four central components to the collective acceptance view of social objects and social object kinds: creationism, intentionalism, functionalism, and collective representationalism. Creationism says that social objects are essentially made or created, as opposed to found or discovered in nature. As in the case of artifacts, creationists may wish to allow for “found” social objects such as river borders or tokens of seashell money, which are created with little or no physical manipulation of pre-existing material. Intentionalism says that social objects are essentially tied to the intentional states of members of the relevant community. In particular, they are essentially the intended products of intentional activity. Moreover, their continued existence essentially depends upon the intentional states of members of the relevant community. Functionalism says that social objects are essentially tied to intentional states with a specific content, namely states that involve the ascription of some purpose or function. Moreover, it says that social objects have a proper function in virtue of this ascription of function, and social object kinds are individuated by their associated proper functions. Finally, collective representationalism says that social objects are essentially created and maintained in a distinctively social manner, namely through a community’s collective acceptance of certain sorts of rules or principles which involve a concept of the relevant kind of social object. The precise form and content of these rules or principles is a matter of debate in the social ontology literature,39 as is the nature of collective intentionality.40 Having explicated the collective acceptance view of social objects and social object kinds, let us now consider some of the challenges and alternatives to this view. I will focus on what I take to be the three most important and distinctive challenges and on an alternative view of social objects which aims to avoid these challenges.41 The first challenge, raised by Thomasson, pertains to intentionalism (Thomasson 2003a). Thomasson argues that some social entities are unintended byproducts of intentional activity as opposed to intended products. Her examples include recessions and racism: we evidently do not intend to create recessions through our collective economic activity, nor do we intend to create racism through our collective attitudes, practices, and behaviors.42 It may be objected that recessions and racism are not social objects because they are not objects or things in the relevant sense. Rather, they are entities which belong to some other ontological category, such as event or activity. However, there are other examples of unintended byproducts which are clearly social objects in the relevant sense. Consider, for example, a boundary between the good and the bad parts of town which emerges gradually over time without anyone explicitly intending to create this boundary; or a corporation that is unintentionally created through the filing of the requisite paperwork. A second important challenge, raised by Francesco Guala, concerns functionalism as well as collective representationalism (Guala 2016: 167–71). Guala argues that in the case of institutional kinds such as money, the essential property which unifies the members of the kind is not some intended or proper function, but some actual or fulfilled function. For example, the essential property which unifies the members of the kind money is fulfilling the functions of being a medium of exchange, unit of account, and store of value. In support of this view, Guala points out that this is how money is standardly defined in economics textbooks (Guala 2016: 35). Guala further argues that given this alternative functionalist conception of institutional kinds, collective representationalism should be rejected because collective acceptance of Searlean constitutive rules or the like is not necessary for the fulfillment of a given function. For example, cigarettes may fulfill the characteristic functions of money in a prisoner of war camp without there being collective acceptance of rules or principles of the requisite sort. Finally, Åsa Burman presses another important objection to an idea which is at the heart of collective representationalism, namely that the creation and maintenance of social reality rests

269

Asya Passinsky

upon cooperation and consensus (Burman 2023: intro, chs. 1–2). Burman points out that while this may hold true of the simple and imaginary cases that oftentimes serve as paradigm examples in the social ontology literature, it is not true in a vast array of real-world cases which involve significant conflict and contestation. For example, gender and race do not appear to be created and maintained through harmonious collective acceptance of rules or principles which everyone agrees to, but rather through oppressive social practices that privilege some and disadvantage others.43 And the same may be said for many other social entities which clearly belong to the category of social objects. For example, many political borders have been created through war and subsequent proclamations by the victors. In such cases, genuine collective acceptance is arguably lacking. Instead, those with less power are simply compelled to go along with the will of those who have more power. Asya Passinsky develops an alternative to the collective acceptance view of social objects and social object kinds which aims to avoid these challenges (Passinsky 2021). The central idea is that social objects are essentially normative entities whose existence is partly a matter of the existence of certain kinds of norms, namely norms of conventional or political morality, legal norms, or prescribed or practiced social norms. Construed as a view of the full or complete essence of being a social object, this normative account says that it is essential to x’s being a social object that its existence is partly a matter of the existence of norms of conventional or political morality, legal norms, or prescribed or practiced social norms. The correlative view of the full or complete essence of being a social object of kind K says that it is essential to x’s being a member of social object kind K that its existence is partly a matter of the existence of moral, legal, or social norms N1, … , Nk (or a sufficient number thereof). Thus, the kind social object is understood in normative terms rather than functionalintentionalist terms, and kinds of social objects are individuated by their associated norms rather than their associated proper functions. Passinsky’s normative view of social objects and social object kinds avoids Thomasson’s challenge because moral, legal, and social norms can be created unintentionally as well as intentionally. For example, social norms can emerge gradually over time as certain behaviors come to be viewed as appropriate within a community, without anyone explicitly intending to create such norms. And legal norms can be created unintentionally through the unwitting exercise of legal powers. In cases where the relevant norms are created unintentionally, the corresponding social objects are created unintentionally. The view also avoids Burman’s and Guala’s objections to collective representationalism because the existence of moral, legal, and social norms need not be grounded in collective acceptance of rules or principles—or in agreement or consensus more generally. This is evident from the fact that throughout history, there have been many extant laws and social norms which were not collectively accepted by members of the relevant society, including laws against interracial marriage, sodomy laws, and gender norms. In cases where the existence of the relevant norms is not grounded in collective acceptance, the corresponding social objects are not created or maintained through collective acceptance. Finally, by essentially tying social object kinds to norms rather than functions, the view avoids taking a stand on whether the functions typically associated with kinds of social objects are best construed as proper functions or fulfilled functions.

17.5 Conclusion We have thus far considered various views on the essence of artifacts and social objects, respectively. A question which remains—and which has been underexplored in the existing literature—concerns the precise relation between artifacts and social objects: Are social

270

Artifacts, Artworks, and Social Objects

objects a kind of artifact? Or conversely, are artifacts a kind of social object? Or is neither a subkind of the other? Of course, one’s position on this question will depend on one’s preferred view of artifacts and social objects. But let me mention two salient alternatives here. The first combines the functional-intentionalist view of artifacts with the collective acceptance view of social objects. This leads to the view that social objects are a kind of artifact, namely artifacts which are created and maintained through the collective acceptance of rules or principles. The study of social objects then turns out to be a branch of the study of artifacts. The second alternative combines the functionalintentionalist view of artifacts with a normative view of social objects that allows for more kinds of normativity to ground the existence of these objects. This leads to the view that artifacts are a kind of social object, namely social objects whose existence is partly a matter of norms of use tied to proper function. The study of artifacts then turns out to be a branch of the study of social objects. It remains to be seen which of these approaches—if either—provides an adequate account of the relation between artifacts and social objects.44 Related topics: Race (Chapter 24); Sex and Gender (Chapter 25); Social Justice (Chapter 26); Social Construction (Chapter 31).

Notes 1 Artworks are standardly classified as a kind of artifact in both the literature on artifacts and the literature on artworks. See, e.g., Dickie (1983), Baker (2007: ch. 3), Levinson (2007), Thomasson (2014), and Evnine (2016: ch. 4). 2 See Fine (1994) for an influential argument against modal analyses of essence. 3 For a helpful discussion of essentialism about natural kinds, see Khalidi (2013: ch. 1). See also Tahko (this volume). For a helpful discussion of essentialism about social categories, see Mallon (2007). See also Mallon (this volume) on racial essentialism and Griffith (this volume) for a survey of different versions of essentialism and their applicability to socially constructed kinds. 4 Cf. Raven (2022: 134). 5 See, e.g., Witt (2011: 5–6). See also Marabello (this volume) on essences of individuals. 6 See Correia (2006) and Fine (2015). 7 The non-triviality constraint is meant to rule out essentialist truths such as “It is essential to x’s being a table that x is self-identical”, which does not even partially individuate being a table from any other ways of being. 8 See Elder (2007) for an opposing view according to which the dependence of artifacts upon minds is causal and not constitutive. 9 See Devitt (1991: §2.2) and Thomasson (2003b: §1) on mind-independence as a criterion for realism. See also Khalidi (2015: §4) and Mason (2016: §4, 2020) for discussion of the mindindependence criterion in relation to social kinds. 10 See Khalidi (2013: §1.3) for a discussion of the idea that only real or natural kinds have essences and an examination of the criteria that essential properties must meet. 11 See Griffith (this volume) for a related argument which challenges essentialism about socially constructed kinds on the grounds that these kinds are mind-dependent. 12 Cf. Mason (2016: §4, 2020). 13 See, e.g., Thomasson (2003b) and Evnine (2016: ch. 3). 14 See Searle (1995: ch. 2) on the self-referentiality of social concepts. Cf. Passinsky (2020) on the response-dependence of social objects. 15 See Passinsky (2020: §4) on the objectivity of social objects. 16 See Pettit (1991: 588–90) and Johnston (1993: 106) on the link between objectivity and reality. 17 See Khalidi (2015: §4) and Mason (2020) for further arguments in support of the view that the mind-dependence exhibited by social kinds does not compromise their reality. 18 This example suggests that essence precedes existence in some sense, since the kind witch has an essence despite not having any instances. Philosophers who are committed to the view that existence precedes essence may therefore have to reject this example.

271

Asya Passinsky 19 For further arguments in support of the view that social entities have essences, see Mason (2021: 3986–7) and Raven (2022: 135–7). See also Griffith (this volume) and Stoljar (this volume). 20 Similar definitions may be found in Baker (2004: 99) and Preston (2018: para. 4), though Preston herself rejects this view in favor of an alternative use-based approach. See Hilpinen (1992, 1993) and Baker (2004, 2007: ch. 3) for developments of the view. 21 See Hilpinen (1993: §4, §6), Baker (2007: ch. 3), and Evnine (2016: ch. 3). 22 See Korman (2015: 155, 2020: §3.3) for an opposing view. Korman argues that in such cases, no new object is brought into existence. Rather, an old object acquires a new property. 23 See, e.g., Thomasson (2003b: 592, 2009: §2) and Hilpinen (2011: §1). 24 See, e.g., Hilpinen (1993: 156, 2011) and Thomasson (2009: §2). 25 See Baker (2007: 51–3) on proper function. 26 See Baker (2007: ch. 3) for a development of this view. 27 This example is from Preston (2018: §1). 28 Thomasson’s view also remains committed to an intentionalist conception of artifact function. For an alternative non-intentionalist conception of artifact function, see Preston (1998). See also Preston (2009) for a helpful survey of theories of artifact function. 29 See Koslicki (2018: §§8.4.2–8.4.3) for further objections to what she calls “author-intention-based accounts” of artifactual kinds. 30 Koslicki borrows this example from Carrara and Vermaas (2009: 135). See Kornblith (2007) for another case involving carabiners. 31 See Koslicki (2023) for another objection to Evnine’s proposal. 32 See Preston (2013) for a use-based approach to artifacts. See also Koslicki (2023) for a discussion of Preston’s view. 33 The proponent of a use-based view could appeal to the capacities or dispositions of the prototype in cases where the prototype functions properly. However, such an appeal would not secure the desired classification in cases where the prototype malfunctions. See Koslicki (2023: 221–6) for further discussion. 34 See, e.g., Kivy (1983, 1987) and Dodd (2000, 2007). 35 See Kivy (1983: 112–9) and Dodd (2007: §5.4). 36 See Fine (2003) for an influential monist reply and pluralist rebuttal. See King (2006) for a reply to Fine. See also Scarpati (this volume) for a discussion of the monism-pluralism debate and essentialism. 37 Note that Searle formulates his theory of social reality as a theory of social and institutional facts, not social and institutional objects. While it is not entirely clear what his view of social objects is, he is plausibly construed as denying creationism both in the case of artifacts and in the case of social objects. Thus, he should not be regarded as a proponent of what I call the “collective acceptance view of social objects”. However, he may be regarded as endorsing a view in the ballpark, namely that it is essential to x’s being a social object that it is an object which has been intentionally assigned a new status with an associated function, through the collective acceptance of constitutive rules of the form “X counts as Y in context C” ( Searle 1995: ch. 2). 38 See also Thomasson (2003a, 2003b), though note that Thomasson rejects the idea that social objects must be the intended products of intentional activity (see below). Cf. Tuomela (2002) for another influential collective acceptance approach to social practices and social institutions. 39 Two prominent alternatives are those of Searle and Thomasson. Searle maintains that what is collectively accepted is constitutive rules of the form “X counts as Y in C”, where X is an object or class of objects, Y is a status with an associated function, and C is a context ( Searle 1995: ch. 2). Thomasson maintains that it is principles describing sufficient conditions for the existence of members of the relevant kind ( Thomasson 2003a). 40 The main debate here concerns whether collective intentionality is reducible to individual intentions plus mutual knowledge or belief. For a reductionist view, see Bratman (1999). For non-reductionist views, see Gilbert (1990) and Searle (1990). 41 Other prominent alternatives to a Searlean view include Guala’s (2016) rules-in-equilibrium approach and Ásta’s (2018) conferralist framework. However, it should be noted that neither of these authors develops a theory of social objects. Guala’s theory pertains to social institutions, while Ásta’s theory pertains to social properties of individuals. 42 Races and genders are further salient examples of social entities which are arguably unintended byproducts. See Mallon (this volume), Rosario (this volume), and Stoljar (this volume) for a discussion of these social entities.

272

Artifacts, Artworks, and Social Objects 43 See Haslanger (2000). 44 I would like to thank audiences at the University of Barcelona and the University of Neuchâtel, where versions of this material were presented. I would also like to thank Kathrin Koslicki and Mike Raven for helpful comments and discussion.

References Ásta. (2018) Categories We Live By: The Construction of Sex, Gender, Race, and Other Social Categories. New York: Oxford University Press. Baker, Lynne Rudder. (2004) “The Ontology of Artifacts”. Philosophical Explorations 7, 99–112. Baker, Lynne Rudder. (2007) The Metaphysics of Everyday Life: An Essay in Practical Realism. Cambridge: Cambridge University Press. Bloom, Paul. (1996) “Intention, History, and Artifact Concepts.” Cognition 60, 1–29. Bratman, Michael. (1999) Faces of Intention: Selected Essays on Intention and Agency. Cambridge: Cambridge University Press. Burman, Åsa. (2023) Nonideal Social Ontology: The Power View. New York: Oxford University Press. Carrara, Massimiliano and Vermaas, Pieter. (2009) “The Fine-Grained Metaphysics of Artifactual and Biological Functional Kinds.” Synthese 169, 125–143. Correia, Fabrice. (2006) “Generic Essence, Objectual Essence, and Modality.” Noûs 40, 753–767. Devitt, Michael. (1991) Realism and Truth. Second edition. Princeton: Princeton University Press. Dickie, George. (1983) “The New Institutional Theory of Art.” Proceedings of the 8th Wittgenstein Symposium 10, 57–64. Dodd, Julian. (2000) “Musical Works as Eternal Types.” British Journal of Aesthetics 40, 424–440. Dodd, Julian. (2007) Works of Music: An Essay in Ontology. Oxford: Oxford University Press. Dodd, Julian. (2008) “Musical Works: Ontology and Meta-Ontology.” Philosophy Compass 3, 1113–1134. Elder, Crawford L. (2007) “On the Place of Artifacts in Ontology.” In Eric Margolis and Stephen Laurence (eds.), Creations of the Mind: Theories of Artifacts and their Representation. New York: Oxford University Press (pp. 33–51). Evnine, Simon J. (2016) Making Objects and Events: A Hylomorphic Theory of Artifacts, Actions, and Organisms. Oxford: Oxford University Press. Evnine, Simon J. (2022) “The Historicity of Artifacts: Use and Counter-Use.” Metaphysics 5, 1–13. Fine, Kit. (1994) “Essence and Modality.” Philosophical Perspectives 8, 1–16. Fine, Kit. (1995a) “Ontological Dependence.” Proceedings of the Aristotelian Society 95, 269–290. Fine, Kit. (1995b) “Senses of Essence.” In Walter Sinnott-Armstrong (ed.), Modality, Morality and Belief: Essays in Honor of Ruth Barcan Marcus. Cambridge: Cambridge University Press (pp. 53–73). Fine, Kit. (2003) “The Non-Identity of a Material Thing and Its Matter.” Mind 112, 195–234. Fine, Kit. (2015) “Unified Foundations for Essence and Ground.” Journal of the American Philosophical Association 1, 296–311. Gilbert, Margaret. (1990) “Walking Together: A Paradigmatic Social Phenomenon.” Midwest Studies in Philosophy 15, 1–14. Guala, Francesco. (2016) Understanding Institutions: The Science and Philosophy of Living Together. Princeton: Princeton University Press. Haslanger, Sally. (2000) “Gender and Race: (What) Are They? (What) Do We Want Them To Be?” Noûs 34, 31–55. Hilpinen, Risto. (1992) “On Artifacts and Works of Art.” Theoria 58, 58–82. Hilpinen, Risto. (1993) “Authors and Artifacts.” Proceedings of the Aristotelian Society 93, 155–178. Hilpinen, Risto. (2011) “Artifact.” In Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Winter 2011 Edition). Johnston, Mark. (1993) “Objectivity Refigured: Pragmatism without Verificationism.” In John Haldane and Crispin Wright (eds.), Reality, Representation, and Projection. Oxford: Oxford University Press (pp. 85–130). Khalidi, Muhammad Ali. (2013) Natural Categories and Human Kinds: Classification in the Natural and Social Sciences. Cambridge: Cambridge University Press. Khalidi, Muhammad Ali. (2015) “Three Kinds of Social Kinds.” Philosophy and Phenomenological Research 90, 96–112.

273

Asya Passinsky King, Jeffrey. (2006) “Semantics for Monists.” Mind 115, 1023–1058. Kivy, Peter. (1983) “Platonism in Music: A Kind of Defense.” Grazer Philosophische Studien 19, 109–129. Kivy, Peter. (1987) “Platonism in Music: Another Kind of Defense.” American Philosophical Quarterly 24, 245–252. Korman, Daniel Z. (2015) Objects: Nothing Out of the Ordinary. Oxford: Oxford University Press. Korman, Daniel Z. (2020) “The Metaphysics of Establishments.” Australasian Journal of Philosophy 98, 434–448. Kornblith, Hilary. (2007) “How to Refer to Artifacts.” In Eric Margolis and Stephen Laurence (eds.), Creations of the Mind: Theories of Artifacts and their Representation. New York: Oxford University Press (pp. 138–149). Koslicki, Kathrin. (2012) “Essence, Necessity, and Explanation.” In Tuomas E. Tahko (ed.), Contemporary Aristotelian Metaphysics. Cambridge: Cambridge University Press (pp. 187–206). Koslicki, Kathrin. (2018) Form, Matter, Substance. Oxford: Oxford University Press. Koslicki, Kathrin. (2023) “Artifacts and the Limits of Agentive Authority.” In Miguel Garcia-Godinez (ed.), Thomasson on Ontology. Cham, Switzerland: Palgrave Macmillan (pp. 209–241). Levinson, Jerrold. (1980) “What a Musical Work Is.” Journal of Philosophy 77, 5–28. Levinson, Jerrold. (2007) “Artworks as Artifacts.” In Eric Margolis and Stephen Laurence (eds.), Creations of the Mind: Theories of Artifacts and their Representation. New York: Oxford University Press (pp. 74–82). Lowe, E. J. (2008) “Two Notions of Being: Entity and Essence.” Royal Institute of Philosophy Supplement 62, 23–48. Lowe, E. J. (2018) “Metaphysics as the Science of Essence.” In Alexander Carruth, Sophie Gibb, and John Heil (eds.), Ontology, Modality, and Mind: Themes from the Metaphysics of E.J. Lowe. Oxford: Oxford University Press (pp. 14–34). Mallon, Ron. (2007) “Human Categories Beyond Non-essentialism.” Journal of Political Philosophy 15, 146–168. Mason, Rebecca. (2016) “The Metaphysics of Social Kinds.” Philosophy Compass 11, 841–850. Mason, Rebecca. (2020) “Against Social Kind Anti-Realism.” Metaphysics 3, 55–67. Mason, Rebecca. (2021) “Social Kinds Are Essentially Mind-Dependent.” Philosophical Studies 178, 3975–3994. Passinsky, Asya. (2020) “Social Objects, Response-Dependence, and Realism.” Journal of the American Philosophical Association 6, 431–443. Passinsky, Asya. (2021) “Norm and Object: A Normative Hylomorphic Theory of Social Objects.” Philosophers’ Imprint 21, 1–21. Pettit, Philip. (1991) “Realism and Response-Dependence.” Mind 100, 587–626. Preston, Beth. (1998) “Why is a Wing Like a Spoon? A Pluralist Theory of Function.” Journal of Philosophy 95, 215–254. Preston, Beth. (2009) “Philosophical Theories of Artifact Function.” In Anthonie Meijers (ed.), Philosophy of Technology and Engineering Sciences. Amsterdam: North Holland. Preston, Beth. (2013) A Philosophy of Material Culture: Action, Function, and Mind. New York: Routledge. Preston, Beth. (2018) “Artifact.” In Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Fall 2018 Edition). Raven, Michael J. (2022) “A Puzzle for Social Essences.” Journal of the American Philosophical Association 8, 128–148. Searle, John R. (1990) “Collective Intentions and Actions.” In P. Cohen, J. Morgan, and M.E. Pollack (eds.), Intentions in Communication. Cambridge, MA: MIT Press (pp. 401–415). Searle, John R. (1995) The Construction of Social Reality. New York: Free Press. Searle, John R. (2010) Making the Social World: The Structure of Human Civilization. New York: Oxford University Press. Thomasson, Amie, L. (2003a) “Foundations for a Social Ontology.” Protosociology 18-19, 269–290. Thomasson, Amie L. (2003b) “Realism and Human Kinds.” Philosophy and Phenomenological Research 67, 580–609. Thomasson, Amie L. (2009) “Artifacts in Metaphysics.” In Anthonie Meijers (ed.), Handbook of Philosophy of the Technological Sciences. Elsevier Science (pp. 191–212).

274

Artifacts, Artworks, and Social Objects Thomasson, Amie L. (2014) “Public Artifacts, Intentions, and Norms.” In Maarten Franssen, Peter Kroes, Thomas A.C. Reydon, and Pieter E. Vermaas (eds.), Artefact Kinds: Ontology and the Human-Made World, Synthese Library 365. Cham: Springer (pp. 45–62). Tuomela, Raimo. (2002) The Philosophy of Social Practices: A Collective Acceptance View. Cambridge: Cambridge University Press. Witt, Charlotte. (2011) The Metaphysics of Gender. New York: Oxford University Press.

275

18 BIOLOGICAL SPECIES 1 Ingo Brigandt

Anti-essentialism has been particularly vivid in the case of biological species. Like in many other specific cases, e.g., gender or race (see Mallon, this volume and Rosario, this volume), what one always needs to bear in mind is that—given the background of particular philosophical debates—“essentialism” and “anti-essentialism” mean something specific to the case and context at hand. This made the ascription or rejection of essences meaningful for a case like species. My survey of the nexus of essentialism and biological species starts with a primer on biological systematics, which is important to understand how species are individuated. Then I address the question of whether an individual is essentially a member of its species and whether a species essentially belongs to its higher taxa. Section 18.3 features the central debate, which led to the prominence of anti-essentialism about species and the position that species are not natural kinds but individuals, despite the more recent position that a revised notion of essence is applicable to biological species and other biological kinds. The final section looks at literature and new philosophical topics from the last decade, including a neo-Aristotelian essentialism about species, a post-essentialism about human nature (an issue relevant to humans as a species), and considerations about the temporal persistence conditions of species stemming from species de-extinction efforts. A noteworthy trend is that both these discussions about species persistence and recent philosophy of science accounts of natural kinds in general could have been conducted—and usually were conducted—without relying on the notion of essence, possibly on the grounds that the label “essence” retains problematic connotations.

18.1 A Primer on Contemporary Biological Systematics Species are grouped into higher taxa, such as genera, families, orders, and kingdoms. It is still common to assign a Linnean rank to each higher taxon, although different taxa of the same rank (e.g., a bird family, a primate family, and an herbaceous plant family) can differ significantly in the number of species they contain and even in their evolutionary age. During the 1960s and 1970s, biological systematics (also called taxonomy) witnessed a dispute between three schools of classifying species into higher taxa (Hull 1988). According to numerical taxonomy (also called phenetics), only the overall phenotypic similarity between species matters. This was based on the philosophical orientation that biological taxonomy is nothing but a convenient way to store information, and should be a practice that can be

276

DOI: 10.4324/9781003008750-22

Biological Species

conducted independently of presupposing substantial theories (e.g., from evolutionary biology). In contrast, the longstanding approach of evolutionary systematics used to classify in terms of common phylogenetic ancestry, while also recognizing morphological change as resulting in different evolutionary grades that can count as distinct taxa. Finally, the phylogenetic systematics (also called cladistics) proposed by Willi Hennig in 1950 classifies exclusively in terms of common ancestry (Hennig 1966). Only the splitting of a phylogenetic lineage into separate species, resulting in different branches (clades) of the phylogenetic tree, can make for new taxa, while evolutionary change along a lineage that does not split is of no taxonomic relevance. A complete branch of the tree of life, consisting of an ancestral species (called a stem species) and all of its descendants is called a monophyletic group; such monophyletic groups are the only taxonomic groups that phylogenetic systematics recognizes as genuine taxa. For example, the traditional Linnean taxon of fish (Pisces)—although a distinct evolutionary grade for evolutionary taxonomy—is not monophyletic, as the most recent common ancestor of fish also has as its descendants all the land-living vertebrates, which the alleged taxon fish fails to include. Since the 1980s, phylogenetic systematics has become the only accepted approach, so contemporary systematics defines higher taxa as monophyletic groups in terms of common ancestry, and Pisces and some other Linnean taxa are no longer recognized. Although nowadays there is only one account of what constitutes a higher taxon, there are actually many different species concepts in use. Biologist Richard Mayden (1997) distinguished 22 different species concepts, and more recently philosopher John Wilkins (2018) provided his count of 28 conceptions as part of his detailed historical and philosophical discussion. Since these concepts rely on different conditions of what constitutes a species, the plurality of species concepts will have metaphysical implications in some of the following sections. To illustrate the motivation for using one species concept, which at the same time has drawbacks over other species concepts, it suffices to mention only some of the concepts as examples. The so-called biological species concept defines a species as a group of (potentially) interbreeding natural populations that are reproductively isolated from other such groups (Dobzhansky 1935; Mayr 1942). Given that interbreeding is a major mechanism of gene flow within a species, one major theoretical advantage is that the biological species concept ties into causal explanations of how speciation occurs. If two populations become geographically isolated, diverge genetically and phenotypically, and eventually lose the ability to interbreed and exchange genes, then these populations are now two different species. The biological species concept has also been historically instrumental in understanding the distribution of malaria, by means of being able to distinguish different species of Anopheles mosquitos (some of which transmit malaria to humans). At the same time, this species concept faces wellknown theoretical problems. One oddity is that when each of two geographically adjacent populations can interbreed, organisms of this species on different parts of the globe need not have the ability to even potentially interbreed. More problematic is that even proponents of the biological species concept admit that hybridization is quite common in plants and some animals—and hybridization is by definition an instance of interbreeding between different species. Finally, interbreeding consists in sexual reproduction, so that for (the vast number of) asexual species the biological species concept is utterly inapplicable and another species concept must be used. Another widely used concept is the ecological species concept (Simpson 1961; Van Valen 1976). It defines a species as a lineage (or a closely related set of lineages) that occupies an adaptive zone (ecological niche) minimally different from any other lineage in its range and that evolves separately from all lineages outside its range. The ecological species concept is beneficial in tracking the evolutionary adaptation of populations to local ecological

277

Ingo Brigandt

conditions, and it may be able to distinguish two populations as different species even when they are potentially able to interbreed. There are also different variants of the phylogenetic species concept (Cracraft 1983; Hennig 1966; Mishler and Brandon 1987). And the phenetic species concept and the morphological species concept—both construing species in terms of phenotypic similarity—are still relevant today, not as theoretical accounts of what constitutes a species, but as tools for classifying species in practice (Wilkins 2018). Whereas some decades ago some deemed their preferred species concept as the only important or fundamental one, nowadays the vast majority of biologists and philosophers are pluralists about species concepts (but see Barker 2019; de Queiroz 1999), where biologists may also make joint use of several species concepts as part of their work (Bzovy 2017). Not only are there several ontological processes that generate overall biodiversity (Ereshefsky 1992), but there are different epistemic purposes and scientific aims, each of which may require a distinct species concept (Kitcher 1984). We have seen that while the biological species concept is a useful tool for explaining speciation, it is ill-suited for classifying asexual organisms. Different species concepts entail not only different conditions of what constitutes a species but can also result in different boundaries between species and different counts of the number of species in a habitat. This can have major practical implications; for instance, while one species concept counts the Alabama sturgeon as a distinct species, which is critically endangered, another species concept would consider the Alabama sturgeon as a mere part of a species, which as a species is not endangered and thus not worthy of protective measures. Because of this, it is often recognized that various nonepistemic aims, such as conservation purposes or the medical aim of preventing malaria transmission, can legitimately be used when choosing a species concept and delineating what species there are (Conix 2019; Ludwig 2016).

18.2

Essentialist Statements Involving Species

Although most of the debates among biologists and philosophers concerning essentialism have centered on the question of whether species are natural kinds with essences (to which I turn in the next section), there are other essentialist statements involving species that one can investigate for their truth or falsity. These have only been occasionally considered, and in this section, I follow the book-length discussion by Joseph LaPorte (2004), among others. As tigers (Panthera tigris) are mammals, one may wonder whether the species Panthera tigris is essentially mammalian, or at least whether it necessarily belongs to the taxon Mammalia (see Torza, this volume and Correia, this volume). Likewise, one may ask whether Tony the Tiger is necessarily a tiger. Regarding the latter issue, the answer arguably is no. LaPorte (1997) as well as Okasha (2002) point out that this answer is at odds with the assumption found from ancient philosophy to modern analytic metaphysics (e.g., Wiggins 1980) that a person is essentially a human being and that any organism belongs necessarily to its species (see also Marabello, this volume). A clear way to arrive at Laporte’s and Okasha’s claim is if one adopts the orthodox account of phylogenetic systematics, according to which upon an ancestral species Δ splitting into two lineages Τ and Φ, species Δ thereby becomes extinct (ceases to exist) and two new species Τ and Φ originate (LaPorte 2004: Ch. 2, Sect. II.2.a). And if Tony the Tiger belongs to lineage Τ, he is only contingently a member of species Τ. For the existence of Τ as a species hinges on the branching event that consists in some lineage Φ branching off. Had this contingent event not happened, species Δ would still persist to the present and be the lineage that contains Tony.

278

Biological Species

This conclusion about an organism not essentially belonging to its species also holds if one doesn’t adopt the extinction upon splitting convention, while relying on the biological species concept that focuses on the potential to interbreed or alternatively the ecological species concept that pertains to occupying the same ecological niche (LaPorte 2004: Ch. 2, Sect. II.2.b; Okasha 2002). These properties for species identity are contingently possessed by an organism. For instance, the potential to interbreed can change if other organisms were to change their mate recognition behavior, and the potential of flowers to interbreed is even dependent on changes in their pollinators. Imagine a large population Δ from which a splinter group becomes geographically isolated. Assuming that the members of the splinter group either lose the ability to interbreed with the members of the larger population Δ (or alternatively come to adapt to a new ecological niche), the splinter group would form a separate species Τ. (Species Δ would not thereby become extinct, and contain ancestral as well as contemporary members.) In this scenario, Tony the Tiger actually belongs to species Τ. But this could have been different. Had the loss of interbreeding (or the adaptation to a different ecological niche) by the overall splinter group containing Tony not happened, there would only be species Δ and Tony would instead belong to Δ. So Tony the Tiger is not necessarily—and thus not essentially—a tiger. What about the quite different question of whether the species Panthera tigris necessarily belongs to the taxon Mammalia, or to the taxon Animalia? Now LaPorte (2004: Ch. 2, Sect. II.1) claims that a species (or a higher taxon) necessarily belongs to any more encompassing taxon to which it actually belongs. We have seen that phylogenetic systematics (as the only currently accepted approach) only recognizes higher taxa that are monophyletic groups (clades), consisting of an ancestral species and all its descendants. Consequently, mammals are those organisms that descended from species G—this historical ancestry could be considered as the essence that defines mammals (see also Robertson Ishii, this volume). There are possible worlds where the tree of life has a different branching structure, with some branches (clades) missing, including tigers or even mammals not existing. Still, every possible world that contains tigers—which actually descended from ancestral species G—also contains their ancestor G. LaPorte’s reasoning is that whatever organisms descended from G in such a possible world (e.g., tigers) count as mammals, hence his tenet that the species Panthera tigris necessarily is a mammalian species (and necessarily an animal species). However, Margarida Hermida (2022) has more recently argued that LaPorte is wrong on this issue and that all tigers are merely contingently mammals and animals. Interestingly, her argument is based on the situation that any particular tiger contingently belongs to the tiger species—which as we have seen even LaPorte grants. Hermida argues that individual tigers could have been non-mammals if they were present in a possible world where mammals did not exist at all. The reason is that a particular tiger (such as Tony the Tiger) exists in a possible world whenever his ancestors are present in this world as well, while the organisms that form species G (and found the clade Mammalia) in the actual world need not belong to this species in the possible world, in which case she contends that this world has no mammals (in contrast to LaPorte’s assumption that the counterfactual descendants of G in every case count as mammals). To my mind, these discussions exhibit a potential ambiguity: for the most part, LaPorte (2004) focuses on the taxon Panthera tigris when claiming it to be necessarily a mammalian species, while Hermida (2022) convincingly argues that all individual tigers are contingently mammals. Does a necessity still hold for a biological taxon as such (even if it doesn’t hold for any individual member of the taxon)? I am not sure that is the case either, at least if organisms forming ancestral groups that found taxa in the actual world can belong to different species in possible worlds. Even the very type specimen that is the individual used to

279

Ingo Brigandt

anchor a species name does not necessarily belong to its species—LaPorte (2003) diagnoses this as an instance of the contingent a priori. In spite of endorsing the origin essentialist claim that mammals are defined by deriving from ancestral species G, LaPorte (2004: Ch. 3) argues that such essences are not discovered, in contrast to the vision popularized by Kripke and Putnam (see also Mallozzi, this volume). This obtains in cases of scientists finding that what they took to be a kind does not actually make up a natural kind, where the finding entails having to choose among different options of how to align the traditional natural kind term with one of the genuinely existing natural kinds—or to choose to discard the term. (LaPorte’s discussion also covers chemical kinds, so that his tenet that essences are not discovered does not hinge on the historical essences that characterize biological taxa.) A good example is rodents. Rabbits were once considered to be rodents, but it turned out that rabbits form a lineage that branched off before the origin of all the other rodents. This separate rabbit lineage then acquiring characteristics that are not found among the other rodents was one reason to taxonomically exclude the rabbits from the rodents altogether—resulting in one account of what the ancestral group H is that is the most recent ancestor of all rodents. But in principle, one could also have decided that the rodents are the more encompassing monophyletic group that includes rabbits (in addition to the descendants of H that we now consider rodents). This would have resulted in a different group H’ being the ancestor that figures in the essence defining rodents. More interestingly, in the 1990s evidence emerged that suggested Guinea pigs are more closely related to seals, horses, and primates than they are to mice and rats. If this is the true phylogeny,2 then mice, rats, and Guinea pigs as paradigmatic rodents do not form a monophyletic group—unless one makes the in principle possible choice that rodents also include seals, horses, and primates. If one eschews this counterintuitive option, one has to make a further decision about which of what were considered to be paradigmatic rodents are actually rodents (and which are not). The need for such scientific choices bolsters LaPorte’s contention that the essences of at least some natural kinds are not just discovered, and that the reference of the corresponding natural kind term is not pre-determined by prior language use plus the natural kind structure of the world (see also Brigandt 2013; LaPorte 1996).

18.3 Species as Natural Kinds vs. Species as Individuals The vast majority of discussions on species and essentialism have turned on the question of whether a species is a natural kind or an individual. Around the same time that the rise of the causal theory of reference for natural kind terms featured tigers and lemons as examples of natural kinds (Kripke 1980; Putnam 1975; see Tahko, this volume), biologist Michael Ghiselin (1974) and philosopher of biology David Hull (1978) argued that a species is not a class or kind, but instead an individual. This novel perspective on the metaphysics of species had been preceded by the charge that for too long biological taxonomy had been governed by essentialism, which is at odds with modern evolutionary theory (Hull 1965). Essentialism about species came to be viewed as effectively identical to what previously had been dubbed typological thinking, as opposed to population thinking (Mayr 1959).3 Population thinking emphasizes genetic and phenotypic variation within populations because natural selection acts on phenotypic differences between individuals and genetic variation makes gradual evolutionary change possible. From this perspective, individual organisms and their properties are primary. The properties of species such as most common phenotypic traits or quantitative trait averages are merely derivative of the properties of individuals, and subject to evolutionary change. Sober (1980) articulates the contrary typological thinking (or essentialism about

280

Biological Species

species) as the adoption of a natural state model for explaining variation, analogous to Aristotelean physics. This way of thinking postulates the existence of a natural phenotypic state of a species (grounded in the species’ essence), where deviations from this natural state are seen as being due to intervening (e.g., environmental) forces. This provides a way to nominally acknowledge the readily observable within-species variation, while taking variation to be of no scientific significance—unlike the species essence that has explanatory impact. Such a natural state model about species is clearly inconsistent with Darwinian evolutionary theory. As a result, many evolutionary biologists were critical of such a form of essentialist thinking, which formed the background for subsequent discussions of the specific ontological status of species. The new argument against the traditional construal of species as classes or kinds pointed out that each species is a particular object, not a universal (Ghiselin 1974; Hull 1978, 1989). A species is a spatiotemporally restricted object, which comes into being at a certain point in time, exists extended in time at certain regions of space, and may become extinct. The most compelling objection to species as kinds is that defining a species in terms of necessary or sufficient properties or an essence consisting of shared molecular, physiological, or anatomical traits is impossible (see Dumsday, this volume). Not only is there significant variation among the members of a species at any one point in history, but the very opportunity of subsequent evolutionary change—and of in principle unlimited evolutionary transformation—makes it moot to point to certain phenotypic traits as defining this species. And, pace Kitts and Kitts (1979), invoking some genetic traits as the species’ essence will not fare any better (see also Mallon, this volume, Rosario, this volume and Stoljar, this volume). In contrast, the speciesas-individuals (SAI) thesis was upheld as doing justice to the nature of species as units of evolutionary transformation. In this approach, the species Panthera tigris is an individual, which is made up of various organisms as its physical parts (as opposed to organisms being members of a class). Analogous to how an organism is made up of cells (including different cell types), variation and differences among these parts are very well possible. And an individual is a concrete object, coming into being at a certain point in time and existing during a particular period of history. Most importantly, an individual can undergo substantial change over time, without ceasing to be this individual.4 The species-as-individuals thesis also meshes with the fact that biological species are denoted by proper names. Soon after its formulation, it became the dominant metaphysical position among biologists and philosophers of biology. Apart from species, higher taxa often came to be construed as individuals as opposed to kinds (Jenner 2006).5 It was not until the turn of the century that a new vision of natural kinds was articulated that also attempted to capture species and other kinds studied by biology and other special sciences. This is Richard Boyd’s (1999a, 1999b) notion of natural kinds as homeostatic property cluster (HPC) kinds. The first important ingredient is to replace the idea that a natural kind is defined by necessary and sufficient conditions by mere correlations among properties. An HPC kind is a cluster of properties—properties which kind members tend to share, but need not universally share. The presence of variation within a species thereby becomes no genuine obstacle for construing this species as an HPC kind. (Wilson et al. 2007 call this the “intrinsic heterogeneity” of HPC kinds.) Based on his realism about natural kinds, Boyd views this clustering of properties across kind instances not just as a function of some language game, but something qualifies as an HPC kind only to the extent to which there is some real mechanism that accounts for the clustering, which Boyd dubs a “homeostatic” mechanism. The second important move is to point out that traditional discussions of natural kinds (at least tacitly) assumed that the essences defining a kind are intrinsic properties of kind

281

Ingo Brigandt

members, such as microstructure or genes (see Torza, this volume and Griffith, this volume). But there is no reason to exclude relational properties as also being constitutive of some kinds (Boyd 1999a; Griffiths 1999; Okasha 2002; see Tahko, this volume). Historical relations of ancestry are important to species as evolutionary entities (and more generally to other historical kinds; Godman 2021; Millikan 1999; see Robertson Ishii, this volume). We learned in Section 1 that a higher taxon is simply a monophyletic group (a clade), the identity of which consists in common ancestry from some ancestral group of organisms G. And being descended from G is not an intrinsic, but a relational property. While common ancestry as a relation defines which organisms are members of this higher taxon as a natural kind, this is fully consistent with genetic, phenotypic, and other intrinsic properties of organisms significantly diverging within the taxon, and is thus compatible with evolutionary change. And while common ancestry and the forming of phylogenetic lineages likewise matter to species, some species concepts feature further relational properties. In Section 18.1 we encountered the so-called biological species concept, which defines a species as a group of potentially interbreeding natural populations (which are reproductively isolated from other such groups). “Being able to interbreed with” is not an intrinsic property of a particular organism, but a relation between organisms that also hinges on the internal features of (and genetic compatibility with) potential mates. Since interbreeding leads to gene flow within the species, any novel genetic variant (e.g., a mutation) introduced in one subpopulation can spread across the species. In this sense, the interbreeding relation does yield a certain cohesion of the species (Wilson et al. 2007). But this genetic and phenotypic cohesion is fully compatible with the ongoing evolutionary change of this species. Likewise, the ecological species concept focuses on an organism occupying the same adaptive zone as others, again a relational property. This yields phenotypic adaptation to the same ecological niche, while fully permitting evolutionary change (where even the ecological niche can gradually change).6 Pointing to species concepts including relational properties therein is part and parcel of the naturalism guiding the HPC approach (Brigandt 2009). A theory of natural kinds has to capture kinds as studied in biology, and it is ultimately an empirical question of what constitutes the ontological character of a given species (see also Mallon, this volume). Proponents of the notion of HPC kinds have used the notion of an “essence” for what determines the identity of such a kind, often by viewing the homeostatic mechanism underlying the property clustering as the essence, which is explanatorily more fundamental than the clustering (Boyd 1999a; Griffiths 1999; see also Godman et al. 2020). At the same time, it has been emphasized that such essences need not be traditional (e.g., microstructural) essences at all (see Griffith, this volume). An HPC kind essence can include relational and other non-intrinsic properties. The essence can also be complex, consisting of many properties or be multiply realized. Such an essence may also change across time, e.g., what physically constitutes the ability to interbreed or the sharing of an ecological niche may be subject to change.7 Based on the introduction of the HPC kinds approach many philosophers have come to agree that species (and higher taxa) can be considered natural kinds after all, and even a few biologists have adopted the HPC approach (Assis 2011; Franz 2005; Rieppel 2005, 2009; Wagner 2014). It has also been suggested that a species can be both a natural kind and an individual and that these ontological categories do not exclude each other (Boyd 1999a; Brigandt 2009; LaPorte 2004; Okasha 2002; Rieppel 2007, 2013). From this perspective, the kinds vs. individuals question is less important than the features that make diverse organisms belong to one species and that account for the species’ persistence and identity across time. Still, the species-as-individuals thesis remains the dominant position among biologists. Some

282

Biological Species

philosophers also continue to maintain that species as evolutionary entities can only be individuals, and cannot possess essences or be natural kinds (Ereshefsky 2007, 2010).8 Ironically, in the history of metaphysics, the category of an individual has likewise been characterized by an essence (Okasha 2002; see also Marabello, this volume), where an individual’s essence accounts for why the individual’s parts are not just a heap of objects but form a unified whole and how the individual can persist through changes (Witt 2011). While the proponents of species as HPC kinds have pointed to properties and processes figuring in species concepts as what makes a species a unified whole, philosophers favoring the speciesas-individuals thesis have only emphasized evolutionary transformation but (curiously) never addressed the obvious question of what conditions would account for the identity of any such species-individual across evolutionary time (see also Scarpati, this volume). Beyond ontological issues, another imbalance between the two sides of the individual or kind debate is that only proponents of species as natural kinds have made points of epistemological significance. The very motivation for the HPC account of natural kinds is that scientists rely on natural kinds because knowledge of genuine kinds permits scientific inferences and explanations (Boyd 1999b), something also echoed by other and more recent accounts of natural kinds in philosophy of science. Reliable inferences can be made to the extent to which similarities exist among kind members and different properties associated with the kinds are correlated. And to the extent to which knowledge about a natural kind includes underlying mechanisms and other causal relations, scientific explanations become possible. Philosophical attention to scientific aims matters as even in the domain of evolutionary biology and systematics different kinds are used for different specific scientific purposes, which can have implications for the nature of each such kind, e.g., regarding the relevance of intrinsic or of relational properties (Brigandt 2009; see also Bolker 2013). A naturalistic approach needs to capture the diversity of kinds found in the special sciences and the different ways they figure in scientific reasoning. In contrast, philosophers of science upholding the species-as-individuals position have been comparatively silent on matters of scientific theorizing and practice.

18.4

From the Death of Species Essences to the Rebirth of Species?

Among philosophers, debates about species being natural kinds or individuals have been revived by Michael Devitt’s (2008) call for a strong role for intrinsic properties within a species essence, on the grounds that the relational properties (e.g., ability to interbreed) used by species concepts are insufficient to identify a particular species taxon as that taxon (rather than another species). Predictably, critical responses to such an intrinsicalism have been raised (Barker 2010; Ereshefsky 2010; Godman and Papineau 2020; Slater 2013). Very recently, Christopher Austin (2019) has put forward a construal of species as natural kinds with (purely) intrinsic essences. What makes his contribution highly original is the use of a neo-Aristotelian framework, in terms of the members of a given species sharing the disposition to develop in a certain fashion. The intrinsic essence consists in an organism’s developmental modules. Importantly, Austin avoids any claim that the disposition would consist in developing one (species-typical) phenotype. Instead, he relies on the disposition to generate a restricted but whole range of phenotypic outcomes that are possible for this species’ members. From this perspective, developmental modules yield a natural kind with the capacity to produce various phenotypic manifestations (including non-actual ones), depending on the environmental and other circumstances of individual organisms. Austin’s neo-Aristotelian account has a hylomorphic component, in that the form consisting in a species-wide goal-directed disposition can across individual organisms be multiply realized in matter.

283

Ingo Brigandt

While a philosophical focus on dispositions holds great promise to understand theorizing in evolutionary biology that employs explanations capturing the possible and impossible (Austin and Nuño de la Rosa 2021; Brigandt 2015; Brigandt et al. 2023), in my view, it is less obvious that this is suitable for specifically construing species as natural kinds with intrinsic essences.9 Austin (2019: Ch. 2) explicitly separates the question of what makes an organism a member of a species (answered in terms of an intrinsic essence) from the question of what makes a taxon a species as opposed to some other category (answered in terms of species concepts, including relational properties). But this cuts off a metaphysical account that is supposed to be of the nature of species from all traditional considerations (discussed in Section 3) that delineate different species and provide the basis for a species’ historical cohesion combined with the potential for evolutionary change. A more cautious neoAristotelian and teleological essentialism (than Austin’s) was previously put forward by Denis Walsh (2006), which focuses on the natures of individual organisms, not the natures of species. Looking beyond the restricted topic of species, on the one hand, the last decade has seen a reinvigorated and ongoing interest in natural kinds among philosophers of science (Ludwig 2017, 2018). On the other hand, these accounts do not rely on the notion of an essence (Franklin-Hall 2015; Khalidi 2013; Magnus 2012; Slater 2015; see also Griffith, this volume). These new accounts of what natural kinds are also do not adopt Boyd’s prominent identification of natural kinds with homeostatic property clusters, on the grounds that many kinds in biology and other special sciences (possibly even species) need not have an underlying homeostatic mechanism (see Brown, this volume). But they still continue Boyd’s naturalistic project of attempting to philosophically capture the diversity of kinds found in the special sciences. Given this recent proposal of actually different accounts of which properties make for a natural kind combined with the recognition that across science there are different types of kinds, some have also suggested that we should abandon the search for the one unique theory of natural kinds and that the mere statement that something is a natural kind—without further details about this kind—is not very informative (Brigandt 2022; Ludwig 2018). The motivation for many contemporary philosophers of science having turned away from the label “essence” appears to be that the notion tends to suggest outdated tenets or obscures several philosophical issues that are better clarified in different terms. This perspective also aligns with Maria Kronfeldner’s (2018) post-essentialist treatment of human nature—a notion traditionally related to the idea that the human species has an essence. She argues that the (“essentialist”) tradition promised to offer three things through one human nature (see also Mallon, this volume and Griffith, this volume): a classificatory nature (what makes humans members of their species), a descriptive nature (the properties of human beings studied in biology, cognitive science, and social science), and an explanatory nature (in terms of developmental resources of humans that are biologically or socially inherited). Kronfeldner convincingly argues that a single account of human nature is inadequate to accomplish all three tasks and tends to yield a problematic picture of at least some of us. Instead, as part of a pluralist approach to human nature, she develops three separate accounts, yielding the classificatory, descriptive, and explanatory nature, respectively. An analogous point may well be made about species essences. If a natural kind’s essence is to provide membership conditions for kind members and ground a causal or mechanistic explanation of the kind’s characteristic properties—as has often been assumed—while possibly also encompassing the kind’s characteristic properties or property cluster, then ambiguously many different intrinsic and relational properties have to be stuffed into the essence of a complex biological kind such

284

Biological Species

as a species. While advocating for three concepts of human nature, Kronfeldner (2018) vigorously eschews any normative conception of human nature that would claim some human biological, cognitive, or behavioral traits to be normal or ideal while considering deviations as abnormal or inferior. Such problematic but practically inevitable connotations would also have to be kept in mind when assigning an essence to the human species (see also Ritchie, this volume, Rosario, this volume, Stoljar, this volume and Tannenbaum & Glesakos, this volume). A novel metaphysical question about species stems from scientific efforts toward the deextinction of species. One motivation is to enhance species conservation programs by restoring a species that has recently gone extinct, for instance, through back-breeding (breeding and interbreeding of extant species to generate organisms similar to past instances). More fanciful are speculations about recreating long-extinct species such as the woolly mammoth, possibly through somatic cell nuclear transfer or by using DNA partially preserved in permafrost. While such scientific efforts face serious if not unsurmountable practical obstacles, philosophers have offered conceptual reasons concerning the metaphysical impossibility or possibility of de-extinction (Finkelman 2018; Piotrowska 2018; Siipi and Finkelman 2017). The very possibility of restoring a species that hitherto was extinct hinges on the persistence and spatiotemporal identity conditions of a species. At the same time, such philosophical discussions of what could count as the same species as represented by past organisms have been conducted without using the notion of a species essence. To a first approximation, the species concept adopted impacts the potential for the de-extinction or rebirth of species (Finkelman 2018). If one used a phenetic or morphological species concept, which construes belonging to the same species in terms of phenotypic similarity, then any newly created organism that is sufficiently similar would fully count as a member of the species that was considered extinct. And the same would hold for a genetic species concept. However, the more commonly used species concepts do not just provide operational criteria for distinguishing species but a more substantive account of the nature of species. And they typically view a species as some phylogenetic lineage of organisms linked by reproduction, combined with additional considerations of what counts as a splitting of a lineage into different species (e.g., not being able to interbreed any longer or occupying different ecological niches). The requirement of a continuous lineage imposes serious challenges to the idea of deextinction being conceptually coherent at all. Philosophers have recently been exploring ways in which the notion of “reproduction” could be construed and have been evaluating whether certain biological techniques pursued (e.g., somatic cell nuclear transfer) would then count as a reproductive link to ancestral organisms (Piotrowska 2018). The conceptual possibility of the species as such still being extinct while having presently existing members has also been considered (Siipi and Finkelman 2017). Taking stock, recent insightful discussions about what metaphysically would count as a species’ continued existence (under counterfactual modifications) did not have the need to employ the notion of an essence, and may be based on reasons for avoiding a historically loaded or contextually specified notion of essentialism. To a significant extent, this also holds for the recent enthusiasm about natural kinds among naturalistic philosophers of science. And even though the late Richard Boyd was happy to use the notion of essence as part of his theory, from the outset he also argued that while retaining some realism about kinds, social constructivists were right in many respects (Boyd 1999b). On his “bicameralism” thesis, something is a natural kind not only in virtue of the structure of reality but also in virtue of such structures fulfilling the inferential and explanatory demands of a disciplinary matrix. And such demands are up to our human interests and purposes. Subsequently, it has been acknowledged that the naturalness of a

285

Ingo Brigandt

kind is relative to a scientific domain (Khalidi 2013; Magnus 2012) or scientists’ aims, interests, and norms (Franklin-Hall 2015; Slater 2015). The choosing of one species concept over another—and thus the metaphysical conditions determining a species’ identity—is likewise contingent on scientists’ interests. Section 18.1 already indicated that conservation, medical, and other non-epistemic purposes can play an important role in species concept choice (Brigandt 2022; Conix 2019; Ludwig 2016; for an analogous case see Brown, this volume). So if one chooses to construe species as having essences, chances are that such essences are co-determined by human values (see also Vaidya & Wallner, this volume).

18.5 • • • • • • • • •

Related Topics

Marabello, M. (this volume) “Essences of Individuals” Tahko, T. (this volume) “Natural Kind Essentialism” Robertson Ishii, T. (this volume) “Origin Essentialism” Ritchie, K. (this volume) “Language of Essence” Brown, D. (this volume) “Psychiatric Kinds” Mallon, R.. (this volume) “Race” Rosario, E. (this volume) “Sex and Gender” Stoljar, N. (this volume) “Social Justice” Griffith, A. (this volume) “Social construction”

Notes 1 I thank the participants of the Routledge Handbook of Essence Workshop for their comments on a synopsis of my essay. I am particularly indebted to Kathrin Koslicki and Mike Raven for their detailed comments on a draft of this chapter. The work on this essay was supported by the Social Sciences and Humanities Research Council of Canada (Insight Grant 435-2016-0500) and the Canada Research Chairs Program (CRC-2018-00052). 2 Since 2005 substantial evidence has been accumulating that guinea pigs are actually more closely related to rats, mice, and other rodents, so that the traditional taxon considered to be rodents can be retained. 3 Such a lumping together of Platonic types and Aristotelean essences immediately suggests carelessness about the views of Aristotle and other historical figures ( Lennox 2001). But it has also been revealed that even pre-Darwinian taxonomists, including Linnaeus, were not essentialists ( Amundson 2005; Brigandt 2021; Wilkins 2013; Winsor 2003). 4 Brogaard (2004) suggests that the failure to distinguish between a three-dimensional (endurantist) and a four-dimensional (perdurantist) account of species has led to some of the issues debated, e.g., whether species can be construed as mereological sums. Among the species-as-individuals proponents, Hull (1989: 187) merely states that a species name refers “both to a spatiatemporally extended lineage and to a time-slice of that lineage”. Rieppel (2008, 2013) favors a four-dimensional interpretation. 5 Likewise, homologues came occasionally to be viewed as individuals ( Ereshefsky 2009; Grant and Kluge 2004). A homologue is the same character (bodily part) across all the organisms making up a species or higher taxon. Traditionally, the relation of homology (character x in one species is homologous to y in another species) was considered an equivalence relation, in line with these characters forming a kind. 6 While some have objected to using the HPC account for biological taxa on the grounds that “whatever is ‘homeostatic’ cannot, by definition, evolve” ( Kluge 2003: 234), this takes Boyd’s label “homeostatic” mechanism too literally. While common ancestry and interbreeding account for some cohesion among the members of a taxon, they do not prevent evolutionary change. 7 Although Boyd (1999a) is not fully clear on whether the essence of an HPC kind is the property cluster or the homeostatic mechanism accounting for the clustering (or both), he explicitly views the

286

Biological Species property cluster as well as the homeostatic mechanism as subject to change. While the HPC account of natural kinds recovered a notion of essentialism compatible with evolutionary theory, based on the field of evolutionary developmental biology there have also been arguments that a sort of typology (or typological thinking) can legitimately be used in evolutionary contexts ( Amundson 2005; Austin 2017; Brigandt 2007, 2021; Lewens 2009; Love 2009; Walsh 2006). 8 While biologist Olivier Rieppel (2006) pointed out that phylogenetic systematists (including those viewing species as individuals) are committed to essentialism in the form of origin essentialism, philosopher Marc Ereshefsky (2007) responded that since “qualitative essentialism” and origin essentialism are two types of essentialism, the proponents of the species-as-individuals thesis (committed to origin essentialism) are not committed to essentialism at all. The blatancy of this non-sequitur illustrates how some preferring the SAI position have been less reconciliatory than those HPC proponents who acknowledge that species can at the same time be individuals and kinds. 9 For problems with dispositions in other scientific contexts, see Lam, this volume.

References Amundson, R. (2005) The Changing Role of the Embryo in Evolutionary Thought: Roots of Evo-Devo, Cambridge: Cambridge University Press. Assis, L. C. S. (2011) “Individuals, Kinds, Phylogeny and Taxonomy,” Cladistics 27: 1–3. Austin, C. J. (2017) “Aristotelian Essentialism: Essence in the Age of Evolution,” Synthese 194: 2539–2556. Austin, C. J. (2019) Essence in the Age of Evolution: A New Theory of Natural Kinds, New York: Routledge. Austin, C. J. and L. Nuño de la Rosa (2021) “Dispositional Properties in Evo-Devo,” in L. Nuño de la Rosa and G. Müller (eds.), Evolutionary Developmental Biology: A Reference Guide, Cham: Springer, pp. 469–481. Barker, M. J. (2010) “Specious Intrinsicalism,” Philosophy of Science 77: 73–91. Barker, M. J. (2019) “Eliminative Pluralism and Integrative Alternatives: The Case of Species,” British Journal for the Philosophy of Science 70: 657–681. Bolker, J. (2013) “The Use of Natural Kinds in Evolutionary Developmental Biology,” Biological Theory 7: 121–129. Boyd, R. (1999a) “Homeostasis, Species, and Higher Taxa,” in R. A. Wilson (ed.), Species: New Interdisciplinary Essays, Cambridge, MA: MIT Press, pp. 141–185. Boyd, R. (1999b) “Kinds as the “Workmanship of Men”: Realism, Constructivism, and Natural Kinds,” in J. Nida-Rümelin (ed.), Rationality, Realism, Revision: Proceedings of the 3rd International Congress of the Society for Analytic Philosophy, Berlin: de Gruyter, pp. 52–89. Brigandt, I. (2007) “Typology Now: Homology and Developmental Constraints Explain Evolvability,” Biology & Philosophy 22: 709–725. Brigandt, I. (2009) “Natural Kinds in Evolution and Systematics: Metaphysical and Epistemological Considerations,” Acta Biotheoretica 57: 77–97. Brigandt, I. (2013) “A Critique of David Chalmers’ and Frank Jackson’s Account of Concepts,” Protosociology 30: 63–88. Brigandt, I. (2015) “Evolutionary Developmental Biology and the Limits of Philosophical Accounts of Mechanistic Explanation,” in P.-A. Braillard and C. Malaterre (eds.), Explanation in Biology: An Enquiry into the Diversity of Explanatory Patterns in the Life Sciences, Dordrecht: Springer, pp. 135–173. Brigandt, I. (2021) “Typology and Natural Kinds in Evo-Devo,” in L. Nuño de la Rosa and G. Müller (eds.), Evolutionary Developmental Biology: A Reference Guide, Cham: Springer, pp. 483–493. Brigandt, I. (2022) “How to Philosophically Tackle Kinds without Talking About “Natural Kinds”,” Canadian Journal of Philosophy 52: 356–379. Brigandt, I., C. Villegas, A. C. Love and L. Nuño de la Rosa (2023) “Evolvability as a Disposition: Philosophical Distinctions and Scientific Implications,” in T. F. Hansen, D. Houle, M. Pavličev and C. Pélabon (eds.), Evolvability: A Unifying Concept in Evolutionary Biology?, Cambridge, MA: MIT Press, pp. 55–72. Brogaard, B. (2004) “Species as Individuals,” Biology and Philosophy 19: 223–242.

287

Ingo Brigandt Bzovy, J. (2017) Species Pluralism: Conceptual, Ontological, and Practical Dimensions, Dissertation, Western University. http://ir.lib.uwo.ca/etd/4309 Conix, S. (2019) “Radical Pluralism, Classificatory Norms and the Legitimacy of Species Classifications,” Studies in History and Philosophy of Biological and Biomedical Sciences 73: 27–34. Cracraft, J. (1983) “Species Concepts and Speciation Analysis,” in R. F. Johnston (ed.), Current Ornithology, New York: Plenum Press, pp. 159–187. de Queiroz, K. (1999) “The General Lineage Concept of Species and the Defining Properties of the Species Category,” in R. A. Wilson (ed.), Species: New Interdisciplinary Essays, Cambridge, MA: MIT Press, pp. 49–89. Devitt, M. (2008) “Resurrecting Biological Essentialism,” Philosophy of Science 75: 344–382. Dobzhansky, T. (1935) “A Critique of the Species Concept in Biology,” Philosophy of Science 2: 344–355. Ereshefsky, M. (1992) “Eliminative Pluralism,” Philosophy of Science 59: 671–690. Ereshefsky, M. (2007) “Foundational Issues Concerning Taxa and Taxon Names,” Systematic Biology 56: 295–301. Ereshefsky, M. (2009) “Homology: Integrating Phylogeny and Development,” Biological Theory 4: 225–229. Ereshefsky, M. (2010) “What’s Wrong with the New Biological Essentialism,” Philosophy of Science 77: 674–685. Finkelman, L. (2018) “De-Extinction and the Conception of Species,” Biology & Philosophy 33: 32. Franklin-Hall, L. R. (2015) “Natural Kinds as Categorical Bottlenecks,” Philosophical Studies 172: 925–948. Franz, N. M. (2005) “Outline of an Explanatory Account of Cladistic Practice,” Biology and Philosophy 20: 489–515. Ghiselin, M. T. (1974) “A Radical Solution to the Species Problem,” Systematic Zoology 23: 536–544. Godman, M. (2021) The Epistemology and Morality of Human Kinds, New York: Routledge. Godman, M., A. Mallozzi and D. Papineau (2020) “Essential Properties Are Super-Explanatory: Taming Metaphysical Modality,” Journal of the American Philosophical Association 6: 316–334. Godman, M. and D. Papineau (2020) “Species Have Historical Not Intrinsic Essences,” in A. Bianchi (ed.), Language and Reality from a Naturalistic Perspective: Themes from Michael Devitt, Cham: Springer, pp. 355–367. Grant, T. and A. G. Kluge (2004) “Transformation Series as an Ideographic Character Concept,” Cladistics 20: 23–31. Griffiths, P. E. (1999) “Squaring the Circle: Natural Kinds with Historical Essences,” in R. A. Wilson (ed.), Species: New Interdisciplinary Essays, Cambridge, MA: MIT Press, pp. 208–228. Hennig, W. (1966) Phylogenetic Systematics, translated by D. D. Davis and R. Zangerl, Urbana: University of Illinois Press. Hermida, M. (2022) “Cats Are Not Necessarily Animals,” Erkenntnis. doi: 10.1007/s10670-022-005 88-w. Hull, D. L. (1965) “The Effect of Essentialism on Taxonomy – Two Thousand Years of Stasis,” British Journal for the Philosophy of Science 15: 314–326 and 16:1-18. Hull, D. L. (1978) “A Matter of Individuality,” Philosophy of Science 45: 335–360. Hull, D. L. (1988) Science as a Process: An Evolutionary Account of the Social and Conceptual Development of Science, Chicago: University of Chicago Press. Hull, D. L. (1989) The Metaphysics of Evolution, Albany: State University of New York Press. Jenner, R. A. (2006) “Unburdening Evo-Devo: Ancestral Attractions, Model Organisms, and Basal Baloney,” Development Genes and Evolution 216: 385–394. Khalidi, M. A. (2013) Natural Categories and Human Kinds: Classification in the Natural and Social Sciences, Cambridge: Cambridge University Press. Kitcher, P. (1984) “Species,” Philosophy of Science 51: 308–333. Kitts, D. B. and D. J. Kitts (1979) “Biological Species as Natural Kinds,” Philosophy of Science 46: 613–622. Kluge, A. G. (2003) “On the Deduction of Species Relationships: A Précis,” Cladistics 19: 233–239. Kripke, S. (1980) Naming and Necessity, Cambridge, MA: Harvard University Press. Kronfeldner, M. (2018) What’s Left of Human Nature? A Post-Essentialist, Pluralist, and Interactive Account of a Contested Concept, Cambridge, MA: MIT Press. LaPorte, J. (1996) “Chemical Kind Term Reference and the Discovery of Essence,” Noûs 30: 112–132.

288

Biological Species LaPorte, J. (1997) “Essential Membership,” Philosophy of Science 64: 96–112. LaPorte, J. (2003) “Does a Type Specimen Necessarily or Contingently Belong to Its Species?,” Biology and Philosophy 18: 583–588. LaPorte, J. (2004) Natural Kinds and Conceptual Change, Cambridge: Cambridge University Press. Lennox, J. G. (2001) Aristotle’s Philosophy of Biology: Studies in the Origins of Life Science, Cambridge: Cambridge University Press. Lewens, T. (2009) “Evo-Devo and “Typological Thinking”: An Exculpation,” Journal of Experimental Zoology Part B: Molecular and Developmental Evolution 312B: 789–796. Love, A. C. (2009) “Typology Reconfigured: From the Metaphysics of Essentialism to the Epistemology of Representation,” Acta Biotheoretica 57: 51–57. Ludwig, D. (2016) “Ontological Choices and the Value-Free Ideal,” Erkenntnis 81: 1253–1272. Ludwig, D. (2017) “Indigenous and Scientific Kinds,” British Journal for the Philosophy of Science 68: 187–212. Ludwig, D. (2018) “Letting Go of “Natural Kind”: Toward a Multidimensional Framework of Nonarbitrary Classification,” Philosophy of Science 85: 31–52. Magnus, P. D. (2012) Scientific Enquiry and Natural Kinds: From Planets to Mallards, New York: Palgrave Macmillan. Mayden, R. L. (1997) “A Hierarchy of Species Concepts: The Denouement in the Saga of the Species Problem,” in M. F. Claridge, H. A. Dawah and M. R. Wilson (eds.), Species: The Units of Biodiversity, London: Chapman and Hall, pp. 381–424. Mayr, E. (1942) Systematics and the Origin of Species from the Viewpoint of a Zoologists, New York: Columbia University Press. Mayr, E. (1959) “Darwin and the Evolutionary Theory in Biology,” in B. J. Meggers (ed.), Evolution and Anthropology: A Centennial Appraisal, Washington, D.C.: Anthropological Society of Washington, pp. 1–10. Millikan, R. G. (1999) “Historical Kinds and the “Special Sciences”,” Philosophical Studies 95: 45–65. Mishler, B. D. and R. N. Brandon (1987) “Individuality, Pluralism, and the Phylogenetic Species Concept,” Biology and Philosophy 2: 397–414. Okasha, S. (2002) “Darwinian Metaphysics: Species and the Question of Essentialism,” Synthese 131: 191–213. Piotrowska, M. (2018) “Meet the New Mammoth, Same as the Old? Resurrecting the Mammuthus primigenius,” Biology & Philosophy 33: 5. Putnam, H. (1975) “The Meaning of ‘Meaning’,” in Mind, Language and Reality: Philosophical Papers, Vol. 2, Cambridge: Cambridge University Press, pp. 215–271. Rieppel, O. (2005) “Monophyly, Paraphyly, and Natural Kinds,” Biology and Philosophy 20: 465–487. Rieppel, O. (2006) “The Phylocode: A Critical Discussion of Its Theoretical Foundation,” Cladistics 22: 186–197. Rieppel, O. (2007) “Species: Kinds of Individuals or Individuals of a Kind,” Cladistics 23: 373–384. Rieppel, O. (2008) “Origins, Taxa, Names and Meanings,” Cladistics 24: 598–610. Rieppel, O. (2009) “Species as a Process,” Acta Biotheoretica 57: 33–49. Rieppel, O. (2013) “Biological Individuals and Natural Kinds,” Biological Theory 7: 162–169. Siipi, H. and L. Finkelman (2017) “The Extinction and De-Extinction of Species,” Philosophy & Technology 30: 427–441. Simpson, G. G. (1961) Principles of Animal Taxonomy, New York: Columbia University Press. Slater, M. H. (2013) Are Species Real? An Essay on the Metaphysics of Species, New York: Palgrave Macmillan. Slater, M. H. (2015) “Natural Kindness,” British Journal for the Philosophy of Science 66: 375–411. Sober, E. (1980) “Evolution, Population Thinking, and Essentialism,” Philosophy of Science 47: 350–383. Van Valen, L. (1976) “Ecological Species, Multispecies, and Oaks,” TAXON 25: 233–239. Wagner, G. P. (2014) Homology, Genes, and Evolutionary Innovation, Princeton: Princeton University Press. Walsh, D. (2006) “Evolutionary Essentialism,” British Journal for the Philosophy of Science 57: 425–448. Wiggins, D. (1980) Sameness and Substance, Cambridge, MA: Harvard University Press. Wilkins, J. S. (2013) “Essentialism in Biology,” in K. Kampourakis (ed.), The Philosophy of Biology: A Companion for Educators, Dordrecht: Springer, pp. 395–419.

289

Ingo Brigandt Wilkins, J. S. (2018) Species: The Evolution of the Idea, Boca Raton: CRC Press. Wilson, R. A., M. J. Barker and I. Brigandt (2007) “When Traditional Essentialism Fails: Biological Natural Kinds,” Philosophical Topics 35: 189–215. Winsor, M. P. (2003) “Non-Essentialist Methods in Pre-Darwinian Taxonomy,” Biology and Philosophy 18: 387–400. Witt, C. (2011) “What Is Gender Essentialism?” in C. Witt (ed.), Feminist Metaphysics: Explorations in the Ontology of Sex, Gender and the Self, Dordrecht: Springer, pp. 11–25.

290

19 IDENTITY, PERSISTENCE, AND INDIVIDUATION 1 Maria Scarpati

Persistence is mainly discussed by metaphysicians in either of two senses. First, there is the question of how things manage to persist through time per se—i.e., of what happens when something exists at different times. Second, there is the issue of persistence conditions—of which changes a given thing can undergo while surviving, i.e., while staying the same. Although both debates have connections with issues of essentialism and essentialist claims, these are significantly more manifest and direct in the case of the latter debate. This chapter will thus mainly focus on the issue of conditions of persistence. Such an issue presents metaphysicians with several puzzles and theoretical problems. Among these, the focus here will be on two of such riddles that are most tightly connected with essentialism—in particular, on two that can be answered exactly by endorsing some essentialist claim. The discussion of those riddles on the one hand, and of their essentialist solutions on the other will also give us a chance to appreciate the distinction between the notion of an essential property and that of an individual essence. The latter notion will be related to the issue of individuation, for which it plays a crucial role.

19.1

Persistence and Persistence Conditions

Persistence is mainly discussed by metaphysicians in either of two senses. First, there is the question of how things manage to persist through time—i.e., to exist at different times. How is it that you, for instance, existed as a newborn a certain number of years ago, and still exist as a grownup now? And is the claim that the newborn you and the grownup you are numerically identical really viable, given that being a newborn and being a so-and-so-many years old grownup are incompatible features?2 Metaphysicians who hold that it is in fact viable typically defend a form of Endurantism—the idea that whenever a thing x persists, it does so by existing, numerically identical, at different times or, as Lewis (1986: 202) put it, “by being wholly present at more than one time”. Those who take that idea to be ultimately untenable, by contrast, typically endorse either a form of Perdurantism or what is known as the Stage View (the latter also goes under the label “Exdurantism”). Perdurantists hold (as Lewis did) that some things do persist and that anything that persists does so “by having different temporal parts, or stages, at different times, though no one part of it is wholly present at more than one time” (ibid.). Exdurantists hold instead that nothing, strictly

DOI: 10.4324/9781003008750-23

291

Maria Scarpati

speaking, persists. You are the so-and-so-many-years-old grownup that exists at this very instant, and the abovementioned newborn that we would tend to identify with you has not strictly speaking persisted at all—not even by being a temporal part of something (you) that has a grownup temporal part existing now. Nothing has both those human beings as parts. Every object is an instantaneous stage. Second, there is the issue of persistence conditions—of which changes a given thing can undergo while surviving, i.e., while staying the same. (One would of course be tempted to say: while staying numerically the same. But that is correct only if Endurantism is true. In effect, a perdurantist might wish to hold that while you persist by having different temporal parts at different times, all your temporal parts must be, e.g., human beings, so that you would cease to exist if you were, for instance, to be turned by some magician into a manatee.) In other words, given a thing with features F1, F2, … , Fn, which of these features are such that the thing cannot lose any of them without going out of existence? Note that an answer to the second question about persistence need not be an answer to the former—nor viceversa. You and I might agree that, e.g., being human is the only feature a human being cannot lose without going out of existence, and yet disagree on how human beings persist through time. You might e.g. be an endurantist: you will then hold that every human being persists by being wholly present at each instant of its existence. I, as a perdurantist, will hold instead that every human being persists by having a proper temporal part at each instant of its existence. Still, as I said, we agree that any human being will cease to persist if it stops being human, but will survive any other change in its features. Conversely, suppose you and I are both endurantists about human beings. This does not entail that we agree on which features are such that a human being cannot lose any of them without going out of existence. You might hold, for instance, that being human is one such feature, while I hold instead that it is not—so that you would, indeed, survive being turned into a manatee. Both debates have connections with essentialism and with specific essentialist claims. However, these are significantly more manifest and direct in the case of the latter debate. As a matter of fact, if the modal account of essence is accepted and unless something like Leibniz’s superessentialism is true, the essential features of a thing are a proper subclass of the features the thing cannot lose without ceasing to exist (cf. the chapter “Modal Conceptions of Essence” in this volume). The reason why the two classes of features cannot be taken to simply coincide is the following. Some features are such that, whether or not they are essential or even necessary to the things that have them, they cannot, once had, be lost. For instance, you were born from your parents, and you cannot cease having been born from them, quite independently from whether you essentially were born from them, and even from whether being born from them is a necessary—though perhaps inessential—feature of yours. Thus, one might hold that you could have been born from different parents, or even brought into existence by God a 30-year-old and out of thin air, in which case you would have not been born from any parents whatsoever, and still maintain that having been born from your parents is a feature you cannot lose. For the thesis that one essentially has the parents one actually has, see the chapter “Origin Essentialism” in this volume. Moreover, the connections between essentialist theses and the former debate I mentioned— that as to how things persist—become evident only once endurantists, perdurantists, and exdurantists take a stance as concerns certain issues that are crucially entwined with that of persistence conditions. Some of these issues are typically approached in the form of certain longstanding puzzles and theoretical problems that arise in treating them. I will use the word “riddles” to refer to all such puzzles and problems, and to two of these, I will now turn.

292

Identity, Persistence, and Individuation

19.2

Persistence Conditions and Their Riddles

Most of us get to philosophy with at least some rather sharp pre-theoretical beliefs about persistence conditions. For instance, we tend to think that Brian May would survive losing a lock of hair, and that the statue Psyche Revived by Cupid’s Kiss by Antonio Canova would not survive being smashed. However, these and other pre-theoretical opinions that we tend to view as unquestionably true are among the sources of certain quite challenging riddles in the metaphysics of material objects. While there are many such riddles, in what follows I will introduce only two of them. Those two riddles, I take, are among those that are most clearly related to essentialism in that each one of them might be answered exactly by endorsing some given essentialist thesis. The essentialist theses in question will be presented in the next section.

19.2.1 The Statue and Its Matter You might be surprised to hear that a certain big portion of marble is worth a round trip to Paris, a ticket for the Louvre Museum, and 90 minutes queuing. Your astonishment would probably fade if I added that the portion of marble in question was chiseled by Antonio Canova into Psyche Revived by Cupid’s Kiss, one of the most celebrated masterpieces of Neoclassical sculpture. But then you might be even more bewildered now than you were at the idea of going to Paris just to see a portion of marble. Is the statue really the same thing as the portion of marble? Pre-theoretical beliefs about persistence conditions seem to suggest otherwise. One might well think that the statue would not survive being smashed, while the portion of marble would—though scattered, the very same portion of marble would still exist after the unfortunate event. Thus, one would say, the statue and the portion of marble (“Psyche” and “Marble”, henceforth) are not numerically identical. For they differ in their features: to mention but one way they do, Marble is such that it would survive being smashed, while Psyche is not. Hence, that they are two—that they are numerically different—seems to be dictated by Leibniz’s Law, i.e., by the Indiscernibility of Identicals, which states that for any x and for any y, if x is numerically identical to y then x and y do not differ in any respect. This line of thought led many to conclude that a material object and its matter are, indeed, not identical. If they are right, there are as a matter of fact two things—a statue, Psyche, and a portion of marble, Marble—under one and the same inventory number at the Louvre Museum. More importantly, there are two things at exactly the same place at the same time, and these also seem to have exactly the same parts. Now, even if we were to simply take it as unproblematic that two things can thus occupy the very same region of spacetime and have the same parts as well, the claim that Psyche is numerically different from Marble has worrisome consequences and some still hold that, popular beliefs about persistence conditions notwithstanding, a material object and its matter are one and the same after all. For instance, Psyche has a given weight—of, say, n kilograms. If asked what the weight of Marble is, one would say it is n kilograms, as well. But then those who take Psyche and Marble to be two things (“two-thingers” henceforth) should at least explain away the intuition that, if they are right, then what occupies r (where r is the spacetime region occupied by Psyche—and of course by Marble) should have a total weight of 2xn kilograms—not, as it happens, of only n kilograms (Lewis 1986: 252). After all, twothingers literally hold there to be two things each of which weighs n kilos at r! To mention but one more difficulty for the two-thinger, once it is accepted that there is more than one thing occupying the whole spatiotemporal region r, it is hard to see why one

293

Maria Scarpati

should stop at holding there to be exactly two such things, as the two-thinger aims to do. In effect, the argument the two-thinger delivered goes from the acknowledgment of two distinct “persistence conditions profiles” possessed at r to positing two distinct entities at r via Leibniz’s Law. (Roughly, the persistence conditions profile of a thing would be the collection of all its persistence conditions, and it would specify all the changes that the thing can survive and all those that the thing cannot survive). But what if more than two such profiles are instantiated at r? If so—if, say, 2+n persistence conditions profiles are instantiated at r (for n any number equal to or greater than 1), the two-thinger will have to posit, by their same reasoning, 2+n entities at r. (See e.g. Sosa 1999; Fairchild 2019. Cf. also Hirsch 1982: 32–3; Yablo 1987).

19.2.2

The Grounding Problem

Be that as it may, many stick to the idea that at least two things—Psyche and Marble—occupy the entirety of the abovementioned region r. But then a further problem arises. Two-thingers or, more generally speaking, multi-thingers typically endorse both of the following claims: i Psyche and Marble differ with respect to some of what Bennett (2004) would call their “sortalish properties”.3 In particular, they differ in their persistence conditions and have, more generally, different modal profiles. ii Psyche and Marble do not differ in any of their “non-sortalish properties”. They have, in particular, exactly the same mereological and microphysical structure. Two comments here. 1 “Sortalish properties” are, for Bennett (2004: 341), “persistence conditions, particularly modal properties like being essentially shaped about like so”, “kind or sortal properties”, and “properties that things have partially in virtue of their instantiation of properties” which belong to either of the former two classes. All properties that are not sortalish in this sense are non-sortalish. 2 Saying that a and b have different modal profiles is to say that not everything that is necessarily true of the one is also necessarily true of the other, and that not everything that is possibly true of the one is possibly true of the other (the two claims are inter-definable and they entail one another). A difference in persistence conditions is a difference in modal profile. For instance, Psyche would not survive being smashed inasmuch as Psyche must have by and large the shape Canova made it into—what is not the case for Marble. But there are differences in modal profile that are not differences in persistence conditions. For instance, Marble could have been chiseled by someone other than Canova, or even by Canova himself, into an utterly different shape than Psyche’s shape—intuition suggests that this is not the case for Psyche. That is a difference in modal profile while not being a difference in persistence conditions as well. More precisely, it concerns Marble’s persistence conditions (for Marble could have survived being chiseled into the shape of, e.g., a pumpkin) but not Psyche’s (for it seems that Psyche would have never existed had Canova decided to chisel Marble into the shape of a pumpkin. Thus, issues of Psyche’s persistence conditions would never have arisen). Now, there seems to be a tension—or, some might rather say, a gap—between i. and ii above. If ii. is true, i.e., if Psyche and Marble are alike in absolutely every non-sortalish respect, there

294

Identity, Persistence, and Individuation

seems to be nothing non-sortalish about them that can ground the differences in their modal profiles and, more generally, the sortalish differences between them—nothing those differences can be based upon. This issue is often referred to as “the grounding problem”, but this does not mean that it need be phrased or answered in terms of the technical notion of grounding as construed and formalized in works such as Correia and Schnieder (2012), Fine (2012), and Bliss and Trogdon (2021). Rather, in the present context saying that the modal (and, more generally, sortalish) differences between Psyche and Marble seem to lack a ground is simply to say that they seem to lack something like a support in the non-sortalish features of Psyche and of Marble—that no non-sortalish difference-maker seems available to produce the sortalish differences in question. (As pointed out in Wasserman 2002, the problem was also phrased in terms of supervenience (Rea 1997), indiscernibility (Olson 2001) and, as in Wasserman’s own preferred approach, explanation. I chose the label “grounding problem” because it seems to be the most commonly employed in recent literature; see for instance Bennett 2004; Sidelle 2014; Koslicki 2018b; Kriegel 2019. Cf. Zimmerman 1995). The grounding problem thus demands of the many-thinger that she either points out what she takes to ground the differences mentioned in i. above, or bites the bullet and admits that such differences lack a ground, in which case she must provide support for the idea that this is not a hard row to hoe after all—but see Koslicki (2018b: 335) for the idea that the onethinger also faces a version of the grounding problem and Bennett (2004: 340) for an even broader take on the issue. See also deRosset (2011) for the idea that under no characterization the grounding problem is both a legitimate one and a genuine challenge to the multi-thinger. Now, one might think that the grounding problem is staggeringly exacerbated if one is not just a two-thinger, but a many-thinger of roughly the sort I started to sketch above—one who holds, in particular, that for every modal profile possibly instantiated at a region r there is a thing that instantiates that modal profile at r, and thus that for n different modal profiles instantiated at a region r there are n different entities co-located at r, where n is greater (perhaps even significantly greater) than 2. However, as I will make clear in the next section, a way to bite the bullet as concerns the grounding problem and to support the idea that this makes for a tenable view takes its lead exactly from the acceptance of that very many-thinger attitude in its most extreme version.

19.3

Essentialist Takes on the Riddles

As I said, certain essentialist views provide solutions to—or at least viable attitudes towards—the riddles that persistence conditions face us with.4 I will now present some essentialist takes on the two riddles considered above.

19.3.1

A Statue, Its Matter, and Their Kinds

A by now classical take on the puzzle concerning Psyche and Marble has it that there are indeed two things at r—the spacetime region at which Psyche Revived by Cupid’s Kiss is located. Psyche and Marble are such things. That they differ in their persistence conditions, which via Leibniz’s Law proves (the thought goes) that they are numerically different, is only to be expected inasmuch as the two also differ in kind. Psyche is a statue, while Marble is not: it is a portion of marble. Psyche, in turn, is not a portion of marble.

295

Maria Scarpati

The view I am considering holds that some things—among them, those things that can coincide with others, belong to their kinds essentially (that is, they satisfy the thesis called “essentiality of kind membership”—cf. Bird and Tobin 2022) and that different kinds can bestow different persistence conditions on their members. It is the case for Psyche’s kind—statue—and for Marble’s kind—amount of marble. While I refer the reader to the chapter “Natural Kind Essentialism” for an extensive treatment of the topic, two clarifications are in order here. First, the approach I am considering must be one that ascribes essential features to those things that are the members of kinds—in particular, to Psyche and Marble, and not only to kinds themselves such as the kind statue and the kind amount of marble. Thus, the stress it puts on talk of kinds should not lead one to think that kinds themselves are ascribed certain essential properties while the individuals belonging to them are not (cf. the chapter “Essences of Individuals” in this volume). Second, in order to provide an answer to the puzzle concerning Psyche and Marble, advocates of this view must take at least some kinds that are not natural kinds to bestow persistence conditions on their members and to be such that said members belong to them essentially. Otherwise, the view would have per se nothing to say about statues inasmuch as statue is an artefactual—as opposed to a natural—kind. (For essence and artifacts, see the chapter “Artifacts, Artworks, and Other Social Kinds” from this volume. Cf. also, for artefactual kinds, Franssen, Kroes, Vermaas, and Reydon eds. 2013 and literature therein.) The debate on the view I am considering and on related ones was sometimes phrased in terms of sortals rather than in terms of kinds: these are, indeed, the terms in which parts of the landmark contributions of David Wiggins on the topic were phrased (see in particular Wiggins 1968 and Wiggins 2001). However, the phrasing in terms of kinds is the one that is widely employed in recent literature. Moreover, I take it that the idea, which is nowadays considered rather contentious, of there being a systematic correspondence between kinds and sortals largely motivated Wiggins and others to employ the latter notion. That idea seemed in effect to be implicitly endorsed by Wiggins when in the locus classicus for two-thingers Wiggins (1968) he offered the following restriction on the principle that would ban numerically different things located at exactly the same place at the same time: “No two things of the same kind (that is, no two things which satisfy the same sortal or substance concept) can occupy exactly the same volume at exactly the same time”.5 Now, by presenting the view I have been describing as a viable answer to the riddle of Psyche and Marble, I do not mean in any way to suggest that it is a solution to it if a solution to the riddle is expected to resolve all of its problematic aspects and all the potential weaknesses of two-thingism. Despite not being a solution in that sense, however, the conjunction of (a.) essentialism of kind membership and (b.) the claim that different kinds bestow different persistence conditions on their members offers a framework that is particularly congenial for two-thingers (and to multi-thingers more generally). In effect, even supposing that the conjunction of (a.) and (b.) is accepted, there are still open questions such as that of why the weight of Psyche and that of Marble do not add up. More importantly, the two-thinger who endorses essentialism of kind membership and holds that different kinds bestow different persistence conditions on their members still owes us answers concerning: • the grounding problem: how is it that Psyche and Marble, alike in every non-sortalish way, differ in sortalish ones—in particular, how is it that they belong to different kinds?, and:

296

Identity, Persistence, and Individuation

• the arbitrariness worry: if there is more than one thing at r, why two and not, say, three or twenty or ninety-nine? If one-thingism is to be deserted, how can one be sure that exactly two persistence conditions-profiles are instantiated—and thus that exactly two things exist—at r? In the next subsection, I will consider not only some attempts to ground the sortalish differences between Psyche and Marble, but also a view that was proposed as a viable reaction both to the grounding problem and to the arbitrariness worry, albeit one that bites the bullet, and thus denies that the sortalish differences between statue and matter are grounded at all. For the issue concerning the weight of Psyche and Marble, I refer the reader to Lewis 1986: 252, Zimmerman 1998: 293–4, Wasserman 2002, Walters 2019, and Lipman 2021.

19.3.2

Parts, Forms, and Plenitude.

The grounding problem is widely viewed as the most serious threat to the two-thinger’s view—to the point of having been labeled “the standard objection” against it (Wasserman 2002). (To be precise, Wasserman presents the grounding problem as the standard objection to the endurantist two-thinger. In effect, for reasons that I cannot delve into here, perdurantists and exdurantists are typically one-thingers, and endurantists—or at least contemporary ones—are typically two-thingers or, more generally, multi-thingers. However, a multi-thinger could at least in principle endorse either perdurantism or the stage view. I submit that the metaphysician in question would face the grounding problem no less than their endurantist counterparts do). Early cases to the effect that the sortalish differences between Psyche and Marble (and of course between the members of any pair in an akin predicament) are grounded after all were strongly rebutted—see Olson’s (2001) answers to Baker (1995); (1997), Rea (1997), and Shoemaker (1999); see also Bennett (2004): § 2–3. (I should mention that in approaching the topic of coincident objects Olson is chiefly interested in cases concerning a person and their body—and in arguing for the claim that the members of any such pair are numerically identical. Cf. Olson (1997). For essence and persons, see the chapter “Persons” in this volume). In the current debate, some substantive attempts to ground the sortalish differences between distinct coincident objects focus on mereological considerations, to the effect that (e.g.) Psyche and Marble “differ with respect to their parts”, as in Wasserman (2002: 197), or even fail to share any part, as in Lipman (2021). For present purposes, an answer to the grounding problem that was advanced in the context of hylomorphism has a particular relevance. Contemporary hylomorphism is, generally speaking, a neo-Aristotelian view that takes objects to be literally composed of matter and form. Fine (1999), (2008) and Koslicki (2008); (2018a); (2018b), in particular, maintain that the form of a structured thing is one of the thing’s parts.6 The hylomorphic response to the grounding problem is especially interesting in the present context inasmuch as its tenability crucially rests on essentialist tenets and on a specific nonmodal conception of essentialist notions. Very roughly, the idea is that hylomorphism allows one to hold that Psyche and Marble, despite being perfectly alike in all their non-sortalish features, differ in as much as some such features figure in Psyche’s essence but not in Marble’s, and vice versa. For instance, it can be held that Psyche essentially has the shape of two creatures who are about to kiss, while Marble has such a shape only accidentally, and that Marble essentially weighs exactly n kilos, while Psyche only accidentally does. By accepting a

297

Maria Scarpati

non-modal conception of essentialist notions, the hylomorphist can deny that such essential differences between Psyche and Marble are modal, and thus she can take them to ground the modal differences—and eventually all sortalish differences—between them. As Koslicki (2018b: 360) put it, “[a] response to the Grounding Problem which utilizes a non-modal conception of essence will of course not serve to explain all the differences which obtain between numerically distinct spatiotemporally coincident objects in terms of other more basic features; rather, it assumes that all such explanations must eventually bottom out once we have reached certain basic non-modal facts about essences”. Presumably, the thought is that the hylomorphist can take the availability of such facts to be guaranteed by the coincident objects’ being mutually different compounds of form and matter, where the form of a thing plays a crucial role in determining which features of the thing figure in its essence, and thus in the basic, non-modal facts about it. But I do not mean to suggest that this is literally what Koslicki—or any hylomorphist—defends; in particular, I certainly cannot do justice to all the subtleties of Koslicki (2018b) here. The idea that hylomorphism can indeed provide a “distinctive solution to the grounding problem” was challenged by Sidelle (2014), to which the abovementioned Koslicki (2018b) responded in detail. Be that as it may, hylomorphism is a demanding view and the multithingers who are not independently attracted to its theses are, unfortunately, rather unlikely to accept them for the sake of solving the grounding problem alone. I would like to conclude this part by considering what was proposed, in Bennett 2004, as a way for the multi-thinger to make sense of primitivism—the claim that the sortalish differences between distinct spacetime coincident objects are brute. Bennett suggested that the multi-thinger can offer “an explanation of the primitiveness” (Bennett 2004: 354) of sortalish facts about distinct coincident objects if she accepts what Sosa (1999: 133) called “explosion of reality”—although the view currently goes by more sympathetic labels such as “material plenitude” (Fairchild 2019: 144) and “essentialist plenitude” (Kriegel 2019: 16). I take the following one by Kriegel (2019: 18) to be the neatest characterization of the claim to this day: “For any region R occupied by object O with non-sortalish properties P1, … , Pn and, such that there are m logically permissible but metaphysically impermissible distributions of essentiality and accidentality across P1, … , Pn, (i) there are in R exactly 2n–m coincident objects, each with its unique metaphysically permissible distribution of essentiality and accidentality across P1, … , Pn, and (ii) 2n vastly outnumbers m”. Presumably, the permissible distributions are those that result in modal profiles which can (as a matter of metaphysical possibility) be instantiated. However that idea is to be characterized, it is crucial for plenitude to hold that (ii) from Kriegel’s fraising is true, i.e., that “when we compare the number of metaphysically permissible distributions of essentiality/accidentality to the number of metaphysically impermissible distributions, we find that the former vastly outnumber the latter” (ibid.: 17). Informally, the plenitude claim in question is an extreme version of a thesis I started to sketch earlier. It holds that for n non-sortalish properties instantiated at a region r, every metaphysically permissible distribution of essentiality and accidentality over those properties makes for a modal profile that is indeed instantiated at r and that there is a one-one correspondence between the modal profiles instantiated at r and the entities that instantiate a modal profile (and thus exist) at r. (This is how the view is standardly stated. As a matter of fact, it seems to me that it might as well be phrased in terms of necessity and contingency instead of essentiality and accidentality. The view would be one that holds that for every permissible distribution of necessity and contingency over the non-sortalish properties instantiated at a region there is

298

Identity, Persistence, and Individuation

exactly one object at that region. However, the standard formulation in terms of essentiality and accidentality might open the way to a view that as far as I can see was never attempted before: one that defends both essentialist plenitude and the idea that non-modal facts about the essence of objects can ground the modal differences between things—à la Koslicki 2018b. It might indeed be interesting to see whether the resulting view would have any theoretical edge over other accounts of coincidence). Very roughly, Bennett’s suggestion is that, far from exacerbating the grounding problem for multi-thingers, essentialist plenitude allows the multi-thinger to dissolve the sense of mystery which made one look for a support to the sortalish differences between distinct coincident objects in the first place. At least, that is what I take the gist of her suggestion to be: “The idea is that because all of the complete modal profiles possible in a given spatiotemporal location are instantiated there, there is no contrast to be drawn between those that are instantiated and those that are not. We cannot expect there to be anything special about some particular modal profile M in virtue of which it (and not others) is instantiated in that place. That is why it is primitive that M is instantiated there”. Opinions may of course differ as to whether the view depicted (and, I should be clear, not endorsed) by Bennett is one that does not itself owe any further answer to the grounding problem. What is certainly the case, however, is that such a view does not face the arbitrariness worry I mentioned earlier. In effect, its advocate takes there to be a one-to-one correspondence between the modal profiles instantiated at a region and the objects existing at that very region. Thus, she certainly cannot be accused of positing distinct objects when faced with certain differences in modal profile while sticking to one object when faced with other differences in modal profile. For the idea that the only solid takes on material coincidence are one-thingism and plenitude, see Bennett (2004: 355), Johnston (2006: 697–8) and Leslie (2011: 278–80). One might object that plenitude still leaves a point unexplained: i.e., the contrast between those distributions of essential (necessary) and accidental (contingent) over non-sortalish properties that are metaphysically permissible and those that are not. In other words, plenitude does not per se explain the fact that certain modal profiles can be instantiated at a region while others cannot. It merely has it that every modal profile that can be instantiated is instantiated by exactly one object. Opinions may certainly diverge on this, but I think that the plenitude endorser should be granted at least two important points. First, that particular form of arbitrariness that affects the two-thinger’s view is indeed avoided under plenitude. For, among all the modal profiles that can be instantiated at a region, the two-thinger has it that only two are instantiated while all others are not. Plenitude, by contrast, does not have any bias for certain possibly instantiated modal profiles over others. Second, it is far from clear that other approaches towards the grounding problem need not count on the same distinction between permissible and non-permissible modal profiles that plenitude accepts. Setting Meinongianism aside, there are ways something can be, and ways nothing can. As metaphysicians, we typically take this to be as basic a fact about reality as any. It would be somewhat unfair to hold against the plenitude endorser that they cannot explain that fact—what we do not hold against metaphysicians of other tendencies—merely because plenitude entails that every way something can be in a region is a way something is in that region.

19.4 Individuation and Individual Essence The above allows one to appreciate a distinction that is crucial when dealing with essentialist theses—in particular, with such theses in as much as they relate to issues of identity.

299

Maria Scarpati

The distinction in question is that between an essential feature and an individual essence. If essence is understood as the collection of all the features of a thing that the thing possesses essentially, a related distinction must be made between an essence and an individual essence as well. To appreciate the point, recall the essentialist response to the puzzle of Psyche and Marble considered above. The advocate of such a response ascribes essential features to the individual Psyche and to the individual Marble, respectively—where the essential features of the former do not coincide with those of the latter, and the essential differences between Psyche and Marble include essential differences in persistence conditions. But it is only in as much as Psyche belongs to a certain kind—the kind statue—that Psyche has the essential properties the view in question ascribes to it, and the same goes, mutatis mutandis, for Marble, its kind, and its essential features. Thus, it is compatible with the essentialist view in question that every statue has the very same essential features every other statue has and, more generally, that any two things of a kind K have exactly the same essential features. If Psyche has the very same essential features that every other statue has, then Psyche certainly fails to have an individual essence. Roughly, an individual essence is a feature or a collection of features that is both essential to its possessor and necessarily sufficient for identity with its possessor. A little more precisely, we can define the claim that F is an individual essence of a thing a as follows: (IE) F is an individual essence of a =df F is a feature or a collection of features such that (iIE.) F is essential to a, and (iiIE.) necessarily, for any x, if x has F then x = a. (For characterizations of the notion of an individual essence, see Plantinga 1974; 1976, 1979, Adams 1979; 1981, Forbes 1985; 1994: fn. 23, Mackie 2006, Lowe 2008, Roca-Royes 2011, Mackie and Jago 2017). If F is an individual essence of a, then one might be tempted to say that a is individuated by its having F—that having F is what makes it be a. More precisely, the idea is that having F individuates a (it is an individuator of a) if and only if having F is what bestows on a not the feature of being identical to itself, which any entity whatsoever possesses, but a feature that a must have and that nothing else can possibly have: the feature of being identical to a, of all entities. However, there are properties that satisfy (IE) and that would do poorly as individuators of a. For instance, the feature of being identical to a seems to respect (IE), but the claim that being identical to a is what makes something be identical to a does not look like one that can duly answer the question of what makes a be the very thing it is, i.e., a. In other words, that claim seems one that cannot establish a principle of individuation for a. I should mention that the problem of providing a “principle of individuation” was understood in at least three different ways in the history of philosophy. First, it was viewed—particularly in the Middle Ages—as the problem of what makes something be an individual, as opposed to a thing that is not an individual but, for instance, a universal. For this construal, I refer the reader to Gracia 1988 and literature therein. Second, providing a principle of individuation was at times meant as an epistemic task—roughly, as that of figuring out which conditions are necessary and sufficient for us to correctly re-identify a thing as one and the same over its temporal career; see in particular Strawson 1959. Third, and last, a principle of individuation is nowadays rather standardly understood—solely—as an answer to the abovementioned metaphysical question as to what makes a given thing be the very thing it is, in particular, as opposed to every other thing belonging to the same kind; see e.g.

300

Identity, Persistence, and Individuation

Lowe 2003; 2008; 2012. For instance, providing a principle of individuation for Brian May is to answer the question of what makes Brian May be Brian May, as opposed to every other individual, every other human being, and so on. Individuation in the third sense, and in that one alone, is relevant for my present aims. For a traditional take on the connection between the second understanding of “individuation” I mentioned and the third, see Wiggins (2001). Now, to be clear, there is no inconsistency in the idea that being identical to a is an individual essence of a, nor in the idea that such a property is the only individual essence of a.7 However, if the latter is the case, and if it is also the case that a property must be an individual essence of a in order to individuate a, then what we should say is that a is primitively individuated. For a’s being the very entity it is should then be viewed as a brute fact—as a fact that is not brought about by any other fact, or, at least, by any other fact about a itself. Features of being identical to a given thing in particular, such as that of being identical to Brian May, play a prominent role in certain contributions of Robert Adams which are landmark for the debate on metaphysical Haecceitism—roughly, on whether things are individuated by their qualitative profiles. Following Adams himself, such features are referred to as the “thisnesses” of their possessors. Adams (1979) argues that at least some possible things are not individuated by their qualitative profiles inasmuch as they fail to have an individual essence that is purely qualitative. The notion of a thisness is characterized in further detail and duly distinguished from that of a haecceity in Adams (1981); see Plantinga (1976) and Rosenkrantz (1993) for comparison. Adams’s latest take on the topic of thisness is the focus of Chapter 8 in his What Is, and What Is in Itself. The important point here is that one might very well hold that a’s thisness is the only individual essence of a without thereby having to deny that a has other essential features. To make the point more vivid, consider Brian May. One might well hold that Brian May’s thisness is Brian May’s sole individual essence, and still ascribe to Brian May several essential features on top of that one. It might be held, for instance, that Brian May is essentially human, essentially a mammal, essentially born from his parents, essentially capable of rationality, and so on. If by “the essence of Brian May” one means the collection of all the features that Brian May possesses essentially, then will in this case simply be the essence of Brian May. However, under the approach in question none of Brian May’s essential features (other than, of course, his thisness) would satisfy condition (iiIE.) of (IE). And it might well be that, if we were to remove being identical to Brian May from Brian May’s essence, the resulting class of features would not fulfill that condition either. If we put aside for a moment the idea that Brian May’s thisness is an individual essence of Brian May—for instance by accepting, for the sake of discussion, that Della Rocca (1996) and Denby (2014) are right in holding that such a feature is not even essential to Brian May—it can be appreciated that one might even hold that Brian May has no individual essence whatsoever, and still ascribe to him a number of essential features such as the ones mentioned in the previous paragraph. In the resulting view, none of such features and no collection of such features would be necessarily such that anything that has it is Brian May. Thus, none would be suitable for individuating Brian May: in effect, it would be possible for something to have any and even all of them without being Brian May. This would be a view according to which nothing in the essential profile of Brian May is sufficient to individuate him. But I have not yet addressed the question whether a feature of a thing—or a collection of features thereof—must be an individual essence of that thing in order to individuate it. I submit that such a feature or collection of features must indeed be an individual essence of the thing if the notion of an individual essence is understood—as it is, for instance, in Mackie

301

Maria Scarpati

(2006)—in merely modal-existential terms, i.e., if an individual essence is characterized as follows: (IEME) F is an individual essenceME of a =df F is a feature or a collection of features such that, necessarily, Fx if and only if x = a. For a principle of individuation for a thing must indeed provide conditions that are necessarily both necessary and sufficient for identity with that thing. As for the first characterization I offered, (IE), opinions may vary: while a principle of individuation must certainly satisfy condition (iiIE.), I take it that, if “essentially” in condition (i.) is understood in a “post-Finean”, not merely modal or even utterly non-modal fashion then it is unclear that a feature must satisfy (iIE.) in order to serve as an individuator for its possessor. Now, as I said, the doctrine that holds that (1.) things belong to their kind essentially and (2.) different kinds bestow different identity on their members is compatible with things lacking an individual essence on top of their thisnesses. What about the other essentialist claims I presented above—hylomorphism with a non-modal view of essence on the one hand, essentialist plenitude on the other? Do they commit their advocates to any particular claim concerning individual essence? While I believe that the answer is “no” in both cases, this is a two-fold substantive philosophical question that demands substantive investigation.

Related Topics Correia, Fabrice, “Non-Modal Conceptions of Essence” Kovacs, David, “Essence, Grounding, and Explanation” Marabello, Marco, “Essences of Individuals” Robertson Ishii, Teresa, “Origin Essentialism” Sidelle, Alan, and Livingstone-Banks, Jonathan, “Conventionalism” Tahko, Tuomas, “Natural Kind Essentialism” Torza, Alessandro “Modal Conceptions of Essence”

Notes 1 This chapter benefitted from the comments and suggestions of many people. I would like to thank, in particular, the editors of this volume Kathrin Koslicki and Mike Raven, and all the participants in the Routledge Handbook of Essence Workshop. Special thanks also go to Marco Marabello, Gonzalo Rodriguez-Pereyra, and Achille Varzi, and very special ones to Claudio Calosi. My thoughts on essence are deeply indebted to the work of, and to my conversations with, Penelope Mackie, to whom I shall always be thankful. 2 Note that, contra what this way to introduce the issue might seem to suggest, the significance of this first question about persistence need not rely on issues of persistence through change. The question is, first and foremost, a matter of of sustained persistence, and it is meant to arise without regard to whether the object at issue changes or remains unchanged over time. It is, moreover, not obvious that time could pass in the absence of change; cf. Shoemaker (1969). 3 I should mention that while I generally prefer talk of features over talk of properties because I want it to be clear that none of what I say is committed to any form of realism about properties, I shall here go along with Bennett’s phrasing, which is in terms of properties. I take it that it was no intention of Bennett (2004) to phrase the debate as one that is open to realists alone, either. 4 For a very strong case against the idea that issues of persistence support the acceptance of essentialist metaphysics, however, see Sullivan (2016).

302

Identity, Persistence, and Individuation 5 Wiggins (1968): 93. 6 Other views of what it is for a structured entity to have a form are of course available to the hylomorphist. See e.g. Johnston (2006) where form is understood as a principle of unity. 7 Della Rocca (1996) and Denby (2014) argue for definitions of essentiality which happen to count the feature of being identical to a as non-essential to a. Under either of those definitions, such a feature thus fails to satisfy (IE). However, neither of those contributions provides a conclusive argument for the idea that the feature in question should not be considered—as it traditionally has been—essential to its possessor.

References Adams, R. M. (1979) “Primitive Thisness and Primitive Identity”, The Journal of Philosophy 76 (1): 5–26. Adams, R. M. (1981) “Actualism and Thisness”, Synthese 49 (1): 3–41. Baker, L. R. (1995) “Need a Christian Be a Mind/Body Dualist?”, Faith and Philosophy 12 (4): 489–504. Baker, L. R. (1997) “Why Constitution Is Not Identity”, The Journal of Philosophy 94: 599–621. Bennett, K. (2004) “Spatio-Temporal Coincidence and the Grounding Problem”, Philosophical Studies 118 (3): 339–371. Bird, A. and Tobin, E. (2022) “Natural Kinds”, The Stanford Encyclopedia of Philosophy (Spring 2022 Edition), Edward N. Zalta (ed.), forthcoming URL = < https://plato.stanford.edu/archives/spr2022/ entries/natural-kinds/>. Bliss, R. and Trogdon, K. (2021) “Metaphysical Grounding”, The Stanford Encyclopedia of Philosophy (Winter 2021 Edition), Edward N. Zalta (ed.), URL = < https://plato.stanford.edu/archives/win2021/ entries/grounding/>. Correia, F. and Schnieder, B. (2012) “Grounding: An Opinionated Introduction”, in Fabrice Correia and Benjamin Schnieder(eds.), Metaphysical Grounding: Understanding the Structure of Reality, Cambridge: Cambridge University Press. Della Rocca, M. (1996) “Essentialism: Part 1”, Philosophical Books 37 (1): 1–13. Denby, D. (2014) “Essence and Intrinsicality”, in Robert M. Francescotti (ed.), Companion to Intrinsic Properties. De Gruyter. DeRosset, L. (2011). “What Is the Grounding Problem?”, Philosophical Studies 156 (2): 173–197. Fairchild, M. (2019) “The Barest Flutter of the Smallest Leaf: Understanding Material Plenitude”, The Philosophical Review 128 (2): 143–178. Fine, K. (1999) “Things and Their Parts”, Midwest Studies in Philosophy 23 (1): 61–74. Fine, K. (2008) “Coincidence and Form”, Aristotelian Society Supplementary Volume 82 (1): 101–118. Fine, K. (2012) “A Guide to Ground”, in Fabrice Correia and Benjamin Schnieder (eds.), Metaphysical Grounding: Understanding the Structure of Reality, Cambridge: Cambridge University Press. Forbes, G. (1985) The Metaphysics of Modality, Oxford: Oxford University Press. Forbes, G. (1994) “A New Riddle of Existence”, Philosophical Perspectives 8: 415–430. Gracia, J. E. (1988) Individuality: An Essay on the Foundations of Metaphysics, State University of New York Press. Hirsch, E. (1982) The Concept of Identity, Oxford: Oxford University Press. Johnston, M. (2006) “Hylomorphism”, The Journal of Philosophy, 103 (12): 652–698. Koslicki, K. (2008) The Structure of Objects, New York: Oxford University Press. Koslicki, K. (2018a) Form. Matter, Substance, Oxford: Oxford University Press. Koslicki, K. (2018b) “Towards a Hylomorphic Solution to the Grounding Problem”, Royal Institute of Philosophy Supplements to Philosophy, 82: 333–364. Kriegel, U. (2019) “Essentialist Plenitude and the Semantics of Proper Names”, Metaphysics 2 (1): 16–25. Leslie, S.-J. (2011) “Essence, Plenitude, and Paradox”, Philosophical Perspectives 25 (1): 277–296. Lewis, D. (1986) On the Plurality of Worlds. Oxford: Blackwell. Lipman, M. (2021) “In defense of Disjointism”, Inquiry, DOI: 10.1080/0020174X.2021.2006773 Lowe, E. J. (2003) “Individuation”, in Michael J. Loux and Dean Zimmerman (eds.), The Oxford Handbook of Metaphysics, Oxford: Oxford University Press.

303

Maria Scarpati Lowe, E. J. (2008) “Two Notions of Being: Entity and Essence”, Royal Institute of Philosophy Supplement, 62: 23–48. Lowe, E. J. (2012) “Asymmetrical Dependence in Individuation”, in Fabrice Correia and Benjamin Schnieder (eds.), Metaphysical Grounding: Understanding the Structure of Reality, Cambridge: Cambridge University Press. Mackie, P. (2006) How Things Might Have Been. Oxford: Oxford University Press. Mackie, P. and Jago, M. (2017) “Transworld Identity”, The Stanford Encyclopedia of Philosophy (Winter 2017 Edition), Edward N. Zalta (ed.), https://plato.stanford.edu/archives/win2017/entries/ identity-transworld/. Olson, E. (1997) The Human Animal: Personal Identity Without Psychology, Oxford: Oxford University Press. Olson, E. (2001) “Material Coincidence and the Indiscernibility Problem”, Philosophical Quarterly 51 (204): 337- 335. Plantinga, A. (1974) The Nature of Necessity, Oxford: Clarendon Press. Plantinga, A. (1976) “Actualism and Possible Worlds”, Theoria 42 (1-3): 139–160. Plantinga, A. (1979) “De Essentia”, Grazer Philosophische Studien 7: 101–121. Rea, M. (1997) “Supervenience and Co-Location”, American Philosophical Quarterly 34 (3): 367–375. Roca-Royes, S. (2011) “Essential Properties and Individual Essences”, Philosophy Compass 6 (1): 65–77. Rosenkrantz, G. (1993) Haecceity. Dordrecht: Kluwer. Shoemaker, S. (1969) “Time without Change”, Journal of Philosophy 66 (12): 363–381. Shoemaker, S. (1999) “Self, Body, and Coincidence”, Aristotelian Society Supplementary Volume 73 (1):287–306. Sidelle, A. (2014) “Does Hylomorphism Offer a Distinctive Solution to the Grounding Problem?”, Analysis 74 (3):397–404. Sosa, E. (1999) “Existential Relativity”, Midwest Studies in Philosophy 23 (1): 132–143. Strawson, P. F. (1959) Individuals: An Essay in Descriptive Metaphysics, London: Routledge. Sullivan, M. (2016) “Are There Essential Properties? No”, in Elizabeth Barnes (ed.), Current Controversies in Metaphysics, New York: Routledge. Walters, L. (2019) “Are the Statue and the Clay Mutual Parts?”, Noûs 53 (1): 23–50. Wasserman, R. (2002) “The Standard Objection to the Standard Account”, Philosophical Studies 111 (3): 197–216. Wiggins, D. (1968) “On Being in the Same Place at the Same Time”, The Philosophical Review 77 (1): 90–95. Wiggins, D. (2001) Sameness and Substance Renewed, Cambridge: Cambridge University Press. Yablo, S. (1987) “Identity, Essence, and Indiscernibility”, Journal of Philosophy 84 (6): 293–314. Zimmerman, D. (1995) “Theories of Masses and Problems of Constitution”, The Philosophical Review 104 (1): 53–110. Zimmerman, D. (1998) “Criteria of Identity and the Identity Mystics”, Erkenntnis 48 (2–3): 281–301.

304

20 ESSENCE, GROUNDING, AND EXPLANATION David Mark Kovacs

20.1

Introduction

How is essence related to laws and explanation? A number of things could be meant by this question. One issue concerns the role of essences qua explanantia, i.e. the ability of facts about essence to explain certain other facts. This potential role of essence is discussed in many other chapters of this collection (see, for example, the contributions of Scarpati on persistence and individuations, Loets on persons, Brigandt on biological species, and Vaidya and Wallner, Griffith, Passinsky, Rosario, and Mallon on various aspects of social ontology). A different question is whether essences underlie any distinctive type of explanation. That is, do essences play any role in explanations not qua explanantia (that which explains) but as the link that connects some explanantia to an explanandum (see Schaffer 2017 and Kappes 2021 for this distinction)? But here, too, we can ask at least two different questions. One concerns the role of essences in scientific explanation. Various aspects of this issue occupy other chapters of this volume (see the contributions of Tahko on natural kinds essentialism, Dumsday on scientific essentialism, and Lam on dispositional essentialism; and in the special sciences, Brigandt on biological species and Brown on psychology and psychiatry). The present chapter’s main focus will be the role of essence in metaphysical explanation: the kind of explanation that philosophers often appeal to when making non-causal “in virtue of” and “because”-claims. Before proceeding, let me state a few assumptions that I will take for granted throughout the chapter, unless noted otherwise. First, I will generally assume a non-modal (“Fine”grained) notion of essence, for the simple reason that this is what most contributors in the relevant literature presuppose (one exception, as we will see in section 1.2, is Meghan Sullivan’s context-sensitive explanatory analysis of essence). However, I won’t make further choices at this stage of the inquiry, for example between constitutive and consequential essence (Fine 1994) or objectual and generic essence (Correia 2006). I will take note of these differences where appropriate. Second, and again in line with most of the literature, I won’t worry much about the relation between metaphysical explanation and grounding. I will usually talk as a unionist: i.e., for brevity of expression I will often use the word “grounding” interchangeably with “metaphysical explanation”. In the few places where the difference is relevant (e.g. Glazier’s

DOI: 10.4324/9781003008750-24

305

David Mark Kovacs

distinction between grounding-based and essence-based metaphysical explanation in section 3), I will explicitly flag it. With these qualifications in mind, we can distinguish four possible approaches to essence and grounding: 1 Unity. Essence and grounding belong to a unified set of explanatory concepts. 2 Supplementation. Essence and grounding don’t mark two separate distinctive types of explanation. Rather, they both contribute (in their own way) to the same one distinctive type of explanation. 3 Independence. Essence is explanatory, but it’s a sui generis notion with no straightforward conceptual links to grounding. 4 Irrelevance. Essence isn’t explanatory at all and lumping it together with explanatory notions is a mistake. Irrelevance has few adherents in the contemporary literature on essence. One recent exception is the trio of Duncan, Miller and Norton (2021: 398–9), who suggest that our intuitions about asymmetric dependence and determination are explained away by asymmetries about essence – which, however, doesn’t itself imply any kind of determination or dependence (nor, presumably, explanation). If we take for granted the standard assumption that modal notions aren’t by themselves explanatory, it’s plausible to think that many advocates of the modal conception of essence (on which see Torza’s entry to his volume) also tacitly subscribe to Irrelevance. Moreover, in the present taxonomy we can count anti-essentialists who deny that there are non-trivial essences as adherents of Irrelevance, too. For the rest of this entry, however, I will focus on Unity, Supplementation, and Independence, which have been far more prevalent in the literature over the past two decades.

20.2 Unity: Essence and Grounding as Closely Unified Unity can be understood broadly as the view that there is a systematic link between essence and grounding, which is strong enough to count them as belonging to the same family of concepts. The link may be reducibility (of one of these notions to the other), or it may be something weaker, such as deep structural similarities between essence and grounding. It’s safe to say that in the contemporary literature on grounding and essence, Unity has been the most commonly adopted position. Several authors have endorsed weaker versions of it since the early days of the grounding literature. These authors don’t propose definitions of either essence or grounding in terms of the other (or of both notions via the same vocabulary), but they nonetheless posit systematic and informative links between them. One thesis that has been particularly influential is Gideon Rosen’s “grounding-reduction link”, which goes as follows (here and in what follows, I use the “[‘, ‘]” notation as a device to form names of facts, so that “[p]” means “the fact that p”): Grounding-Reduction Link: If p reduces to q and p is true, then [q] grounds [p] (2010: 122) Rosen understands reduction in a distinctively metaphysical sense: the phrase “p reduces to q” means that for it to be the case that p is for it to be the case that q. And in turn, the expression “is for” expresses real definition: for p to be the case (partly) consists in q’s being

306

Essence, Grounding, and Explanation

the case; or, q’s essence is part of p’s. Although in later work Rosen (2015) offers a groundtheoretic account of real definition, by default we can treat it as an essentialist locution. Some authors who reject the Grounding-Reduction Link still posit a tight link between essence and grounding. Audi (2012), for example, rejects the link on the basis that grounding is an irreflexive relation between coarse-grained facts, and that p could reduce to q only if [p] = [q], which is inconsistent with irreflexivity. However, he still believes that there is a tight link between essence and grounding: relations of grounding obtain because of the essence of properties involved in the grounding relata. That is, if [Fa] grounds [Gb], this is because it’s in the nature of F that if F instantiated, then something also instantiates G. Dasgupta (2014: 568) floats a similar idea (although without committing himself to it) to answer the question of what grounds the grounding facts. For instance, he suggests that if the fact that e contains people engaged in C-activities grounds the fact that e is a conference, then this complex fact is in turn grounded by e’s containing people engaged in certain activities and its being essential to a conference that if an event contains people engaged in such and such activities, it’s a conference. In short, grounding facts (facts of the form, [Fa] grounds [Gb]) are grounded in facts about the essences of the individuals that occur in the grounding relata (a and b).1 The primary objection to this proposal isn’t specific to the relation between essence and grounding, but rather affects its general structure: why do facts about essence explain facts about grounding? Dasgupta argues that this connection (between essences and grounding facts) isn’t apt to be grounded, whereas Audi seeks to avoid the regress by claiming that the relevant connection between the essences of properties and a grounding fact isn’t itself the grounding relation. Dasgupta, and less explicitly Audi too, appeal here to the widely shared intuition that essentialist facts are in some sense “rock-bottom”: we cannot sensibly ask why something is essentially so-and-so, or why what it is for something to be the case is for something to be so-and-so.2 A further example of a hypothesized close connection between essence and grounding can be found in Trogdon’s (2013) defense of Grounding Necessitarianism, the view that if [p1] … [pn] ground [q], then necessarily, if p1 … pn then q. According to Trogdon, grounding implies lack of cognitive significance with respect to connection questions about the ground and the grounded. That is, if [p1] … [pn] ground [q] then the question of why they do so lacks cognitive significance; and this, in turn, is possible only if there are essential truths about p1 … pn, q or the entities they involve according to which if p1 … pn then q (2013: 467). Similarly to the essentialist account of what grounds the grounding facts, Grounding Necessitarianism (as well as Trogdon’s defense of it) remains a controversial thesis.3 As we can already see from this quick overview, some form of Unity is widely assumed in much of the grounding literature. However, even stronger forms have been defended in the specialized literature on essence, to which we will turn now. I will consider three possibilities: (20.2.1) grounding can be defined in terms of essence; (20.2.2) essence can be defined in terms of grounding; and (20.2.3) both grounding and essence can be defined using the same primitive concepts.

20.2.1 Essentialist Definitions of Grounding Perhaps nobody did as much to explore the logical space of possibilities regarding the conceptual connections between grounding and essence as Fabrice Correia (see also 2.3). Correia’s preferred approach is to express both grounding and essence via an operator rather than a predicate. That is, instead of construing grounding as a relation between facts, he uses the sentential connective “because” to express connections of ground. Similarly, he uses the

307

David Mark Kovacs

already familiar phrase “it is part of what it is to be the case” to express essence. After considering a number of proposals (using different combinations of predicational vs. operational views of grounding and essence, and the notions of objectual vs. generic essence – see Koslicki and Raven’s Introduction to this volume), he tentatively comes down in favor of the following definition: (Grounding-to-essenceCorreia) P because p1, p2, … because

def

i p1&p2& … , and ii It is part of what it is for it to be the case that P that if p1, p2, … , then p (2013: 292) Correia’s account is not without predecessors; Fine (2012) considers and rejects a similar definition. One worry, he argues, is that the account cannot distinguish between conjunctive and plural grounds, since there’s no difference between its being the case that p1, p2, … and its being the case that p1&p2& … (2012: 78). In response, Correia argues that the operator “it is the case that” makes all the difference here: although the plurality of p1, p2 … and the conjunction p1&p2& … are intensionally equivalent, they aren’t substitutable salve veritate in the definition (290–1). Another potential problem is that (Grounding-to-essenceCorreia) has implausible consequences about ontological dependence. Suppose, for example, that existentially quantified facts are grounded in their witnesses. Then [Someone is a philosopher] is grounded in [Socrates is a philosopher]. By (Grounding-to-essenceCorreia), part of what it is for it to be the case that someone is a philosopher is for it to be the case that Socrates is a philosopher. However, essential connections of this sort are often thought to imply a kind of ontological dependence, essential dependence: that is, [Someone is a philosopher] essentially depends on [Socrates is a philosopher]. Fine finds this implausible; the fact that someone is a philosopher “knows nothing”, as he puts it, of Socrates (2012: 75). However, Correia argues that the dependence claim is in fact quite intuitive. For it simply follows from this claim that the nature of the fact that someone is a philosopher involves that if Socrates is a philosopher, then someone is; it doesn’t have the implausible consequence that nobody is a philosopher unless Socrates is (292–3). That is, essential dependence as Correia understands it doesn’t imply existential dependence. This isn’t the last word on (Grounding-to-essenceCorreia). For example, Carnino (2014) offers a reassessment of Fine’s objections and also gives additional ones, which, he argues, overall strengthen (Grounding-to-essenceCorreia) as opposed to alternative formulations that presuppose the relations-between-facts view of grounding. Even with the complications Carnino raises, (Grounding-to-essenceCorreia) remains a serious contender, and critical discussion of it is sure to continue in the years to come. Correia is not the only one to have offered an essence-theoretic reduction of grounding. In a recent paper, Justin Zylstra offered an alternative version of this approach. Zylstra operates with objectual and (in Fine’s sense) constitutive essences. However, unlike Correia, he conceives of grounding as a relation between facts. His proposed definition of grounding in terms of essence goes as follows (lower-case letters are singular and capitalized ones are plural variables bound by quantifiers over facts): (Grounding-to-essenceZylstra) f is grounded in G iffdf For some R, it is essential to f that (f exists only if (For some F, R(fF) and F exist) and R(fG) and G exist) (Zylstra 2019: 5144)

308

Essence, Grounding, and Explanation

To show how the account is supposed to work, Zylstra mentions the case of determinates and determinables as an example. Suppose that [the pen is blue] grounds [the pen is colored]. Plausibly, it’s essential to [the pen is colored] that if it obtains, there are some facts that are determinates of [the pen is colored]. Moreover, indeed [the pen is blue] exists and bears the determinate-determinable relation to [the pen is colored] – just as the definition requires. Zylstra argues that his reductive account of grounding has a number of further advantages, such as explaining several principles about the interaction between grounding and the logical constants (“the impure logic of grounding”), helping us make sense of symmetric cases of metaphysical dependence, and also addressing certain worries about the coarse-grainedness of grounding (cf. Koslicki 2015). Moreover, Zylstra argues, the account is at no disadvantage compared to (Grounding-to-essenceCorreia) when it comes to problem cases about ontological dependence. Essentialist definitions of grounding are of significant interest not only to theorists of essence but also to theorists of grounding, since grounding is standardly understood as an irreducible notion that resists definition in other terms. There is a lot more work to be done in this area, and the prospects of an essentialist analysis of grounding are yet to be thoroughly assessed.

20.2.2

Explanation- and Ground-Theoretic Definitions of Essence

Above we have discussed essentialist definitions of grounding. But one could also try to achieve conceptual economy by going in the other direction and attempting to reduce essence to grounding or, more broadly, explanation. One such recent proposal has been offered by Meghan Sullivan. Unlike most authors in the literature, Sullivan works with a modal conception of essence, i.e. she assumes that an object o has a property P essentiality just in case o has P necessarily. Her starting point is an antirealist view of essence, according to which “essential property ascriptions only hold relative to semantic or metasemantic facts” (2016: 53). This view, she argues, is untenable; however, another account in its vicinity is more promising. According to this view, the ascription of essential properties is relative to explanatory frameworks, as follows: (Explanation-Relative Essentialism): An object o is essentially P relative to framework E iff i o has P ii in any good explanation of type E which involves o, o has P, and iii there are objective norms governing explanations of type E (2016: 56) That essential property ascriptions have a relativistic logical form doesn’t immediately entail that they in fact vary with the framework, but it certainly allows for such variation. Sullivan’s view is only moderately antirealist because what’s framework-relative according to it is which of a thing’s properties are essential, not what properties it has. It’s also worth noting that (Explanation-Relative Essentialism) is not a ground-theoretic reduction of essence per se. Rather, the proposal is to define essence in terms of the general notion of explanation, of which grounding (or metaphysical explanation) is just one species. To illustrate the account, Sullivan gives the following example. Suppose a coin has properties such as conducting electricity and being a unit of account in the financial market. There are objective norms guiding physical explanations, and according to these norms any

309

David Mark Kovacs

physical explanation involving the coin (for example, that it completes an electric circle) will cite its conductivity; however, such explanations won’t cite its being a unit of account. On the other hand, economic explanations are also guided by objective norms, and these norms require that any economic explanation involving the coin (for example why it’s worth ten cents) will have to cite the coin’s being a unit of account, but not its conductivity. Thus (Explanation-Relative Essentialism) predicts that the coin is essentially conductive relative to a physical (but not relative to an economic) explanatory framework, and essentially a unit of account relative to an economic (but not relative to a physical) explanatory framework.4 The view also leaves room for properties that are essential in a distinctively metaphysical sense: for example, the fact that an object has proper parts is metaphysically explained partly by its being extended in space and time but not by its being an automobile. Accordingly, relative to a metaphysical-explanatory framework the object is essentially extended in space but not essentially an automobile. Sullivan’s explanation-relative essentialism is intriguing, but it has also been subject to criticism. Kris McDaniel and Steve Steward (2016) argue that the account is extensionally inadequate: there are properties that are indispensable relative to an explanatory framework, but which intuitively aren’t essential. For example, relative to a framework that explains educational attainments, the fact that Jasmine studies very hard will be a part of any explanation of her 4.0 GPA; yet even relative to that framework, she could stop studying hard without ceasing to exist. So, (ii) isn’t sufficient for a property’s being essential. Moreover, not being a poached egg is an essential property of mine (at least according to the modal view of essence that Sullivan presupposes), yet it doesn’t seem to play an indispensable role in all explanations that involve me relative to any explanatory framework. So, (ii) isn’t necessary for a property’s being essential either (2016: 73–4). Aaron Segal and Noga Gratvol (2021) raise a different kind of objection to Sullivan’s explanatory analysis: the view implies an objectionable mismatch between the logic of essence and the logic of explanation. For (at least on the modal view of essence, which Sullivan presupposes) being an essential property is closed under entailment: if Fa is essentially F and Fa entails Ga, then a is essentially G, too. Explanation, however, doesn’t work this way: it’s not the case that if F is indispensably mentioned in every explanation of a certain type relative to a framework, and Fa entails Ga, then G is also indispensably mentioned in every explanation involving F relative to the same framework. In response, Sullivan could switch from the modal conception of essence to a “Fine-grained” one. However, Segal and Gratvol argue that such a heavy-weight view would not sit comfortably with Sullivan’s frameworkrelative approach. On the other hand, more sophisticated modal accounts (which define essence in terms of necessity and some further condition, such as sparseness or intrinsicality) don’t solve the problem they raise for Explanation-Relative Essentialism, since these views don’t deny that essence is closed under entailment. An explanation-based analysis of essence that (unlike Sullivan’s) isn’t context-relative is Michael Gorman’s foundational account. In formulating this view, Gorman relies on a pretheoretical notion of essence that encompasses a thing’s most central, most crucial features, where, however, this notion of centrality is emphatically not interest-relative (Gorman 2014: 121). In laying out his account of essence, Gorman relies on the purportedly explanatory notion of support. As far as I can tell, this isn’t different from what most authors call “grounding”, and Gorman indeed says as much (2014: 134 n1). Another important concept in Gorman’s account is that of a foundational feature, which he defines as follows:

310

Essence, Grounding, and Explanation

“F is a foundational feature of a iff a is F and there is no G such that a’s being G supports a’s being F F is a non-foundational feature of a iff a is F and F is not a foundational feature of a”. (2014: 129) Finally, essence is defined in terms of foundational features, as follows: “To be an essential feature of a is to be a foundational feature of a. To be an accidental feature of a is to be a non-foundational feature of a”. (2014: 130) When we put together these definitions and interpret “support” as grounding, we get that F is an essential feature of a iff there’s no G such that a’s being G grounds a’s being F. Importantly, this definition doesn’t demand that a’s being G be fundamental in the grouns‐theoretic sense. It may well be that there is a further fact that grounds and explains why a has some essential property, G; it’s just that this fact cannot consist in the same property-bearer, a, having some other property (2014: 129). Another important qualification is that by “support” Gorman means only partial support, i.e. partial grounding. That is, for the feature of an entity to be accidental (non-essential), it suffices if it has a partial ground in terms of the same entity’s having some other feature.5 Unlike Sullivan, Gorman works with a non-modal notion of essence. However, as he himself notes, his account doesn’t even require the necessity of essential properties as a necessary condition. After all, nothing in the notion of a foundational property demands that things have their foundational properties necessarily. This means that (in a traditional Aristotelian vein) Gorman maintains that the distinction between essential and accidental properties is orthogonal to that between necessary and contingent ones: at least so far as the definitions go, not only could there be necessary accidental properties (as the Fine-inspired literature has it), but also contingent essential ones. As an example, Gorman offers a certain conception of Jesus Christ that has been popular in the scholastic tradition, according to which he has two natures, human and divine, but only one of them necessarily (2014: 131). More prosaically, if an object contingently has a certain microstructure, that may also be a contingent essential property on Gorman’s view. What are the prospects of an explanation- or grounding-based analysis of a non-modal, context-insensitive notion of essence that nonetheless accepts that essentiality is sufficient (even if not necessary) for necessity? Julio De Rizzo (2022) has recently offered an account of essence that attempts to do just that. His analysis has two stages. First, he adopts Correia’s ground-theoretic analysis of (rigid) ontological dependence: ∀x∀y (x depends ontologically on y ←→ ∃F□(x exists → (x exists because y is F))), where “because” is to be understood in the ground-theoretic sense (Correia 2005: 66). In the second stage, he offers a multi-clause definition that states the essence of a proposition according to its logical form. Going through each clause would be too involved for a brief survey like this, but the basic idea is that a truth about a certain entity E is essential to that entity just in case it is a necessary truth that is either atomic or the negation an atomic truth, and it contains reference solely to that E or entities on which E ontologically depends. De Rizzo gives the following simple example: it’s essential to {Socrates} that Socrates is a member of {Socrates}. {Socrates} ontologically depends of Socrates, which is as it should be given De Rizzo’s rule for atomic essentialist statements (2022: 36–37). It is important for De Rizzo’s account that he defines essence in terms of simple rigid dependence rather than essential dependence. This is necessary in order for the analysis to be

311

David Mark Kovacs

reductive (we cannot understand essential dependence without a prior grasp on essence). However, those who find the distinction between these two notions of dependence intuitive may worry that De Rizzo’s analysis conflates them, and that this is a disadvantage of analyses that take essence to be primitive and attempt to define grounding in terms of it. Either way, De Rizzo’s paper is an important recent addition to ground-theoretic accounts of essence, and of the three accounts considered in this section, it is the one that remains most faithful to the Finean tradition.

20.2.3

Shared-Basis Definitions of Essence and Grounding

A third possible way to increase ideological parsimony is to give reductive definitions of both essence and grounding in terms of the same primitive concepts. This is exactly what Fabrice Correia and Alexander Skiles attempt to do in a recent paper. A basic concept that plays a central role in their account is what they call “generalized identity”. The notion is perspicuously expressed by the locution “For … is for …” (Correia and Skiles 2019: 643). Sometimes this phrase functions as an essentialist locution itself, and this is how I read Correia’s earlier work in section 1.1. However, in this more recent paper Correia and Skiles understand the phrase as expressing identity: for example, for something to be a water molecule is (the same thing as) for it to be an H2O molecule; for something to be a bachelor is (the same thing as) for it to be an unmarried man; and for me to have knowledge is (the same thing as) for me to have true, justified, non-Gettiered belief. Importantly, these identity statements are generalized. That is, Correia and Skiles ask us to understand them as being clear enough to speak for themselves, without the need to paraphrase them into objectual identities that would commit us to entities in the world that serve as the relata of the identity relation (as in “The property of being water is identical to the property of being H2O”, a sentence about properties, or “Every bachelor is identical to an unmarried man, and every unmarried man is identical to a bachelor”, a sentence about people). Correia and Skiles’ analyses rely on the idea that generalized identity can be partial. They distinguish between two types of parthood. Conjunctive parthood (the notion we would normally identify with parthood simpliciter) corresponds to merely necessary conditions in a generalized identity statement. For example, part of what it is for someone to be a bachelor is for him to be a man, and part of what it is to be Socrates is to be human. Disjunctive parthood is less familiar. Intuitively, we could say that a disjunctive partial identity statement specifies a merely sufficient condition of the right-hand side of that statement. For example, for it to be rainy is a disjunctive part of what it is to be rainy or snowy. It’s not hard to see why conjunctive parthood will be relevant to essence and disjunctive parthood to grounding. Intuitively, each part of the full essence of something is itself essential. On the other hand, not each ground of a fact is a necessary condition of that fact; a fact can be “grounding-overdetermined” so that any set within a set of sets of full grounds is sufficient for its obtaining. The details of Correia and Skiles’ full proposal resist economic presentation because they distinguish several different notions of essence (full and partial; generic, objectual and factual). Here, I just mention the most basic pair of definitions they provide, for generic essence: (Full Generic Essence) Being F is what it is to be G in full iff: for a thing to be G is for it to be F (648) (Partial Generic EssenceC&S) Being F is partially what it is to be G iff: there is some H such that for a thing to be G is for it to be both F and H (649)

312

Essence, Grounding, and Explanation

A short clarification is in order here. As is clear from the definition, Correia and Skiles identify full generic essence with generalized identity itself. However, they don’t propose (Full Generic Essence) as an analytic truth that holds simply in virtue of stipulation. Rather, the idea is that we have some antecedent understanding of both essence and generalized identity, and the definition captures the core cases of the former well enough to be plausible. (This is why I classified Correia and Skiles’ approach as one that reduces both grounding and essence to the same primitive notion, rather than as a reduction of grounding to essence taken as primitive.) What about grounding? Here, we need to use the notions of both conjunctive and disjunctive parthood. When grounding is understood as a relation between facts or propositions, the definition goes as follows: (GroundingC&S) p1, p2, … ground q iff: (i) p1, p2, … are disjunctive parts of q and (ii) q is not a conjunctive part of a disjunctive part of any of p1, p2, … Clause (ii) is less straightforward to understand than clause (i); we can better grasp it by considering an example Correia and Skiles provide. Suppose the facts [a is red] and [a is round] jointly ground [a is red and round]. Now, for the disjunctive fact [a is red and round or a is red and round] has [a is red and round] as a disjunctive part; and so it has [a is red] and [a is round] as disjunctive parts too. However, the conjunctive fact [a is red and round] is obviously not a conjunctive part of any (disjunctive part of) either [a is red] or [a is round]. So, according to the definition, [a is red] and [a is round] indeed ground their conjunction, as they should. Correia and Skiles give a number of further variations on this definition. As they note, the notion of grounding so defined is non-factive (a non-factive grounding sentence can be true even if the sentence on the left side of the “because”, i.e. the one expressing the grounded fact, is false); but, they argue, the view could be easily extended to the more familiar notion of factive grounding. They also develop a parallel set of definitions for those who find the notions of conjunctive and disjunctive parthood insufficiently clear in terms of “subsumption”, where p1, p2, … are subsumed by q just in case p1, p2 … are ways for q to hold. In the present chapter, I will forgo discussing these complications. Clearly, the concept of generalized identity has a wide range of applications, and its use in reductive accounts of grounding and essence is yet to receive the critical attention that it deserves.

20.3

Supplementation

Some authors think that there is an intimate relation between grounding and essence in that they both play roles, albeit distinct ones, in the same kind of explanation. This is Boris Kment’s view. Kment (2014) develops his account of metaphysical explanation in the context of an ambitious general theory of modality. According to this view our modal concepts are rooted in counterfactual reasoning, and they are derivative from explanatory notions, which can be used to analyze them. Accordingly, metaphysical necessity can be analyzed in terms of grounding. The relation between grounding and essence, which (in line with the neoAristotelian / Finean tradition) Kment distinguishes from necessity, is less straightforward. Kment maintains that metaphysical explanations have both a grounding component and an essence component (note that this also implies that on his view, grounding is distinct from metaphysical explanation). He conceives of the role of essence in metaphysical explanations as analogous to the role of the laws of nature in scientific explanations; indeed, he identifies metaphysical laws with essentialist truths. Take, for example, the following case:

313

David Mark Kovacs

1 It’s essential to being a gold atom that all atoms with atomic number 79 are gold atoms 2 a is an atom with atomic number 79. 3 a is a gold atom. (Kment 2014: 163) According to Kment, (1)–(2) are both among the metaphysical explanantia of (3). However, only (2) is a ground of it; (1) is a metaphysical law that states an essential truth about the object of the metaphysical explanation. This example points to an important difference Kment sees between the laws of nature and the laws of metaphysics: while the former are general in nature, the latter concern particular entities. Kevin Richardson also assigns a distinctive role to essence, albeit only in a restricted set of metaphysical explanations. His main focus is on generic grounding claims, which he argues we should understand as generics rather than as universally quantified sentences. For example, “An act is right in virtue of maximizing happiness” shouldn’t simply be understood as universally quantifying over all acts. This is because universal generalizations are usually taken to be (at least partially) grounded in their instances, yet one could coherently hold that the generic is true but has the status of a primitive moral law (Richardson 2020: 79). Instead of understanding generic grounding in terms of universal quantification, Richardson offers the following definition in terms of (particular) grounding and essence (I slightly reformulated it to fit the rest of this chapter): (General Grounding) [x is G] is generically grounded in [x is F] iff it’s in the nature of being G that if x is G or F, then the fact that x is G is fully grounded in the fact that x is F) (2020: 82) Richardson notes as an advantage of his account that it doesn’t posit primitive metaphysical laws (2020: 84). But someone who seeks to keep this notion could plausibly identify them with essentialist claims, as we have seen does Kment. Richardson argues that his proposal has a number of further advantages: it explains the semantics of generic grounding; it gives us the conceptual tool to distinguish between reductive and non-reductive theories, where the former express generics while the latter express mere regularities about grounding; moreover, the view allows us to stay neutral about several substantive disputes about grounding.

20.4

Independence: Grounding and Essence as Two Distinct Modes of Explanation

Perhaps the attempt to find deep, systematic connections between grounding and essence is a futile exercise. Perhaps these are two fundamentally different explanatory notions that show important similarities and may even show a certain level of unity, but which should nonetheless be understood in their own right. Several authors take this view. They all agree that both grounding and essence are sui generis explanatory notions, but they differ in the extent to which they think they lend themselves to parallel treatment. Fine (2012), as mentioned in section 1.1, raised a number of objections to the view that grounding could be understood in terms of essence. But in his later work, Fine (2015) has come to think of essence and grounding as two fundamentally different but nonetheless complementary explanatory concepts, corresponding to constitutively necessary and constitutively sufficient conditions, respectively. That is, the essence of A consists of its constitutively necessary conditions, while A’s grounds are its constitutively sufficient conditions. Fine expresses some optimism that essence and grounding so understood would lend themselves to parallel formal treatments, possibly also shedding some light on the logic of grounding.

314

Essence, Grounding, and Explanation

Yet Fine’s approach has been subject to criticism. Correia and Skiles (2019) ask in what sense we can say that the necessary conditions that essence involves and the sufficient ones that grounding does are “constitutive”. Fine cannot simply have in mind necessary and sufficient conditions in the modal sense, since this interpretation leads to absurd results (for example that something’s being a number is what it is for it to be a number and being blue-or-not-blue). Alternatively, constitutively necessary and sufficient conditions could be understood non-modally, as sui generis notions. Indeed, this is what Fine has in mind, as he claims that these concepts cannot be defined in terms of each other (2015: 307). But in that case, it’s unclear that Fine’s treatment of essence and grounding is unified in anything but name. This is one of the motivations Correia and Skiles offer to prefer their account to Fine’s (2017: 664). Fine considers essence and grounding separate explanatory notions that are nonetheless strongly unified (as we have seen above, whether his account lives up to this ambition remains controversial). Some authors go further and maintain that grounding and essence provide two entirely different types of explanation, with no attempt to place them in a unified framework. Thus, Martin Glazier argues that there are two fundamentally different kinds of metaphysical explanation: grounding explanations and essentialist explanations. In Glazier’s use of the phrase, an explanation counts as essentialist only if it’s of the form “A because t is essentially such that A” (2017: 2873). So, not every explanation that appeals to an essentialist fact is an essentialist explanation in Glazier’s sense. Explanations of this sort, he argues, are ultimate in the sense that the explanans in an essentialist explanation cannot be the explanandum of a further essentialist explanation (although it might still be explainable in some other way). For example, the singleton set of Socrates contains Socrates because it’s in the nature of the singleton set to contain Socrates. Then by the ultimacy principle, there is no essentialist explanation of why it’s in the nature of the singleton set to contain Socrates. Glazier offers two arguments for why essentialist explanations cannot be thought of as a kind of grounding explanation. First, a fact could be ungrounded but essential; what’s in the nature of a thing might lack grounds altogether. For example, perhaps it’s a fundamental (in the sense “ungrounded”) fact about electrons that they have negative charge. This fact could still have an essentialist explanation if it’s in the nature of electrons to have negative charge. Second, the hypothesis that essentialist explanation is just a form of grounding explanation sits ill with some popular principles of the impure logic of grounding, for example, that disjunctions are grounded in their true disjuncts. For example, it may be that a Boolean variable in a computer program essentially has either the value 0 or the value 1. But it’s false that the variable essentially has the value 0, and it’s also false that the variable essentially has the value 1. Nonetheless, it could be that it has either the value 0 or the value 1 because it’s essentially such that it has one of those two values. The problem is that “a fact can ground a disjunction only if it does the grounding work of a true disjunct” (Glazier 2017: 2877), and this condition isn’t satisfied. For while the contingent fact that the variable has the value 1 grounds the disjunction that the variable has the value 1 or grass is green, the essential (and thus necessary) fact that it’s in the nature of the variable to have the value 0 or the value 1 cannot ground it. According to the widely accepted thesis of Grounding Necessitarianism, grounds necessitate what they ground. If so, then necessary facts cannot ground contingent ones, as they don’t entail them. Richardson (2021) takes a similar approach, although his taxonomy is different from Glazier’s. Instead of distinguishing grounding-based from essentialist metaphysical explanations, he differentiates between two types of grounding: “why-grounding”, which is what most people (including Glazier) usually mean by “grounding” without any qualification, and

315

David Mark Kovacs

“what-grounding”, which seems similar to essentialist explanation. (Skiles [2019] also draws a similar distinction between two different types of metaphysical explanation.) Despite the (rough) pairwise correspondence between why-grounding and grounding on the one hand, and what-grounding and essentialist explanation on the other, some of the differences between Glazier’s and Richardson’s views seem substantive. For while Glazier presupposes Grounding Necessitarianism (see two paragraphs above), Richardson thinks that the distinction can help adjudicate an ongoing debate over this thesis. Namely, he argues that Grounding Necessitarianism is true of what-grounding (precisely due to its connection to essence) but not of why-grounding. The latter relation, he argues, should be understood in terms of difference-making and bears no interesting connections to essence. As we have seen in section 1, reducing either grounding or essence to the other (or both of them to a third primitive) is a formidable task. By contrast, a non-reductive approach that doesn’t attempt to offer such definitions is presumably the default approach that has many possible variants other than those offered by Glazier and Richardson.

20.5 Conclusion Research on the relation between essence on the one hand and metaphysical explanation and laws on the other hand is still in its early stages. But from the budding literature it can already be gathered that there is a wide variety of options concerning the explanatory nature of essence: whether it’s explanatory at all; if so, whether its explanatory role is distinctive or similar to that of grounding; and moreover, whether essence and grounding or metaphysical explanation can be analyzed either in terms of each other or in terms of the same primitive notions. Besides the intrinsic interest that these questions hold, they may shed light on more general themes about metaphysical methodology. In the contemporary literature, grounding and (non-modal notions of) essence came to the forefront around the same time, between the mid-90s and the mid-2000s, in the context of the “hyperintensional revolution” (Nolan 2014). Both are often seen as useful conceptual tools that can be invoked to fill theoretical roles for which modal concepts are inadequate. They are also often considered “partners in crime”: for philosophers who harbor suspicion about one of these notions it’s natural to be skeptical about the other as well. However, a closer look reveals that even philosophers who see the hyperintensional revolution as progress have a variety of attitudes to what such progress consists in. Some, for example E.J. Lowe (2018), have been working toward a very robust rehabilitation of neo-Aristotelian metaphysics, conceiving of metaphysics as the “science of essence”. Others are considerably more reserved. For example, despite also being inspired by Aristotle, Schaffer sees questions of what grounds what central to metaphysics (2009) but seeks to give an account of grounding that doesn’t rely on the notion of essence, which he considers suspect (2016: 83). Koslicki, by contrast, sees essence as valuable and even central to metaphysics (2012) but considers grounding too coarse-grained to do useful theoretical work (2015). These different attitudes are often reflected in one’s view about the relation between grounding and essence. It can be said that almost all (although not only) advocates of Unity and Supplementation are full-throated advocates of the hyperintensional revolution. Offering an account of either essence of grounding that substantively relies on one’s account of the other notion is a risky endeavor unless one is highly confident that both notions are in good standing. Independence and Irrelevance are much more congenial to theorists like Koslicki and Schaffer, who embrace certain aspects of the hyperintensional revolution but have reservations about others. The author of this chapter counts himself among these more

316

Essence, Grounding, and Explanation

reserved theorists, having worked extensively on grounding and metaphysical explanation but remaining sympathetic to a purely modal, pre-Finean account of essence. Either way, much work is yet to be done on both on the general question of proper methodology and on what analytic connections, if any, there are between essence and explanation. Further developments in these debates will be crucial to our understanding of the role of essence in post-modal metaphysics.

Related Topics For connections to explanation and laws in the philosophy of science, see Chapter 13 by KaHo Lam (“Dispositional Essentialism”), Chapter 10 by Tuomas Tahko (“Natural Kind Essentialism”), and Chapter 12 by Travis Dumsday (“Scientific Essentialism”). See also Chapter 7 by Alessandro Torza (“Modal Conceptions of Essence”) and Chapter 8 by Fabrice Correia (“Non-Modal Conceptions of Essence”).

Notes 1 See also O’Conaill (2018) and Wallner (2018) for similar views. 2 Cf. also Glazier (2017) and Raven (2020) as well as section 3 of this chapter on the ultimacy of essentialist explanations. 3 See Leuenberger (2014) and Skiles (2015) for criticism and O’Conaill (2018) for some pushback. 4 This is Sullivan’s own example. One might worry that it’s not plausible that the coin essentially conducts electricity even according to a physical framework; rather, it’s merely (non-essentially) constituted by a piece of metal that essentially conducts electricity according to a physical framework. However, I think this is an artifact of this particular example. We can replace conductivity with the property of, say, having an effigy on the obverse. Intuitively, any numismatic explanation involving the coin must appeal to this feature, but not to the coin’s being a unit of account (coins of numismatic interest are typically no longer units of account); whereas any economic explanation involving the coin must appeal to its being a unit of account, but not to there being an effigy on its obverse (the Central Bank may well have issued the same coins with the same face value but with State Arms rather than an effigy on their obverse). Thanks to Kathrin Koslicki and Mike Raven for raising this concern. 5 Another aspect of Gorman’ view is that its essential properties explain many of an entity’s other properties. As Gorman notes, this is not new to his account but is in line with the Aristotelian tradition. For further references, see Gorman (2014): 135 n27.

References Audi, Paul (2012), “Grounding: Toward a theory of the in-virtue-of relation,” Journal of Philosophy, 109: 685–711 Carnino, Pablo (2014), “On the reduction of grounding to essence,” Studia Philosophica Estonica, 7:56–71 Correia, Fabrice (2005), Existential Dependence and Cognate Notions, Philosophia Verlag. Correia, Fabrice (2006), “Generic essence, objectual essence, and modality,” Nous, 40: 753–767 Correia, Fabrice (2013), “Metaphysical grounds and essence,” M. Hoeltje, B. Schnieder and A. Steinberg (eds.), Varieties of Dependence: Ontological Dependence, Grounding, Supervenience, Response-Dependence, Munich: Philosophia, 271–296. Correia, Fabrice and Alex Skiles (2019), “Grounding, essence, and identity,” Philosophy and Phenomenological Research, 98: 642–670 Dasgupta, Shamik (2014), “The possibility of physicalism,” Journal of Philosophy, 111: 557–592 De Rizzo, Julio (2022), “A ground-theoretical modal definition of essence,” Analysis, 82: 32–41 Duncan, M., K. Miller and J. Norton (2021), “Ditching determination and dependence: or, how to wear the crazy trousers,” Synthese, 198: 395–418 Fine, Kit (1994), “Essence and modality,” Philosophical Perspectives, 8: 1–16

317

David Mark Kovacs Fine, Kit (2012), “Guide to ground,” In Fabrice Correia and Benjamin Schnieder (eds.), Metaphysical Grounding: Understanding the Structure of Reality, 37–80, Cambridge: Cambridge University Press Fine, Kit (2015), “Unified foundations for essence and ground,” Journal of the American Philosophical Association, 1: 296–311 Glazier, Martin (2017), “Essentialist explanation,” Philosophical Studies, 174: 2871–2889 Gorman, Michael (2014), “Essentiality as foundationality,” in Novotný, Daniel B. and Lukáš Novák (eds.), Neo-Aristotelian Perspectives in Metaphysics, pp. 129–147, New York: Routledge, Kappes, Yannic (2021), “Explanation by status as empty-base explanation,” Synthese, 199: 2575–2595 Kment, Boris (2014), Modality and Explanatory Reasoning, Oxford: Oxford University Press Koslicki, Kathrin (2012), “Essence, necessity, and explanation,” in Tuomas E. Tahko (ed.), Contemporary Aristotelian Metaphysics, Cambridge: Cambridge University Press, pp. 187–206 Koslicki, Kathrin (2015), “The Coarse-grainedness of grounding,” Oxford Studies in Metaphysics, 9: 306–344 Leuenberger, Stephan (2014), “Grounding and necessity,” Inquiry, 57: 151–174 Lowe, E. J. (2018), “Metaphysics as the science of essence,” In Alexander Carruth, Sophie Gibb & John Heil (eds.), Ontology, Modality, and Mind: Themes from the Metaphysics of E. J. Lowe, Oxford: Oxford University Press, pp. 14–34 McDaniel, Kris and Steve Steward (2016), “Are there essential properties? Yes”, in Elizabeth Barnes (ed.), Current Controversies in Metaphysics, London: Routledge, 62–78 Nolan, Daniel (2014), “Hyperintensional metaphysics,” Philosophical Studies 171 (1):149–160 O′Conaill, Donnchadh (2018), “Grounding, physicalism, and necessity,” Inquiry, 61:713–730 Raven, Michael (2020), “Explaining essences,” Philosophical Studies, 178: 1043–1064 Richardson, Kevin (2020), “On what (In General) grounds what,” Metaphysics, 2: 73–87 Richardson, Kevin (2021), “Grounding is necessary and contingent,” Inquiry, 64: 453–480 Rosen, Gideon (2010), “Metaphysical dependence: Grounding and reduction,” in Bob Hale and Aviv Hoffman (eds.), Modality: Metaphysics, Logic, and Epistemology, 109–136, Oxford: Oxford University Press Rosen, Gideon (2015), “Real definition,” Analytic Philosophy, 56: 189–209 Schaffer, Jonathan (2009), “On what grounds what,” D. Manley, D. Chalmers & R. Wasserman (eds.), Metametaphysics: New Essays on the Foundations of Ontology, Oxford University Press. pp. 347–383 Schaffer, Jonathan (2016), “Grounding in the image of causation,” Philosophical Studies, 173: 49–100 Schaffer, Jonathan (2017), “The ground between the gaps,” Philosophers’ Imprint, 17 (11): 1–26 Segal, Aaron and Noga Gratvol (2021), “Essence and explanation: A logical mismatch,” Inquiry, 64: 1038–1050 Skiles, Alexander (2015), “Against grounding necessitarianism,” Erkenntnis, 80: 717–751 Skiles, Alexander (2019) “There is no Haeccitistic Euthryphro problem,” Analysis, 79: 477–484 Sullivan, Meghan (2016), “Are there essential properties? No”, in Elizabeth Barnes (ed.), Current Controversies in Metaphysics, New York: Routledge, 45–61 Trogdon, Kelly (2013), “Grounding: Necessary or contingent,” Pacific Philosophical Quarterly, 94: 465–485 Wallner, Michael (2018), “The ground of ground, essence, and explanation,” Synthese 198 (Suppl 6): 1257–1277 Zylstra, Justin (2019), “The essence of grounding,” Synthese, 196: 5137–5152

318

21 THE “REDUCTION” OF NECESSITY TO NON-MODAL ESSENCE Kathrin Koslicki

21.1

Introduction

Non-modalists about essence reject the idea that metaphysical modality is prior to essence, e.g., in the sense that the latter can be reduced to or defined in terms of the former. On the contrary, according to these theorists, the explanation, if anything, proceeds in the opposite direction: metaphysical modality does not explain, but is instead explained in terms of, essence.1 Thus, for non-modalists like Aristotle, Kit Fine and E. J. Lowe, one of the primary theoretical roles of essence is to provide an explanatory basis for metaphysical necessity and possibility. When Lowe speaks of essences, in this connection, he has in mind a doctrine he calls “serious essentialism”, according to which essences are not further entities, distinct from the entities whose essences they are; essence precedes existence; and essences are the ground of all metaphysical possibilities and necessities, not only of existing things but also of nonexisting things (Lowe 2008: 45–6). Kit Fine famously argues against the contemporary assimilation of essence to modality and instead adopts an Aristotelian, definitional conception of essence. According to Fine, a statement of the essence is a non-modal truth which specifies what it is to be an entity; the entity’s modal profile, in turn, is thought to “flow from” its essence (Fine 2005b: 348).2 This chapter explores a series of as of yet unanswered questions which appear to be central to the non-modalist’s program of explaining metaphysical modality in terms of essence. We begin, in Sections 21.2.1–2, by considering cases in which the proposition embedded within the metaphysical necessity operator (“It is metaphysically necessary that … ”) is either the same as, or logically entailed by, the proposition embedded within the essence operator (“It is part of the essence of __ that … ”). Next, in Sections 21.2.3–5, we take up cases from several domains (viz., mathematics, metaphysics, and natural science), in which the proposition embedded within the necessity operator is neither the same as nor logically entailed by the proposition embedded within the essence operator. In each case, it turns out that, in order to derive a metaphysically necessary truth from a statement of essence, additional facts concerning the essences of related entities must be brought in that go beyond those explicitly appealed to in the initial essentialist premise-set. Our discussion in Sections 21.3.1–2 of extant proposals indicates that no logical “one-size-fits-all” strategy of widening the relevant class of essentialist facts appropriately appears to be feasible. Rather, non-modalists must proceed by

DOI: 10.4324/9781003008750-25

319

Kathrin Koslicki

means of a non-logical case-by-case engagement with specific cases, if they hope to fill this important lacuna in the non-modalist’s proposed explanation of metaphysical necessity in terms of essence.

21.2

Explanatory Connections Between Essence and Metaphysical Modality 21.2.1 The Simple Case 3

Let’s begin our survey of the different types of explanatory connections which may obtain between essence and metaphysical modality with what I will call “the simple case”, as illustrated in (1) and (2)4: (1) □Socrates(Socrates is human) (2)□(Socrates is human) In the simple case, the proposition that is embedded within the essentialist operator is the same as the proposition that is embedded within the modal operator.5 Thus, according to (1), it is part of Socrates’ essence to be human, while (2) states that Socrates is necessarily human. (1) concerns the essence of just a single object, viz., Socrates. But we can also generalize the simple case to allow for instances in which the essentialist truth in question is a collective claim concerning the essences of a plurality of entities, taken together, as illustrated in (3) and (4): (3)□Socrates, the Eiffel Tower[(Socrates is human) & (the Eiffel Tower is a tower)] (4)□((Socrates is human) & (the Eiffel Tower is a tower)) (3) states that it is part of the essence of Socrates and the Eiffel Tower, taken together, that Socrates is human and the Eiffel Tower is a tower. According to (4), the same conjunctive proposition, viz., that Socrates is human and the Eiffel Tower is a tower, holds necessarily. In this case, the collectivity in (3) can be factored out into two separate essentialist claims concerning Socrates and the Eiffel Tower, individually, viz., that it is part of the essence of Socrates to be human and that it is part of the essence of the Eiffel Tower to be a tower. However, we may allow for the possibility that some statements concerning the essences of a plurality of entities taken together are ineliminably collective. Moreover, even in cases like (3) and (4), where the collectivity can be factored out, only certain ways of pairing metaphysically necessary truths with essentialist claims are available. For example, the metaphysically necessary truth that Socrates is human must be paired with the essentialist claim that it is part of Socrates’ essence to be human, rather than with the essentialist claim that it is part of the essence of the Eiffel Tower to be a tower; and similarly for the second conjunct of (3) and (4). With respect to the simple case, non-modalists commonly (though not universally) accept that a one-way entailment holds between the essential truth at issue and the corresponding metaphysically necessary truth, i.e., in this case, between (1) and (2).6 Thus, (Fine 1994), for example, remarks as follows: My objection to the modal accounts will be to the sufficiency of the proposed criterion, not to its necessity. I accept that if an object essentially has a certain property then it is necessary that it has the property (or has the property if it exists); but I reject the converse. (Fine 1994: 4)7 We can thus, for now, proceed on the assumption that, in the simple case, the inference from the essentialist claim to the corresponding metaphysically necessary truth is thought to be licensed by way of a general one-way entailment principle of the form stated in (EN) (where “EN” abbreviates “Essence-Necessity”):

320

The “Reduction” of Necessity to Non-Modal Essence

(EN) One-Way Entailment from Statements of Essence to Metaphysically Necessary Truths:8 □x,y,

…A

⊢ □A

If (EN) itself is thought to hold necessarily, then non-modalists are presumably committed to the idea that the source of the necessary truth of (EN) can in turn be located in the essences of some entities. We will return to this issue below.

21.2.2

Logical Connections

But what happens in a case in which the proposition that is embedded within the essence operator is not the same as the proposition that is embedded within the necessity operator and yet the necessary truth in question is nevertheless thought to follow in some way from the corresponding essentialist claim?9 Scenarios of this kind can arise in a variety of different ways. One such possibility is what I will call “the logical case”, illustrated here by means of (6) and (7): (6) □Socrates’ singleton set(Socrates is a member of Socrates’ singleton set) (7) □(Socrates’ singleton set contains some member or other) According to (6), it is part of the essence of Socrates’ singleton set that Socrates is a member of Socrates’ singleton set. (7) states that Socrates’ singleton set necessarily contains some member or other. In this case, while the proposition that is embedded within the essence operator in (6) (viz., that Socrates is a member of Socrates’ singleton set) is not the same as the proposition that is embedded within the necessity operator in (7) (viz., that Socrates’ singleton set contains some member or other), the second proposition is logically entailed by the first proposition.10 One way in which non-modalists might understand the explanatory connection between essence and metaphysical necessity in the logical case is by reference to a distinction Fine draws between “constitutive essence” and “consequential essence”11: A property belongs to the constitutive essence of an object if it is not had in virtue of being a logical consequence of some more basic essential properties; and a property might be said to belong to the consequential essence of an object if it is a logical consequence of properties that belong to the constitutive essence (a similar account could be given for the case in which the essence is conceived in terms of propositions rather than properties). Thus the property of containing Socrates as a member will presumably be part of the constitutive essence of singleton Socrates, whereas the property of containing some member or another will presumably only be part of its consequential essence. (Fine 1995c: 276) Assuming that it is part of the constitutive essence of Socrates’ singleton set that it has Socrates as a member, as stated in (6’), it will thus be part of the consequential essence of Socrates’ singleton set that it has some member or other, as stated in (8) (where “□CONST:x,y, …” abbreviates “it is part of the constitutive essence of x, y, … ” and “□CONSQ:x,y, …” abbreviates “it is part of the consequential essence of x, y, … ”): (6’) □CONST:Socrates’ singleton set(Socrates is a member of Socrates’ singleton set) (8) □CONSQ:Socrates’ singleton set(Socrates’ singleton set contains some member or other) At this point, we have arrived at a situation in which the proposition that is embedded within the consequential essence operator in (8) is the same as the proposition that is

321

Kathrin Koslicki

embedded within the necessity operator in (7). As long as the one-way entailment between statements of essence and metaphysically necessary truths in (EN) can be assumed to hold for statements concerning both constitutive and consequential essence, the logical case can thus be assimilated to the simple case. The assimilation of the logical case to the simple case, however, requires additional assumptions: for the derivation of the consequential essence of an entity from its constitutive essence itself turns on the fact that the proposition embedded within the consequential essence operator is logically entailed by the proposition embedded within the constitutive essence operator. In response to the question of how this connection itself might be justified, Fine offers the following comments: … [W]e should think of the nature of the logical concepts (or the meanings of the logical terms) as being given, not by certain logical truths, but by certain logical inferences. Thus what properly belongs to the nature of disjunction is the inference from p to (p or q) rather than the fact that p implies (p or q). (Thus this is a case in which one might want to think of the nature of something as being nonpropositional in character). That Socrates is a man or a mountain will then follow from certain propositions by means of certain rules. The concept of consequence is not presupposed but is already built into the rules. (Fine 1995b: 58) Applying this idea to the particular case at hand, then, the derivation of (8) from (6’) can, in part, be traced to the fact that it lies in the nature of the existential quantifier to license the inference from the proposition that Socrates’ singleton set contains Socrates as its sole member to the proposition that Socrates’ singleton set contains some member or other. In this way, the logical connection which underlies the derivation of the consequential essence of an entity from its constitutive essence is itself explained by reference to the essences of certain entities, viz., in this case, the logical operations that are relevant to the inference in question.12

21.2.3 Mathematical Connections Next, consider the following mathematical example discussed by E. J. Lowe (Lowe 2012: 105–7): (9) An ellipse is the locus of a point moving continuously in a plane in such a fashion that the sum of the distances between it and two other fixed points (viz., the ellipse’s foci) remains constant. (10) An ellipse is the closed curve of intersection between a cone and a plane cutting it at an oblique angle to its axis greater than that of the cone’s side. According to Lowe, (9) gives a real definition or statement of the essence of an ellipse: it specifies what it is to be an ellipse by capturing its “generating principle”, i.e., by revealing what it takes for an ellipse to come into being (ibid., 105). In contrast, (10) provides a necessary but non-essential characterization of an ellipse as a type of conic section. We can thus rephrase (9) as (9’) and (10) as (10’): (9’) □CONST:Ellipse(An ellipse is the locus of a point moving continuously in a plane in such a fashion that the sum of the distances between it and two other fixed points remains constant). (10’) □(An ellipse is the closed curve of intersection between a cone and a plane cutting it at an oblique angle to its axis greater than that of the cone’s side).

322

The “Reduction” of Necessity to Non-Modal Essence

(9’) makes explicit that it is part of the constitutive essence of an ellipse to meet the condition specified in (9), while (10’) states that the condition described in (10) applies to ellipses necessarily. In this case, the proposition embedded within the constitutive essence operator in (9’) is not the same as the proposition embedded within the necessity operator in (10’); nor is the second logically entailed by the first. Thus, the mathematical connection between (9’) and (10’) cannot be assimilated to the simple case, even with an additional appeal, as in the logical case, to the natures of the relevant logical operations. Nevertheless, the overall success of the nonmodalist program of explaining metaphysical necessity in terms of essence would seem to require that the necessary truth concerning ellipses specified in (10’) can in some way be derived from the specification of what it is to be an ellipse stated in (9’), perhaps in conjunction with additional truths concerning the essences of other relevant entities. In response to the question of how the necessary truth concerning ellipses stated in (10’) might be traced to its proper essentialist source, Lowe argues, in effect, that the proposition embedded within the necessity operator in (10’) does not itself express an essential truth concerning ellipses, since it “characterises an ellipse in terms that are extrinsic to its nature as the particular kind of geometrical figure that it is” (Lowe 2012: 105). In order to trace the necessary truth in (10’) to its proper essentialist source, so Lowe argues, facts concerning the essences of other relevant entities, viz., in this case, cones, must also be taken into consideration. But even when these are added into the mix, the proposition embedded within the necessity operator in (10’) itself still cannot be understood as expressing an essential truth: for it is neither part of the essence of ellipses or cones alone that they satisfy the condition specified in (10’); nor is it part of the essences of ellipses and cones, taken together, that they satisfy the condition specified in (10’). The most we can say about this case, in Lowe’s view, is that the necessary truth in (10’) “holds in virtue of” (or “because of” or “is grounded in”) the relevant essential truths concerning ellipses and cones. The mathematical case therefore exhibits an explanatory connection of some sort between statements of essence and necessary truths, but not one that can be assimilated either to the simple case or to the logical case. The details of how non-modalists might approach the explanatory connection encountered in the mathematical case remain yet to be filled in.

21.2.4

Metaphysical Connections

Explanatory connections between necessary truths and statements of essence can also occur in metaphysical contexts, as can be illustrated, for example, by the following instance of what is known as “the Grounding Problem”13: (11) a. b. (12) a. b.

□CONST:Lumpl(Lumpl is a lump of clay) □CONST:Goliath(Goliath is a statue) □(Lumpl can survive being squashed) □(Goliath cannot survive being squashed)

According to (11.a) and (11.b), it is part of the constitutive essence of Lumpl to be a lump of clay and part of the constitutive essence of Goliath to be a statue. (12.a) and (12.b) bring out a particular difference between Lumpl’s and Goliath’s persistence conditions, which can be assumed to hold of these objects necessarily, namely that the former can, while the latter cannot, survive being squashed. The Grounding Problem more generally asks how objects that are apparently numerically distinct but spatiotemporally coincident, such as a statue and the lump of clay that constitutes it, can differ from one another with respect to their modal, temporal, or other characteristics, despite the fact that they are otherwise so alike.

323

Kathrin Koslicki

How (if at all) can the necessary truths concerning Lumpl and Goliath in (12.a) and (12.b), respectively, be derived from the partial statements of their constitutive essence in (11.a) and (11.b)? Presumably, any such derivation would have to take recourse to additional premises which spell out further why it is incompatible with Goliath’s nature, as a statue, but compatible with Lumpl’s nature, as a lump of clay, to take on the sort of shape an object acquires as a result of being squashed. Suppose, for example, as stated in (13), that it is part of the constitutive essence of a lump of clay to be composed of clay parts arranged in the shape of a lump: (13) □CONST:Lumps of clay(A lump of clay is composed of clay parts arranged in the shape of a lump) Moreover, for the sake of simplicity, let’s assume that nothing further is required of the material parts of a lump-shaped object than that these material parts are in close spatial vicinity to one another. (14) describes this condition as being part of the constitutive essence of the shape-property, being lump-shaped: (14) □CONST:Being lump-shaped(The material parts of a lump-shaped object are in close spatial vicinity to one another) Now, given (13) and (14), it is easy to see why Lumpl, as a lump of clay, can survive being squashed: for even after an object has been squashed, its material parts may still be in close spatial vicinity to one another; and this (so we are assuming) is the only kind of arrangement that Lumpl’s material parts are required to exhibit as part of Lumpl’s nature. The compatibility in question is therefore explained in part by reference to the fact that the general determinable shape-property, being lump-shaped, has among its manifestations some of the determinate shape-properties objects can acquire as a result of having been squashed. Without entering too far into the complex question of what might be constitutively essential to Goliath, as a statue, we can safely assume that, on any plausible essentialist conception of artworks, Goliath’s nature will require more of its material parts than merely that they are in close spatial vicinity to one another. It is therefore natural to expect that, if Goliath were to be squashed, its material parts would no longer meet these more stringent requirements, whatever they are, and Goliath would go out of existence as a result of being squashed. To see how the incompatibility in question can arise, let’s assume, for the sake of specificity, that it is constitutively essential to Goliath (the statue) that its material parts are arranged in such a way as to represent Goliath (the Philistine biblical giant allegedly slain by David), as stated in (15): (15) □CONST:Goliath-the-statue(Goliath-the-statue is composed of material parts arranged in such a way as to represent Goliath-the-biblical-giant) And let’s assume further, for the sake of simplicity and specificity, that in order for an object to have the power to represent Goliath-the-biblical-giant, it must at least resemble Goliath-thebiblical-giant in certain relevant respects (leaving open what these relevant respects might be). The presumed connection between representation and resemblance is stated in (16) as a general claim concerning the constitutive essence of the representational property, representing Φ: (16) □CONST:Representing Φ(An object that represents Φ resembles Φ in relevant respects) And while these assumptions are of course only intended as a “toy theory” concerning the nature of artworks and representation, they do serve to bring out how, with the help of additional essentialist premises, the incompatibility in question might arise between Goliath’s nature, as a statue, and its ability to take on the sort of shape an object acquires as a result of having been squashed: for an object that has been squashed presumably loses its ability to resemble in the relevant respects, and therefore to represent, Goliath-the-biblical-giant. Again,

324

The “Reduction” of Necessity to Non-Modal Essence

as in the case of Lumpl, the explanatory connection at issue between the necessary truth concerning Goliath-the-statue’s persistence conditions in (12.b) and the partial statement of its constitutive essence in (11.b) has been traced in part to facts concerning the essences of other relevant entities, viz., in this case, the representational, resemblance and shape-properties that are pertinent to the case at hand. In the present instance, the incompatibility in question is thought to arise from the fact that the determinate shape-property an object might acquire as a result of having been squashed is not among the determinate shape-properties an object might manifest if it is to resemble, and therefore to represent, a certain object, viz., in this case, Goliath-the-biblical giant. As noted, deriving the necessary truths concerning both Lumpl’s and Goliath-the-statue’s persistence conditions in (12.a) and (12.b) from the respective partial statements of their constitutive essence in (11.a) and (11.b) required, in both cases, an appeal to additional premises concerning the essences of other entities, viz., the various shape-, resemblance, and representational properties that are relevant to their respective persistence conditions; moreover, in both cases, the derivation in question went well beyond the patterns of inference observed in either the simple case or the logical case.

21.2.5

Scientific Connections

The final type of explanatory connection I want to consider is taken from the realm of natural science and can be illustrated by means of the following example14: (17) □CONST:Water(Water is predominantly composed of H2O-molecules) (18) □(Water boils at 100 degrees Celsius) According to (17), it is part of the constitutive essence of water to be predominantly composed of H2O-molecules, while (18) states that water necessarily boils at 100 degrees Celsius. Again, as in the mathematical and metaphysical cases, the proposition embedded within the necessity operator in (18) is neither the same as nor logically entailed by the proposition embedded within the constitutive essence operator in (17). How (if at all) can the necessary truth concerning water in (18) be derived from the partial statement of its constitutive essence in (17)? In the case of a scientific explanatory connection, such as that illustrated in (17) and (18), non-modalists have the option of applying a strategy that is different from those already encountered, namely to argue that the explanatory connection in question does not in fact instantiate a more general schema in which a metaphysically necessary truth follows from some essential truths. If some version of this strategy can be made to work, then the relation between (17) and (18), regardless of how exactly it is to be understood, does not straightforwardly fall under the commitment incurred by non-modalists to explain metaphysical necessity in terms of essence. Variants of this strategy are adopted by both E. J. Lowe and Kit Fine.15 According to Lowe, the proposition embedded within the necessity operator in (18) does not express a metaphysically necessary truth concerning water, but rather states a contingent natural law consisting in the dispositional predication of a non-substantial universal (viz., the property, boiling at 100 degrees Celsius) of a substantial kind (viz., the kind, water) (Lowe 2006: 127–8). The necessity operator in (18) might be taken to denote what is commonly referred to as “natural” or “physical” necessity; but this type of necessity, in Lowe’s view, is “at best a species of ‘relative’ necessity: a matter of what is necessarily the case given that some contingent truth obtains” (ibid., 133). In addition, given Lowe’s approach, whether the proposition embedded within the constitutive essence operator in (17) can in fact be accurately characterized as a partial statement of the constitutive essence of water depends

325

Kathrin Koslicki

on whether we understand the term, “water”, in (17) as being used in the way in which theoretical chemists might use it to denote a certain chemical compound or in the way in which ordinary speakers might use it to denote a certain kind of liquid identifiable by its macroscopically detectable features, such as transparency, colorlessness and tastelessness (Lowe 2008: 44–5). Given the first (scientific) usage of the term, “water”, it is in fact correct to say, according to Lowe, that it is part of the essence of the chemical compound in question to be predominantly composed of H2O-molecules. Given the second (non-scientific) usage of the term, “water”, however, it is at most correct to say that the macroscopically identifiable liquid which figures in our ordinary non-scientific discourse is, as a matter of natural law, predominantly composed of H2O-molecules. But this latter statement expresses merely that it is overwhelmingly unlikely, given certain contingent states of affairs, that the transparent, colorless, tasteless liquid which fills our rivers, lakes and oceans could have a chemical composition very different from the chemical composition it actually has (where the occurrence of “could” here expresses mere physical or natural, and not metaphysical, possibility). As a result, given that, in Lowe’s view, (18) does not express a metaphysically necessary truth concerning water and given that (17) only expresses a partial statement of the constitutive essence of water on certain construals but not others, the relation between (17) and (18) does not exhibit a pattern in which a metaphysically necessary truth follows from a statement of essence. (Fine 2002) argues that there are three main and irreducible varieties of necessity, viz., metaphysical, natural, and normative necessity. The proposition embedded within the necessity operator in (18), on Fine’s approach, would be classified as naturally but not metaphysically necessary and thus similarly escapes the non-modalist’s commitment to trace metaphysically necessary truths to their proper essentialist source. What explanatory connection, if any, might there be, on Fine’s view, between the constitutively essential truth concerning water in (17) and the naturally necessary truth in (18)? In Fine’s view, even though (18) cannot be derived from (17), even with an appeal to additional essentialist premises, what we can say concerning the explanatory connection in question is that any apparently de re naturally necessary fact concerning a particular quantity of water, to the effect that it boils at 100 degrees Celsius, reduces to (i) the metaphysically necessary (and constitutively essential) de re fact that the quantity in question belongs to the kind, water, and is therefore, as part of its nature, predominantly composed of H2O-molecules; and (ii) the naturally necessary de dicto fact that chemical compounds of this kind boil at 100 degrees Celsius (Fine 2002: 243). The latter naturally necessary de dicto fact is not further derivable from any metaphysically necessary or essential truth, though it might of course be derivable from other more fundamental de dicto natural necessities concerning the connection between chemical composition, molecular structure and temperature. But the kind of derivability that is at issue in the relation between more fundamental and less fundamental de dicto natural necessities is of course different from that which figures in the non-modalist program of deriving metaphysical necessity from essence.

21.3 Explanatory Strategies 21.3.1

The One-Way Entailment Between Statements of Essence and Metaphysical Necessary Truths

Let’s return now to a question that was left open earlier in our discussion of the simple case, i.e., the case in which the proposition embedded within the necessity operator is the same as

326

The “Reduction” of Necessity to Non-Modal Essence

the proposition embedded within the essence operator. As noted in Section 21.2.1, our two representative non-modalists, Kit Fine and E. J. Lowe, both accept that, in the simple case, a one-way entailment, stated in (EN), holds between statements of essence and metaphysically necessary truths: according to (EN), if it is part of the essence of a certain entity or entities that A (where “A” abbreviates a sentence), then it is also metaphysically necessary that A. Among the questions that arise with respect to (EN), the following seems especially pressing for nonmodalists: assuming that the one-way entailment from statements of essence to metaphysically necessary truths itself holds with metaphysical necessity, can this metaphysical necessity also be traced to the essences of some entities and, if so, how? Suppose that any particular application of the schema stated in (EN) results in a metaphysically necessary truth of the following form: (N) □(□x,y,

…A→□A)

According to (N), the following holds with metaphysical necessity: if it is essential to some entities, x, y, … , that A, then it is metaphysically necessary that A. The proposition embedded within the outer metaphysical necessity operator in an instance of (N), by hypothesis, expresses a metaphysically necessary truth and therefore, by the non-modalist’s lights, itself needs to be traced to the essences of some entities. But which entities could act as the proper essentialist source from which metaphysically necessary truths of the form stated in (N) can be derived? It is implausible to think, in general, that the entities, x, y, … , whose identity serves as the explanatory basis for the essential truth of the proposition that A, could also act as the proper essentialist source by means of which the metaphysically necessary truth of the embedded conditional statement (“If it is part of the essence of x, y, … , that A, then it is metaphysically necessary that A”) can be explained. For suppose, for the sake of illustration, that “A” abbreviates the sentence, “Socrates is human” and that the inference in question is that from (1) to (2). In that case, the essential truth of “Socrates is human” is assumed to have its source in Socrates’ constitutive essence; and, by (EN), this essential truth in turn entails the metaphysically necessary truth of “Socrates is human”. But presumably there is nothing in Socrates’ nature which speaks to the fact that the following holds with metaphysical necessity: if it is part of Socrates’ essence to be human, then it is metaphysically necessary for Socrates to be human. For, unlike a logical operation, say, Socrates, as a human being, is not the right sort of entity whose nature could license an inference, such as that stated in (EN), or the metaphysically necessary truth of a conditional statement of the form stated in (N) which expresses the inference in question, viz., “If it is part of Socrates’ essence to be human, then it is metaphysically necessary for Socrates to be human”. Given Socrates’ unsuitability to act as the proper essentialist source in this particular case, we therefore also cannot expect that, in general, the metaphysically necessary truth of a statement of the form, “If it is part of the essence of x, y, … , that A, then it is metaphysically necessary that A”, can be explained by reference to the essences of the entities, x, y, … , mentioned therein. Presumably, the best hope for non-modalists, in response to the question at hand, is to follow the strategy proposed by Fine for the logical case. Recall that, in the logical case, Fine’s strategy is to explain the logical connection between a consequentially essential truth and the constitutively essential truth from which it follows in part by reference to the nature of the logical operations that are relevant to the case at hand. Thus, the inference from the constitutively essential truth that Socrates’ singleton set contains Socrates as a member to the consequentially essential truth that Socrates’ singleton set contains some member or other, on this conception, is explained in part by reference to the fact that it lies in the nature of the

327

Kathrin Koslicki

existential quantifier to license inferences which instantiate the rule of existential generalization. Similarly, applying this strategy to the present case, non-modalists who (like Fine and Lowe) accept the one-way entailment in (EN) in the simple case from statements of essence to the corresponding metaphysically necessary truths may opt to appeal to the nature of the (constitutive or consequential) essence operator to explain the validity of inferences which follow the pattern described in (EN), from a statement of essence to the corresponding metaphysically necessary truth. The same essentialist basis would then also serve to explain the metaphysically necessary truth of instances of the schema stated in (N).16

21.3.2

The “Cosmic” Strategy

As our preceding discussion has brought out, the non-modalist’s program of explaining metaphysical necessity in terms of essence often requires widening the class of relevant essences beyond those to which the initial essentialist premise-set is explicitly indexed. In the simple case, for example, facts concerning the nature of the (constitutive or consequential) essence operator were brought in to explain why the one-way entailment stated in (EN) from statements of essence to the corresponding metaphysically necessary truths itself might be thought to hold with metaphysical necessity. In the logical case, the derivation of a consequentially essential truth from a constitutively essential truth was justified in part by appeal to the nature of the various logical operations licensing the inference in question. In the mathematical case, the derivation of a metaphysically necessary truth concerning one type of geometrical entity (viz., ellipses) was seen to require an appeal to the nature of other relevant types of geometrical entities (viz., cones). Finally, in the metaphysical case, the class of relevant essences needed to explain facts about the persistence-conditions of certain objects (viz., statues and the lumps of clay constituting them) was broadened to include facts about the nature of various relevant properties (viz., shape-, resemblance, and representational properties). Only in the scientific case, the non-modalist’s strategy diverged from the pattern just cited, since in this case the necessary truth in question was thought to be physically or naturally, rather than metaphysically, necessary, and therefore does not instantiate the non-modalist pattern in which a metaphysically necessary truth in some way follows from some essential truths. Given the general need to broaden the class of entities whose essences might be relevant to the derivation of any given metaphysically necessary truth beyond those to which the initial essentialist premise-set is explicitly indexed, it is perhaps not surprising that Fine opts for what might be called the “cosmic strategy”, according to which the metaphysically necessary truths are true in virtue of the nature of all objects whatsoever: Indeed, it seems to me that far from viewing essence as a special case of metaphysical necessity, we should view metaphysical necessity as a special case of essence. For each class of objects, be they concepts or individuals or entities of some other kind, will give rise to its own domain of necessary truths, the truths which flow from the nature of the objects in question. The metaphysically necessary truths can then be identified with the propositions which are true in virtue of the nature of all objects whatever. (Fine 1994: 9)17 In the context of Fine’s logic of essence, we can see why the move towards the cosmic strategy might seem to be warranted. For, first, Fine’s system operates with consequential, rather than constitutive, essence and therefore already casts its net very widely, by including all the logical consequences of any given constitutively essential truth in the collection of truths that are said to

328

The “Reduction” of Necessity to Non-Modal Essence

be essential to any given entity, provided that no additional objectual content is introduced. Secondly, there appears to be no general logical strategy by means of which the class of relevant essences can be widened beyond those to which an initial essentialist premise-set is explicitly indexed, in order to arrive at exactly those that are relevant to the derivation of any given metaphysically necessary truth. For, as we saw in the preceding sections, filling in the details that might underlie any particular case requires establishing non-logical connections between related phenomena, e.g., between ellipses and cones or between shape-properties and representational properties. By widening the class of relevant essences to include those of all objects whatsoever, Fine’s cosmic strategy therefore might seem to guarantee that enough essences are available to license any particular transition from a statement of essence to a metaphysically necessary truth. Given that the non-modalist’s derivation of metaphysical necessity from essence is meant to be explanatory, however, the cosmic strategy appears to overshoot its intended target. For, unless the metaphysically necessary truths at issue are in fact absolutely general and concern the natures of all objects whatsoever, it seems that Fine’s cosmic strategy will be guilty of loading up the explanatory basis from which any given metaphysically necessary truth is supposed to follow with entities whose essences do not directly contribute, and are therefore explanatorily irrelevant, to the case at hand. To illustrate, suppose the aim is to derive the metaphysically necessary truth that Socrates is an animal in part from the constitutively essential truth that Socrates is human. In that case, it would seem that only facts concerning the essences of human beings and animals are directly relevant to the explanatory connection at issue. There is thus no need to bring in the natures of all objects whatsoever; and doing so will actually diminish the explanatory power of the essentialist premise-set from which the statement, “It is metaphysically necessary for Socrates to be an animal”, is supposed to be derived. For, in general, if a truth of the form, “It is (constitutively or consequentially) essential to x, y, … , that A”, contributes to the derivation of a metaphysically necessary truth of the form, “It is metaphysically necessary that B”, there is no guarantee that expanding the essentialist premise-set by adding further essential truths will increase or even preserve the explanatory power of the initial premise-set. For example, the fact that it is constitutively essential to the Eiffel Tower to be a tower has nothing to contribute to any proposed derivation of “It is metaphysically necessary for Socrates to be an animal” from statements concerning the essences of human beings and animals. Thus, adding the statement, “It is constitutively essential to the Eiffel Tower to be a tower”, to the essentialist premise-set from which the statement, “It is metaphysically necessary that Socrates is an animal”, is supposed to follow, in effect weakens the explanatory power of the essentialist premise-set in question, since it now includes irrelevant information concerning the essences of unrelated entities, viz., the Eiffel Tower. Expanding the relevant class of essences to include those of all objects whatsoever, as the cosmic strategy proposes to do, will of course only worsen the problem just cited.18 When it comes to the non-modalist’s program of explaining metaphysical necessity in terms of essence, there thus appears to be no logical “one-size-fits-all” method of supplying a suitable class of entities whose essences are relevant to the derivation of any given metaphysically necessary truth. Rather, the process unfolds, at least in part, by non-logical means and requires a case-by-case engagement with the specific types of entities whose essences are thought to contribute to the derivation in question.

329

Kathrin Koslicki

21.4 Conclusion Our aim in this chapter was to bring out, with respect to various types of cases and explanatory connections, how non-modalists may proceed with their goal of explaining metaphysical necessity in terms of essence. In each case, we found that, in order to derive a metaphysically necessary truth from a statement of essence, it was necessary to bring in additional facts concerning the essences of related entities beyond those that are explicitly appealed to in the initial essentialist premise-set. It appears, however, that there is no logical “one-size-fits-all” strategy by means of which the relevant class of essentialist facts can be widened in the appropriate way. Thus, a successful execution of the non-modalist’s program requires a non-logical engagement with specific cases to make good on the promise that metaphysically necessary truths can, in all cases, be derived from statements of essence.

21.5

Related Topics

Correia, Fabrice, “Non-Modal Conceptions of Essence” Kovacs, David, “Essence, Grounding, and Explanation” Litland, Jon, “Logic of Essence” Tahko, Tuomas, “Natural Kind Essentialism” Torza, Alessandro, “Modal Conceptions of Essence”

Notes 1 Strictly speaking, non-modalists need not accept that essence is prior to metaphysical modality. A further (though uncommon) option for non-modalists would be simply to accept that neither is prior to the other. Given the state of the literature, however, I will address versions of non-modalism which do accept that essence is, in some sense, prior to metaphysical modality; and that the latter is, in some way, to be explained in terms of the former. 2 See Torza (this volume) and Correia (this volume) for a discussion of modal and non-modal conceptions of essence, respectively. 3 See also Correia (this volume, Sections 8.5– 8.7) for discussion relevant to what I call “the simple case”. 4 Unless otherwise noted, I assume that the type of necessity at issue, represented by “□”, is metaphysical necessity. To state essentialist claims, here and in what follows, I avail myself of the indexed operator notation introduced in Fine 1995a: for each predicate, “F”, there is an operator, “□F”, to be read as “it is true in virtue of the nature of the objects that are F that … ”, which denotes an unanalyzed relation between the objects that are F and a proposition. Essentialist statements of the form, “□FA”, in this framework, are explicitly relativized to their source, viz., in this case, the objects that are F in virtue of whose identity the proposition that A is true. If a proposition is true in virtue of the nature of a single object, y, a predicate which applies to y uniquely may be formulated using lambda-abstraction, viz., “λx(x=y)”; similarly for pluralities of objects. For reasons of simplicity, I side-step the predicate-notation in the main text and follow Fine’s informal convention of speaking of objects directly as the source of essentialist truths. Fine uses the terms, “essence”, “identity”, and “nature” interchangeably ( Fine 1995b: 69, n. 2) and takes definitions to be statements of essence ( Fine 2015: 308). See Litland (this volume) for more discussion of Fine’s notation. 5 For the sake of simplicity, I formulate these claims, here and in what follows, directly in terms of propositions, rather than in terms of sentences and the propositions they express. 6 See for example Leech (2018) for arguments to the effect that an inference like that from (1) to (2) may be held to be objectionable. According to Leech, “what something is does not tell us—absent further assumptions—what something must be” (320). 7 A similar sentiment is also voiced for example in ( Lowe 2012: 106). 8 “⊢” is intended to stand for an informal notion of entailment, rather than for deducibility in some

330

The “Reduction” of Necessity to Non-Modal Essence

9

10

11 12 13 14 15

16

17

18

formal system. See Ditter 2022, for discussion of whether the variables, “x, y, … ” (or other expressions) to which the essence-operator is indexed must occur in the proposition that is embedded within the operators in question. If the propositions in question are not the same, then the relevant relation of “following from” is no longer that specified in (EN), which, as formulated, requires that the propositions in question be the same. We might try to generalize (EN) by allowing that a suitable replacement proposition may be substituted for the proposition embedded within the necessity-operator. The question before us then becomes what counts as a “suitable replacement” in this context, since the propositions in question cannot vary independently. For example, it does not follow, in the relevant sense, from “Socrates’ singleton set has Socrates as its sole member” that 2 is a number, even if we assume that the former is an essentialist truth and the latter holds with metaphysical necessity. Of course, given (EN), the proposition embedded within the essence operator in (6) also holds with metaphysical necessity. Since the proposition embedded within the essence operator in (6) logically entails the proposition embedded within the necessity operator in (7), it may be tempting to think that the connection between (6) and (7) can be captured by appeal to (EN), together with the idea that necessity is closed under logical consequence. However, given non-modalism, the latter idea should presumably itself be explained in some way by appeal to some essentialist facts. See also Correia (this volume) and Litland (this volume) for further discussion of the distinction between constitutive and consequential essence. See also Correia (2012) for a further elaboration of the idea that the essence of a logical concept is nonpropositional and is given, instead, by certain rules of inference. For further discussion of the Grounding Problem, see for example Koslicki (2018) and the references therein. See also Tahko (this volume) for relevant discussion of statements of essence and metaphysically necessary truths involving natural kinds. Other theorists, however, employ a different strategy from that illustrated here and propose instead that statements expressing laws of nature hold with metaphysical necessity. (See for example the defense of dispositional essentialism by Bird (2001) and others.) These theorists, then, assuming that they also accept non-modalism, would be committed to the idea that (18) can in some way be derived from (17). See also, for example, Wallner (2019) for useful discussion of the question of whether non-modalists are committed to unexplained metaphysical necessities; and Ditter (2022) for discussion of the proposal that metaphysical necessity should be construed as truth in virtue of the nature of all propositions. The cosmic strategy, which is here characterized informally, is also embedded within Fine’s logic of essence by means of a monotonicity principle, stated in Axiom II (v), according to which a proposition which is true in virtue of the nature of the Fs is also true in virtue of the nature of the Gs, provided that the Fs are also Gs ( Fine 1995a: 247). See Litland (this volume) for further discussion of monotonicity as well as Zylstra (2019). As pointed out by Robert Michels (see Michels 2018: 8), Fine anticipates this worry in a parenthetical remark (italicized here): “[a] necessary truth can be taken to be a proposition that is true in virtue of the identity of all objects […] (not that all objects […] need be relevant)” ( Fine 1995b: 56). We should, in this context, distinguish between the following two aims: (i) deriving all metaphysically necessary truths from all statements of essence taken together; and (ii) deriving some particular metaphysically necessary truth from some particular statements of essence. It very well may be the case that Fine’s cosmic strategy helps us with the first goal; in relation to the second goal, however, which has been my main concern, the cosmic strategy overshoots its explanatory target by introducing potentially irrelevant information into class of essentialist facts from which a particular metaphysically necessary truth is supposed to follow.

References Bird, Alexander (2001) “Necessarily, Salt Dissolves in Water,” Analysis 61 (4) (October 2001) 267–274. Correia, Fabrice (2012) “On the Reduction of Necessity to Essence,” Philosophy and Phenomenological Research 84 (3) 639–653. Ditter, Andreas (2022) “Essence and Necessity,” Journal of Philosophical Logic 51 653–690.

331

Kathrin Koslicki Fine, Kit (1994) “Essence and Modality,” Philosophical Perspectives 8 (Logic and Language), 1–16. Fine, Kit (1995a) “The Logic of Essence,” Journal of Philosophical Logic 24 241–273. Fine, Kit (1995b) “Senses of Essence,” in Walter Sinnott-Armstrong, Diana Raffman, and Nicholas Asher, eds., Modality, Morality, and Belief, Essays in Honor of Ruth Barcan Marcus, New York: Cambridge University Press, 53–73. Fine, Kit (1995c) “Ontological Dependence,” Proceedings of the Aristotelian Society 95, 269–290. Fine, Kit (2002) “The Varieties of Necessity,” in Tamar Szabo Gendler and John Hawthorne, eds., Conceivability and Possibility, Oxford: Oxford University Press, 253–281, reprinted in Kit Fine (2005a) Modality and Tense, Oxford: Clarendon Press, 235–260. Fine, Kit (2005b) “Necessity and Non-Existence,” in Kit Fine (2005a) Modality and Tense, Oxford: Clarendon Press, 321–354. Fine, Kit (2015) “Unified Foundations for Essence and Ground,” Journal of the American Philosophical Association 1 (2) (Summer 2015) 296–311, URL=< 10.1017/apa.2014.26>. Koslicki, Kathrin (2018) “Towards a Hylomorphic Solution to the Grounding Problem,” Royal Institute of Philosophy Supplements to Philosophy 82 (2018), Metaphysics, 333–364. Leech, Jessica (2018) “Essence and Mere Necessity,” Royal Institute of Philosophy Supplement 82 309–332. Lowe, E. J. (2006) The Four-Category Ontology: A Metaphysical Foundation for Natural Science, Oxford: Clarendon Press. Lowe, E. J. (2008) “Two Notions of Being: Entity and Essence,” in Robin Le Poidevin, ed., Being: Developments in Contemporary Metaphysics, Cambridge: Cambridge University Press, 23–48. Lowe, E. J. (2012) “Essence and Ontology,” in Lukáš Novak, Daniel D. Novotný, Prokop Sousedik, and David Svoboda, eds., Metaphysics: Aristotelian, Scholastic, Analytic, Berlin: De Gruyter, 93–111. Michels, Robert (2018) “How (Not) to Define Modality in Terms of Essence,” Philosophical Studies, published online 25 January 2018, URL=< 10.1007/s11098-018-1040-8>. Wallner, Michael (2019) “The Structure of Essentialist Explanations of Necessity,” Thought. URL= < 10.1002/tht3.436>, 1–10. Zylstra, Justin (2019) “Collective Essence and Monotonicity,” Erkenntnis 84 (5) 1087–1101.

332

22 PERSONS Annina Loets

Like everything, I am something. I am also someone, a person. Many philosophers think that this additional fact about me is of great importance. Some think that by virtue of being a person I have a distinctive moral standing. Some think that by virtue of being a person I could, at least in principle, survive death. Claims such as these are naturally interpreted as being about the nature or essence of persons. In this chapter, I’ll explore various themes in the literature on persons, how they interact with one another, and what implications, if any, they have for wider debates in metaphysics and beyond.

22.1 Three Lockean Ideas Much contemporary thinking about persons can be traced back to three theses about persons formulated by John Locke in his Essay Concerning Human Understanding. Cognitive Capacity: Accountability:

Identity:

“Person stands for (…) a thinking intelligent Being, that has reason and reflection, and can consider itself as itself, the same thinking thing in different times and places” (Locke 1689: 2.27.9). “Person is a forensic term appropriating actions and their merit; and so belongs only to intelligent agents capable of a law, and happiness and misery” (Locke 1689: 2.27.26). “Consciousness always accompanies thinking, and ‘tis that, that makes everyone to be, what he calls self. (…) in this alone consists personal identity, i.e., the sameness of rational Being” (Locke 1689: 2.27.9).

The first two theses characterize personhood in broadly functional terms. Roughly, glossing over subtleties of Locke exegesis, the first idea is that persons are all (and presumably only) those things with certain cognitive capacities. The second idea is that persons are all (and presumably only) those things which can be appropriately held to account for their actions. It is an interesting question whether these two “person-roles” coincide. The third thesis concerns the conditions under which persons are identical, setting out a broadly psychological

DOI: 10.4324/9781003008750-26

333

Annina Loets

approach to personal identity. It is usually assumed that all three theses hold of persons not just contingently, but of necessity. An interesting feature of these Lockean theses about persons is that they leave it open how personhood is realized. Consequently, Lockean views of persons are hospitable to an idea many have found compelling, i.e., that beings of radically different internal constitution could all be persons. To the extent that they play the person-role, octopuses, gods, angels, Martians, perhaps even large language modules, could, in principle, all be persons. The three Lockean ideas still inform many contemporary debates about persons. It’ll be instructive to explore how these ideas are developed and related to one another in the current literature.

22.2

Distinctive Cognitive Capacities

The first Lockean idea is to think of persons as beings with distinctive cognitive capacities. Many have approached the question of which cognitive capacities are distinctive of persons by considering which capacities distinguish human from non-human animals, the assumption being that the latter are not persons. One consequence of putting the question in this way is that capacities which don’t sharply distinguish human from non-human animals receive less attention in the literature, even if they make for plausible necessary conditions on being a person. For instance, David Lewis might well be right in suggesting that it is part of “our common-sense theory of persons” that persons are rational agents, beings whose behavior can be given rationalizing explanations (Lewis 1974: 337). Yet, since many authors are happy to attribute a form of instrumental rationality to the more advanced nonhuman animals, rationality figures less centrally in discussions about the nature of persons. Locke’s claim that persons “can think themselves as themselves” is naturally interpreted as suggesting that persons are distinguished from non-persons in not just being conscious, but in being self-conscious. This thought is developed at length by Lynne Rudder Baker (2000: ch.3) who claims that “what distinguishes persons from all other beings” is “their capacity for (…) a first-person perspective” (ibid.: 4). Such a perspective involves not just the ability to “make self-reference” by thinking “I”-thoughts of which one is the subject, e.g., thoughts expressible by a sentence like “I am tall” (ibid.: 64). In addition, one must be able to “attribute selfreference”, by thinking “I”-thoughts of which one is the object, e.g., thoughts expressible by a sentence like “I wish that I were tall” (ibid.: 64).1 Beings with a first-person perspective so understood have an important precondition for the kind of reflective capacities some have deemed to be distinctive of persons. A prominent advocate of this sort of view is Harry Frankfurt (1971). According to Frankfurt, what distinguishes persons from non-persons is that persons are capable of attitudes towards their own attitudes: they have beliefs about their desires, desires about their desires, and as a result, they might be motivated to change their desires. Frankfurt places particular emphasis on the capacity to have desires about one’s own will, i.e., desires about which desires should be effective in motivating one’s actions. Where such higher-order desires are thwarted, the will is not free. In light of these considerations, Frankfurt proposes that a person is “a type of entity for whom the freedom of its will may be a problem” (Frankfurt 1971: 14). While Frankfurt and Baker stress the significance of the capacity to take attitudes towards one’s own attitudes, Daniel Dennett (1976) highlights the importance of attitudes about attitudes more generally, whether these are one’s own or those of others. According to Dennett, it is distinctive of persons that their behavior can at times only be appropriately explained by attributing to them attitudes about someone’s attitudes (Dennett 1976: 181). For instance, to the

334

Persons

extent that intentionally deceiving someone requires beliefs and desires about their attitudes, behavior that is best explained as an intentional deception could only be the behavior of a person. Further, both Dennett (1976: 186) and Lewis (1969, 1974, 1975) take the capacity to have attitudes about attitudes—one’s own and those of others—to be a precondition for another phenomenon many take to be distinctive of persons: having a language. According to Lewis (1969, 1975), a language L is the language of a particular group of people if, and only if, the group exhibits a convention of truthfulness and trust in L, where such a convention’s being in place involves group members having specific kinds of beliefs about their own beliefs, desires, and reasons, as well as specific kinds of beliefs about the beliefs, desires, and reasons of others. A more demanding version of the idea that a person can take attitudes towards their own attitudes can be found in Tim Scanlon’s work. According to Scanlon, it is distinctive of persons that they are “capable of holding judgment-sensitive attitudes” like anger or respect (Scanlon 1998: 179), i.e., attitudes which, were one ideally rational, one would form (give up) if and only if one judged there to be sufficient reason to form (give up) such an attitude (Scanlon 1998: 21). Plausibly, having judgment-sensitive attitudes implies having attitudes towards one’s attitudes, though the converse does not hold. Scanlon’s take on the distinctive capacities of persons is then strictly more demanding than that of Frankfurt. We have considered different ways of developing the Lockean idea that a person “can think itself as itself”. But this cuts the first Lockean thesis short, which says that a person “can think itself as itself, the same thinking thing in different times and places”. Several authors have placed emphasis on diachronic self-awareness in their discussions of what distinguishes persons from non-persons. For instance, Robert Nozick highlights the importance of having “an overall conception of [one’s] life, and what it adds up to” (Nozick 1974: 49).2 Combining the sorts of reflective abilities discussed previously with such an overall conception of one’s life, Nozick proposes that a person has “the ability to regulate and guide its own life in accordance with some overall conception it chooses to accept” (ibid.). Thus, Nozick can be seen as giving the Lockean conception of a person a Kantian twang: personhood is linked to autonomy, the ability to self-govern.3 Finally, Peter Singer suggests that one factor distinguishing persons from non-persons is how “future-oriented in their preferences” persons are (Singer 1980: 80). Interestingly, Singer (1980: 74–6) explicitly connects these thoughts to the Lockean understanding of a person as a being that “can consider itself as itself, the same thinking thing in different times and places”. Set aside for a moment the question of which capacities are distinctive of persons. An important choice point is whether having those capacities is necessary for being a person, or whether some weaker condition might do. For instance, some philosophers have argued that an iterated capacity—the capacity for these capacities—is necessary and sufficient for being a person (see Baker 2000: 92).4 Others have suggested that membership in a kind the typical members of which have the relevant cognitive capacities is sufficient (though perhaps not necessary) for being a person (cf. Wiggins 1976: 158). As one might expect , the philosophical implications of being a person will vary depending on how this choice is resolved.

22.3 Accountability The second Lockean thought is that “person” is “a forensic term”, one having to do with “appropriating actions and their merit”. It is often construed as the view that persons are the sorts of beings that are, at least in principle, accountable for their actions; persons are appropriate objects of reactive attitudes like praise or blame.

335

Annina Loets

As mentioned above, it is natural to think that this second functional characterization of persons is not independent of the first. In particular, it is tempting to think that persons are, at least in principle, accountable because they are persons, i.e., because they are beings which possess the kinds of cognitive capacities required for accountability. Notice though that if being a person only requires an iterated capacity for the relevant capacities, or membership in a kind the typical members of which have these capacities, being a person would not imply accountability. The idea that person is a forensic notion also connects in interesting ways with the third Lockean idea about personal identity. After all, one is not normally accountable for things which someone else did.5 And so we’d expect judgments about accountability to often walk in lockstep with judgments about identity.6

22.4

Identity

The third Lockean idea is that it is “sameness of rational being alone” that makes for personal identity. It is widely assumed to give voice to a broadly psychological approach to personal identity which still dominates much of the contemporary literature on the topic. To a first approximation, the debate about personal identity concerns the question of under what conditions people are identical. A more precise formulation of the question reveals important background assumptions and choice points. To begin with, the question is usually understood as concerning numerical identity, a relation by which we count.7 If you and I had phones that were numerically identical, we’d only have one phone between the two of us, a shared phone. There is of course a perfectly good sense of “identical” in which you and I can have identical phones despite each of us having a phone of our own. However, authors in the debate about personal identity usually set aside those looser notions of identity.8 Secondly, at issue is usually identity over time, perhaps better called “persistence”. It is widely assumed that identity over time is compatible with change: I used to be shorter than I am now. It would be a mistake to conclude that numerical identity does not imply qualitative identity. Rather, what we have is an instance of the “Problem of Change”.9 One of the most basic principles in the logic of identity is The Indiscernibility of Identicals, according to which if a is the same as b, a has a property if, and only if, b has it, too. But for something to change, it seems, is for that thing to have a property at one time which it lacks at another time. So how is change possible? Instead of giving up on the idea that identity implies indiscernibility, most extant solutions revise their ideas about what change consists in. Thirdly, as Olson (1997: 25) observes, the scope of the question about personal identity is not always entirely clear. Many authors say explicitly that their aim is to complete an analysis of the following form: Identity between Persons: Necessarily, for every persons x and y, x = y if, and only if …

So understood, personal identity is a relation which is defined only over persons. But many authors also claim to be investigating the persistence conditions of persons. For instance, Parfit (1984: 202) commits to answering the following question: “what is necessarily involved in the continued existence of each person over time?” Identity between persons does not answer this question. In particular, it is silent on the question whether a person could persist as anything other than a person, i.e., whether the following principle is true: Perpetuity of Personhood: Necessarily, every person is always a person.

336

Persons

The principle is certainly not obviously true. One might think that, like being a professor, being a person is a property which, if one has it at all, one has only temporarily.10 If an answer to the question of what personal identity consists in is to imply general persistence conditions for persons, it would be better to formulate it as follows: Persistence of Persons: Necessarily, for every person x and every y, x = y if, and only if …

Put in this way, the question under discussion in debates about personal identity is the question what necessary and sufficient conditions are for a person at some time to be something at another time, i.e., to exist at all at that time.11 Should the Perpetuity of Personhood turn out to be true, continued existence and continued existence as a person will coincide. But it seems wise not to prejudge the matter. Finally, many of those engaged in the literature on personal identity appear to strive for a reductive analysis of the identity of persons over time. To see this, notice that it is still considered an important objection to a memory condition on personal identity that it would render the resultant account of personal identity circular; after all it is impossible to remember anyone else’s experiences.12 It is a striking fact that although few people would expect there to be a reductive analysis of numerical identity at a time, our concept of identity being too basic to allow for an analysis in independent terms, many in the literature on persons seem optimistic about the project of giving an adequate reductive analysis of numerical personal identity over time.13 With these observations in place, we can now turn to the debate about personal identity. As noted above, the third Lockean idea according to which “sameness of rational being” makes for personal identity is usually associated with a broadly psychological approach to personal identity. As Parfit (1984: 203) notes, a standard assumption about material objects is that their persistence over time requires a certain amount of spatiotemporal continuity sustained by appropriate causal connections. By contrast, psychological approaches to personal identity are unified by the idea that a different kind of continuity, psychological continuity (perhaps sustained by appropriate causal connections), is necessary and (given certain conditions are met) sufficient for the persistence of persons. Psychological continuity could be cashed out in terms of the continued existence of some entity like a soul or an immaterial mind. More common is the idea to think of psychological continuity as continuity in certain mental states and capacities. Consider Locke’s claim that “the identity of a person” reaches back as far as “the consciousness [which] accompanies thinking” does (Locke 1689: 2.27.9), a claim often interpreted as the view that a person x at one time is only identical to some y at an earlier time if x remembers y’s experiences (or, to block the circularity charge, “quasi”-remembers those experiences, where “quasi-memory” is a relation functionally analogous to memory but which doesn’t presuppose identity to the subject of the experience, see Parfit 1971). On this view, for x to be y, x needs to be psychologically related to y. However, it is common today to require continuity in a broader set of psychological states and capacities, including memory, intention, character dispositions, etc. If personal identity is to be reduced to psychological continuity, psychological continuity needs to be an equivalence relation, i.e., a reflexive, symmetric, and transitive relation. A specification of the relevant psychological relations which involves temporally directed relations needs to ensure that personal identity ends up being a symmetric relation. For instance, I am psychologically related to a certain postdoc by virtue of remembering many of her experiences. That I remember much of what the postdoc did does not imply that the

337

Annina Loets

postdoc remembers any of the things I am currently doing. So, the relation I bear to the postdoc is not symmetric. Likewise, that the postdoc’s intentions explain many of my current actions does not imply that my intentions explain any of the postdoc’s actions. So, the relation the postdoc is bearing to me is not symmetric. Lewis (1976: 61) suggests that this problem can be overcome by defining two non-symmetric relations: forward psychological connectedness and backward psychological connectedness. Psychological connectedness simpliciter is then the disjunction of those two relations, a symmetric relation. Identity is a transitive relation, but the psychological relations which many take to matter for personal identity do not seem to be. Focusing on memory, we can adapt a famous example from Thomas Reid (1785: 248f.) to illustrate the point. Within a single life, a professor might remember much of what the graduate student did, and the graduate might remember much of what the undergraduate student did, although the professor does not remember much at all of what the undergraduate student did. Other examples of failures of transitivity in forwardlooking and backward-looking relations are easy to produce. To ensure that personal identity is a transitive relation, proponents of the psychological approach therefore typically focus on the transitive closure of the relevant psychological relation. Taking Lewis’s notion of psychological connectedness simpliciter as a basis, we could then say that a is psychologically continuous with b if, and only if, there is a finite chain of things psychologically connected to one another which links a to b (cf. Parfit 1984: 206). An important choice point is whether psychological continuity needs to be sustained by appropriate causal connections, and if so, which causal connections are appropriate. Parfit (1984: 208) distinguishes between a “narrow” view on which psychological continuity over time must have its normal cause, a “wider” view on which psychological continuity over time must have a reliable cause, and “the widest” view on which psychological continuity over time could be sustained by any cause. This point is relevant, if we wish to contrast a psychological approach to personal identity with other approaches, e.g., the view that personal identity (at least of human persons) consists in identity of the body (Thomson 2007), identity of the organism (Olson 1997; Snowden 1990), or identity of the brain (Nagel 1986; Unger 1990). If there are no normal cases in which psychological continuity is sustained across different bodies, organisms, or brains, then the narrow psychological approach might imply such alternative views on personal identity, at least in all actual cases.14 By contrast, the wider views on what are appropriate causes of psychological continuity can in many cases be expected to yield significantly different verdicts from alternative non-psychological views. What motivates a psychological approach to personal identity? Some have sought to give theoretical arguments. For instance, Lewis (1976: 56f.) can be seen as arguing for a psychological approach of personal identity by suggesting that only such an approach manages to reconcile the following two plausible theses about what matters most to us in survival15: 1. What matters most to us in survival is that our mental lives should flow on. 2. What matters most to us in survival is that we ourselves should continue to exist. Lewis (1976) proposes a four-dimensional ontology of persons geared towards vindicating the claim that for a person to continue to exist just is for their mental life to flow on, in a sense he makes precise. However, by far the most influential arguments for the third Lockean thesis are by way of thought experiments. Particularly prominent are cases which seem to suggest that it is possible to move persons from one body to another by somehow instilling their mental lives in another

338

Persons

body. Locke himself presented a “body switching” case to support a psychological criterion of identity, the case of the Prince and the Cobbler (Locke 1689: 2.27.15). Here is a version from Judith Jarvis Thomson (2007: 160): Here are two people, Brown and Robinson, and let us give the names ‘Brown-body’ and ‘Robinson-body’ to their bodies. (…) We transplant the brain of Brown-body into Robinson-body, destroying the rest of Brown-body. Let us suppose that the operation succeeds, and that the survivor thinks he is Brown, wants much of what Brown wanted, seems to remember a good bit of what Brown experienced, and so on. Many people have the strong intuition that such an operation would have the effect of moving Brown from Brown-body into Robinson-body. However, this judgement, in itself, does not yet support a psychological criterion of personal identity. It is consistent with a view on which personal identity consists in identity of the brain, or at least those parts of the brain responsible for cognition.16 Much of the literature on personal identity is concerned with probing judgments about apparent body switching cases such as these, as well as with understanding to what details in the structure of the cases our judgments are sensitive.17 A serious challenge to psychological approaches (though perhaps not to such approaches alone) stems from so-called “fission cases”. Suppose you are a proponent of a (narrow) psychological account of personal identity. You acknowledge that a transplant of your cerebrum (the part of your brain responsible for cognition) to another body could, in principle, preserve psychological continuity along a sufficiently normal causal path. Consequently, you should think that you could, in principle, move from one body to another by way of a cerebrum transplant. Furthermore, to the extent that even a transplant of only one of your cerebral hemispheres could preserve psychological continuity along a sufficiently normal causal path, you should also think that you could, in principle, move from one body to another by way of a partial cerebrum transplant, transplant of only one of the cerebral hemispheres.18 The puzzle now is this: what should we say about a case where someone’s cerebrum is split in half, the left hemisphere is grafted onto the brain stem of one unconscious human organism and the right hemisphere is grafted onto the brainstem of another unconscious human organism?19 Could the initial person survive the operation? If so, who are they? There do not seem to be any good answers to these questions for a proponent of the psychological approach. Consider the two people waking up, call them “Lefty” and “Righty”, respectively. Each of them is psychologically continuous with the initial person: they seem to remember a good deal of what the initial person experienced, they are eager to carry out a good deal of the initial person’s intentions, they are just as impatient as the initial person etc. Assuming a psychological approach to personal identity, if Lefty and Righty are each psychologically continuous with the initial person, both are predicted to be identical to the initial person. Since identity is an equivalence relation, this would imply that Lefty is identical to Righty. But that seems absurd. Lefty and Righty are two people, not one. They are located in different rooms of the hospital, one may be lying down, the other may be seated, many of their thoughts, even if otherwise indistinguishable, would be about different objects. So how could they be numerically identical?20 On the other hand, if only one of Lefty and Righty were the initial person, which one would it be? Isn’t any answer going to be entirely arbitrary? After all, what sorts of facts could make it the case that the initial person is Lefty rather than Righty?21

339

Annina Loets

Neither does it seem entirely satisfactory to say that neither Lefty nor Righty is the initial person; to the extent that there is no other good candidate around, we’d get the conclusion that the initial person simply does not survive the operation. But that seems odd. Had we transplanted only the left hemisphere, and destroyed the right hemisphere, Lefty’s gaining consciousness after the operation and continuing the life of the initial person would have seemed to be a clear case of survival, and a massive success. Why should a successful second transplantation make any difference here? As Parfit puts it aptly: “how could a double success be a failure?” (Parfit 1971: 5). A lot has been written about fission. Some have argued that fission does not produce two people. Rather, in cases where fission into two will occur, there are already two people prior to fission, ready to be separated by the operation (see Perry 1972; Lewis 1976; Noonan 1989; Robinson 1985). The proposal raises interesting questions about counting as well as about the contents and rationality of various first-person thoughts.22 Others have considered the view that the initial person survives as an object composed of Righty and Lefty. The initial person ends up being something like a group agent, at least so long as a certain amount of cooperation remains between Lefty and Righty.23 Parfit (1971, 1984) famously adopts a more radical response, claiming that identity, and hence continued existence, is not ultimately what matters to us in survival. Parfit claims that “the relation of the original person to each of the resulting people contains all that interests us—all that matters—in any ordinary case of survival” (1971: 10). Fission and other forms of duplication, although incompatible with continued existence, are “as good as ordinary survival” (Parfit 1984: 201). Our judgments about what matters practically track relations of psychological continuity—or would at least, if we were rational. Pure identify facts, e.g., whether it was me who did something, whether it is me who will wake up after the operation, whether it is me you will marry, do not matter practically and morally. Instead, only facts about psychological continuity do, an interesting though by no means obvious view. As Parfit readily admits, the view is deeply revisionary and would require rethinking many of our practices which are built around the assumption that one person does not survive as two.24 An interesting question is how dialectically stable this position is for someone who adopts the psychological approach to personal identity primarily because of their intuitions in body switching cases, but who is also sympathetic to the Parfitian line on what matters practically. As Olson (1997: 56f.) points out, such a person should at least be open to the idea that their intuitions about body switching cases simply don’t track personal identity, but rather track that which matters. It might then turn out that the transplant intuitions are entirely compatible with broadly physical approaches to personal identity according to which psychological continuity is neither necessary nor sufficient for identity. Independently of its revisionary character, the Parfitian line may thus turn out to be dialectically unstable. There is reason to think that fission cases pose a challenge not only to psychological approaches to personal identity. With enough creativity it should be possible to devise credible fission cases for any view on which personal identity is preserved through the identity of an entity that can divide, whether it is organisms, bodies, or brains. By contrast, views on which personal identity consists in the identity of an indivisible entity (e.g., a soul) seem to be safe from fission problems. Yet, such views face problems of their own.25 We have seen in this section how the three Lockean ideas about persons mentioned at the outset have been taken up and developed in the contemporary literature on persons. But questions remain: Are Lockean views of persons plausible? And what, if anything, turns on one’s theory of persons?

340

Persons

22.5

Philosophical Implications

I began this article with the seemingly uncontroversial claim that I am a person. To see how controversial Lockean views of persons can be, notice that, at least on some ways of developing the Lockean package, my being a person does not seem to sit easily with other popular views about the nature of a human person, e.g., Essentiality of humanity: Necessarily, every human being is essentially human. Essentiality of origins: Necessarily, every human being has its origins essentially.

If psychological continuity along a reliable causal path were sufficient for my continued existence, I could, in principle, be turned into something that is not made of flesh and blood at all, but of wires and metal (Baker 2000: 19; 123).26 Such a being would not be human, so if I could be that being, I could be non-human. If death is a biological notion, Lockean persons would then also be beings which could, in principle, survive death. Similarly, if what matters for being me is just appropriate psychological unification across time, one might wonder whether there are any material conditions on my origins such as originating from a particular sperm and egg (see Dasgupta 2018: 560; Dorr et al. 2021: 324).27 By contrast, Lockeanism lends some credence to the previously considered principle: Perpetuity of Personhood: Necessarily, every person is always a person.

The crucial premise in an argument for the Perpetuity of Personhood (PP) would be that since only a person could have my kind of mental life, only a person could be psychologically continuous with me. But as noted above, the PP is certainly not uncontroversial. Olson (1997: ch.4) argues that the PP sits uneasily with the apparent Moorean fact that I once was in my mother’s womb; after all it is doubtful that anything in my mother’s womb had the relevant kinds of cognitive capacities. Notice that it is of no help to adopt one of the weaker versions of the first Lockean thesis, according to which a capacity for such capacities, or membership in a kind the typically members of which have such capacities, is sufficient for being a person. Perhaps on such conceptions of being a person, a later stage embryo could be a person. Whether such a later stage embryo could be psychologically continuous with me, by contrast, is a different matter, and is indeed doubtful. Some may be prepared to give up the third Lockean thesis, while holding on to the first and second theses. Lockeanism Light, as we might call such a package of views, would be compatible with the idea that one’s being a person is a contingent and temporary matter. Others might be willing to bite the bullet. Baker (2000: 205), for instance, argues that while perhaps I was never in my mother’s womb, I’m now constituted by something that was in my mother’s womb. Why think our intuitions on this matter are sufficiently fine-grained to distinguish between identity and constitution? Proponents of biological approaches to personal identity, like Snowden (1990) and Olson (1997, 2007), claim that the idea that human persons are constituted by, yet distinct from, human animals, raises more problems than it can solve. If I’m a person constituted by, but distinct from, a human animal, am I an animal? If no, Lockeanism seems to be at odds with the Moorean (or Darwinian) fact that I am a human animal. If yes, are there two human animals here where we would have expected at most one? Likewise: if there is a human animal here which constitutes a person, is that animal a person? If no, why not? Doesn’t it have all the capacities distinctive of a person? If yes, are there two persons here where we would have

341

Annina Loets

expected at most one? It should be flagged that these are objections to constitution without identity more generally and have little to do with anything specific to persons. (Compare: if a piece of tin constitutes a tin soldier, is it itself a tin soldier? If not, why not? If yes, don’t we have too many tin soldiers? Or: if a tin soldier weighs 30g and the piece of tin it is made from does too, don’t we have two times 30g and hence 60g worth of tin soldier?). Appeals to property inheritance, which are standard in the literature on material constitution, might be useful in addressing questions such as these.28 Many philosophers take the view that personhood is morally significant. Consider the widely held view that, if one had to choose between taking the life of a human being and a non-human animal, taking the human life would normally be morally worse. What, other than blatant speciesism, could justify such a claim?29 An initially tempting idea is to ground the moral status of human beings in their being persons, beings with particular cognitive capacities. For instance, Nozick (1974: ch.3) and Singer (1980: ch.4) draw explicitly on the diachronic self-awareness of which persons are capable to argue that being killed would be worse for a person than for a non-person. And Scanlon (1998) argues that it is by virtue of our being persons, beings capable of judgmentsensitive attitudes, that we “have strong reason to want our actions to be justifiable to [one another]” and hence to treat one another “in accordance with principles that [one] could not reasonably reject” (Scanlon 1998: 180)—the basic principle of his Contractualism. However, notice that on standard ways of developing the Lockean view of persons, not all human beings are persons. Many have the strong intuition that nevertheless, it would normally be worse to take the life of a new-born human baby or that of a severely cognitively impaired human being, than to take that of a chimpanzee, say.30 Is this view compatible with a higher-moral status of persons over non-persons? Some have explored the view that what matters morally is potentiality, the capacity to develop the capacities distinctive of persons, which matter morally (Harman 2003). Others have argued that it is our kind membership after all, our all being human beings, which matters morally (see Cohen 1986: 866; Scanlon 1998: 186; Finnis 1995: 48).31 Yet others have argued that the relevant human beings do, in fact, have the kinds of sophisticated cognitive capacities which on some accounts matter morally, though they are incompletely realized (Jaworska & Tannenbaum 2014). As noted earlier, one could take the view that it is sufficient for being a person that one has an iterated capacity for certain cognitive capacities, or that one is a member of a species whose typical members have these capacities. If one also thought that those properties are morally significant, one could hold on to the idea that it lies in the nature of persons to have a distinctive moral standing. However, what ultimately does the work in different accounts of the grounds of moral status are views on the question of which properties are normatively significant. The question of whether the normatively significant properties are the properties all and only persons have is secondary.

22.6 Concluding Remarks Theories of personhood can be seen as concerned with the question of what makes something someone. We’ve explored different ways of developing a Lockean view of persons, and we’ve considered what they imply about the essence of persons, e.g., whether by virtue of being a person I could survive death, or whether by virtue of being a person I have a distinctive moral standing. Ultimately, it emerges that many of the questions of interest can be asked and answered without mentioning persons at all. Instead of asking whether a person can survive biological death, we could ask whether something which persists by way of psychological

342

Persons

continuity can survive biological death. And instead of asking whether all and only persons have a distinctive moral standing, we can ask whether all and only things with these and those properties have a distinctive moral standing. While some might feel the urge to dispense with “person-talk” altogether and to instead ask the relevant questions directly, others might like the idea of coordinating their views on the relevant topics in such a way that such questions are answered by an account of what a person is.

22.7 Cross-References • On p. 7, I raise issues about persistence and change. For more on this topic see Chapter 19 (“Identity, Persistence, and Individuation”) in this volume. • On p. 17, I mention the essentiality of origins. For more on this topic see Chapter 11 (“Origin Essentialism”) in this volume. • On p. 19, I draw out possible ethical implications of personhood. For more on the relation between essence and value see Chapter 28 (“Ethical Value”) in this volume.

Notes 1 See also Castañeda (1966). According to Baker (2000: 6), the “merely perspectival phenomena” discussed by John Perry (1979) and Lewis (1979) are only “weak first-person phenomena” which do not constitute a first-person perspective. 2 See also Baker (2000: 11; 147). 3 A particularly demanding view of personhood results from combining the two previous ideas and claiming that persons are capable of both judgment-sensitive attitudes and of self-governance. This approximates the conception of personhood set out by Kant (1785) in his Groundwork of the Metaphysics of Morals, and further developed in contemporary Kantian ethical theory. On this view, persons are beings with rational, autonomous wills: they can discern what is substantively (rather than just instrumentally) valuable, determine, based on rational reflection alone, what is morally required, and impose those requirements on themselves. See, for instance, Korsgaard (1996) for a development of these ideas. 4 I’m assuming that although the capacity for a capacity does not in general imply that capacity, the converse holds: a capacity implies the capacity for that capacity. 5 I’m glossing over special cases like responsibility for the doings of one’s children, employees, etc. 6 Though see Olson (1997: ch.3). 7 The claim that we count by numerical identity is not undisputed; see Lewis (1986, 1993) and Liebesman (2015) for arguments to the contrary. Relatedly, numerical identity may not be the only relation by which one can count; other equivalence relations might do. 8 Though see Baxter (1988). It is important not to conflate such looser notions with the notion of qualitative identity. Pace Parfit (1984: 201; 2012: 5), qualitative identity requires indiscernibility with respect to every qualitative condition. No matter how similar our two phones look, they would not be qualitatively identical if yours was bought online while mine was bought in a shop. 9 See Haslanger (2003) for a helpful introduction to the debate. See also Chapter 19 in this volume (“Identity, Persistence, and Individuation”). 10 Baker (2000) adopts a principle which is stronger (assuming that things have their essential properties throughout their existence), namely the Essentiality of Personhood, according to which necessarily every person is essentially a person. 11 Necessitists may wish to distinguish between being something and being concrete, see Williamson (2013). 12 The objection goes back to Bishop Butler (1736: 298). 13 Though see Shumener (2017, 2020). While not in the business of offering a conceptual analysis of identity, Shumener explores a view on which identity and distinctness facts are not fundamental.

343

Annina Loets 14 What the predictions of a narrow psychological approach would be under various counterfactual suppositions depends on whether what is normal is contingent. Parfit (1984) seems to tacitly assume that what’s normal is necessarily normal, or at least normally normal. But that’s not obvious. See Carter (2019) and Loets (2021) for discussion. 15 Though note that neither claim is uncontroversial. Williams (1970) and Thomson (2007) deny (1), while Shoemaker (1970), Parfit (1971, 1984), and Olson (1997) deny (2). 16 Both Olson (1997: 9f.) and McMahan (2002: 34) stress the importance of being clearer on which parts of the brain are transplanted. 17 Williams (1970) considers how our judgments differ from a first-person future-looking perspective. Thomson (2007) and Parfit (2007) consider numerous variations of “body switching” cases to figure out whether intuition favours a psychological or a bodily criterion of personal identity. Olson (1997: 9) asks whether a person would go where their cerebrum goes, or stay behind with the irreversibly unconscious, yet living human organism. 18 There is some pressure to say this. According to Olson “hemispherectomy”, a procedure in which one of the cerebral hemispheres is removed, is an accepted treatment of severe epilepsy and is “routinely performed in hospitals around the world” ( Olson 1997: 50). An account of personal identity which predicts such operations to end the patient’s existence might then be too revisionary to be acceptable. 19 To alleviate potential ethical qualms, imagine that two people had accidents that left them in persistent vegetative states. Prior to these accidents both had signed documents confirming that in the event of such an accident, they would like to donate their remains to people seeking to leave diseased bodies by way of cerebral transplants. Those attracted to bodily or biological approaches to personal identity will of course insist that what they signed off on is that they themselves would be given different minds. But recall that we are considering the case under the supposition of a psychological approach to personal identity. 20 Could the initial person be both lying down and sitting up in the same way as a person with two cars could be parked both inside a garage and outside of a bar? Nothing on the psychological approach rules out such a view. Notice, however, that if psychological continuity is necessary and sufficient for identity over time, the puzzle arises as soon as Lefty and Righty cease to be psychologically continuous with one another while each remaining psychologically continuous with the initial person. 21 Perhaps only one of Lefty and Righty is the initial person, though it is indeterminate which? Why assume that there need to be any independent explanation of such haecceitistic facts? This position might be less comfortable for someone engaged in the heroic (or quixotic) task of formulating a reductive view of personal identity. 22 For discussion see, for instance, Parfit (1976), Lewis (1976, 1983), Sider (2001), Moss (2012). 23 See Parfit (1984: 256) and Unger (1990: 264) for critical discussion. Whether a group agent could play (parts of) the person-role is itself a topic of great controversy. For discussion see, for instance, List and Pettit (2011), Briggs (2012), List (2016), Tuomela and Mäkela (2016), Backes (2021) and Lovett and Riedener (2021). 24 For instance, the intention or promise to become the first woman president of the United States of America cannot be fulfilled by two fission-generated successors ( Baker 2000: 129). But fulfilment by only one successor will not always be enough for fulfilment of the initial person’s promise or intention—just think of such promises as the promise not to cheat, not to lie, or not to kill. 25 For discussion see, for instance, Baker (2000: ch. 5.5), McMahan (2002: ch. 1.2), and Olson (2007: ch.7) 26 If what is normal is contingent, the Essentiality of humanity would also look doubtful on the narrow view psychological view of personal identity. 27 For more on the essentiality of origins, see Chapter 11 in this volume (“Origin Essentialism”). 28 Baker (2000: ch.7–8) uses the distinction between derivative and non-derivative properties to respond to these sorts of objections to constitution views in general and objections to a constitution view of human persons in particular. See also Liebesmann and Magidor (2017). 29 See Chapter 28 in this volume (“Ethical Value”) for more on the relation between essence and ethics. 30 Though see Singer (1980: 101) for the opposite view. 31 For critical examination, see McMahan (2005).

344

Persons

References Backes, M. (2021) “Can Groups Be Genuine Believers? The Argument from Interpretationism,” Synthese 199 (3-4): 10311–10329. Baker, L. R. (2000) Persons and Bodies: A Constitution View, New York: Cambridge University Press. Baxter, D. (1988) “Identity in the Loose and Popular Sense,” Mind 97 (388): 575–582. Briggs, R. (2012) “The Normative Standing of Group Agents,” Episteme 9 (3): 283–291. Butler, J. (1736) [1842] The Analogy of Religion, Natural and Revealed, to the Constitution and Course of Nature, London: James, John, and Paul Knapton. Reprinted New York: Robert Carter. Carter, S. (2019) “Higher-order Ignorance Inside the Margins,” Philosophical Studies 176 (1): 1789–1806. Castañeda, H.-N. (1966) “‘He’: A Study in the Logic of Self-Consciousness,” Ratio 8: 130–157. Cohen, C. (1986) “The Case for the Use of Animals in Biomedical Research,” New England Journal of Medicine 315: 865–870. Dasgupta, S. (2018) “Essentialism and the Nonidentity Problem,” Philosophy and Phenomenological Research 96 (3): 540–570. Dennett, D. (1976) “Conditions of Personhood,” in A. Oksenberg-Rorty (ed.) The Identity of Persons, Berkeley, CA: University of California Press, 175–196. Dorr, C. and Hawthorne, J. with Yli-Vakkuri, J. (2021) The Bounds of Possibility, Oxford: Oxford University Press. Finnis, J. (1995) “The Fragile Case for Euthanasia: A Reply to John Harris,” in J. Keown (ed.), Euthanasia Examined, Cambridge: Cambridge University Press, 23–35. Frankfurt, H. (1971) “Freedom of the Will and the Concept of a Person,” Journal of Philosophy 68 (1): 5–20. Harman, E. (2003) “The Potentiality Problem,” Philosophical Studies, 114 (1–2): 173–198. Haslanger, S. (2003) “Persistence Through Time,” in Loux, M. and Zimmerman, D. (eds.), The Oxford Handbook of Metaphysics, New York: Oxford University Press, 315–354. Jaworska, A. and Tannenbaum, J. (2014) “Person-Rearing Relationships as a Key to Higher Moral Status,” Ethics, 124: 242–271. Kant, I. (1785) The Groundwork of the Metaphysics of Morals, in P. Guyer and A. Wood (eds.), The Cambridge Edition of the Works of Immanuel Kant, Cambridge: Cambridge University Press, 1996, 37–108. Korsgaard, C. (1996) The Sources of Normativity, New York: Cambridge University Press. Lewis, D. (1969) Convention: A Philosophical Study, Oxford: Blackwell. Lewis, D. (1974) “Radical Interpretation,” Synthese 23: 331–344. Lewis, D. (1976) “Survival and Identity,”in A. O. Rorty (ed.), The Identities of Persons, Berkeley: University of California Press, 17–40. Lewis, D. (1975) “Languages and Language,” Minnesota Studies in the Philosophy of Science, Minneapolis: University of Minnesota Press, 3–35. Lewis, D. (1979) “Attitudes de dicto and de se,” The Philosophical Review 88 (4): 513–543. Lewis, D. (1983) “Postscript to Survival and Identity,” in Philosophical Papers, Volume 1, New York: Oxford University Press, 73–77. Lewis, D. (1986) On the Plurality of Worlds, Oxford: Blackwell. Lewis, D. (1993) “Many But Almost One,” in Lewis, D. (ed.), (1999): Papers in Metaphysics and Epistemology, Cambridge, UK: Cambridge University Press. Liebesman, D. (2015) “We Do Not Count by Identity,” Australasian Journal of Philosophy 93 (1): 21–42. Liebesmann, D. and Magidor O. (2017) “Copredication and Property Inheritance,” Philosophical Issues 27 (1): 131–166. List, C. and Pettit, P. (2011) Group Agency: The Possibility, Design, and Status of Corporate Agents, New York: Oxford University Press. Edited by Philip Pettit. List, C. (2016) “What is it Like to be a Group Agent?” Noûs: 295–319. Locke, J. (1689/1694) An Essay Concerning Human Understanding, in Peter H. Nidditch (ed.), Oxford: Oxford University Press, 1975. Loets, A. (2021) “Choice Points for a Theory of Normality,” Mind 131 (521): 159–191. Lovett, A. and Riedener, S. (2021) “Group Agents and Moral Status: What Can We Owe to Organizations?” Canadian Journal of Philosophy 51 (3): 221–238.

345

Annina Loets McMahan, J. (2002) The Ethics of Killing: Problems at the Margins of Life, New York: Oxford University Press. McMahan, J. (2005) ‘“Our Fellow Creatures”,’ The Journal of Ethics 9 (3–4): 353–380. Moss, S. (2012) “Four-Dimensionalist Theories of Persistence,” Australasian Journal of Philosophy 90 (4): 671–686. Nagel, T. (1986) The View from Nowhere, New York: Oxford University Press. Nozick, R. (1974) Anarchy, State, and Utopia, New York: Basic Books. Noonan, H. (1989) Personal Identity, London: Routledge. Olson, E. (1997) The Human Animal, New York: Oxford University Press. Olson, E. (2007) What Are We? A Study in Personal Ontology, New York, Oxford University Press. Parfit, D. (1976) “Lewis, Perry, and What Matters,” in A. O. Rorty (ed.), The Identities of Persons. The location is Berkeley, CA: University of California Press (1976). Parfit, D. (1984) Reasons and Persons, Oxford, UK.: Oxford University Press. Parfit, D. (2007) “Persons, Bodies, and Human Beings,” in T. Sider and D. Zimmerman (eds.), Contemporary Debates in Metaphysics, Malden, MA: Blackwell., 177–208. Parfit, D. (2012) “We Are Not Human Beings,” Philosophy 87 (1): 5–28. Parfit, D. (1971) “Personal Identity,” The Philosophical Review 80: 3–27. Perry, J. (1972) “Can the Self divide?” Journal of Philosophy 69: 463–488. Perry, J. (1979) “The Problem of the Essential Indexical,” Noûs 13 (1): 3–21. Reid, T. (1785) Essays on the Intellectual Powers of Man, abridged and reprinted, in J. Walker (ed.), Cambridge, MA: John Bartlettt, 1850; second edition, 1851. Robinson, D. (1985) “Can Amoebae Divide without Multiplying?” Australasian Journal of Philosophy 63: 299–319. Scanlon, T. (1998) What We Owe to Each Other, Cambridge, MA: Harvard University Press. Shoemaker, S. (1970) “Persons and Their Pasts,” American Philosophical Quarterly 7: 269–285. Shumener, E. (2017) “The Metaphysics of Identity: Is Identity Fundamental?” Philosophy Compass 12 (1): 1–13. Shumener, E. (2020) “Explaining Identity and Distinctness,” Philosophical Studies 177 (7): 2073–2096. Sider, T. (2001) Four Dimensionalism. An Ontology of Persistence and Time, Oxford, UK: Oxford University Press. Singer, P. (1980) Practical Ethics, New York: Cambridge University Press. Snowden, P. F. (1990) “Persons, Animals, and Ourselves,” in C. Gill (ed.), The Person and the Human Mind, Oxford: Clarendon Press, 83–107. Thomson, J. J. (2007) “People and their Bodies,” in T. Sider and D. Zimmerman (eds.), Contemporary Debates in Metaphysics, Malden, MA: Blackwell, 156–176. Tuomela, R. and Mäkelä, P. (2016) “Group Agents and Their Responsibility,” The Journal of Ethics 20 (1–3): 299–316. Unger, P. (1990) Identity, Consciousness, and Value, New York: Oxford University Press. Wiggins, D. (1976) “Locke, Butler and the Stream of Consciousness: and Men as a Natural Kind,” Philosophy 51 (196): 131–158. Williams, B. (1970) “The Self and the Future,” The Philosophical Review 79: 161–180. Williamson, T. (2013) Modal Logic as Metaphysics, Oxford: Oxford University Press.

346

23 PSYCHIATRIC KINDS Danielle Brown

23.1

Introduction

The development of the psychological professions is fraught and convoluted. Historical investigation of these fields reveals the degree to which the conceptual territory describing our internal experiences has changed over time, as well as the entities and mechanisms to which we appeal when things go awry. In the modern day, the diagnoses in the Diagnostic and Statistical Manual of Mental Disorders (DSM) have undergone frequent shifts, encompassing greater or fewer symptoms, as well as changes in the hypothesized mechanisms thought to underlie them. The diagnoses we use today do not appear as eternal, unchanging categories so much as the products of a complex historical interplay of scientific advancement, therapeutic aims, and progressive changes of social norms and values. To grapple with this history requires an examination of whether psychiatric kinds possess essences, and if so, determine of what these essences consist. By “essence”, I am not referring to individual essences sought by those interested in questions of the persistence of individual entities over time or alteration ,(see: Marabello (Chapter 9)) but rather, the essences of kinds. A kind essence is what we are looking for when we ask in virtue of what every member or instance of a kind belongs to that kind. In the case of a psychiatric kind (e.g. schizophrenia), what is it that makes certain patterns of cognition or behavior fall under that category? Whatever conditions we point to in order to answer this question could be said to constitute the essence of the kind. The question then turns to how we can discover these essences. Where we look for essences depends on the entities we are interested in, but for the entities that science is concerned with, we are typically looking for natural essences—those which can be discovered in the natural world via empirical investigation, in contrast to the essences of human-made artifacts (see: Passinsky (Chapter 17)) or those of the social world. The central question of this chapter is whether psychiatric kinds possess these natural kind essences (see: Tahko (Chapter 10)). “Natural kind” is a philosophical term referring to groupings of particulars thought to reflect the structure of the natural world, as opposed to assembled, conventional (see: LivingstoneBanks (Chapter 30)), or socially constructed groupings (see: Griffith (Chapter 31)) which function as stable targets of prediction and explanation. What sorts of properties could be considered essential for psychiatric kinds has been a matter of debate, as there is

DOI: 10.4324/9781003008750-27

347

Danielle Brown

considerable variance in symptom presentation for any mental disorder and the boundaries between different diagnoses as well as the threshold between disorder and normalcy are often blurry (Cooper 2013). A place to start in terms of locating the essences of psychiatric kinds is the Boydian homeostatic property cluster (HPC) account (Boyd 1999a, Brigandt (Chapter 18)), or, as it has been recently dubbed, the mechanistic property cluster (MPC) account (Kendler et al. 2011). According to this view, natural kinds are co-occurring clusters of properties united by a causal mechanism and are often discussed without appealing to any notion of essence at all. Nevertheless, these sorts of views can be understood as employing a conception of essence—namely, causal essences, which require only that a given cluster of properties is reliably produced or sustained via a certain causal mechanism. However, considering the difficulty of discovering the causal mechanisms underlying psychopathology, the HPC/MPC model creates the potential for conflict if the set of kinds included in the DSM—identified via clinical observation and characterized in folk-psychological terms—are not precisely undergirded by the causal mechanisms thought to be responsible for them. This has led to debate about the practicality of etiologically-based approaches for psychiatric classification, or whether the current set of psychiatric kinds ought to be revised in order to better conform to these causal structures (Murphy 2006). These discussions raise questions about what it is that we want out of a psychiatric ontology. I discuss possibilities for revised ontologies before moving on to alternatives to the HPC/MPC view that allow kind concepts which fulfill our inductive and explanatory aims without reduction of each kind to a single causal mechanism (Khalidi 2013; Borsboom & Cramer 2018). A challenge to the idea that psychiatric kinds possess essences and that these essences are natural comes from Hacking (1999), who contends that psychiatric kinds lack the stability necessary to possess essences and that the social mechanisms which produce this instability compromise their naturalness as well. He argues that kinds in the social sciences—psychiatry, sociology, economics, etc.—are interactive in that the members of a given kind interact with their classification, confirming or subverting generalizations, and thus altering the kind itself. I discuss several responses to Hacking with regard to whether interactivity is unique to human kinds and whether it presents an insurmountable challenge to the idea of psychiatric kinds as possessing natural essences (Cooper 2004; Khalidi 2013; Laimann 2020). Finally, I examine the view that psychiatric kinds are, to some degree, socially constructed, which is thought to complicate, if not outright disqualify them from natural kind status (see: Griffith (Chapter 31)). Traditionally, natural kinds are conceived of as mind-independent in the sense that their nature does not depend on human values. Looking to the history of psychiatry shows us examples of diagnoses that are now recognized as the products of misguided values producing unjustified behavioral expectations for persons or groups. Such diagnoses as “hysteria” and “drapetomania”1 we think have been rightfully relegated to the realm of historical curiosity following changes to the social attitudes that produced them, but concern is not unwarranted that social and moral values continue to shape many psychiatric kinds that persist today. I explain the motivation behind such concerns in the case of the Cluster B personality disorders, which stand as a prominent example of where psychiatry intertwines with our social and moral values. I end with an examination of Zachar (2014), whose conception of “practical kinds” treats kinds in psychiatry as an intersection of epistemic, practical, and social interests.

23.2

The Mechanistic Property Cluster (MPC) View and Its Problems

One plausible candidate view for capturing the essences of kinds in the “special sciences” is property cluster views, the most notable example being Boyd’s “homeostatic property

348

Psychiatric Kinds

clusters” (HPC kinds) (Boyd 1999a). The HPC view has been applied to kinds in biology, where it has proved advantageous for capturing biological terms like “species”, “gene”, andhigher order biological taxa (Boyd 1999b; Wilson et al. 2007). On the HPC view, kinds are defined by “a cluster of often co-occurring properties and by the (“homeostatic”) mechanisms that bring about their co-occurrence” (Boyd, 1999a: 17). By “homeostatic” mechanism, what is meant is any kind of similarity generating mechanism that causally brings about instances of the kind. For elaboration on this view as applied in the biologial sciences and science more generally, see: Brigandt (Chapter 18) and Tahko (Chapter 10), respectively. The suggestion that a similar property cluster view could be applied to psychiatric kinds comes from Kendler, Craver and Zachar (2011), whose view differs from the Boydian HPC picture mostly in its application to psychiatry rather than biology and a greater emphasis on mechanistic explanations, hence the revised label of “mechanistic property clusters” (MPC kinds).2 The authors describe their view as “encourage[ing] the thought that there are robust explanatory structures to be discovered underlying most psychiatric disorders. However, it cautions us to expect that they will be messy and that it will take hard work and some degree of idealization and abstraction to bring them into focus” (Kendler, Craver, & Zachar 2011: 1146). The MPC view of kinds is not without its challenges. One pressing problem for the MPC view as it pertains to psychiatric kinds concerns the potential for mismatches between so-called “diagnostic” kinds and their causal mechanisms (Tabb 2018). Given that, on the MPC view, the validity of a kind is predicated on the identification of some unique causal mechanism, the search for such mechanisms in psychiatry is of paramount importance. The search for causal mechanisms for mental disorders at the genetic, neurological, and developmental levels remains ongoing, and while progress has been made, the fact remains that most of the diagnoses listed in the DSM were formulated prior to such discoveries on the basis of clinical observation rather than any assumptions about underlying etiological mechanisms. In light of this, there is little reason to expect that the causal explanations uncovered will vindicate the set of kinds currently used in psychiatry. Even for psychiatric disorders that can be reliably addressed through medication, the mechanisms behind these interventions often remain unclear. However plausible it may be to believe that similar clusters of symptoms arise from a common cause, this is not guaranteed and may necessitate revisions to the DSM classification system and our broader understanding of psychiatric kinds. Murphy (2006) uses the example “delusion” of a psychiatric kind in which heterogeneous causes underlie a set of superficially similar symptoms. The DSM-5 characterizes delusions as “fixed beliefs that are not amenable to change in light of conflicting evidence” (APA 2013). However, it is not just any fixed belief that can be described as such. Religious beliefs are often resistant to change but are not commonly considered delusional. This discrepancy points to a finer distinction between ordinary and delusional beliefs, specifically that a delusional belief is one that is not acquired via rationally or culturally-prescribed avenues of belief formation. The psychiatric concept of delusion, then, is grounded in folk-epistemological concepts of belief acquisition. Whatever the content of a particular belief, the feature marking its status as delusional consists in a deviation from these folk epistemological models of belief acquisition, and thus, the vindication of “delusion” as a kind requires the identification of a cause for a breakdown in belief formation. To adopt this characterization as a guiding principle for neuroscientific investigation into the nature of delusions would be a mistake, however. Neuroscientific evidence points to multiple heterogeneous causal factors involved in delusions, many of which appear specific to the content of particular delusions. It is possible that the similarities observed between delusional beliefs in the clinical setting may prove to be superficial

349

Danielle Brown

at the neurological level, leaving us with little justification for considering delusions, as a whole, to be of a kind. On this basis, Murphy (2006) recommends an eliminativist stance with regard to the current set of psychiatric kinds and suggests we turn to the available neuroscientific evidence to produce a set of kinds that serve our explanatory aims as the bedrock of a new classificatory structure. Psychologists and philosophers in support of what has been dubbed “the precision turn” in psychiatry share this outlook and have pioneered initiatives intended to investigate the genetic and neurological correlates of psychopathology to produce more empirically-informed taxonomies of mental disorders. The Research Domain Criteria (RDoC) developed by the NIMH aims to identify foundational mental functions informed by cognitive science in order to facilitate a precise mapping of functional domains over increasingly complex units of analysis (e.g. molecules, genes, cells, neural circuits). While not explicitly reductionist in outlook, proponents of the RDoC object to the use of the clinically-derived DSM-5 constructs in the research context and seek to produce taxonomies that more accurately reflect the causal mechanisms implicated in psychopathology (Insel et al. 2010) with the hope that these taxonomies will also function in a clinical capacity. Another project with revisionist potential is the Hierarchical Taxonomy of Psychopathology (HiTOP), developed by an independent consortium of nosologists within the psychological professions. Where the RDoC attempts to trace the causal factors of psychopathology from a neuroscientific perspective, the HiTOP aims for a restructuring of the DSM’s organization via the use of hierarchical dimensional constructs that capture significant commonalities between recognized disorders (Kotov, et al. 2017). This strategy, which incorporates elements of etiology such as shared risk factors, rates of comorbidity, and treatment responses, largely preserves the set of existing psychiatric kinds while pursuing revisions within the broader organizational structure. If either of these initiatives proves successful, we may see future psychiatric taxonomies consisting of better candidates for kinds on the MPC picture. In their discussion of this issue, Kendler, Craver, and Zachar (2011) admit of the potential for “multiple realizability” of MPC kinds in that “the same cluster of symptoms might arise from different etiological, underlying, or sustaining mechanisms in different cases” (Kendler, Craver, & Zachar, 2011: 1148), however, as the considerations presented in this section demonstrate, it remains unclear how to go about individuating such clusters. For psychopathology, in which the underlying mechanisms for symptom clusters are often complex, multivariate, and probabilistic, there is often no strict one-to-one causal relation between a given mechanism and its effects. Craver (2006) questions how we can gauge similarity between mechanisms that result in the same set of properties, as well as what to do when a single causal mechanism produces different sets of properties. To settle these disputes, we must go up a level and ask whether there are kinds of mechanisms, leading to a regress. This is the point where one might appeal to differing epistemic interests, as the set of kinds that best serve empirical purposes for neuroscientists may not be the same set that best serves the purposes of clinicians. However, many would be reluctant to admit such a degree of interestrelativity when it comes to kinds and instead propose similar causally driven accounts designed to capture the advantages of the MPC view while avoiding the drawbacks.

23.3

Alternatives to the MPC View

In response to objections leveled at the HPC/MPC account of kinds, there have been alternatives proposed which aim to address these problems. I focus on two types of

350

Psychiatric Kinds

alternatives: (1) more permissive property cluster accounts which share Boyd’s central insight that kinds are constituted via some combination of clustered properties and/or causal relationships, but do away with the need for a singular mechanism (Khalidi 2013; Borsboom & Cramer; 2018) and (2) pluralist accounts that accept the interest-relativity of kinds and question the demand for natural kinds in psychiatry in the first place (Haslam 2002; Tabb 2018). On Khalidi’s causal network account, kinds are “nodes” within causal networks, where the network as a whole represents the total causal structure of the world. What counts as a kind within a network is any “set of properties whose co-instantiation causes the instantiation of other properties” (Khalidi 2013: 80), and this can come in degrees. The “naturalness” of a kind is measured by the number of causal relations it is involved in, the amount of inductive generalizations it permits, and how strongly the properties within a cluster internally hold together. This serves to demarcate kinds from arbitrary or conventional groupings while allowing a high degree of flexibility. The most important distinction between a causal network account and the MPC account is that kinds are individuated on the basis of the property cluster itself and the causal role it plays within the network, rather than the causal mechanism that brings it about. We can consider Murphy’s example through this lens and conclude that even if delusions are causally heterogeneous, the cluster of properties associated with the clinical concept of delusion may permit further predictions such as prognosis or treatment response, regardless of the causal history of the particular delusion, thereby justifying its status as a kind without the need to appeal to diverging epistemic interests. Borsboom and Cramer (2018) put forward a causal network theory that places the emphasis on the network itself rather than the “nodes” within it. Formulated specifically as a non-reductionist model designed to accommodate the unique challenges presented by psychiatric kinds, the authors propose a network structure in which mental disorders are not posited as a latent causal variable that underlies a cluster of symptoms; rather, mental disorders simply are networks of causally-related symptoms in the sense of identity. This distinction serves to differentiate this network approach from alternatives like Khalidi’s, where the ability to single out properties within a cluster as more or less causally basic risks reintroducing a notion of hierarchy. On Khalidi’s account, the naturalness of a kind depends in part on the internal structure of the kind, which properties can be considered causally basic and the strength of the relations between them and derivative properties. In contrast, Borsboom and Cramer’s network account does not admit to this distinction, nor does it posit any notion of hierarchy. In accordance with the complex nature of mental disorders, in which biological, psychological, historical, environmental, and cultural factors interact in myriad ways, the authors argue that mental disorders “likely involve feedback loops that cross all the traditional divides between levels of explanation, none of which can claim the status of ’basis’ for the others” (Borsboom & Cramer 2018: 41). By denying notions of causal priority as it applies to mental disorders, the authors gesture toward a more radical shift in our understanding of the basic units of analysis of psychopathology. Instead of focusing on disorders as latent variables underlying symptom networks, we should look to the symptoms themselves which may feature in the networks of multiple disorders, leaving us with a patchwork classification of fuzzy disorder kinds. The question motivating each of these causal network accounts is the strength of causal relations and stability required to consider some set of properties a natural kind. This question presumes the necessity of identifying natural kinds as the proper objects of psychiatric inquiry, but this assumption, too, has been subject to question. That the categories and classifications that function as the targets of psychiatric interest must meet a threshold of

351

Danielle Brown

naturalness to be worth investigating has been challenged by Haslam (2002), who argues for a pluralist view in which there is no single standard for naturalness. Presented in order of strictness, Haslam identifies “practical kinds” which he characterizes as dimensional or continuous constructs (e.g. personality dimensions), “fuzzy kinds”, which possess discontinuities but not sharp ones (e.g. Borderline Personality Disorder), “discrete kinds”, which exhibit sharp discontinuities from other concepts, but do not possess essential properties (e.g. Depression), and “natural kinds”, which do possess what may be thought of as essential properties (e.g. William’s Syndrome). Haslam’s account admits of degrees and demonstrates how concepts that fail to meet this or that standard of naturalness can still be of scientific relevance. However, while Haslam’s pluralism accurately describes the landscape of psychiatry in terms of the range of kind concepts in use, this account does not broach the more pressing issue that some kinds may be interest-dependent and so does not directly address the concerns motivating revisionist views. A version of psychiatric kinds pluralism that accounts for the interest-dependence of kinds and proposes a means of navigating the challenge of interdisciplinary collaboration comes from Tabb (2018). Drawing from Ankeny and Leonelli’s (2015) work on “scientific repertoires”, Tabb argues for a pluralism that takes as its foundation the social-epistemological practices of the psychological sciences in order to foster a “coordinated pluralism”3 that encourages collaboration between different disciplines and formulating means of translation across different psychiatric ontologies. As this pertains to the status of clinically-derived versus laboratory-derived kinds and the status of the set of kinds included in the DSM, Tabb argues that the potential for retention or revision of these categories will depend on their utility as “epistemic hubs” within nonreductionist network theories previously described, and that these categories “are best conceived of as “robust patterns” revealed by the juxtaposition of diverse stratifications of the patient population produced by different theoretical perspectives” (Tabb 2018: 18). Which set of kinds is used within a given circumstance, then, is not a reduction of one discipline’s ontology to another’s, but a judgment about which ontology provides the most utility within a given domain according to collectively recognized disciplinary practices, norms, and aims.

23.4

Hacking on Interactive Kinds

When talking about the “stability” of psychiatric kinds in the previous section, the quality at issue was how tightly a kind’s properties hold together in terms of the regularity of their cooccurrence and their internal causal relations. There is, however, another sense of “stability” to which psychiatric kinds pose an issue. This sense of stability refers to the persistence of kinds over time and the mechanisms by which they undergo change. Ian Hacking (1999) contends that there is a qualitative distinction between kinds in the so-called “natural” sciences and those of the “social” or “human” sciences in that human kinds exhibit “interactivity” between the classification and the entities that are classified. By this, he means the fact that members of a given kind may respond to being classified as members of that kind, either to confirm or subvert that classification, thereby altering the characteristics of the kind itself. For example, the kind term “woman refugee” is interactive in the sense that a person classified as a “woman refugee” interacts with her classification in the form of conforming to or subverting the beliefs, allowances, and expectations others have of her on the basis of that classification. If enough of the characteristics of the kind “woman refugee” are subverted, then the classification itself may change to reflect this, producing a “looping effect”.4 Hacking contrasts this with the “indifferent” kinds of the natural sciences which, although not static, do not change in response to the act of classification.

352

Psychiatric Kinds

On an epistemological level, interactivity would seem to compromise a kind’s utility as a stable target of knowledge, and on a metaphysical level, lends itself to anti-essentialist thinking if the properties of a kind that are subject to change are essential ones. As psychiatric diagnoses are applied to human beings, many psychiatric kinds are interactive in this way. Individuals diagnosed with mental disorders can and do interact with their classification in both explicit ways (e.g. making use of medical and social resources designed to treat, manage, or compensate for their condition), and tacit ways (e.g. altering one’s behavior to mask disorder symptoms) that have the potential to change the properties of the kind. We may ask, however, whether Hacking’s distinction between interactive and indifferent kinds captures what he intends and whether he is justified in the conclusion that there is a unique epistemic or metaphysical difficulty with regard to the stability of kinds in the social sciences that does not apply to kinds in the natural sciences. These objections can be articulated by asking (1) whether interactivity is unique to human kinds, and (2) whether the causal mechanism of interactivity poses a more substantial obstacle to our epistemic practices than other sources of kind instability and (3) whether interactive kinds are incompatible with essentialism about kinds. Some have offered negative responses to the first question (Khalidi 2013), arguing that many kinds in the natural sciences can be understood as interactive, and thus Hacking’s distinction fails to identify any unique difficulty with human kinds. Cooper (2004) answers the first question in the affirmative, but denies that interactivity produces a unique obstacle to our epistemic practices. Laimann (2020) takes a different approach, interrogating the mechanisms of interaction to determine what quality Hacking seeks to capture, and concluding that the particular mechanism of interactivity that Hacking describes in his discussion of human kinds does serve to demarcate a real difference between human and non-human kinds and that this mechanism by which interactivity operates is less predictable than other sources of change to a kind. Khalidi interprets Hacking’s notion of interactivity in a broad sense, arguing that interactions between classifications and the entities that fall under that classification are not restricted to the human realm. Taking seriously Hacking’s insistence that awareness of the classification on the part of the classified is unnecessary for interactivity, Khalidi understands the mechanism behind interactivity to be causal mind-dependence in which “deliberate human intervention plays a significant causal role in determining the properties associated with a certain kind of individual, sometimes culminating in the creation of an entirely new kind of individual” (Khalidi 2013: 149–50). On this characterization of interactivity, kinds such as pedigree dogs may be candidates for interactive kinds, as dog-breeders’ knowledge of the various characteristics associated with the classifications of different breeds affect their decisions about which dogs to breed, thus producing changes in the dogs and, over generations, to the breed itself. As Hacking does not specify how proximal a causal role classifications must play in the modification of their members, Khalidi proposes that certain chemical compounds may be construed as interactive due to changes arising from human intervention. If Khalidi is correct that the mechanism of interactivity can be understood as causal mind-dependence and that non-human kinds can exhibit interactivity, then Hacking’s distinction fails to identify any feature of human kinds that renders them uniquely epistemically challenging. That Hacking explicitly rejects the idea of non-human kinds exhibiting interactivity has led others to conclude that the mechanism of interactivity must be a narrower concept than causal-mind dependence, leading to speculation over what sort of mechanism this is and whether it creates unique epistemic issues for human kinds. Cooper (2004) describes this mechanism as “idea-dependence” that operates via cultural feedback. Cooper observes that

353

Danielle Brown

many human kinds carry a normative valence in the sense that “being classified a certain way [carries] institutionalized benefits or costs” (Cooper 2004: 78), that motivates individuals to alter their behavior and, over time, change the classification itself.5 That idea-dependence operates on the basis of normative cultural influence restricts interactivity to human kinds, however Cooper denies that interactivity poses any special challenge to our epistemic practices. Unless there is special metaphysical significance associated with idea-dependence, it is not sufficient to demonstrate that certain kinds are affected by a causal mechanism to which other kinds are invulnerable to conclude that those kinds cannot function as adequate epistemic targets. As we saw in Khalidi’s examples, there are non-human kinds that are similarly destabilized by feedback mechanisms. Additionally, there are other causal mechanisms which can render a kind a “moving target” that are unrelated to feedback (e.g. rapid mutations in bacteria in response to selection mechanisms). This leaves the rate of change as the factor distinguishing human kinds from non-human kinds in terms of their epistemic utility, and even if it could be shown that idea-dependence prompts more rapid changes to kinds than other mechanisms of instability, the distinction would remain a matter of degree. The role of normativity in interactivity bears further examination. Laimann (2020) argues that normativity is a problem for the stability of kinds not only in the sense that it prompts rapid change to a kind over time, but also that these changes are less predictable than other sorts and are largely resistant to theoretical understanding. Laimann presents interactive kinds as possessing a unique ontological structure consisting of a “base” kind, which consists of the properties of the kind, and a corresponding “status” kind, which consists of the social costs and benefits associated with kind membership. To be treated in a certain way on the basis of kind membership prompts changes in an individual’s self-concepts and alters their behavior to either confirm or undermine the classification in order to avoid negative social consequences or bring about positive ones (see: Stoljar (Chapter 26), Tannenbaum and Glezakos (Chapter 18), and Griffith (Chapter 31)). The trouble this presents with regard to our epistemic practices is not that these changes are too substantial or frequent for the kind to function as a stable target of knowledge, but that both the destabilizing and stabilizing effects of this process may obscure our understanding of other causal processes affecting that kind. In her discussion of the sex/gender distinction, Laimann argues that we can view these concepts as a base/status kind pair in which sex acts as a base kind while gender is a status kind consisting of the norms and expectations associated with a certain sex. This effect of interactivity can become a confounding factor when researching sex differences. (See: Rosario (Chapter 25)). The social pressures for people of a certain sex to perform their gender according to their society’s standards has the potential to mislead in a scientific context. People making an effort to conform to their gender within a research context may exaggerate perceptions of sexual differences, and likewise, people making an effort to subvert their gender may minimize what differences do exist, thereby producing biased conceptualizations of the base kind that obscure other relevant causal processes. In this way, stabilizing and destabilizing feedback alike pose an obstacle to our epistemic practices. As for the metaphysical question about whether interactivity is compatible with an essentialist view of kinds, there are two considerations to address. The first concerns whether the instability in properties brought about by interactivity threatens an essentialist conception of kinds, and straightforwardly, if the properties subject to change are non-essential, then this poses no problem at all as long as the essentialist account is paired with an account of how new kinds come into existence. For example, if interactivity causes the essential property to change, then what we end up with would be the elimination of the first kind and the emergence of a second, which, while epistemically challenging, would not necessarily be metaphysically problematic.

354

Psychiatric Kinds

The second consideration has to do with normativity as part of the mechanism of interactivity and whether this is compatible with the idea of psychiatric kinds possessing natural essences. If a kind can come into or out of existence or change as a result of what people believe about it, this raises questions about whether that kind is natural. In particular, interactivity challenges the notion that natural kinds must be mind-independent. However, it seems clear that the sort of mind-dependence operant in interactivity is merely a specific type of causation in which social factors such as culture and institutions influence the behavior of the individuals subject to them in ways that are normatively charged. Though Laimann (2020) correctly points out that causal mechanisms in the social sciences are less predictable than those in the natural sciences, there is no reason to conclude that these causal mechanisms could not ground a kind in the same way as any other.6 This is not to say that there are no reasonable concerns about the naturalness of psychiatric kinds, only that interactivity is not the source of these worries.

23.5

Psychiatric Kinds as Socially Constructed

The idea of psychiatric kinds as value-laden is hardly a novel insight. In the most basic sense, we can view psychiatric kinds as value-laden in the same way as all medical kinds are valueladen—namely, that they are negative evaluations relative to a state of physical or mental health. I expect many will agree that it is undesirable to be ill. Matters become more complicated when we interrogate what it means to be “ill” and how social and cultural values can come to bear on our understanding of illness and disorder. Unlike somatic disorder, where we at least have a widely agreed upon evaluative metric in the form of premature death or detriments to quality of life to judge whether a condition is harmful to the bearer and thus merits negative evaluation as well as scientific investigation to discover effective treatments or preventative measures, the standards employed to determine what constitutes a mental disorder are considerably less clear and often permeated with values.7 This value-ladenness has led to discussion about whether psychiatric disorders may be, to some degree, socially constructed. To say that something is a social construct in the epistemically problematic sense is not, as previously discussed, to say that something is merely causally dependent on social factors (e.g. values, institutions, collective beliefs) for its existence. That psychiatric disorders arise and take the forms they do may be the result of aspects of the social world affecting people’s lives. The type of social construction that challenges the idea of psychiatric kinds as natural is constitutive social construction, in which social factors play a constitutive role in defining what a kind is (see: Griffith (Chapter 31)). In the case of psychiatric kinds, the social factors at issue are the social values and norms in virtue of which we categorize certain behaviors as pathological. If it is the case that psychiatric kinds are constitutively socially constructed, this raises questions regarding their epistemic efficacy, as there is no reason to think that our non-epistemic norms and values function to pick out relevant similarities between cases of non-normative experiences or behavior that could facilitate scientific investigation or the discovery of natural kinds. Our non-epistemic norms and values are also historically contingent and subject to frequent reexamination, lending weight to worries that the contemporary norms and values that influence psychiatric diagnoses may be perpetuating harm or injustice via stigma, social disapprobation, or the coercive power of medical authority. The history of psychiatry furnishes examples of former “disorders” which, following a shift in social or cultural values, were determined to be either (1) non-pathological variations in human experience, or (2) loose groupings of behaviors deemed undesirable that did not

355

Danielle Brown

adequately function as a target of the biomedical sciences. The removal of homosexuality from the DSM-II in 1973 stands as an example of a former disorder that was subsequently recognized as a non-pathological variation of human romantic and sexual behavior. This shift in values did not, however, affect its status as a kind within the realm of the social sciences; it merely ceased to be pathologized. For the latter category, we can take the pervasive but historically slippery diagnosis of hysteria which, while common throughout the 19th and early-20th centuries, eventually fell out of clinical favor in the latter half of the 20th century due to a number of factors. For one, the expansive and variable mixture of dissociative, conversion, and somatoform symptoms associated with hysteria were determined to be more productively conceived as separate targets of inquiry (Skull 2009). The shift away from Freudian psychoanalytic explanations and toward the description-focused neo-Kraepelinian model as the dominant paradigm of psychiatry also played a role. Secondly, it would be remiss not to acknowledge that, up until the end of the First World War,8 hysteria was diagnosed almost exclusively in women and widely perceived as a women’s illness due to sexist beliefs about women and expectations regarding feminine behavior. This fact, combined with the vagueness of the diagnosis, allowed hysteria to be wielded as a tool of gender norm enforcement, identifying women who deviated from sexist standards behavior and subjecting them to medical interventions. The diagnosis of hysteria has since come to be understood as a social construct made up of an assortment of qualitatively and etiologically distinct symptoms, united only by deviance from widely held social norms and a justification for the use of coercive authority by medical institutions (see: Tannenbaum and Glezakos (Chapter 28)). Worries about the social functions of psychiatric classification has motivated examination of the current set of psychiatric kinds. While it may be impossible to eliminate the influence of values within a science that deals directly with human lived experience, it may be feasible to sort out the classifications that pick out some set of correlated symptoms underwritten by stable etiological mechanisms and those which do not (Tsou 2016). Even if the former classifications are causally dependent on social factors in such a way that interactive feedback obscures these mechanisms, they may still serve our epistemic purposes. Tsou lists depression, schizophrenia and bipolar disorder as good candidates for natural kinds insofar as they are some of the most historically consistent and comprehensively studied diagnoses with known biological correlates.9 However, even these cases are not insulated from extra-scientific interests and there is reason to doubt that the current diagnostic constructs designed to capture these illnesses successfully pick out the relevant phenomena. Horwitz and Wakefield (2007) describe how the interests of the pharmaceutical industry have contributed to expanding the diagnostic criteria of Major Depressive Disorder in accordance with what broad-spectrum anti-depressants are capable of addressing and thereby failing to adequately differentiate depression from mood disturbances following adverse life events. Cooper (2015/ 2018) makes the case that the DSM and ICD diagnostic criteria are affected by coordination problems that result in “path-dependence” or “lock-in” diagnoses, wherein certain diagnoses or diagnostic criteria may be revised or retained due not only due to advancements in scientific understanding, but also the needs of various interest groups: clinicians, advocacy groups, insurance companies, and patients. What can be gleaned from this is that even the most likely candidates for natural kinds in psychiatry are bound to diagnostic constructs which are subject to competing interests, both epistemic and social. The situation is more dire for other psychiatric disorders, especially those for which stable etiological mechanisms are still largely unknown. One of the strongest examples of psychiatric kinds that appear to be heavily socially dependent both in their construction

356

Psychiatric Kinds

and diagnosis are the personality disorders, specifically the Cluster B personality disorders. Personality disorders in general are characterized by enduring patterns of experience or behavior that markedly differs from the expectations of an individual’s culture (APA 2013: 646). In other words, these disorders are measured in terms of deviation from cultural norms rather than an individual’s baseline level of functioning, as is the case with more “episodic” disorders such as depression. The Cluster B personality disorders, or the “dramatic cluster”, are particularly problematic, as they often involve deviation from not only cultural norms, but also social and moral norms. Antisocial, Borderline, Narcissistic, and Histrionic personality disorders, to various degrees, include morally-loaded diagnostic criteria. For example, the diagnostic criteria for Antisocial Personality Disorder in the DSM-5 includes “a pervasive pattern of disregard for and violation of the rights of others” and “failure to conform to social norms with respect to lawful behaviors” (APA 2013: 659). Though all of these diagnoses also include a number of non-moral criteria, concerns have been raised that these disorders exist primarily to medicalize what would otherwise be considered morally disvalued behavior. Charland (2004/2006) advances a strong version of the argument that the Cluster B personality disorders are not properly medical but moral in nature. His so-called “NeoSzaszian Critique”—named for Thomas Szasz, an influential figure in the anti-psychiatry movement of the 60s and 70s—argues that the Cluster B personality disorders are not pathologized because of the potential for harm they present to the patient, but to others in his or her life. Charland presents two primary arguments for this claim: (1) that the morallyloaded language in the diagnostic criteria of the Cluster B disorders precludes anyone who could be described as a morally praiseworthy individual from being diagnosed with one, and (2) any treatment for a Cluster B personality disorder would have to provoke what amounts to a “moral transformation” in the patient to be gauged successful. As such, he argues, these disorders should not be regarded as medical problems and would be more aptly understood as deficiencies in one’s moral character and should be classified as something closer to the Szaszian notion of “problems in living” and should be addressed via explicitly moral forms of therapy and rehabilitation. Rashed and Bingham (2014) make a similar case that the inclusion of the Cluster B personality disorders conflicts with the guiding principles of the DSM-5, which aims to be value-neutral and specifically includes a clause against diagnoses made solely in virtue of deviance from social norms to guard against the risk of pathologizing conflicts with society. Sadler (2009) takes a similar line and includes Antisocial Personality Disorder among what he calls “vice-laden” diagnoses, that is to say, disorders primarily characterized by certain kinds of criminal misconduct. As regarding their status as natural kinds, personality disorders are where the specter of hysteria looms largest. Lacking a stable empirical basis and heavily stigmatized even within the psychological professions (Sheehan 2016), the combination of limited epistemic utility and overtly moral content in the diagnostic criteria, concern about the legitimacy of personality disorders as natural kinds is not unwarranted. However, not everyone is as pessimistic about the prospect of admitting value-laden kinds into our psychiatric ontology. In accordance with broader trends within the philosophy of science and the recognition of the role of science in serving social aims as well as epistemic ones, it has been argued that attempts to eliminate the intrusion of values in science are counterproductive, and the focus should instead be on ensuring that these values are the right ones. Brigandt (2020) extends this approach to natural kinds, putting forward an account of how kinds may legitimately serve moral and social aims as well as epistemic. With regard to psychiatric kinds, Zachar (2014) introduces the concept of “practical kinds”, an account that aims to recognize psychiatric kinds as simultaneously inductive tools and targets of inquiry, historical concepts, and normative concepts, with the goal of balancing these competing

357

Danielle Brown

interests. To understand psychiatric kinds as a normative concept, however, demands that we be explicit about the normative theories to which we are appealing. Zachar and Potter (2010) provide an account of how this might be accomplished with the Cluster B personality disorders. They argue that the conceptual resources of virtue ethics offer a way to bridge the moral and the medical elements of personality disorders. The development of the DSM-5’s Alternative Model of Personality Disorders, which conceptualizes personality disorders in terms of personality traits which already possess a moral valence may lend itself to this approach and present a viable path forward. The idea of choosing the values we want to revise the current set of psychiatric kinds dovetails with recent efforts to develop a revisionist approach to kinds through the creation and acknowledgment of kinds with constructed, or non-natural essences (see: Rosario (Chapter 25) and Stoljar (Chapter 26)). Contrary to the cases detailed previously, in which erroneous essentialist beliefs about various groups have negatively contributed to the construction of socially damaging kinds (e.g. women and hysteria), it has been pointed out that essentialist ideas may play a positive role in the thoughtful and deliberate development of kinds that serve to ground our epistemic practices as well as advancing our social and moral aims for mental healthcare.

23.6 Conclusion Psychiatric kinds present challenges for essentialist accounts of natural kinds. This has required the exploration of alternative accounts of essence, away from sharply defined categories and toward causal networks and property clusters. It also necessitates an account of kinds that allows for differences of epistemic interests and acknowledgment that the set of kinds that serve our empirical purposes in one domain may not serve for another, dividing the same conceptual territory along different lines. Even more troublesome are the effects of interactive feedback mechanisms, which challenges our epistemic practices by muddying the waters with regard to the causal mechanisms affecting a given kind. Finally, the legacy of mental disorder as a history of social constructs that rise and fall in accordance with changes in our social and moral values raises the question of whether the current set of psychiatric kinds are tools of human interests rather than windows into the structure of the world. Recent viewpoints have emerged that have questioned this picture, offering approaches potentially capable of balancing scientific discovery with human needs and carving a space for psychiatric kinds.

23.7

Related Topics

Chapter 10, Natural Kind Essentialism (Tahko) Chapter 18, Biological Species (Brigandt) Chapter 26, Social Justice (Stoljar) Chapter 28, Ethical Value (Tannenbaum and Glezakos) Chapter 31, Social Construction (Griffith)

Notes 1 A hypothesized mental illness posited in the 19th century to explain the tendency of African slaves to flee captivity. 2 Exactly what this amounts to is unclear in Kendler, Craver, & Zachar (2011), but is explicated in Craver (2006) as a requirement for explanations that “include all of the relevant features of the

358

Psychiatric Kinds

3 4

5

6 7

8 9

mechanism, its component entities and activities, their properties, and their organization” ( Craver 2006: 367). This is to be contrasted with “uncoordinated pluralism”, which Sullivan (2017) argues has characterized the psychiatric professions thus far. This example may appear to suggest that subjective awareness of the classification on the part of the classified agent is a necessary condition for interactivity, but Hacking insists this is not the case. Interactivity may arise even in cases where the agent is unaware that they have been classified in a particular way and are responding to how they are treated by other agents on the basis of the classification. While this account seems close to what Hacking has in mind, he does not provide a comprehensive psychological account of how this process takes place. Tekin (2014) offers a more detailed psychological model of interactivity in terms of how classification influences patient self-concepts in the case of mental disorder. Godman (2013) analyzes a case study of a culture-bound psychiatric disorder (Pervasive Refusal Syndrome) to demonstrate how a specific combination of environmental and cultural conditions may produce a robust and unified set of symptoms. This is not to say that determinations of physical disorder are without complication, especially when it comes to disability. In this domain, discussions arise with regard to what extent quality of life is diminished as a result of the condition itself and what may be more directly attributed to difficulties in meeting social expectations and a lack of both social and infrastructural accommodations on the part of a society constructed for the able-bodied ( Amundson 2000). Specifically, returning male veterans were diagnosed with hysteria for somatoform symptoms relating to their combat experiences. This was later relabeled as “shell-shock”, and even later conceptualized as Post-Traumatic Stress Disorder. Autism was once regarded as an “ideal” candidate for a natural kind within the psychiatric kinds literature for similar reasons, however recent developments have revealed that autism is much more heterogeneous than previously thought, to the point where it is difficult to capture on the HPC/MPC model ( Weiskopf 2017).

References American Psychiatric Association. (2013) Diagnostic and Statistical Manual of Mental Disorders (5th ed.) American Psychiatric Association Publishing: Washington, DC. Amundson, R. (2000) “Against Normal Function,” Studies in History and Philosophy of Biological and Biomedical Sciences, vol. 31, no. (1) (2000), 33–53. Borsboom, D. & Cramer, A. (2018). “Brain Disorders? Not Really … Why Network Structures Block Reductionism in Psychopathology Research,” Behavioral and Brain Sciences, vol. 42, 1–54. Boyd, R. (1999a) ““Kinds as the Workmanship of Men”: Realism, Constructivism, and Natural Kinds,” in J. Nida-Rümelin & W. De Gruyter (eds.) Rationality, Realism, Revision: Proceedings of the 3rd International Congress of the Society for Analytic Philosophy, 52–91. Boyd, R. (1999b) “Kinds, Complexity, and Multiple Realization” Philosophical Studies, vol. 95, pp. 67–98. Brigandt, I. (2020) “How to Philosophically Tackle Kinds Without Talking About Natural Kinds,” Canadian Journal of Philosophy, 1–24. Charland, L. (2004) “Character: Moral Treatment and the Personality Disorders,” in J. Radden (ed.), The Philosophy of Psychiatry: A Companion (64–77). Oxford University Press : New York. Charland, L. (2006) “Moral Nature of the Cluster B Personality Disorders,” Journal of Personality Disorder, vol. 20, no. (2) (2006), 116–125. Cooper, R. (2004) “Why Hacking is Wrong About Human Kinds,” British Journal for the Philosophy of Science, vol. 55, 73–85. Cooper, R. (2013) “Avoiding False Positives: Zones of Rarity, the Threshold Problem, and the DSM Clinical Significance Criterion,” The Canadian Journal of Psychiatry, vol. 58, 606–611. Cooper, R. (2015) “Why is the Diagnostic and Statistical Manual of Mental Disorders so Hard to Revise? Path-dependence and “Lock-In” in Classification,” Studies in History and Philosophy of Biological and Biomedical Sciences, vol. 51 (2015), 1–10.

359

Danielle Brown Cooper, R. (2018) “Understanding the DSM-5: Stasis and Change,” History of Psychiatry, vol. 29, no. (1), 49–65. Craver, C. (2006) “When Mechanistic Models Explain,” Synthese, vol. 153 (2006), 355–376. Godman, M. (2013) “Psychiatric Disorders Qua Natural Kinds: The Case of the “Apathetic Children”,” Biological Theory, vol. 7, no. (2), 144–152. Hacking, I. (1999) The Social Construction of What? Harvard University Press: Cambridge. Haslam, N. (2002) “Kinds of Kinds: A Conceptual Taxonomy of Psychiatric Categories,” Philosophy, Psychiatry & Psychology, vol. 9, 203–217. Horwitz, A. & Wakefield, J. (2007) The Loss of Sadness: How Psychiatry Transformed Normal Sorrow into Depressive Disorder. Oxford University Press: New York. Insel, T., et al. (2010) “Research Domain Criteria (RDoC): Toward a New Classification Framework for Research on Mental Disorders,” American Journal of Psychiatry, vol. 167, 748–751. Kendler, K., et al. (2011) “What Kinds of Things are Psychiatric Disorders?” Psychological Medicine, vol. 41, no. (6), (July 2011), 1143–1150. Khalidi, M. (2013) Natural Categories and Human Kinds: Classification in the Natural and Social Sciences. Cambridge: Cambridge University Press. Kotov, R., et al. (2017) “The Hierarchical Taxonomy of Psychopathology (HiTOP): A Dimensional Alternative to Traditional Nosologies,” Journal of Abnormal Psychology, vol. 126, no. (4), 454–477. Laimann, J. (2020) “Capricious Kinds,” The British Journal for the Philosophy of Science, vol. 71, no. (3), 1043–1068. Murphy, D. (2006) Psychiatry in the Scientific Image. The MIT Press: Cambridge. Rashed, M. & Bingham, R. (2014) “Can Psychiatry Distinguish Social Deviance From Mental Disorder?” Philosophy, Psychiatry, & Psychology, vol. 21, no. (3), (Sept 2014), 243–255. Sadler, J. (2009) “Vice and the Diagnostic Classification of Mental Disorders: A Philosophical Case Conference,” Philosophy, Psychiatry, & Psychology, vol. 15, no. (1) (March 2008), 1–17. Sheehan, L., et. al. (2016) “The Stigma of Personality Disorders,” Current Psychiatry Reports, vol. 18, no. (11) (January 2016). Skull, A. (2009) Hysteria: The Disturbing History. Oxford University Press. Sullivan, J. (2017) “Coordinated Pluralism as a Means to Facilitate Integrative Taxonomies of Cognition,” Philosophical Explorations, vol. 20, no. (2) (Cognitive Ontologies in Philosophy of Mind and Philosophy of Psychiatry), 129–145. Tabb, K. (2018). “Philosophy of Psychiatry After Diagnostic Kinds,” Synthese, vol. 196,2177–2195. Tekin, S. (2014). “The Missing Self in Hacking’s ‘Looping Effects,’” In H. Kincaid & J. Sullivan (eds.), Mental Kinds and Natural Kinds. MIT Press (2014), 227–256. Tsou, J. (2016). “Natural Kinds, Psychiatric Classification, and the History of the DSM,” History of Psychiatry, vol. 27, no. (5), 406–424. Weiskopf, D. (2017). “An Ideal Disorder? Autism as a Psychiatric Kind,” Philosophical Explorations, vol. 20, no. (2), 175–190. Wilson, R., et al. (2007). “When Traditional Kinds Fail: Biological Natural Kinds,” Philosophical Topics, vol. 35, no. (1 & 2), (Spring/Fall, 2007), 189–215. Zachar, P. (2011). “The Clinical Nature of Personality Disorders: Answering the Neo-Szaszian Critique,” Philosophy, Psychiatry, & Psychology, vol. 18, no. (3), (Sept 2011), 191–202. Zachar, P. (2014). A Metaphysics of Psychopathology. Cambridge: The MIT Press. Zachar, P. & Potter, N. (2010). “Personality Disorders: Moral or Medical Kinds—Or Both?,” Philosophy, Psychiatry, & Psychology, vol. 17, no. (2) (June 2010), 101–117.

360

24 RACE Ron Mallon

Racial essentialism is the view that human races exist, and they are constituted by, or grounded in, individuals’ possession of racial essences. Racial essentialism, thus conceived, is a sort of kind or type essentialism in that it explains membership in a type or kind as opposed to being a claim about what is essential to an individual. In this way (among others) it is analogous to doctrines that suggest gender kinds or biological kinds have essences, and it contrasts with questions about individuals, for example, the question of whether an individual person’s race is essential to that individual’s personal identity. What are racial essences? In philosophy, essences are often conceived of modally, as those properties that are necessary features of an object or kind, or nowadays, more narrowly, as those necessary properties that give the nature or definition of the kind (Fine 1994; cf. Correia, this volume). Understood in this latter way, racial essences would simply be the properties possession of which constitute an individual as a member of a kind or that define membership in the kind, perhaps in the way that H20 is said to be the essence of water and also to be the meaning of “water” (Putnam 1975). Some recent work on race seems to similarly conceive of itself as giving an (externalist, a posteriori) definition of “race”, often one that is at odds with various features of folk understandings of race. Thus, these accounts of race would count as essentialist on either a modal or a definitional understanding (e.g., Andreasen 1998; Haslanger 2000, 2003, Mallon 2016).1 However, in much social theory and parts of cognitive psychology, a “racial essence” is understood in a different, narrower way, and it is this narrower view that I use “racial essentialism” to label here. On this narrower use, racial essentialism is the view that races possess or are characterized by an essence where that is a property or set of properties such that, E1. The property or properties are simple, and possession of the property or properties is necessary and sufficient for membership in the kind. E2. The property or properties are intrinsic (nonrelational) and natural (nonsocially contingent) features of individual members of the kind. In its conceiving of properties as “simple”, E1 is ruling out wildly disjunctive properties shared by all and only members of a set of things as counting as candidate essences.2 The properties of concern are conceived of as naturally or logically simple. In understanding them

DOI: 10.4324/9781003008750-28

361

Ron Mallon

as “intrinsic”, E2 reflects an intuition that essences are “inside” the sorts of things of which they are the essence and independent of other things. Nonessentialism about a category is simply the denial that category members share an essence, and here I use “nonessentialism” narrowly to pick out views that deny essences of the sort characterized by E1 and E2.3 Racial essences are often supposed to play at least three important explanatory roles. R1. They explain what it is to be a member of a race. Members of a race are all and only those who possess the essence. R2. They explain the nonessential, but race-typical, properties possessed by members of a race. Race-typical features are (other things equal) caused by possession of the essence. Possession of a racial essence is also, at least normally, taken to be acquired by and only by those in some very special relationship with other genuine instances of the kind (typically, one’s parents), and so members of an essentialist racial kind share a lineage in which later instances of the kind are “offspring” or “descendants” of earlier ones. In this way, racial essentialism seems of a piece with other sorts of essentialism that involve specific lineages or causal histories (e.g., U.S. currency must come from the U.S. Mint, Ford F150 trucks must be made by the Ford Motor Company, Picasso paintings must have been caused by Picasso in the right way). Attention to causal histories or lineages draws attention to a third explanatory role for racial essences, they: R3. Explain why some people and not others possess the essence. Members of a race are those who stand in an “offspring-relation” to others who possess the essence. Here I use “offspring-relation” as a placeholder for whatever the relevant relation is, or relations are, by which one acquires an essence from a possessor. While for biological kinds, it is natural to understand this relation (or relations) biologically, there may be other relevant ways to understand such descent in the case of race or other human kinds.4 While racial essentialism seems common enough in everyday thought, there is now considerable agreement among many in the humanities, social sciences, and biological sciences, both that such racial essentialism is a false view, and also that it is a bad view (cf. Stoljar, this volume). As a result, most contemporary accounts of race are joined in touting their nonessentialist bona fides, despite the considerable other differences among them. It is regarded as a false view because, as nearly everyone agrees, there are no racial essences of the sort imagined. And it is a bad view because the false belief that such essences do exist and play the proposed explanatory roles seems to provide a metaphysical basis for generalizing about members of a race in pernicious ways. Because denying invidious and pernicious generalizations has long been a component of anti-racist thought, nonessentialism about racial kinds has become a standard feature of anti-racist thought as well.5 In what follows, I first briefly explore why the conception of racial essentialism I have just sketched plays such a central role in theorizing, namely that it articulates commitments of a widely held folk theory of race. In Section 24.2, I review reasons to believe that the folk theory is false, that there are no racial essences of the sort imagined. In Section 24.3, I consider the most obvious response to the failure of racial essentialism: the rejection of a belief in race itself. Then, in Sections 24.4 and 24.5, I consider a range of strategies for offering realist though nonessentialist accounts of race, some centered upon accounts of race that incorporate

362

Race

denial of E1, and others on accounts of race that incorporate denial of E2. If any of these accounts is successful, then it is possible to be a realist about race without a belief in essences. In Section 24.6, I return to consider the badness of essentialism, and I consider whether the nonessentialist accounts of race of the sort considered in Sections 24.4 and 24.5 escape the problems believed to plague essentialism. In Section 24.7, I note an emerging strategy for combatting the badness of essentialism, one that focuses upon psychological tendencies rather than metaphysical bases. In Section 24.8, I conclude.

24.1 Racial Essentialism is a Folk Theory Why do social theorists tend to think of racial essentialism in something like the way I have characterized it above? Several converging lines of evidence suggest that racial essentialism is widely assumed in folk-theorizing about race, at least in some cultures, and perhaps among humans in general. To begin with, there are many discussions of the concept of race in the European and American context that suggest racial essentialism is a central strand of race-thinking. Consider Kwame Anthony Appiah, who notes, in the European-American context, the widespread, underlying assumption of racialism, the view that: we could divide human beings into a small number of groups, called “races”, in such a way that the members of these groups shared certain fundamental, heritable, physical, moral, intellectual, and cultural characteristics with one another that they did not share with members of any other race. (1996: 54) When we look to historical accounts, Michael Banton and Jonathan Harwood argue that in the nineteenth century [writers] started to describe men as belonging to races and to maintain that differences between men and peoples stemmed from race. It was then said that some men belonged to races which had an inborn tendency to expand their territory and rule over others who belonged to lesser strains and from whom less could be expected. (1975: 11) And the historian George Fredrickson writes of the historical emergence of the assumption that a negative characteristic of a group “was organic and carried in the blood” (2002, 25). Similarly, the French sociologist Colette Guillaumin has suggested that contemporary racethinking emerged with the belief that groups are naturally diverse, because of endogenous characteristics which are determining factors in themselves, independently of history or economics” (1980; italics hers) Looking further afield than the European-American context for evidence of something like racial essentialism, the historian Susan Bayly writes of Indian castes that: both in the past and today, those sharing a common caste identity may subscribe to at least a notional tradition of common descent, as well as a claim of common geographical origin, and a particular occupational ideal … . In theory … the central characteristic of ‘caste society’ has been for many centuries the hierarchical ranking of castes or birth groups. The implication here is that to be of high or low caste is a matter of innate

363

Ron Mallon

quality or essence. … even people who came to reject caste principles either recently or in the more distant past are at least likely to have been familiar with these notions of corporate moral essences or qualities, meaning that in ‘caste society’, gradations of rank and precedence are innate, universal and collective. The implication would be that all who are born into so-called clean castes will rank as high, pure, or auspicious in relation to those of unclean or ‘untouchable’ birth, regardless of wealth, achievement or other individual circumstances. (Bayly 1999: 10) These sorts of conceptual and historical accounts suggest the idea that racial categories are distinguished by some sort of internal difference that is both category-defining and explanatory. But where does such racial essentialist thinking come from? Here, there are two important, but prima facie sharply different sorts of answers. On the one hand, some developmental psychologists suggest that a tendency towards essentializing some human groups emerges rather early (e.g., Hirschfeld 1996), and a range of cross-cultural evidence suggests that essentializing of human groups occurs across a range of cultures, even ones purported to lack a concept of race (Astuti et al. 2004; Gil-White 2001a, 2001b; Jones 2009). Such evidence has led some to suggest that essentialist thinking about human groups is “innate”, or at least “culturally canalized”—that is, it develops across a range of cultures as a kind of default assumption about some sorts of human groups (Machery and Faucher 2005; Mallon 2013). On the other hand, this claim has proven controversial (Dikötter 2015; Jackson 2017). For one thing, there is a substantial humanist literature to the effect that racial essentialist thinking was invented, along with race itself, by Europeans sometime since the Renaissance. Fredrickson, for instance, writes that: It is the dominant view among scholars who have studied conceptions of difference in the ancient world that no concept truly equivalent to that of ‘race’ can be detected in the thought of the Greeks, Romans, and early Christians. (2002: 17) Before this time, according to this consensus, our contemporary concept of race did not exist. Who is right? The questions here are very complex. For instance, some are inclined to treat cultural variation in the embrace of essentialist thinking as cases in which human-typical essentialist psychological proclivities (which are, after all, only tendencies) are overridden by local teaching about human groups, but others treat cultural variation as evidence of a cultural basis of racial essentialism. Deciding between these interpretations may require evidence that we do not have and perhaps will never have. For now, I bracket the question of the sources of folk essentialism, and I turn to assessing the content of the folk theory.

24.2

Racial Essentialism is False

It has long seemed plausible to think of folk racial kinds as putatively biological kinds, something that many accounts of such kinds (with their reference to differences “in the blood”), as well their connection to dehumanizing narratives about kind members (Smith 2011), seem to support. The most common reasons for thinking that racial essentialism is false rest upon the disqualification of biological candidates for such essential properties. Essentialism about biological kinds like species or subspecies is nearly universally rejected in biology, and via at least two different lines of thought (cf. Brigandt, this volume). The first reason is that biological variation is ubiquitous in the natural world, and this suggests that there are no simple, intrinsic properties shared by all and only members of a

364

Race

biological kind that play the imagined explanatory roles. Thus, essentialism has been generally rejected for species kinds, and a fortiori, for racial kinds: Just as our species, like other species, consists of a varied population of individuals, so too do groups within a species. Human beings form overlapping pools of genetic variation, not distinct races, each with its own distinctive genome. Because our genetic material dates back to the beginning of the evolutionary process, and because human populations have typically been separate for only tens of thousands of years, only a small proportion of variation is distinctive of particular human populations. (Sterelny and Griffiths 1999: 9) Since the advent of genetics, for instance, it has been tempting to think of the genes of biological kinds as candidates for serving as essences. But genes typically vary almost as much within racial groups as they do between racial groups, making them poor candidates for category-defining, explanatory kinds. These facts about variation do not (as anyone who has used home genetic ancestry tests knows) mean that genes carry no information about, say, the geographic locations of one’s ancestors. But such systematic variation does not seem to neatly align with racial boundaries and in some cases reflects very specific environmental adaptations (e.g., as in skin color to climate) instead of the broad array of traits suggested by the racialist (Templeton 2013). All this is to say that simple, explanatory essences of the sort classically imagined—essences shared by all and only members of a racial group—are unlikely to be found. Put schematically, this variation argument goes something like this: V1. The best candidate for an essence for race is some intrinsic biological property. V2. Intrinsic biological properties tend vary within racial populations, and crosscut them. For example, genetic variation is nearly as great within populations as it is across populations. V3. So, it is not the case that there is a simple biological property shared by all and only members of racial groups that explains category membership and race-typical properties (from V2). V4. E1 is false of biological kinds. Racial essences do not exist. A different way of rejecting essentialism also depends upon the thought that racial kinds are implicitly conceived as biological kinds, as well as the thought that biological kinds are appropriately conceived as populations. According to such “population thinking”, instances of putative biological kinds like species and races are to be identified via the relationship of an individual organism to a population of organisms. One family of views identifies populations with particular, spatio-temporally extended individuals that are comprised of many individual organisms across space and time (Ghiselin 1974, Hull 1978), giving rise, perhaps, to a sort of historical essentialism about species (Griffiths 1999). Such a view allows us to account for the sorts of things that we say of species: that they come to exist, evolve, and go extinct, for instance. We can simply stop there and identify species with segments of an accurate phylogenetic tree that indicates speciation and extinction events. A somewhat different way of thinking about populations is to see them as comprised of a group of individual organisms that stand in certain relationships to other organisms in the population—for example, individuals that are part of a reproductively isolated breeding community, one whose members breed with one

365

Ron Mallon

another at a significantly higher rate than with other organisms. On such a view, “the species is not defined by intrinsic, but by relational properties” (Mayr 1984: 535). Both conceptions conceive species as historical particulars whose essential features are fixed by relations to particular historical events or particular populations. So, a familiar thought experiment goes, suppose you traveled to Mars and found an indigenous, bird-like creature there that looked exactly like Cardinalis cardinalis (the Northern Cardinal), and suppose you investigated further and found that it was indistinguishable from an instance of Cardinalis cardinalis on earth—at the cellular level, and even at the genetic level, say. Still, if you had conclusive evidence that this organism had arisen entirely independently of Cardinalis cardinalis, and so its similarity was a bizarre and wildly improbable accident, then you should not classify it as a member of Cardinalis cardinalis. (It would not be a part of the same individual, nor a member of the same breeding population.) This population argument can be expressed as follows: P1. If races are biological kinds, then they are some sort of population. P2. Whether an individual organism is a member of a population does not depend only upon its intrinsic properties. It depends also on its relational properties. P3. E2 is false of biological kinds. Racial essences do not exist. In short, arguments from variation and from population thinking have suggested that there are no simple, intrinsic, natural properties shared by all and only members of a race that play the roles that essences are supposed to play, and they have led to widespread rejection of the sort of racial essentialism that has long been a part of folk theorizing. While essentialism about biological kinds still has a few defenders (Devitt 2008), the rejection of essentialism is the subject of a striking level of consensus within philosophy of biology and dovetails with a much broader nonessentialism consensus about race that spans much of the humanities, social sciences, and natural sciences.

24.3

Alternatives to Folk Essentialism: Skepticism

If folk essentialism about race is wrong, if there are no racial essences, then how should we think about race? Perhaps the most straightforward option is to believe that races do not exist; the rejection of essentialism is a way of rejecting the existence of race. In many domains, rejection of key aspects of a folk theory is taken to be sufficient for rejecting the folk theory too (e.g., Stich 1983). And, indeed, racial anti-essentialism has been a core argument in favor of racial skepticism, the view that races (as conceived by the folk) do not exist (e.g., Appiah 1996; Zack 1993, 2002; Blum 2002; Glasgow 2009). However, many others have resisted this conclusion. For one thing, the existence of races seems obvious. For another, race seems to be among the most common and explanatory variables in the social sciences (Root 2000). For a third, race seems to play an important role in many people’s identities, and in cultural projects and collective projects of social organization that aim to resist racial oppression. For all these reasons, many have wondered about the possibility of racial realism without racial essentialism, and a variety of nonessentialist accounts of race have been offered to play this role.

24.4

Denying E1: Nonessentialist Race Without Simple Necessary and Sufficient Features

Racial skepticism is only entailed by the denial of E1 and E2, however, if you think the existence of race requires racial essences of the sort they imagine. But there are nonessentialist

366

Race

ways of thinking about the reality of race. In this section, I highlight one way of being a nonessentialist about race: provide an account of race that denies the existence of simple properties that define category membership.

24.4.1

Races as a Homeostatic Property Cluster Kinds

Perhaps the most prominent development in the theory of natural kinds in recent decades is the development of “homeostatic property-cluster” or HPC accounts of kinds (Boyd 1988, 1991, 1999). On HPC accounts, kinds are characterized not by essences of the sort suggested by E1 and E2, but by a set of imperfectly shared, typical properties, along with some homeostatic mechanism that explains their nonaccidental cooccurrence (see Brigandt, this volume).6 Such accounts of natural kinds allow for variation among kind members and explain how reference to the kind supports prediction and explanation. For this reason, some have thought it a promising account of species kinds in the wake of the rejection of essentialism for biological kinds (e.g., Boyd 1999). With this in mind, it is a short step to the suggestion that while variation within and among races is common, nonetheless races might be some sort of homeostatic property cluster kind (Mallon 2007, 2016), thus explaining the central role racial kinds seem to play in contemporary social scientific projects while denying racial essentialism (because accounts of such kinds deny, or are consistent with the denial of, E1).

24.4.2

Races are Epistemic Kinds

Still more recent developments in the philosophical literature on natural kinds have retreated from the metaphysical assumptions of the homeostatic property cluster view, noting, instead that many kinds referred to in the special sciences seem to lack clear mechanisms of causal homeostasis, and might be better regarded as “simple causal kinds” or mere “epistemic kinds” (Craver 2009, Khalidi 2013, Slater 2015). The idea here is that we can pick out kinds as causally important (in the sense that we refer to them in causal generalizations in our sciences) without taking a position on metaphysical questions like their underlying structure, their eliminability from our best metaphysics, and so forth. Applying this account to race suggests the possibility that we might refer to racial kinds for specific purposes without commitment to underlying essences, or even to underlying homeostatic mechanisms. Races might just be useful things to pick out in a context, with other, metaphysical questions regarding their status addressed, if at all, later on, in the fullness of time.

24.5

Denying E2: Nonessentialist Race Without Intrinsic and Natural Features

A different way of approaching nonessentialist accounts of race is to deny E2, and focus upon accounts of races that are relational, rather than intrinsic, or social rather than natural.

24.5.1 Races as Populations A natural place to look for models of anti-essentialism about race is in debates over biological species, and some philosophers have identified races with partially reproductively isolated biological populations, populations that are not individuated by intrinsic, essential differences, but rather by historical or relational features (Andreasen 1998; Kitcher 1999, though see 2007; Spencer 2014).

367

Ron Mallon

24.5.2 Races as Social Constructions A different family of positions asserts anti-essentialist accounts of race by insisting race is some sort of product of our socio-cultural practices, and these accounts deny essentialism by denying the naturalness and intrinsicness of the properties that constitute or ground race (cf. Griffith, this volume). Constructionist views may understand races as constituted by relational facts such as our practices of ascription, or by the causal effects of those practices of ascriptions, or both. Such accounts come in a number of forms, including the claim that race is conferred by our social practices (Àsta 2018; Vaidya and Wallner, this volume), or that it is an HPC kind made up of a social role together with the causal effects of such a role (Mallon 2016), or that it is a social role characterized by subordination on the basis of bodily features (Haslanger 2000), or that races are the cultural groups that are the causal result of social practices of racial ascription and differential treatment (Jeffers 2019). These accounts are all nonessentialist in that they deny that race is a natural feature of persons, and at least most also view racial membership as a relational property of persons.

24.5.3 Races as Nonnatural, “Thin” Kinds Still another way of de-essentializing race is to deny the naturalness of race by insisting that racial classifications themselves do not pick out important natural boundaries (cf. Tahko, this volume). One way to do this is to identify races with groups that are specified by a thin but objective set of criteria, criteria that might include visible criteria (like skin color, hair type, body morphology) often used to ascribe persons to racial groups or it might include one’s relationship to some ancestral population or geographic location. (e.g., Glasgow and Woodward 2015, Hardimon 2017). The crucial thing about such “thin” accounts is that they deny that racial kinds are human-independent, intrinsic, natural properties that play important explanatory roles.

24.6

Is Nonessentialism Less Bad?

It seems clear enough that essentialist accounts of kinds support explanations and predictions about category members. Traditional essentialism has it that the typical properties of racial group members are to be explained by appeal to members shared essence, and the proclivities of individual members of a race can be predicted from their possession of such essences. So, essences have been taken to support pernicious generalizations, and generalizations have been taken to be evidence for the assumption of essence. At the outset, I noted that essentialism is widely rejected because it is false, but it is also widely regarded as bad because of the role that essentialism can play in apparently underwriting generalizations about racial category members, especially false and pernicious generalizations about, say, aptitudes or moral dispositions. The idea that racial generalizations can be problematic is familiar enough, though it is difficult to explain whether all generalizations are problematic (say, in virtue of failing to treat individuals as individuals) or whether some more specific kind of generalization is at fault (cf. Beeghly 2018). Allowing that essentialist accounts of racial kinds do encourage and support pernicious generalizations, still, we can ask, do nonessentialist accounts do better? One reason to worry is that nonessentialist accounts of natural kinds like biological species have been proposed to support generalizations, explanations and predictions about kind members. And this makes

368

Race

sense: even if the purple finch (Carpodacus purpureus) is not properly characterized by possession of some unseen, intrinsic, explanatory essence, it is nonetheless the case that individual purple finches tend to have a lot in common with other purple finches, and one can learn about their species-typical features from a good North American bird guide. Insofar as nonessentialist accounts of racial kinds are like (and in some cases perhaps adaptations of) nonessentialist accounts of species, we might expect them to also support various sorts of generalizations, explanations, and predictions about kind members (Mallon 2007). We can see the possibility of such generalization-supporting nonessentialist kinds by focusing upon the first two explanatory roles played by essences. Those held that essences (R1) explain what it is to be a member of a race, and (R2) explain the race-typical properties possessed by members of a race. If we, say, suppose a constructionist, nonessentialist account of race based on a homeostatic property cluster, we might find that: members of a race R are those that, more or less, share a socially constructed social positions in virtue of being classified as an R, and so also share, more or less, the causal effects of being so classified. Such a view holds that there is a “cluster” of nonaccidentally cooccurring properties characterizing members of a race, and that that the existence of this cluster is constituted and/or caused by social facts including practices of racial classification and differential treatment. Such an account is nonessentialist both in that it denies that members of a race share the simple necessary and sufficient conditions imagined by the essentialist, and also in that it denies that membership in a race is constituted by intrinsic, natural features of a person. But on such an account, racial kinds can play explanatory roles previously attributed to essences. Membership in the kind occurs when one instantiates—to a sufficient degree—the properties in the property cluster. And kind-typical properties are explained as consequences of some homeostatic mechanism (e.g., a system of social classification and differential treatment), or as the causal effects of other properties in the cluster. The worry is that the ability to fill these explanatory roles is sufficient to allow such nonessentialist kinds to support probabilistic generalizations about category members. Indeed, homeostatic property clusters were introduced to understand how inductive explanation and prediction is possible in the absence of classic essences (Boyd 1988). Such kinds seem to pose no special barrier to pernicious generalizations about racial kinds understood in a nonessentialist way (Mallon 2007). For instance, if you held with the United States government during the early 1940s that: Japanese Americans are likely to be loyal to Japan and disloyal to the United States in the ongoing war with Japan. Then, given a random American x of Japanese descent, or a random community c of Japanese Americans, you might infer that: x and c are likely to be disloyal to the U.S. in the ongoing war with Japan. Views like these were, indeed, an important cause of the internment of many tens of thousands of U.S. citizens of Japanese descent during the war with Japan.

369

Ron Mallon

But suppose we ask what connects Japanese Americans as a kind in the minds of those who so generalized? While the evidence for folk essentialism reviewed above supports the idea that a belief in essences may well have been important to such thinking, the arguments I have just offered suggest that nonessentialist accounts of racial kinds could also allow for such generalizations. Emphasizing a “more or less” quality of possession of kind-characteristic traits, or a cultural rather than biological bases, does not, by itself, exclude pernicious reasoning about category members.7 Faced with pernicious generalizations based on nonessentialist accounts of racial kinds, opponents of such generalizations cannot rest with the denial of essentialism. They must undertake the more difficult task of showing such generalizations are unjustified or false, a task that requires not simply producing counterexamples to characterizations of essences, but of engaging closely with the social sciences in which racial membership figures as a common variable.

24.7

Combatting Racial Essentialism With Psychological Anti-Essentialism

Nonessentialist accounts of race are attempts to provide more truthful accounts of the metaphysics of race, where this project is in part motivated by interests in truth, and in part by the hope of undermining pernicious generalizations. Recent psychological work has suggested a different approach to short-circuiting pernicious generalizations, though, by disrupting intuitive or default essentialist inferences that folk racial thinking seems to support. In contrast to the metaphysical nonessentialist emphasis on a lack of metaphysical support for essentialism, the psychological anti-essentialist seeks strategies for short cutting the psychological processes that make essentialism and pernicious generalizations cognitively tempting. For instance, Sarah-Jane Leslie has argued that the use of generic terms for racial groups (like “black”, “white”, and “Asian”) seems to activate folk essentialist thinking, leading people to view those groups in an essentialist way. Leslie goes on to suggest there could therefore be benefits of reducing the use of generic terms for racial, ethnic, and religious groups: the evidence suggests that the use of labels and generics contributes to essentialization, and so the converse may also hold: reducing the use of labels and generics for racial, ethnic, and religious groups may reduce the extent to which children grow up essentializing these groups. (2017: 42) For Leslie, as for the tradition of metaphysical nonessentialists, one aim in combatting essentialist thinking is to shortcut pernicious inferences about members of racial groups. But in contrast to nonessentialists emphasis on the denial of essences, Leslie’s argument is psychological, and its success depends upon the patterns of construal and inference that human minds use to make sense of and act within our social world. It replaces the metaphysical question of whether pernicious racial generalizations are still possible without racial essentialism (as I have suggested they are) with the psychological question of whether they would still occur (as frequently or as perniciously) without the use of terms that activate essentialist patterns of reasoning. While it remains to be seen how successful this project will be, the idea that distinguishing essentialist from nonessentialist accounts according to the psychological dispositions they activate, rather than the inferences they license or entail, is a good fit with the idea mooted in Section 24.1 that essentialist thinking emerges from deep-seated human psychological proclivities, a better understanding of which may enable countering of pernicious generalizations.

370

Race

24.8

Summing Up

Racial essentialism is a manifestation of the older and more general idea of essence, one with a long history in philosophy, and whose exact parameters remain contested. Here, I have elucidated a particular account of what racial essentialism is and explained why that account is widely viewed as mistaken. At the same time, I have also suggested that mere recognition of this mistake is not enough to achieve the normative aims of many nonessentialists, since nonessentialist accounts of race can also support pernicious generalizations about members of racial categories. Faced with this, at least two compatible options are available for connecting nonessentialism with the undermining of pernicious generalizations. Nonessentialists can undertake the hard work of showing specific pernicious generalizations are empirically false. And, alternatively, they can show that a psychological and linguistic strategy of undermining essentialist thinking tends to also undermine pernicious generalizing (even if, in principle, such generalizations could be supported in another way).

24.9

Related Topics

Brigandt, Ingo, “Biological Species” (this volume) Griffith, Aaron, “Social Construction” (this volume) Ritchie, Katherine, “Language of Essence” (this volume) Rosario, Esther, “Sex and Gender” (this volume) Tahko, Tuomas, “Natural Kind Essentialism” (this volume) Vaidya, A., and Wallner, M., “Conferralism” (this volume)

Acknowledgments I am indebted to Kathrin Koslicki and Michael J. Raven for helpful feedback on earlier drafts.

Notes 1 Haslanger, for example, offers a definition of socially constructed race ( 2000), and elsewhere, links her definitional understanding of social construction to Fine’s (1994) definitional conception of essence ( Haslanger 1995, endnote 12). 2 For instance, if we take an arbitrary set of things (a1, a2, … an), we can specify a unique property of each member of the set (p1, p2, … pn) and join them disjunctively (p1 or p2 or … pn). The resulting complex, disjunctive property specifies a property shared by all and only members of the set. But this kind of disjunctive property is not the sort imagined to figure in essences. 3 Again, it should be noted that strands of contemporary metaphysics that deny E1 or E2 might nonetheless retain essences in some other sense; for example, they may allow that essences might be relational (e.g., Raven 2022) or complex. Paul Griffiths, for instance, regards Richard Boyd’s work on homeostatic property cluster kinds—kinds that potentially allow the denial of E1 and E2—as “a substantial revision of the traditional ideas of essence and natural kindhood” (1999: 219). Whether such views amount to a revision of mistaken ideas about essences or a rejection of essentialism merits continued investigation (into the essence of essences, as it were), but for present purposes I focus my attention on the conception of essence and nonessentialism stipulated in the text. 4 Is attention to “offspring-relations” inconsistent with E2’s insistence that essences be nonrelational? Some theorists conceive of kinds as having a “historical essence” (e.g., Griffiths 1999; Bach 2012). Griffiths, for instance, sets out an historical view of biological taxa on which “nothing that does not share the historical origin of the kind can be a member of the kind” (1999: 219), and then proceeds to explain why organisms with a common phylogenetic role would tend to share the sorts of structural properties we expect species to possess. In insisting upon relational essences, this way of conceiving of

371

Ron Mallon lineages denies the sort of narrow essentialism E2 embodies. But an alternative interpretation is consistent with the denial of relational essences in E2, holds that the offspring relation is simply causal, and that lineages occur because essence instances are the only reliable, and perhaps the only nomologically possible, cause of other essence instances. 5 A similar pattern can be observed in feminist philosophy. Charlotte Witt once remarked that “showing a position is ‘essentialist’ can function in and of itself as a good reason for rejecting it” ( 1995: 321). However, recent work suggests that different ways of understanding essence might resuscitate the notion. See, for example, Raven (2022) for a defense of the usefulness of the idea of essence to social ontology, and Passinsky (2021) for a defense of a Finean approach to essence in feminist metaphysics. 6 Given the rejection of E1 and E2, we can regard HPC kinds as a nonessentialist account as I do here, though one can also regard it (cf. Griffiths 1999 and fn. 3 above) as a revisionary account of essence not beholden to E1 and E2. 7 It is true that the nonessentialist can insist that some individuals are not typical of a race R, and so do not possess a race-typical property, P. However, the essentialist can say this as well since essentialists typically explain individual variation from type as the result of the essence being incompletely manifested in the individual ( Appiah 1996, pp. 54–5).

References Andreasen, R.O. (1998) ‘A New Perspective on the Race Debate’, British Journal of the Philosophy of Science, 49, pp. 199–225. Appiah, K.A. (1996) ‘Race, Culture, Identity: Misunderstood Connections’, in K. Anthony Appiah and A. Guttmann (eds) Color Conscious: The Political Morality of Race. Princeton, NJ: Princeton University Press, p. 192. Ásta (2018) Categories We Live By: The Construction of Sex, Gender, Race, and Other Social Categories. New York, USA: Oxford University Press. Astuti, R., Solomon, G.E.A. and Carey, S. (2004) ‘Constraints on Cognitive Development: A Case Study of the Acquisition of Folkbiological and Folksociological Knowledge in Madagascar.’, Monographs of the Society for Research in Child Development, 69(3), p. i+v+vii-viii+1-161. Bach, Theodore (2012) Gender Is a Natural Kind with a Historical Essence. Ethics, 122(2), 231–272. Banton, M. and Harwood, J. (1975) The Race Concept. New York: Praeger. Bayly, S. (1999) Caste, Society, and Politics in India. New York: Cambridge University Press. Beeghly, E. (2018) ‘Failing to Treat Persons as Individuals’, Ergo, an Open Access Journal of Philosophy, 5, 687–711. Available at: 10.3998/ergo.12405314.0005.026. Blum, L. (2002) I’m not a Racist But …. Ithaca, NY: Cornell University Press. Boyd, R. (1988) ‘How to Be a Moral Realist’, in G. Sayre-McCord (ed.) Essays on Moral Realism. Ithaca, NY: Cornell University Press, pp. 181–228. Boyd, R. (1991) ‘Realism, Anti-Foundationalism and the Enthusiasm for Natural Kinds’, Philosophical Studies, 61, pp. 127–148. Boyd, R. (1999) ‘Homeostasis, Species, and Higher Taxa’, in R.A. Wilson (ed.) Species: New Interdisciplinary Essays. Cambridge, MA: MIT Press, pp. 141–186. Brigandt, I. (2024) ‘Biological Species’. Koslicki, Kathrin & Raven, Michael editors. The Routledge Handbook of Essence in Philosophy. Routledge. This Volume. Correia, F. (2024) ‘Non-modal Conceptions of Essence.’ Koslicki, Kathrin & Raven, Michael. editors. The Routledge Handbook of Essence in Philosophy. Routledge. This volume. Craver, C.F. (2009) ‘Mechanisms and Natural Kinds’, Philosophical Psychology, 22(5), pp. 575–594. Devitt, M. (2008) ‘Resurrecting Biological Essentialism’, Philosophy of Science, 75(3), pp. 344–382. Dikötter, F. (2015) The Discourse of Race in Modern China. 2nd ed.New York: Oxford University Press. Fine, K. (1994) ‘Essence and Modality’, Philosophical Perspectives, 8 (Logic and Language), 8, pp. 1–16. Available at: 10.2307/2214160. Fredrickson, G.M. (2002) Racism: A Short History. Princeton, N.J.; Oxford: Princeton University Press. Ghiselin, M.T. (1974) ‘A Radical Solution to the Species Problem’, Systematic Zoology, 23(4), pp. 536–544.

372

Race Gil-White, F. (2001a) ‘Are Ethnic Groups Biological “Species” to the Human Brain?’, Current Anthropology, 42(4), pp. 515–554. Gil-White, F. (2001b). Sorting is not Categorization: A Critique of the Claim that Brazilians Have Fuzzy Racial Categories. Journal of Cognition and Culture, 1(3), 219–249. Glasgow, J. (2009) A Theory of Race. New York: Routledge. Glasgow, J. and Woodward, J.M. (2015) ‘Basic Racial Realism’, Journal of the American Philosophical Association, 1(3), pp. 449–466. Griffith, A. (2014) ‘Social Construction.’ Koslicki, Kathrin & Raven, Michael . editors. The Routledge Handbook of Essence in Philosophy. Routledge. This volume. Griffiths, P. (1999) ‘Squaring the Circle: Natural Kinds with Historical Essences’, in R.A. Wilson (ed.) Species: New Interdisciplinary Essays. Cambridge, MA: MIT Press, pp. 209–228. Guillaumin, C. (1980) ‘The Idea of Race and Its Elevation to Autonomous Scientific and Legal Status’, in UNESCO (ed.) Sociological Theories: Race and Colonialism. Paris: UNESCO. Hardimon, M.O. (2017) Rethinking Race: The Case for Deflationary Realism. Cambridge, Mass.: Harvard University Press. Haslanger, S. (1995) ‘Ontology and Social Construction’, Philosophical Topics, 23(2), pp. 95–125. Haslanger, S. (2000) ‘Gender and Race: (What) Are They? (What) Do We Want Them To Be?’, Noûs, 34(1), pp. 31–55. Haslanger, S. (2003) ‘Social Construction: The “Debunking” Project’, in F. Schmitt (ed.) Socializing Metaphysics: The Nature of Social Reality. Lanham, MD: Rowman adn Littlefield, pp. 301–325. Hirschfeld, L.A. (1996) Race in the Making: Cognition, Culture, and the Child’s Construction of Human Kinds. Cambridge, MA: MIT Press (Learning, Development, and Conceptual Change). Hull, D.L. (1978) ‘A Matter of Individuality’, Philosophy of Science, 45(3), pp. 335–360. Available at: 10.1086/288811. Jackson, J.P. (2017) ‘Cognitive/Evolutionary Psychology and the History of Racism’, Philosophy of Science, 84(2), pp. 296–314. Available at: 10.1086/690720. Jeffers, C. (2019) ‘Cultural Constructionism’, in What Is Race?: Four Philosophical Views. New York: Oxford University Press. Available at: https://oxford.universitypressscholarship.com/view/10.1093/ oso/9780190610173.001.0001/oso-9780190610173-chapter-3 (Accessed: 29 October 2020). Jones, D. (2009) ‘Looks and Living Kinds: Varieties of Racial Cognition in Bahia, Brazil’, Journal of Cognition and Culture, 9, pp. 247–269. Khalidi, M.A. (2013) Natural Categories and Human Kinds: Classification in the Natural and Social Sciences. New York: Cambridge University Press. Kitcher, P. (2007) ‘Does “Race” Have Future?’, Philosophy and Public Affairs, 35(4), pp. 293–317. Kitcher, P. (1999) ‘Race, Ethnicity, Biology, Culture’, in L. Harris (ed.) Racism. New York: Humanity Books, pp. 87–120. Leslie, S.-J. (2017) ‘The Original Sin of Cognition: Race, Prejudice and Generalization’, Journal of Philosophy, 114(8), pp. 393–421. Machery, E. and Faucher, L. (2005) ‘Why do we Think Racially?’, in H. Cohen and C. Lefebvre (eds) Handbook of Categorization in Cognitive Science. Orlando, FL: Elsevier, pp. 1010–1033. Mallon, R. (2007) ‘Human Categories Beyond Non-Essentialism’, Journal of Political Philosophy, 15(2), pp. 146–168. Mallon, R. (2013) ‘Was Race Thinking Invented in the Modern West?’, Studies in History and Philosophy of Science, 44, pp. 77–88. Mallon, R. (2016) The Construction of Human Kinds. Oxford: Oxford University Press. Mayr, E. (1984) ‘Species Concepts and Their Application’, in E. Sober (ed.) Conceptual Issues in Evolutionary Biology: An Anthology. Cambridge, MA: MIT Press, pp. 531–540. Passinsky, A. (2021) ‘Finean Feminist Metaphysics’, Inquiry: An Interdisciplinary Journal of Philosophy, 64(9), pp. 937–954. Available at: 10.1080/0020174x.2019.1669984. Putnam, H. (1975) ‘The Meaning of “Meaning”’, in Mind, Language and Reality: Philosophical Papers. New York: Cambridge University Press, pp. 215–271. Raven, M.J. (2022) ‘A Puzzle for Social Essences’, Journal of the American Philosophical Association, 8(1), pp. 128–148. Available at: 10.1017/apa.2020.48. Root, M. (2000) ‘How We Divide the World’, Philosophy of Science, 67((Proceedings)), pp. 628–639. Slater, M.H. (2015) ‘Natural Kindness’, British Journal for the Philosophy of Science, 66(2), pp. 375–411.

373

Ron Mallon Smith, D.L. (2011) Less Than Human: Why We Demean, Enslave, and Exterminate Others. New York: St. Martins Press. Spencer, Q. (2014) ‘A Radical Solution to the Race Problem’, Philosophy of Science, 81(5), pp. 1025–1038. Sterelny, K. and Griffiths, P.E. (1999) Sex and Death: An Introduction to Philosophy of Biology. Chicago, Ill.: University of Chicago Press (Science and its conceptual foundations). Stoljar, N. (2024) ‘Social Justice.’ Koslicki, Kathrin & Raven, Michael. editors. The Routledge Handbook of Essence in Philosophy. Routledge. This volume. Stich, S.P. (1983) From Folk Psychology to Cognitive Science: The Case Against Belief. Cambridge, MA: MIT Press. A Bradford Book. Tahko, T. (2024) ‘Natural Kind Essentialism.’ Koslicki, Kathrin & Raven, Michael. editors. The Routledge Handbook of Essence in Philosophy. Routledge. This volume. Templeton, A.R. (2013) ‘Biological Races in Humans’, Studies in History and Philosophy of Science Part C: Studies in History and Philosophy of Biological and Biomedical Sciences, 44, pp. 262–271. Vaidya, A., and Wallner, M. (2024) ‘Conferralism.’ Koslicki, Kathrin & Raven, Michael The Routledge Handbook of Essence in Philosophy. Routledge This volume. Witt, Charlotte (1995) Anti-Essentialism in Feminist Theory. Philosophical Topics, 23, 321–344. Zack, N. (1993) Race and Mixed Race. Philadelphia: Temple University Press. Zack, N. (2002) Philosophy of Science and Race. New York: Routledge.

374

25 SEX AND GENDER Esther Rosario

25.1

Introduction

This chapter deals with the idea that sex and gender are by their natures essentially distinct. That sex and gender are distinct was the prevailing view in Anglo-American feminist thought in the wake of Simone de Beauvoir’s The Second Sex (1949) until the popular reception of Judith Butler’s Gender Trouble (1990) and Bodies that Matter (1993), which challenge the sex/gender distinction in ways unarticulated previously. Following Butler’s influential work, two related debates dominated the feminist literature throughout the turn of the century: the distinction debate and the essentialism/anti-essentialism debate. The first debate concerns whether sex and gender are essentially distinct, while the second pertains to whether sex and gender have essences. By 2006, Linda Martín Alcoff argued that the essentialism/antiessentialism debate in feminist philosophy had become “passé” with most believing antiessentialists had won (152). I shall outline why neither debate was resolved in the early 2000s and how they have shifted in recent years to focus on self-identification. In order to address what the sex/gender distinction consists in and why it is contested, this chapter also deals with the matter of what sex and gender are, which lacks consensus in the philosophical literature. In what follows, I characterize the perspectives I survey as externalist, contextualist, and internalist. Internalist approaches rely on self-identification as a necessary and sufficient condition for having a sex or a gender. In contrast, externalist approaches do not treat self-identification with a sex or a gender as necessary or sufficient. Instead, having a sex or a gender (at least partly) depends on biological or social factors independent of self-identification. In contrast, contextualist approaches treat sex and gender as property clusters, incorporating external factors, self-identification, and other psychological features. The philosophical study of sex and gender has generated a rich and diverse literature. I’ll take two passes at characterizing the sex/gender distinction and assess the viability of essentialist and anti-essentialist approaches to the issue. The theories I’ll survey do not exhaust the broader literature and are a snapshot of the major issues I take to be the most salient, particularly at the present time. I’ll also characterize three problems that essentialist and anti-essentialist accounts face: the Inclusion Problem, the Definition Problem, and the Exclusion Problem.

DOI: 10.4324/9781003008750-29

375

Esther Rosario

The Inclusion Problem concerns the difficulty of, for example, including all women in the category woman without identifying necessary conditions for being a member of that category.1 The Definition Problem points to the worries that arise when there is no clear target for theorizing about (or otherwise addressing) feminist aims (Heyes 2000; Stone 2004; Mikkola 2017; Bogardus 2022). I’ll argue that anti-essentialists about gender face the Inclusion Problem, and anti-essentialists about sex and/or gender encounter the Definition Problem. The Exclusion Problem, on the other hand, highlights difficulties that both essentialists and anti-essentialists confront in offering theories of sex and gender essences that do not account for intersex, trans, two-spirit, or genderqueer2 categories or identities, or that exemplify sexist stereotypes. The Exclusion Problem relates to Sally Haslanger’s (2000) normativity problem for gender—the worry that “any definition of ‘what a woman is’ is value-laden and will marginalize certain females, privilege others and reinforce current gender norms” (37). In short, I shall unpack how the Inclusion and Definition Problems highlight difficulties with vagueness in anti-essentialist theses about sex and gender. I’ll also explore how the Exclusion Problem can render theses about sex and gender unacceptably narrow.

25.2

Sex and Essentialism

In this section, I’ll outline three types of essentialist position about sex—biological determinism, sexual difference, and socio-historical-linguistic appeals. These types of essentialist position are not exhaustive but offer a useful contrast. Feminist theories that endorse the sex/gender distinction usually target biological determinism. Biological determinism is a form of biological essentialism, which treats biology as destiny (Beauvoir 1949). Biological determinist views about sex differences have a long history (Taylor Merleau 2003; Mikkola 2022). Such views include the notion that differences in metabolic state determine sex-based psychosocial and behavioral traits. This sort of biological determinism was developed by biologist Patrick Geddes and naturalist J. Arthur Thompson in The Evolution of Sex (1889). Their metabolic theory contrasts significantly with Charles Darwin’s (1871) theory of sexual selection. Rather than positing that successful reproductive strategies shape the appearance and behavior of females and males as Darwin does, Geddes and Thompson postulate females as anabolic (energy conserving) and males as katabolic (energy spending) types of individual. On their view, sex differences are metabolic differences that evolve over time (especially in terms of appearance and behavior) via mate selection. The activity levels of these individuals, which Geddes and Thompson note permit of exceptions, generally form a “constitutional contrast” (Geddes and Thompson 1889: 249–50). In other words, having an anabolic or katabolic nature is a matter of fundamental constitution rather than selection pressure. On their view, morphological and biobehavioral differences are expressions of metabolic differences such that “the same general habit of body … results in the production of male elements in the one case or female elements in the other” (Geddes and Thompson 1889: 19). Geddes and Thompson’s metabolic theory appears to posit dispositional differences in energy expenditure whereas chromosomal accounts of sex difference invoke genetic information as the microstructural blueprint for sex differentiation along stages of sex development. The agenda to explain sex differences in biologically deterministic terms remains popular (Fausto-Sterling 1992; Kourany 2010; Meynell 2012). Although Geddes and Thomson’s metabolism-based dispositional analysis has been abandoned, analyses of brain-based essential differences between the sexes have expanded from the early modern period to contemporary brain studies (Malebranche Lennon 1674; Fine 2010). Psychologist Simon Baron-Cohen (2003) is one prominent sex researcher who attributes differences in human

376

Sex and Gender

female and male behavior to differences in innate brain structure. He outlines five levels of sex: genetic, gonadal, genital, brain type, and sex-typical behavior. Brain type and behavior are ultimately determined by genetic sex (XX or XY chromosomes structures). On his view, females develop empathizing brains and males develop systematizing brains. These two brain types purportedly cause individuals to engage in and excel at empathizing or systematizing behavior. Empathizing involves proficiency in communication and “mind-reading” (BaronCohen 2003: 29–84). Systematizing concerns proficiency in mathematical and logical reasoning and organizational skills. Like other brain-based sex researchers (Gurian and Stevens, 2005; The Gurian Institute 2009; Brizendine 2006, 2010), Baron-Cohen endorses the organizational-activational hypothesis. This hypothesis asserts that sex steroids prenatally organize some brain regions in a dimorphic fashion, and then later (e.g., during and after puberty) act again to activate brain circuits so as to result in sex-specific behaviors (Arnold 2009). Testosterone and estradiol are the hormones leading to the brain found in males, whereas estrogen is instrumental in brain development in females. For Baron-Cohen and others who endorse the organizational-activational hypothesis, exposure to pre-natal testosterone marks the “essential difference” in language skills and sociability between females and males with higher amounts leading to systematizing and lower amounts to empathizing. Biological determinism is considered an essentialist view of sex because it proffers that females and males have certain types of biological features that are intrinsic and necessary to being a member of the kind female or male. Metabolic and brain-based accounts of sex are worth highlighting because they purport to explain how an intrinsic feature like metabolism or chromosome type determines, on average, female and male behavior. That some or all behavior in females and males is metabolically or genetically determined rather than socially learned is a controversial claim in the distinction and essentialism/antiessentialism debates. Moreover, the idea that there are binary developmental trajectories in humans, and that intersex conditions are exceptions that do not undermine this rule is also controversial (Ainsworth 2017; van Anders et al. 2017). Biological determinism is typically championed outside of feminist theorizing, which Cressida Heyes calls “anti-feminist biological essentialism” (2000: 33). Biological essentialisms that reject biological determinism but endorse the sex/gender distinction are heralded by some feminist philosophers, namely of the sexual difference variety. Sexual difference approaches assume there is a biological reality that exists outside of socialization. On this view, biological sex exists prior to social expectations about females and males, and is treated as a natural kind (Tahko, this volume). The usual target of feminist anti-essentialist arguments is biological essentialism within feminist theory rather than the biological determinism I sketched above (Heyes 2000). Biological determinist and sexual difference accounts appeal to an externalist notion of sex where sex exists in virtue of biological properties independent of human attitudes. Both of these views face the Exclusion Problem because they make universal claims about females and males, which may exclude persons who do not have typical female or male attributes or whose self-identification does not match such attributes. Biological essentialism, whether in the form of biological determinism or sexual difference, is not the only type of essentialism about sex. Accounts that attribute rigid socio-historical meanings to sex terms, arguably, appeal to a type of essence that is linguistic and fixed by socio-historical context. Instead of being a property that exists independently of attitudes or language use, what sex is is how it is talked about over time and place. What makes this account essentialist is the particular ways sex is thought and talked about historically

377

Esther Rosario

establishes necessary and sufficient conditions for what it is. An example of sex treated as a socio-historical linguistic matter is the idea that biological sex cannot be conceptually separated from traditional social expectations about the sexes. In other words, any present or future invocation of the concept sex or the term “sex” independent of an invocation of social attitudes and practices cannot be said to invoke sex. This type of essentialism is externalist based not only in social practices but in the history of language use. Butler’s (1993) theory that sex is a regulatory ideal can be interpreted as positing a sort of socio-historical essence because they argue that sex is not a “bodily given” but a “cultural norm which governs the materialization of bodies” which has a social history (Butler 1993: 236–8). Butler and philosophers responding to their work treat this view as anti-essentialist since they reject the idea that biological sex precedes the social norms that pertain to it (Butler 1993; Witt 1995; Heyes 2000). Butler’s view contrasts importantly with Simone de Beauvoir’s and Sally Haslanger’s externalist positions discussed in the next section.

25.3 The Sex/Gender Distinction: First Pass I shall offer my initial pass at the sex/gender distinction by invoking two emblematic philosophical statements. The first is Beauvoir’s (1949) claim that, “One is not born, but rather, becomes woman” (283). And the second is Haslanger’s (2000) reference to the slogan, “gender is the social meaning of sex”3 (Haslanger 2000: 37; Butler 1993: 238). These are distinct statements but both notably capture the idea that the categories of woman and gender are social and thereby external. First, let’s start with Beauvoir’s analysis of what it means to be a woman. Beauvoir claims that “not every female human being is necessarily a woman … If the female function is not enough to define woman … we then have to ask: What is a woman?” (Beauvoir 1949: 3–5). Beauvoir’s question, “What is a woman?” continues to trouble feminist philosophy (Bogardus 2020; Heyes 2000; Jenkins 2016) and is central to the Definition Problem. Beauvoir frames her analysis from her position as a particular instance of the category woman by drawing on her experience as a woman (Bauer 2001: 43; Beauvoir 1949:5). For Beauvoir, being a woman is not a “natural fact”, rather, it is a result of someone’s social context and personal history. In particular, someone becomes a woman in virtue of the history of her childhood, i.e., what is socially expected of her and how she is treated by her family and community. In this sense, being a woman for Beauvoir is social and historical, which she refers to as a “woman’s situation” (Bauer 2001: 226; Beauvoir 1949). This situation is a “state of affairs” not determined by biology, but by social, legal, and economic conditions (Beauvoir 1949: 9, 21). Beauvoir also rejects the notion that women have intrinsic mental features that explain their unequal social condition, and instead, views their situation as a contingent social matter. Beauvoir takes a functional approach to sex, viewing hormonal and physiological differences between females and males as playing a functional role in differences in bodily capacity, e.g., reproductive capacity or upper body strength. Although she sees biological sex as constraining females and males in different ways, biological sex does not account for the social situations of woman and man. Haslanger (2000) takes a different approach to endorsing the sex/gender distinction. Unlike Beauvoir, Haslanger’s (2000) inquiry begins with the question “What is gender?” and notes that the question “What is it to be a man or a woman?” is related to this first question (Haslanger 2000: 32). Rather than taking a dual descriptive and normative approach to answering “the woman question” as Beauvoir does, Haslanger (2000, 2004, 2005, 2006) adopts an “analytical/ameliorative” approach to defining gender in terms of hierarchical

378

Sex and Gender

definitions of the categories woman and man. An ameliorative analysis of a concept deviates from standard definitions, and instead, offers a new concept for particular aims. Haslanger’s (2000) account construes gender in terms of social class and incorporates differentials in social and economic status between women and men, ultimately, to eliminate those differentials. On her view, someone’s presumed sex (based on observed bodily features) marks them as a target for hierarchical social treatment. According to Haslanger (2000), women are sexually marked as subordinate along an economic, political, legal, or social dimension. Women’s “presumed” sex marks them as a target for systematically subordinated treatment, whereas, men’s “presumed” sex marks them as a target for privileged treatment. On this view, if women were defined as merely those persons observed to be female, then such a definition would not be able to capture what Haslanger argues are systematic differentials in social, political, economic, and legal resources. The aim of Haslanger’s analysis is to define women and men as oppressed or privileged in order to target those social and political properties for ameliorative purposes. Beauvoir and Haslanger both understand women and men to occupy social positions that depend on, at least, perceived sex difference yet are distinct from being female or male. Being born female does not cause someone to have a subordinate social position, rather the ways that we value persons creates unequal social statuses or hierarchical social classes. Beauvoir and Haslanger take an externalist approach to the sex/gender distinction where sex is biological and gender (understood as woman and man) is social (or social-historical), economic, political, and legal. The externalism I attribute to Beauvoir and Haslanger, based in an understanding of gender as social role or social position, exemplifies the classical framing of the sex/gender distinction. Beauvoir’s and Haslanger’s theories face the Exclusion Problem. Beauvoir’s theory is generally considered to do so since she offers a general account of the situation of woman as Other, which arguably leaves out differences among women along the dimensions of race, class, or geographical location. Haslanger is criticized for offering definitions of woman and man based on observed or imagined sex characteristics. According to critics, Haslanger’s view is trans-exclusive (Jenkins 2016). Next, I’ll survey antiessentialist positions about sex from externalist frameworks.

25.4

Sex and Anti-Essentialism

Anti-essentialist accounts of sex take externalist (Beauvoir 1949; Butler 1990, 1993) and contextualist approaches (Alcoff 2006; Bettcher 2013; van Anders 2022). Although it is possible to take a purely internalist approach to sex, which would appeal to self-identification only, such perspectives are not well-represented in the literature. Most anti-essentialist perspectives that include self-identification or first-person experiences of embodiment also integrate externalist views. For this reason, I’ll survey anti-essentialist approaches by first characterizing externalist accounts before contrasting these approaches with contextualist views. Technically, contextualist approaches are a type of externalist approach, but they differ in scope. Contextualist approaches tend to localize explanations of sex or gender to individuals or subgroups; whereas, externalist approaches posit more general explanations. It is worth noting that most social constructionist (Griffith, this volume) positions about sex are putatively anti-essentialist, while some social constructionist approaches are regarded as essentialist (Irigaray 1985, 1993). Anti-essentialist views on sex typically reject the idea that there are definitive biological properties which constitute an individual’s sex. These anti-essentialist perspectives counter biological essentialist approaches which propose that sex is fundamentally biological.

379

Esther Rosario

Such anti-essentialist accounts, beginning in the philosophical tradition with Monique Wittig (1982) and Judith Butler (1990, 1993), challenge the classical feminist framing of sex. Rather than proposing that a category female exists independently of social norms and social/ linguistic practices and that the term “female” merely picks out a group of people with biological characteristics, these perspectives challenge the very idea that sex is natural and prior to language and culture. In what follows, I’ll contrast Beauvoir’s anti-essentialist perspective on sex with Butler’s, which will set the stage for an analysis of contextualist perspectives. Drawing on Wittig’s “The Category of Sex”, Butler (1993) develops an account of sex that is socio-linguistic, which builds on their (1990) account of gender as a performance. What sets Beauvoir’s externalism apart from Butler’s is the latter’s emphasis on the function of language in producing a phenomenon through repeated representation in speech. On Butler’s socio-linguistic approach, gender norms constitute sexes as the natural foundations that serve to legitimate and “reproduce” binary gender (1993: 12–3). For Butler, sex is socio-linguistic in the sense that it is a regulatory ideal that “materializes” and “naturalizes” gender in the human body through discourse and repetitive practices. From this anti-essentialist perspective, sex is socially constructed in a way that aims to buttress the compulsory performance of restrictive binary gender norms. In this way, biological sexes are produced through rule-governing discursive practices that secure binary gender categories (1993: 23). Through discourse bodies are materialized as natural surfaces to fit gender categories, but for Butler that fit is never wholly successful. Gender establishes sex as its prelinguistic and pre-socio-political foundation. But if sex really is, let’s say, the enforcement branch of a binary gender ideology, then the sex/gender distinction is a mechanism that naturalizes the gender system too. Beauvoir’s and Butler’s analyses of the sex/gender distinction were arguably the most ground-breaking and influential philosophical contributions to this topic in the 20th century, and appear to have informed many 21st century responses to this issue. Social neuroendocrinologist Sari van Anders (2022) presents what I call a “contextualist” anti-essentialist account of sex. Similar to Ásta’s (2011, 2018) conferralism (Vaidya and Wallner, this volume), contextualist approaches are informed by some interplay of (external) social norms and socio-linguistic practices with (internal) self-identification. The particularities of each interplay depend on the individual involved and their social and personal circumstances in time and place. The upshot is that sex has multiple biological, social, and social-psychological constitutive features and in each case some features may be more salient than others.4 Van Anders offers a contextualist analysis insofar as she views sex as involving “biological/evolved”, “biomaterial”, and “bodily/physical” constitutive elements that are located in a “within-bodily outline” (2022: 3). For van Anders, questions about what sex is must also be accompanied by questions about context. It is worth mentioning that van Anders (2022) does not endorse a clear demarcation between sex and gender. Instead, she views sex, gender, and sexuality to be partially coconstitutive. Van Anders (2013) also sees sex and gender as sometimes invoked or operating in a composite, which she labels “gender/sex” (202). Her definition of sex incorporates externalist elements in its biological/evolved features (endogenous biological properties) and biomaterial features (biological properties that arise from exogenous influences). In addition, van Anders’ account integrates self-identification with bodily/physical features, which she describes as objects, substances, or practices that shape or modify “bodily sex” as the individual understands it (2022:3). The self-identification component of her account becomes clearer when we consider her analysis of the location of sex. Recall that biological

380

Sex and Gender

determinists do not argue that sex has a location, rather they assume there are some universal properties that are instantiated on the basis of a more fundamental property that all females and males share, respectively. Those fundamental biological properties could be a metabolic disposition or a microstructural feature, i.e., XX or XY chromosomes. These essential features have a “location” in the sense that they are parts of a human organism but biological essentialist theses of sex-based properties are usually theorized in a way that aims to shed light on what can be said of females, males, or humans in general rather than specified in terms of what can be said about a particular individual’s attitude or circumstances. Van Anders’ within-bodily outline conceptualizes sex within a mind-body boundary that involves diverse mental, physical, and behavioral phenomena that include acting, responding, sensory processing, believing, perceiving, and imagining (2022: 3). According to van Anders, the advantages of her account are that the within-bodily outline doesn’t universalize or homogenize sex. Rather, through incorporating self-identification, her position represents a range of individuals’ first-person embodied experience.

25.5 The Sex/Gender Distinction: Second Pass In our first pass, we looked at the classical externalist framing represented in Beauvoir’s existential-phenomenological theory and Haslanger’s ameliorative analysis. In the wake of Butler’s challenge to the sex/gender distinction and its influence on social constructionist critiques of sex as biological, another pass at the distinction is in order. In addition to Butler’s challenge, the critical impact of intersectional analyses developed within Black feminist theory (hooks 1984; Lorde 1984; Collins 1990; Crenshaw 1989, 1991) continues to influence accounts of gender. Intersectional theories analyze how social position, social identity, and inner-life experience are shaped by social factors including gender, race, sexuality, class, age, and disability. Generally, for intersectional approaches, offering accounts of gender that do not incorporate other social factors yield dubious and incomplete analyses. Arguably, van Anders (2022) offers an intersectional treatment of sex and gender, but recall that she does not view sex as fundamentally biological. For those theories that take the biological essentialist route, offering an intersectional account of sex seems less plausible. From the late 20th century onwards, intersectional, trans, and genderqueer perspectives featured more centrally in feminist philosophical accounts of sex and gender (Stoljar 1995; Butler 2004; Heyes 2000; Saul 2006, 2012; Alcoff 1988, 2006; Bettcher 2007; Diaz-Leon 2016Bettcher 2007) which aimed to “solve” the Exclusion Problem faced by feminist essentialists (Haslanger 2000) and by some anti-essentialist endorsements of the sex/gender distinction (Beauvoir 1949). Butler’s (1990) conception of gender as a series of repeated speech acts helped pave the way for the integration of trans philosophy and queer theory in what would become mainstream feminist philosophy, but even their theory faces problems with accounting for diversity along van Anders’ schema. For example, Butler’s view does not appear to account for transgender or genderqueer identities that are not produced through the performance of binary gender norms. Butler (1990, 1993, 2004) addresses this issue by arguing that regulatory ideals inevitably create points of disruption and resistance, but it is debatable whether a more robust psychological component is needed to complement Butler’s theory. In light of such problems with externalist approaches, Katharine Jenkins (2016) offers a trans-inclusive analysis of gender as class, drawing on Haslanger’s (2000) account, and as identity. Jenkins’ view attempts to solve the Exclusion Problem that Haslanger’s account faces. Recall that Haslanger’s (2000) ameliorative account of woman as class excludes some

381

Esther Rosario

trans women from the definition of woman by requiring that a woman be “regularly and for the most part observed or imagined to have certain bodily features presumed to be evidence of a female’s biological role in reproduction” (Haslanger 2000: 42). Jenkins builds on Haslanger’s externalist approach with her inclusion of gender as class, but supplements it with an internalist position that involves a notion of gender identity (Jenkins 2016: 410). She defines gender identity as the “internal map” that allows one to navigate gender class structures. Even though the general idea of an internal map may be promising for transinclusive approaches, some object that Jenkins’ (2016) internalist framework is needlessly strong (Brigandt and Rosario 2020). The reason being that Jenkins’ internal map includes a relevancy criterion. According to this criterion, an individual has the gender identity woman if she takes feminine norms to be relevant to herself (even if she does not comply). This criterion makes Jenkins’ account vulnerable to the Exclusion Problem because it assumes that, e.g., a woman needs to know which gender norms are relevant to her. The worry is that gender norms are culturally specific, and in principle, one can have the experience of being a woman without believing that feminine gender norms are relevant to oneself. In addition, persons with some cognitive disabilities appear to be excluded from having the internal map as an awareness of gender norms is required. As several philosophers have raised recently, internalist accounts of gender, especially those that treat self-identification as necessary and sufficient, appear to be too weak or even incoherent (Stock 2021; Barnes 2022; Bogardus 2022; Lawford-Smith 2022).

25.6

Context, Dispositions & Identity: Essentialist & Anti-Essentialist Perspectives on Gender

In this section, I’ll take a closer look at contextualist theories of gender (Heyes 2000), dispositional (McKitrick 2015), and other essentialist theories of gender (Witt 2011; Jenkins 2016; Byrne 2020) as well as another pass at theories of gender identity (Bettcher 2013; Barnes 2022; Bogardus 2022). Heyes (2000) argues for an account that treats gender as Wittgensteinian (Hamawaki, this volume) family resemblances. She is skeptical of the purported clear divide between essentialist and anti-essentialist accounts of gender, and assesses the limits of anti-essentialist orthodoxy in feminist theory as well as its critiques. In so doing, she offers an account that is a “middle ground” between essentialism and antiessentialist theories wary of generalizations about women (Heyes 2000: 67–72). Heyes’ family resemblance approach is contextualist because it looks to examples of contexts where diverse claims about women can be made by identifying commonalities without assuming that women are all the same. Heyes describes her project as anti-essentialist: Wittgenstein’s anti-essentialism recommends that we “look and see”, … Instead of trying to “get it right” about who women are, we can give examples of contexts where different claims are justified. … I take up the challenge of giving such examples, using my Wittgensteinian feminism to develop an anti-essentialist feminist research method … (Heyes 2000: 75–6). but in ways similar to van Ander’s contextualist analysis, Heyes’ theory of gender could be described as a type of essentialism that defines gender in terms of a range of constitutive features that not all members share in common. For Heyes, when we observe how language is used in practice we can look for connections in deployments of gender terms, and see how, for example, boundaries about who women are are drawn in different contexts.

382

Sex and Gender

Jennifer McKitrick’s (2015) dispositional analysis offers a view of gender as relational and psychosocial. McKitrick does not offer a purely internalist account; rather, she integrates externalist features by positing that dispositions toward gender-based behaviors are products of being part of a social environment. Gender dispositions change across time and place, and are instantiated in various degrees. Although flexible and changeable, gender dispositions are stickier than habituated patterns of behavior. McKitrick interprets her dispositional analysis to accommodate subjective experiences of gender in ways Butler’s performativity does not. Presumably, the reason McKitrick’s internalist picture is better equipped to address conflicts between one’s first-person experiences of gender and the gender norms in one’s social setting is because dispositions don’t require manifestation or external recognition to be what they are. What secures someone’s being a woman even when she is not observed to exemplify the traits associated (or stereotyped) with woman is that she has “sufficiently strong and sufficiently many” dispositions constituting a feminine gender identity even when not manifested (McKitrick 2015: 9). Although McKitrick aims her dispositional account to provide a more robust framework for one’s gender psychology in defining gender, a significant question arises: “How do we draw lines around women’s, men’s, and genderqueer behavior without simply invoking stereotypes or appealing to intrinsic gender essences?” To this question, McKitrick’s view offers two responses. Firstly, acquiring gender-based dispositions is a complex interactive process between an individual and their social environment. Our habits, responses, and desires are shaped by our social contexts; yet, as subjects we are often (but not always) in a better position to know what we are disposed to do and desire even when we are not outwardly expressing it. McKitrick seems to grant some degree of first-person authority to gender dispositions if one is aware of them (2015: 5,10). Secondly, since she views gender as subject to change, relational, and psychologically integrated, one’s dispositions take shape in their respective culture and linguistic community but it is (at least to some extent) up to them whether or how to express those dispositions given their circumstances. Moreover, there isn’t a single set of dispositions that is necessarily associated with each gender category. So, what counts as feminine, masculine, or genderqueer will depend on the norms of one’s social location, and importantly, how one interacts with them. In this way, this account appeals to relational essences that lack indexed contents about what women’s dispositions or “feminine gender identity” looks like in every case. This approach contrasts with other essentialist views like Haslanger’s (2000) and Charlotte Witt’s (2011) uniessentialist view that gender is the unifying essence of social individuals who occupy various social roles (Witt, this volume). On Witt’s account, gender is the “mega social role” that unifies constituent roles of the social individual. Haslanger’s and Witt’s externalist approaches necessitate that woman or gender exemplify particular characteristics. A narrower and more contentious contrast to McKitrick’s dispositional analysis is Alex Byrne’s claim that “Women are adult human females” (2020: 3794–5). Robin Dembroff offers an extensive reply to Byrne’s biological essentialist view, which they label the “natural attitude” (2021: 19–20). The natural attitude about gender makes several assumptions, viz., that there are only two genders, that all persons have one or the other, and that this picture is supported by the biological sciences. Byrne’s analysis of gender can be understood as endorsing the natural attitude. On the flip side, McKitrick’s internalist view of gender might side-step the Exclusion Problem depending on what being aware of one’s own gender dispositions involves exactly. In addition to having advantages over other essentialist accounts, McKitrick’s appears to go

383

Esther Rosario

beyond Jenkins’ account of gender identity by positing a less rigid relationship between the navigating subject and their social environment. From a feminist disability perspective, Elizabeth Barnes (2022) raises a worry concerning the popularization of gender identity, particularly via self-identification, as the unique criterion for gender. Although she views gender identity as important to theorizing about gender categories and crucial to the lived experiences of many, she argues that gender identity should not be the sole determining criterion nor should an account of gender exclude the lived experiences of cognitively disabled persons. Barnes’ view can be contrasted with Talia Mae Bettcher’s (2009, 2013) theory of gender as sincere self-identification and with Jenkins’ claim that the labels “genderqueer”, “woman”, and “man” should be reserved for those who selfidentify with the genders those terms refer to. As a result, the externalist aspect of Jenkins’ view includes all disabled persons but it does not do so in its internalist framing. From a different angle than Barnes, Tomas Bogardus (2019, 2020, 2022) also rejects defining gender in terms of self-identification, pointing out the circularity that results. Like Byrne, he favors a biological essentialist account of gender, and challenges those who endorse the sex/gender distinction to show why gender is social rather than merely a way of referring to biological sex. For Bogardus (and Byrne), gender is a set of linguistic features that serves to pick out members belonging to female and male natural kinds. Unlike Barnes’ challenge to gender as (solely) self-identification, Bogardus’ contributions to this recent phase of the distinction debate have been largely negative, putting the onus on social constructionists to show why gender isn’t just linguistic.

25.7

The Problems

The distinction and essentialism/anti-essentialism debates continue. Contemporary antiessentialists have rightly pointed out problems with feminist essentialist approaches, but anti-essentialist accounts can seem overly fuzzy and ad hoc. Even the most promising postButlerian views (Ásta 2018, 2011; Heyes 2000; van Anders 2022) raise worries about the ability of theses focused on norms and language to account for shared material and psychological realities among persons. Barnes’ (2022) theory of gender offers a useful way to account for material conditions shared by women while being pluralistic. For Barnes, there are real and objective facts about gender but our use of gender terms doesn’t always match these objective facts (2022: 21). She argues that rather than using gender terms to match perfectly with theoretical hallmarks in the metaphysics of gender, viz., matching our use of the term “woman” with only those who identify as women, we should incorporate permissive uses of gender terms. The reason for this is that someone who does not (or is not able to) identify as a woman can still be materially impacted in ways specific to women. Barnes’ account highlights that there isn’t always a one-to-one mapping between a gender category’s extension and those disposed to certain treatment or characteristics of life history based on (perceived) sex. I stated at the outset that this chapter was concerned with whether sex and gender are essentially distinct. Noticeably, many of the positions I’ve surveyed either reject the distinction altogether or atomize the distinction by defining it in terms of local/individual cases. Caution or even skepticism about the distinction is a common feature of essentialist and anti-essentialist approaches that aren’t classically externalist. And even what seem like the most advantageous hybrid approaches (e.g., anti-essentialist contextualism, dispositional essentialism) have difficulties addressing the Problems. For instance, van Anders’ view may lump together related but diverse phenomena (sex, gender, sexuality) yet split common

384

Sex and Gender

features in a way that makes identifying explananda difficult. On the essentialist side, situating dispositional analyses in social location can make it hard to track common features across time and place. That is, very localized accounts of gender can obscure intersecting social patterns and are vulnerable to claims that feminism is no longer needed since conditions, e.g., of women in higher education, can change dramatically. Despite attempts to clearly demarcate essentialist vs. anti-essentialist views, the boundaries between them aren’t always clear since putatively anti-essentialist accounts can make essentialist claims in several ways, some of which have been surveyed here (Butler 1993; Heyes 2000; van Anders 2022). Perhaps a promising way to approach the essentialism/anti-essentialism debate, and the difficulties the Problems present, would be to defuse it by focusing on accounts of sex and gender that are sufficiently fine-grained and inclusive of individual variation while still being able to project useful generalizations.

25.8

Related Topics

Natural Kind Essentialism, Dispositional Essentialism, Social Justice, Unity, Social Construction, Conferralism, Wittgenstein

Notes 1 The Inclusion Problem relates to Haslanger’s (2000) discussion of the commonality problem, which concerns the issue of whether there is anything social that all females share in common beyond body type (if even that). 2 I shall use “genderqueer” as an umbrella term to refer to gender categories that fall outside of the categories woman and man, which include non-binary, gender-nonconforming, third gender, genderfluid, and pangender categories. 3 Butler (1993) characterizes the statement, “gender is the social significance that sex assumes within a given culture” as the Beauvoirian position to the sex/gender distinction (238). 4 Van Ander’s view could be characterized as essentialist because she appeals to sex having certain constitutive features, but it appears that she takes her account to be anti-essentialist.

References Ainsworth, C. (2017) “Sex and the Single Cell.” Nature 550:S6–S8. Alcoff, L. (1988) “Cultural Feminism versus Post-Structuralism: The Identity Crisis in Feminist Theory.” Signs 13:405–436. Alcoff, L. M. (2006) Visible Identities: Race, Gender, and the Self. New York: Oxford University Press. USA. Arnold, A. P. (2009) “The Organizational–activational Hypothesis as the Foundation for A Unified Theory of Sexual Differentiation of All Mammalian Tissues.” Hormones and Behavior 55 (5): 570–578 Ásta. (2011) “The Metaphysics of Sex and Gender.” In C. Witt (ed.), Feminist Metaphysics. Dordrecht: Springer. pp. 47–65. Ásta. (2018) Categories We Live By. Oxford: Oxford University Press. Barnes, E. (2022) “Gender without Gender Identity: The Case of Cognitive Disability.” Mind 131 (523): 836–862. Baron-Cohen, S. (2003) The Essential Difference: The Truth about the Male and Female Brain. New York: Basic Books. Bauer, N. (2001) Simone de Beauvoir, Philosophy, and Feminism. New York: Columbia University Press. Beauvoir, S. d. (1949) The Second Sex. Translated by C. Borde and S. Malovany-Chevallier. New York: Knopf. 2010.

385

Esther Rosario Bettcher, T. M. (2007) “Evil Deceivers and Make-Believers: On Transphobic Violence and the Politics of Illusion.” Hypatia 22(3):43–65 Bettcher, T. M. (2009) “Trans Identities and First-Person Authority.” In L. Shrage (ed.), You’ve Changed: Sex Reassignment and Personal Identity. Oxford: Oxford University Press. Bettcher, T. M. (2013) “Trans Women and the Meaning of ‘Woman’.” In A. Soble, N. Power, and R. Halwani (eds.), Philosophy of Sex: Contemporary Readings. Sixth Edition. L Lanham: Rowman & Littlefield, pp. 233–250 Bogardus, T. (2019) “Some Internal Problems with Revisionary Gender Concepts.” Philosophia 48 (1): 55–75. Bogardus, T. (2020) “Evaluating Arguments for the Sex/Gender Distinction.” Philosophia 48 (3): 873–892. Bogardus, T. (2022) “Why the Trans Inclusion Problem Cannot be Solved.” Philosophia 50 (4): 1639–1664. Brigandt, I., and Rosario, E. (2020) “Strategic Conceptual Engineering for Epistemic and Social Aims.” In H. Cappelen, D. Plunkett, and A. Burgess (eds.), Conceptual Engineering and Conceptual Ethics. Oxford: Oxford University Press. Brizendine, L. (2006) The Female Brain. New York: Three Rivers Press. Brizendine, L. (2010) The Male Brain. New York: Broadway Books. Butler, J. (1990) Gender Trouble: Feminism and the Subversion of Identity. New York: Routledge. Butler, J. (1993) Bodies That Matter: On the Discursive Limits of “Sex”. New York: Routledge. Butler, J. (2004) Undoing Gender. New York: Routledge. Byrne, A. (2020) “Are Women Adult Human Females?” Philosophical Studies 177 (12):3783–3803. Collins, P. H. (1990) Black Feminist Thought: Knowledge, Consciousness, and Politics of Empowerment. Boston: Unwin Hyman, pp. 221–238. Crenshaw, K. (1989) “Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique of Antidiscrimination Doctrine, Feminist Theory and Antiracist Politics.” The University of Chicago Legal Forum 140:139–167. Crenshaw, K. (1991) “Mapping the Margins: Intersectionality, Identity Politics, and Violence Against Women of Color.” Stanford Law Review 43 (6):1241–1299. Darwin, C. R. (1871) The Descent of Man, and Selection in Relation to Sex. London: John Murray. Dembroff, R. (2021) “Escaping the Natural Attitude About Gender.” Philosophical Studies 178 (3): 983–1003. Diaz-Leon, E. (2016) “Woman as a Politically Significant Term: A Solution to a Puzzle.” Hypatia 31:245–258. Fausto-Sterling, A. (1992) Myths of Gender: Biological Theories About Women and Men, Revised Edition. New York: Basic Books. Fine, C. (2010) Delusions of Gender. New York: Norton. Geddes, P. and Thompson, J. A. (1889) The Evolution of Sex. London: Walter Scott. Gurian, M., and Stevens, K. (2005) The Minds of Boys: Saving our Sons from Falling Behind in School and Life. San Francisco: Jossey-Bass. Haslanger, S. (2000) “Gender and Race: (What) Are They and (What) Do We Want Them To Be?” Nous 31 (1):31–55. Haslanger, S. (2004) “Future Genders? Future Races?” Philosophic Exchange 34 (1):1–24. Haslanger, S. (2005) “What Are We Talking About? The Semantics and Politics of Social Kinds.” Hypatia 20 (4):10–26. Haslanger, S. (2006) “What Good Are Our Intuitions: Philosophical Analysis and Social Kinds.” Aristotelian Society Supplementary 80 (1): 89–118. Haslanger, S. (2012) “Social Construction: Myth and Reality.” Resisting Reality: SocialConstruction and Social Critique. Oxford: Oxford University Press, pp. 183–220. Heyes, C. J. (2000) Line Drawings: Defining Women through Feminist Practice. Ithaca: Cornell University Press. hooks, b. (1984) Feminist Theory from Margin to Center. Cambridge, MA: South End Press. Irigaray, L. (1985) This Sex Which is Not One. Ithaca: Cornell University Press. Irigaray, L. (1993) An Ethics of Sexual Difference. Ithaca: Cornell University Press. Jenkins, K. (2016) “Amelioration and Inclusion: Gender Identity and the Concept of Woman.” Ethics 126:394–421. Kourany, J. (2010) Philosophy of Science After Feminism. Oxford: Oxford University Press.

386

Sex and Gender Lawford-Smith, H. (2022) Gender-Critical Feminism. Oxford: Oxford University Press. Lorde, A. (1984) Sister Outsider: Essays and Speeches. Trumansburg, NY: Crossing Press. Malebranche, N. (1674/1997) The Search After Truth: With Elucidations of the Search After Truth. T. A. Lennon and P. J. Olscamp (Eds.). Cambridge: Cambridge University Press. McKitrick, J. (2015) “A Dispositional Account of Gender.” Philosophical Studies 172 (10):2575–2589. Meynell, L. (2012) The Politics of Pictured Reality: Locating the Object from Nowhere in FMRI. In R. Bluhm, A. J. Jacobson, and H. L. Maibom (Eds.), Neurofeminism: Issues at the intersection of feminist theory and cognitive science. New York: Palgrave Macmillan, pp. 11–29. Mikkola, M. (2017) “Gender Essentialism and Anti-Essentialism.” In A. Garry, S. J. Khader, and A. Stone (Eds.), The Routledge companion to feminist philosophy. New York: Routledge, pp. 168–179. Mikkola, M. (2022) “Feminist Perspectives on Sex and Gender.” Stanford Encyclopedia of Philosophy. Section Three: Problems with the Sex/Gender Distinction https://plato.stanford.edu/entries/feminismgender/ Saul, J. (2006) “Gender and Race.” Aristotelian Society Supplementary 80:119–143. Saul, J. (2012) “Politically Significant Terms and Philosophy of Language: Methodological Issues.” In Out from the Shadows: Analytical Feminist Contributions to Traditional Philosophy, edited by S. Crasnow and A. Superson. Oxford: Oxford University Press, pp. 195–216. Stock, K. (2021) Material Girls: Why Reality Matters for Feminism. London: Fleet. Stoljar, N. (1995) “Essence, Identity, and the Concept of Woman.” Philosophical Topics 23 (2): 261–293. Stone, A. (2004) “Essentialism and Anti-Essentialism in Feminist Philosophy.” Journal of Moral Philosophy 1 (2):135–153. Taylor Merleau, C. (2003) “Bodies, Genders and Causation in Aristotle’s Biological and Political Theory.” Ancient Philosophy 23 (1):135–151. The Gurian Institute, Bering, S., & Goldberg, A. (2009) It’s a Baby Girl! The Unique Wonder and Special Nature of Your Daughter from Pregnancy to Two Years. San Francisco: Jossey-Bass. van Anders, S. M., Schudson, Z. C., Abed, E. C., Beischel, W. J., Dibble, E. R., Gunther, O. D., Kutchko, V. J., & Silver, E. R. (2017) “Biological Sex, Gender, and Public Policy.” Policy Insights from the Behavioral and Brain Sciences 4 (2):194–201. 10.1177/2372732217720700 Van Anders, S. (2013) “Beyond Masculinity: Testosterone, Gender/Sex, and Human Social B Behavior in a Comparative Context.” Frontiers in Neuroendocrinology 34 (2013):198–210. Van Anders, S. (2022) “Gender/Sex/ual Diversity and Biobehavioral Research.” Psychology of Sexual Orientation and Gender Diversity. Advance online publication. https://doi.org/10.1037/sgd0000609 Witt, C. (1995) “Anti‐Essentialism in Feminist Theory.” Philosophical Topics 23 (2):321–344. Witt, C. (2011) “What is Gender Essentialism?” In Feminist Metaphysics. Dordecht: Springer, pp. 11–25. Wittig, M. (1982) “The Category of Sex.” Feminist Issues. Springer. 2:63–68.

387

26 SOCIAL JUSTICE Natalie Stoljar

26.1

Introduction

This chapter identifies connections between social justice and philosophical conceptions of essence. There is no literature devoted specifically to this topic, so the chapter offers a preliminary exploration of where these two issues intersect. In broad terms, the goal of social justice is the eradication of oppression. Oppression is defined as a group-based phenomenon; people suffer the harm of oppression in virtue of their membership in social groups like genders or races (Frye 1983; Young 1990; Cudd 2006). Oppression thus corresponds to a hierarchical social structure in which certain groups (e.g., women) are subordinated and other groups (e.g., men) are privileged (Cudd 2006: 25). Because the social justice literature often understood “essence” to correspond to “biological essence”, essentialism was usually thought to be contrary to social justice. Feminists and anti-racists advocated a form of anti-essentialism that repudiated biological essences. This was important to uphold a social justice “debunking project” which claims that, although categories like gender and race might seem to track essential, natural or biological characteristics, in fact they are the result of contingent social practices that benefit dominant groups and subordinate others (Haslanger 2003: 318). However, discussions of the relation between essence and social justice are not confined to questions of biological essentialism. It has been observed that many forms of essentialism are compatible with emancipatory goals (Mason 2021: 3990–1). For instance, the social groups that are presupposed in the analysis of oppression might themselves have essences (Epstein 2019, 4904). Indeed, articulating the essences of social categories potentially plays a role in the debunking project. Social reality is a “disorganized jumble”; identifying the categories that are significant for social justice and articulating their essences can reveal how social reality entrenches oppressive hierarchies (Haslanger 2000: 35). It has also been noted that there are different forms of essentialism that potentially have different implications. For instance, the question of whether social groups have unifying essences should be distinguished from the question of whether genders, races, etc. are essential properties of individual members of these groups (Stoljar 1995). It is possible that, whereas some forms of essentialism could be inimical to social justice, others might not be.

388

DOI: 10.4324/9781003008750-30

Social Justice

Each section of the chapter identifies a different area of overlap between the issues of essence and social justice. The first focuses on the social groups assumed in definitions of “oppression”. Activism aimed at eradicating oppression can be pursued by or on behalf of social groups. How should we understand social groups and, in particular, could social groups have essences? Section 26.2 turns to groups that are characterized in terms of people’s shared identifications with a social category—genders, classes, races, disability groups, etc. There is a debate over whether feminism, anti-racism or other political action on behalf of these groups requires that they are metaphysically unified. Are essences needed to unify these social groups? Section 26.3 turns to the question of individual essence: could gender, race, sexual orientation, etc. be essential properties of the individual members of social groups? If so, should these essences be articulated in modal, non-modal or functional terms? Section 26.4 discusses the role of beliefs about essences in sustaining “ideological” forms of oppression. Unlike repression that is “forced on people through coercive measures”, ideological oppression can be “enacted unthinkingly or even willingly by the subordinated or privileged”, which makes it particularly intractable (Haslanger 2017: 149). For instance, people often unthinkingly endorse stereotypes that falsely attribute natures or essential properties to members of stereotyped groups (Haslanger 2011; Leslie 2017). The last section briefly reviews how one model of essence, the definitional model (Fine 1994), has been employed in recent discussions of gender. On this model, definitions articulate the natures or essences of the things they define. It has been claimed that positing essences in this way privileges those who are included as members of a category while excluding others. This will undermine social justice in cases in which some people are unjustifiably included or excluded thereby reinforcing their marginalization (Jenkins 2016: 398–402).

26.2 Social Groups and Essences The notion of a social group is significant for social justice in two different ways. First, social justice activism is usually undertaken by or on behalf of a social group. Sometimes it is pursued by social groups that are organized around a common political goal, for instance, environmental groups, human rights organizations, neighbourhood activist groups, or groups at protests. Other kinds of activism are pursued on behalf of social groups—races, genders, classes, etc.—and seek to eradicate the injustice suffered by these groups even if the members themselves are not always oriented around this common goal. Second, the philosophical analysis of oppression itself presupposes the notion of a social group. On one prominent account, oppressive harm is understood as “perpetrated through a social institution or practice on a social group” while another social group “benefits from the institutional practice” (Cudd 2006: 25). These social groups have identities that exist “apart from the oppressive harm” (Ibid.). Therefore, the very ideas of social justice and the eradication of oppression presuppose the existence of social groups. The issue of essence arises in this context if the social groups themselves have essences. Katherine Ritchie proposes a broad distinction between “organized groups” (Type 1) and “feature social groups” (Type 2) (2015: 314).1 Type 1 groups, like teams, committees and courts, are characterized by their “structural-functional organization” and the presence of collective intentionality. Feature social groups such as racial or ethnic groups, or gender groups, need not have structural-functional organization or collective intentionality but rather are groups in virtue of some characteristic feature that their members possess (Ibid.). This feature can be “socially constructed”. For instance, Sally Haslanger argues that the concepts

389

Natalie Stoljar

of gender and race should be constructed or stipulated to serve the “fight against injustice” (Haslanger 2000: 36; Griffith, this volume; Mallon, this volume; Rosario, this volume). On her view, it is theoretically and normatively useful for purposes of social justice to posit that (roughly) someone is a woman if and only if they occupy a subordinated position in a social hierarchy and are “marked” for this treatment on the basis of real or imagined female bodily features (Haslanger 2000: 39). People who have this socially constructed property would be members of the feature social group woman. It is possible for both types of social group to have essences (Epstein 2019: 4904). Type 1 groups like human rights organizations or people at a protest might have essential origins, for instance “having been formed by a particular action” (Ibid.; Robertson Ishii, this volume). Also, the features that characterize Type 2 groups could be essences: the feature of being subordinated on the basis of real or imagined female sex could be essential to belonging to the social group woman. Some social groups could have “their members essentially” meaning “that they do not persist through changes of members” (Epstein 2019: 4904). A married couple for instance could comprise a social group for which the members are essential to the group’s existence (cf. Fine 2020: 81). On Ritchie’s view, while “[m]embers of racial groups, gender groups, sexual orientation groups, etc. do not share a natural essence, [and they] cannot be identified with natural kinds”, they are nevertheless social kinds (2015: 316–7; Tahko, this volume).2 Rebecca Mason employs Fine’s non-modal or definitional conception of essence to argue that social kinds, including genders and races, have essences (Mason 2021; Correia, this volume; Torza, this volume). On the definitional conception, “the essential properties of an object are those of its properties that are part of the object’s ’definition’”; that is, providing definitions can be thought of as articulating the real essences of things, namely their “nature or what it is to be that thing” (Robertson Ishii & Atkins 2023, Section 26.2; Correia this volume). Mason argues that social kinds are not “mind-dependent simply because they are modally correlated with our mental states”; rather, they are defined in part as being mind-dependent and hence mind-dependence is part of the nature or essence of social kinds (2021: 3980). There is a further debate over whether genders or races should instead be characterized as natural kinds. In contrast to Ritchie and Mason, Theodore Bach argues that gender and other social categories are “mind-independent structures” that have natural rather than social essences (Bach 2016: 180). He claims that Haslanger’s stipulated social categories are empirically inadequate and therefore politically inadequate—there should be instead a “tight relationship between the correct tracking of causal-explanatory phenomena [empirical adequacy] and moral outcomes” (Bach 2016: 186). On his view, because Haslanger’s stipulated social kinds do not track causal-explanatory phenomena, her account obscures empirical and political possibilities such as “the mutability of social categories like men and women” (Bach 2016: 195). To meet the empirical adequacy requirement, Bach argues that gender should be understood as a historical natural kind: “conscious and unconscious design mechanisms mold infants as reproductions of historical men and women from the moment they are born. Through these processes an individual comes to participate in a lineage of men or women and thereby becomes a member of the historical kind men or women” (Bach 2012: 259). However, while the idea of historical lineages as essences has some initial plausibility (especially when applied to races), it is not obvious how this would generalize to other social groups like disability or sexual orientation. Hence, Ritchie’s notion of feature social groups whose members have shared social essences may have greater explanatory power for the purposes of social justice than that of historical natural kinds. These different approaches agree, however, that it is not inimical to social justice that social groups have essences.

390

Social Justice

26.3 Essences and the Unification of Social Categories There are persuasive arguments, particularly in the feminist literature, that even if members of social groups—genders, races, classes, disability groups and sexual orientation groups—have similar features, they nevertheless do not have exactly the same feature in virtue of which they are members of the group (Stoljar 2011). That is to say, the relevant feature is not a universal. Perhaps the strongest argument that gender, races, etc. are not universals is that of “intersectionality” (Crenshaw 1991; Grillo 1995). Intersectionality arguments claim that each person is situated at the intersection of different group identifications, e.g., “African American” and “lesbian” is a different social position from “Latina” and “heterosexual”. The different social positions respectively generate different experiences and forms of oppression that are incompatible with treating gender (or race) as a unified category, and hence proponents of intersectionality argue that “there is no essence or ‘golden nugget’ that all women have as women” (Spelman 1988: 136). Indeed, the intersectionality framework rejects two possible forms of essentialism. It rejects kind essentialism because, according to intersectionality, there is no single property or set of properties (no universal or essence) that binds people into social kinds. It also rejects individual essentialism, the position that genders, races, etc. are essential properties or “nuggets” of the individual members of the kind (Marabello, this volume). In addition, intersectionality theorists articulate a positive claim that properties like gender and race are particular to each individual: “the experiences of a white woman and a Black woman are different and not just additively … Race and class can never be just ‘subtracted’ because they are in ways inextricable from gender” (Grillo 1995: 18). Each intersecting dimension of a person’s identity—their gender, race, disability, sexual orientation, etc.—affects the qualitative features of all others. (The racial identity of a Black woman with a disability is qualitatively different from that of an able-bodied Black man, and so forth.) Hence, on intersectionality views, “there is a variety of genders [which are] constructed and defined in conjunction with elements of identity such as race, class, ethnicity and nationality rather than separable from them” (Spelman 1988: 175). This conclusion raises the question of whether the relevant social groups are sufficiently unified to provide a basis for activism on behalf of the group. Or are they so heterogeneous that they fall apart or indeed do not exist? There are different possible solutions to unifying social groups. Some feminist scholars draw on Locke’s nominal essence (Schechtman, this volume), which they claim is “especially useful for anti‐essentialist feminists who want to hold onto the notion of women as a group without submitting to the idea that it is ‘nature’ which characterizes them as such” (Fuss 1989: 5). A nominal essence is often thought of as a “definition of a species or sort of thing, which people assemble in their minds out of ideas” (Epstein 2021: Section 26.1.1). Nominal essences delineate categories on the basis of similar overt features, and are useful classifications for certain purposes, but do not assume that the things that are classified together share worldly natures. Nevertheless, nominal essences have been criticized as either too restrictive to capture the differences among women and the liberatory possibilities of feminism, or too weak to provide a unifying foundation. A common objection is that because the definitions that pick out nominal essences articulate precise boundaries of categories, nominal essences inevitably will be too prescriptive, privileging those who count as members of the category and excluding others. On the other hand, it has been thought that any version of nominalism entails that genders or other social categories are fictions made up by us. They are “conceptual” not metaphysical (Mason 2021: 3986). This calls the very existence of genders and other social groups into question: while positing these social groups might be

391

Natalie Stoljar

pragmatically useful, it may not be explanatorily indispensable or ontologically required (E. Taylor 2016: 525). If this is the case, it is unclear how a notion of oppression that presupposes the existence of social groups can be sustained. Subsequent attempts to solve the problem of unifying social categories divide into metaphysical and non-metaphysical approaches. As explained above, one metaphysical approach is the position that genders (and races) are historical kinds (Bach 2012; Jeffers 2013: 405-407). On this view, it is essential to being a woman or to being Black that an individual participates in a certain (real) historical lineage. Bach claims that positing historical kinds secures sufficient unity while at the same time allowing diversity within social groups: “replicative processes … explain why gendered individuals tend to share certain behavioral, psychological, and social properties … [but also allow] individuals who fail to exemplify characteristic gender properties—who fail to be masculine or feminine, or privileged or subordinated—to retain their gender status” (Bach 2016: 193). Similarly, historical essences could provide a metaphysical foundation of racialized kinds. An alternative though non-essentialist metaphysics employs the notion of resemblance nominalism. On this view, there are real resemblances—e.g., along dimensions of selfidentification, experience, social role and political commitment—among members of gender and potentially other social groups. This resemblance structure provides a basis for relatively unified and stable gender (or other) groups without a commitment to a kind essence or universal (Stoljar 1995; Stoljar 2011). Other proposals are non-metaphysical. Tommie Shelby proposes that Black political solidarity requires a notion of Black “collective identity”—but this is constituted by historical and cultural identifications that he argues should be cultivated by the Black community precisely to promote emancipatory goals (Shelby 2002). Iris Marion Young suggests that people can be “passively” united into a “series” according to the objects their actions are oriented around; self‐conscious commitment to common goals is not required. For instance, people waiting in line at a bus stop are passively united in this way (Young 1997). Mari Mikkola argues that feminist activism does not require a “thick” concept of woman that picks out a metaphysically unified class or kind (Mikkola 2016: 44). She claims that talking about women as a class is necessary for emancipatory feminism: “we cannot undertake practical and explanatory tasks without using certain terms … [W]e need the term ‘woman’ to demand legal abortions and to close the gender pay gap … ” (2016: 116). But the class woman is fixed by the “extensional intuitions” of ordinary speakers (2016: 106 ff.). In other words, all and only the individuals that ordinary speakers label “women” comprise the class woman; hence the category may be very heterogenous and include people with significantly different experiences and political goals. On Mikkola’s view, this does not matter, because it is a mistake to think that feminist activism is concerned with the wrongs suffered by women as women. Rather, it is concerned with the wrongs suffered by women as human beings. The notion of dehumanization, not a “thick” concept of woman, “underpins the political and normative claims of emancipatory feminism” (2016: 10). This argument therefore also calls into question the idea of oppression as a group-based phenomenon.

26.4 Social Justice and Individual Essences There is a third set of issues connecting social justice and essence—that of individual essences (Marabello, this volume). People often report that their social group identifications (their gender, race, sexual orientation, disability, etc.) are authentic or essential components of their sense of self (e.g., Rogan 2016). Issues of social justice arise if failing to recognize people’s

392

Social Justice

own conception of their authentic selves constitutes disrespect or reinforces the marginalization of an already marginalized group (Jenkins 2016). Could social properties like gender, Blackness or sexual orientation be essential properties of individuals? I briefly outline different possible approaches to answering this question: a modal conception, a non-modal or definitional conception, and individual essences as functional. As is well-known, Saul Kripke proposes a modal account—an account of the conditions for transworld identity—on which the individual essence of a human being includes their origin in the actual sperm and egg from which they develop (Kripke 1980: 106; Robertson Ishii, this volume). However, even if origin essentialism is generally correct, most agree that it does not imply that properties like gender or race are essential to individuals; one’s gender or race (especially if these are considered social properties) are not necessary outcomes of genetic origins.3 Kripke proposes also that being the same sort of thing (i.e., being of the same kind) is part of individual essence (1980: 106; Scarpati, this volume). Indeed, it is claimed to be “intuitively compelling” that individual members of sorts have the sortal property essentially, for instance, that each dog has the property of being a dog essentially (Robertson Ishii & Atkins 2023: Section 26.1).4 If the intuition is taken to be compelling also for genders, races, disabilities etc., it would follow that transitions from one of these social kinds to another (e.g., from one gender kind to another, or one disability kind to another) would result in the emergence of a new individual.5 This conclusion on its own is not contrary to social justice goals such as respecting people’s authenticity. However, there are independent reasons for thinking that social kinds are not equivalent to the sorts assumed by Kripke’s account. Unlike dogs and other animal kinds, social kinds are not usually thought to be categories classifying biological substances (cf. Byrne 2020). Neither do the standard criteria for persistence of identity through change—bodily continuity and psychological continuity—seem to be violated in cases of (e.g.,) gender transition, or transition from one disability kind to another (Loets, this volume). Thus, even if it is correct that continuing to be the same sort of thing (i.e., the same substance) is an essential property of individual members of substance kinds, there is reason to think that this is not the case for the social kinds we are discussing here. A second approach to articulating individual essences is to adopt Fine’s non-modal conception of essence (Fine 1994; Correia, this volume). As others have pointed out, some objects, particularly individual persons, “do not seem to admit so readily of definition” (Robertson Ishii & Atkins 2023: Section 26.2). Hence, it may not be feasible to attempt to articulate definitions that correspond to the individual essences of members of social groups. On the other hand, it could be said that the members of social groups are defined by their group membership, and hence that they have social natures that correspond to the features that characterize these groups. Individual Black people might have the social property of Blackness essentially, and women might have the social property of womanness essentially, and so forth. If so, losing any of these properties would thus create a new individual. However, as noted above, this conclusion conflicts with the standard idea that persons’ identities persist through change as long as bodily and/or psychological continuity is preserved. Again, this provides some reason for thinking that properties like Blackness or womanness are not part of the definitional essence of the individual members of social groups. A third form of individual essentialism employs functional essences. Charlotte Witt argues that gender is the functional essence of social individuals (Witt 2011; Witt 2011a; Witt, this volume). Just as the components of a house—its bricks, mortar, flooring and roof tiles—are unified into a separate ontological entity (the house) by the functional essence of providing shelter, so the social role of gender unifies the different social roles occupied by individuals into a distinct social individual (e.g., Witt 2011: 6–7). While, in principle, any social role

393

Natalie Stoljar

(e.g., disability, race or sexual orientation) could be the normative principle that organizes practical agency, “engendering is a necessary social function in that it is required for society to persist” (Witt 2011: 100). Hence, gender is the historically and cross-culturally dominant social role and the principle of normative unity in the organization of people’s agency. Witt’s analysis is congenial to the goals of social justice. In particular, it draws attention to the interrelations between the norms associated with social categorization and individual agency (cf. Ásta 2018; Griffith, this volume; Wallner & Vaidya, this volume). However, unification essentialism may not be compatible with certain specific social justice goals. For instance, on Witt’s account, genders are characterized in terms of an “engendering function”: e.g., people are women when they are “recognized as having a body that plays one role in the engendering function: women conceive and bear” (Witt 2011: 40). This analysis may not easily capture a plurality of genders or provide an adequate account of the practical agency of trans, non-binary or other persons whose gender does not fit the gender binary.6 As Witt herself puts it, “the notion of a third gender is compatible with my approach [only] if the third gender itself is described in relation to women and men as I have defined them. … my focus on the engendering function rules out … a third gender that is defined with no relation at all to the engendering function” (Witt 2011: 41–2). Outside of queer communities, social practices do not typically include a plurality of actual social roles (even ones that are defined in relation to the engendering function) and hence, typically, social individuals who identify as trans, genderfluid or non-binary will be recognized as occupying one of the gendered social roles on offer, namely woman or man. To characterize the practical agency of people who are trans as always centrally involving a repudiation of the gender binary may offer only a stunted account of how gender organizes their practical agency. It is also unclear whether Witt’s position can accommodate intersectionality. Recall that, on her view, gender is the dominant social role and hence the principle of normative unity in the organization of people’s agency, whereas intersectionality claims that no social role is dominant because each is “constructed and defined in conjunction with” the other social positions a person occupies (Spelman 1988: 175). A final possibility to be considered is that people’s reports that their gender, race, etc. are “essential” to their sense of self should be explained in epistemic rather than metaphysical terms. Perhaps these properties are core to people’s experience or self-conception rather than part of individual essence? Anthony Appiah observes that the question “Would I still be me?” if I were a different gender or race is both metaphysical and ethical (Tannenbaum & Glezakos, this volume). Even if the answer to the metaphysical question is “yes”, the answer to the ethical question may be “no” because it depends on how, for each person, these properties contribute to “the ethical project of composing a life for oneself” (Appiah 1990: 495). It has been argued that gender, race, etc. could be “volitionally necessary” or “circumstantially inescapable” aspects of a person’s self-conception (Stoljar 2018). Aspects of the will that are “volitionally necessary” are so significant for a person that it would be “unthinkable” to repudiate them or cease to identify with them (Frankfurt 1988: 177–90). Harry Frankfurt describes a fictional aristocrat whose conception of himself as an English gentleman makes it unthinkable for him to pursue a conversation about intimate matters with a person of a lower class (Frankfurt 1998, 183). There are also “circumstantially inescapable” aspects of self-conception “over whose presence in the person’s life and effect on her life she has no say” (Oshana 2005: 83). Marina Oshana writes that “[w]e cannot escape the racialized norms that define us, and that inform our self-concept, even where we regard these norms as alien. Consciously or not, welcome or not, one’s racialized identity contravenes upon most aspects of one’s self-conception” (Oshana 2005: 90). The position that

394

Social Justice

genders, races, etc., are volitionally necessary and/or circumstantially inescapable—that they are essential components of one’s self-conception—avoids the potentially controversial commitments of the forms of individual essentialism discussed above.

26.5

Ideological Oppression and Essentializing Beliefs

The issue of essences arises also in the context of ideological oppression. Ideological oppression is one component of the “social imaginary”, namely “the ways people imagine their social existence, how they fit together with others, how things go on between them and their fellows, the expectations that are normally met, and the deeper normative notions and images that underlie these expectations” (C. Taylor 2004: 23). It has been argued that the negative stereotypes and prejudices that make up the oppressive aspects of the social imaginary are false because they implicitly assert that biological essentialism is true of gender, race or other social categories. It is crucial for achieving social justice to identify these false beliefs about essences and to repudiate them. Stereotypes and prejudices, like women are submissive or Muslims are terrorists, are often taken to be generalizations that are expressed by means of generics—statements that attribute a trait to a category without employing the quantifiers “all”, “some” or “many” (Beeghly 2015: 677; Ritchie, this volume).7 Sarah-Jane Leslie argues that the cognitive mechanism that permits us to generalize is “primitive” and “our most basic and immediate means of obtaining information about categories” (Leslie 2007: 383–4).8 In particular, there is a category of generics—striking property generics—that attribute “harmful, dangerous, or appalling properties to the kind, … the sort of property of which one would be well served to be forewarned” (Leslie 2008: 15). Whereas the disposition to generalize in response to dangers in the physical world serves us well, Leslie argues that the same mechanism leads us to accept false generics about the social world. Further, on Leslie’s analysis, when we encounter harmful or appalling properties, we generalize in a way that essentializes the kind. That is, we tacitly posit “that there is some hidden, nonobvious, and persistent property or underlying nature shared by members of that kind that causally grounds their common properties and dispositions” (2017: 405). This may render striking property generics about the natural world true: e.g., the generic “Sharks attack surfers” is true because it is in the nature of sharks to be disposed to attack surfers and their essence in this respect is fundamentally different from that of other kinds of fish. On the other hand, in social contexts, the striking property generics that we are disposed to accept are typically false. When a person who is Muslim commits an act of terrorism, our cognitive mechanism generalizes in a way that essentializes Muslims as a kind. Muslims are not a group that is unified by a biological essence, but “there are [still] persisting beliefs that Muslims are fundamentally different from other groups of people while also being fundamentally all alike” (2017: 411). Such essentialist beliefs are false and “can be very damaging, and the groups that are most highly essentialized often face the worst forms of social prejudice” (Leslie & Lerner 2022: Section 26.5.1). Hence, one way of reducing prejudice and dismantling ideological oppression is to stop using essentializing generics in social contexts when they are false. However, many stereotypes, e.g., women are submissive, are not striking property generalizations; they do not attribute harmful or dangerous properties to a kind.9 Perhaps these should be analyzed as implicitly making objectionable normative claims, for instance that women ought to be submissive (Haslanger 2011: 200). But what if it turns out that a majority of women in fact have the tendency to be deferential to men? Is the generic statement

395

Natalie Stoljar

“Women are submissive” acceptable because it is true? Haslanger develops an analysis on which generics are criticizable (even if they are true) when they implicate, rather than assert, false claims about essences: “There is a set of problematic generics that introduce implicitly into the common ground a proposition about how … [people] are by nature or intrinsically. These cases are problematic because the proposition introduced is false” (Haslanger 2011: 193). Oppressive stereotypes can induce false beliefs about essences in hearers; e.g., the stereotype that women are submissive “invites the interpretation that it lies in the very nature of women to be submissive—rather than their submissiveness coming about as the result of extrinsic, accidental, and changeable social circumstances” (Leslie & Lerner 2022: Section 26.5.1).10 Dismantling ideological forms of oppression requires us to actively repudiate these implicatures using the device of “meta-linguistic negation” which “blocks the falsehood from entering the common ground” (Haslanger 2011: 189).

26.6

Definitions and Essences

Some philosophers who are interested in promoting social justice posit a connection between definitions and essences. We saw above that definitions can pick out nominal essences. Whereas nominal essences obtain at the conceptual or linguistic level, Fine’s definitional model of essence claims that definitions can also pick out natures or the real essences of things in the world (Fine 1994). Therefore, philosophers who might have seemed at first to be advancing only definitions at the linguistic level might also be providing accounts of real essences. Asya Passinsky suggests that the different definitions of the term “woman” offered in recent literature, such as Haslanger’s, in fact provide competing accounts of “the things themselves as opposed to definitions of our words or concepts for them” (Passinsky 2019: 7–9). Passinsky claims that interpreting debates about definitions as debates over real essences makes sense of the kinds of political explanations that are of interest to these philosophers, e.g., the question of why gender inequalities are so intractable. In particular, Haslanger’s definition explicates why gender oppression is persistent because it entails that “part of what it is to be a woman is to be systematically subordinated” (Passinsky 2021: 9).11 A common objection to definitions is that because they articulate precise boundaries, they risk both inappropriate inclusion and inappropriate exclusion (Rosario, this volume). Inappropriate inclusion or exclusion can be problematic if it is disrespectful (for instance, it fails to recognize trans people who seek inclusion in a gender category other than the one assigned at birth) or if it reinforces the marginalization of an already marginalized group (Jenkins 2016). Indeed, Katherine Jenkins argues that Haslanger’s account of gender faces the “inclusion problem” because it wrongly includes and wrongly excludes some trans people from their preferred gender categories (Ibid.: 398–402). Jenkins’ solution is that they are two different gender concepts: gender defined as a class and gender defined as an identity. These alternative definitions of gender therefore could also be interpreted as articulating competing accounts of gender essences. An alternative approach would to be to deny that the analysis of “woman” at the linguistic level entails that there is a single property or set of properties that it is essential for individuals to have to be members of the category woman. Consider the proposal that “woman” is a cluster concept (Stoljar 1995; Stoljar 2011; Brigandt, this volume; Mallon, this volume). On this view, there is a cluster of features associated with the concept—e.g., anatomical features, social roles, and lived experience associated with bodily characteristics. For an individual to be correctly attributed the property of being a woman, she must satisfy enough of the features in this complex cluster, rather than all and only these features. There is therefore no single set

396

Social Justice

of features in virtue of which an individual is a woman and no real essence of the property woman.12 (People are members of the group woman in virtue of their participation in a resemblance structure rather than their instantiation of a single property or set of properties.) Despite offering a non-essentialist account of kind membership, this approach permits an essentialist analysis at the linguistic level because each of the features of the cluster could be essential components of the concept even if they are not each essential to membership in the kind. A final topic worth noting in the context of definitional essences is the reemergence of biological essentialism regarding sex and gender (Rosario, this volume). It has been claimed that a person is a woman if and only if they are an adult human female (e.g., Byrne 2020). There are both modal and definitional versions of this claim (Mason 2022). On the modal version, “necessarily, S is a woman if and only if S is an adult human female”; on the definitional version, “‘women, by definition, are adult human females’ [which implies that] being an adult human female is part of the real definition of being a woman” (Mason 2022). Since the latter entails the former, Mason argues against both versions by arguing against the modal claim. She shows that there are non-trans women who do not have female sex characteristics (chromosomal sex, gametic sex, phenotypic sex, or some combination) as well as non-trans men who do, and, if so, the modal claim is false. If Mason is right, the notion that women are essentially adult human females is another example of a false essentializing belief that serves to sustain gender oppression. As she points out, it has been used to undermine social justice advocacy for trans people, e.g., to deny the legitimacy of their claims to be included in spaces that conform to their preferred gender identities.

26.7 Conclusion This chapter has surveyed the connections between social justice, oppression, and philosophical conceptions of essence. The definition of oppression assumes the existence of social groups such as genders, races, classes, disability groups and sexual orientation groups. As we have seen, it is not incompatible with either social justice or the eradication of oppression to attribute social or historical essences to these groups or to claim that the properties corresponding to people’s social identifications are essential properties of individual members of groups. There is in fact a wide variety of views according to which positing essences is helpful to answer questions of importance for social justice. Nevertheless, there are also many approaches that conclude that essentialism is not needed to answer these questions. Some argue that the unification of social groups can be achieved without positing essences; others that issues of individual identity and the organization of agency need not be understood in essentialist terms. Importantly however, the eradication of ideological oppression requires the repudiation of stereotypes that falsely attribute biological essences to social groups.13

26.8

Related Topics

Correia, F., “Non-modal Conceptions of Essence” Griffith, A., “Social Construction” Mallon, R., “Race” Rosario, E., ‘Sex and Gender” Marabello, M., ‘Essences of Individuals” Witt, C., “Unity” Robertson Ishii, T., “Origin Essentialism”

397

Natalie Stoljar

Notes 1 Brian Epstein criticizes Ritchie’s distinction, but I set aside his objections here ( Epstein 2019: 4901–3). 2 Even if all social groups are social kinds, not all social kinds are social groups: e.g., money is a social kind but not a social group. 3 It might be thought that the origin condition—being from the same sperm and egg—implies sex essentialism, but this is not so clear. E.g., it is not biologically necessary that a human fetus with an XY chromosome will develop with male sex characteristics. Hence, the origin condition does not seem to imply that being a female or a male is essential to the individual identity of the human organism. See discussions of the idea that genders correspond to biological sex categories in Mason 2022 and Rosario, this volume. 4 See Mackie (1994) for a challenge to the view held by Kripke and others that being in a species (or sort) is essential to being the thing that it is. 5 There are sorts for which the intuition is not plausible. Even if husbands are a sort of thing, it is not plausible that each individual that is a husband is essentially so. Perhaps genders, races etc., are more like husbands. A full discussion would require unpacking the distinction between phase sortals and substance sortals, which is beyond the scope here. (I am grateful to Kathrin Koslicki for these points.) 6 I use the term “trans”, or “transgender” in the most inclusive way possible, to refer to people whose gender identity does not align with the conventional gender binary. I intend my use of “trans” to include, for example, people who are agender, non-binary and genderfluid. Some people use the terminology ‘trans*’ instead of ‘trans’. See the definition in McKinnon 2015: footnote 1. 7 Most of the examples come from Leslie’s work or are variations on her examples. 8 There are alternative analyses of both generics and stereotypes (see Beeghly 2015), but since they do not employ essences, I don’t address them here. 9 Such stereotypes may not fit neatly into Leslie’s typology of generics. I leave aside detailed discussion here. 10 Jennifer Saul critiques the arguments due to Leslie and Haslanger that generics are ‘especially pernicious’ ( Saul 2017). She argues that the focus on generics is misguided, and that the real problem are statements (whether generic or not) that have implicatures of irrelevant associations, e.g., between being Black and criminality. 11 It should be clarified that the definitions offered by Haslanger and others do not reveal anything about the natures of the individuals who are women; they are definitions of the property “woman”. Compare a definition ‘B is a bachelor if and only if B is a man and unmarried’ and suppose this articulates not only a definition at the linguistic level, but also the nature of bachelorhood. Being a bachelor is nevertheless an accidental property of individuals who are bachelors: bachelorhood can be lost simply through a change of marital status without this entailing the emergence of a new individual. 12 Alternatively, one could claim that there is a formal essence even on a cluster concept approach, namely that the essence of being in a certain gender category is satisfying enough of the features in the cluster. 13 I am grateful to Kathrin Koslicki and Mike Raven for comments on previous drafts that considerably improved this chapter.

References Appiah, A. (1990) ‘“But Would That Still Be Me?” Notes on Gender, “Race,” Ethnicity, as Sources of “Identity”’, Journal of Philosophy, 87, pp. 493–499. Ásta. (2018) Categories We Live By. The Construction of Sex, Gender, Race, & Other Social Categories. New York: Oxford University Press. Bach, T. (2016) ‘Social Categories are Natural Kinds not Objective Types (and Why it Matters Politically)’, Journal of Social Ontology, 2(2), pp. 177–201. Bach, T. (2012) ‘Gender is a Natural Kind with a Historical Essence’, Ethics, 122, pp. 231–272. Beeghly, E. (2015) ‘What is a Stereotype? What is Stereotyping?’ Hypatia, 30(4), pp. 675–691. Byrne, A. (2020) ‘Are Women Adult Human Females?’ Philosophical Studies, 177, pp. 3783–3803.

398

Social Justice Crenshaw, K. W. (1991) ‘Mapping the Margins: Intersectionality, Identity Politics, and Violence against Women of Color’, Stanford Law Review, 43(6), pp. 1241–1299. Cudd. A. E. (2006) Analyzing Oppression. Oxford: Oxford University Press. Epstein, B. (2021) ‘Social Ontology’, in The Stanford Encyclopedia of Philosophy (Winter 2021 Edition), E. N. Zalta (ed.), < https://plato.stanford.edu/archives/win2021/entries/social-ontology/>. Epstein, B. (2019) ‘What are Social Groups? Their Metaphysics and How to Classify Them’, Synthese, 196, pp. 4899–4932. Frye, M. (1983) The Politics of Reality. Essays in Feminist Theory. Berkeley: Crossing Press. Fine, K. (2020) ‘The Identity of Social Groups’, Metaphysics, 3(1), pp. 81–91. Fine, K. (1994) ‘Essence and Modality’, Philosophical Perspectives, 8(Logic and Language), pp. 1–16. Frankfurt, H. (1988) The Importance of What We Care About. New York: Cambridge University Press. Fuss, D. (1989) Essentially Speaking: Feminism. Nature and Difference. New York: Routledge. Grillo, T. (1995) ‘Anti‐Essentialism and Intersectionality: Tools to Dismantle the Master’s House’, Berkeley Women’s Law Journal, 10, pp. 16–30. Haslanger, S. (2017) ‘Culture and Critique’, Aristotelian Society Supplementary Volume, xci, pp. 149–173. Haslanger, S. (2011) ‘Ideology, Generics, and Common Ground’, in C. Witt (ed.), Feminist Metaphysics. Explorations in the Ontology of Sex, Gender and the Self. Dordrecht: Springer, pp. 179–207. Haslanger, S. (2003) ‘Social Construction. The “Debunking” Project’, in Schmitt, F.F. (ed.), Socializing Metaphysics. The Nature of Social Reality. Lanham MD: Rowman & Littlefield. Haslanger, S. (2000) ‘Gender and Race: (What) are they? (What) do we want them to be?’ Noûs, 34, pp. 31–55. Jeffers, C. (2013) ‘The Cultural Theory of Race: Yet Another Look at Du Bois’s “The Conservation of Races”’, Ethics, 123, pp. 403–426. Jenkins, K. (2016) ‘Amelioration and Inclusion: Gender Identity and the Concept of Woman’, Ethics, 126, pp. 394–421. Kripke, S. (1980) Naming and Necessity. Cambridge, MA: Harvard University Press. Leslie, S-J. (2017) ‘The Original Sin of Cognition. Fear, Prejudice and Generalization’, Journal of Philosophy, 114(8), pp. 1–29. Leslie, S-J. (2008) ‘Generics: Cognition and Acquisition’ The Philosophical Review, 117(1), pp. 1–47. Leslie, S-J. (2007) ‘Generics and the Structure of the Mind’, Philosophical Perspectives, 21(1), pp. 375–403. Leslie, S-J and A. Lerner (2022) ‘Generic Generalizations’, The Stanford Encyclopedia of Philosophy (Fall 2022 Edition), E. N. Zalta & U. Nodelman (eds.), URL = < https://plato.stanford.edu/archives/ fall2022/entries/generics/>. Mackie, P. (1994) ‘Sortal Concepts and Essential Properties’, The Philosophical Quarterly, 44, pp. 311–333. Mason, R. (2021) ‘Social Kinds are Essentially Mind-Dependent’, Philosophical Studies, 178, pp. 3975–3994. Mason, R. (2022) ‘Women Are Not Adult Human Females’, Australasian Journal of Philosophy, 10.1080/00048402.2022.2149824 McKinnon, R. (2015) ‘Trans*formations’, Res Philosophica, 92, pp. 419–440. Mikkola, M. (2016) The Wrong of Injustice. Dehumanization and its Role in Feminist Philosophy. New York: Oxford University Press. Oshana, M. (2005) ‘Autonomy and Self-Identity’, in J. Christman and J. Anderson (eds.), Autonomy and the Challenges of Liberalism: New Essays. Cambridge: Cambridge University Press. Passinsky, A. (2019) ‘Finean Feminist Metaphysics’, Inquiry: An Interdisciplinary Journal of Philosophy, 64(9), pp. 937–954. Ritchie, K. (2015) ‘The Metaphysics of Social Groups’, Philosophy Compass, 10, pp. 310–321. Robertson Ishii, T. and Atkins, P. (2023) ‘Essential vs. Accidental Properties’, The Stanford Encyclopedia of Philosophy (Spring 2023 Edition), E. N. Zalta & U. Nodelman (eds.), URL = < https://plato.stanford.edu/archives/spr2023/entries/essential-accidental/>. Rogan, M. (2016) ‘Growing Up Trans’, The Walrus, September 12, 2016; https://thewalrus.ca/growingup-trans/, accessed February 20, 2022. Saul, J. M. (2017) ‘Are Generics Especially Pernicious?’ Inquiry. An Interdisciplinary Journal of Philosophy, DOI: 10.1080/0020174X.2017.1285995

399

Natalie Stoljar Shelby, T. (2002) ‘Foundations of Black Solidarity: Collective Identity or Common Oppression?’ Ethics, 112, pp. 231–266. Spelman, E. (1988) Inessential Woman. Problems of Exclusion in Feminist Thought. Boston, MA: Beacon Press Books. Stoljar, N. (2018) ‘Gender and the Unthinkable’, in P. Garavaso (ed.), The Bloomsbury Companion to Analytic Feminism. London: Bloomsbury, pp. 123–143. Stoljar, N. (2011) ‘Different Women. Gender and the Realism–Nominalism Debat’, in C. Witt (ed.), Feminist Metaphysics. Explorations in the Ontology of Sex, Gender and the Self. Dordrecht: Springer, pp. 27–46. Stoljar, N. (1995) ‘Essence, Identity and the Concept of Woman’, Philosophical Topics, 23, pp. 261–293. Taylor, C. (2004) Modern Social Imaginaries. Durham, NC: Duke University Press (Public Planet Books). Taylor, E. (2016) ‘Groups and Oppression’, Hypatia, 31, pp. 520–536. Witt, C. (2011) The Metaphysics of Gender. Oxford: Oxford University Press. Witt, C. (2011a) ‘What is Gender Essentialism?’ in C. Witt (ed.), Feminist Metaphysics. Explorations in the Ontology of Sex, Gender and the Self. Dordrecht: Springer, pp. 11–25. Young, I. M. (1997) ‘Gender as Seriality: Thinking about Women as a Social Collective,’ in I. M. Young (ed.), Intersecting Voices. Dilemmas of Gender, Political Philosophy, and Policy. Princeton, NJ: Princeton University Press, pp. 12–37. Young, I. M. (1990) Justice and the Politics of Difference. Princeton: Princeton University Press.

400

27 UNITY Charlotte Witt

“An essence is by its very nature a kind of unity” Aristotle (Metaphysics 1045b3)

The concept of essence threads through philosophy from Aristotle’s biological and metaphysical works to contemporary debates over essentialism regarding social kinds, like races and genders. (Malink, this volume) One basic question about essences is: How do essences make a living? What question or questions are essences an answer to? What problems do they solve? Some of the proposed questions or problems are familiar. Perhaps essences secure the individuation or the identity of individuals, perhaps they are relevant to an individual’s persistence conditions or to its membership in a natural or social kind? (Marabello, this volume; Tahko, this volume; Scarpati, this volume) The debate between essentialists and anti-essentialists is standardly framed in terms of these issues and questions. And in response to each of these suggestions the anti-essentialist argues that nothing like an essence is needed for the job or that the job does not need to be done or cannot be done or ought not to be done at all. But there is another line of thought about how essences make a living that also stretches back to Aristotle and extends forward to discussions in contemporary metaphysics. This is the idea that the job of essences—how they make a living—is to unify parts into an organized whole, an individual that is not identical to the sum of its parts. Notice that the task of unifying the parts of individuals is not the same job as providing an individual an identity or providing persistence conditions or explaining membership in a natural or social kind. All these jobs presuppose either that an atomic individual exists, or that an individual’s parts have already been unified. In other words, putting to one side the existence of atomic individuals, the other tasks assigned to essence presuppose that the individual has been unified. From this perspective the function of essences as unifiers of composite beings is presupposed by the other functions associated with essence. In order for the other projects to get off the ground, there must exist unified individuals. Even though the notion of essence as unifying, what I sometimes call a “uniessence”, is not as prominent as the other ways of specifying the job of essences itemized above, it may well be more fundamental. In what follows I describe the idea that essences unify individuals beginning with Aristotle and tracing its re-emergence in contemporary metaphysics.

DOI: 10.4324/9781003008750-31

401

Charlotte Witt

27.1

Essence and Unity in Aristotle

Aristotle raises the question of the unity of a composite substance in the Metaphysics where he says “clearly the question is why the matter is an individual thing, e.g. why are these materials a house?” (Metaph. 1041b 5–6) Aristotle contrasts an individual, whose parts (or matter) are unified, like a house, with “heaps” or piles of parts, like windows and doors in a pile on a building site. Aristotle’s answer to the question of unity of a composite individual, say, a house is that the essence of the house is present in the matter (or house parts. (Metaph.1041b 6–7). In this central text on the unity of composite substances Aristotle refers to both the form and the essence of the individual as the principle of unity, and he seems to treat them interchangeably.1 It is by virtue of being a shelter for humans and animals (which is the essence of a house) that the parts are a unified individual rather than a mere “heap” of house parts. What is true of an artifact, like a house, is also true of a living being, like a human being. (Metaph. 1041b 8–10). For Aristotle composite substances, like houses and humans, are hylomorphic individuals; they are composed of form and matter. In this text (and others) there is a close relationship between the concept of essence and the concept of form; indeed, here and in other texts Aristotle uses the expression for essence and the term for form interchangeably.2 It is reasonable to conclude that Aristotle sees form and essence as performing a similar unifying function in the case of composite substances, like animals and artifacts, even if, as we see later, they appear to differ from one another in other contexts. But what notion of unity or being one is relevant to composite substances like animals and artifacts? Typically, Aristotle says that being one is “said in many ways” and we might wonder exactly what the relevant notion of unity is in the case of composite substances. What is the matter? What is the form? And how does the latter unify the former? In the discussion of unity one of Aristotle’s examples is of an artifact, a shoe: “While in a sense we call anything one if it is a quantity and continuous, In a sense we do not unless it is a whole, i.e. unless it has one form; e.g. if we saw the parts of a shoe put together anyhow we would not call them one all the same (unless because of their continuity); we do this only if they are put together to be a shoe and to have thereby one form”(Metaph. 1016b12–15). In this example the matter is the functional parts of the shoe (sole, laces etc), and if the parts are arranged correctly then they realize the shoe function. It is not the continuity of the matter that secures this type of unity, but rather that the parts are arranged to realize a shoe form, (“one form”) and it is by doing so that the substance is unified. The same conception of unity was operative in the example of the house and house parts and the human being and bodily parts. In both these examples there is unity in form or eidos, which is to say the functional parts, the matter, are unified by the form into an individual, a shoe. The functional parts of the shoe (sole, heel, tongue etc) are functional parts, of course, only when they are parts of a shoe, and not when they are a heap of parts on the cobbler’s bench. Aristotle’s general question is “why do these parts form this whole?” (Metaph. 1041b2) and it can arise in any case where material parts and a unified whole can be articulated or distinguished. Both the house and the human are given as examples of essences functioning as principles of unity of individuals. There is a tradition of scholarship on Aristotle that differentiates the unity of living beings from the unity of artifacts, which would bear on the question of whether living beings and artifacts had essences (conceived of as principles of unity) in the same sense or degree. However, we can put this question of interpretation to one side for the moment since, as we have seen Aristotle treats the house and the human as

402

Unity

individuals whose material parts are unified by their essences. Indeed, as we saw above Aristotle’s general characterization of the question of unity to which essences respond is given simply in terms of parts and wholes or parts and individuals, which suggests a broad range of possible applications. If Aristotle thinks that animals, but not artifacts, are individuals whose parts are unified by their essences in some special sense, then the restriction will need to rest on additional arguments or considerations. Or it may be the case that the notion of unity comes in degrees, and that living substances are more unified than artifacts, but the latter are not simply accidental unities. Some scholars use the notion of unity and the issue of the unity of form and matter in the composite substance to differentiate between artifacts and living natural substances. Only in living natural substances are form and matter truly unified, and so only living natural substances, like animals and plants, count as genuine substances. Kosman argues along these lines: “They (artifacts) are introduced into the discussion of the Metaphysics not as examples of substance, but as things whose material and formal principles are more clearly visible precisely because they are nonsubstantial beings, and therefore beings in which the principles of matter and form are independent of one another and separable”.3 Are artifacts and animals parallel exemplars of form or essence as the cause of unity of composite substances as we argued above? Or do artifacts fail to make the cut? We will return to this issue below. It is useful to pause for a moment to consider the relationship among the various jobs assigned to essences in the opening paragraph. What is the relationship between essence as cause of unity of composite individuals, essence as securing the identity or persistence of an individual, and essence as grounding an individual’s membership in a kind? One way to describe the relationship is that both the job of essence to ground the identity or persistence of an individual, and the job of essence to ground kind membership presuppose that an individual exists. (Witt 1989). But absent the unification of parts into an individual, or without the presence of essence as unifier of the material parts, there would be no individual, to classify or to identify or that persists. So, the suggestion is that the role of essence as unifier is prior to, though compatible with, the other roles essence has traditionally been assigned. In order for an individual to be something or other, in order that it persist or be a member of a kind, its parts must be unified into a whole individual. However, the role of essence as unifying the composite is compatible with its other traditional functions. An individual’s principle of unity can at the same time be what accounts for the individual’s identity, persistence, and kind membership. There are, of course, complications to the story. In some contexts, Aristotle connects essence with definition. Indeed, Aristotle describes definition as the statement of the essence. (Topics 102a; Metaph. 1017b20) There is an apparent tension between the idea of essences as unifiers, and the description of essences as the objects of definition. The tension is not that essences are real beings, like forms, and definitions are linguistic and semantic. Recall that Aristotle has a theory of real definitions, in which things or beings are the object of definitions rather than words or meanings. So, it is not simply essence as the object of definition that is problematic. Rather the difficulty is that essences as the objects of definitions would seem to be universal because Aristotelian definitions are formulated in universal terms. But it is difficult to see how a universal definition, or its essence correlate can unify anything. Universals, as Aristotle says, are shared and common; they are predicated of more than

403

Charlotte Witt

one thing (Metaph. 1038b9–15). But as a principle of unity, the form or essence seems to be individual, if it is to do the job of organizing the material parts into an individual. In short, essences as unifying and essences as the objects of definition seem to make a living in two incompatible ways. The question of whether Aristotle’s forms (and hence essences) are universal or individual is still debated among Aristotle scholars—with no consensus in sight. (Frede 1987; Witt 1989; Loux 1991) The role of form or essence as unifier lends support to the individual form or essence interpretation, without, of course, being decisive. However, it is not necessary to untie this scholarly knot to borrow from Aristotle the idea that essences could function as unifying principles whether or not that role is compatible with the notion of essence as the object of definition. What is important for later metaphysicians in the Aristotelian tradition is the idea that form or essence could be unifying principles of individual substances.

27.2

Essence and Unity in Contemporary Metaphysics

The resurgence of philosophical interest in metaphysics during the latter part of the 20th century also marked a rebirth of interest in Aristotle’s metaphysical ideas, including his essentialism. Contemporary metaphysicians turned to Aristotle for ideas about essences and the related notions of identity, persistence, and natural kinds. (Slote 1975; Brody 1980; Wiggins 1980) A second important line of thought attempted a synthesis between Aristotle’s hylomorphism and contemporary mereology: mereological hylomorphism. The basic idea guiding mereological hylomorphism is that form and matter should be understood as parts (in the very same sense of the term) of the individual composite substance. Koslicki traces mereological hylomorphism back to Aristotle: “And while I of course concede that Aristotle speaks less often explicitly of the form as being itself part of the compound than he speaks of the matter as being part of the compound, I nevertheless want to urge that we accept both of these commitments as official Aristotelian doctrine” (Koslicki 2006: 725). Others are critical of the mereological interpretation of hylomorphism. (Marmodoro 2013) There are two central challenges in attributing mereological hylomorphism to Aristotle and both problems concern the unity of the composite substance. Some scholars point to a regress argument according to which Aristotle argues that the unifier of a composite substance cannot be simply an additional part or element (as required by mereological hylomorphism) but must be of a different ontological type. (Metaph. 7. 17). Aristotle argues that just as a syllable is not unified by one of its parts or letters, or by adding another part or letter, so, too a composite substance is not unified by one of its parts. Rather, the syllable is unified by the arrangement of the letters; and the arrangement (the form of the syllable) is not part of the syllable in the very same way a letter is. It is not a part as required by mereological hylomorphism. Other scholars critical of attributing mereological hylomorphism to Aristotle point to his view that the matter of a composite substance does not persist through the destruction of the substance; in Aristotle’s memorable image, a corpse is a human body in name only. But if the matter does not persist through the destruction of a composite substance, then the substance cannot be a just a mereological sum of its parts because they do not exist to be summed up independent of the composite substance. However, quite apart from the issue of whether to attribute mereological hylomorphism to Aristotle, we might wonder about its viability as an ontology of concrete individuals. If form and matter are both parts of the composite, then what accounts for the unity

404

Unity

of the substance? One important criterion for being substance is being unified, and mereological hylomorphism, whether or not it is the best interpretation of Aristotle, needs an account of unity that can do the job. To revert to Aristotelian idiolect, how is a horse or a house more unified than a heap? How are matter and form unified in the composite substance? What story can mereological hylomorphism tell about the unifying function of form or essence? In a recent book Koslicki (2018) proposes that the traditional criterion for being substance, namely ontological independence, should be replaced with the criterion of unity. “The results of the present discussion indicate that, at least for the case of composite entities, unity has at least as much, and perhaps more, to offer than ontological independence, and therefore ought to be explored first by Aristotelians in search of a defensible notion of substancehood” (2018: 190). Koslicki points out there is a serious difficulty facing a hylomorphic understanding of the individual composite using the independence criterion for substancehood. Given the hylomorphic analysis of the composite substance the individual is not independent because it depends on the existence of matter and of form. Hence, given hylomorphism, the composite will turn out not to be substance because it is dependent on its parts—on its form and its matter. On the other hand, if substance is defined in terms of unity rather than independence then the composite might still be a viable candidate for being substance. And, in addition, unity is a criterion that would allow for a principled distinction between natural beings, like animals, and heaps, like a pile of stones, and—arguably—also between natural beings and artifacts. The proposed unity criterion for substance provides a sliding scale, a comparative ranking of fundamentality. For example, it could be that according to a given criterion of unity the hylomorphic natural being is more unified and hence more substantial than either the artifact or the heap. What notion of unity compatible with mereological hylomorphism can provide grounds for a comparative ranking of substantiality? According to Koslicki integrated wholes “derive their unity from the way in which their parts lawfully depend upon each other for the manifestation of certain of their capacities” (2018: 215) For example, the parts of an eye display this type of unity; the parts of the eye depend upon one another, lawfully, in order to realize the capacity of sight. (2018: 211). Koslicki proposes to define unity in terms of “lawful, generic, interactional interdependence” which grounds the unity of the material parts of the composite. Further, the composite substance is an integrated whole whose unity is secured by the “team-work” engaged in by the parts—or by the way in which some parts interact with other parts to enable the composite to realize certain of its capacities. (2018: 210) The unity of the composite rests on its having lawlike “team-work capacities” or on its functional form. In contrast, according to Koslicki, artifacts and heaps are not integrated wholes to the same degree. Heaps clearly do not exhibit the kind of “team-work capacities” that hylomorphic compounds that belong to natural kinds display. However, artifacts do seem to manifest the kind of “team-work capacities” that ground the unity of hylomorphic entities like animals or chemical compounds. Hammers pound nails and they do so via the teamwork of the capacities of wooden handle and iron head. However, Koslicki argues that in the case of artifacts the teamwork is not strictly lawlike, as it is in the case of natural beings, because artifacts require, in addition, certain “conditional normative principles” which govern how the components (handle, head) ought to interact to accomplish goals that are set by agents external to the system. Hence, Koslicki’s criterion of unity allows for a principled ranking of hylomorphic individuals from more substantial natural beings to artifacts dwindling down to mere heaps and piles.

405

Charlotte Witt

An alternative, more egalitarian, view of artifacts sees them primarily in functional terms. According to Lynne Rudder Baker, artifacts have their proper functions essentially. “Thus, an artifact has its proper function essentially: the nature of an artifact lies in its proper function—what it was designed to do, the purpose for which it was produced” (Baker 2008: 102). An artifact’s proper function is the function it was intended by its maker to perform. This notion hearkens back to Aristotle, who distinguished between the proper function of an artifact like a shoe and what it can be used for. The proper function of a shoe is to cover and protect the foot, but it can be used to kill a wasp or as a projectile (Witt 2015). For Baker’s Constitution View, all ordinary material objects are related to their matter or material parts via the unifying relationship of constitution. Both natural beings and artifacts are unified via the relationship of constitution, and they are all unified to the same degree. For Baker constitution is an ontological relationship of unity that holds between individuals of different primary (or essential) kinds.4 It is a relationship that may hold between granite slabs and war memorials, between pieces of metal and traffic signs, between DNA molecules and genes, between pieces of paper and dollar bills—things of basically different kinds that are spatially coincident. (2008: 32) Constitution is a relationship of unity but not identity: The piece of paper is not identical to the dollar bill because it is of a different primary kind with different de re persistence conditions. The notion that every existing object is of just one primary kind, and has its primary kind property essentially, is a core assumption of the Constitution View. The key puzzle facing Baker is to explain how two individuals of different primary kinds can become numerically one. How can the piece of paper become one with the dollar bill if they are members of different primary kinds, with different essential properties and different de re persistence conditions? Aren’t they two individuals contingently occupying the same spatial location? According to some interpretations Aristotle’s response to this question is to deny that the paper is an individual of a primary kind with de re persistence conditions, which is independent of the dollar bill it constitutes. Baker does not follow Aristotle on this point, however, and instead points out that the “two way borrowing of properties” between the piece of paper (it is valuable because it is a dollar bill) and the dollar bill (it is 5” long because the paper is 5” long) is evidence for the claim that constitution is a unity relationship. The difficulty comes not with the idea that constitution is a unity relationship of some kind or other, but with the idea of numerical oneness. How can constitution both relate two individuals of essentially different kinds (two objects) and be a relationship of numerical unity (one object)? Baker refers to Aristotle’s notion of accidental sameness as a historical precedent for attributing numerical oneness to an accidental (or contingent) combination of different individuals like a musical man, which is a combination of a substance and an accident. Beings that are accidentally the same (the musical man, the pale man) are numerically one according to Aristotle. But these beings are not combinations of two individuals of different primary kinds with different de re persistence conditions; they are combinations of substances (= Baker’s individuals) and accidents, and accidents are ontologically dependent upon substances for Aristotle. So, it is not clear that Aristotle does provide a historical precedent for the concept of unity (= numerical oneness) without identity. Another aspect of constitution does provide grounds for a unity without identity claim, and that is the idea that the identity of the constituted object is given by its primary kind and not the primary kind of the constituter.

406

Unity

If x constitutes y at t, there is a unified individual whose identity is provided by y’s primary kind. If a piece of marble constitutes a statue, the piece of marble does not cease to exist, but (I can only put it metaphorically) its identity is encompassed or subsumed by the statue. (2008: 166) This is a crucial step in rendering the unity without identity aspect of constitution coherent, and I do not think that Baker’s metaphors are fully adequate without further explanation. However, the notion of a primary or essential kind connects Baker’s notion of constitution as a unity relation with essentialism or with the idea that ordinary objects, both artifacts and natural kinds, are essentially what they are. Are artifacts less unified and hence less substantial than natural beings as Koslicki thinks? Or are artifacts and natural beings both just ordinary objects woven together (and unified) by means of a relationship of constitution as Baker would have it? Each of these views can claim a legitimate Aristotelian heritage and origin in the connection that we find in Aristotle’s texts between an individual form or essence and a unified substance. In order to adjudicate between the two ways of thinking about artifacts we would need to know more about each. Why does the externality of the normative conditions that govern the “teamwork” in the case of artifacts works against the unity of the artifact for Koslicki? How can objects that belong to two essentially different kinds become unified through the constitution relationship for Baker? Artifacts are just one type of social kind, however, and recently the idea of essence or form as unifying has emerged in relation to human social kinds, like races and genders.

27.3

Social Kinds and Uniessentialism

As we have just seen the ontological status of artifacts is one topic of lively debate in contemporary metaphysics with quite differing theories of their unity and the role played by essences in grounding their unity. However, recent work in social ontology also focuses on exploring the criteria for membership in human social kinds like races and genders. (Mallon, this volume; Scarpati, this volume) To what extent are these social kinds like natural kinds? Are there clear criteria for social kind membership? As we noted above, these questions about the criteria for kind membership demarcate one job often assigned to essences. With regard to human kinds, like races and genders, there is widespread criticism of essentialist views; indeed in this context, to call a view “essentialist” serves—in itself—as a fatal criticism. There is a tendency to equate essentialism with biological essentialism in these criticisms, and the criticism of essentialism is really a rejection of the idea of races and genders as natural kinds. However, as the example of artifacts demonstrates, the fact of a social origin need not rule out essentialism with regard to that entity. But it also does not “rule in” essentialism about artifacts or about human social kinds, which must be argued for or rejected on other grounds. Determining the grounds of kind membership (or arguing that there are none) is only one type of question one might be interested in with regard to essentialism and social kind membership. One might also be interested in exploring the notion of essence as unifying as it applies to members of human social kinds. Recall the earlier discussion of Aristotle and the idea that its form is a principle of unity for a material object like a house. It is by virtue of the presence of form to the material parts that a unified house exists. This idea can be transposed to the normative, social realm. Social individuals, each of us, occupy many social positions, and consequently enact many social roles at a time and over time. We are responsive to complex sets of norms that are not always either partially or wholly compatible with one

407

Charlotte Witt

another. What is the glue that holds us together normatively? How are these norms unified into a more or less coherent whole? What is the principle of normative unity for social individuals? Gender uniessentialism is the position that our gender is the principle of normative unity. Our responsiveness to gender norms trump, order and inflect our other social roles at a time and over time (Witt 2011). The normative centrality of gender reflects the way that human social lives are organized as a matter of fact; hence the status of gender as the principle of normative unity for social individuals is neither fixed nor necessary. Two important criticisms of gender uniessentialism focus on the unifying role attributed to gender. Some argue that we don’t have or need a principle of normative unity; we are just fine without one. Indeed, the focus on the normative unity and coherence of the self is the problem and normative multiplicity is the solution. (Scheman 2020) Others argue that we might need a principle of normative unity, but our gender is not that principle and there are more plausible candidates (Cudd 2012). As we have seen the metaphysical question of how parts are unified into an individual has persisted in the history of metaphysics, taking different forms with different emphases and presuppositions. A central focus here is the Aristotelian hylomorphic individual, a composite of matter and form because it highlights the role of essence (or form) as the principle of unity. The form or essence is the cause of being of the hylomorphic individual. This connection, between essence and unity expands in different directions depending upon how one understands the role of essence and what one means by unity.

27.4

Related Topics

See Malink (this volume), Marabello (this volume), Tahko (this volume), Scarpati (this volume), Mallon (this volume), Rosario (this volume).

Notes 1 Aristotle describes the same example earlier in the chapter; also referring to the essence of the house as what unifies the material parts. (1041a27-28). 2 Aristotle uses essence and form interchangeably in Z 17. Essence at 1041a8 and 1041b6, and form at 1041b8. See also Z 7 1032b1-2; 1032b13-14; Z10 1035b14-16; 1035b32; H 4 1044a36. 3 The Activity of Being: An Essay on Aristotle’s Ontology, Harvard University Press, 2013, p. 93. 4 The following two paragraphs are lightly modified from my Notre Dame Philosophical Review of Lynne Rudder Baker’s The Metaphysics of Everyday Life: An Essay in Practical Realism Cambridge University Press 2008. 2008/07.28

References Baker, L. (2008). The Metaphysics of Everyday Life: An Essay in Practical Realism. Cambridge: Cambridge University Press. Broadie, B. (1980) Identity and Essence, Princeton: Princeton University Press. Cudd, A. (2012) “Comments on Charlotte Witt’s The Metaphysics of Gender” Symposium on Gender, Race and Philosophy, Vol. 8, No. 2: 1–6. Frede, M. (1987) “Individuals in Aristotle” in Essays in Ancient Philosophy ed. by M. Frede, Minneapolis: University of Minnesota Press. pp. 49–71. Koslicki, K. (2006) “Aristotle’s Mereology and the Status of Form” Journal of Philosophy, Vol. 103, No. 12: 715–736. Koslicki, K. (2018) Form, Matter, Substance, Oxford: Oxford University Press. Loux, M. (1991) Primary Ousia: An Essay on Aristotle’s Metaphysics Z and H, Ithaca: Cornell University Press.

408

Unity Marmodoro, A. (2013) “Aristotle’s Hylomorphism without Reconditioning” Philosophical Inquiry, Vol. 36, No. 1–2: 5-22. Ross, W. D. ed. (1924). Aristotle’s Metaphysics. Oxford: Oxford University Press. Scheman, N. (2020) “Disruptive Demands: Messy Challenges to Analytic Metaphysics” Journal of Social Philosophy, Spring 2020: 1–21. Slote, M. (1975) Metaphysics and Essence, Oxford: Blackwell. Wiggins, D. (1980) Sameness and Substance, Cambridge: Harvard University Press. Witt, C. (1989) Substance and Essence in Aristotle, Ithaca: Cornell University Press. Witt, C. (2015) “In Defense of the Craft Analogy: Artifacts and Natural Teleology” in Aristotle’s Physics: A Critical Guide ed. by M. Leunissen. Cambridge: Cambridge University Press. pp. 107–120. Witt, C. (2008) Review of Lynne Rudder Baker’s The Metaphysics of Everyday Life: An Essay in Practical Realism. Notre Dame Philosophical Review, Vol. 7 Witt, C. (2011) The Metaphysics of Gender, Oxford: Oxford University Press.

409

28 ETHICAL VALUE Stavroula Glezakos and Julie Tannenbaum

28.1 Virtue Theory and the Essence of Human Beings Some ethical theorists begin with a fundamental notion of value(s), or ethical intuition(s), and argumentatively proceed to foundational ethical principles (e.g., Kant, G.E. Moore, and William David Ross). Aristotle, on the other hand, starts with non-evaluative claims1 about what is characteristic of, or essential to, being human and concludes that living virtuously (justly, generously, courageously, temperately, etc.) is eudaimonia, i.e., the best human life. Aristotle transitions from essence to ethics in the so-called “The Function Argument” of the Nicomachean Ethics (1097b23–1098a21; see Lawrence 2006, 2009), where he begins by describing the function (ergon) of a human being. The function characterizes or picks out what is essential to a kind, and thus what distinguishes it from other kinds of things.2 Aristotle argues by elimination (1097b35–1098a3) that the function of a human being is to engage in activities that follow or imply reason (1098a2–8, 1102b30–1103a8). Living well for a human being is doing these reason-involving activities well, that is, doing them virtuously. While some think a good life can involve some types of vice (Williams 1972, 1981), there are also compelling defenses of Aristotle’s position (Lawrence 2001; Nussbaum 1995a) as well as contemporary theorists inspired by this approach (Foot 2001; Hursthouse 1999). To establish that living well amounts to a lifetime of action, thought, and feeling in accord with the virtues of justice, generosity, temperance, courage, etc., Aristotle again turns to essence: the essences of reason involving activities (Books II-VII).3 For example, giving involves transferring or sharing something (e.g., one’s money or time). To do this well is to give what one should, when one should, to whom one should, for the reason one should, wholeheartedly, and so on (see 1106b21–23 and BKIV.1). It is precisely the virtues of generosity and justice that make possible giving what one should (e.g., what is needed, not what is stolen), to whom one should (e.g., to those who are in need or who have a right to it), and so on. Much of Aristotle’s ethical system focuses on the nature of human beings, and thus on the virtues that enable human beings to live well regardless of their gender, biological sex, sexual orientation, race, religion, etc. (though, as discussed in Section VII below, Aristotle periodically suggests that what counts as living well for a man, woman, and slave are different due to differences in their nature). Nonetheless his methodological approach can

410

DOI: 10.4324/9781003008750-32

Ethical Value

be applied to any kind of living being: to live well, according to him, is determined by the function (nature) of the kind of being that one is.4 There are other forms of what might be considered a virtue theory, which are based not on claims about the essence of, e.g., human beings, but rather, the denial of essence. We turn to Buddhism to explore one such perspective. A metaphysical position adopted by most schools of Buddhism is that “all things lack an inherent nature … all things are, as it were, “empty” of self-existence” (James 2003: 144). On this view, there is change, impermanence, and interdependence, but no being that endures change or that possesses mental or physical properties (Emmanuel 2015: 6). There is nothing that can be isolated and identified as essential. While some deny that there can be mental states without a self whose mental states they are (Williams 1998), others offer an account of how this is possible (Garfield 2022a). Recognizing the truth of this denial of essence is, according to Buddhism, a key element in avoiding ethical vices like selfishness and greed.5 Being appropriately responsive to experiences beyond the ones that I take to be mine is part of making progress on the path of insight regarding how the world really is (Siderits 2021). This shift in perspective gives one “an orientation to the world that is characterized by attitudes such as friendliness and impartiality” (Garfield 2022b: 151). On the Buddhist view, although there are no selves, there is suffering and actions that can either add to, or reduce, that suffering. For example, unjust acts based on the thought “it should all belong to me” add to suffering, while attentiveness to the effects of harsh words and skillful efforts to minimize their occurrence reduce suffering. It is towards the cultivation of the latter thoughts and actions that the Buddhist ethic directs us. While the Buddhist condemnation of selfishness is shared by western virtue theories, the no-self view appears to be incompatible with other central western ethical notions like respect for autonomy, personal responsibility, guilt, unfairness, and resentment. Since there is no self that can be given more or less than another, there is no appropriate target of fairness or blame. And it is meditation upon the fact that there is no self that supports us in our ethical development: “If resentment arises in him when he applies his mind to a hostile person because he remembers wrongs done by that person … he should try resolution into elements. How? ‘… . what is it you are angry with? Is it head hairs you are angry with? Or body hairs? Or nails? … Or is it urine you are angry with? Or alternatively, is it the earth element in the head hairs, etc., you are angry with?’” (Nanamoli 2010: 293, 301; see also Goodman 2002). Instead of responding to disappointment or perceived wrongdoing with demands, resentment, and punishment, the aim is to cultivate compassion for suffering wherever it exists (Finnigan 2017, 2018; Garfield 2022b).

28.2

Ethics of Enhancement and the Essence of Being Human

According to Juengst (1998), an enhancement is an intervention used for improvement beyond what is necessary to restore or maintain health. Many enhancements begin as treatments for disability, but those same interventions can become enhancements when used by those without the relevant disability. For example, Modafinil was initially a treatment for narcolepsy, but now is used by airline pilots, military personnel, and others to stay awake

411

Stavroula Glezakos and Julie Tannenbaum

for days at a time (Sahakian and Morein-Zamir 2007; Maher 2008; Talbot 2009). Claims about what is essential to being human arise in some of the moral debates about enhancement. In some cases these claims play no crucial role, as in this argument: humans are essentially vulnerable (e.g., to sickness, death, and frustration and sadness—in general, to chance and change); a key good of human life is the struggle to deal with our vulnerabilities; enhancements that eliminate or reduce our vulnerability, and thereby eliminate or reduce the good of struggling with it, are thus bad in this respect (Parens 1995). Even if the features claimed to be essential to human beings (e.g., vulnerability) were in fact merely contingent and widespread, the same conclusion would follow. However, there are other debates about enhancement in which claims about what is essential to human beings are ineliminable from the argument. For example, for an enhancement to benefit a particular individual, the individual must persist post-enhancement (Maria Scarpati, this volume). But if being human is essential to being that individual, and if the enhancement is incompatible with being human (e.g., involves no longer succumbing to disease or being embodied), then the individual would no longer exist post-enhancement, and thus would not be benefited by it (Loets, this volume). Some philosophers nevertheless “favor future people being posthuman rather than human, [because] the posthuman beings would lead lives more worthwhile than the alternative humans would lead” (Bostrom 2003: 496). There are also concerns about whether enhancement might result in the end of the human species and not merely human individuals. This concern becomes increasingly pressing if the enhancement involves germ-line genetic engineering (i.e., altering the genes of eggs or sperms, which are then passed onto offspring), and yet more so if these changes are irreversible. One strategy for objecting to such enhancements would be to establish not only that the existence of the human species is valuable but also that its value justifies forgoing the creation of a new species which perhaps would also be valuable. Another strategy (Annas, Andrews, and Isasi 2002) is to suggest that individual human beings have a right to be a member of the human species and that genetic enhancement (of all human beings) that would eliminate the human species violates this right. However, not only is it unclear what the grounds for such a right would be, as well as why and when such a right might be overridden, but it also seems impossible to violate such a right: so long as an individual human being exists, so does the species. Finally, there is the issue of better understanding what the nature or essence of an enhancement is. If an enhancement improves functioning beyond what is necessary to restore or sustain health, this raises the question: what is health? As we discuss in the next section, a purely medical model cannot appropriately capture the notion of health, and hence also cannot account for the related notions of enhancement and disability (Juengst 1998; Schwartz 2005).

28.3

Disability Ethics and Essence

Both historically and currently, the categorization of a condition as a disability has led to various forms of exclusion and oppression, including confinement, sterilization, and death of the individuals so categorized (e.g., not providing life-saving medical treatment, aborting, or not implanting embryos), as well as to feelings of disgust, pity, and shame (Boorsay 2005; Carlson 2001). Activists and academics have not only challenged this unethical treatment, but also questioned the conceptions of disability on which it relied. As corrective to this oppressive treatment, some have advocated for ethical principles of accommodation (e.g., the installation of ramps, tactile pavement, captioning technology, as

412

Ethical Value

well as the providing of extra time on exams, implementing alternative education models), mitigation (or altered responsibility in light of what a particular person’s body or mind cannot do), and compensation (workers’ compensation for disabilities caused at the workplace, etc.). There are also principles of prioritization that are about, or make use of, the notion of a disability: using scarce resources to treat or develop treatments for disabilities rather than to provide enhancements, and giving scarce resources to those (individuals or countries) who would benefit the most—where this is determined in part by comparing the disability-adjusted life years of the individuals were they to receive vs. not receive the resource (Persad, Wertheimer, and Emanuel 2009). A definition of disability is needed in order to deploy any such ethical principles. Some disabilities are physical, while others are mental, including psycho-social (Brown, this volume) or a combination of the two. Here, we focus on some approaches to specifying the essence of disability in general, and some difficulties with these accounts (Altman 2001; Wasserman and Aas 2022). Clinical (medical) definitions of disability focus on deviations from a biological, chemical, or physiological statistical mean. Assuming such deviations can be reliably measured, questions remain: what degree of deviation is consistent with health, or constitutes a disability, and why? Without an answer, the ethical significance of such deviations is unclear. Feminists point out that some medical norms are often based on young men, which then pathologize women of all ages and conditions (Young 1980; Wendell 1996). Moreover, medical norms have also traditionally enshrined sexual dualism, which has led to harmful intervention in the case of intersex neonates. Lastly, the medical model of disability takes disability to be a problem requiring medical rather than social or environmental intervention or accommodation (e.g., altering a building’s architecture, or developing alternative technological modes of interaction). Others embrace a social definition of disability (Griffith, this volume). One type of social definition focuses on whether a person’s bodily or mental condition prevents or inhibits a certain degree or type of participation in various forms of social (including interpersonal, political, and economic) life. So defined, disability is of ethical significance when combined with the ethical claim that these forms of social life are goods and the ethical principle that being deprived of goods is bad (at least in some respect) for the one who is deprived (Nagel 1979). But just as with the medical model, this type of social definition of disability must answer the question of which types of social activities are paradigmatic (and hence serve as the standard for purposes of defining disability) and why. For example, not being able to see in the dark, echolocate, and mind-read prevent one from engaging in a variety of kinds of economic and social activities unless one has accommodations or assistance (e.g., light). And yet these are not disabilities. A different social model of disability highlights the role of biased practices. According to one version, certain conditions (e.g., blindness) are identified as either pre-social impairments or mere variations in a mismatch between one’s condition and the physical or social environment. The condition is a disability when it is socially responded to in various unfair and oppressive ways (UPIAS 1976; Scotch and Schriner 1997; Morris 2001, Barnes 2016). Some critical disability theorists make no mention of impairment and thus (in spite of eschewing the language of essence) could be thought of as offering a very broad social or political definition of disability: to be disabled is to be labeled as such and in virtue of this label stigmatized, oppressed, and discriminated against (Kafer 2013; Tremain 2017). The description of the essence of disability for these two social models makes use of ethical terms and the remedy involves the undoing of biased practices (via accommodations, civil rights protections, anti-discrimination laws, etc.).

413

Stavroula Glezakos and Julie Tannenbaum

A final way in which the issue of essence arises in the discussions of disability is in connection to personal identity. Are some disabilities essential to a person’s identity? If so, then a person cannot be benefited by having their disability eliminated, as they would cease to exist and someone new, without the disability, would come into existence. This, of course, depends on the conditions of identity over time (Loets, this volume).

28.4 The Essence and Ethics of Killing Some philosophers doubt the moral relevance of the distinction between killing and letting die (Tooley 1972, 1994; Rachels 1975; Bennett 1981). However, whether one has killed or not is central to various ethical discussions. Here we focus on euthanasia and abortion. What is at issue are the essential difference(s) between killing and letting die.6 In a doctor-patient relationship, patients have both a right not to be killed by their doctor (e.g., not to be given a lethal injection) and a right to life-sustaining aid, that is, a right not to be let die (e.g., a right to antibiotics to treat pneumonia). When a patient has waived the latter right but not the former, would removing a ventilator violate the patient’s remaining right? Is the removal killing, letting die, or some third type of action? Questions like this have inspired a large literature on the nature of killing vs letting die (a good starting point is Foot 1977, 1984; McMahan 1993; and relatedly Woollard and Howard-Snyder 2016). To avoid violating their patients’ rights, and hence acting impermissibly, doctors need to know whether they are killing or letting die when they withdraw life support. This issue arises not only for doctors but also for families and friends caring for someone suffering from an imminent terminal illness who has not expressed a wish to be killed. The difference between killing and letting die also arises in debates about abortion. According to Judith Jarvis Thomson (1971), abortion is permissible even if we concede that a fetus has a right to not be killed. Thomson implicitly suggests that, when one aborts, one does not kill the fetus but lets the fetus die; and the fetus has no right to life-sustaining aid from those who do not intentionally become pregnant.7 So, the fetus’ rights are not violated. However, paradigmatic cases of letting die differ in important ways from abortion: in the former cases, one is under threat independently of the existence of the agent who lets die, while in pregnancy the threat the fetus is under depends on it being in, and hence on the existence of, the pregnant person.

28.5

Essence of Identity and Moral Responsibility

Moral responsibility can come into question in light of the essence of a person’s identity. Consider the non-identity problem, which arises in a variety of contexts such as disease prevention, environmental ethics, and reparations. It can be illustrated with the following case (Parfit 1984: 358–60): A 14 year old girl becomes pregnant (not via in vitro fertilization), gives birth, and raises the child (Tom). Because she is so young and lacking in support, Tom has difficulties throughout his life, but he nevertheless has a life worth living. At first glance, one might judge that the 14 year old girl has harmed Tom and is morally responsible to some degree for the challenges that he faces in life that are due to how she raised him. And yet it is difficult to account for why he is harmed, because Tom is no worse off being born and having a life worth living than had she waited to get pregnant (and given birth

414

Ethical Value

to a child other than Tom). The state of affairs in which she waits is one in which Tom does not exist; that state of affairs is not better for Tom than the state of affairs in which he does exist with a life worth living. And so there seems to be no way to say that he has been made worse off in virtue of being born and raised as he was. There is thus no harm for which she is responsible. The non-identity problem arises in part given two plausible views one could have about identity. First, if having the same DNA, or a certain amount of that DNA, is necessary for having the same identity over time, then had the young girl gotten pregnant years later via sex, she would have had a child with different DNA than the child that she gave birth to at 14, and thus, Tom would not have existed. Alternatively (and compatibly), perhaps having significantly different experiences is sufficient for a difference in identity. Suppose the 14 year old girl had created and frozen an embryo. Even though the DNA would be the same, if she waits years to implant the embryo, the experiences of the life of a person who would be born would be sufficiently different from those of a person who would be born were she to implant now that they would not be the same person (Loets, this volume). What is essential to a person’s identity also plays a role in determining the moral responsibility of an agent of harm. A necessary condition for being responsible at t2 for what one did at t1 is that one is relevantly the same now as then. Some changes an agent undergoes are irrelevant for responsibility (e.g., changes in one’s hair color, the city in which one lives). But consider late-stage Alzherimer’s, which eliminates both memory and the ability to be a practically rational agent (Beebee 2018). While some might say, “when she was young, she got into all kinds of trouble”, and so she is the same person in one sense of “identity” (though not Locke’s 1689/1975) sense of identity, which relies on psychological continuity), is this the kind of identity that is relevant to being responsible for earlier wrongdoing? And what if the changes undergone are to character (e.g., from callousness to gentleness and generosity). Does such a substantial alteration result in a fundamentally different person, one who is no longer a proper target of moral criticism for past wrongdoing?

28.6 Essence and the Ethics of Sexual Activity One contentious area where issues of essence and ethics intersect (usually with a contribution from religious commitments) is on the topic of sexual activity. From claims about which sex acts are “unnatural”, to arguments about the fundamental purpose of sexual activity, essence is used by some to draw conclusions about sexual morality. Edward Feser (2013) argues as follows: (i) the function or essence of sexual activity is twofold: procreation and union (forming close emotional bonds)8 and (ii) all and only penile penetration of a vagina that does not protect against pregnancy (regardless of whether pregnancy can or will result) is consistent with the procreative function. Feser’s argument is broadly Aristotelian in that it connects claims about moral (im)permissibility with the function of a human activity. Feser concludes that the only permissible sexual activity is in the context of a marriage between one male and one female, where this involves uncontracepted penile penetration of a vagina; all other forms of sexual activity are impermissible (e.g., sex between those with the same sex organs, intercourse involving a constructed vagina, sex with contraception, and masturbation). To reach this conclusion, the argument needs two additional premises: (iii) the unitive function can only be accomplished in marriage and (iv) doing that which frustrates or interferes with fulfillment of an activity’s function is morally impermissible.

415

Stavroula Glezakos and Julie Tannenbaum

However, contrary to (iii), the unitive function seems fully realizable in a committed relationship (which can occur among more than two people), regardless of whether explicit vows are made as in a marriage ceremony. And against (iv) Melissa Moschella argues that, while the application of antiperspirant to one’s armpits prevents the “sending of pheromonal cues regarding one’s biological suitability” to potential mates, the use of antiperspirant is surely not a morally problematic act (2019: 266). Finally, as John Corvino (2013) and Andrew Koppelman (2002) have pointed out, (i) is not plausible. Some people’s sexual activity is centered only on the clitoris, which has no reproductive function. In fact many other non-reproductive body parts can provide aimed-for sexual pleasure and satisfaction (whether the person is alone or with multiple others), which supports the claim that, if anything is essential to sexual activity, it is pleasure and satisfaction (Goldman 1977). Other arguments about the ethics of sexual activity begin with a definition of marriage as a comprehensive union, which is then claimed to be composed of exactly two people who achieve “organic bodily union” (Girgis, Anderson, and George, 2012: 255; also Finnis 2001). The latter requires a “bodily good or function toward which their bodies can coordinate, reproduction being the only candidate” (255). The argument continues thus: marriage so understood is a “profound human and social good” and broadening the definition of marriage beyond one male and one female is “weakening” the goodness of marriage and is thus a “tragedy” (George and Lopez 2011). Hence it is “morally obligatory that the state promote and protect [this variety of] marriage” (Lee, George, and Bradley 2011). John Finnis claims that to think other forms of sexual activity are permissible is to be “ordered [i.e., directed] toward an intrinsic moral evil” (2001: 62). It is unclear why “only an act of the generative kind (in the sense just specified) truly unites the spouses at all levels, biologically as well as at the level of feelings and intentions” (Finnis 2001: 65). There are all sorts of ways of uniting biologically—why privilege one over all others? Moreover, even if one grants that marriage (as defined above) is a good, why does believing that other types of unions are also goods (and/or legally or morally permissible) amount to being directed toward an evil or entail a weakening of the goodness of marriage?

28.7

Essence and the Ethics of Sex and Gender

Much of Aristotle’s virtue theory focuses on the nature of human beings. In Metaphysics X.9 and 1058a 29–31, Aristotle suggests that the difference between male and female is not one of essence (form), but rather a difference in matter (or body) (Malink, this volume). And yet, in other writings, he states the opposite, differentiating the essence of male and female human beings,9 and on that basis maintaining that virtues for each differ (Politics III.4). Some contemporary feminists agree with these latter claims, while others maintain that the notion of essence—at least, when it is invoked in accounts of gender—is anti-feminist and oppressive. This last position is not simply a view that is skeptical of essence talk (à la Sullivan 2015); it is the view that essentializing women is harmful to women. Indeed, recent social science indicates it might be harmful to all because it can make us both epistemically and morally worse (see Leslie 2017, as well as Ritchie and Rosario, this volume). Mary Daly (1978) assumes that every human is either male (and so a man) or female (and so a woman), and that the essential nature of women differs from that of men. Although Daly recognizes the existence of trans individuals, she denies that trans women are women (68). Thus, for her, the essence of womanhood appears to be inextricably tied to (unchangeable) biological characteristics.

416

Ethical Value

Daly exhaustively reviews the ways that men have, through their works and institutions, perpetuated the perspective that women’s virtues lie in “femininity”: being chaste, agreeable, and mothering within a constrained domesticity. She emphasizes that “patriarchy attempts to enforce motherhood by bombarding women with propaganda that this is their inevitable destiny” (291). In contrast, Daly takes the true essence of women to be what she calls “the natural (wild) state of femaleness” (31). She exhorts women to: use their awakened gynaesthesia - “a kind of multidimensional/multiform power of sensing/ understanding [their] environment” (316); to hear “the call of the wild” (343), and to seek “out the wild question” (345). “Wild” has many meanings for her: “living in a state of nature: inhabiting natural haunts: … It means ‘not subjected to restraint or regulation: UNCONTROLLED, INORDINATE, UNGOVERNED’” (343). Daly’s essentialist position has faced various criticisms (see Norlock’s 2019 overview). Some feminist authors charge Daly with not supporting her contention that women have (what Allison Jagger 1983 labels in her critique) “special powers” (94). Additional concerns, highlighted in a letter by Audre Lorde (1979/2012) to Daly (1978/2012), include the way in which Daly’s accounts of women’s lives focus almost exclusively on the experiences of women of European ancestry (except when she is describing varieties of victimhood). Her portrayals of aspirational and inspirational Wild Women and Goddesses are similarly limited to those found in European traditions. This call to pay attention to more than white or privileged women’s lives echoes Sojourner Truth’s critique in “Ain’t I a women?” (1851/2013). Further objections target Daly’s denial that trans women are women. Talia Bettcher both denies that biology is essential to gender (2015) and argues (2009) that there is an ethical dimension (a lack of respect) to rejecting someone’s firstperson authority when they sincerely and reflectively maintain, for example, “I am a woman” (see Rosario, this volume). A different way that essence of gender is invoked in feminist thought comes from those who endorse purely social forms of essentialism.10 Like Daly, these feminists (Noddings 1984; Ruddick 1989; Held 1993) identify ways that patriarchal systems and perspectives “construct” a feminine nature that all women share in virtue of being (identified as or socialized as) women. However, unlike Daly, these theorists see “feminine” traits and qualities like caring and nurturance as being positive. They go on to claim that, if men were to acquire these traits (and thus, essentially change), society would be fundamentally different, and better (Chodorow 1978; for alternative social constructionist approaches see Rosario, this volume). In order to ground claims that women are treated in ways that are unjust and intolerable, some authors appeal to a shared human nature (Antony 2022; Nussbaum 1995b). Sally Haslanger, on the other hand, develops an account of ameliorative concepts: concepts for categories like woman that both include the social realities that exist for members of the categories, and can also serve as tools to lead us to an “emancipatory” future. (Haslanger 2000, 2005, 2020; Stoljar, this volume). Finally: many feminist theorists fully reject essence in their theorizing about gender. Simone de Beauvoir (2010) notably denies that anyone is “born” a woman or man, and instead maintains that all human beings are radically free, and thus, able to shed or choose alternate ways of being. Other feminists cite moral reasons for shunning the notion of essence when discussing gender. For example, Stone (2004) points out that when certain features are taken to be paradigmatic of all women, some (e.g., trans women and women of color) are unjustly minimized, judged, or ignored.

417

Stavroula Glezakos and Julie Tannenbaum

Partly from recognition of the harms perpetuated by such essence-based doctrines, “… the term ‘essentialisť has become a term of abuse in feminist writing as well as in university classes in feminist theory” (Stoljar 1995: 261). These moral concerns prompt the question: must we abandon essentialism? Or is there an ethically defensible way to engage in an essentialist project?11

28.8 Ontological Oppression As noted above, some theorists consider disability and gender to be essentially socially constructed kinds (Griffith, this volume). Robin Dembroff and Kevin Richardson have argued that ontological oppression is a distinctive form of wrongdoing that can arise for socially constructed kinds—and so, also for essentially socially constructed kinds. According to Dembroff, ontological oppression “occurs when the social kinds (or the lack thereof) unjustly constrain (or enable) persons’ behaviors, concepts, or affect due to their group membership” (2018: 26). Richardson (2023) suggests there are two forms that ontological oppression can take: ontological exclusion and ontological erasure. Ontological exclusion (Rosario, this volume) wrongs an individual in virtue of unjustly excluding them from membership in a socially constructed group (e.g., when Smith College deemed that trans women were not women and thus ineligible for admission). Ontological erasure wrongs an individual in virtue of leaving it indeterminate whether they are a member of a socially constructed group (e.g., Bryn Mawr College’s admissions policy left it unclear whether trans women were eligible for admission; “the category of trans identity is ignored entirely” [Richardson 2023: 603]).12 Dembroff’s definition of ontological oppression leaves open the location of the injustice. Is it in the membership criteria of a socially constructed group (e.g., one must have certain sex organs to qualify as a member of the kind “woman”), the characterization of that socially constructed group (e.g., to be a woman is to serve the sexual desires of men and thus women cannot be raped), or in how others relate to one in virtue of one’s being a member of that group (denying one the right to vote or access to education)? On Richardson’s definitions of ontological exclusion and erasure, it is the fact that membership conditions exclude, or neither determinately apply nor determinately exclude, that “constitutes an injustice” (2023: 608, see also 603, 604, and 616). And yet, in the examples of Smith and Bryn Mawr, it would seem that the injustice is not merely in failing to be categorized as a woman but also in not being eligible for admission because of this miscategorization.13

28.9 Conclusion In some of the debates discussed in this chapter, the ethical conclusions rely on independent claims about the essence of non-evaluative entities. For example, Aristotle’s conclusion that we should live virtuously depends on his claim about the essence of human beings, and Feser’s conclusions about which type of sexual activity is permissible depends on his claim about the nature of sexual activity as being for procreation. But in other cases, it is ethical commitments that motivate claims about the essence (or non-essence) of non-evaluative entities. For example, working to call attention to and fight against the oppression of women led some feminists to develop alternative conceptions of woman’s essence or to deny that women have an essence. Similarly, the unfair treatment of the disabled motivated some social constructionist views of disability.

418

Ethical Value

28.10

Related Topics

Tahko, Tuomas, “Natural Kind Essentialism” Brigandt, Ingo, “Biological Species” Scarpati, Maria, “Identity, Persistence, and Individuation” Loets, Annina, “Persons” Brown, Danielle, “Psychiatric Kinds” Griffith, Aaron, “Social Construction” Malink, Marko, “Ancient” Rosario, Esther, “Sex and Gender” Stoljar, Natalie, “Social Justice” Ritchie, Katherine, “Language of Essence”

Notes 1 We acknowledge that drawing a distinction between the evaluative and non-evaluative is notoriously difficult ( Foot 1958 and 1959; Putnam 2002; Dancy 1995). 2 Thompson (1995) and Foot (2001: 27, 31–2, 25–37) claim that function has a normative component, e.g., that we look to the function of a kind when making judgments about whether an individual member of that kind is disabled, not in its proper environment, or not yet fully formed. This conception of nature in terms of function allows for the possibility that not every member of a kind can do its function (contrary to some contemporary views described in Tahko, this volume), and yet does not characterize the function as merely a cluster of homeostatic properties (Ingo Brigandt, this volume) 3 In Book X of the Nicomachean Ethics, Aristotle surprisingly suggests that contemplation in accord with its virtue (sophia) is the best life, which seems to be in conflict with the broader account of the best life (living in accord with all the virtues that pertain at any given moment) he arrives at in Books I-VII. His view might be that some virtuous activity is better than others—not in the sense that one can act unjustly in order to have more time for contemplation, but in the sense that if contemplating at t1 is not contrary to any of the other virtues, then that is how one should spend one’s time. 4 Such a transformation might be so gradual that, at some points, it would be metaphysically vague whether the being undergoing it is still a human being. 5 Hume also denied the existence of selves ( Hume 1739–40/1978: 252), but this is not the basis of, nor always consistent with, his metaethical views. In some passages, he offers an error theory of morality (e.g., Hume 1751/1975: 294), which is to deny that there are any true moral claims—not something that a Buddhist would agree with. In other passages, his moral theory seems to be in tension with his no-self view: “when you pronounce any action or character to be vicious, you mean nothing, but that from the constitution of your nature you have a feeling or sentiment of blame from the contemplation of it.” (1739–17/1978: 468–9). 6 While some philosophers in the debate about the moral importance of killing vs letting die do not use the language of essence or nature, others do (e.g., Bennett 1981; Foot 1984; McMahan 1993; Kamm 2007). Regardless, what is at issue are not merely generalizations about each, or even contingent necessary conditions, but rather, their essential difference(s). 7 Does a pregnant person have a parental duty to provide life-sustaining aid to the fetus inside of them? Four grounds of the parental duty to provide life-sustaining aid to children - genetic, laborbased, voluntarist, and causal - and the difficulties with each are examined in Brake and Millum (2021). 8 See also Alexander Pruss, who claims that it is “through biological union that interpersonal union occurs” ( 2012: 6). Sexuality, on his view, is “an objective one-body union” (91) and the “natural goal” of sexual union is “the biological good” of reproduction (132). 9 See Generation of Animals Book I and Politics I.5 and 1.13, as well as Horowitz (1976). 10 As Allison Stone (2004) notes, “Having identified femininity as socially constructed, these theorists sought to identify an invariant set of social characteristics that constitute femininity and that all women, qua women, share. Possibilities included women’s special responsibility for domestic,

419

Stavroula Glezakos and Julie Tannenbaum affective, or nurturant labour (as Nancy Hartsock argued), their construction as sexual objects rather than sexual subjects (according to Catherine MacKinnon) or their relational, contextual and particularist style of ethical and practical reasoning (Carol Gilligan)” (139). 11 See Witt (2011) for a feminist defense of essentialism. 12 Katharine Jenkins’ notion of “ontic injustice” is narrower than Richardson’s interpretation of “ontological oppression.” Jenkins’ notion involves only ontological exclusion, not ontological erasure, and the exclusion must be systematic, significant, and pervasive ( 2020: 191, 202), whereas Richardson does not require these features of ontological exclusion. Ásta’s notion of “categorical injustice” (2019) is silent about the role of ontology of social kinds and is neither applicable to all cases of ontological exclusion nor to any cases of ontological erasure. 13 Richardson also claims that ontological oppression is a form of agential identity discrimination (a concept introduced by Dembroff and Saint-Croix 2019–2020). One’s agential identity is a selfidentification that one intends to externalize (i.e., have others to take up). Trans women applicants to Smith and Bryn Mawr were, as Richardson puts it, “discriminated against on the basis of attempting to externalize” their self identity (13). This way of putting it implies that the injustice was not merely the institution’s refusal to take up another’s self identity. A question worth pursuing is what types of refusals to take up another’s self-identity count as wrongdoing (e.g., consider the difference between self-identifying as a woman as opposed to as a dominator or genius).

Works Cited Altman, B. (2001) “Definitions, Models, Classifications, Schemes, and Applications,” in G. Albrecht, K. Seelman, & M. Burry (eds.) Handbook of Disability Studies, Thousand Oaks: Sage Publications, 97–122. Annas, G., Andrews, L., & Isasi, R. (2002) “Protecting the Endangered Human: Toward an International Treaty Prohibiting Cloning and Inheritable Alterations,” American Journal of Law and Medicine 28(2/3), 151–178. Antony, L. (2022) “‘Human Nature’ and Its Role in Feminist Theory,” in Only Natural: Gender, Knowledge, and Humankind, Oxford: Oxford University Press, 269–298 Aristotle (1984) The Complete Works of Aristotle, J. Barnes (ed), vols. 1 and 2, Princeton, NJ: Princeton University Press. Ásta. (2019) “Categorical Injustice,” Journal of Social Philosophy 50(4), 392–406 Barnes, E. (2016) The Minority Body: A Theory of Disability, Oxford: Oxford University Press. Beebee, H. (2018) “Should People Be Punished for Crimes They Can’t Remember Committing?” Iai news, March 20. < https://iainews.iai.tv/articles/should-people-be-punished-for-crimes-they-cantremember-committing-what-john-locke-would-say-about-vernon-madison-auid-1050?access=ALL? utmsource=Reddit>. Bennett, J. (1981) “Morality and Consequences,” in S. McMurrin (ed.) The Tanner Lectures on Human Values, vol II, Cambridge: Cambridge University Press, 45–116. Bettcher, T. (2009) “Trans Identities and First-Person Authority,” in L. Shrage (ed.) You’ve Changed: Sex Reassignment and Personal Identity, Oxford: Oxford University Press, 98–120. Bettcher, T. (2015) “Intersexuality, Transgender, and transsexuality,” in L. Disch & M. Hawkesworth (eds.) The Oxford Handbook of Feminist Theory, Oxford: Oxford University Press, 407–427. Boorsay, A. (2005) Disability and Social Policy in Britain Since 1750: A History of Exclusion, Basingstoke: Palgrave. Bostrom, N. (2003) “Human Genetic Enhancements: A Transhumanist Perspective,” The Journal of Value Inquiry 37(4), 493–506. Brake, E. & Millum, J. (2021) “Parenthood and Procreation,” in E. Zalta (ed.) The Stanford Encyclopedia of Philosophy < https://plato.stanford.edu/archives/sum2021/entries/parenthood/>. Carlson, L. (2001) “Cognitive Ableism and Disability Studies: Feminist Reflections on the History of Mental Retardation,” Hypatia 16(4), 124–146. Chodorow, N. (1978) The Reproduction of Mothering, Berkeley: University of California Press. Corvino, J. (2013) What’s Wrong with Homosexuality? New York: Oxford University Press. Daly, M. (1978) Gyn/Ecology: The Metaethics of Radical Feminism, Boston: Beacon Press. Dancy, J. (1995) “In Defense of Thick Concepts,” Midwest Studies in Philosophy 20(1), 263–279. De Beauvoir, S. (2010) The Second Sex, New York: Knopf Doubleday Publishing Group.

420

Ethical Value Dembroff, R. (2018) “Real Talk on the Metaphysics of Gender,” Philosophical Topics 46(2), 21–50. Dembroff, R. & Saint-Croix, C. (2019-2020) “‘Yep, I’m Gay’: Understanding Agential Identity,” Ergo 6(20), 571–599 Emmanuel, S. (2015) “Introduction,” in S. Emmanuel (ed.) A Companion to Buddhist Philosophy, Chichester: Wiley-Blackwell, 1–10. Feser, E. (2013) “The Role of Nature in Sexual Ethics,” The National Catholic Bioethics Quarterly 13(1), 69–76. Finnigan, B. (2017) “Buddhism and Animal Ethics,” Philosophy Compass 12(7), 1–12. Finnigan, B. (2018) “Madhyamaka Ethics,” in D. Cozort & J. Shields (eds.) The Oxford Handbook of Buddhist Ethics, Oxford: Oxford University Press, 162–183. Finnis, J. (2001) “Reason, Faith and Homosexual Acts,” Catholic Social Science Review 6, 61–69. Foot, P. (1958) “Moral Arguments,” Mind 67(268), 502–513. Foot, P. (1959) “Moral Beliefs,” Proceedings of the Aristotelian Society 59(1), 83–104. Foot, P. (1977) “Euthanasia,” Philosophy and Public Affairs 6(2), 85–112. Foot, P. (1984) “Killing and Letting Die,” in J. Garfield & P. Hennessy (eds.) Abortion: Moral and Legal Perspectives, Amherst: University of Massachusetts Press, 177–185. Foot, P. (2001) Natural Goodness, Oxford: Clarendon Press. Garfield, J. (2022a) Losing Ourselves: Learning to Live Without a Self, Princeton: Princeton University Press. Garfield, J. (2022b) Buddhist Ethics, Oxford University Press George, R. & Lopez, K. (2011) “Sex and the Empire State,” National Review, June 28 < https://www. nationalreview.com/2011/06/sex-and-empire-state-interview/>. Girgis, S., Anderson, R., & George, R. (2012) What Is Marriage? Man and Woman: A Defense, New York: Encounter Books. Goldman, A. (1977) “Plain Sex,” Philosophy & Public Affairs 6(3), 267–287. Goodman, C. (2002) “Resentment and Reality: Buddhism on Moral Responsibility,” American Philosophical Quarterly, 39(4), 359–372 Haslanger, S. (2000) “Gender and Race: (What) Are They? (What) Do We Want Them To Be?” Noûs 34 (1), 31–55. Haslanger, S. (2005) “What are we talking about? The semantics and politics of social kinds,” Hypatia 20(4), 10–26. Haslanger, S. (2020) “Going On, Not in the Same Way,” in C. Burgess and Plunkett (eds) Conceptual Engineering and Conceptual Ethics, Oxford: Oxford University Press, 230–260. Held, V. (1993) Feminist Morality: Transforming Culture, Society, and Politics, Chicago: University of Chicago Press. Horowitz, M. (1976) “Aristotle and Woman,” Journal of the History of Biology 9(2), 183–213. Hume, D. (1751/1975) Enquiries Concerning Human Understanding and Concerning the Principles of Morals, L. A. Selby-Bigge (ed), revised by P. H. Nidditch, 3rd edition, Oxford: Clarendon Press. Hume, D. (1739–1740/1978) A Treatise of Human Nature, L. A. Selby-Bigge (ed), revised by P. H. Nidditch, 2nd edition, Oxford University Press, 1978. Hursthouse, R. (1999) On Virtue Ethics, Oxford: Clarendon Press. Jaggar, A. (1983) Feminist Politics and Human Nature, Totowa: Rowman and Allanheld. James, S. (2003) “Zen Buddhism and the Intrinsic Value of Nature,” Contemporary Buddhism 4(2), 143–157. Jenkins, K. (2020) “Ontic Injustice,” Journal of the American Philosophical Association, 6(2), 188–205. Juengst, E. (1998) “What Does Enhancement Mean?” in E. Parens (ed.) Enhancing Human Traits: Ethical and Social Implications, Washington, DC: Georgetown University Press, 29–47. Kafer, A. (2013) Feminist, Queer, Crip, Bloomington: Indiana University Press. Kamm, F. (2007) Intricate Ethics, New York: Oxford University Press. Koppelman, A. (2002) The Gay Rights Question in Contemporary American Law, Chicago: University of Chicago Press. Lawrence, G. (2001) “The Function of the Function Argument,” Ancient Philosophy 21(2), 445–475. Lawrence, G. (2006) “Human Good and Human Function,” in R. Kraut (ed.) The Blackwell Guide to Aristotle’s Nicomachean Ethics, Oxford: Oxford University Press, 37–75. Lawrence, G. (2009) “Is Aristotle’s Function Argument Fallacious? Part 1, Groundwork,” Philosophical Inquiry 31(1/2), 191–224.

421

Stavroula Glezakos and Julie Tannenbaum Lee, P., George, R., & Bradley, G. (2011) “Marriage and Procreation: Avoiding Bad Arguments,” Public Discourse, March 30 < https://www.thepublicdiscourse.com/2011/03/2637/> Leslie, S. J. (2017) “The Original Sin of Cognition,” The Journal of Philosophy 114(8), 395–421. Locke, J. (1689/1975) An Essay Concerning Human Understanding in P. Nidditch (ed.) The Clarendon Edition of the Works of John Locke, Oxford: Oxford University Press. Lorde, A. (1979/2012) Sister Outsider: Essays and Speeches, Trumansburg: Crossing Press. Maher, B. (2008) “Poll Results: Look Who’s Doping,” Nature 452(7188), 674–675. McMahan, J. (1993) “Killing, Letting Die, and Withdrawing Aid,” Ethics 103(2), 250–279. Moschella, M. (2019) “Defense of the New Natural Law,” The National Catholic Bioethics Quarterly 19(2), 251–278. Morris, J. (2001) “Impairment and Disability: Constructing an Ethics of Care that Promotes Human Rights,” Hypatia 16(4), 1–16. Nagel, T. (1979) “Death,” in Mortal Questions, Cambridge: Cambridge University Press, 1979. 1–10. Nanamoli, B. & Buddhaghosa, B. (2010) Visuddhimagga: The Path of Purification, Buddhist Publication Society. Noddings, N. (1984) Caring: A Feminine Approach to Ethics and Moral Education, Berkeley: University of California Press. Norlock, K. (2019) “Feminist Ethics,” in E. Zalta (ed.) The Stanford Encyclopedia of Philosophy < https://plato.stanford.edu/entries/feminism-ethics> Nussbaum, M. (1995a) “Aristotle on Human Nature and the Foundations of Ethics,” in J. Altham, & R. Harrison (eds.) Mind, and Ethics: Essays on the Ethical Philosophy of Bernard Williams, Cambridge: Cambridge University Press, 86–131. Nussbaum, M. (1995b) “Human Capabilities, Female Human Beings”, in M. Nussbaum and J. Glover (eds.) Women, Culture and Development: A Study of Human Capabilities, Oxford and New York: Clarendon Press and Oxford University Press, 61–104 Parens, E. (1995) “The Goodness of Fragility: On the Prospect of Genetic Technologies Aimed at the Enhancement of Human Capacities,” Kennedy Institute of Ethics Journal 5(2), 141–153. Parfit, D. (1984) “The Non-Identity Problem.” in Reasons and Persons, Oxford: Oxford University Press, 351–379. Persad, G., Wertheimer, A., & Emanuel, E. (2009) “Principles for Allocation of Scarce Medical Interventions,” Lancet 373(9661), 423–431. Pruss, A. R. (2012) One body: An Essay in Christian Sexual Ethics, Indiana: University of Notre Dame Press. Putnam, H. (2002) The Collapse of the Fact/Value Dichotomy and Other Essays, Cambridge: Harvard University Press. Rachels, J. (1975) “Active and Passive Euthanasia,” The New England Journal of Medicine 292(2), 78–80. Richardson, K. (2023) “Exclusion and Erasure: Two Types of Ontological Oppression,” Ergo 9(23), 603–622. Ruddick, S. (1989) Maternal Thinking: Towards a Politics of Peace, New York: Ballantine Books. Sahakian, B., & Morein-Zamir, S. (2007) “Professor’s Little Helper,” Nature 450(7173), 1157–1159. Schwartz, P. (2005) “Defending the Distinction between Treatment and Enhancement,” American Journal of Bioethics 5(3), 17–19. Scotch, R. & Schriner, K. (1997) “Disability as Human Variation: Implications for Policy,” The Annals of the American Academy of Political and Social Science 549(1), 148–159. Siderits, M. (2021) Buddhism as Philosophy, Indianapolis/Cambridge: Hackett Publishing Company. Stoljar, N. (1995) “Essence, Identity, and the Concept of Woman,” Philosophical Topics 23(2), 261–293. Stone, A. (2004) “Essentialism and Anti-Essentialism in Feminist Philosophy,” Journal of Moral Philosophy 1(2), 135–153. Sullivan, M. (2015) “The Irrelevance of Essence,” Philosophy and Phenomenological Research 91(2), 499–507 Talbot, M. (2009) “Brain Gain,” The New Yorker, April 27, 32–43. Thomson, J. (1971) “A Defense of Abortion,” Philosophy & Public Affairs 1(1), 47–66. Thompson, M. (1995) “The Representation of Life,” in R. Hursthouse, G. Lawrence, & W. Quinn (eds.) Virtues and Reasons, Oxford: Clarendon Press, 247–296. Tooley, M. (1972) “Abortion and Infanticide,” Philosophy and Public Affairs 2(1), 37–65.

422

Ethical Value Tooley, M. (1994) “An Irrelevant Consideration: Killing Versus Letting Die,” in A. Norcross, & B. Steinbock (eds.) Killing and Letting Die, New York: Fordham University Press, 103–111. Tremain, S. (2017) Foucault and Feminist Philosophy of Disability, Ann Arbor: University of Michigan Press. Truth, S. (1851/2013) “Ain’t I a Woman?” in W. Kolmar & F. Bartkowski (eds.) Feminist Theory: A Reader, New York: McGraw-Hill. UPIAS. (1976) Fundamental Principles of Disability, London: Union of the Physically Impaired Against Segregation. Wasserman, D. & Aas, S. (2022) “Disability: Definitions and Models,” in E. Zalta (ed.) The Stanford Encyclopedia of Philosophy < https://plato.stanford.edu/archives/sum2022/entries/disability/>. Wendell, S. (1996) The Rejected Body: Feminist Philosophical Reflections on Disability, London: Routledge. Williams, B. (1972) Morality: An Introduction to Ethics, Cambridge: Cambridge University Press. Williams, B. (1981) “Moral Luck” in Moral Luck, Cambridge: Cambridge University Press, 20–39. Williams, P. (1998) Altruism and Reality: Studies in the Philosophy of the Bodhicaryavatara, London: Routledge. Woollard, F. & Howard-Snyder, F. (2016) “Doing vs. Allowing Harm,” in E. Zalta (ed.) The Stanford Encyclopedia of Philosophy < https://plato.stanford.edu/archives/win2016/entries/doing-allowing/>. Young, I. (1980) “Throwing Like a Girl: A Phenomenology of Feminine Body Comportment, Motility and Spatiality,” Human Studies 3(1), 137–156.

423

PART 4

Anti-essentialist Challenges Kathrin Koslicki and Michael J. Raven

Introduction Since the notion of essence was introduced into philosophical discourse, it has been subject to continuous skeptical challenges. These challenges may come in many various forms. One form challenges the intelligibility of the notion of essence. A second form challenges whether essences are real, as opposed to being mere projections of our language, mind, or conventions onto reality. A third form challenges whether essences are knowable, at least by creatures like us. And a fourth form challenges whether there is any point or usefulness in appealing to essences. Each of these skeptical challenges alone may serve to motivate anti-essentialism. The challenges may also be combined in various ways. They may even be viewed in progression: even if it is granted that the notion of essence is intelligible, it may be doubted whether essences are real; even if essences are real, it may be doubted whether they are knowable by us; and even if essences are knowable by us, it may be doubted whether appealing to them serves any useful purpose. Not all skeptical challenges neatly fit into just one, or even a combination, of the forms above. But most are plausibly regarded as drawing from them in some way or another. In Part 4, no attempt is made to survey comprehensively all of the skeptical challenges that have, or could, arise. The focus, instead, is on some especially influential skeptical challenges prominent in the contemporary literature. In Chapter 29, Kit Fine’s “Quinean Anti-essentialism” focuses on one of the most prominent opponents of essentialism in contemporary times: W. V. Quine. In a series of influential works, Quine presented various skeptical challenges to what he called “Aristotelian essentialism”. Quine, like many philosophers of his generation, assumed that essence was a modal notion. His empiricist scruples against modality largely motivated his skepticism of essence. Fine himself, along with Kaplan, Kripke, and others, gave influential responses to Quine’s challenges. Fine’s chapter focuses on what’s most distinctive of Quine’s skeptical challenges. Ironically, Fine suggests that Quine’s own skepticism of modality may have required him to appeal to a nonmodal notion of essence. In Chapter 30, Jonathan Livingstone-Banks and Alan Sidelle’s “Conventionalism” focuses on conventionalism about essences. This view allows for truths about the essences of things, but demotes them as merely conventional truths. The essentialist truths are ultimately, in

DOI: 10.4324/9781003008750-33

Kathrin Koslicki and Michael J. Raven

some way or another, projections of our own languages, minds, or conventions upon the world. Older versions of conventionalism were often formulated in terms of the notion of analyticity which, at least back then, was thought to be equivalent to necessity and a priority. But these formulations have needed updating ever since Kripke’s and Putnam’s influential work disentangling analyticity from necessity and a priority. Livingstone-Banks and Sidelle discuss some prominent updates of conventionalism. They also argue that these updated formulations are more resilient to familiar objections often raised against conventionalism. In Chapter 31, Aaron Griffith’s “Social Construction” focuses on social construction and whether it poses a challenge to essence. In recent decades, it has become increasingly common to find theorists claiming that kinds (races, genders), institutions (nations, courts), or objects (money, borders) are socially constructed. The claim that some item is socially constructed is often (although not always) intended to convey that it, or its presumed hallmark features, are not inevitable. Had society gone differently, the item would have been socially constructed in a different way, or to have different hallmark features, or perhaps would not have been constructed at all. Many have thought that this shows that socially constructed items do not have essences. Griffith’s chapter explores the notion of social construction, considers its alleged conflict with essence, and suggests ways to reconcile social construction with essentialism. In Chapter 32, Anand Vaidya and Michael Wallner’s “Conferralism” focuses on an approach to essence that allows for essentialist truths but takes them to be radically contextsensitive. The mechanism of context-sensitivity is conferral: whether an item essentially has some feature depends on whether, in a given context, that feature is conferred upon it as essential. A feature conferred as essential in one context may not be conferred as essential in some other context. And none of an item’s features will be essential independently of some such context of conferral. Although this approach allows for essentialist truths, it may seem that it takes an anti-realist stance toward them precisely because of their dependence on context-sensitive conferrals. This conferralist approach has been defended in print by Ásta, whose account is also the focus of Vaidya and Wallner’s chapter. They also consider whether it, in particular, is an anti-realist view of essence and, more generally, what differentiates between realist and anti-realist views of essence. In Chapter 33, Arata Hamawaki’s “Wittgenstein” focuses on Wittgenstein’s influence on essence. Wittgenstein’s work is often (although not uncontroversially) divided into an early epoch exemplified in Tractatus Logico-Philosophicus and a later epoch exemplified in Philosophical Investigations. The notion of essence figures, at least implicitly, in both but in different ways. In Wittgenstein’s earlier work, essence seems to concern the form or structure that must be shared by language and the world, if language is to represent the world. In Wittgenstein’s later work, essence seems to be a confused ideal refuted by his famous discussion of “family resemblances”. Hamawaki’s chapter discusses how essence figures into Wittgenstein’s earlier and later thought, and whether the considerations of family resemblance pose a skeptical challenge to essence in the way they are often thought to do so.

426

29 QUINE ON ESSENCE Kit Fine

We may distinguish two kinds of skepticism about modality. One may be a de dicto skeptic and object to the application of modal notions to dicta, or sentences; or one may be a de re skeptic and object to the application of modal notions to res, or things. Quine was skeptical on both counts. He would have found it problematic to say “the sentence ‘all bachelors are unmarried’ is necessarily true” (or “it is necessary that all bachelors are unmarried”) and he would also have found it problematic to say “the predicate ‘is rational’ is necessarily true of Socrates” (or “for any x identical to Socrates, it is necessary that x is rational”). However, for the purposes of arguing for de re skepticism, Quine was willing to concede the case for de dicto modality. His point, then, was that in addition to the reasons for being skeptical of de dicto modality, there were further reasons for being skeptical of de re modality. It is the second of these two forms of skepticism that is most relevant to the topic of essence. For on the standard modal construal of essence, to say Socrates is essentially rational is to make a de re modal claim; it is to say that for any x identical to Socrates, it is necessary that x is rational or, at least, necessary that x is rational if x exists. However, let me first make a few remarks about the de dicto notion, which will help set the subsequent discussion in context (Section 29.1). I shall then discuss Quine’s views on the de re notion (Section 29.2) and finally discuss two topics which perhaps have not been given the attention they deserve: Quine’s views on the non-strict modalities (Section 29.3), and the bearing of his views on the non-modal notion of essence (Section 29.4). There are already many excellent detailed accounts of Quine’s views on de re and de dicto modality,1 and my principal concern in the present chapter is not to add to them or to survey the field but to point to certain general features of Quine’s views which, in the fog of detail, may easily be overlooked.

29.1

De Dicto Modality

Hume was the first great skeptic of modality in modern philosophy. For him, necessity resides not in the world but in the mind, not in “matters of fact” but in “relations of ideas”. The logical positivists gave a linguistic twist to Hume’s skepticism. They replaced Hume’s relations of ideas with analytic truths, sentences true in virtue of meaning. For them, necessity resided not in the mind, but in language. Quine took the retreat from Hume one step further.

DOI: 10.4324/9781003008750-34

427

Kit Fine

For he questioned the very distinction between analytic and synthetic truths that the logical positivists had relied on; once the analytic truths had been denuded of meaning, all that remained by way of necessity were the logical truths. Quine’s criticisms of the analytic/synthetic distinction in his paper “Two Dogmas of Empiricism” (Quine 1951) have received a great deal of flak. But I believe the impetus behind his criticisms has often been misunderstood. One of the most important things about the paper is its title. For Quine wrote the paper as a friend, not a foe, of empiricism; he was saying that, as a good empiricist, we should not accept these dogmas. It would not do, for example, to explain analyticity or some form of necessity as immunity from revision since, even if this idea could be stated in empirically acceptable terms, there would be no good reason for an empiricist to think that analytic or even logical truths would be immune from revision. Failing an empirically acceptable characterization of the analytic/synthetic distinction, Quine’s response was “so much the worse for the distinction”, though my own response would have been “so much the worse for empiricism, at least in the form envisaged by Quine”. It is worth noting in this connection that it is not even clear that, for Quine, a logical truth, such as “if it is raining then it is raining”, is a genuine form of a necessary truth—of a truth that, in a robust sense of the phrase, could not be otherwise. For the logical truth of “if it is raining then it is raining” simply amounts for him to the truth of all substitution-instances of “if p then p”. It is therefore a paradigm of an empiricist reduction of what might appear to be something modal to a form of regularity, with universality in place of necessity. Of course, someone might believe that each logical truth in Quine’s sense is genuinely necessary, that it could not have failed to have been otherwise. But this connection with necessity is not built into the notion itself, and it is open to someone impervious to modal distinctions to reject the connection simply on the grounds that there is no meaningful notion of necessary truth. Perhaps something similar could be said of Quine’s notion of an analytic truth, were it taken to be in good standing. For an analytic truth for Quine is one that arises from a logical truth upon the substitution of synonyms for synonyms; if the logical truth is not genuinely necessary then it is not clear why any of the analytic truths that arise from it should be genuinely necessary. It might even be maintained that the notion of synonymity, or the related notion of definition, are not themselves genuinely modal; they merely concern the actual relations between two expressions and do not in any way presuppose a robust distinction between what could or could not be so. In this case too, it is then open for someone to accept the Quinean notion of analytic truth and yet reject the notion of necessary truth. If this is right, then it points to an important difference between Hume’s and the logical positivists’ rejection of substantive necessary truths and Quine’s rejection of the analytic/ synthetic distinction. In both cases, the rejection is fueled by empiricist concerns. But in the former case, these concerns can be seen to arise from modal misgivings, from the idea that there can be no genuine form of necessity in the world, while, in the latter case, they can be seen to arise from semantical misgivings, from the idea that there can be no genuine form of meaning within language. It is in large measure this shift in focus from an empiricist critique of modality to an empiricist critique of meaning that makes Quine’s version of empiricism so interesting.

29.2

De Re Modality

I turn to Quine’s critique of de re modality. Many philosophers prior to Quine had also been skeptical of de re modality; and in this regard, I might refer the reader to the quotation

428

Quine on Essence

from Locke in Anat Schechtman’s chapter and to the quotation from James in Howat Andrew’s chapter. What made Quine’s criticisms distinctive and breathed fresh life into the controversy was the development of quantified modal logic (cf. Chapter 6). For the symbolism of quantified modal logic suggested that if we had an understanding of the modal operator and an understanding of the quantifier then the two could combine to provide us with an understanding of de re modality. Quine then argued that this apparently compelling line of argument would not work since our understanding of the modal operator might be limited to its application to the closed sentences of the language and might therefore provide us with no understanding of its application to the open sentences of the language. Quine in fact developed two lines of argument against de re modality.2 The first should be seen to hold against the very formulation of a de re modal attribution and to hold regardless of which notion of necessity might be in question. The second is more in accord with the traditional critiques and should be seen to hold against the very idea of a de re modal attribution regardless of how it might be formulated. According to the first line of argument, singular terms within modal contexts are subject to failures of substitution. Consider the sentence: (1) necessarily, 9 > 7 Upon substituting “the number of planets” for “9”, we obtain: (1)ʹ necessarily, the number of planets > 7. But (1) is true and (1)ʹ is false. Moreover, the identity statement: (2) 9 = the number of planets (we may suppose) is also true. We therefore have a substitution failure; truth is not preserved under the substitution of co-referential terms. It may then be argued (as in Quine 1953, 1961—hereafter referred to as TGMI and RM) that in the face of such a failure, we cannot meaningfully quantify into the position occupied by the given term. Thus we cannot, in the given case, meaningfully say either: (3) ∃x(x = 9 & □(x > 7)), or (3)ʹ ∃x(x = the number of planets & □(x > 7)) since, given the identity in (2), (3) and (3)ʹ should have the same truth value and yet it remains undetermined whether we should say (3) is true on the basis of the truth of (1) or that (3)ʹ is false on the basis of the falsity of (1)ʹ. My own view (Fine 1989, 1990) is that, despite its seeming plausibility, this line of argument should be rejected; others, such as Kaplan (1968, 1986) have agreed. But we should note that it is at best an argument against the standard symbolism of quantified modal logic, in which we are allowed to quantify into a sentential context governed by □. It has no force against the more natural formulations of de re modal claims, in which we would say, not that some object x identical to 9 is such that necessarily x is greater than 7 (∃x(x = 9 ∧ □(x > 7)), but that 9 is necessarily greater than 7, with the idiom of necessity governing the copula or the predicate rather than the whole sentence (□λx(x > 7)[9]); for the same reason, it will not work against the more natural formulations of essentialist claims, in which we talk of the essential or accidental properties of some object rather than of something being so in virtue of the nature of that object.

429

Kit Fine

Quine’s second line of argument operates specifically against the “strict” modalities. “The general idea of strict modalities”, he explains (RM, 143), “is based upon the putative notion of analyticity as follows: a statement of the form ‘Necessarily …’ is true if and only if the component statement which ‘necessarily’ governs is analytic”. As will become clear, the arguments Quine uses against the strict analytic modality work equally well against the strictly logical modality, under which a statement of the form “‘Necessarily …’ is true if and only if the component statement which ‘necessarily’ governs is logically true”. Thus, even though Quine would have looked much more favorably on the strictly logical than on the strictly analytic modalities, he would still have had the same misgivings about their de re application. It is important to recognize that the above equivalences will not fix the use of the modal claims on the left in terms of the claims of analytic or logical truth on the right. This might have been so had we regarded “it is necessary that P” as merely elliptical for, or abbreviatory of, “P is analytic” (a suggestion considered by Quine in TGMI, 162).3 Treating the modal operator in this way would have made short work of the argument against quantifying in, since this would have made it a case of quantifying into a quotational context; Quine is not so naive as to think that the argument against his opponents could be so straightforward (RM, 144; TGMI, 172). For this reason, presumably, he merely insists upon a material equivalence; he does not even specify how, on the right, the “component statement” is to be designated, which would have been required had we had a genuine ellipsis. This means that the equivalences in question will only determine the truth values of the non-iterated de dicto claims, not the iterated de dicto claims.4 Furthermore, almost any determination of the truth values of the de dicto claims, iterated or not, will be compatible with different determinations of the de re truths.5 Thus, the equivalences will, at best, provide a constraint on the intended sense of the modal operators, but one which is compatible with many different extensions of the class of non-iterated de dicto truths to the iterated de dicto truths or to the de re truths. I presume that not all of these extensions could plausibly be regarded as consonant with the interpretation of □ as a strict modality. Under one such extension, for example, ∀x(φ ⊃ □(∃y(y = x) ⊃ φ) would be true for any condition φ.6 This would effect the collapse of actual predication (∃y(y = x) ∧ Fx) to essential predication (□(∃y(y = x) ⊃ Fx)), though not the more general collapse of actual truth to necessary truth. However, I take it that we would not want it to be necessary that the world be as it is if the individual Socrates exists under the strict analytic sense of necessity, or under any sense of necessity for that matter. Quine’s second line of argument comes out of his attempt to avoid any appeal to singular terms in the first line of argument (this creates the impression that the second line of argument is merely a version of the first whereas, in my opinion, it is better seen as an independent line of argument). In RM, p. 149, he writes “Necessary greaterness than 7 makes no sense as applied to a number x; necessity attaches only to the connection between ‘x > 7’ and the particular method (32) [x = √x + √x + √x ≠ x], as opposed to (33) [x is the number of planets], of specifying x”. But although Quine here has the formulation □(x > 7) in mind, it is clear that the same criticism would apply if “□” attached to the predicate and we wanted to make sense, not of □(x > 7), but of □λx(x > 7). Indeed, he later writes “necessary fulfilment of (34) [‘if there is life on the Evening star there is life on x’] makes no sense”, where the use of □ either as a sentence operator or predicate modifier is not even in question. Later in the same paper, when responding to Smullyan, he comes across as a little more concessive. He writes (p. 155), “the only hope [of sustaining quantified modal logic] lies in … insisting … that the object x in question [the number 9] is necessarily greater than 7”.

430

Quine on Essence

However, this turns out not to be much of a concession since it “means adopting an invidious attitude towards certain ways of uniquely specifying x … as better revealing the ‘essence’ of the object”. If Quine were right in claiming that the proponent of quantified modal logic would have to insist that the object 9 was necessarily greater than 7, then the game would be up for her. For in the intended sense of necessity as a strict analytic modality, it is clear that 9 is not necessarily greater than 7. For numbers have no meaning and therefore do not differ in meaning; so there could be no consideration of meaning that could assure us that the predicate “greater than 7” was true of the number 9 yet not true of the number 5. But there is, of course, no good reason the proponent of quantified modal logic who had the strict conception of modality in mind would have to accept that 9 was necessarily greater than 7, plausible as this may be for other conceptions of modality. I suspect that what Quine has in mind at this point of the argument is the following. A discriminatory attitude towards different ways of specifying an object would not be justified under the intended strict interpretation of the modalities. It is, as he puts it, “abruptly at variance with the idea … of explaining necessity by analyticity” (RM, 155; also TGMI, 174). Therefore the only hope of justifying such an attitude would be to adopt some other, nonstrict, interpretation of the modalities and, presumably under such an interpretation, 9 would be necessarily greater than 7. But such a re-interpretation of the modalities would be bad, not only because it gave up on the idea of explaining necessity by analyticity, but because it would require a “reversion to Aristotelian essentialism”, under which “an object, of itself, and by whatever name or none, must be seen as having some of its traits necessarily and others contingently”. But why does Quine think that the discriminatory attitude towards specifying an object would not be justified under the strict interpretation of the modalities? He distinguishes two ways of specifying an object—by singular terms and by conditions (RM, 149). Given an open statement φ(x) in one free variable x, we might take it to be analytically true of the object a under the first method, when φ(t) is analytic for t the term by which a is specified or we might take φ(x) to be analytically true of the object a under the second method when the statement ∀x(ψ(x) ⊃ φ(x)) is analytic for ψ(x) the condition (in one free variable x) by which a is specified. The second method appears to be more general than the first, since when a has been specified by the term t under the first method, it can equally well be specified by the condition x = t under the second method while, when a has been specified by the condition ψ(x) under the second method, it is not so clear that it can equally well be specified by the description ιxψ(x) under the first method, since the condition ψ(x) might be true of many objects or of no objects at all. In any case, let us for now confine our attention to the second method. Then, in order to secure the truth of ∀x∀y(x = y ⊃ □(x = y)), which would seem to be required for a proper understanding of the quantifiers (RM, 156; TGMI, 173), the condition □(x = y) should be analytically true of the pair of objects a, a. Under the natural extension of the second method to conditions containing several variables, this would then require that the sentence ∀x∀y(ψ(x) ∧ ψ(y) ⊃ x = y) be analytic, for ψ(x) the condition by which a is specified. Given that ψ(x) is true of the object a, it follows that it must be uniquely true of a, and so the possible difference between the two methods disappears, just as Quine had presupposed when he talked of “uniquely [my italics] specifying x”. But given that the condition ψ(x) must be uniquely true of the object, it is hard to see how the choice of ψ(x) over any other uniquely specifying condition χ(x) could be justified. If one takes ψ(x) to be the condition x = t, for example, then why choose t rather than some other term by which the object could be designated? The invidious choice of condition will lead,

431

Kit Fine

moreover, to an invidious discrimination in what is necessarily true, for whereas ∀x(ψ(x) ⊃ □ψ(x)) will be true under the choice of ψ(x), it may well not be true under the equally good choice of the condition χ(x). Thus, even if Quine might have been mistaken in thinking that his opponent is obliged to specify the number 9 in one particular way, it still seems that he was correct in thinking that she would be forced into making an invidious distinction in whichever way was chosen. This defense of Quine, however, overlooks a lacuna in his account of how the objects are to be specified. For Quine assumes that the objects are to be specified one by one, by means either of a singular term or a monadic condition. But when several objects are in question, when we are talking of the necessary fulfilment of a condition φ(x, y, …) by a number of objects a, b, …, then there would appear to be no good reason why the objects should be specified by means of monadic conditions φ1(x), φ2(x), … which apply separately to the objects a, b, …, rather than by a polyadic condition φ(x, y, …) which applies jointly to a, b, … . It was perhaps because Quine initially considered specifying the objects by means of singular terms and then, in an attempt to eliminate reference to singular terms, turned to the corresponding specification of the objects by means of monadic conditions—or it was perhaps because he had the model of substituting synonyms for synonyms in the definition of analytic truth—that he overlooked the possibility of specifying the objects by means of a polyadic condition.7 But, once we are allowed to specify the objects jointly rather than separately, the equivalence between the two methods—by means of singular terms or by means of conditions—disappears. Indeed, there would then appear to be a non-invidious way of specifying the objects so as to secure the truth of ∀x∀y(x = y ⊃ □(x = y)). For surely, when analytic or logical truth is in question, there is nothing invidious in allowing ourselves to specify the objects in question by means of purely logical predicates and by means of the identity predicate in particular. So let it be granted that the pair a, a may be specified by means of the condition (x = y).8 Indeed, this condition constitutes a full logical description of the pair a, a, although one that applies equally to any other pair b, b. The truth of ∀x∀y(x = y ⊃ □(x = y)) is then assured since ∀x∀y(x = y ⊃ x =y) is a logical truth.9 Not only is the use of logical predicates in specifying objects not invidious in this case, it also strikes me as extremely natural and not in the slightest “at variance with the idea … of explaining necessity by analyticity” (RM, 155). Thus, it seems to me that Quine’s second line of argument against quantifying into the strict modalities also fails to get off the ground.

29.3 The Non-strict Modalities The last two pages of RM (158–9) contain a remarkable concession. For Quine seems open to admitting that one can quantify into a modal context as long as the modality is one of “physical necessity and possibility” (see Chapters 8, 11, 12, 13 and 19 for related discussion). It is actually rather hard to discern how open he is. For he seems to admit, on the one hand, that for the purpose of doing physics we need such locutions as “necessarily, if x is in water then x dissolves” and seems to allow that quantification into such contexts might not “precipitate an ontological crisis”. But he also goes on to state that “we do not yet know whether there is a suitable sense of ‘necessarily’ into which we can so quantify” and to suggest that “the only useful modes of composition susceptible to unrestricted quantification are the truth functions”. The admission, for what it is worth, is remarkable in a number of ways. Any such necessity, if it existed, would presumably be some form of necessity in the world, rather than in language or the mind. Its acceptance would therefore go against the primary motivation

432

Quine on Essence

empiricists have had for rejecting genuine forms of modality. Thus, we may see this more concessive Quine, not as supplementing the traditional empiricist concerns about modality with semantical concerns, but as relinquishing those concerns in favor of the semantical concerns and as of only being opposed to necessity in so far as it is explained in semantic terms. But odder still is Quine’s attempt to sequester this other sense of “necessarily” from the reach of his previous arguments. He begins the last paragraph on p. 158 by stating “what has been said of modality in these pages relates only to strict modality”. Yet is this really so? The first line of argument applies to any sense of modality which gives rise to opaque contexts and, almost as if to emphasize this point, Quine ends the paper by giving a general reason for thinking that it is only truth-functional contexts that will not fall prey to opacity. The second line of argument might be thought to fare a little better. For Quine had previously argued that the de re application of the strict modalities called for an invidious attitude towards how a given object was to be specified; this is because there is nothing in language, or the notion of analytic truth, that could be seen to favor any one specification over another. But now, when it comes to this other “physical” modality, he can maintain that the “world”—the micro-structure of the physical objects, for example—can tell us how the object is to be specified. I do not dispute the reasonableness of this position. But it is to buy into a version of Aristotelian essentialism which Quine had previously excoriated and one might well wonder: if some salt can be soluble in water, i.e. be such that necessarily it dissolves when in water, then why can the number 9 not be such that necessarily it is greater than 7? Indeed, I very much doubt whether, at the end of the day, the one kind of de re necessary truth can be detached from the other. If, furthermore, the question is “what room is there for skepticism about de re modality, especially skepticism of an empiricist variety?”, then Quine’s tentative position at this point strikes me as being the exact reverse of the truth. For as we have seen, when it comes to the strict modalities, there are no reasonable grounds for skepticism, once relational methods of specifying the objects are taken into consideration. However, when it comes to the physical modalities, there do exist serious reasons for being skeptical of their de re application. For even if the empiricist is willing to accept the physical necessity of physical laws, these laws are all de dicto in content and provide no basis for its de re application. It may be a matter of necessity that any salt will dissolve when in water, but that is not to say, or in any way to explain how it is, that any salt will, as a matter of necessity, dissolve when in water. Thus, it is here, in connection with the non-strict modalities, that Quinean skepticism gets some traction. It is sometimes thought that Kripke’s introduction of the notion of metaphysical necessity into philosophy provided the answer to Quine’s skeptical doubts about de re modality. Substitute metaphysical necessity for analytic necessity and these doubts disappear. But if I am right, it is with physical necessity, or with what is possibly the stricter notion of metaphysical necessity, that Quinean doubts are at their strongest. This is not to say, of course, that de re modality of the metaphysical sort is not in good order, but merely that there exist grounds for questioning its coherence which cannot be so readily brushed aside as in the case of analytic or logical necessity.

29.4

Essence

Quine, like all others who worked within the framework of quantified modal logic, adopted a modal conception of essence. For Socrates to be essentially rational is simply for whatever

433

Kit Fine

is Socrates to be such that necessarily it is rational or, at least, such that necessarily it is rational if it exists. So, although Quine and others in that tradition might sometimes have talked of essence and accident, such talk is to be understood as a colorful way of making ordinary de re modal attributions. Still, we may ask: what if Quine had distinguished between modality and essence, what then would have come of his objections? There is no doubt that essentialist operators (as used in Fine (1995), for example) create opaque contexts and so his first line of argument would still apply.10 However, the application of the second line of argument and even of his general skepticism towards de dicto necessity is less clear. Quine was against meaning and against analyticity, and he thought of them as belonging to the same circle of ideas. But as we have seen, Quine does not appear to have had the same hostility towards the notion of physical necessity, even in its de re application. Thus, he might have been prepared to admit that objects have natures even if words do not have meanings. Of course, this does not mean that he would have been happy with the idea of the essential nature of a thing where this was somehow tied to the thing itself and was not simply treated as a matter of its necessarily having some property. Indeed, just as he regarded analyticity and meaning as belonging to the same circle of ideas, he might likewise have regarded physical necessity and physical nature as belonging to the same circle of ideas. But what I find interesting is that Quine’s way of thinking of the connection between the de dicto and the de re leads naturally to a distinction between de re modality and essence. For the general idea is that we get de re necessity out of de dicto necessity by plugging in suitable properties, or specifications, for the things. The problem then is to say what those properties are and on what basis the choice of some over others might be justified. But assuming that this problem can be solved, then we not only get a notion of de re necessity out of de dicto necessity but also a distinctive notion of essence or nature. For I can help myself to the properties (and perhaps also the relations) associated with some of the objects but not the others, and then say that something is essential to those objects if it can be seen to be necessary by only taking those properties (and relations) into account.11 Something similar can be said of David Lewis. For Lewis (1968) gets de re modality out of de dicto modality by means of the counterpart relation, which is essentially the same idea as Quine’s specifications but with the difference that the properties or specifications are allowed to vary from context to context and, within a context, from world to world.12 But this then gives us a more discerning notion of essence, since we need only take into account the counterpart relation in so far as it applies to certain of the objects and not to the others.13 Thus, a distinction between essence and modality is already built into the framework of Lewis’ counterpart theory in a way that is not true of the standard models for quantified modal logic, in which the transworld identity of all individuals is simply taken for granted. I myself would not wish to endorse Quine’s and Lewis’s way of thinking about de re modality. I share with Kripke (1980) the conviction that the notion of de re modality needs no underpinning in terms of how the objects in question are to be given. But still, it is interesting that an attempt to deal with the problem of de re skepticism in the very terms in which the problem had been posed by Quine would have immediately led to the recognition of an independent notion of essence.

29.5

Related Topics

Torza, “Modal Conceptions of Essence” [Chapter 7]

434

Quine on Essence

Correia, “Non-modal Conceptions of Essence” [Chapter 8] Marabello, “Essences of Individuals” [Chapter 9] Lam, “Dispositional Essentialism” [Chapter 13]

Notes 1 The older literature includes Kaplan (1968, 1986), Fine (1989, 1990), and Burgess (1997). The more recent literature includes Shieh (2013), Divers (2017) and Kripke (2017). 2 For a recent discussion on whether Quine did indeed have one or two arguments, see Hale (2020) and Fine (2020). 3 But even this is not quite right, for if we disabbreviate “it is necessary that it is necessary that P” starting with the innermost occurrence of the modal operator, we get “P is analytic”, while if we disabbreviate starting with the outermost occurrence of the operator, we get “it is necessary that ‘P’ is analytic”, where ‘it is necessary that P’ is merely an abbreviation for, and not the very same sentence as, “P’ is analytic”. Presumably, the intention is that we should proceed from the innermost occurrences of the modal operator. 4 Bacon and Fine (2024) contains a discussion of the propositional modal logics that arise from different conceptions of logical necessity. 5 For suppose ◊(∃x(x = x) ∧ φ) and ◊(∃x(x = x) ∧ ¬φ) are among the de dicto truths for some dicto sentence φ. Then using the model theoretic criterion for being de dicto in Fine (1978), we may show that there is a model for the de dicto truths in which ∃x(φ ∧ ◊(∃y(x = y) ∧ ¬φ)) is true and also one in which it is false (I here assume, of course, that the language does not contain “rigid” constants). The de re modal truths may also fail to distinguish between different individuals, as shown in Parsons (1969). 6 This result may then be established by letting all the individuals be world-bound but leaving the isomorphism-types of the worlds the same. 7 Curiously, the original formulation of counterpart theory in Lewis (1968) is subject to a similar defect (as pointed out by Hazen 1979), since the counterparts of several objects are taken one by one rather than jointly; it is conceivable that Lewis took over this way of looking at the matter from Quine. 8 Strictly speaking, when we specify several objects by means of a polyadic condition, we should say which objects go with which variables. 9 A related line of thought is pursued in §2 of Fine (1989) and pp. 106–12 of Fine (1990). 10 Though under certain interpretations of the essentiality operator, the contexts would not permit the substitution of logical equivalents salve veritate and so it is not clear to what his extent his more general argument for the opacity of the context would still have application in this case. 11 Marko Malink has pointed out to me that such a procedure might give us a way of getting a constitutive notion of essence from the specifications and even getting the de dicto necessities, although the latter would require us to attach appropriate specification of conditions to the predicates of the language as well as to the singular terms. We might in this way see de dicto necessity as a form of de re necessity and Quine’s criticism of de dicto necessity as an instance of his criticism of de re necessity, one in which no specification of the ‘property’ signified by a predicate is privileged over any other. 12 Cf. Lewis (1983, 43). See also Divers (2007) and Beebee and MacBride (2015) for further discussion of how Lewis’ approach to modality is Quinean in spirit. See also Marabello [this volume]. 13 Probably the simplest way to factor our the influence of the other objects is to take their counterparts to be as broad as possible.

References Bacon A., Fine K. (2024) ‘The Logic of Logical Necessity’, forthcoming in a volume for Saul K ripke, ed. by Y. Weiss. Beebee H., MacBride F. (2015) ‘De Re Modality, Essentialism and Lewis’ Humeanism’, in ‘A Companion to David Lewis’ (eds. B. Loewer & J. Schaffer), New York: Blackwell Wiley, pp. 220–36.

435

Kit Fine Burgess, J. (1997) ‘Quinus ab omni naevo vindicatus’, Canadian Journal of Philosophy Supplementary Volume, Volume 23: Meaning and Reference, pp. 25–65; also in Ali A. Kazmi (ed.), ‘Meaning and Reference’, University of Calgary Press. pp. 25–66 (1998), and in ‘Mathematics, Models and Modality’, Cambridge: Cambridge University Press (2008): 203–35. Divers, J. (2007) ‘Quinean Scepticism about De Re Modality after Lewis’, European Journal of Philosophy, vol. 15.1, 40–62. Divers, J. (2017) ‘De Re Modality in the Late 20th Century: The Prescient Quine’, in ‘The Actual and the Possible: Modality and Metaphysics in Modern Philosophy’ (ed. M. Sinclair), Oxford, England: Oxford University Press, pp. 217–36. Dumitru, M. (2020) ‘Metaphysics, Meaning and Modality: Themes from Kit Fine’, Oxford: OUP. Fine, K. (1978) ‘Model Theory for Modal Logic Part I—The De Re/De Dicto Distinction’, The Journal of Philosophical Logic, vol. 7, 125–56. Fine, K. (1994). ‘Essence and Modality’, Philosophical Perspectives, vol. 8, 1–16. Fine, K. (1995) ‘The Logic of Essence’, The Journal of Philosophical Logic, vol. 24, 241–273. Fine, K. (2005) ‘Modality and Tense: Philosophical Papers’, Oxford: Oxford University Press. Fine, K. (1989) ‘The Problem of De Re Modality’, in ‘Themes from Kaplan’ (eds. J. Almog et al.), Oxford: Clarendon, pp. 197–272; reprinted as chapter 2 of Fine [2005] (references are to the chapter). Fine, K. (1990) ‘Quine on Quantifying In’, in ‘Proceedings of the Conference on Propositional Attitudes’ (eds. C. A. Anderson & J. Owens), Stanford CSLI, pp. 1–26; reprinted as chapter 3 of Fine [2005], 115–30 (references are to the chapter). Fine, K. (2020) ‘Comments on Bob Hale’s “The Problem of de re Modality”’, pp. 456–60 of Dumitru [2020]. Hale, B. (2020) ‘The Problem of de re Modality’, pp. 234–46, in Dumitru [2020] Hazen, A. (1979) ‘Counterpart-Theoretic Semantics for Modal Logic’, Journal of Philosophy, vol. 76, 319–38. Kaplan, D. (1968) ‘Quantifying in’, Synthese, vol. 19, no. 1/2, 178–214. Kaplan, D. (1986) ‘Opacity’, in ‘The Philosophy of W. V. Quine’ (eds. L. E. Hahn & P. A. Schilpp), Open Court, pp. 229–89. Kripke, S. (1963) ‘Semantical Considerations on Modal Logic’, Acta Philosophica Fennica, vol. 16, 83–94. Kripke, S. (1980) ‘Naming and Necessity’, Oxford: Blackwell. Kripke, S. (2017) ‘Quantified Modality and Essentialism’, Nous, vol. 51, no. 2, 221–34. Lewis, D. (1968) ‘Counterpart Theory and Quantified Modal Logic’, Journal of Philosophy, vol. 65 (1968), 113–26. Lewis, D. (1983) ‘Postscript to ‘Counterpart Theory and Quantified Modal Logic’, in ‘Philosophical Papers, vol. 1’, New York: Oxford University Press, pp. 39–46. Parsons, T. (1969) ‘Essentialism and Quantified Modal Logic’, Philosophical Review, vol. 78, 35–52. Quine, W. V. O. (1961) ‘Reference and Modality’, in ‘From a Logical Point of View’, New York: Harper and Row, 2nd edition, pp. 139–59. Quine, W. V. O. (1953) ‘Three Grades of Modal Involvement’, Proceedings of the XIth International Congress of Philosophy, Volume 14, Amsterdam: North-Holland; reprinted in Quine [1976] 158–76. Quine, W. V. O. (1951) ‘Two Dogmas of Empiricism’, Philosophical Review, vol. 60, 20–43. Quine, W. V. O. (1976) ‘The Ways of Paradox and Other Essays’, revised edition, Cambridge, MA: Harvard University Press. Shieh, S. (2013) ‘Modality’, in ‘The Oxford Handbook of the History of Analytic Philosophy’ (ed. M. Beaney), Oxford: Oxford University Press, pp. 1043–81.

436

30 CONVENTIONALISM Jonathan Livingstone-Banks and Alan Sidelle

30.1

Introduction

Conventionalism about essence is the view that truths about what is (and isn’t) essential to things are based upon talk and thought about the world, rather than mind-independent facts.1 Some restrict the title “conventionalism” to views that claim essence facts depend upon choices (perhaps implicit) of speakers or thinkers, so if, for instance, the linguistic or conceptual facts which base the essential facts are hard-wired into us by evolution, linguistic rules expressing this would not be conventional. We use the term more broadly, applying it to any view where the key contribution to essence-making comes from us. In some sense, even in the above evolutionary case, this involves what can be considered “metaphysically (though perhaps not causally) arbitrary selection”, and this motivates our use of the term. Other “challenges to essentialism” presented in later chapters: Conferralism and Social Construction, may emphasize differences in the mechanism of essence-generation with linguistic conventions or rules—and we do not deny these may be important differences. But to us, these are intra-familial disputes. Others may disapprove of this lexical choice. Nonetheless, we focus on the version of mind-dependence that appeals to linguistic or conceptual rules. Conventionalism also contrasts with anti-essentialism (see chapter: Quine’s AntiEssentialism), though it shares many of the negative motivations for that position. The anti-essentialist, however, denies there are any (positive) facts about essence, while the conventionalist allows them, simply maintaining that they are not mind-independent. This is in part because the conventionalist finds at least some essentialist arguments or claims compelling. That they find it difficult to understand these (purported) facts of essence as mindindependent does not require denying them outright. The conventionalist finds it more plausible to accept such facts, but deny their mind-independence, just as someone may believe there are no mind-independent facts making mountains beautiful, but still accept their beauty.2

30.2 Traditional Conventionalism Conventionalism about essence, and necessity, takes different forms. The most familiar “traditional” form holds that necessary truths obtain in virtue of the meanings or

DOI: 10.4324/9781003008750-35

437

Jonathan Livingstone-Banks and Alan Sidelle

definitions of the terms involved. Suppose that “vixen” means female fox. Applications of “vixen” are constrained—according to the conventional rules of our language—by whether something is female and a fox. Given that “female” and “fox” mean female and fox,3 the sentence “All vixen are female foxes” is true—true “by definition”, and as the definitions are given by conventions, it is true “by convention”.4,5 We speakers of the language use these words, and so may truly say that all vixen are female foxes—that this is a fact. Next, if asked “what is it for something to be a vixen?”, these meanings require the answer: being a female fox. Meanings constrain the application of a term unrestrictedly—whether talking about how things actually are or how they might be. Thus, being a female fox is essential to being a vixen. If this really is what “vixen” (conventionally) means, then using the word, we may truly assert that this is the essence of vixenhood. This involves nothing from the world, aside from our conventions, beyond the triviality of female foxes having to be female foxes. This, according to traditional conventionalism, is how all essential truths should be understood.6 Suppose one challenges the idea that word meanings can generate essential truths: “But this is just the meaning of the word ‘vixen’, and tells us nothing about the essence of the property or kind that is picked out by ‘vixen’”. Ask how these come apart. If “vixen” means “female fox”, then when we inquire into vixenhood, the conditions for belonging to that kind are: being female and a fox. Asserting that something might be a vixen while failing these conditions, or that something else may make something a vixen, is tantamount to denying that “vixen” really does mean “female fox”. This is not to defend conventionalism, but to explain what it (traditionally) says, and defend the idea that word meanings can (in principle) generate essentialist truths.7 This form of conventionalism can be supported by considering how we address “What is the essence of vixenhood?” or “What is it that makes something a vixen?”. Reflection on a general level, and about the full range of possible cases, suggests that it requires something be female and a fox. But we may also think this merely tells us how we use “vixen”—the conditions by which we constrain our application of the term. We are not talking about the term—we are using it. But given these rules for use, we properly say that nothing could possibly be a vixen without being a female fox, and this is what makes it a vixen. As this fits our understanding of essence, we can assert that being a female fox is the essence of vixenhood. Similar reasoning motivated Locke’s “nominal essentialism”. For Locke (1689), the “nominal essence” of F is the meaning of “F”—and this is what sets the boundary for being an F. Asking “What is the essence of F?”, is the same as asking “What is the meaning of ‘F’?”, except we use “F” rather than mentioning it. It seems more “world-directed”— and in a way, it is—but what is generating the truths are our meanings, rather than “real essences” of the referents of the terms.8 For more discussion of Locke, whom we see as a major influence on 20th century (and later) conventionalists, see Schechtman’s chapter in this volume. Conventionalism is thus often presented as the view that all necessary truths are analytic, or true by convention. That is, where “necessarily p” is true, “p” is analytic. Some even represent it as holding that “necessarily p” just means “‘p’ is analytic” (Quine 1951). However, this does not seem right. It is not simply that we have another word—”necessarily”—that means the same as “in virtue of meaning” or “analytically”. Rather, meanings constrain the application of terms in any possible situation. Necessity is a product of meaning. In these cases, “p” is true in virtue of the meaning of the components of “p”—“necessarily p” is true in virtue of that and the meaning of “necessarily”. Given that meaning comes from us, we describe this as “truth-by-convention”.9,10

438

Conventionalism

Central motivations for this view are (a) the a priori character of our justification for belief in both p and necessarily p,11 and (b) an air of metaphysical extravagance about any more robust sense of necessity, and puzzlement about what such real necessity or essence could really be. For the first, our knowledge of necessary truths does not involve empirical investigation: it seems available just from our understanding of the concepts involved, and our ability to reason from them. We often argue for the necessity of p by trying (and failing) to conceive of a situation in which p would be false. Alternatively, we prove p from something necessary, which usually traces back to such thought experiments, or “direct intuitings”. If we don’t need to look at the world to know that p (or necessarily p) is true, it seems plausible that it is nothing about the empirical world that makes it true: it is instead analytic.12 In many cases—like bachelors’ being unmarried—it seems independently plausible that this is so in virtue of the (conventional) meanings of terms.13 On the metaphysical side, one may ask: what is the difference between the fact that all bachelors are unmarried, and that all bachelors are Earthlings, such that the former is necessary, but not the latter? Does one have some extra, metaphysical “oomph” such that it can’t fail to obtain, which the other lacks? It is hard to gather what this might be. (While modern essentialists insist that essences are “actual”, “being essentially human” goes beyond simply “being human” in the same worrisome ways, both epistemologically and metaphysically.) That we know a priori that the former, but not the latter, is necessary (even assuming both are true), suggests we are simply marking a difference that results from our concepts, and not some mind-independent feature distinguishing the two. Essence seems mysterious when considered metaphysically, but it falls right into place when one supposes that essential or necessary truths result from meanings. If some essential claims seem less obvious, it is only because it is less obvious what the meanings are. (This fits the epistemology as well, including the defeasibility of initial intuitions, and the need to probe unexplored scenarios with thought experiments.)

30.3

Traditional Conventionalism and Necessity a Posteriori

In the early 1970s, Kripke (1980) and Putnam (1975 and elsewhere) argued for necessary truths (and truths about essence) knowable only a posteriori. Being empirical, they seem to reflect facts about the world, and not our linguistic conventions, the contents of which are knowable a priori. This puts necessity and essence squarely in the world, contrary to conventionalism. Kripke’s initial examples are empirical identities where both terms are names: Hesperus=Phosphorus. More obviously essentialist examples include the essentiality of material and biological origin (table T is originally made of wood, Chelsea is a biological daughter of Bill and Hillary); gold is an element; water is H2O, and statements of kind membership (Obama is human).14 If there are necessary truths that are not a priori—and thus not analytic—then conventionalism appears at least partially mistaken. That water is H2O is a scientific discovery; there is no contradiction in the possibility of scientists proclaiming that water is in fact an element, or has different components. But thought experiments—most famously, Putnam’s (1975) Twin Earth case, taken counterfactually—support the claim that it is nonetheless necessary. If it is necessary, but not a priori and so not analytic, it is not true by convention, and conventionalism is false. However, a closer look is needed. Does the fact that “water is H2O” is not analytic—not itself true by convention—mean that “water is essentially H2O” must represent a wholly mind-independent fact? We (among others) suggest that a conventionalist approach is not only still possible, but preferable.15

439

Jonathan Livingstone-Banks and Alan Sidelle

The conventionalist’s epistemological and metaphysical concerns are still in play. Nothing in the presentation or defense of these new necessity and essence claims takes the mystery out of the “what is special about these properties?” question. Adding “that is what water is” merely reasserts the essence claim. Why is “being watery16” not “what water is”? The supporting arguments are again conceivings and thought experiments. They involve empirical elements—that Obama is human, or that water (or: the stuff in lakes, etc.) is composed of H2O—but the arguments all involve reflection on conditional or hypothetical claims, which do not commit to these empirical claims: “If Obama is human, he must be human” or “Assuming water is composed of H2O, it cannot fail to be so/XYZ would not be water”.17 It seems clear that when there is empirical input into a defense of a necessity or essence claim, there is a “bridge” principle linking a non-modal empirical claim with the conclusion, and this principle is epistemically independent of that empirical claim (though perhaps not others—see note 17). Our knowledge of the necessity as necessary, or as essential, still depends on these intuitings, conceivings or thought experiments. We have as much reason to think what we are discovering here are (the products of) our conventions or meanings as we did pre-Putnam and Kripke. Of course, the conclusions are not themselves true by convention, nor simply in virtue of meaning (and logic)—they come from a combination of a conditional principle and an empirical antecedent. But the “extra” force—the necessity or essentiality—is provided by the principle, the a priori element. And if the principle—say, “If x is human, then x is essentially (necessarily) human”—is analytic, then we again have a “naturalistic” answer to what the “extra oomph” is that being human has, but being born in Hawaii lacks: it is simply a conceptually imposed constraint on how we identify something in hypothetical situations.18 Put another way, the epistemic and metaphysical reasons to accept conventionalism before the advent of a posteriori necessities and truths of essence are still available afterward. The data simply point to conventions of a hypothetical form: “if (these samples of) water is composed of H2O, then something is water in any situation (if and) only if it has that composition” or “If Obama is human, nothing counts as Obama in any situation unless it is human”. This allows the conclusions to be non-analytic and empirical, but requires no more commitment to mind-independent essences (subject to the caveat in note 18), and provides explanation of (a) the evidential role of non-empirical methods and (b) the difference between properties that are and are not essential. This shows that conventionalism is not committed to the claim that “necessarily p” is true just in case “p” is analytic (much less that this is what “necessarily p” means). This was never the central commitment of conventionalism, but only a natural form for it to take when all the (plausible) examples of necessary and essential truths under consideration were (plausibly) analytic (or more exactly: as plausibly analytic as they are plausibly necessary). In those cases, note, both “p” and “necessarily p” can be supported by the same sorts of thought experiments. But when we have a posteriori necessities, “p” and “necessarily p” come apart epistemically. Next, we give a more detailed account. This is not the only way to develop a post-KripkePutnam conventionalism, but we think it is a promising one.19

30.4

Conventionalism Post-a Posteriori Necessity

Because conventionalism was developed before philosophers focused on the (purported) difference between essence and (merely) necessary properties, it will be easier to present the account as developed, in modal terms. We will then see a natural enough way to account for non-modal conceptions of essence. (On the debate concerning modal and non-modal

440

Conventionalism

conceptions of essence, see chapters by Torza and Correia respectively.) The modal essentialist characterizes essentialist claims as being some form of modal claim. To say that Curie is essentially a woman is to say (something like) if she exists, Curie must be a woman.20 As such, in order to provide a theory of essence, all the conventionalist need do is provide an adequate theory of necessity (as sketched above). Since, in all these cases, the empirical input requires that whatever a certain sort of empirically discovered property is, it will be essential (but see note 21), or that this sort of finding will be essential as well to other things of the same kind, we can suggest that (generally), when there is an a posteriori necessary truth, there is a general principle of individuation (GPI) for things of the relevant sorts. These can be presented schematically as: GPI Schema: (x) (If x belongs to kind K, then (if p is x’s P-property, then it is necessary that x is p))21 Where x is a variable for individuals, kinds, or properties; P-property is a kind of property that is taken to be definitive of kind K, and p is a specific instance of that property. For example, if x = water, K = chemical kind, P-property = elemental composition, and p = H2O, then the resulting GPI would be: WATER: If water is a chemical kind, then if H2O is water’s elemental composition, then water necessarily has the elemental composition H2O. The reasoning that secures the necessity of water having the elemental composition H2O continues: P1: Water is a chemical kind. (Premise) C1: If H2O is water’s elemental composition, then water necessarily has the elemental composition H2O. (By modus ponens from WATER and P1) P2: Water has the elemental composition H2O. (Empirical premise) C2: Water necessarily has the elemental composition H2O. (By modus ponens from C1 and P2) This presents a plausible account of the reasoning behind any necessity a posteriori, regardless of whether one is a realist or antirealist about essence. What sets the conventionalist apart is the claim that GPIs are not facts about the way the world is (as the realist holds), but rather are object language formulations of our conventional practices. The practice described by a GPI is the means by which we carve up the world—that is, by which we identify things: apply “is the same thing” or “an instance of the same kind” or “the same substance” to both actual and hypothetical items. According to realists, these judgments may be true or false, depending upon the identities or natures of the objects in question. According to conventionalists, these judgments are determinative of, or reflect, our decisions to count things as the same or of the same sort, and so, to apply these terms in new cases. The conventions themselves govern how we use our terms for the subjects of the GPIs. That we tend to do so according to regular patterns allows us to speak of kind terms (terms we introduce with the intention to use in accordance with the GPIs for a specific kind). A K-term is a term governed by criteria for the kind K. For example, I introduce the term “rose” near some flowers. I intend the term to apply to other things that are like these biologically or botanically, rather than, say, decoratively, so I introduce it as a biological kind term (or more specifically, as a flora term). In doing so, I resolve that hypothetical and counterfactual applications of “rose” are governed by certain (conventionally selected) aspects of that original application, such as petal formation, or lineage, rather than being

441

Jonathan Livingstone-Banks and Alan Sidelle

pretty or aromatic22. The aspects we select as important are the P-properties. A kind term L is a K-term if the P-property for the kind K (for instance, elemental composition for a chemical kind) is found in most of the things we apply L to (or “the relevant things”),23 and governs our counterfactual use of L. In this way, P-properties are definitive of kinds because we resolve not to count something as of the same kind if the intended referent does not have the P-property for that kind term. For example, “water” is a chemical kind-term because whether we correctly apply the term “water” in any given instance is determined in relation to the P-property selected by the GPI for chemical kinds (in this case elemental composition). The elemental composition of “the relevant (note 23) items to which we apply the term” is H2O, so we resolve only to apply the term “water” to things with the elemental composition H2O (Sidelle 1989, p. 44). Likewise, “chair” might be a furnitureterm, governed by the GPI for furniture; “chemical kind” might be a natural kind-term, and “furniture” might be an unnatural/artifactual kind-term. Focusing on the linguistic or conceptual rules rather than their object level products, we can present the conventional schema for GPIs as: GPI Conventional Schema: If ‘x’ is a K-term then if p is the P-property of the (relevant) things to which we apply ‘x’, then ‘x’ applies to something in any possible situation only if it is (has) p. (Sidelle 1989, p. 44) Our use of a term L is governed by a GPI.24 The kind of kind term we introduce L as determines which GPIs govern it. Introducing L as a K-term commits us to governing the use of L in terms of the GPI for kind K. This effectively makes L an instance of K. In this way, we use our conventions to create kinds. By introducing “water” as a chemical kind term, “water is a chemical kind” is true by convention, and so, speaking the language: water is a chemical kind and necessarily, if water is H2O, then nothing that is not H2O is water.25 Our presentation has focused on modally-conceived essence, but this account is perfectly suited to the more fine-grained essentiality championed by non-modalists, who conceptualize essence in terms of the “definitions” of objects. Under conventionalism, objects do have definitions, as provided by GPIs, and the associated criteria of individuation for referring terms. For singular terms introduced as rigid in the normal way, the kind which applies and is intended (human, say) will be, in an extended sense, part of the “definition” of Bill (or what have you), and thus not only necessary, but essential. Being distinct from the Eiffel Tower, or such that 2 + 2 = 4, or a member of the singleton {Bill}, will not. Not only does this yield exactly the results Fine wants, but does so in a way that avoids the epistemological difficulties and metaphysical extravagance of realism. For more detail, see Sidelle (1989), ch. 4 and (Livingstone-)Banks (2014).

30.5

Other Views

Various philosophers, while not calling themselves “Conventionalists”, are skeptical of a realist view of necessity or essence, while allowing essentialist truths (note 15). We include them in our broader sense of “conventionalism”, though their views differ in interesting and important ways from what we present here. While we lack space to discuss them in any detail: Chalmers 1996, 2002, 2014 offers a structurally similar account, using a two-dimensional semantics and emphasizing that there are no “strong necessities”; Thomasson 2013, 2020 presents modality as an object-level expression of rules; Sider 2011 ch. 7, 2013, section 30.9 suggests the relevant practices are our somewhat arbitrary uses of the modal terms themselves

442

Conventionalism

(Cameron 2008, 2009, 2010b offers a similar account); Goswick 2018 and Asta 2008, 2013 see modality and essence as based in our responses to objects, and Clarke-Doane 2019a, 2019b questions the objectivity of metaphysical modal claims. Asta is discussed in Wallner and Vaidya’s chapter: Conferralism.

30.6

Potential Challenges to and Questions for a Conventionalist Theory of Essence 30.6.1

The Contingency Problem

A familiar and long-standing objection to conventionalism stems from the observation that our conventions—by their very nature!—could have been other than they are; indeed, any convention we have could have failed to obtain. This seems to undermine the ability of conventions to explain essence or necessities: how can that which cannot fail to obtain be accounted for by something that can? One way to articulate the worry starts: (A) If our conventions had been suitably different, then a different range of necessities would have been true. The concern is that this entails that what is necessary might not have been necessary, and consequently, the S4 axiom: What is necessary is necessarily necessary, is false. Some see here an objection to conventionalism, as the S4 axiom (and indeed, the stronger S5 axiom: what is possible is necessarily possible) seems very plausible with respect to metaphysical necessity. Others see this as a bullet the conventionalist ought to bite, and indeed as question-begging to take the S4 claim for granted (Hale 2002). Unfortunately for the conventionalist, rejecting S4 offers a poor reply. It is premised on the assumption that if we didn’t have the convention that “bachelor” applies only to men, then it would not have been necessary for bachelors to be male: there could then possibly have been female bachelors. But then, not only would it be possibly possible that there be female bachelors: as there would have been unmarried women, there would have been female bachelors. If different conventions would have produced different essences and necessities, then our actual conventions are compatible with the falsity of what is (by hypothesis) necessary or essential—there could have been female bachelors, and so, being male is not essential to being a bachelor. Thus, the conventionalist cannot avoid the objection by simply rejecting the S4 axiom, and the worry about S4 is not the best way to formulate the challenge from the contingency of conventions. (See van Cleve 2006; Sidelle 2009). However, the presupposition in this formulation of the objection seems mistaken. It is wrong to think that if we had used “bachelor” so as to include unmarried women, then being male would not be essential to being a bachelor. If we assume it is a convention that something is a bachelor only if male, then when describing any situation—even situations with different conventions—nothing is a bachelor unless male. Thus, “if we had different conventions, there would have been female bachelors” and “if we had different conventions, it would have been possible for there to be female bachelors” both turn out false (see Wright 1986; Sidelle 2009). However, this reply points towards another way to articulate the problem. If differences in our conventions would not generate differences in what is necessary or essential, it is

443

Jonathan Livingstone-Banks and Alan Sidelle

unclear how convention plays a role in determining necessity. But the claim that it does is central to conventionalism! We can express this commitment as: • (LINK) There must be a clear link between our conventions and the true necessitation claims. To satisfy LINK, Einheuser 2006 distinguishes between our conventional practices (the behavioral regularities—such as calling all female foxes “vixens”—that constitute our conventions) and the carvings (the theoretical counterpart to our conventional practices) that correspond to them. As such, talk of “conventions” is ambiguous. It can mean either conventional practices or the corresponding carvings.26 This allows for two dimensions along which modal change can occur: change in the pre-conventional facts (i.e. conventional practices), and change in carvings. We can interpret A as inviting us to jump to a world different from ours. If A jumps to a world with different conventional practices, then, as above, A is false. If A jumps to a world with a different carving, then A is true. This is enough to satisfy the intuitive pull of LINK, but the truth of A relies on worlds that are not possible relative to standard modality, so represents no counterexample to S4. Simply put, A is true on one reading and false on another, and no threat to S4, or conventionalism, under either. For other versions of the contingency problem, see van Cleve 2006, Blackburn 1986 and Lange 2008. For more replies, see Tennant 1987, Sidelle 2009, Livingstone-Banks 2017 and Wildman 2021.

30.6.2

Truth-by-Convention and Analyticity

Analyticity and truth-by-convention are often conflated. However, there are multiple roles convention might play in the truth of a sentence, beyond that of analyticity. P might be true iff some convention C holds (where C does not feature in the meaning of P, for instance if P asserts we hold convention C). P might be true iff some fact F holds, where C determines that the truth of P is contingent upon F. P might be true iff F holds, where what counts as an instance of F is determined by C. Each might be in some sense considered truth-byconvention. Quine (1951) identifies two classes of analytic statement: those that are logically true (e.g. “no unmarried man is married”, which is true for any candidate meanings of “man” and “married”), and those that can be transformed into logical truths (e.g. “no bachelor is married”, which becomes a logical truth when “bachelor” is substituted for its synonym “unmarried man”). Quine accuses the second class of showing up the circularity of analyticity, arguing that the relevant notion of synonymy cannot be understood without appeal to analyticity itself. Appeals to definition, interchangeability, and semantical rules are each considered and rejected as potential explanations of synonymy without appeal to analyticity, leaving Quine to reject analyticity on the grounds of circularity. As for the first class, Quine’s attack takes two forms.27 First, that convention cannot possibly account for the truth of all necessities. Quine (1936, p. 4–5) asserts that word-sized conventions (definitions) cannot found truths, only of transform them: “what is loosely called a logical consequence of definition is therefore more exactly describable as logical truth definitionally abbreviated”. As such, conventionalism must rely on sentence-sized conventions, rules stipulating the truth of (infinitely large) classes of necessities. However, any

444

Conventionalism

attempt to account for all necessities falls victim to Carroll-style regress (Carroll 1895), where for any necessity to follow from such a class requires a further necessity that justifies that inference. The second criticism argues that conventions just aren’t the kind of things that can make sentences true in any meaningful way. Just as definitions can only transform truth (by determining what a word means) and not found it, no convention can found truth, but can only determine the meaning of a sentence (what proposition it expresses); whether that proposition is true is always a further matter beyond the conventions governing the sentence. Finally, Quine takes aim at the idea that analytic statements are those that are “vacuously confirmed”—true “come what may” in experience, and therefore unrevisable without a change in meaning. Quine rejects this, espousing a moderate holism, whereby experience does not confirm/disconfirm statements individually, but only within one holistic totality. Recalcitrant experience can thus result in the revision of any statement. We confuse unrevisability in principle with the greater convenience in revising other statements. We lack space for a full discussion of analyticity and truth-by-convention, but there are two paths of reply the conventionalist might take. First, they can defend analyticity/truth-byconvention, thereby rendering reliance on it harmless. There are many such defences against both the “definitional circle” argument and the “everything is revisable” argument. Generally speaking, the former seems overly restrictive and based in a behavioristic semantics, while the latter seems an unsufficiently supported overgeneralization (e.g. Grice and Strawson 1956; Lewis 1969; Searle 1969; Sidelle 1989, ch. 5; Warren 2020). Further, as per note 13, our position is compatible with a non-conventionalist view of logic. Second, they can demonstrate that conventionalism need not rely on (a troublesome form of) truth-by-convention. Several contemporary conventionalists count themselves free from truth-by-convention (e.g. Cameron 2009; Sider 2011), and (Livingstone-)Banks 2014, ch.2 argues that the conventionalist theory presented above can be formulated without reliance on a problematic kind of truth-by-convention. For a related contemporary concern, see Russell 2010 and for a response, (Livingstone-)Banks 2014.

30.6.3

Does Conventionalism about Essence Require Conventionalism (or Constructivism) about the Things That Have Essences?

On one understanding of conventionalism, the world mind-independently contains objects and kinds, and our conventions merely select some of their properties as essential. After all, the worries that motivate conventionalism arise in connection with the “extra” the essences and modal features have—they don’t obviously concern the objects themselves. Nonetheless, some propose that conventionalism about essences requires this further commitment. The central thrust can be found in Locke: given that essence is what something is, that which is responsible for the essence is also responsible for the object (Locke 1689 Book III, ch. iii, sec, 13, 14; ch. vi, sec. 35). One way to develop this is given below. For more, see, e.g. Sidelle 1989 ch. 3, 2010; Rea 2002, ch. 7; Einheuser 2011. (This need not be an objection to Conventionalism, though some treat it so (as Rea and Oderberg (below)). If conventionalism does have this commitment, it may be an argument for a more thoroughgoing constructivism (as Locke, Sidelle and Einheuser argue)). Assuming conventionalism about essence, suppose Bert is essentially human. Could Bert be a mind-independent object to which an essence is “added”? In considering his essence conventional, we allow that our conventions may instead have focused on mental properties. Supposing they did, would Bert have been essentially a person, and only accidentally human (or at least, capable of occupying another human body, via a cerebrum transplant)?

445

Jonathan Livingstone-Banks and Alan Sidelle

Surely not—Bert is essentially human, and so has human persistence conditions. In a world where we had different conventions, and Bert’s cerebrum was put into another body, Bert would not move into another body: Bert’s persistence conditions entail this. So this being (“Ernie”?) who would be capable of changing bodies must not be Bert.28 Bert cannot be some mind-independent object to which we apply our conventions, generating an essence: that which has the essence is itself, then, mind-dependent. In short, we cannot treat essences as something we can “detach” from the objects that have them, in the way we can “detach” something’s being a work of art, or a doorstop, by the way we use it. Essences individuate objects and determine their trajectory through spacetime and modal space, and so cannot be subject to such variation (Wiggins 1979). One may respond using counterpart theory (Lewis 1971; Sider 2001, ch. 6.4) (or an “Aberlardian” view of modal predication, Noonan 1991). On such views, “a is essentially F” is interpreted relative to a certain sort, and “Ernie” might look to a person-counterpart relation while “Bert” to human counterparts, even if Ernie and Bert are identical. Thus, we cannot infer from “Ernie is essentially a person” and “Bert is not essentially a person” that Ernie is not Bert, as we did above. While we lack space to pursue this in detail, consider: if Ernie/Bert does have his cerebrum transferred, has that thing changed bodies? If there is just one thing, there should be one answer—but the essences seem to require different answers for Ernie and Bert. In reply, a thoroughgoing counterpart theorist may apply counterpart theory to temporal and modal predications (Sider 2001), denying that aside from some way of thinking about this mind-independent thing, it makes sense to ascribe temporal properties to it—here, that it does or does not survive some change. So there remains scope to discuss whether conventionalism about essence precludes realism about the things that have the essences.

30.6.4

Does Conventionalism Presuppose Objects with Mind-Independent Essences?

The analyticity of general principles of individuation yields the analyticity of “If Obama is human, he is essentially human”. But this seems to require that “Obama” can apply to something only if it is already, mind-independently, essentially human. Unless there are objects with mind-independent essences, the terms governed by such conventions will simply have no referents. (For something like this concern, see Yablo 1990 and Cameron 2020.) While we lack space for a full conventionalist account (see Sidelle 1989, ch. 3, 1992a, 1995), we can sketch the basics of a reply. How this should go depends on how we resolve whether conventionalism can be combined with a mind-independent view of objects. Let us start with the supposition that it can, and then we can apply the basic strategy to the more throughgoing conventionalist position. If Obama’s being essentially human can be conventional, while Obama himself is not, the story seems straightforward. Mind-independently, there is Obama, with his various “purely actual” properties. Our conventions select from among these which will be essential or individuating—which will determine the application of “Obama” (and “this guy”). This probably occurs via a general default when introducing names in given contexts. Thus, e.g. “Obama” may be an “organism” name, as we have a default tendency to count something as Obama—as “the same thing as this”—only if it is of the same species as this thing.29 Or, we may instead use it as a person name, whereby application requires psychological continuity with that to which the name has been introduced, which will not require physiological continuity, or even that the thing be an organism. (How we use it is rarely if ever a matter of explicit stipulation—it is a function of how we use the term, in particular in modal

446

Conventionalism

(and temporal) contexts.) In one case we have “Obama is essentially human” and in the other “Obama is essentially a person”. Presumably, we don’t assert both of these pairs in any case—but we might have different terms which (on the current assumption) refer to the same actual object, and ascribe it different (conventional) essences. Once we have introduced these rules, the essential predications are true, and it is true that Obama is essentially human (and also that he was essentially human before we named him, and would have been so even without our conventions—see discussion of contingency above). The only input was Obama; his being essentially human was not presupposed. We do need to ask: is the Obama to which we applied these conventions the referent of “Obama” that bears the essential properties? If Bill uses “Obama” as a person-name, and Sarah uses it as an organism-name (i.e. if their counterfactual judgments are so guided), can their terms have the same referent? We might think not: they have conflicting essential properties. However, in the above Abelardian/counterpart theoretic approach, we avoid contradiction. The mind-independent object-plus-counterpart-theory approach is thus one way conventionalism can generate essential truths without presupposing mind-independently essence-laden objects. An alternative that does not require the Abelardian/counterpart view might propose that while there is a mind-independent object, it is not that which, after the application of our conventions, is essentially human. Rather, this mind-independent object always lacks any essential features (Goswick 2018). Our conventions generate a new non-mind-independent object, which is essentially a human, and for the person-name, there is a different new object (coincident with the first30), which is essentially a person. The objects are distinct—both from each other and the “pre-modal” object from which they were each constructed, since they differ modally/essentially. Now, while “If Obama is human, he is essentially human”, is true, it is not a perspicuous representation of the convention. The input into the convention is neither Obama nor his species—it is the species (or whatever) of the pre-modal thing. Once again, we generate Obama’s essential humanity without presupposing an object (much less Obama) that has a mind-independent essence.31 Some may object that nothing is human unless essentially human. So the “pre-modal” urObama cannot be human. One reply is to deny that whatever is human is essentially so: Ernie—the person—is human, but not essentially so. An alternative reply introduces a new predicate—human*—which can be met by something accidentally. It requires (say) an actual sort of organic composition. Presumably, we understand a thing’s being F, before we inquire whether it is essentially so. The rule then can be that if ur-Obama is human*, “Obama” applies only to what is human. We can repeat this for any property to which a conventionalist appeals, which the anti-conventionalist insists has modal import. Finally, what if one denies there are such pre-modal objects? What if any object requires persistence conditions and hence some essential features, some constraints on what is possible for that object? (Wiggins 1980). What if for every property there is “what it is” to have that property? What, then, in addition to our individuative conventions, can be the worldly input for the generation of objects with essences? The above replies point in a natural direction. Look at the tree outside your window. On the view in question, there being a tree there requires that something meets certain essential conditions. According to the conventionalist, what is there, independent of our conventions, is, then, not a tree. But something is there, pressing itself upon us, generating our visual experience. It stands to a tree much as being human* stands to being human: it is the non-modally involved worldly matter, which looks, tastes and feels just like a tree. Sidelle 1989 calls it “stuff”. Putnam 1992 in contrast suggests it is a mistake to try to say

447

Jonathan Livingstone-Banks and Alan Sidelle

anything about this mind-independent input, echoing Kant’s approach to noumena. But it seems the conventionalist should say enough to show that they need not presuppose a modality-and-essence infused world, in order for the conventions to not simply be empty.32 The above discussion hopefully illustrates that conventionalism has significant room to maneuver here.

30.6.5

Does Conventionalism Presuppose Convention Makers?

If conventionalism entails—or is combined with—the view that objects with essences are also (partially) products of convention, one might think conventionalism is doomed to a vicious circle. Oderberg 2007 formulates the objection: We are the source of our conventions, and if conventionalism were true in general it would have to be true of us. But then, our conventions would have to be logically prior to us; but on the contrary, we are logically prior to our conventions. Hence, conventionalism about us could not be true … If conventionalism is not true in respect of us, why should it be true in respect of anything else? (p. 44) (see also Elder 2004, p. 17) If one can be a conventionalist about essence but not the relevant objects, this is not a concern. But such a connection is plausible, so a defense is required. First, the conventionalist is independently likely to find that the general concerns supporting conventionalism apply—indeed, very palpably so—to people. There are many candidates for what is essential to people, each fundamentally a priori and based on thought experiments (even if “prompted” by empirical facts). The same reasons we think these sorts of judgments, if true, reflect conceptual and linguistic decisions, apply here, leading us to suspect that no realist account adequately explains why exactly one of these essential profiles is true of us.33 One could hold an alternate view of what is essential to people and not be making a mistake, except insofar as one is not using “person” and “same person” as one’s peers do. This by itself does not avoid the circularity. But it shows the Conventionalist is unlikely to lament: “Oh no! We forgot about ourselves!”. And it suggests a natural approach—namely, the same approach taken for other objects. As we have seen, the relevant sort of Conventionalist thinks that to understand the construction of a statue that essentially has a certain shape, we should not appeal to a mind-independent statue of which certain properties are “selected” as essential. Rather, we have principles of individuation, established conventionally, that determine that when certain non-modal conditions are met—say, when some material is arranged by a certain intention—there is an object with a certain essence, and nothing counts as that object without meeting those essential conditions. Similarly for people. The “matter” for people may consist of mental states that stand in certain causal relations, or the material substrate underlying them (brain stuff), or broader matter that interacts with these in certain ways (our bodies), and by selecting among these and a much broader range of possibilities (like brain transplants and teleporters), principles of individuation are selected, thereby “constructing” people with persistence conditions and essences.34 But who or what is doing the constructing? That very psychological stuff: those judgments about actual and counterfactual cases. The conventionalist account here is uniform throughout: to know what the world contributes to objects with essences, simply look to

448

Conventionalism

the world and try to abstract away the modal/essenced features (i.e. being essentially F—not simply being F). As it is with trees, so it is with people and with conventions themselves. Perhaps calling it a “convention” or “rule” implies attributing it to people or a community; the conventionalist can be fine with this. Overall, the very flexibility of our concepts of person, and of sameness of person, is palpable enough that the idea of conventions or decisions settling what it takes to be the same person, and what is essential to a given person (or to personhood) should seem a candidate view to take seriously whether or not one is drawn to conventionalism in general.

30.6.6

Can Conventionalism Allow “Better and Worse” Conventions?

One last objection appears more often in conversation than print. Consider water. We use the term “water” as applied to many samples. We have good reason to guide counterfactual (and future) applications by explanatory microstructure; for instance, inductive inferences from statements containing “water” are more likely to be true, making the referent of “water” a (more) explanatory kind. Past practice with other terms provides further reason to continue conforming to this microstructure-based practice. It is certainly better than using “grue-like”, or even “thwater”-like conditions. Because we have reason to favor some conventions over others, some conclude that conventionalism is false. Why would this be?35 One thought is that conventions must be arbitrary, while the above shows that it is not arbitrary how we should continue using “water”, nor what sorts of terms we should have. However, we have many conventions for which we have good reasons—choosing red for stop lights, for instance, could be preferable to yellow because it is striking and easily seen. Conventions need not be arbitrary in this sense—at most, there must be options, choices. Even if there is good reason to use terms for whatever shares a microstructure, it is certainly open to us to use “water” for whatever shares more superficial features. On brief reflection, it should be clear that admitting there are “better” semantic conventions would be problematic only if they pick out objects or kinds according to real, mind-independent essences. But one need not think this to believe we may have good reasons to have certain sorts of conventions. There is no valid inference from some G’s providing a good explanation for why things that are F behave in similar ways, to the conclusion that being G is essential to being F. And similarly for other good reasons for governing the application of a term by a certain feature.36

30.7 Conclusion Conventionalism offers intuitive accounts of the epistemology and metaphysics of essence and de re modality, where it may seem that Realism is on shakier ground. Despite this, Conventionalism is often treated dismissively, or taken to be subject to easy and quickly decisive objections. Some of these objections represent an almost wilful misrepresentation of Conventionalism. For the other, more forceful concerns, conventionalists have a range of replies plausible enough for the view to be taken quite seriously.

30.8

Related Topics

Social Construction Conferralism

449

Jonathan Livingstone-Banks and Alan Sidelle

Notes 1 This comes with the usual caveat for understanding realist/anti-realist disputes: mere “minddependence” does not capture it. Even realists about psychology, sociology, etc. believe the “real facts” depend on our mental states. The proper sort of mind-independence is a substantive topic in its own right, and isn’t crucial here. See Passinski’s chapter on Artifacts, Artworks, and Social Objects; Mallon on Race; Rosario on Sex and Gender; and Griffith on Social Constructivism. 2 Conventionalism about essence, however, may be importantly disanalogous to conventionalism about beauty in other respects, due to the role of essence in something’s being what it is. See below “Does conventionalism about essence entail conventionalism about objects?” also Sidelle 2010. 3 Strictly, all that is needed is that they refer to these things, and for the necessity, that they refer rigidly. For those concerned about conventionalism’s ability to appeal to rigid terms, see Sidelle 1989, 1992a and 1995. 4 Throughout, “conventions” means “semantic conventions” (though they may apply to mental concepts as well as words). As will hopefully be clear, not everything true by convention (more broadly) is necessary— it is not necessary that one should drive on the right in the US. It is specifically semantic conventions that generate necessities, because they provide constraints on the application of terms in all (particularly, modal) contexts. 5 Two caveats. First, in our rather capacious use of “convention”, we are neutral on whether the semantic rules are given basically in idiolects, or in more public languages (though it would affect some details). Admittedly, some might not count the rules as “conventions” in the former case. Second, the truth of this sentence may depend on the prior fact that all female foxes are female foxes. If that is not true by convention, then our statement is true by convention plus logic. A fully thoroughgoing conventionalism would thus depend on a conventionalist account of logic, which we do not here present (see note 13). However, we believe (logic aside), essences are typically taken to “come from the world”—the “nature of things”—in a more robust manner. So an account of essence in terms of conventions-plus-logic is still a recognizably conventionalist position. 6 Historically, the account is usually of necessary truths and properties. But if one thinks not all necessary properties are essential, the approach is naturally adapted as above. 7 One might worry how word meanings can generate de re, and not merely de dicto necessary truths. This involves terms having associated identity criteria—rules for applying “the same thing” or “is identical to”, and not merely application conditions. For detail, see Sidelle 1989 ch. 3, 1992a, 1995. 8 Locke 1689 book III, ch. vi introduces “nominal essence” for essences as determined by the meanings of words, so his position may be called “nominal essentialism”. For a 20th century variation, see Black 1958. 9 For an important presentation of this sort of conventionalism, see Carnap 1947. 10 For the classic critique of analyticity and truth-by-convention, see Quine 1951, 1936. There are countless replies to Quine’s arguments, e.g. Grice and Strawson 1956, and Putnam 1983. 11 Recall we are presenting traditional conventionalism here, where the “commonly agreed upon” examples fit this mold. We address a posteriori necessities below. 12 Not everyone who believes modal epistemology is fundamentally a priori agrees that what is known is analytic or based in convention. See Bealer 1999, 2002 and Bengson 2015. Others question whether modal epistemology is fundamentally based in conceivings, intuitions and imaginings ( Biggs 2011; Biggs and Wilson 2019). For an overview of a variety of epistemological views, some of which purport to allow knowledge of mind-independent essences and modal facts, see Mallozzi, Vaidya and Wallner 2021, and Mallozzi’s chapter on the Epistemology of Essence. 13 Some worry that even here, independent necessity is presupposed: It is necessary that bachelors are unmarried only if it is already necessary that whatever is unmarried is unmarried. Some also worry that we must presuppose logic (and its necessity) to get from conventions to necessary truths ( Pap 1958; Lewy 1976; Quine 1936; Stroud 1981). See note 5: if all essentialist facts reduce, via convention, to facts of logic, they are hardly “deep metaphysical facts” about (say) water and humans. However, there are plausible replies to these traditional arguments. See Wright 1986, Tennant, 1987 and Warren 2015, 2020. 14 For more discussion of these cases, see Marabello’s chapter on Essences of Individuals; RobertsonIshii’s Origin Essentialism; and Tahko’s Natural Kind Essentialism. 15 While the following detail draws from Sidelle 1989, see Mackie 1974, Coppock 1984, Chalmers 1996, 2002, 2010, Chalmers and Jackson 2001, Jackson 1998, Asta, 2013, 2018, and

450

Conventionalism

16 17

18

19 20 21

22 23

24

25

26 27 28 29

Thomasson 2007, 2013, 2020 for similar accounts, though not all consider themselves conventionalists. For an alternative account, see Sider 2011, ch. 12. That is, “having the superficial properties by which we typically recognize something as water”— being clear, potable, filling lakes, etc. Perhaps there are more empirical presuppositions involved here—e.g. we would not judge that XYZ is not water if it turned out that being H2O was not as fundamentally explanatory as we believe. But as Chalmers 2012 argues, any such presuppositions are in principle detectable a priori and may be “frontloaded” into the antecedent of the conditional. Figuring out the basic principles involved is, of course, a non-trivial philosophical and linguistic investigation—just as finding the meaning of any term introduced non-stipulatively. (This is how we can ascribe the relevant intentions to historical users of “water” (and its translations) before the advent of chemistry. See Putnam 1975.) For the conventionalist, it is important that “being essentially human” is not a criterion of application for “x”, or for “is identical to/the same substance as x” (since, independently of convention, nothing meets such a condition). Rather, being human is such a constraint—but concerning all contexts, particularly modal and counterfactual ones. That is the proposed rule, and how being F gets “elevate” to “being essentially F”. See below: “Does Conventionalism Presuppose Objects with Mind-Independent Essences?” For similar views, see note 15. Different modal essentialists add various further criteria, such as naturalness, intrinsically, or necessity conditional on existence. See Torza’s chapter: Modal Conceptions of Essence. Sidelle 1989, p. 34. The principles need not take exactly these forms. For example, on plausible versions of the essentiality of origin (see Robertson’s chapter), if O is originally composed of matter m, what is essential to O is not this, but that it is not composed of matter wholly distinct from m. What is essential need not be what is actual—it only depends upon what is actual. The traditional representation of the “derivation” in Kripke: 1. P. 2. If P, then necessarily P, therefore 3. Necessarily P - need not always be correct: the more general form would be 1. P. 2*If P, then necessarily Q, therefore 3*. Necessarily Q. In the section “Does conventionalism presuppose objects with mindindependent essences?”, we show how the object spaces (represented by “x” in the GPI) also need not be the same throughout. Such “resolutions” are rarely explicit. They can be discerned from hypothetical judgments the speaker (or relevant community members) makes, similar to other aspects of meaning. See, e.g. Chalmers 1996. Different views (and maybe different words) vary on just what samples are relevant for general/kind terms. Rarely will there be a “one-shot” baptism like “what is in this glass is water” (notice that introductions of terms can use, rather than mention the term); rather, there will be a pattern of use and in different circumstances, we may look at “the majority” or require a large majority, or may only focus on certain paradigm uses. As there is no “one size fits all”, our general schema is meant to be neutral on which samples matter, so “the relevant” samples is probably best. The order of explanation can go in either direction. If we already are aware of having certain sorts of terms, we can (implicitly) directly intend a term to be a K-term (“L is a chemical kind term”/“X is a furniture term”); in other cases, it is a K-term only because we constrain the application of the term by the same feature we do for a wide range of other terms. Sidelle 1989, p. 46; this depends on there being a relevant chemical similarity. More strictly, a term is introduced as a “chemical kind candidate” term. If there is no such commonality, we revert to what Putnam (1975, p. 142) calls “fall back conditions”. This often happens with names for diseases ( Livingstone-Banks 2018). For more detail, see Einheuser 2006, Livingstone-Banks 2017. For an alternative approach without accepting (A), see Sidelle 2009. Quine 1936. See also Dummett 1959, Hale 2002, 2013. One might think this illicitly assumes that what is necessary is necessarily necessary—that what is essential is necessarily essential—whereas conventionalists should deny this S4 axiom. This is a mistake. See above “The contingency problem”. The details will be fiddly—presumably, if Obama turns out to be a robot, we will judge him essentially a robot, so it isn’t just organism type. We might try “basic sort” as our general term, but it is unclear what that means, beyond how we constrain our counterfactual judgements about whether

451

Jonathan Livingstone-Banks and Alan Sidelle

30 31 32 33 34 35 36

something is Obama, or “this thing”. As Chalmers (2006) emphasizes, intensions need not be easily summarizable. Strictly, this need not be the case. Introducing a person-name underdetermines the physical extent of that thing: it can be reckoned as the size of a brain, or a human, or some sub-portion of a brain. Similarly, see Einheuser 2011. Alternatively, see Blackson 1992. For similar considerations, though not explicitly formulated in support of conventionalism, see Parfit 1984 Part 3, White 1989 and Kovaks 2016 and 2019 (among many others). Also, see Loets’ chapter on Persons. For detailed discussion of how essenced-objects may be generated from conventions and non-modal materials, see Sidelle 1989, ch. 3 and Einheuser 2011. Also the section: Does Conventionalism presuppose objects with mind-independent essences? Conventionalists themselves allow that there can be better/worse conventions. E.g. Locke 1689, e.g. Book III, ch. 6, sec 39; Carnap 1950; Sidelle 1989 and elsewhere. The classic work on Convention, Lewis 1969 clearly allows for better and worse conventions, though our usage is wider than his.

References Asta (2008) “Essentiality Conferred,” Philosophical Studies 140 (1): 135–148. Asta (2013) “Knowledge of Essence: The Conferralist Story,” Philosophical Studies 166 (1): 21–32. Banks, Jonathan Edward (2014) Antirealist Essentialism. PhD thesis, University of Leeds. Bealer, George (1999) “A Theory of the a Priori” Philosophical Perspectives 13: 29–55. Bealer, George (2002) “Modal Epistemology and the Rationalist Renaissance,” in T. Gendler & J. Hawthorne (Eds.), Conceivability and Possibility (Oxford: Oxford University Press): 71–125. Bengson, John (2015) “Grasping the Third Realm,” Oxford Studies in Epistemology 5: 1-38. Biggs, Stephen (2011) “Abduction and Modality,” Philosophy and Phenomenological Research 83 (2): 283–326 Biggs, Stephen and Wilson, Jessica (2019) “Abduction versus Conceiving in Modal Epistemology,” Synthese 198 (Suppl 8): 2045–2076. Black, Max (1958) “Necessary Statements and Rules,” The Philosophical Review 67(3): 313–341. Blackburn, Simon (1986) “Morals and Modals,” in G. Mcdonald & C. Wright (Eds.), Fact, Science, and Morality: Essays on A.J. Ayer′s Language, Truth, and Logic (Oxford: Blackwell): 119–141. Blackson, Thomas (1992) “The Stuff of Conventionalism,” Philosophical Studies 68 (1): 65–81. Cameron, Ross (2008) “Truthmakers and Modality,” Synthese 164 (2): 261–280 Cameron, Ross (2009) “What’s Metaphysical about Metaphysical Necessity?” Philosophy and Phenomenological Research 79 (1): 1–16 Cameron, Ross (2010) “On the Source of Necessity,” in Bob Hale & Aviv Hoffman (Eds.), Modality: Metaphysics, Logic and Epistemology (Oxford: Oxford University Press): 137–152. Cameron, Ross (2020) “Modal Conventionalism,” in O. Bueno & S. Shalkowski (Eds), Routledge Handbook of Modality (London: Routledge): 136–145. Carnap, Rudolf (1947) Meaning and Necessity (Chicago: The University of Chicago Press). Carnap, Rudolf (1950) “Empiricism, Semantics and Ontology,” Review Internationael de Philosophie 4 (11): 20–41 Carroll, L (1895) “What the Tortoise Said to Achilles,” Mind 4(14): 278–280. Candlish, Stewart (2019) “Private Language,” in Stanford Encyclopedia of Philosophy https://plato. stanford.edu/entries/private-language/ Chalmers, David (1996) The Conscious Mind (Oxford: Oxford University Press). Chalmers, David (2002) “Does Conceivability Entail Possibility?” in T. Gendler & J. Hawthorne (Eds.), Conceivability and Possibility (Oxford: Oxford University Press): 145–200. Chalmers, David (2006) “Two-Dimensional Semantics,” in E. Lepore & B. Smith (Eds.), The Oxford Handbook for the Philosophy of Language (Oxford: Oxford University Press): 574–606. Chalmers, David (2012) Constructing the World (Oxford: Oxford University Press). Chalmers, David (2014) “Strong Necessities and the Mind-Body Problem: A Reply,” Philosophical Studies 167 (3): 785–800

452

Conventionalism Chalmers, David & Jackson, Frank (2001) “Conceptual Analysis and Reductive Explanation,” Philosophical Review 110 (3): 315–361 Clarke-Doane, Justin (2019a) “Metaphysical and Absolute Possibility,” Synthese 198 (Suppl 8): 1861–1872 Clarke-Doane, Justin (2019b) “Modal Objectivity,” Noûs 53: 266–295. Coppock, Paul (1984) “Review of Nathan Salmon Reference and Essence,” Journal of Philosophy 81 (5): 261–270. Cowling, S (2013) “The Modal View of Essence,” Canadian Journal of Philosophy 43(2): 248–266. Denby, David (2014) “Essence and Intrinsicality,” In Robert M. Francescotti (Ed.), Companion to Intrinsic Properties (De Gruyter): 87–109. Dummett, M 1959 “Wittgenstein’s Philosophy of Mathematics,” Philosophical Review 68(3): 324–348. Einheuser, Iris (2006) “Counterconventional Conditionals,” Philosophical Studies 127 (3): 459–482 Einheuser, Iris (2011) “Towards a Constructivist Solution to the Grounding Problem,” Nous 45 (2): 300–314. Elder, Crawford (2004) Real Natures and Familiar Objects (Cambridge: Bradford, MIT Press). Fine, Kit (1994) “Essence and Modality,” The Second Philosophical Perspectives Lecture. Philosophical Perspectives 8: 1–16. Goswick, Dana (2018) “Ordinary Objects are Nonmodal Objects,” Analysis and Metaphysics 17: 22–37. Grice, H.P., & Strawson, Peter (1956) “In Defense of a Dogma,” The Philosophical Review 65(2): 141–158. Hale, Bob (2002) “The Source of Necessity,” Philosophical Perspectives 16: 299–319. Jackson, Frank (1998) From Metaphysics to Ethics (Oxford: Clarendon Press). Kovaks, David Mark (2016) “Self-Made People,” Mind 125: 1071–1099 Kovaks, David Mark (2019) “Diachronic Self-Making,” Australasian Journal of Philosophy 98: 349–362. Kripke, Saul (1980) Naming and Necessity (Cambridge: Harvard University Press). Lange, Marc (2008) “Why Contingent Facts Cannot Necessities Make,” Analysis 68: 120–128. Lewis, David (1969) Convention (Wiley-Blackwell). Lewis, David (1971) “Counterparts of Persons and their Bodies,” Journal of Philosophy 68 (7): 203–211 Lewy, Casimir (1976) Meaning and Modality (Cambridge: Cambridge University Press). Locke, John (1689) Essay Concerning Human Understanding (New York: Oxford University Press). Livingstone-Banks, Jonathan (2017) “The Contingency Problem for Neo-Conventionalism,” Erkenntnis 82 (3): 653–671 Livingstone-Banks, Jonathan (2017) “In defence of Modal Essentialism,” Inquiry 60(8): 816–838, DOI: 10.1080/0020174X.2016.1276855 Livingstone-Banks, Jonathan (2018) “The Case for a Meta-Nosological Investigation of Pragmatic Disease Definition and Classification,” Journal of Evaluation in Clinical Practice 2018 (24): 1013–1018. 10.1111/jep.13012 Mackie, J. L. (1974) “De What Re Is De Re Modality?” Journal of Philosophy 71 (16): 551–561 Mallozzi, Antonella, Vaidya, Anand, & Wallner, Michael (Fall 2021 Edition) “The Epistemology of Modality”, The Stanford Encyclopedia of Philosophy, Edward N. Zalta (ed.), forthcoming URL = https://plato.stanford.edu/archives/fall2021/entries/modality-epistemology/>. Noonan, Harold (1991) “Indeterminate Identity, Contingent Identity and Abelardian Predicates,” The Philosophical Quarterly 41(163): 183–193. Oderberg, David (2007) Real Essentialism (New York: Routledge). Parfit, Derek (1984) Reasons and Persons (Oxford: Oxford University Press). Pap, Arthur (1958) Semantics and Necessary Truth (New Haven: Yale University Press). Putnam, Hilary (1975) “The Meaning of ‘Meaning,’” Minnesota Studies in the Philosophy of Science 7: 131–193. Putnam, Hilary (1992) Replies, Philosophical Topics 20:347–408. Quine, W. V. O. (1983) “‘Two Dogmas’ Revisited,” in Philosophical Papers 3 (Cambridge: Cambridge University Press): 87–97. Quine, W. V. O. (1936) “Truth by Convention,” Philosophical Essays for Alfred North Whitehead (London: Longmans, Green & Co.): 90–124. Quine, W. V. O. (1951) “Two Dogmas of Empiricism,” Philosophical Review 60 (1): 20–43.

453

Jonathan Livingstone-Banks and Alan Sidelle Rea, Michael (2002) World without Design: The Ontological Consequences of Naturalism (Oxford: Oxford University Press). Searle, John (1969) Speech Acts: An Essay in the Philosophy of Language, Vol. 626 (Cambridge University Press). Sidelle, Alan (1989) Necessity, Essence and Individuation (Ithaca: Cornell University Press). Sidelle, Alan (1992) “Rigidity, Ontology and Semantic Structure,” Journal of Philosophy 89 (8): 410–430. Sidelle, Alan (1995) “A Semantic Account of Rigidity,” Philosophical Studies 80(1): 69–105. Sidelle, Alan (2009) “Conventionalism and the Contingency of Conventions,” Nous 43(2): 224–241. Sidelle, Alan (2010) “Modality and Objects,” Philosophical Quarterly 60 (238): 109–125 Stroud, Barry (1981) “Evolution and the Necessities of Thought,” in L. Sumner, J. Slater & F. Wilson (eds.), Pragmatism and Purpose (University of Toronto Press): 236–247. Sider, Theodore (2001) Four Dimensionalism: An Ontology of Persistence and Time (Oxford, GB: Oxford University Press). Sider, Theodore (2011) Writing the Book of the World (Oxford: Oxford University Press). Tennant, N (1987) “Conventional Necessity and the Contingency of Convention,” Dialectica, 41(1/2): 79–95. Thomasson, Amie (2007) Ordinary Objects (Oxford: Oxford University Press). Thomasson, Amie (2013) “Norms and Necessity,” Southern Journal of Philosophy 51 (2): 143–160. Thomasson, Amie (2020) Norms and Necessity (Oxford: Oxford University Press). Van Cleve, James (2006) “Descartes and the Destruction of the Eternal Truths,” Ratio 7: 58–62. Warren, Jared (2017) “Revisiting Quine on Truth by Convention,” Journal of Philosophical Logic 46 (2): 119–139. Warren, Jared (2020) Shadows of Syntax: Revitalizing Logical and Mathematical (Oxford University Press). White, Stephen (1989) “Metapsychological Relativism and the Self,” Journal of Philosophy 86 (6): 298–323. Wiggins, David (1979) “Ayer on Monism, Pluralism and Essence,” in Graham F. Macdonald (ed.), Perception and Identity (Ithaca: Cornell University Press): 131–160. Wiggins, David (1980) Sameness and Substance (Cambridge: Harvard University Press). Wildman, Nathan (2013) “Modality, Sparsity, and Essence,” The Philosophical Quarterly 63(253). 10.1111/1467-9213.12059 Wildman, Nathan (2021) “A Note on Lange on Contingent Necessity-Makers,” Erkenntnis 86: 762–771. Wright, Crispin (1986) “Inventing Logical Necessity” in J. Butterfield (ed.), Language, Mind and Logic (Cambridge): 187–209. Yablo, Steven (1990) “Review of Necessity, Essence and Individuation: A Defense of Conventionalism,” Philosophical Review 101 (4): 878–881

454

31 SOCIAL CONSTRUCTION Aaron M. Griffith

Reverse-engineering what makes it tick Dissecting the fine-tuned mechanism Rack and barrel, spring and pin Its synchronous characteristics To kill what makes it spin — Meshuggah, “Clockworks”

31.1

Introduction

Two sorts of claims are ubiquitous in philosophy: claims that something is essentially the way it is and claims that something is a social construction. Water is essentially H2O, heat essentially mean molecular motion. Race, gender, class, and disability are socially constructed. Given the ubiquity of these sorts of claims, it is an important philosophical task to understand how essentialist and social constructionist claims interact. The default position has been to view being a social construct as incompatible with having an essence or essential properties. Constructionists about gender and race, for example, have long warned about the dangers of essentialist thinking. However, there are many different notions of essence and therefore it is an open question whether being a social construct is compatible with having an essence or essential properties. The purpose of this essay is to explore the relation between essentialist and social constructionist claims. Among the questions considered are: can socially constructed things have essences or essential properties? Can humans construct essences? Are essences somehow fixed, necessary, or otherworldly such that nothing socially constructed can have them? Do claims about essences undermine the social and political importance of claims about social constructions? In Section 31.2, I outline a number of views about the nature of social construction. In Section 31.3, I outline a number of views about essence. In Section 31.4, I consider ways in which certain claims about social construction may be thought to challenge certain claims about essences. Section 31.5 then offers rejoinders to these challenges and attempts to point the way toward reconciling constructionist and essentialist claims.

DOI: 10.4324/9781003008750-36

455

Aaron M. Griffith

31.2

Social Construction

The metaphor of “construction” connotes building up from or out of something. But many alleged social constructions do not exhibit this sort of building or putting together. Nevertheless, what lies behind the metaphor of social construction is the idea of dependence. To say that something is a social construction is to say, at least, that it is dependent upon social factors. By “social factors” let us mean, generally, facts concerning collective human interaction. That includes collective beliefs, patterns of action, and practices. It also includes the ways that humans organize themselves and how they regulate their behavior through laws, customs, cultures, and norms, etc. Social constructions are dependent on—created, determined, produced, or controlled by—social factors.1 The focus of this chapter is on socially and politically significant constructions. According to Hacking (1999) constructionist claims are socially and politically significant when some entity that contributes to an unsatisfactory situation, e.g., injustice, is shown to be a noninevitable social product that is subject to change or eradication through social action. Claims of social construction are meant to reform, critique, and liberate. They have the function of revealing, unmasking, or debunking what is taken to be inevitable as social, alterable, and under our control. The social constructionist, says Haslanger, seeks to identify “levers for social change” (2012: 184). It is standard to distinguish between (a) what is constructed, (b) what does the constructing, and (c) how it is constructed (Mallon 2019). Regarding (a), constructionists disagree about the scope of construction. Is everything constructed or only certain kinds or domains of things? It is important to distinguish between the construction of “worldly” entities—objects, facts, properties, kinds, groups, institutions, states of affairs—and the construction of representations—concepts, theories, ideologies, ideas, beliefs, etc. (See Ritchie this volume Chapter 15 on representation and essence.) Scientific theories making reference to quarks, bosons, and electrons have a history involving human choices that constructionists have traced. While there are important issues surrounding how concepts, theories, and ideas are produced or selected, our focus will be on the construction of objects. The construction of things, properties, and kinds, in particular, are among the more controversial cases of construction. Once it’s been fixed what has been constructed, then focus shifts to (b), what does the constructing. What sorts of social factors are responsible for the existence or properties of the constructed entity? Suppose, for instance, the fact that a certain piece of paper is a US dollar bill is socially constructed. The piece of paper is an artifact that humans have created, but the status of the paper as a dollar bill is determined, on certain theories, by our collective attitudes towards it or bills like it (Thomasson 2003, Searle 2010, Epstein 2015). In other cases, human action, convention, and patterns of interaction are the producers. For example, recessions and racism seem to be produced by patterns of human interaction, even when the agents involved are unaware of their role in the production (Thomasson 2003, Khalidi 2015). In still other cases, material things are thought to be contributing to the construction of the object. Racial categories, for instance, are thought by some to be maintained and enforced by material infrastructure, e.g., the placement of highways and sources of pollution or the distribution of resources across a city (Sundstrom 2003, Taylor 2013, Mallon 2018, Liao and Huebner 2021). The next question is (c), how the entity has been constructed. The question here is the form which construction takes. Constructionists distinguish between causal and constitutive social construction. The distinction, roughly, is one between an object being the causal product of

456

Social Construction

social factors and the object (somehow) consisting in social factors. Environmental pollution may be regarded as a causal construction: smog is caused to exist and have its features by the collective activities of human beings. X is socially constructed causally iff social factors play a central role in causing the existence of X or X’s properties. (Cf. Haslanger 2003: 317 and Mallon 2019: 1.3). Some feminists, for example, have suggested that certain sex-typical physiological traits of males and females, such as height and physical strength, are causal products of centuries of patriarchal social arrangements.2 On social learning views of gender, for instance, gender (or gender characteristics) is the causal result of socialization or social learning. One is socialized from a very young age to accept certain masculine or feminine gender norms (appropriate temperament, character, interests, etc.) by one’s parents, peers, and culture.3 Gendered treatment of children and their subsequent conformity with gendered norms is causally responsible for one’s gender on these views. On the other side, something is said to be socially constructed constitutively when it is in some sense made up of (rather than caused by) social factors: X is socially constitutively constructed iff social factors play a central role in defining what X is (cf. Haslanger 2003: 318) Social kinds such as being a husband, being a war criminal, or being a landlord, are candidates for constitutive social constructions. Being a landlord, for instance, involves standing in social relations with tenants, contractors, and city officials, among other things. Recently, some have argued that constitutive social construction should be construed in terms of metaphysical grounding (Epstein 2015, Schaffer 2017, Griffith 2018a).4 Grounding is a non-causal form of dependence in which the grounded item exists/obtains in virtue of its grounds. Grounds non-causally generate or produce that which they ground. In the case of social construction, the grounds are taken to involve social factors, the grounded is the socially constructed object, and the grounding connection is the relation of social construction. One motivation for framing constructionist claims in terms of grounding is that it brings such claims into a more general framework of metaphysical dependence and structure.5 From here on out I’ll sometimes drop the “social” in “social construction”. Even though there are non-social kinds of construction, e.g., set construction, it should be understood that all references to construction are references to social construction.

31.3

Essence

Locke famously says, “Essence may be taken for the very being of any thing, whereby it is, what it is” (1975: III.iii.15). Talk of essence is talk of the nature of a thing, i.e., about what that thing is. While many doubt that essence can be analyzed in terms of modality, most agree that essentialist claims entail necessities. (See Correia (this volume, Chapter 8) on non-modal conceptions of essence.) If x is essentially F, then necessarily x is F (Fine 1994). It is important to distinguish between claims about the essence of individual things and those about the essence of kinds. (In this volume see Scarpati (Chapter 19) and Robertson Ishii (Chapter 11) on individual essentialism and chapters by Tahko (Chapter 10), Brigandt (Chapter 18), and Mallon (Chapter 24) on kind essentialism.) The former concern the nature of individuals whereas the latter concern the nature of kinds. Witt (1995, 2011) argues convincingly that the

457

Aaron M. Griffith

two kinds of essentialism are distinct. While there are many interesting questions regarding individual essentialism (e.g., “Do I essentially have the gender I do?”), our focus here will be on kind essentialism. The reason for this is that many of the most interesting and important social constructionist claims are about kinds (categories, types) of things rather than individual members of kinds. One way to conceive of an essence is as the set of properties that make a thing what it is (Oderberg 2007). Another approach to essence is to view a statement of the essence as giving a “real definition” of a kind.6 This is not merely a definition of the word denoting the thing, but of the very thing itself (Fine 1994, 1995). According to Kit Fine, essentialist claims have the form “It lies in the nature of x that p”. For example, it lies in the nature of heat to be mean molecular kinetic motion. The essence of an object, on Fine’s view, is the collection of propositions that are “rendered true” by the very nature of the object. I’ll frame things in terms of essential properties as we go, but those who like Fine’s framework should feel free to translate talk of essential properties into talk of essentialist truths and talk of essences into talk of sets of truths rendered true by the very nature of the object. There are a variety of forms that kind essentialism can take, some stronger than others. In its most robust form, a statement of a kind K’s essence will identify, a Properties that are each necessary and jointly sufficient for membership in K. b Properties that distinguish K from other kinds. c Properties that explain and allow us to predict the characteristic features of Ks because they either cause or constitute those features. d Properties of the following sort: • Universals that all members of K share, • Objective, natural, and mind-independent properties, e.g., properties studied by the natural sciences, • Intrinsic properties, i.e., properties that their bearers would have if they were isolated from all other things, • Non-disjunctive properties. Let Strong Essentialism about a kind K be the most robust form of essentialism: Strong Essentialism: For a kind K, there are properties that satisfy (a) through (d). Obviously, there are a variety of views that are weaker than Strong Essentialism. Weaker forms of essentialism may meet some but not all the conditions in (a) through (d). For example, a view might allow that the properties of K are necessary and sufficient for kind membership but deny that they allow us to predict characteristic features of Ks. Or a view might allow some of the properties in K’s essence to be relational, extrinsic, or minddependent properties of their bearers. So let’s acknowledge that there is a spectrum of views between Strong Essentialism and what can be called Weak Essentialism: Weak Essentialism: For a kind K, K has some essential properties. This form of essentialism only requires the kind to have at least one essential property. The properties mentioned in (a) through (d) are examples of the sorts of essential properties K may have, though they are not limited to them.

458

Social Construction

31.4 Social Constructionist Challenges to Essentialism In this section, we’ll consider several arguments for thinking that socially constructed kinds have no essences. We won’t consider arguments that cast doubt on essentialism in general that would thereby also apply to constructed things. Rather the focus will be on arguments that attempt to show that being a social construct is incompatible with having an essence.

31.4.1 Argument 1: Necessary and Sufficient Conditions 1 K is socially constructed. 2 If K is socially constructed, then there are no necessary and sufficient conditions for membership in K. 3 Therefore, there are no necessary and sufficient conditions for membership in K. 4 If K has an essence, then there are necessary and sufficient conditions for membership in K. 5 Therefore, K does not have an essence. The crux of the argument is premise (2). If there were necessary conditions for belonging to a kind, then all members of the kind would have these properties in common. Constructionists about gender, for example, worry that specifying necessary conditions for womanhood would require all women to have certain physical features, beliefs, experiences, self-conceptions, etc. But this would ignore and obscure the vast differences between women (Spelman 1988). A related worry is that specifying necessary conditions for kind membership would inevitably exclude from membership people who ought to be included in the kind.7 Regarding gender, trans women would be excluded from womanhood if being a woman required having certain anatomical features or certain early life experiences. The more general worry is that members of constructed kinds exhibit too much diversity to share a rigidly defined essence. (See Rosario this volume Chapter 25 for discussion of sex, gender, and essence.) Constructionists who want to push back on the argument could reject (2). They can allow disjunctive or gerrymandered properties to be part of the essence of a constructed kind K. That would allow them to say that it is necessary and sufficient for being a K that one has F or G or H, etc. This strategy offers prospects for identifying necessary and sufficient conditions for being a K without the implication that all members of K share specific, non-disjunctive, intrinsic properties. On the other hand, the constructionist might reject (4). They could point out that only Strong Essentialism demands that essences involve properties that are necessary and sufficient for membership in the kind. Some weaker essentialisms are compatible with the view that there is some threshold for membership: if x has enough of the features then x is a member without any particular set of features being necessary.8 Views that treat kinds as homeostatic property clusters (HPC) have it that members of a kind tend to share features. (For further discussion, see Brigandt (Chapter 18), Mallon (Chapter 24), and Dumsday (Chapter 12) in this volume.) These properties are reliably exemplified together due to some mechanism (causal or otherwise), such that they individuate the kind and allow for useful explanation and prediction about the kind, without requiring that all members of the kind have the same properties (Boyd 1999, Mallon 2016, and Brown this volume Chapter 23). As long as there is a stable cluster of these properties due to some mechanism which sustains the clustering of the properties (due presumably to social factors in the case of social kinds), we have a kind with a nature. HPC social kinds would still allow us to demarcate the boundaries of the kinds and

459

Aaron M. Griffith

allow for generalization and inferences about members of the kind without requiring universal necessary features of all members (cf. Stoljar (1995: 283). Even though Weak Essentialism only requires that the kind have an essential property, it may not get the constructionist out of the concern at the root of Argument 1. The concern is that essentialist claims about constructed kinds entail that there is something in common, indeed, something necessary to all members of the kind. After all, if Ks (qua Ks) are essentially Fs, then Ks are necessarily Fs. That there are necessary features of Ks may appear to be at odds with the constructionist insistence that constructed kinds are not inevitable, need not be as they in fact are, and are “not determined by the nature of things” as Hacking says.9 So there is no guarantee that weaker essentialisms about socially constructed kinds avoids the worries about accounting for diversity and avoiding exclusion. Can essentialism do without necessary conditions on kind membership? Can an essence be constituted by a set of merely sufficient properties? One could have a view of social kinds on which the essence of the kind is made up of a set of properties that are each sufficient for kind membership. Such a view is consistent with some accounts of essential properties, e.g., A property P is an essential property of being an F iff anything is an F partly in virtue of having P. A property P is the essence of being an F iff anything is an F in virtue of having P. The essence of being F is the sum of its essential properties. (Devitt 2008: 345). Devitt’s definition allows for a range of properties P1, … , Pn to belong to the essence of F if for each property, Pi, something is an F in virtue of having Pi. It need not follow from the possibility that several properties may be sufficient without any being necessary for kind membership that the kind would be purely disjunctive, e.g., to be K is to be either P1, or P2, or P3. Khalidi (2013: 16ff.) discusses “polythetic” kinds, i.e., kinds that are not defined in terms of a set of individually necessary and jointly sufficient properties. Some polythetic kinds are defined by “complex constructions of properties, such as K1 = P1 & (P2 v P3), or K2 = P1 & ((P2 V P3) & P4)” (2015: 16). Other polythetic kinds may be such that any two members of the kind K will share properties but where there is no single (non-disjunctive) property had by all members of K. So while K is necessarily P if K is essentially P, P itself may be a complex property that involves disjunction. This allows different members of K to display important differences. It remains to be seen whether any constructed kinds exhibit essences like this, but it is a metaphysical possibility that the essence of constructed kinds need not involve a set of individually necessary and jointly sufficient properties that rigidly demarcate the kind.

31.4.2 Argument 2: Mind-(In)dependence Our second argument pits the mind dependence of constructed kinds against the mind independence of essences. 6 7 8 9 10

K is socially constructed. If K is socially constructed, then K is mind dependent. Therefore, K is mind dependent. If K has an essence, then K is mind independent. Therefore, K does not have an essence.10

460

Social Construction

The notions of mind dependence and mind independence admit of many different characterizations. Therefore, there is room for nuance when it comes to evaluating Argument 2. A kind may be mind-dependent in different senses. In a well-known paper, Khalidi (2015: 9) outlines three forms of mind-dependence kinds might exhibit. i K depends neither for its existence nor the instantiation of its instances on our having propositional attitudes toward K or its instances, respectively. E.g., racism and recessions. ii K depends for its existence, but not the instantiation of its instances, on our having propositional attitudes towards K or its instances, respectively. E.g., war and money. iii K depends for its existence and for the instantiation of its instances on our having propositional attitudes towards K and its instances, respectively. E.g., permanent resident, prime minister. Kinds of type (i) are mind independent with respect to propositional attitudes toward the kind itself. Yet, that is compatible with being dependent on a variety of social factors, including minds and beliefs (those not directed at the kind). So, until the proponent of Argument 2 specifies the sense in which K is or isn’t mind dependent, there is no real challenge to thinking that K being socially constructed is incompatible with K’s having an essence. So far, Argument 2 has been discussed as if there is no distinction between K and K’s essence. But the constructionist might insist that once the distinction is granted, there is no tension between K itself being socially/mind dependent and K’s essence being mind independent. We can hold that for K to exist and to have members depends on social factors, including collective beliefs. But K’s essence—what it is to be a K (generically)—is independent of any social factors whatever. Even if K’s essence involves (i.e., its definition mentions) social factors, e.g., to be a plumber is to perform a certain function and to stand in certain social relations, it does not follow that K’s having the essence it does depends on social factors (see Mason 2020, 2021). And that, even if K’s having any members depends on contingent social factors.

31.4.3

Argument 3: Otherworldliness 11

Our next argument is related to the previous one. Here, the argument attempts to find an incompatibility between the social dependency of constructed kinds and the independence of essences from worldly circumstances. 11 K is socially constructed. 12 If K is socially constructed, then K is embedded, i.e., K exists in virtue of circumstances in the world. 13 Therefore, K is embedded, i.e., K exists in virtue of circumstances in the world. 14 If K has an essence, then K is otherworldly, i.e., K exists regardless of circumstances in the world. 15 Therefore, K does not have an essence. Socially constructed kinds are products of the social world. They seem to exist because of specific circumstances in the concrete world and the basis for their instantiation is worldly circumstances. Such kinds seem to be “embedded”, as Raven (2022) says. Essences—and the kinds that have them—on the other hand, are conceived of as “detached” and “otherworldly”, i.e., as things that exist regardless of circumstances in the concrete world

461

Aaron M. Griffith

(Fine 2005). Presumably, their otherworldliness is what makes them objective, mindindependent features of the world as it is in itself. Hence, Argument 3 could supplement Argument 2 in supporting the conclusion that constructed kinds don’t have essences. One response to this argument would be to insist on a distinction between the existence of K and K’s instantiation (or its having actual members). One could then accept (14)—kinds exist and have essences regardless of circumstances in the world—but reject (12). It is not that socially constructed kinds exist because of circumstances in the world. Rather, they come to be instantiated in virtue of the circumstances in the world; things are constructed as members of kinds. Thus, kinds could have otherworldly essences, but embedded instantiation conditions. That would resolve the apparent tension between kinds being embedded and otherworldly. But this response comes at a cost for the constructionists about kinds. Kinds themselves would no longer be constructed but be otherworldly, existing independently of human existence or activity. The kind US citizen would, for instance, exist and have its nature independently of the existence of the US or the activity of its government and citizens. Humans would not even have a choice about the existence, persistence, or membership conditions of kinds. The only thing that would really be constructed would be whether anyone is in fact a member of these kinds. Alternatively, one could distinguish between the kind and its having members on the one hand, and its essence on the other. The existence of the kind and its having members is embedded, yet the essence itself is otherworldly. Being money, for example, looks to be the sort of kind whose instantiation requires the social world to be a certain way. But the essence of being money, i.e., serving the function of being a medium of exchange, looks to be fixed, not based in worldly circumstances (Passinsky 2019 cf. Mason 2020: 60ff.). Resistance to this response to Argument 3 is rooted in the conviction that humans, through our beliefs, actions, and practices, can in fact produce kinds, and not just their instantiation or membership conditions. We decide those conditions ourselves. Discussing institutional kinds like being a US citizen or being a dollar bill, Thomasson writes, Our acceptance of a set of conditions C as sufficient for being K is constitutive of what conditions suffice for being K, so what conditions there are is determined by what conditions we accept. As a result, we could not turn out to be mistaken—our acceptance of the set of conditions C declaratively establishes the conditions for being K rather than attempting to describe pre-existing and independent conditions for being K. (2003: 588–9). If setting the membership conditions for a kind establishes its essence, then, on Thomasson’s view, we can fix the essence of certain kinds, namely the ones whose membership conditions are up to us.12 There may be other ways to establish the nature of social kinds. According to Epstein (2015), we can “anchor” kinds by setting up grounding conditions of the kind. While he doesn’t frame things in terms of essence, anchoring could be interpreted as the construction of the essence (or essential properties) of kinds. If the grounding conditions of K help determine what K is, its nature, and we anchor these conditions through our beliefs, practices, and actions, then the nature or essence of K would seem to be constructed by us. So there is a strong intuitive pull to the embeddedness of social kinds. We seem to require strong arguments for thinking that essences of constructed kinds are otherworldly and not embedded. But none of the conditions (a) through (d) listed above entail that essences

462

Social Construction

are otherworldly or that humans cannot play a role in constructing essences themselves. (Of course, certain constraints on the properties that make up essences, e.g., those enumerated by (a)–(d) above, may have this implication.) The conviction that constructed kinds are embedded seems to be rooted in the thought that constructed items are not inevitable, a thought that leads us to Argument 4.

31.4.4

Argument 4: Inevitability

16 K is socially constructed. 17 If K is socially constructed, then K is a contingent, culturally bound product that need not have existed and may be changed or eradicated given the appropriate social changes. 18 Therefore, K is a contingent, culturally bound product that need not have been and may not be given the appropriate social changes. 19 If K has an essence, then K is inevitable. 20 Therefore, K does not have an essence. The non-inevitability of constructed kinds is at the heart of constructionist claims (Hacking 1999). Constructed kinds need not have existed or need not have the features they do. Moreover, it shores up the social justice motivation for constructionism: if a kind (or the characteristics associated with it) is not inevitable, but rather up to us, then we can change or eradicate the kind. The non-inevitability of constructed kinds also supports their fluidity and their possibility of change over time. Like the worry expressed in Argument 1, the worry expressed in Argument 4 is rooted in the constructionist aims of justice. If a kind K has an essence such that the essence is inevitable—in the sense of being determined, fixed, and unavoidable through social change—then the constructionist debunking strategy will fall flat. There would be no political point in highlighting the real but covert nature of K if membership in K has unjust consequences and there was nothing we could do about it. The power of constructionist claims is supposed to lie in their exposing the social world for what it is in order to point the way towards positive social change. Argument 4 turns on premise (19), the claim that essences are “inevitable”. If a kind is inevitable, then it either exists come what may or it has its characteristics come what may. Essences, according to (19), are fixed and unchanging. Inevitability entails a lack of control on our part: it is not up to us whether the thing exists or what it’s like. Being inevitable in this sense is straightforwardly incompatible with being a social construct. The question then is whether essences are inevitable in this sense. One reason, connected to Argument 3, to think they are, is if essences are otherworldly, Platonic entities. Of course, the essences themselves may exist come what may, but it may be in our control whether such essences get instantiated. So even if the nature of some kinds is not up to us, whether it is part of the social world may be. Another reason to think that essences are inevitable is that—at least according to Strong Essentialism—the properties that can make up essences are natural, fixed, intrinsic, and mind-independent. That entails a robust sense of inevitability: the nature of kinds is simply not up to us in the sense of being dependent upon us for its existence or features. While the strongest form of Strong Essentialism would secure the win for Argument 4, Weaker Essentialisms do not have this result. Weak Essentialism does not entail that the properties in an essence need be natural or mind-independent. Moreover, these constraints on essences

463

Aaron M. Griffith

may seem unmotivated. There is little reason to deny that the kind shopkeeper has a nature that consists in relational, social, and mind-dependent properties. There is something it is to be a shopkeeper and it is in our control whether there are shopkeepers and what they are like. However, Argument 4 cannot be so easily dispatched. One further motivation for the incompatibility between inevitable essences and non-inevitable constructs is that essentialities entail necessities.13 As was discussed above, if Ks are essentially F, then necessarily, members of K, qua members of K, are F. So there is a clear sense in which the features of Ks are inevitable: necessarily, if there are Ks, then Ks are F. (I am not assuming that K is necessarily instantiated or that being a K is essential to individual members of K.) Another reason for thinking there is a genuine tension between K having an essence and K being socially constructed in terms of the inevitability of essence is the “looping effect”. Hacking writes, “People classified in a certain way tend to conform to or grow into the ways that they are described; but they also evolve in their own ways, so that the classifications and descriptions have to be constantly revised” (1995: 21). Kinds like multiple personality disorder, widow, refugee, and others are “interactive kinds”, according to Hacking, insofar as the kind modifies it occupants, who in turn modify the kind itself. (See Brown this volume Chapter 23 on essence and psychiatric kinds.) If essences are inevitable—fixed, otherworldly, unchangeable through social means—then interactive kinds lack essences. The reason is that the essence of one and the same kind K cannot change due to the activities of K’s members. At best, the members of K at t1 could construct and come to constitute the members of another kind K* at a later time t2. K and K* would not be identical kinds (despite possibly sharing members at different times). At any rate, there would be no real looping effect, but rather only the instantiation of one kind at one time and the instantiation of another kind at another time. A final complication for Argument 4 is that “inevitability” admits of various aspects and dimensions. Social constructions can be inevitable in some respects and non-inevitable in others. Race constructionists might think that it is inevitable that differences in skin color, physical morphology, and ancestry will always have some social meaning. That is, there will always be some social construction out of these features, though not necessarily the current construction we observe. Race is not inevitable insofar as the social arrangements, practices, and beliefs produced around different morphologies are liable to be changed. (See Mallon this volume Chapter 24) on racial essentialism.) Such claims are important to achieving social justice insofar as they reveal the nature of the categories with which we are concerned. When an account of the nature of social categories explains what social arrangements constitute the category, it helps pinpoint the specific aspects of the social world that need to change for the category to be eradicated (or at least un-instantiated). Moreover, we would not know how to identify the causes of a social category without understanding the nature and constitution of the category first. So what is the relevant sense of “inevitable” that would make constructed kinds ineligible for having essences? I think it is this: if essences are inevitable in the sense that we have no social control over their existence, instantiation, and their constitutive properties, then constructed kinds have no essences. If a kind exhibited no dependence (at any time) on our individual or collective beliefs, actions, and practices, then it would seem to be inevitable in a way that rules out its being a social construct. What Argument 4 comes down to is whether the properties in an essence are such that we could have this sort of control over them (either their instantiation or them being part of the essence). Conditions (a) through (d) alone do not rule out essences that we control, but only the added

464

Social Construction

constraints on properties (natural, fixed, intrinsic, mind-independent) in an essence stipulated by strong essentialism. As far as I can tell, weaker essentialisms allow for socially constructed kinds to have essences that are non-inevitable (in some sense relevant to constructionist aims).

31.5

Social Kind Essentialism

We have just looked at several arguments for thinking that being a social construct is incompatible with having an essence. In this section, we will investigate the positive case for socially constructed kinds having essences (or at least some essential properties), i.e., social kind essentialism.

31.5.1

Constructionist Claims and Aims

The first reason in favor of social kind essentialism is that constructionists themselves have often, even if not explicitly, expressed their constructionist claims in terms of essence (Passinsky 2019: 10). Consider Haslanger’s definition of constitutive construction: Y is social constructed constitutively as an F iff Y is of a kind or sort F such that in defining what it is to be F we must make reference to social factors (or: such that in order for Y to be F, Y must exist within a social matrix that constitutes Fs). (Haslanger 2003: 318) One way to interpret what she is saying is that a kind is constitutively constructed just in case social factors are part of its essence, part of “what it is to be” F. Another, antiessentialist, reading is that Haslanger is simply talking about the definition of the term “F”. But Haslanger herself indicates that what she has in mind is the definition of kind F—its real definition—rather than the term “F” (Haslanger 2013: 31). On Haslanger’s view, being constructed is itself defined in terms of essence. So, there couldn’t possibly be a conflict between being a social construct and having an essence on this definition of being constructed. Moreover, the debunking project that is so central to many constructionists aims appears to involve identifying social essences. Understood in terms of essence, the debunking project would aim at revealing that things we believed to have natural essences in fact have social essences. As Ásta says, debunking work consists in “exposing the beast for what it is” (2018: 36 emphasis added).14 Assuming (plausibly) that these authors are not ignorant of the relation their claims have to essentialist views, we should interpret them as making essentialist claims about constructed kinds. A second reason for social essentialism is that many of the most influential constructionist accounts of kinds like gender and race seem to entail that these constructions have essential properties. Commenting on Catherine MacKinnon’s (1989) influential definition of women as sexual objects, Allison Stone writes, My claim that theorists such as MacKinnon are essentialists might sound odd, given the frequent contrast between essentialism and social constructionism. Yet social constructionists can readily be essentialists if they believe—as do these influential feminist theorists—that a particular pattern of social construction is essential and universal to all women. (2004: 40)

465

Aaron M. Griffith

Similarly, in her well-known ameliorative account of womanhood, Haslanger offers an explicit definition of womanhood, i.e., necessary and sufficient conditions that define womanhood: S is a woman iffdf S is systematically subordinated along some dimension (economic, political, legal, social, etc.), and S is ‘marked’ as a target for this treatment by observed or imagined bodily features presumed to be evidence of a female’s biological role in reproduction. (2012: 230). Of course, each of these accounts has faced worries about them being “essentialist” in problematic ways.15 The point is, however, that many avowed social constructionists have taken themselves to be offering accounts of the essences of constructed kinds. That means that far from there being an incompatibility between construction and essence, these thinkers take the constructionist project to fundamentally involve articulating essences. A related line of argument is that an essentialist framework is uniquely conducive to making sense of certain debates in social metaphysics. Asya Passinsky (2019) argues that Fine’s notion of essence is helpful in clarifying and rigorously framing the debate between Haslanger (2000) and Jenkins (2016) over womanhood. Passinsky claims that this debate is fundamentally about the essence or real definition of womanhood. Framing the debate in terms of essence helps clarify what sort of explanation Haslanger and Jenkins are attempting to give of womanhood, namely an essentialist explanation of womanhood rather than causal, rationalizing, or functional explanations. According to Passinsky, the essentialist framework also explains disagreements about modal claims regarding social kinds. A debate about whether members of a certain constructed kind could have or lack certain features makes perfect sense in terms of disagreements about the real definitions of the kind. Since real definitions entail necessities, i.e., if being F is part of the real definition of x, then necessarily, x is F, different views about whether it is possible for members of a social kind to have or lack a certain property are ultimately disagreements about the real definitions of these kinds. Conversely, one could argue that anti-essentialism about constructed kinds actually undermines constructionist claims and aims.16 Suppose we believe that kind K is socially constructed and deny that K has an essence (in a strong or weak sense). And suppose that K is a politically relevant kind (e.g., woman, LGTBQ+, refugee, undocumented immigrant). On these assumptions, we could not even say that K is essentially a social kind. That looks problematic given the constructivist aim of debunking the apparent naturalness of K. Moreover, constructionists aim to identify (more or less) unified groups about which socially and politically helpful generalizations can be made. Essentialisms of various forms offer ways to unify and generalize about social kinds. To this extent, some form of social essentialism may help these constructionist aims. The more anti-essentialism undermines our ability to identify commonalities among members of a group and say something general about the group, the harder it is for anti-essentialism to support unified political coalitions (cf. Zack 2005: ch. 2). Some unity among Ks is required to form a coalition and if there are no generalizations to make about members of K (as Ks), then it would be difficult to advocate for all members of K. Successful social movements rely on some dimension of commonality among and projectability to the members of the kind. Now whether that requires a commitment to Strong or some weaker Essentialism is debatable. Even if one is not convinced that a commitment to some form of essentialism is required for constructionist political projects, it could be argued that essentialism could be an auxiliary help to

466

Social Construction

social movements insofar as it would help demarcate the social groups that are seeking to establish solidarity and political representation. Still another case to be made for social essentialism is the argument that essentialist claims are in fact central to the social justice aims of constructionists. (See Stoljar (this volume Chapter 26) on social justice and essence and Rosario (this volume Chapter 25) on sex and gender.) Above, we saw that a number of constructionists (mainly about gender, sex, and race) seem to frame their constructionist claims as essentialist claims. That’s no accident since the debunking project is aimed at revealing certain kinds for what they really are, i.e., revealing their essence, or at giving politically expedient definitions of kinds. Debunking claims are supposed to reveal to us that some kind is contingent, non-inevitable, liable to be changed, and under our control. This is central to the social justice aims of constructionists because the revelation about the nature of a kind is meant to aid in eradicating or altering that kind. More specifically, constitutive constructionist claims look to advance our understanding of the building blocks, structures, and mechanisms that create and sustain constructed kinds. That has the potential to open ways for activists to challenge unjust social kinds and social structures.

31.5.2 Weak Social Kind Essentialism Another class of arguments for social essentialism appeals to the differences between the forms of essentialism outlined in section 31.3. The argument is that while social kind essentialism may be incompatible with Strong Essentialism (because of the constraints it puts on the properties belonging to an essence), it is not obviously incompatible with weaker forms of essentialism that fall between Strong and Weak Essentialisms. Skepticism about essentialism is primarily rooted in worries about the strongest form of essentialism and the requirements it sets on essences, e.g., that essence involve necessary and sufficient conditions on membership, intrinsic, natural, and mind-independent properties. Each of these claims about essences seems optional, not essential to essences. Prima facie, constructed kinds could have essences that are made up of relational, mind-dependent, social properties. Weak Essentialism, recall, simply says that a kind has some essential property. Weak social essentialism, therefore, avoids most of the serious worries about essentialism. Moreover, attending to the objectual/generic distinction (see Koslicki and Raven Introduction to this volume) can help make the case for a version of Weak Social Kind Essentialism. That distinction, recall, is between what is essential to an object and what is essential for being of a certain kind. In an objectual sense—where the question is “what is a?”—it may be straightforward to establish Weak Social Kind Essentialism. The “object” in this case is the kind K itself. If it can be shown that K is essentially socially constructed, essentially mind dependent (Mason 2021) or essentially historical (Bach 2012), then Weak Essentialism is true of social kinds in the objectual sense. Generic essentialism appears to be the more contentious version of essentialism to maintain, though. Most anti-essentialists about social kinds are not concerned about the status of the kind itself, but with what it takes to be a member of the kind. Weak Social Kind Essentialism about K in the generic sense must maintain that there is at least one property essential for being a K. Such a view faces the worries raised above (section 31.4) about specifying what property all members of K have in common. Perhaps this worry can be mitigated, however, if the property(s) essential for being K is generic enough to admit of being realized in different ways, e.g., Haslanger’s notion of being systematically subordinated “along some dimension”. Multiply realizable or determinable essential properties might allow

467

Aaron M. Griffith

enough variability among the members of the kind to avoid requiring problematic commonality and universality among K’s members.17

31.5.3

General Metaphysical Considerations

The final case for social kind essentialism I’ll consider appeals to general metaphysical considerations about essences. One line of thought is that everything has an essence (or essential properties), including, eo ipso, socially constructed kinds. Rebecca Mason, for instance, expresses the view like this: [T]here is no obvious reason to deny that social kinds have essential properties. If a kind, K, exists, then there is something that it is to be K. Moreover, the properties that specify what it is to be K are the essential properties of that kind. This is so whether the kind in question is social, psychological, biological, chemical, or physical, and so forth. (2016: 844) This argument connects K existing, with there being something it is to be K, with K having essential properties. Presumably, the argument is that for everything that exists, each thing is what it is and not something else. And if each thing is what it is, then, there is something it is to be that thing. Whatever it is to be that thing is the essence of that thing. These considerations are general enough that, if correct, they would entail that constructed kinds have essences. Passinsky (in conversation) pushes a similar line, arguing that if anything has an essence then everything does. (She adopts the Finean conception of essence on which an essence is a collection of truths of the form “it lies in the nature of x that p”.) Her reason is that it would be problematically arbitrary for some things to have essential properties while other things completely lack them. She argues that any way of explaining this asymmetry (existence/nonexistence, mind-independence/mind-dependence) violates essentialist intuitions many have. For instance, can-openers are mind-dependent, but they do not seem to lack essential properties, e.g., the function of opening cans, for that reason.

31.6 Conclusion What I’ve done here is provide a summary of the main questions and options about how essence and social construction relate. What we’ve found is that there is a solid case to be made that the strongest form of essentialism is incompatible with social constructionism. But there are many nuanced and weakened versions of essentialism that may be compatible with, and even advantageous for, social constructionism. Obviously, there is much more to say about each of the arguments raised here. I hope that the present work aids in structuring further discussion about essence and social construction.

31.7

Related Topics

Passinsky, Asya, “Artifacts, Artworks, and Social Objects”, Chapter 17 Loets, Annina, “Persons”, Chapter 22 Brown, Danielle, “Psychiatric Kinds”, Chapter 23 Mallon, Ron, “Race”, Chapter 24 Rosario, Esther, “Sex and Gender”, Chapter 25

468

Social Construction

Stoljar, Natalie, “Social Justice”, Chapter 26 Vaidya, Anand and Wallner, Michael, “Conferralism”, Chapter 32

Notes 1 Classic texts on social construction include Berger and Luckman (1966), Latour and Woolgar (1979), and Hacking (1999). 2 See Butler (1993) and Fausto-Sterling (2000). This example is discussed by Diaz-Leon (2013: 1144). 3 See Millett (1971) and Kimmell (2000). Discussed by Mikkola (2017). 4 For concerns about the grounding approach to social construction see Barnes (2014), Pagano (2021), and Passinsky (2019). 5 Another form of non-causal construction is conferralism, defended by Ásta (2013 and 2018). See the entry on conferralism and essence from Vaidya and Wallner (this volume Chapter 32). Others have used the notion of response-dependence to understand social construction. See Pettit (1991), Hindriks (2006), and Passinsky (2020). 6 The connection between essence and real definition (as well as identity and grounding) remains a matter of discussion. Fine (1994, 2015) and Dasgupta (2014) associate giving essences with giving real definitions. Others have analyzed real definitions in terms of ground ( Rosen 2015). See Correia (2013, 2017), Correia and Skiles (2019), and Koslicki (2012) for further discussion. 7 See Jenkins (2016), Dembroff (2020), and Ásta (2018) who raise this worry for definitions of womanhood. 8 See Mallon (2007: 157), Mason (2016: 844), Stoljar (1995), and Taylor (2013). 9 Marques (2017: 18) voices a worry like this about Haslanger’s account of gender. 10 See Heyes (2000), Stone (2004), and Witt (1995) for discussions of this sort of worry regarding gender and sexuality. Mason (2021) argues that social kinds are essentially mind-dependent. 11 This argument is inspired by and framed in terms of Raven (2022). 12 One might admit that the membership conditions commonly associated with a kind K are up to us, but deny that those conditions establish the essence of K. Indeed, the stipulated membership conditions may be seen as constituting the nominal, but not real, essence of K. See Passinksy (manuscript), Dembroff (2018), and Barnes (2020) for views on which common classificatory practices (do or should) come apart from actual membership conditions of a social kind. 13 Though see Mackie (this volume?) who denies the entailment. 14 Although not all debunking projects need aim at revealing a social essence where we thought there was a natural essence. One could debunk some claim or ideology by showing that there is in fact no such kind in the world at all. This sort of anti-realist debunking project would not entail any commitment to social essentialism. See Appiah (1996) and Glasgow (2009) for influential anti-realist accounts of race. 15 See Jenkins (2016) for concerns over Haslanger’s (2000) account wrongly excluding trans women from being women. 16 See Mallon (2007) and Bach (2012). See Stubblefield (1995) for a discussion about the tension between non-essentialist views of race and the goal of legitimating black solidarity. 17 See Griffith (2018b) for a view on which some forms of social construction can be understood in terms of realization.

References Alcoff, L. (2006). Visible Identities. Oxford: Oxford University Press. Appiah, A. (1996). “Race, Culture, Identity: Misunderstood Connections,” in Color Conscious, Anthony Appiah and Amy Gutmann. Princeton, NJ: Princeton University Press. Ásta. (2013). “The Social Construction of Human Kinds.” Hypatia 28(4): 716–732. Ásta. (2018). Categories We Live By. New York: Oxford University Press. Bach, T. (2012). “Gender is a Natural Kind with a Historical Essence.” Ethics 122(2): 231–272. Barnes, E. (2014). “Going Beyond the Fundamental: Feminism in Contemporary Metaphysics.” Proceedings of the Aristotelian Society CXIV(3): 335–351. Barnes, E. (2020). “Gender and Gender Terms.” Nous 54(3): 704–730.

469

Aaron M. Griffith Berger, P. and Luckman, T. (1966). The Social Construction of Reality. Garden City, NY: Anchor Books. Boyd, R. (1999). “Homeostasis, Species, and Higher Taxa”, in Species: New Interdisciplinary Essays, edited by R. Wilson, 141–185. Cambridge: MIT Press. Correia, F. (2013). “Metaphysical Grounds and Essence,” in Varieties of Dependence. Ontological Dependence, Grounding, Supervenience, Response-Dependence, edited by M. Hoeltje, B. Schnieder and A. Steinberg, 271–296. München, Philosophia: Basic Philosophical Concepts Series. Correia, F. and Skiles, A. (2019). “Grounding, Essence, and Identity.” Philosophy and Phenomenological Research 98(3): 642–670. Dasgupta, S. (2014). “The Possibility of Physicalism.” The Journal of Philosophy 111(9/10): 557–592. Dembroff, R. (2018). “Real Talk on the Metaphysics of Gender.” Philosophical Topics 46(2): 21–50. Dembroff, R. (2020). “Beyond Binary: Genderqueer as Critical Gender Term.” Philosophers’ Imprint 20(9): 1–23. Devitt, M. (2008). “Resurrecting Biological Essentialism.” Philosophy of Science 75: 344–382. Diaz-Leon, E. (2013). “What Is Social Construction?” European Journal of Philosophy 23(4): 1137–1152. Epstein, B. (2015). The Ant Trap. New York: Oxford University Press. Fausto-Sterling, A. (2000). Sexing the Body. New York: Basic Books. Fine, K. (1994). “Essence and Modality,” in Philosophical Perspectives 8: Logic and Language, edited by J. E. Tomberlin, Atascadero, CA: Ridgeview Publishing Company. Fine, K. (1995). “Ontological Dependence.” Proceedings of the Aristotelian Society 95: 269–290. Fine, K. (2005). ‘Necessity and Non-existence’ (ed.), Modality and Tense: Philosophical Papers. New York: Oxford University Press. Fine, K. (2015). “Unified Foundations for Essence and Ground.” Journal of the American Philosophical Association 1(2): 296–311. Glasgow, J. (2009). A Theory of Race. New York: Routledge. Griffith, A. M. (2018a). “Social Construction and Grounding.” Philosophy and Phenomenological Research 97(2): 393–409. Griffith, A. M. (2018b). “Social Construction: Big-G Grounding, Small-g Realization.” Philosophical Studies 175(1): 241–260. Hacking, I. (1999). The Social Construction of What? Cambridge: Harvard University Press. Hacking, I. (1995). “The Looping Effects of Human Kinds,” in Causal Cognition, edited by D. Sperber, D. Premack, and A. J. Permack. Oxford: Clarendon Press. Haslanger, S. (2000). “Gender and Race: (What) Are They? (What) Do We Want Them to Be?” Noûs 34: 31–55. Haslanger, S. (2003). “Social Construction: The ‘Debunking’ Project,” in Socializing Metaphysics, edited by F. Schmitt, 301–325. Lanham, MD: Rowan & Littlefield. Haslanger, S. (2012). Resisting Reality. New York: Oxford University Press. Haslanger, Sally (2013). Race, Intersectionality, and Method: A Reply to Critics. Philosophical Studies 171: 109–119. Heyes, C. (2000). Line Drawings. Ithaca & London: Cornell University Press. Hindriks, F. (2006). “Acceptance-Dependence: A Social Kind of Response Dependence.” Pacific Philosophical Quarterly 87: 481–498. Jenkins, K. (2016). “Amelioration and Inclusion: Gender Identity and the Concept of Woman.” Ethics 126(2): 394–421. Kimmel, M. (2000). The Gendered Society. New York: Oxford University Press. Khalidi, M. (2013). Natural Categories and Human Kinds: Classification in the Natural and Social Sciences. Cambridge: Cambridge University Press. Khalidi, M. (2015). “Three Kinds of Social Kinds.” Philosophy and Phenomenological Research 90(1): 96–112. Koslicki, K. (2012). “Varieties of Ontological Dependence,” in Metaphysical Grounding, edited by F. Corriea, and B. Schnieder, 186–213. Cambridge: Cambridge University Press. Latour, B. and Woolgar, S. (1979). Laboratory Life. Beverly Hills CA: Sage. Liao, S. and Huebner, B. (2021). “Oppressive Things.” Philosophy and Phenomenological Research 103(1): 92–113. Locke, J. (1975). An Essay Concerning Human Understanding, P. H. Nidditch (ed.). Oxford: Clarendon Press.

470

Social Construction MacKinnon, C. (1989). Toward a Feminist Theory of the State. Cambridge, MA: Harvard University Press. Mallon, R. (2007). “Human Categories Beyond Non-Essentialism.” The Journal of Political Philosophy 15(2): 146–168. Mallon, R. (2016). The Construction of Human Kinds. Oxford: Oxford University Press. Mallon, R. (2018). Constructing Race: Racialization, Causal Effects, or Both? Philosophical Studies 175: 1039–1056. Mallon, R. (2019). “Naturalized Approaches to Social Construction.” Stanford Encyclopedia of Philosophy, edited by E. Zalta. http://plato.stanford.edu/archives/win2013/entries/socialconstruction-naturalistic/ Marques, T. (2017). “The Relevance of Causal Social Construction.” Journal of Social Ontology 3(1): 1–25. Mason, R. (2016). “The Metaphysics of Social Kinds.” Philosophy Compass 11: 841–850. Mason, R. (2020). “Against Social Kind Anti-Realism.” Metaphysics 3(1): 55–67. Mason, R. (2021). “Social Kinds are Essentially Mind-Dependent.” Philosophical Studies 178: 3975–3994. Mikkola, M. (2017). “Feminist Perspectives on Sex and Gender.” The Stanford Encyclopedia of Philosophy (Fall 2017 Edition), edited by E. N. Zalta, forthcoming < https://plato.stanford.edu/ archives/fall2017/entries/feminism-gender/>. Millett, K. (1971). Sexual Politics. London: Granada Publishing Ltd. Oderberg, D. (2007). Real Essentialism. New York: Routledge. Pagano, E. (2021). “What Social Construction Isn’t.” Philosophia 49: 1651–1670. Passinsky, A. (2019). “Finean Feminist Metaphysics.” Inquiry 64(9): 937–954. Passinsky, A. (2020). “Social Objects, Response-Dependence, and Realism.” Journal of the American Philosophical Association 6(4): 431–443. Passinsky, A. (manuscript). “Social Essentialism.” Pettit, P. (1991). “Realism and Response Dependence.” Mind 100(4): 587–626. Raven, M. (2022). “A Puzzle for Social Essences.” Journal of the American Philosophical Association 8(1): 128–148. Rosen, G. (2015). “Real Definition.” Analytic Philosophy 56(4): 189–209. Schaffer, J. (2017). “Social Construction as Grounding; Or: Fundamentality for Feminists, a Reply to Barnes and Mikkola.” Philosophical Studies 174(10): 2449–2465. Searle, J. (2010). Making the Social World. New York: Oxford University Press. Spelman, E. (1998). Inessential Woman. Boston: Beacon Press. Stoljar, N. (1995). “Essence, Identity, and the Concept of Woman.” Philosophical Topics 23(2): 261–329. Stone, A. (2004). “Essentialism and Anti-Essentialism in Feminist Philosophy.” Journal of Moral Philosophy 1(2): 135–153. Stubblefield, A. (1995). “Racial Identity and Non-essentialism About Race.” Social Theory and Practice 21(3): 341–368. Sundstrom, R. (2003). “Race and Place: Social Space in the Production of Human Kinds.” Philosophy and Geography 6(1): 83–95. Thomasson, A. L. (2003). Foundations for a Social Ontology. ProtoSociology 18: 269–290. Taylor, P. (2013). Race: A Philosophical Introduction. Malden, MA: Polity Press. Witt, C. (1995). “Anti-Essentialism in Feminist Theory.” Philosophical Topics 23(2): 321–344. Witt, C. (2011). The Metaphysics of Gender. New York: Oxford University Press. Zack, N. (2005). Inclusive Feminism. Lanham, MD: Rowman and Littlefield.

471

32 CONFERRALISM Anand Jayprakash Vaidya and Michael Wallner

This article is about Ásta’s conferralist account of essence, which she has developed in a series of papers (2008, 2013). Conferralism provides an anti-realist account of essence. In order to understand conferralism, it is important to first understand the difference between realist and anti-realist accounts of essence. In this article, we cast the difference between realism and anti-realism as a difference over how an in-virtue-of-question is answered. The question is: why (or in virtue of what) does an essentialist fact hold? Anti-realist positions, such as conventionalism (see Livingston-Banks & Sidelle this volume) and conferralism take the essentialist fact to hold in virtue of our conceptual commitments and practices (Section 32.1).1 In Section 32.2, we present Ásta’s notion of a conferred property by introducing some examples that have become seminal in the debate. Section 32.3 is where we sketch Ásta’s account, which we subject to critical inquiry in Section 32.4.

32.1

Realism vs. Anti-realism about Essence

Ásta (2008: 136) makes it clear that her conferralist account of essence is based on … the idea that an object’s having the essential properties that it has is a reflection of our values and interest, as expressed in our conceptual practices, as opposed to essentiality residing in independent reality. (Ásta 2008: 136–7) It seems obvious that this makes the view anti-realist. But what does this mean exactly? Realism about some x is standardly taken to contain at least two claims: an existence claim—x exists—and an independence claim—x exists independently (from us). (See, e.g., Miller 2021.) Accordingly, there are at least two ways of being an anti-realist: one can deny the existence claim and one can deny the independence claim. Since Ásta’s characterization of the guiding idea behind her view acknowledges essential properties, her view does not seem to deny the existence claim concerning essence. So, it is not a form of what could be called the “nihilistic” brand of anti-realism. It is indeed the independence claim that is rejected, as the latter part of the characterization of the guiding idea makes clear. Ásta does not think that essentiality resides in independent reality but that it is a “reflection of our

472

DOI: 10.4324/9781003008750-37

Conferralism

values and interests, as expressed in our conceptual practices”. To see what this means exactly, it is best to take an example. Think of a triangle T. Suppose that among the many properties T has, there are the following two: (a) it is red; (b) it has three sides. Orthodoxy has it that (b) is an essential property of T, while (a) is a mere accidental property. Why? What makes (b) essential for T? Why is (b)—as opposed to (a)—essential to T? Ásta’s brand of anti-realism can be read as answering this question in the following way: because of our values and interests, as expressed in our conceptual practices. “[A] property is essential to an object because we value certain things over others, not the other way around” (Ásta 2008: 139). We need to unpack two things to clarify that statement and to get a better grasp on Ásta’s brand of anti-realism: first, how, precisely, can we understand the value that is assigned to some but not to other properties; second, what is the contrast class that Ásta is alluding to by using the words “not the other way around”? Let’s start with the first point. Ásta is rather elusive about what “our values and interest, as expressed in our conceptual practices” actually refers to. If we stick to our example and to orthodoxy, then (b), i.e., the property of having three sides, is essential to T while (a), i.e., the property of being red, is merely accidental to T. Now, according to Ásta, this difference is supposed to lie in “our values and interest, as expressed in our conceptual practices”. What is it about the property of having three sides that we value more or that interests us more than the property of being red? This, we think, can only be answered if we relate this to the object, T, i.e., the bearer of those properties. (b), as opposed to (a) is what makes the object the (kind of) object it is, what defines the object. Being three-sided, as opposed to being red, is definitional of what T is. Moreover, while T can stop being red, and still remain the (kind of) object it is, this does not seem to be the case for the property of having three sides. So, we think that the different value we assign to or the difference in interest we show with regard to different properties of specific (kind of) objects has to do with what could be called the definitional and modal importance of that property for the specific (kind of) object. Depending on whether you are working with a so-called “modal” or a so-called “non-modal” account of essence, you will assign different priority relations between the modal and the definitional features of a property. See Corriea this volume and Torza this volume. Moreover, Ásta does not think that (b) is modally and/or definitionally important for T because of some objective fact residing in independent reality. Rather, she thinks of our values and interests as being “expressed in our conceptual practices”. This brings us to the second issue to be unpacked. The relevant contrast class in her claim: “[A] property is essential to an object because we value certain things over others, not the other way around” (Ásta 2008: 139; emphasis added). We can take Ásta’s anti-realist position as a specific answer to the following Euthyphro style question: (E) Do we (1) value (or are we especially interested in) property P of object o because P is essential to o or (2) is P essential to o because we value (or we are especially interested in) property P in relation to o? It is difficult to understand these vague and under-defined terms, “value” and “interest”, especially with respect to the key roles they are supposed to play. As a consequence, we can formulate an analogous question replacing those terms with what we can call the modal and definitional importance. (E*) Do we (1) assign modal and/or definitional importance to property P with regard to object o because P is essential to o or (2) is P essential to o because we assign modal and/or definitional importance to P with regard to o?

473

Anand Jayprakash Vaidya and Michael Wallner

Ásta’s anti-realist position goes for the second option, (2). Ásta’s position is an answer to the question as to what accounts for the difference between essential and accidental properties of some object o. Her answer is that rather than modal and definitional importance lying in some objective fact residing in independent reality, the ground of this difference lies in our values and interests that are expressed by our conceptual practices. (According to our interpretation of this somewhat enigmatic sentence, it lies in the fact that our conceptual practices assign modal or definitional importance to P with regard to o.) In contrast to that, the first option, (1), is more common with realists about essence. The realist thinks that what accounts for the difference between essential and accidental properties of some object o is indeed an independent worldly fact about o.2 In summation, we can pinpoint the difference between Ásta’s brand of anti-realism about essence and some realist positions about essence in their respective answer to the following in-virtue-of-question: (V) In virtue of what is a property essential to an object? Ásta’s anti-realist answer: in virtue of our values and interest, as expressed in our conceptual practices Our best interpretation of Ásta’s anti-realist answer: in virtue of our conceptual practices, which assign modal and definitional importance to properties with regard to objects But what is the realist’s answer concerning the source of essentiality? According to Ásta the realist holds that essentility lies “in the nature of things, as it is independent of human thought and practices” (Ásta 2008, 136, emphasis added).3 It should be noted that this answer is not very substantive, given that “essence” and “nature” are often used interchangeably. So, we might want to ask whether there is a more interesting question. (V*) In virtue of what does an essentialist fact of the form “it is essential to x that p” hold? While realists can answer (V) by just referring to the essence or nature of x, (V*) asks for a more substantive explanation, source, or ground of essentiality. With regard to (V*), many realists about essence hold that there is (in some sense or other) no explanation, ground or source of essence, hence, taking (V*) to be unanswerable in a sense. Examples include Dasgupta (2014, 2016), Glazier (2017), Wallner & Vaidya (2020), and Wallner (2020). Dasgupta (2014, 2016), e.g., finds it plausible that (facts about) essences do not have a ground, a view that can also be found in Aristotle’s Posterior Analytics, for they are not the kind of things that can be grounded. Dasgupta likens essentialist facts to axioms in mathematics. Just like it does not make sense to ask for a proof of axioms, it does not make sense to ask for a ground of essentialist facts. The latter are what Dasgupta calls “autonomous facts”. Glazier (2017) argues that essences figure in specific sui generis metaphysical explanations, called “essentialist explanations” as explanantia. E.g., the fact that {Socrates} contains Socrates as a member can be explained by the fact that it is essential to {Socrates} to contain Socrates as a member. However, while Glazier leaves it open whether essences have some kind of explanation, he maintains that essences do not themselves have essentialist explanations. More precisely, for every essentialist fact f, there is no object o, such that o is essentially such that f holds. Wallner & Vaidya (2020) and Wallner (2020) argue that while there is an explanation of the necessity of essentialist facts or propositions, essentialist claims themselves cannot be explained in the same way, thereby taking essences to be “buck-stoppers”.4

474

Conferralism

While these broadly realist thinkers discuss reasons to believe that (V*) is unanswerable, some anti-realists about essence provide a substantive answer to (V*). On both, conventionalist (see Livingstone-Banks & Sidelle this volume) and conferralist accounts of essence, a property is essential to an object in virtue of some conceptual practices and commitments. The precise difference between conventionalism and conferralism lies in how the details are cashed out. More on this below.5 For now it is important that realists and anti-realists about essence disagree on how the Euthyphro-style question (E) from above should be answered. Realists about essence tend to pick option (1): the reason why P seems so important for o is that P is essential to o. Anti-realists about essence, like conferralists and conventionalists, pick option (2): the reason why P is essential to o is that our conceptual commitments and practices assign special value to P with regard to o.

32.2 What Is a Conferred Property? At the heart of Ásta’s conferralism lies the notion of a conferred property. In this section, we briefly introduce this notion by using Ásta’s own examples that have become seminal in the debate. Consider Ásta’s definition of a conferred property: I call a property of an object ‘conferred’ if it is in virtue of some attitude of subjects that the object has the property. We can say in that case that the attitudes of the subjects confer the property on the object. (Ásta 2008: 136–7) Now take the following example: a baseball pitcher’s pitch possessing the property of being a strike in baseball. In baseball the umpire’s verdict about whether or not a pitch is a strike is final.6 The umpire’s verdict is to track the physical fact whether the ball traveled through the strike zone. However, even if it didn’t, if the umpire calls a strike, the pitch is a strike, even though everybody else agrees that it should not have been one. So, there is a baseball-property of being a strike and that property is conferred by the umpire. However, there is also a physical property of having traveled through the strike zone, which, ideally, the umpire’s verdict should track. This physical property, however, is not conferred. Consider a different example: an act’s being pious. In Plato’s (1984) dialogue Euthyphro, Socrates and Euthyphro debate whether an act is pious because it is loved by the gods (Euthyphro’s position) or whether the gods love an act because it is pious (Socrates’ position). In Euthyphro’s position, being pious is a conferred property. It is conferred on to the act by the gods’ love. Every conferred property needs a subject (or group of subjects) who does (do) the conferring. It can also be specified what attitudes or states of mind of the subject(s) matter in the conferral. And, finally, it can be specified under what conditions the property is conferred. Ásta schematizes these specifications as follows: Property: Who: What: When:

what property is conferred, e.g., being pious who the subjects are, e.g., the Greek gods what attitude, state, or action of the subjects matter, e.g., their love under what conditions the conferral takes place, e.g., normal, ideal, or some specified conditions (Ásta 2008: 139)

475

Anand Jayprakash Vaidya and Michael Wallner

With this schema in mind, we can move on to the next section where we explain what it means for the property of essentiality to be conferred.

32.3

Conferralism about Essence

In this section, we sketch Ásta’s conferralist account of essence by specifying what it means for the property of essentiality to be conferred. Let’s begin by looking at the schema as applied to essentiality. Property: essentiality (being an essential property of a particular object) Who: ideal subjects, i.e., ideal versions of us concept users What: their finding it inconceivable that the object not have the property When: at the limit of enquiry into how we use concepts (Ásta 2008: 140) We will discuss each of these points in turn.

32.3.1

Property: Essentiality

It is important to be clear about what the conferred property is. Suppose it is essential to Franz Kafka that he is human. Being human, then, is an essential property of Kafka. Now, conferralism about essence is not the view that the (essential) property of being human is conferred on to Kafka. Rather, it is the view that the property of essentiality is conferred on to the property of being human (that Kafka possesses). Call a property of an object that is not itself a property a first-order property and a property of a property a second-order property. So, what is conferred is not the first-order property of being human on to Kafka, but the second-order property of being essential on to Kafka’s property of being human. But this latter sentence is still ambiguous, in as much as it is not entirely clear about what exactly the second-order property of essentiality is being conferred on to. Does Ásta mean that the (second-order) property of being essential is conferred on to the property of being human itself or on to its instantiation by Kafka? In order to get a better grip on this question and on what is at stake, we have to take note that it does not make sense to speak of a property as being essential, period. A property P is essential (or non-essential) for a specific object o, which has P. So, essentiality is not just a property of a property, but a property of a property that is had by an object. This is in line with Ásta’s characterization of essentiality as “the property of being an essential property of an object” (Ásta 2008: 136; emphasis added). So, what is conferred here is the second-order property of essentiality onto Kafka’s being human, i.e., on to the property of being human of Kafka, in other words, onto Kafka’s instantiation of that property. This is also the place to draw a comparison between the conferralist scheme as Ásta uses it in her account of essence and her later development of the conferralist scheme in her Categories We Live By (Ásta 2018). In her later work, Ásta uses conferralism to provide an account of our social categories such as sex, gender, race, disability, religion, and LGBTQ categories. (See also Rosario this volume, Griffith this volume, and Passinsky this volume.) Strikingly, in this later work, the conferralist schema, contains a fifth point, besides the Conferred property, the Who, the What and the When: the so-called Base property. Using the baseball example again, Ásta complements the conferralist scheme by adding the following: Base property: what the subjects are attempting to track (consciously or not), if anything; the physical trajectory of the ball (Ásta 2018: 8)7

476

Conferralism

There are or might be more or less clear answers to the question as to what the base property of conferred properties like being a strike, or sex, gender, or race is. In the case of the property of being a strike, as we have seen, it is the physical trajectory of the ball. The base properties of social categories like the above might have to do with either some phenotypical appearances or with social roles that subjects are perceived to have. The question that arises here is whether there are substantial reasons for Ásta not to include a base property in the conferralist scheme in her account of essence. It is important to note that Ásta (2018: 8) adds the “if anything” to the characterization of the base property as “what the subjects are attempting to track (consciously or not), if anything”. We think that there are some conferred properties that do track a base property and some that just don’t. Suppose, one is siding with Euthyphro, taking the piety of an act to be a property conferred by the gods. In this case, the gods’ love which is supposed to confer moral properties on to the act does not attempt to track anything at all. Either the gods love an act or they don’t. If they would love it for a specific reason, e.g., the fact that it maximizes happiness, it would be the latter that would function as the ground for the moral property and not the gods’ love. If they would love it for its piety, being pious would not be a conferred property at all. So it seems possible to have a conferred property without a base property that is supposed to be tracked. Consider another example. Being money is plausibly a conferred property of objects like coins and bills. In contrast to the pitch’s property of being a strike, it does not seem as if there would be anything that the conferred property of being money would track. It certainly is not value, since money is not money because it is valuable. Quite the opposite is the case: money is valuable because it is money—because someone has decreed it to be valuable, thereby making it money. The conferred property of being money also does not seem to track properties like being scarce or hard to get. The production of bills and coins is actually so easy that we have to regulate it. And, again, money is not money because it is hard to get. It is rather hard to get because it is money. These examples make it plausible that there are conferred properties without a base property that is supposed to be tracked. The crucial question here however seems to be whether essentiality is such a property. Is there a base property that is supposed to be tracked in the conferral of essentiality? It is very difficult to answer this question. But why is it so difficult? Why isn’t being human the base property of essentiality in this case, one might ask. Note that the base property is supposed to be the property that the conferring subjects “are attempting to track” (Ásta 2018: 8). Being human here is rather the property (of the specific object in question) upon which essentiality is conferred.8 Here is a thought about a possible base property of essentiality: It seems that if the conferred property is to track the base property, the minimum requirement for being a base property is that it is a property of the same object (i.e., had by the same object) as the conferred property. In case of essentiality, the object of this conferred property is itself a property. So, it seems plausible that whatever the base property of essentiality is, it will also be a property of a property, hence, a second-order property. But which second-order property could fit the bill? Coming back to our Kafka-example: what (if any) property of Kafka’s property of being human is essentiality supposed to track? Maybe it is the property of being indispensible or definitional? Being human is indispensable and/or definitional for Kafka. Note however, that this definitional and/or modal importance is not rooted in worldly facts for Ásta, but rather in our conceptual commitments. So maybe the base property that the conferral of essentiality is supposed to track is related to our conceptual practices. This would also square nicely with Ásta’s brand of anti-realism. Settling this matter, however, is beyond

477

Anand Jayprakash Vaidya and Michael Wallner

the scope of this paper. We submit a determination of the base property of essentiality viz. the question whether essentiality has a base property at all to further research.

32.3.2

Who: Ideal Subjects, i.e., Ideal Versions of Us Concept Users

This point should inform us about who is actually doing the conferring. However, as opposed to straight-forward cases like the baseball example, where the umpire (or the referee team) is doing the conferring, in the case of essentiality we have what Ásta calls “ideal subjects”. By “ideal subjects” Ásta refers to idealizations, i.e., ideal versions of us concept users that are “smarter, less forgetful, better at rational deliberation, and maximally knowledgeable about how we use the concept under consideration” (Ásta 2008: 141). However, can we really speak of such idealizations actually conferring a property? How literal are we to take that proposal? It seems that for these idealizations to actually confer a property, they would have to exist. However, even on realist accounts of abstracta and idealizations on which ideal subjects do in fact exist, their existence does not seem to be such that they could, in the literal sense, actually confer a property. Ásta suggests that the introduction of ideal subjects is rather to be taken as a metaphor or heuristic “spelling out what it is for our conceptual commitments to do the conferring” (Ásta 2008: 141). Ásta (2008: 140) speaks of the ideal subjects’ act of conferring as a hypothetical act: “Pedro’s being human is essential because ideal versions of us would find it inconceivable that Pedro not be human”. So, on this view, where a hypothetical act confers essentiality, ideal subjects do not need to exist, strictly speaking. Given the heuristic function ideal subjects fulfill, Ásta’s claim that essentiality is conferred in such a hypothetical act can be read as the claim that the essentiality of Pedro being human is metaphysically explained by (or grounded in) our conceptual commitment.9,10

32.3.3

What: Their Finding It Inconceivable that the Object not Have the Property

What precisely are the attitudes, states, or actions of the ideal subjects that matter in their conferral of essentiality? Admittedly, given the heuristic and metaphoric role those ideal subjects seem to play, asking about their concrete attitudes and states seems a bit odd. However, there is a clear answer that pertains to the idea that essentiality is conferred in a “hypothetical act”. To repeat Ásta’s (2008: 140) quote: “Pedro’s being human is essential because ideal versions of us would find it inconceivable that Pedro not be human”. So, it is the ideal subjects’ finding it inconceivable that the object, o, not have the property, P, that confers P onto the object. Yet, what does it mean that ideal subjects find o without P inconceivable? Given that ideal subjects are defined as ideal versions of us concept users that are “maximally knowledgeable about how we use the concept under consideration” (Ásta 2008: 141), this ideal inconceivability of o without P just amounts to the conceptual impossibility of o without P. Again, Ásta’s account seems to boil down to the idea that essentiality is metaphysically explained by (or grounded in) conceptual commitments. We will be discussing the relation between conceptual commitments, ideal subjects, and conferralism more when critically assessing Ásta’s account in 32.4.1.

32.3.4

When: At the Limit of Enquiry into How We Use Concepts

We have already seen that in order to get a clear grasp on the Who of the conferralist schema concerning essentiality, we need to also understand the What: it is their finding

478

Conferralism

it inconceivable that o is not P that makes the ideal subjects confer essentiality onto o’s property P. Similarly, it is obvious that to fully grasp the notion of “ideal subjects” in this connection, we also need to understand this When condition. Ásta’s idea is that “the property of being essential to an object is conferred by the ideal representatives of us concept users at the ideal limit of a procedure of correcting for cognitive limitations” (Ásta 2008: 140; emphasis added). This When condition (i.e., the italicized part of the quote), i.e., the fact that conferral takes place at the limit of enquiry into how we use concepts, just expresses the idea that the conferring ideal subjects are maximally knowledgeable, infallible even, about our conceptual commitments. In other words, the When condition simply specifies the notion of ideality that is at play in the Who condition of the scheme. The most important thing to note here is that those idealized subjects are not ideal with regard to their infallible sensitivity concerning the objective fabric of the world but with regard to their omniscience concerning our conceptual commitments. If it were the former, conferralism would, arguably, be more realist than anti-realist. With these specifications of Ásta’s account in the background, we are now ready to move on to a critical appraisal of conferralist essentialism.

32.4

Critical Questions

The following critical questions serve a dual purpose. Thinking through them should, on the one hand, make the reader more aware of the systematic problems and open questions the conferralist about essence faces; on the other hand, it should also provide the reader with a deeper and more detailed understanding of the view itself.

32.4.1 Who or What Really Does the Conferring in (Constructivist) Conferralism? We have seen that in order to make sense of what it means for ideal subjects to confer essentiality, Ásta assigns this notion a heuristical and metaphorical role, that is supposed to express the idea that essentiality is actually conferred by (or explained by, grounded in) our conceptual commitments. The ideal subjects serve as a construction, so to speak. Accordingly, Ásta takes her conferralism to be […] constructivist in that in spelling out what it is for our conceptual commitments to do the conferring we make use of idealizations of us and our epistemic conditions; i.e., a construction, which is to capture our actual conceptual commitments, does the conferring. (Ásta 2008: 141) In this passage Ásta says that our conceptual commitments do the conferring. Yet, she also says that one way to spell this fact out is to claim that a construction, i.e., ideal subjects, do (es) the conferring. (For a discussion of social construction, see Griffith this volume.) So we have two claims: (i) Our conceptual commitments confer essentiality. (ii) Ideal subjects confer essentiality. (ii) is supposed to be a way to express (i). But we have already seen in 32.3.2 that it is difficult to see how ideal subjects could actually confer a property. They either do not

479

Anand Jayprakash Vaidya and Michael Wallner

actually exist, or they exist in a way that makes it mysterious how conferring would be something that they could actually do.11 Ásta is aware of that, which is why she thinks there is not an actual but a hypothetical act involved. The fact that essentiality is conferred upon property P of object o just means that ideal subjects would not find o without P conceivable. Now, it is the ideal subjects’ hypothetical inconceivability-judgment that actually confers essentiality. After all we want to say that P is actually essential to o. Ásta does not find it particularly mysterious that hypothetical acts can actually confer a property. As a precedence, she cites Hume’s account of aesthetic properties: o is beautiful because experts would find it pleasing. (Ásta 2008: 143) One worry that might arise from the analogy between conferralism about essence and Hume’s “conferralism” about aesthetic properties is the following: one might ask whether the conferralist idea is more suited for aesthetic properties than essence. In addition, and more importantly, a conferralist (or, generally, an anti-realist) story might be more suited for some essences (such as some social kinds) (see also Passinsky this volume) than for other objects (such as natural kinds) (see also Tahko this volume). One need not think that there is even a hard boundary between these kinds, only that in some cases our conceptual commitments are regulated by factors external to us, while in other cases our conceptual commitments are all there is to say on the matter. In sum, a worry one might have is that conferralism itself is more suitable to some domains than to others when it is understood as a theory where only conceptual commitments do the work. Putting this big-picture question aside, here is a more systematic worry for this brand of constructivist conferralism, coming from Ásta’s commitment to both (i) and (ii). (ii) is taken to be a heuristic, a way to spell out (i). However, even if we grant that hypothetical acts of ideal subjects can indeed do the actual conferring of essentiality, i.e., even if we grant that (ii) is not problematic, (ii) is taken to be just a heuristic and a way to spell out (i), and (i) might turn out to be particularly worrisome. One might find it particularly difficult to conceive of our conceptual commitments actually conferring a property if those commitments are, as some of Ásta’s (2008: 139) remarks suggest, encoded in dispositional or counterfactual facts. Can dispositional or counterfactual facts (or commitments as such) confer properties? Given that conferring was introduced as something done by subjects, how exactly facts/commitments can confer properties needs further explanation. So, either the conferralist provides us with a story about how facts/commitments can confer properties, or she fully embraces the idea that hypothetical actions of ideal subjects are not only a heuristic but the precise story of how conceptual commitments are encoded. Summing up, there seems to be somewhat of a dilemma for the constructivist conferralist about essence: (D1) Either, first horn, conferralism leans into the fact that ideal subjects are just a heuristic and a metaphor, in which case it needs to tell a story about how conceptual commitments are “officially” encoded and of how conceptual commitments can confer the property of essentiality (since there does not seem to be any kind of subject involved in the official story). (D2) Or, second horn, conferralism takes those ideal subjects more seriously and makes it the “official” way in which conceptual commitments are encoded and not just a heuristic. This would then allow for a straight-forward account of conferring essentiality, because now there are officially subjects involved. But it would also place the burden on conferralism to explain how exactly the commitments can be encoded using ideal subjects. How exactly is that done? So, on either of those horns, there is more explanatory work to be done by the conferralist.

480

Conferralism

32.4.2

What Sets Conferralism Apart from Conventionalism?

Due to space-limitations we can only sketch Ásta’s (2008: 144–8) detailed answer to this question. The most important distinction is that, according to conventionalism, it is something about us, our conventions, that actually confer(s) essentiality, while, according to (constructivist) conferralism, a construction, i.e., a hypothetical act of ideal subjects confers essentiality. (See also Sidelle and Livingstone-Banks this volume.) Ásta takes this difference to yield a crucial advantage of conferralism over conventionalism, which she conveys by the following thought experiment: Suppose, at time t, before the dawn of conferring subjects, there lived a dinosaur, Dino. Arguably, Dino was essentially a dinosaur. On conventionalism, essentiality is grounded in actual conventions. Since, at t, there were no subjects, hence no conventions, the fact that Dino was essentially a dinosaur lacks its metaphysical source.12 Not so on (constructivist) conferralism, for, since a construction does the conferral, “there is no point in time at which the conferral takes place” (Ásta 2008: 145). Such a strategy, of course, raises questions about the details of how hypothetical acts are able to actually confer properties (see 32.4.1).

32.4.3

On Conferralism, Can Essence Ground Modality?

Fine (1994) and others popularized the view in modal metaphysics that essence is the metaphysical ground, source, or metaphysical explanation of modality (instead of the other way around). Given the popularity of this view, one might ask whether it is available to the conferralist about essence. We want to make two points with respect to this question. (We shall use the grounding idiom to make those points.) First, it is plausible that, on conferralism, essence is grounded in conceptual commitments. Combining this with the view that modality is grounded in essences,13 per transitivity of grounding, we get that modality is grounded in conceptual commitments, thus yielding a plausibly anti-realist view of (metaphysical) modality. Second, there might be a principled reason why essence cannot ground modality on conferralism. Again, it is plausible that, on conferralism, essence is grounded in conceptual commitments. Ásta (2008: 139) sometimes suggests that those conceptual commitments are encoded in dispositional or counterfactual facts. If we take those to be modal facts, it seems that, by transitivity of grounding, essence-facts are grounded in modal facts, and, on pain of a violation of the asymmetry of ground, cannot themselves ground modality.

32.4.4

Does Conferralist Anti-Realism about Essence Entail Relativism?

On conferralism, essentiality is conferred by (explained by, grounded in) our conceptual commitments. In other words, essentiality is conferred upon properties by a hypothetical act of ideal versions or us concept users. Now, there might be a different group of concept users with conceptual commitments different from us. Ideal versions of them might not confer essentiality upon, say Kafka’s property of being human. Thus, whether or not a property is essential to an object seems to be relative to the community of concept users that one finds themselves in.14 It seems conferralism entails some kind of relativism. Of course, some antirealists about a domain D are happy to buy into the package deal with relativism about D, but it is neither clear whether those two need to go together, nor whether one having motivation for the former also has motivation for the latter.

481

Anand Jayprakash Vaidya and Michael Wallner

32.4.5

How (Anti-)Realist Does (Anti-)Realism Need to Be?

We have treated conferralism about essence as an anti-realist account of essence precisely because it holds that essentiality is conferred by (explained by, grounded in) our conceptual commitments. However, we might want to ask where our conceptual commitments are coming from. Is this just a matter of conventions? Even if it is the case, we can ask whether our conceptual conventions and commitments are influenced by the world, i.e., by the fabric of reality. In Section 32.1, we have seen that conferralism answers the question as to why we take P to be essential to o by appealing to our conceptual commitments. But we can ask further why we have the conceptual commitments we actually have. If our conceptual commitments are partially grounded in the fabric of reality, conferralism might not be as anti-realist as one initially thought. In general, we can ask whether the opposition between realism and anti-realism is strictly dual or whether there is room for a continuum of more or less realist and/or more or less antirealist views. Realism about some domain of entities, D, has traditionally been understood as something like a purity thesis, where D is pure and uncontaminated by human thought, meaning that the existence and reality of (the entities in) D is in no way dependent on human thought.15 Anti-realism, on the other hand, seems to allow for a continuum of more or less (anti-)realist views, depending on the respective contribution of the world and the mind concerning the grounds of D. (For more on issues concerning mind-(in-)dependence, see Griffith this volume.) It is an interesting question as to whether this continuum view of anti-realism also suggests that realism comes in degrees or kinds or whether the label “realism” should be reserved exclusively for the extreme pole of this spectrum. Consider the following view: The source of essentiality lies both in human interest and practices, and the world. Should this kind of view be regarded as anti-realist, for its “contamination” by human thought, or might a small enough contamination still warrant the classification “realist”?

32.4.6

What Is the Motivation for Conferralism?

It is safe to say that there are two purported motivations, one ontological, one epistemological: On the ontological side, the conferralist might argue that her view has less ontological commitment than realism about essence (cf. Ásta 2013: 22), taking this at least as a motivation for anti-realism about essence.16 On the epistemological side, the conferralist argues that it is easier to give an account of the epistemology of essence than on any realist account. Ásta issues the following prima facie challenge for any realist account of the epistemology of essence. (For a sketch of some contemporary accounts of the epistemology of essence, see Mallozzi this volume.) If essentiality, being the peculiar property it is, is real, and has nothing to do with us or our conceptual powers, how is exercising those powers in thought experiments or ordinary discourse to be a justified method of gaining knowledge of the essences of things? […] If essentiality were somehow dependent or linked to our conceptual powers, then perhaps we could acknowledge that exercising those powers in thought experiments could give us knowledge. But no such story is available to the realist. (Ásta 2013: 23–4)

482

Conferralism

Since conferralism grounds essentiality in our conceptual commitments and practices, Ásta takes her view to have the upper hand, compared to realism about essence, when we are discussing the epistemological story. Ásta (2013: 24–5) briefly considers possible responses to the prima facie challenge on behalf of the realist. The realist might, e.g., resort to intuitions about essence or to science for their epistemological story of essence. However, Ásta maintains that even if there might be some such answer to the prima facie challenge available for the realist, the conferralist account still has the edge over the realist. This is because she thinks that the conferralist account of essence can best make sense of our practices of using thought experiments in our pursuit of knowledge of essence. Conferralists argue that when we are conducting thought experiments in order to answer questions like whether a specific table could have had one more leg or whether a specific tiger could have lacked its tail, what we are testing are not our intuitions about an objectivistically construed essence of the tiger or the table but our intuitions about our conceptual commitments (Ásta 2013: 30–1). In this way the conferralist, contrary to the realist, can avoid the prima facie challenge and neatly account for our epistemic practices concerning essence. Note that Ásta (2013: 23) assumes that any epistemology of essence must involve thought experiments. This assumption, however, might be questioned. (See Mallozzi this volume for an overview of some contemporary accounts in the epistemology of essence.) Is it true that employing thought experiments is essential in any pursuit of knowledge of essence? If it is not, the two responses to the prima facie challenge that Ásta herself grants the realist to have at their disposal, i.e., resorting to intuition or to science, suddenly look significantly better. So the question is whether Ásta might have unjustifiably stacked the deck against the realist by conceiving of our knowledge of essence as essentially depending on thought experiments. In addition, there might be a general problem with the epistemological motivation for constructive conferralism. Depending on how serious the appeal to ideal subjects is in constructive conferralism, we can critically ask how this appeal to those ideal subjects really facilitates the epistemology of essence. Going anti-realist, as Ásta claims, has the purported advantage of making the epistemology of essence easier. However, if we are to track the judgements of ideal subjects in order to get to know what they would find (in)conceivable, the worry might be coming from the fact that we will never truly be “ideal” subjects. So, the epistemic gap between us ordinary subjects and real or objectivistically constructed essences, that was supposed to be closed by the anti-realist move, seems to be replaced by a gap between us ordinary subjects and the judgements and competence of ideal subjects.

32.4.7

Can Conceptual Commitments Even Be the Metaphysical Source of Essentiality?

Arguably, conceptual commitments are contingent. There might have been different conceptual commitments and practices than there are actually. So, on conferralism, the metaphysical source of essentiality is contingent. Now, there are, standardly, two views available concerning the relation between essence and modality. So-called modalists (see Torza this volume) collapse essence and (de re) necessity. The Finean essentialists (see Correia this volume) take essence to be different from (de re) necessity. On the latter view it is standardly assumed that essence is the ground, source or metaphysical explanation of modality. Either way, conferralism entails that the metaphysical source of modality is contingent. However, how can something contingent account for or explain necessity? It seems that contingent facts

483

Anand Jayprakash Vaidya and Michael Wallner

lack the required “oomph” to explain why something is necessary. We cannot go into the details of this criticism. For an argument to the effect that the metaphysical source of necessity must itself be necessary, see, e.g., Cameron (2010: 139–40). It should be mentioned, however, that this criticism is not specific to conferralism. If it hits, it equally hits other anti-realist accounts of essence that ground essence (and modality) in something contingent, like, e.g., conventionalism. (See also Sidelle and Livingstone-Banks this volume.)

32.5 Conclusion The goal of this contribution was to outline Ásta’s (constructivist) conferralist account of essence. In Section 32.1, we clarified in what sense the account is to be considered an antirealist one. In Section 32.2, we introduced her notion of a conferred property by using some examples from the literature as well as Ásta’s own schema of how to specify concrete qualifications of a conferred property. In Section 32.3, we discussed how Ásta is applying this schema to essence or essentiality, in order to argue for the fact that essentiality is conferred. Finally, in Section 32.4, we introduced some critical questions whose discussion was supposed to serve a dual purpose: to provide a deeper understanding of the conferralism about essence as well as to indicate some problems and open questions of the view. Some of the most interesting questions, in our opinion, occur with regard to Ásta’s conception of ideal subjects, as well as with regard to the proper understanding of the distinction between realism and antirealism concerning essence.17

32.6

Related Topics

The following chapters in this handbook are related to the issues discussed in the present chapter: Modal Conceptions of Essence Non-Modal Conceptions of Essence Natural Kind Essentialism Epistemology of Essence Artifacts, Artworks, and Other Social Kinds Sex and Gender Conventionalism Social Construction

Notes 1 Note that by casting the difference between realism and anti-realism about essence in these terms, we are disregarding the nihilistic form of anti-realism, which rejects any form of essentialist facts. 2 Note that the characterization of the difference between realism and anti-realism about essence here is by no means meant to exhaustively include all possible realist and anti-realist positions. For one, the nihilist anti-realist about essence denies the existence claim concerning essences or essentialist facts and, consequently does not bother answering a question like (V) or (V*). Neither does the expressivist about essence, who is also standardly considered an anti-realist, since in their view, claims about essence are not fact-stating. However, we find the Euthyphro style questions (E) and (E*) as well as the in-virtue-of-questions (V) and (V*) a good way to characterize the difference between Ásta’s brand of anti-realism and the kind of realism that she is up against. Even with this restriction of scope in mind, however, one might have the following worry about our characterization of this difference between realism and anti-realism about essence:

484

Conferralism

3

4

5 6 7

8

9 10 11

12 13

Questions (E) and (E*), respectively, seem to presuppose that we do in fact value o’s being P. It may, however, be controversial whether we do, and it may be highly contingent that we do. For these reasons, it seems inadvisable to hitch the issue, i.e., the difference between realism and anti-realism, to what we value. In response, we partially reject that characterizing the difference between Ásta’s anti-realism and the realism she is up against presupposes that we actually value o’s being P. Characterizing the kind of realism at play by opting for option (1) on (E) and (E*), respectively, does not entail that the realist needs to actually assign special value or any modal or definitional importance to P (with regard to o). Even if we would not care at all about the value of P (with regard to o), all that a characterization of realism by means of option (1) in (E) or (E*) says, is the following: if we were to care about the special value that is the modal or definitional importance of P (with regard to o) it would be because of P’s being essential to o—and this is totally compatible with us not caring at all. So much for the realist side, i.e., for option (1). It is true, however, that on the anti-realist side, i.e., option (2), where essentiality lies in us valuing P, our valuing P (with regard to o) is required for P being essential (to o). But this, we claim, is the intended outcome, inasmuch as it is the precise point about this brand of anti-realism. According to Ásta’s anti-realist convictions, essentiality is grounded in our conceptual commitment, which expresses our values and interests. So, the contingency of our values and interests seems to carry over to the contingency of essentiality. More on this in Section 32.4.2 and in Section 32.4.7. Whether the realist about essence really needs to hold that essentialist facts or their sources are independent of human thought and practices is especially contentious when it comes to the essences of mind-dependent entities like social kinds. For a discussion see Griffith this volume and Passinski this volume. However, within the realist camp we also find essentialists that might be interpreted as to provide a substantive answer to (V*). Correia & Skiles (2019) argue that the notion of essence can be analyzed in terms of the notion of generalized identity. While this clearly amounts to a view according to which essence is not conceptually primitive, it is unclear whether this implies that essentialist facts are not metaphysically fundamental, i.e., whether there is a substantive answer to (V*). Due to space reasons, we cannot settle this question here. Note that expressivism about essence, which is also considered to be an anti-realist position, does not answer (V*), since claims about essence, on this view, are not fact-stating. In fact, since the video assistant referee has been introduced to baseball, strictly speaking, it is no longer the umpire’s verdict that is final but the referee team’s verdict still is, one might argue. For the sake of simplicity, we will be referring to the umpire’s verdict in the text. Note that it is important for Ásta’s account of social categories or properties like sex, gender, and race, that it does not matter whether those perceived to belong to these categories, i.e., those onto which these social properties are conferred, actually have the respective base property. All that matters is whether the conferrers take them to have it. One might also wonder whether the conferring subjects’ finding it inconceivable that Kafka lacks the property of being human is the base property of essentiality. But, as will become clearer in Section 32.3.3, that is what about the conferring subjects’ attitudes does the conferring, not the base property the conferring is supposed to track. Note that one does not need to commit to the notion of “grounding” (nor to the notion of “metaphysical explanation” for that matter) to make this point. Any suitable (explanatory) metaphysical determination relation will do. We will be coming back to the heuristic function of ideal subjects and the question about who actually does the conferring in Section 32.4.1. Some readers might find our assumption that it might be problematic that ideal (non-actual) subjects can actually confer properties unfair. After all, in debates about morality and rationality (morally or rationally) ideal subjects are regularly appealed to and their non-actuality does not seem to be that much of a problem there. Why is it here? The reason we think this might be especially problematic here is because of the way the conferralist idea is introduced. Conferring is supposed to be an action, done by subjects and it is introduced using the model of actual subjects committing actual conferring actions. For a conventionalist response to this problem, see Sidelle (1989, 54–5). We do not know of any actual proponent of this combination of views in the literature. We just thought that, given the relative popularity of the Finean idea that essences ground necessity, the question about the consistency of such a combination would be interesting.

485

Anand Jayprakash Vaidya and Michael Wallner 14 At this point, one might wonder whether Ásta’s considerations of the ideal conceptual agents might remove the community-relative features. However, this is not the case. The point about Ásta’s notion of ideal subjects is that they are ideal in virtue of being maximally knowledgeable, infallible even, about our conceptual commitments. Since our conceptual commitments are relative to our community (other communities might have different conceptual commitments), considerations of such ideal subjects cannot transcend or remove community-relative features. Thanks to Mike Raven for pushing us on this point. 15 Although this is indeed controversial and hard to apply to mental domains, as Rosen (1994) and others have pointed out, we can take this as an, indeed, coarse-grained sketch of what many people understand under “realism”. Note that we do in fact proceed to cast some doubt on this view. 16 However, it is not entirely clear how precisely Ásta’s point should be understood here. If the point is that (conferralist) anti-realism does not need to reify “spooky” essences, one might argue that many contemporary accounts of essence that have been classified as realist accounts should be able to claim the same advantage for their explicit attempt to avoid such reification. 17 We are grateful to the participants of the Routledge Handbook of Essence Workshop in July 2021 and, especially, to Kathrin Koslicki, Mike Raven and Ásta for insightful comments on earlier drafts of this paper.

References Ásta (2008) “Essentiality Conferred,” Philosophical Studies, 140, 135–148. (published under Ásta Sveinsdóttir). Ásta (2013) “Knowledge of Essence: The Conferralist Story,” Philosophical Studies, 166, 21–32. (published under Ásta Kristjana Sveinsdóttir). Ásta (2018) Categories We Live By: The Construction of Sex, Gender, Race, and Other Social Categories, Oxford: Oxford University Press. Cameron, R. (2010) “On the source of necessity,” in B. Hale & A. Hoffman (eds.), Modality. Metaphysics, Logic, and Epistemology, Oxford: Oxford University Press, 137–151. Correia, F. & Skiles, A. (2019) “Grounding, Essence, and Identity,” Philosophy and Phenomenological Research, 98 (3), 642–670. Dasgupta, S. (2014) “The Possibility of Physicalism,” Journal of Philosophy, 111 (9/10), 557–592. Dasgupta, S. (2016) “Metaphysical Rationalism,” Noûs, 50 (2), 379–418. Fine, K. (1994). “Essence and modality,”, Philosophical Perspectives, 8, 1-16. Glazier, M. (2017) “Essentialist Explanation,” Philosophical Studies, 174, 2871–2889. Miller, A. (Winter 2021 Edition), “Realism”, The Stanford Encyclopedia of Philosophy, E. N. Zalta (eds.), < https://plato.stanford.edu/archives/win2021/entries/realism/>. Plato (1984) Euthyphro, Apology, Crito, Meno, Gorgias, Menexenus. The Dialogues of Plato, Volume 1, translated with comment by R. E. Allen. New Haven and London: Yale University Press. Rosen, G. (1994) “Objectivity and modern idealism: What is the question?” in O’Leary‐Hawthorne, J. & Michaelis, M. (eds.), Philosophy in Mind. The Place of Philosophy in the Study of Mind. Dordrecht: Kluwer Academic Publishers. 277–319. Sidelle, A. (1989) Necessity, Essence, and Individuation: A Defense of Conventionalism, Ithaca and London: Cornell University Press. Wallner, M. & Vaidya, A. J. (2020) “Essence, Explanation, and Modality,” Philosophy, 95 (4), 419–445. Wallner, M. (2020) “The Structure of Essentialist Explanations of Necessity,” Thought: A Journal of Philosophy, 9 (1), 4–13.

486

33 WITTGENSTEIN Arata Hamawaki

In this chapter, I will take up the concept of essence as it figures in Wittgenstein’s Tractatus Logico-Philosophicus and Philosophical Investigations.1 Both works could be said to concern what could be called the possibility of meaning, and in both works it is in articulating the conditions of the possibility of meaning that an appeal to essence plays a pivotal role. Early on in the Tractatus Wittgenstein argues that it is a condition of the possibility of propositions—of saying something, true or false—that there exist simple objects that have certain “internal properties” or essences and that constitute “the substance of the world”, only to claim later in the work that that very metaphysical view is an attempt to say what cannot be said. In the Investigations Wittgenstein criticizes the view that our concepts track essences in reality and urges that essence should be understood as an expression of the “grammar” of our language. While vastly differing in both form and content, both works undertake to criticize the drive to conceptions of meaning that fund the metaphysical appeal to the concept of essence.

33.1 Part I: The Tractatus 33.1.1

“The Substance of the World”

In Philosophical Investigations, Wittgenstein writes, Thought is surrounded by a halo.—Its essence, logic, presents an order, in fact the a priori order of the world: that is, the order of possibilities, which must be common to both world and thought. But this order, it seems, must be utterly simple. It is prior to all experience, must run through all experience; no empirical cloudiness or uncertainty can be allowed to affect it—It must rather be of the purest crystal … We are under the illusion that what is peculiar, profound, essential, in our investigation, resides in trying to grasp the incomparable essence of language (PI, §97). In speaking here of logic as giving us what he calls “an a priori order of the world”, the “order of possibilities” which must be “common to both world and thought”, Wittgenstein is giving expression to the conception of logic that he tried to articulate and clarify in the

DOI: 10.4324/9781003008750-38

487

Arata Hamawaki

Tractatus Logico-Philosophicus. For Wittgenstein logic does not determine whether something is the case, but rather determines what is possible. (T, 1.21) A fact, what is the case, is the obtaining of a possibility. Thus, logic, in some sense, “underlies” facts. As he puts it a fact is in “logical space” (T, 1.31). Wittgenstein conceived of the sense of a proposition as its truth conditions. Thus, he identified what he here calls “the order of possibilities”2 with the sense of a proposition.3 In articulating the “order of possibilities” Wittgenstein saw himself as stating how the world must be if there is to be sense. However, as the Tractatus proceeds Wittgenstein begins to consider the question of the coherence of that very project. That project assumes that there are necessary truths about the constitution of the order of possibilities, that that constitution is something one can describe correctly or incorrectly. But is it coherent to think of the order of possibilities as something we can describe? Is it possible to state how the world must be if there is to be sense? It turns out that answering that question is really the central aim of the Tractatus, for in seeing that that project is not coherent, we will see, Wittgenstein thinks, what it is to say something, or to think something. All this is going to need some explanation. Let us begin with the opening two remarks: “The world is all that is the case. The world is the totality of facts not of things” (T, 1-1.1). Why so? He says: “what is the case—a fact—is the existence of states of affairs. A state of affairs (a state of things) is a combination of objects (things)” (T, 2-2.01). But if a fact is the existence of a state of affairs and a state of affairs is a combination of things, why not say that the world is the totality of things? To answer this question let us consider T, 1.13, which says, “the facts in logical space are the world”. What does he mean by speaking here of “logical space”?4 Let us suppose that “Socrates sits”, “Socrates walks”, and “Socrates talks” are states of affairs, ways that things could be.5 Whether any of these states of affairs exist is contingent, but their possibility, Wittgenstein claims, is not, for “in logic nothing is accidental: if a thing can occur in a state of affairs, the possibility of the state of affairs must be written into the thing itself” (T, 2.012). Whether Socrates sits is a matter of fact, but whether “Socrates sits” is a possible state of affairs is a matter of logic as Wittgenstein understands it. What is necessary for the truth-evaluability of propositions cannot be contingent. Wittgenstein writes, “a speck in the visual field, though it need not be red, must have some colour: it is, so to speak, surrounded by colour-space. Notes must have some pitch, objects of the sense of touch some degree of hardness, and so on” (T, 2.0131). Employing this analogy, we can say that things are necessarily in “logical space”, a “space of possible states of affairs. This space I can imagine empty, but I cannot imagine the thing without the space” (T, 2.013). A thing essentially occurs in the range of those states of affairs in which it “fits”. What is left out, then, in saying that the world consists of things is their fitting one another in states of affairs.6 But, importantly, their fitting one another in states of affairs is not itself a further fact; this was the point of the remark “in logic nothing is accidental”. Wittgenstein says, “objects fit one another like the links of a chain” (T, 2.03). A link of a chain is not conceivable apart from the chain in which it is a link; similarly an object is not conceivable apart from the states of affairs in which it fits. Thus, the world is the totality of facts, not of things.7 I said that for Wittgenstein objects have as their internal properties their fitting into the states of affairs they fit into. What else can we say about them? Wittgenstein claims that there is nothing else that one can say about them: they are simple. We can also see here why Wittgenstein thought that the meaning of logically proper names must be the objects they refer to and so such names must be without sense. What is Wittgenstein’s argument for this? He writes, Objects make up the substance of the world. That is why they cannot be composite. If the world had no substance, then whether a proposition had sense would depend on

488

Wittgenstein

whether another proposition was true. In that case we could not sketch any picture of the world (true or false) (T, 2.021–2.012). The argument seems to be this. First, a proposition represents things in the world as being some way. That presupposes that what it represents as being a certain way exists. Thus, the sense of a proposition would depend on the existence of the object(s) the proposition is about. Second, if the object is complex its existence depends on the existence of its parts. Thus, the sense of a proposition that is about a complex object would depend on the truth of other propositions. Third, but whether a proposition has sense is not a contingent matter and so cannot depend on whether another proposition is true (which Wittgenstein assumes would be a contingent matter). Therefore, the meaning of a (logical proper) name must be a simple object. Once again, the crucial premise is the third, the proposition that whether a proposition is truth evaluable is not a contingent matter: it cannot depend on whether another proposition is true. Put that way, the claim is surely too strong. I think that we should read Wittgenstein as saying that whether a proposition is truth-evaluable can’t in general depend on whether another proposition is true. The argument here seems to be a regress argument. If in order to determine the sense of any proposition we would first need to determine whether certain other propositions are true, we could never get around to comparing our propositions with the world, to “sketching any picture of the world (true or false)”. There must be some propositions whose sense is independent of what happens to be true, that is, guaranteed to have sense in themselves. There must, therefore, be propositions whose names are connected immediately with their objects, without the mediation of a description, and the objects of such names must be simple. Wittgenstein writes, “it is obvious that an imagined world, however different it may be from the real one, must have something – a form – in common with it. Objects are what constitute this unalterable form” (T, 2.022–2.023). Such objects would have to be simple, for their existence could not depend on the existence of states of affairs, but rather conditions the possibility of states of affairs. Thus, it is possible to sketch a picture of the world true or false only if the names in elementary propositions stand immediately for simple objects, or so Wittgenstein argues. A qualification here before moving on. It is important to bear in mind that Wittgenstein is not saying that what we would call ordinary names are not logically complex or that their objects are not complex. Wittgenstein was strongly influenced by Russell’s analysis of proper names as “definite descriptions”. An implication of this analysis is that an ordinary proper name such as “Moses” can be replaced under analysis by a description which is thought to be true only of Moses, such as perhaps, “the prophet who led the Israelites out of Egypt”, or some such. However, he thought—as Russell also thought—that in order to understand how our thought could be about the world at all, there must in our language be names whose meanings are their referents, that is, names that wouldn’t have a meaning at all if their referents did not exist: these would be logically proper names. Like Russell, he thought that a full analysis of propositions that contain definite descriptions would reveal an underlying stratum of elementary propositions consisting of senseless names that stood for simple objects, although unlike Russell, he did not hold that the simple objects are sense-data with which we are immediately acquainted.8

33.1.2

Picturing and the Dissolution of Metaphysics

We can see how the metaphysical account of the possibility of sense just sketched begins to unravel by looking at Wittgenstein’s account of how thought and language represent facts, an

489

Arata Hamawaki

account that employs an analogy between propositions and pictures. A central idea that the analogy is meant to illuminate is the difference between a name and a proposition. As we saw, a name, (a name in the strict sense) represents what it does by standing for it, but for Wittgenstein there is nothing, no entity, that a proposition stands for. For one thing, if it were to stand for the fact that the proposition represents then a proposition could not represent something that is not the case.9 The problem of understanding the nature of a proposition is the problem of understanding the possibility of false propositions. How does a picture represent what it does? The short (but, as we will see, in the end misleading) answer for Wittgenstein is that it does so because the elements in the picture—elements that stand for certain objects in the world—are arranged in a certain way. In other words, a picture isn’t just a complex object that stands for something in the world—that would be to think of a picture as akin to a complex name. Rather, as Wittgenstein puts it, “a picture is a fact” (T, 2.141). It is the fact that the elements in a picture are arranged in a certain way that represents the fact that the things in the world that correspond to the pictorial elements are arranged in a certain way. Consider a map that represents the relative distances and spatial positions of three cities. It is the fact that the dots on the map (“pictorial elements”) which stand for the cities are arranged as they are that represents that the cities those dots stand for are arranged in a corresponding way. Similarly, Wittgenstein argues, a proposition represents a certain state of affairs because the elements of the proposition, the names in it, are arranged in a certain way. It is the fact that they are arranged as they are that represents that the objects the names refer to are arranged in a certain way. Thus, unlike a name, whose meaning is an entity the name stands for, the sense of a proposition is in the proposition, is in the arrangement of the signs.10 Thus, Wittgenstein writes, “Instead of, “the complex sign ‘aRb’ says that a stands to b in the relation R”, we ought to put, “That ‘a’ stands to ‘b’ in a certain relation says that aRb”(T, 3.1432).11 Wittgenstein writes, “if a fact is to be a picture, it must have something in common with what it depicts. There must be something identical in a picture and what it depicts, to enable one to be a picture of the other at all” (T, 2.16–2.17). Consider again the example of a map that depicts the relative distances and orientation of certain cities. That is possible only because the map shares with what it depicts what Wittgenstein calls “pictorial form”. In this case, the common form is spatiality. Because this is so, any arrangement of pictorial elements in a map corresponds to some possible situation of cities.12 Wittgenstein writes, “a picture can depict any reality whose form it has. A spatial picture can depict anything spatial, a coloured one anything coloured, etc.” (T, 2.171). The commonality of form guarantees that any arrangement of elements in the picture (what Wittgenstein calls “pictorial structure”) represents a possible arrangement in the reality that the picture depicts. But this raises a question once we move from pictures to propositions. In the case of a map, any arrangement of the pictorial elements on the page represents some corresponding arrangement of the objects of these elements, the spatial situation of towns, but signs can be combined any old way, and not any way of combining signs manages to represent. This is right, of course, as far as the physical marks are concerned, but not if the marks are employed according to a “rule of projection”. Consider the following toy model of the rules of projection for the marks “a”, “b”, and “R”. According to this rule, if “a” left flanks “R”, and “b” right-flanks “R”, as in “aRb”, the proposition says that a loves b. If “b” left flanks “R” and “a” right-flanks “R”, as in “bRa”, the proposition says that b loves a. It is important here to distinguish between a mark and a sign, where a sign is a mark that figures in a complex of marks that “stands in a projective relation to the world”. A proposition is neither a certain arrangement of marks nor a certain arrangement of signs, for a sign is a sign only insofar as it signifies as part of a whole

490

Wittgenstein

proposition. While it is possible to arrange marks thus, “Rab”, such an arrangement of marks says nothing, does not stand in a projective relation to the world according to the above rules of projection: those marks do not figure as signs. Wittgenstein writes, “a picture cannot, however, depict its pictorial form; it displays it”, (T, 2.172), since “a picture cannot place itself outside its representational form” (T, 2.174). Consider again the example of the map that represents the spatial situations of three towns by virtue of the arrangement of three dots on a page. The arrangement of the dots on the page “picture” the situations of the three towns, but it doesn’t at the same time picture that in virtue of which the picture pictures what it does. It does not in Wittgenstein’s words, “depict its pictorial form”. Nor could we do so by means of any other map. As Gilbert Ryle put it, “you cannot make a map of what another map says or of how it says it … The significance-conditions which an ordinary map exemplifies are not stated by these or any other maps”.13 What a proposition represents isn’t itself something that the proposition, or any other proposition, can state. Rather, “a proposition shows its sense. A proposition shows how things stand if it is true. And it says that they do so stand” (T, 4.022). This can seem to be obviously false, for can’t I just say what the sentence “Antony loves Cleopatra” means?14 It means: Antony loves Cleopatra. But notice that in saying what the sentence means I have used the sentence “Antony loves Cleopatra”. Saying ““Antony loves Cleopatra” means Antony loves Cleopatra” conveys information only to someone who understands the meaning of the sentence, “Antony loves Cleopatra”. I could use another language to give the meaning of the sentence, say, Japanese: ““Antony loves Cleopatra” no imi wa Antony ga Cleopatra o aishitemasu”. But my saying that conveys information only to someone who knows Japanese, who knows the rules of projection for the arrangement of signs in that language. The point is that stating the sense of a proposition requires that you are already able to discern the sense of the proposition that gives the sense. In that sense, in giving the meaning of “Antony loves Cleopatra” I am not stating its sense, I am “displaying” it, “showing” it, for someone who has a mastery of the rules of projection for discerning sense in the language. There is no explanation of the meaning of a sentence, of the possibility that the sentence represents, that doesn’t already presuppose, as Thomas Ricketts puts it, “the discrimination of possible situations in logical space that constitutes the understanding of language”.15 We see, then, that there is an important disanalogy between the way propositions represent and the way pictures, such as maps and scores do. In the case of a map or a score, we can state what objects in the world the pictorial elements correspond to, and thereby set up what Wittgenstein calls “pictorial relations”. We can do this without yet being able to “read” the map or the score. Thus, in those cases the setting up of the pictorial relations and the discernment of pictorial form can be conceived of as two separate steps that are involved in picturing. However, in the case of representation in language, these two steps have to occur in unison or not at all, which is to say that for the elementary propositions there is no such thing as “setting up” the relation between pictorial elements and the objects they stand for. In Philosophical Remarks, Wittgenstein writes, Any kind of explanation of a language presupposes language already. And in a certain sense, the use of language is something that cannot be taught, i.e., I cannot use language to teach it in the way in which language could be used to teach someone to play the piano.—And that of course is just another way of saying: I cannot use language to get outside of language (PR, 54).16

491

Arata Hamawaki

And isn’t the account of the possibility of meaning that takes up the parts of the Tractatus we have so far discussed an attempt to “use language to get outside language”? The idea that we can use language to represent the possibilities of sense presupposes that we use language without presupposing a grasp of the rules of projection that constitute the possibilities of sense. That is what we would be doing if we thought we could set up the pictorial relation between name and object in a separate, initial step. That would be, in effect, as Wittgenstein puts it, “to station ourselves outside of language”: Propositions can represent the whole of reality, but they cannot represent what they must have in common with reality in order to be able to represent it—logical form. In order to be able to represent logical form, we should have to be able to station ourselves with propositions somewhere outside logic, that is to say outside the world (T, 4.12). What happens when we try to do so? Our words turn to gibberish, unsinn.17 Thus, when earlier in the Tractatus, Wittgenstein said such things as the objects for which names in elementary propositions stand must be simple and have certain internal properties, it looked for all the world as though he were stating necessary truths about the world. It now emerges that those were failed attempts to say something, illusions of thought. It is at this point that we come up against a crucial division among interpreters. There are those who hold that there really is something that we can’t state but that is nevertheless “shown”. We can call this the “ineffability view”: the view that while there are no meaningful metaphysical statements, what they inadequately point to, gesture at, is what must be so, the logical structure that language and world share. On this view there is something—metaphymetaphysical truths—that we would have failed to state: what that something is is nevertheless shown.18 On the opposing interpretation such a view is “irresolute”, what Cora Diamond memorably calls “chickening out”: as I suggested earlier, it fails to own up to what the Tractatus is fully committed to, namely, that the metaphysical statements are as nonsensical as “Twas brillig and the slithy toves did gyre and gimbel in the wabe”.19 How if the metaphysical statements fail to say anything at all can we still hold onto the idea that there is something that they attempt but fail to say? If we understand Wittgenstein correctly, we must, on this reading, throw away the whole kit and caboodle, including the idea that, as Diamond puts it, “there are features of reality that cannot be put into words but show themselves”. What is at stake in this dispute? The ineffability reading holds onto the idea that there is something, a structure that both thought or language and world share, that underlies and explains the possibility of thought or language. This “something”, however, cannot be stated but only shown. But on the resolute view the very idea that there is something to explain belies the nature of thought and language, for that idea presupposes that we are able to conceive of thought and language “from outside” thought and language. If thought and language can only be understood from within thought and language, there is no standpoint from which we can formulate the question what explains how thought and language can in general be about the world, where an answer to that question would need to avoid assuming mastery of the language. The entire enterprise of trying to ask and then trying to answer that question rests on a misunderstanding of the nature of thought and language. There is simply no question here whose answer can be stated or shown. According to the resolute reading, understanding that is grasping the nature of thought and language.20

492

Wittgenstein

33.2

Part II. The Philosophical Investigations

The Philosophical Investigations opens with a passage from Augustine’s Confessions, which Wittgenstein takes to give expression to what he calls “a particular picture of the essence of human language”: It is this: the individual words in language name objects—sentences are combinations of such names.—In this picture of language we find the roots of the following idea: Every word has a meaning. This meaning is correlated with the word. It is the object for which the word stands (PI, §1). I want to begin here by taking up Wittgenstein’s discussion of one element of this picture, the idea that the meaning of a name is the object it stands for. As we saw, this is how Wittgenstein conceived of the meaning of names in the Tractatus, a conception of their meaning that is connected with the metaphysical view that the objects of (genuine) names are simples that essentially occur in the states of affairs they occur in. This conception is put under some pressure in the discussion of “ostensive definition”. Now, to be sure, nowhere in the Tractatus does Wittgenstein appeal to ostension to explain how names come to stand for objects. In fact, as we saw, what he says there precludes such an account.21 However, the appeal to ostensive definition can be understood as one manifestation of thinking of naming as what Wittgenstein calls an “occult process” (PI, §38), a charge that could perhaps be leveled at the Tractatus as well. We will see that the antidote to so thinking involves reconceiving of essence as something that is expressed in what Wittgenstein calls the “grammar” of our language.

33.2.1

Ostensive Definition

Assuming, with Augustine and the Tractatus, that the meaning of a word is the object it refers to, there arises the question how words get their meaning. If this is to be understood as a philosophical question, it must be understood as being posed in full generality. Obviously, we can explain the meaning of any individual word by using other words, whose meanings we already understand. Such explanations would not help us in giving a philosophical explanation of meaning, for a philosophical explanation of meaning would need to be fully general, it would have to explain how someone could understand the meaning of words without already assuming that the person already had an understanding of language.22 The idea that meaning can be explained by ostensive definition promises to be just what we need to deliver on the sort of understanding of meaning that would answer such a philosophical question: it would be through ostensive definition that language as a whole is anchored to the world. Wittgenstein writes, The definition of the number two, “That is called ‘two’”—pointing to two nuts—is perfectly exact.—But how can two be defined like that? The person one gives the definition to doesn’t know what one wants to call “two”; he will suppose that “two” is the name given to this group of nuts … And he might equally well take the name of a person, of which I give an ostensive definition, as that of a colour, of a race, or even of a point of the compass. That is to say: an ostensive definition can be variously interpreted in every case (PI, §28). Wittgenstein’s point here isn’t just that the mere act of pointing while repeating a certain sound pattern can be variously interpreted. He is making the more specific point that the

493

Arata Hamawaki

pointing finger isn’t sufficient all on its own to pick out what kind of thing is being pointed to, for example, a number, a colour, a race, a point on the compass and that unless we can, so to speak, slot the object under the right kind, we will not be able to succeed in defining a word through ostension. He goes on: Perhaps you say: two can only be ostensively defined in this way: “This number is called ‘two’”. For the word “number” here shews what place in language, in grammar, we assign to the word. But this means that the word “number” must be explained before the ostensive definition can be understood.—The word “number” in the definition does indeed shew this place; does shew the post at which we station the word (PI, §29). Wittgenstein offers the following analogy: if one is explaining chess to someone and says pointing to a chessman, “this is the king” those words “are a definition only if the learner already ‘knows what a piece in a game is’” (PI, §31). Otherwise the learner will not know whether what is being pointed to is a piece in a game or to the physical thing that serves as a piece in a game. Just as ostensively defining a chessman as a king can succeed only if the learner has enough familiarity with games like chess to know what a piece in a game is, so an ostensive definition of a word can succeed only if the learner has sufficient mastery of language to know in which “grammatical” category to “station” the word: “so one might say: the ostensive definition explains the use—the meaning—of the word when the overall role of the word in language is clear” (PI, §30). Wittgenstein considers the objection that mastery of a language isn’t necessary to understand an ostensive definition: “all you need—of course!—is to know or guess what the person giving the explanation is pointing to” (PI, §33). The objection is that while the physical gesture of pointing is variously interpretable, what gives determinacy to the physical act of pointing is an inner act of meaning the color, the number, the shape, and so on.23 Wittgenstein, in characteristic fashion, does not attempt to give an argument that would purport to rule out the possibility that there can be such inner acts of meaning, but asks how this is to be done. Just how do we attend to the color as opposed to the shape? And do we always do the same thing when we attend to the color of something? He acknowledges that there may well be characteristic processes that occur whenever one attends to the shape: “you attend to the shape, sometimes by tracing it, sometimes by screwing your eyes so as not see the colour clearly, and in many other ways” (PI, §33). However, he points out that while such processes could, and sometimes do, accompany the state of attending to the shape rather than the color, they needn’t do so, and hence don’t constitute what it is to attend to the shape rather than the color.24 It is important to see that Wittgenstein does not offer a competing account of what constitutes attending to, or intending, the shape rather than the color but rather draws our attention to certain conditions that must be in place for such acts to take place. In this connection he makes two points. First, he suggests that what one points to will depend on the circumstances, that is, on what happened before and after the pointing. (PI, §35) Second, he suggests that it is only on the condition that there is agreement across different particular cases in which we use the phrase “this counts as pointing to a shape” that there is such a thing as pointing to a shape at all. This is implied by passages such as the following: “just as a move in chess doesn’t consist simply in moving a piece in such-and-such a way on the board—nor yet in one’s thoughts and feelings as one makes the move: but in the circumstances that we call ‘playing a game of chess’, ‘solving a chess problem’, and so on” (PI, §33). The point Wittgenstein is making here is a central lesson that we are to draw from the analogy between

494

Wittgenstein

language and games. Whether someone makes a move in a game isn’t a matter of discerning on the basis of what she does the intention that lies behind the act and that accounts for its being a move in a game. It is rather only because there is a shared practice in which performing certain physical actions counts as making a particular move that there is the possibility of making a move in a game at all. Similarly, it is only because there is a practice of pointing to the shape as opposed to the color that there is such a thing as pointing to the shape as opposed to the color. And such a practice depends on there being agreement on what counts as pointing to the shape as opposed to a color.25 It is, I take it, agreement in such things, as opposed to agreement in truth-evaluable judgments, that Wittgenstein is speaking of when he famously says, “If language is to be a means of communication there must be agreement not only in definitions but also (queer as this may sound) in judgments. This seems to abolish logic, but does not do so … ” (PI, §243) To appreciate Wittgenstein’s points concerning ostensive definition, it is useful to place his discussion against the background of two prevailing ways of understanding reference. Reference can be understood as being mediated by a description of the object a word refers to, or it can be understood as “immediate”, or “direct”, that is, unmediated by a description. Ostensive definition—and later, the idea of a private language—treats the reference of at least some terms as “immediate”.26 He can be seen here as challenging both direct reference and description theories of the reference of names, although we should be careful to guard against ascribing an alternative theory of reference to him. His point is that reference can be achieved only against the background of an understanding of what he calls “the overall use of the word in the language”, that is, an understanding of the categorical or “grammatical” kind to which the word belongs. That in turn is, as we have seen, a matter of agreement on the use of the word (“and how he takes the definition is seen in the use that he makes of the word defined” (PI, §29)), and is hence dependent on a social practice of using the word. Thus, a grasp of what could be called differences in essence, say, between color words and shape words and number words, is presupposed by the reference of those words, where such a grasp is a matter of using the words in accordance with the practice of using words that express those kinds. If Wittgenstein is right, that would imply that essence is not something that is discovered as a scientific fact is discovered, but rather the essence of what are traditionally regarded as “metaphysical” kinds belongs to the logical grammar of our language. That is, essences cannot explain the categorical distinctions in language, since those distinctions belong to the framework in which reference to objects is possible in the first place.27 It is important to remember that Wittgenstein isn’t giving a reductive explanation of what it is to mean something by a word, one that appeals to behavioral dispositions to use words rather than to mental states of intention. In fact, his point is something like the opposite of that, for he is saying that any account of the meaning of how words obtain meaning, such as those that appeal to ostensive definition, presupposes a prior mastery of language. To use the phrase from the Philosophical Remarks cited earlier, Wittgenstein maintains that there is no explanation of language “from outside language”. Only in PI, he broadens his conception of “agreement in the language we use” so as to encompass not just agreement on the meanings of words but on what he calls “agreement in form of life” (PI, §242). This suggests that in describing the learning of language as a matter of learning the names of things, Augustine radically under-describes what children learn from us when they learn language. As Stanley Cavell puts it, in “learning language” you learn not merely what the names of things are, but what a name is; not merely what the form of expression is for expressing a wish is, but what

495

Arata Hamawaki

expressing a wish is; not merely what the word for “father” is, but what a father is … you do not merely learn the pronunciation of sounds, and their grammatical orders, but the “forms of life” which make those sounds the words they are., do what they do—e.g., name, call, point, express a wish or affection, indicate a choice or an aversion, etc. (Cavell (1979): 177–8)

33.2.2

Family Resemblance

I have been claiming that Wittgenstein is, in Stanley Cavell’s words, “retrieving” the concept of essence, rather than rejecting it outright.28 How does this square with the anti-essentialist remarks in the passages on “family resemblance”. It is to these passages that I now turn. As we saw, Wittgenstein is fond of drawing an analogy between speaking a language and making a move in a game. He writes, “here the term ’language-game’ is meant to bring into prominence the fact that the speaking of language is part of an activity, or of a form of life” (PI, §23).29 Wittgenstein takes up the objection that while he discusses many examples of language-games, he nowhere says what “the essence of a language-game, and hence of language, is: what is common to all of these activities, and what makes them into language or parts of language” (PI, §65). It is in response to this objection that Wittgenstein elaborates on his famous point about “family resemblance”. He writes, Consider for example the proceedings that we call “games”. I mean board-games, cardgames, ball-games, Olympic games, and so on. What is common to them all?—Don’t say: “There must be something common, or they would not be called ‘games’”—but look and see whether there is anything common to all.—For if you look at them you will not see something common to all, but similarities, relationships, and a whole series of them at that. To repeat: don’t think, but look! (PI, §66). We might try to compile a list of features that it appears all games have in common. We might propose, to begin, that a game is a rule-governed competition between two or more players, with a resolution that ends in one or more players being declared the victor. It is, however, easy to come up with counter-examples, for there are games that don’t involve multiple players, don’t have a winner, aren’t competitions, and don’t have a resolution. And even if there were no extant games that are counter-examples to the definition, it seems that we can’t rule-out the possibility of constructing a game that would be. Nevertheless, despite the challenging nature of the enterprise, one might persist, as philosophers have done with questions concerning the essence of art, of life, of ethical virtue. Certainly, Wittgenstein says nothing to convince someone engaged in such an enterprise of the futility of the quest. What, then, is the point of the discussion? He seems to propose an alternative model for understanding why we apply the same word or concept to a wide range of different instances. The model is this: “instead of producing something in common to all that we call language, I am saying that these phenomena have no one thing in common which makes us use the same word for all,—but that they are related to one another in many different ways. And it is because of this relationship, or these relationships, that we call them all “language” (PI, §65). On the alternative model he proposes there is a “complicated network of similarities, overlapping, and criss-crossing: sometimes overall similarities, sometimes similarities of detail” (PI, §66). Wittgenstein calls such a network of similarities “family resemblances”. But again, suppose we do accept this model in the case of the concept of “game”. Is Wittgenstein suggesting that we do so for all of our concepts, and if not for all, then for which

496

Wittgenstein

ones? He says nothing to address such questions. Furthermore, is Wittgenstein proposing that we think of family resemblance as an alternative explanation of why we apply the same word across different instances? That would surely just push the question back, for it could be asked what is the essence of family resemblance? What is to count as “a complicated network of similarities, overlapping, and criss-crossing: sometimes overall similarities, sometimes similarities of detail?” What, in other words, is the definition of “family resemblance”? More importantly, an appeal to family resemblance would leave unexplained which similarities count as the ones that matter. That is, any appeal to family resemblance would presuppose that we already knew which resemblances mattered out of the indefinite number of resemblances there are between the instances of a concept. Surely, then, the idea of family resemblance isn’t meant to do what the idea of universals or essences was supposed to do, as some have thought Wittgenstein was arguing.30 What, then, is the point of that idea, provocative though it may be? I take it that the idea is meant to be employed to help us dispel the assumption that there must be something that all of the instances of a concept have in common, some essence that gives the meaning of the concept, a universal or Platonic form that explains why the concept subsumes the particular cases it does. This seems to be the point of his admonition “to repeat: don’t think, but look!” This interpretation is confirmed by Wittgenstein’s treatment of the following, to his mind feeble, objection: “But what if someone were to say: “There is something common to all these constructions—namely the disjunction of all their common properties”—I should reply: Now you are only playing with words” (PI, §67). The reason that this would be playing with words is that while it may be literally true that such a disjunction would display what all games had in common (assuming that it could be completely specified), it could not do the work that essences or universals were meant to do, namely, explain why we apply the same word or concept to the different cases we apply it to. The target of Wittgenstein discussion, then, is someone who feels the need for something to do the work that essence or universals are meant to do. Wittgenstein makes a number of points by way of exposing this felt need as a false one. First, he argues that generally our concepts don’t have rigid boundaries, don’t need rigid boundaries to be useful, and may even suffer from having rigid boundaries (PI, §68–70). Second, he points out that any general definition can be misunderstood. Whether a word is to be applied to a new case would depend on how the words, or images, that are employed in the general definition are themselves to be taken. He writes, “here giving examples is not an indirect means of explaining—in default of a better … ” (PI, §71). Third, he argues that what I know in knowing what it is to be an instance of a certain concept is exhausted by the examples and explanations I give: “isn’t my knowledge, my concept of a game, completely expressed in the explanations that I could give? That is, in my describing examples of various kinds of game, … ” (PI, §75). What I intend in giving examples and in giving the explanations I do is borne out in the way I use the word, in this case, the word “game”. That point can sound trivially true.31 Of course, my only evidence for what a person intends are the examples and explanations he gives. But Wittgenstein’s point is that what I intend is entirely exhausted by what is shown in the examples and explanations I give, something that one is able to discern if one has sufficient familiarity with the “language-games” we play with the word “game”. How someone intends the examples to be taken is not something that lies behind the examples, but is something that is essentially and completely displayed in one’s giving the examples. There are three important observations to make in connection with the last point. First, I alluded earlier to the role that perceptions or judgments of pertinence play in our having the concepts we do. That is something that I think the idea of family resemblance is meant to bring out. What explains our sharing concepts, such as the concept of a game, is that we agree

497

Arata Hamawaki

on what resemblances between instances of the concept are relevant for membership in a kind. It is only because we share our perception of which resemblances are relevant that we are able to see the intended commonality between the different examples. Here, Wittgenstein is insisting on the contingency that is built into our possession of concepts. As Frege and others have stressed, concepts, unlike perceptions and judgments, are essentially, not accidentally, shared. Wittgenstein is not disagreeing with Frege about this, but he is pointing out that certain contingent facts are a condition of sharing concepts, for example, the contingent fact that we generally agree that the instances of “game” count as instances of “game”. Wittgenstein is taking issue here with what he maintained in the Tractatus: that whether a proposition has sense cannot depend on whether another proposition is true. Second, as suggested by the idea I just invoked of the intended commonality, there isn’t in the end any difficulty in saying that there is something that all games have in common after all. What games have in common is displayed by my using a single word to refer to the individual instances I use it to refer to. After all, it is not, so to speak, a mere orthographical accident that I use the same word to refer to the different examples. If so, mustn’t there be something that they all have in common, albeit not something that we can state? We can see Wittgenstein here as exposing the dichotomy between conceptual realism and nominalism as a false one. To give up on the idea that a universal explains our use of a word is not to default to the view that what a word refers to is a matter of convention or decision.32 What the examples have in common is displayed in the way that I use the word, “game”, and that is something that is discernible only to someone who shares the judgments of pertinence that I do. And this is where the idea of family resemblance can have a significant role to play, for instead of being invoked as an explanatory replacement for universals, family resemblance can be used to model what someone who grasps the commonality among different games knows. Third, the last point is related to Wittgenstein’s remark, “Essence is expressed by grammar” (PI, §371). That remark both rejects a conception of essence in explaining why a word refers to what it refers to and incorporates a conception of essence freed from that explanatory role.

33.2.3

“Essence is Expressed by Grammar”

The analogy with games suggests that the rules that govern the meanings of words are, in some sense, conventional: just as it is a matter of convention that the piece called the “king” is one that can legitimately move in certain ways and not others, so it is a matter of convention that this is called a “game”, or this is called “being in pain”, or this is called “red”. Of course, there is an obvious sense in which language is conventional, for in English we use the sign “game” to mean game, and not the sign, “pain”. It could very well have been the other way around. But that is not the suggestion. The suggestion is that we could have had a different concept of pain, one that would still count as a concept of pain but is different from our concept of pain.33 The word “pain” would still mean pain, but what Wittgenstein calls the “grammar” of the word would be different. Here, the analogy with games seems to break down, for there is no analog for this in the case of a game. In chess, what the king is is simply defined by the moves it can (legitimately) make: the normative rules governing the piece constitute it as the piece that it is in the game. In that sense what I am calling a “piece” here is a role. If we were to adopt different rules governing the king then we would simply have a different piece/role. In the new game there is no longer the piece/role that in the earlier game we called a “king”. The rules that constitute the pieces in chess are simply conventional: they do not have what Wittgenstein calls “grammar”, which has to do with what does and does not make sense.34

498

Wittgenstein

Wittgenstein says that it belongs to the grammar of our concept of pain that “it makes sense to say about other people that they doubt whether I am in pain; but not to say it about myself”(PI, §246).35 To say, then, that grammar, in this case the grammar of “pain”, is conventional implies that it is possible to imagine, that is, to find intelligible, a concept of pain that did not have this grammar, on which it does make sense to say about myself, “I doubt that I am in pain”. We can grant to Wittgenstein that that is not our concept of pain, but could we find intelligible a concept of pain that did not have that grammar? Wittgenstein, I think, suggests that the answer is no.36 Now Wittgenstein, as we have seen, rejects the idea that our concepts track essences. Thus, he rejects the idea that the concepts we have are the correct ones.37 But he also rejects the view that they are merely products of convention, for example, the idea that we have simply adopted as a convention (perhaps for pragmatic reasons?) that nothing is to count as my doubting that I am in pain because we have as a matter of convention simply laid it down that I am supremely authoritative with regard to whether I am in pain.38 In Zettel, Wittgenstein asks, “then is there something arbitrary about the system? Yes and no. It is akin both to what is arbitrary and to what is non-arbitrary” (Z, §358). The grammar of our concepts is arbitrary in the sense that they don’t track Platonic essences. However, they are not arbitrary because they are not conventional—we cannot make sense of an alternative grammar, even though what does and does not make sense depends on facts about our nature, facts about our “form of life”. This is perhaps another respect in which our language might be thought to be similar to a game, for despite the fact that games are based on conventions, the fact that we play the games that we do seems “akin both to what is arbitrary and not arbitrary”.

32.3

Related Topics in this Volume

Howat, Andrew, “Pragmatism” Livingstone, Jonathan, and Sidelle, Alan, “Conventionalism” Malink, Marko, “Ancient” Michels, Robert, “Contemporary: Analytic Tradition” Ritchie, Katherine, “Language of Essence” Schechtman, Anat, “Modern” Anand Vaidya and Michael Wallner, “Conferralism”

Notes 1 I am indebted to helpful comments on an earlier version from Kathrin Koslicki and Mike Raven. I am also indebted to conversations with Kelly Jolley on topics covered in this chapter. 2 “Order”, here, I think, means something like “level” or perhaps “register”. There is the level of fact, of what is or is not the case. A fact is a truth. So facts presuppose the level of possibilities, the level of what must be the case if there are to be truths, facts. 3 A terminological note: Wittgenstein generally restricts the term “sense” to propositions. Propositions have sense, but, contra Frege, no reference. As we will see, names for Wittgenstein are senseless. They have meaning (Bedeutung) but no sense (Sinn). Their meanings are the objects they stand for. Predicates for Wittgenstein do not have meaning, that is, reference. They could be said to have sense in that they constitute the unity of a proposition. 4 In my exposition of Wittgenstein’s answer to this question, I am indebted to H. O. Mounce (1981) 5 As we will see, these propositions don’t really represent “states of affairs” in Wittgenstein’s technical sense, but they will do for our purposes here. 6 See Mounce (1981): 18

499

Arata Hamawaki 7 In saying that states of affairs are in logical space, Wittgenstein doesn’t mean that their logical connections with one another are essential to individuating them. A state of affairs is the content of what Wittgenstein called an “elementary proposition”, and elementary propositions are logically independent of one another: “each item can be the case or not the case while everything else remains the same” (T, 1.21). Wittgenstein thought that all other propositions are truth-functions of elementary propositions. States of affairs are in one sense independent of one another, since one can exist irrespective of whether any of the others do, but they are in another sense interdependent, since they are composed of objects that essentially occur in the states of affairs they occur in. 8 Note that Wittgenstein must employ something like a distinction between accidental and essential predication, here with respect to existence, for when we say that Moses existed, what we mean is that Moses was actual. But that is not what Wittgenstein means when he says that the objects of logical names exist. He thus says not that substance exists but that it “subsists” (bestehen): “substance is what subsists independently of what is the case” (T, 2.024). For discussion of this distinction with respect to other figures in the philosophical tradition, see the contributions in this volume by Marko Malink, “Ancient”, and by Anat Schectman, “Modern”. 9 To get around this problem we might suppose that a proposition stands for a possibility rather than a fact. But, then, in order to represent a fact we would have to add some content to what a proposition represents, such as the content that the possibility represented by the proposition obtains. But then someone who asserts that some proposition, p, is true would not be representing what is represented by someone, who merely thinks that p, without asserting it. 10 Here, again, my exposition of the point is indebted to Mounce. See Mounce (1981), 20-26. 11 It is important to note here that what Wittgenstein is calling a “proposition” is not what Frege called a “thought”. For Frege a thought is expressed by a sentence, but it isn’t itself a sentence, and furthermore, its expression by a sentence is incidental to it. For Wittgenstein a proposition is, in a sense, a sentence, that is, a string of signs that stand in what Wittgenstein calls a “projective relation to the world”. 12 Another example is a musical score that represents a piece of music. But in this case the arrangement of the notes is spatial and the arrangement of what it represents is temporal and so representation and represented don’t naturally share the same possibilities of combination, as in the case of a map. However, what they share is a certain isomorphism: the left-right ordering of the notes is isomorphic to the before-after ordering of the sounds. 13 Ryle (1957): 273. 14 To use an example from Ricketts (1996). 15 Ricketts (1996): 91. I am indebted to Ricketts’s exposition of this point here. 16 Barry Stroud has written a number of papers on Wittgenstein that stress this point from the Philosophical Remarks. I follow Stroud in stressing this point throughout the reading of both the Tractatus and the Investigations here. See Stroud (2018). 17 Wittgenstein distinguishes between two ways in which a proposition can be lacking in sense. It can be senseless (sinnlos), or it can be nonsense (unsinn). He argues that the propositions of logic are tautologies—hence, sinnlos. The propositions of metaphysics are by contrast unsinn. Neither are genuine propositions, but the propositions of logic “show” the limits of that wherein there can be sense. 18 For an example of this approach see P.M.S. Hacker (1972): 20–4. 19 See Cora Diamond (1988) and James Conant (1989). 20 Is Wittgenstein saying that objects have the logical form that they do because it is a projection of the logical form of our language so that the world is the world-as-we-represent-it-in-our-language? Think of this as a linguistic version of Kant’s “transcendental idealism”. It should be clear that that is no improvement as far as the resolute reading is concerned. The resolute reading doesn’t maintain that we are to be idealists about the ontological talk, it urges us to jettison the ontological talk altogether, to realize that it is a failed attempt, whether in the realist’s or in the idealist’s hands, to say something that can’t be said or thought. For an idealist reading of the Tractatus see P.M.S. Hacker (1972) 21 For further development of this point see Ishiguro (1969). 22 See Michael Dummett (1975), who conceives of a philosophical explanation of meaning along these lines. John McDowell (1998), in a Wittgensteinian vein, raises doubts about whether such an explanation is to be had.

500

Wittgenstein 23 Wittgenstein writes, “and we do here what we do in a host of similar cases: because we cannot specify any one bodily action which we call pointing to the shape (as opposed, for example, to the colour), we say that a spiritual [mental, intellectual] activity corresponds to these words” (PI, §36). 24 cf. PI, §35. 25 Of course, these are large claims that would need to be made out, a challenge that Wittgenstein takes up throughout PI, for example, in the rule-following discussion (PI, §185–242) and the muchdiscussed passages concerning the idea of a private language (PI, §243–63). 26 Although he does not mention it here, a causal theory of reference would be another example of such a view. 27 For further discussion of this general approach to essence see Vaidya and Wallner, this volume. 28 See Cavell (1979): 186. 29 Wittgenstein gives a list of examples of language-games: “Giving orders, and obeying them—Describing the appearance of an object, or giving its measurements— Constructing an object from a description (a drawing)—Reporting an event—Speculating about an event—Forming and testing a hypothesis … ” (PI, §23). 30 For an opposing reading see Renford Bambrough (1961). 31 I thank Mike Raven for pressing me on this point. 32 In this connection see Livingstone and Sidelle, this volume. 33 In a similar vein, Wittgenstein asks, “Can’t we imagine that people do not have our color concepts and that they have concepts which are related to ours in such a way that we would want to call them “color concepts”? (RC, III, §154). 34 Wittgenstein calls his investigation a “grammatical one” (PI, §90). 35 Another example: “could someone have a feeling of ardent love or hope for the space of one second—no matter what preceded or followed this second?” (PI, §583). 36 Wittgenstein poses similar questions about other concepts, e.g., could we imagine a color system that allowed for the possibility of reddish-green? 37 Wittgenstein writes, “if anyone believes that certain concepts are absolutely the correct ones, and that having different ones would mean not realizing something that we realize—then let him imagine certain very general facts of nature to be different from what we are used to, and the formation of concepts different from ours would become intelligible to him” (PI, p. 241). 38 This point is also made by Jolley and Long (2010): 169–75, to whose helpful discussion I am indebted here.

References Brambrough, Renford (1961) “Universals and Family Resemblances,” Proceedings of the Aristotelian Society 61, pp. 207–222. Cavell, Stanley (1979) The Claim of Reason, Oxford University Press. Conant, James (1989) “Must We Show What We Cannot Say?” in R. Fleming and M. Payne, eds., The Senses of Stanley Cavell, Bucknell University Press. Diamond, Cora (1988) “Throwing Away the Ladder: How to Read the Tractatus,” Philosophy 63 (243), pp. 5–27. Dummett, Michael (1975) “What is a Theory of Meaning? (I),” in Samuel Guttenplan, ed., Mind and Language, Oxford University Press, pp. 97–138. Hacker, P. M. S. (1972) Insight and Illusion, Oxford University Press. Ishiguro, Hide (1969) “The Use and Reference of Names,” in Peter Winch, ed., Studies in the Philosophy of Wittgenstein, Routledge, pp. 20–50. Jolley, Kelly and Long, Roderick (2010) “Grammatical Investigations,” in Kelly Jolley, ed., Wittgenstein: Key Concepts, Routledge, pp. 169–175. John, McDowell (1998) “In Defense of Modesty,” in John McDowell, ed., Meaning, Knowledge, and Reality, Harvard University Press, pp. 87–107. Mounce, H. O., (1981) Wittgenstein’s Tractatus: An Introduction, University of Chicago Press. Ricketts, Thomas (1996) “Pictures, Logic, and the Limits of Sense in Wittgenstein’s Tractatus,” in Hans Sluga and David Stern, eds., Cambridge Companion to Wittgenstein, Cambridge University Press, pp. 59–99.

501

Arata Hamawaki Ryle, Gilbert (1957) “Review of Ludwig Wittgenstein: ‘Remarks on the Foundations of Mathematics,’” Scientific American, v. CXVII, 1957, reprinted in his Collected Papers, v. 1, Routledge, 2009, p. 267–277. Stroud, Barry (2018) “Meaning and Understanding,” in Barry Stroud, ed., Seeing, Knowing, Understanding, Oxford University Press, pp. 233–255. Wittgenstein, Ludwig (1953) Philosophical Investigations (PI), G. E. M. Anscombe and Rush Rhees, eds., G. E. M. Anscombe, transl., Third Edition, MacMillan. Wittgenstein, Ludwig (1961) Tractatus Logico-Philosophicus (T), David Pears and Brian McGuinness, translators, Routledge. Wittgenstein, Ludwig (1970) Zettel (Z), G. E. M. Anscombe and G. H. von Wright eds., G. E. M. Anscombe, transl., University of California Press. Wittgenstein, Ludwig (1975) Philosophical Remarks (PR), R. Hargreaves and Roger White, eds. and transls, Blackwell. Wittgenstein, Ludwig (1978) Remarks on Colour (RC), G. E. M. Anscombe, ed., Linda L. McAlister and Margarete Schättle, transls., University of California Press.

502

INDEX

a priori 46, 54, 72, 80, 87–88, 93, 160, 163, 210–223, 280, 439–440, 448, 450–451, 487 Aas, S. 413 abortion 392, 414 accidental property 1–2, 4–6, 30–32, 35, 37, 43, 50, 62, 88–89, 105, 108, 113, 118, 120, 143, 146–147, 150, 188, 194, 197, 200, 216, 221, 230–231, 235, 297, 299, 311, 396, 403, 406, 429, 445, 447, 473–474, 488, 498, 500 Adams, M. 191 Adams, R. M. 143, 145, 148, 300–301 Ainsworth, C. 377 Ajdukiewicz, K. 80 Albritton, R. 28 Alcoff, L. 375, 379, 381 Almeder, R. 57 Almog, J. 93, 95, 120, 137, 178 Altman, B. 413 Amundson, R. 286 analyticity 90, 215, 217, 428, 430–432, 434, 444–446, 450 Anderson, L. 230, 235 Anderson, R. 416 Andreasen, R. O. 361, 367 Andrews, L. 412 Angelelli, I. 78–79 Annas, G. 412 anti-essentialism 4, 53–55, 59–61, 63, 87, 95, 211, 222, 276, 366–367, 370, 375, 379, 382, 384, 388, 437, 466 anti-realism 182, 472–474, 477, 481–482, 484–485 Antony, L. 417, 491 Appiah, A. 3, 11, 363, 366, 394 Aristotelian essentialism 2, 89–91, 95, 185, 261–262, 276, 431, 433 Aristotelian metaphysics 316

Aristotle 2–3, 20–28, 30–31, 33–34, 50, 84–85, 88–89, 95, 106–107, 111, 115, 137–138, 143, 150, 190, 209, 211–213, 218–219, 222, 286, 316, 319, 401–407, 410, 416, 418–419, 474 Armstrong, D. 187, 191, 194, 196–197 Arnold, A. P. 377 artifact 10, 150, 153, 160, 169, 209, 232, 261–272, 296, 347, 402–403, 405–407, 442, 450, 456, 468, 484 artificial 50, 55, 58 artwork 153, 160, 261–263, 265–267, 269, 271, 296, 324, 450, 468, 484 Assis, L. C. S. 282 Ásta 55, 272, 300, 368, 380, 384 Astuti, R. 364 Atkin, A. 54 Atkins, P. 63, 143, 390, 393 Audi, P. 307 Austin, C. J. 162, 191, 203, 283–284 Ayer, A. J. 55, 92, 95 Ayers, M. R. 151 Azzano, L. 202, 204 Bach, T. 371, 390, 392, 467 Backes, M. 344 Bacon, A. 2 Bader, R. M. 118 Baker, L. R. 3, 149, 160, 264, 270–271 Baldwin, T. 85–86, 95 Ballarin, R. 95, 172, 176 Banaji, M. R. 234 Banks, J. E. 112, 211, 302, 347, 437 Banton, M. 363 Barcan formula 90–91, 250 Barcan, R. C. 84, 90–91, 95, 108, 138, 250 Barker, M. J. 278, 283 Barker, S. 203–204

503

Index Barnes, E. 382, 384, 413 Barnes, J. 22, 28 Baron-Cohen, S. 376–377 Bartels, A. 191 Barth, E. M. 79 Bartol, J. 162, 164–165 Bastian, B. 229 Bauer, N. 378 Bauer, W. 191 Baxter, D. 343 Bayly, S. 363–364 Bealer, G. 210, 215, 450 Becker, O. 79 Beebee, H. 197, 205, 415 Beeghly, E. 368, 395 Bender, S. 50 Bengson, J. 210, 215, 218, 450 Bennett, J. 150, 414, 419 Bennett, K. 294–295, 297–299 Berger, P. 469 Bernstein, R. J. 54, 61 Berto, F. 121 Bettcher, T. M. 379, 381–382, 384, 417 Bhaskar, R. 182 Bigaj, T. 204 Bigelow, L. 191 Biggs, S. 450 Bingham, R. 357 biological determinism 376–377 biological essentialism 3, 162, 189, 191, 227, 376–377, 388, 395, 397, 407 biological species 3, 31, 38, 119–120, 153, 157, 161–162, 172, 178, 186, 189, 276–279, 281–283, 285, 305, 358, 367–368, 371, 419 biological systematics 276 Bird, A. 1, 11, 74, 76, 144, 156–159, 164, 191, 194, 196–199, 201–205, 229, 296 Black, M. 229, 381, 391–393 Blackburn, S. 444 Blackson, T. 452 Bliss, R. 295 Bloom, P. 228, 266 Blum, L. 95, 366 Bogardus, T. 376, 378, 382, 384 Boghossian, P. 211, 215, 217 Boisvert, R. D. 60, 62 Bolker, J. 283 BonJour, L. 210, 215, 217 Boorsay, A. 412 Borsboom, D. 348, 351 Bostock, S. 191 Bostrom, N. 412 Bovey, G. 94, 118, 127, 136 Boyd, R. 158, 161, 164, 281–282 Brake, E. 419 Brandom, R. N. 53 Bratman, M. 272

Brigandt, I. 3, 11, 143–144, 153, 161, 189, 191, 305, 348–349, 357–358, 364, 367, 371, 382, 396, 419, 457, 459 Briggs, R. 344 Brizendine, L. 377 Broackes, J. 191 Brody, B. 152, 182, 404 Brogaard, B. 113–114, 127, 286 Bronstein, D. 28 Brown, D. 50 Brown, H. 199 Brown, J. R. 218 Brown, L. 28 Buddhism 411 bundle theory 181, 186, 188–190 Burgess, J. 96, 453 Burke, M. B. 152–153 Burman, A. 269–270 Burnyeat, M. F. 28 Burton-Roberts, N. 231 Butler, J. 3, 11, 343, 375, 378–381, 383 Byrne, A. 382–384, 393, 397 Bzovy, J. 278 Cahn, S. M. 64 Cameron, R. 176–177, 216, 443, 445 Caplan, B. 247 Carlson, L. 412 Carnaghi, A. 234 Carnap, R. 53, 63, 80, 84, 87–92, 450 Carnino, P. 308 Carrara, M. 272 Carroll, L. 445 Carter, S. 344 Cartwright, N. 182 Cartwright, R. 2, 11, 89 Castañeda, H.-N. 343 Casullo, A. 94, 127, 200, 210, 213–214 categorical property 181, 186–188, 194–196, 198–201, 204–205 categoricalism 194–199, 202–205 categories 22–23, 26–28, 150, 157–158, 185–186, 219, 227–228, 230, 232, 234–235, 262, 265, 271, 282, 347, 351–352, 358, 364, 371, 376, 378–380, 384, 388, 390–393, 395–396, 417, 456, 458, 464, 476–477, 485 causal mechanism 164–165, 348–351, 353–355, 358 causation 25, 31, 59, 85, 112, 202, 212, 218–219, 355 Cavell, S. 495–496 Cavendish, M. 42–44, 49–50 Chakravartty, A. 158, 165, 204 Chalmers, D. 442, 450–451 Chandler, H. S. 120, 177 Chang, H. 58 change 1–2, 4–5, 23, 61, 85, 157, 171, 188, 199,

504

Index 205, 218–219, 226, 235, 277, 279–284, 286, 291–292, 294, 302, 336, 343, 352–355, 383, 390, 393, 411–412, 415, 417, 444–446, 456, 463–464 Charland, L. 357 Charles, D. 24, 28, 79, 376 Chiaradonna, R. 28 Chisholm, R. M. 93, 107, 109, 148, 177 Chodorow, N. 417 Christy, A. G. 227 Chudnoff, E. 210, 215 Cimpian, A. 227, 230 Clarke-Doane, J. 443 classical pragmatism 53 Cleland, C. E. 162 Coates, A. 120, 204 Code, A. 28, 78 Cohen, A. 231 Cohen, C. 342 Cohen, S. 28, 376–377 coincidence 43, 152, 221, 297–299, 323, 406, 447 Coleman, M. 191 Coley, J. D. 234 collective essence 8, 246 Collins, P. H. 381 color 23, 32, 38, 43, 68, 70–74, 86–88, 92, 214, 228, 365, 368, 415, 417, 464, 494–495 Conant, J. 500 conceptual engineering 11 conferralism 63, 95, 125, 137, 158, 211, 222, 371, 380, 437, 443, 449, 472–473, 475–485, 499 Conix, S. 278, 286 Connell, S. M. 28 consequential essence 7, 88, 127, 129, 241, 244, 247, 305, 321–322, 328, 331 constitution 11, 47–48, 84, 133, 152–153, 158, 164, 211, 221, 268, 334, 341–342, 344, 376, 406–407, 419, 464, 488 constitutive construction 465 constitutive essence 129, 241, 246, 248, 308, 321–327 contextualism 384 contingentism 94, 134–135, 245 conventionalism 50, 95, 125, 137, 158, 211, 222, 302, 437–451, 472, 475, 481, 484, 499 converse Barcan formula 250 Conway, A. 42, 44, 49–50 Cooper, R. 53, 171, 300, 348, 353–354 Coppock, P. 450 Corkum, P. 28 Correia, F. 78–80, 94–95, 108, 115–116, 143, 153, 165, 178, 190, 209, 227, 241–250, 253, 271, 278, 295, 302, 305, 307–308, 311–313, 315, 330, 361, 390, 393, 397, 441, 457, 483, 485 Corvino, J. 416

counterpart theory 93, 109, 120, 434, 446 Cover, J. A. 51 Cowling, S. 94, 100, 112, 116, 127, 145 Cracraft, J. 278 Cramer, A. 348, 351 Craver, C. 349–350, 358, 367 Crenshaw, K. 381, 391 Crivelli, P. 28 Cross, R. 38 Cross, T. 55, 145, 191, 343 crossworld identity 108–109 Cudd, A. E. 388–389 Cuneo, T. 215, 218 Dahl, O. 230 Daly, M. 416–417 Damnjanovic, N. 176 Dancy, J. 419 Darwin, C. R. 376 Dasgupta, S. 137, 144, 248, 307, 341 de Beauvoir, S. 3, 227, 375, 378, 417 de dicto 89–91, 106, 109–111, 326, 427, 430, 433–434, 450 De Freitas, J. 227 de Melo, T. X. 117–118, 127 de Queiroz, K. 278 de re 70, 89–91, 94, 106, 108–111, 120, 124, 146, 148–149, 209, 243, 326, 406, 427–430, 433–434, 449–450, 483 de Rizzo, J. 127, 311–312 de Spinoza, B. 44 De, M. 108–109, 120 debunking 60, 112, 388, 456, 463, 465–467 definition 1–2, 8, 20–22, 25–28, 33–34, 45, 53, 56–59, 63, 79, 84–89, 93–94, 133, 137, 209–211, 213–222, 246, 261, 306–309, 322, 330, 361, 371, 375–376, 378–380, 389–391, 393, 396–397, 403–404, 413, 428, 438, 442, 444–445, 458, 465–467, 493–497 Della Rocca, M. 2, 50, 120, 144, 301 Dembroff, R. 235, 383, 418 Denby, D. A. 94, 108, 127, 145, 301 dependence 31, 33, 37, 43, 45, 68, 77, 80, 93, 130, 163, 191, 202, 240–241, 243, 246–247, 251–252, 263, 271, 306, 308–309, 311–312, 352–356, 390, 437, 450, 456–457, 460–461, 464, 468, 482 deRosset, L. 173, 175–176, 295 Descartes, R. 6, 42–44, 49–50, 54–55, 80 Devitt, M. 55, 158, 162, 191, 200, 271 Dewey, John 53–55, 57, 60–63 Di Bella, S. 50–51 Diamond, Cora 492, 500 Diaz-Leon, E. 381 Dickie, George 271 Diekemper, J. 145 Dikötter, F. 364

505

Index Dillingham, E. M. 228 direct reference 91, 495 disability 381–382, 384, 389–394, 397, 411–414, 418, 455, 476 disposition 183, 188, 191, 194, 196, 202–203, 205, 221, 230, 272, 283–284, 337, 368, 370, 381–383, 395, 495 dispositional essentialism 94, 157, 159, 182, 189, 191, 194–195, 197, 199, 201–205, 305, 384 dispositional monism 194–199, 202–205 dispositional property 181, 183, 186, 194–197, 199–202, 204–205 Ditter, A. 94, 132, 137, 200, 242, 244 Divers, J. 435 Dixon, T. S. 134 Dobzhansky, T. 277 Dodd, J. 268, 272 Dorr, C. 128, 242, 341 Dretske, F. 187, 197 Dummett, M. 120, 451, 500 Dumsday, T. 162, 203, 281, 305, 459 Duncan, M. 306 Duncan, S. 50 Dunham, Y. 228, 231 Dunn, J. M. 112, 138 Dupré, John 3 Dutilh Novaes, C. 53 Earman, J. 197 Edwards, S. 38 Einheuser, I. 444–445, 451 Elder, C. L. 219–221, 271, 448 Ellis, B. 157–159, 162, 181 Emanuel, E. 413 Emmanuel, S. 411 Engelhard, K. 195 epistemic kinds 367 epistemic possibility 164 epistemology of essence 80, 109, 120, 209, 211, 213, 215–217, 219, 221, 450, 482–484 Epstein, B. 388, 390–391, 456–457 Ereshefsky, M. 3, 158, 278, 283, 286 Ernst, D. 229 essence see accidental property; collective essence; consequential essence; constitutive essence; essence operator; essential property; full essence; general essence; generic essence; immediate essence; individual essence; intrinsic essence; iterated essence; logic of essence; mediate essence; partial essence; predicational essence; species essence; uniessence essence operator 319, 321–323, 325, 327–328 essential property 5–8, 41–48, 50, 55, 61, 68, 85, 91, 93, 107–108, 111, 126–127, 129–131, 138, 143, 145–147, 150, 157, 159–161, 164–165, 178, 185–186, 189, 209–210, 212, 216, 218–220, 231, 269, 271, 291, 296, 300,

309–311, 321, 343, 352, 354, 364, 388–391, 393, 397, 406, 447, 455, 458, 460, 462, 465, 467–468, 472–473, 476 essentialism see Aristotelian essentialism; biological essentialism; constitutive essence; dispositional essentialism; generic essentialism; microstructural essentialism; psychological essentialism; racial essentialism; serious essentialism; social essentialism; strong essentialism; weak essentialism ethical value 343–344, 358, 410–411, 413, 415, 417, 419 ethics of enhancement 411 etiological kind 162 euthanasia 414 Evnine, S. J. 264–265, 267, 271–272 evolution 11, 57, 62, 89, 376, 437 evolutionary biology 277, 283–284 evolutionary theory 233, 280–281 existence 1, 9, 24, 30, 33, 35–37, 45, 54, 61–62, 70, 114–117, 119–120, 126, 128–129, 149, 151–152, 171, 178, 185, 213–214, 240, 245, 251, 263–264, 267–272, 278, 281, 285, 291–292, 319, 336–338, 340–341, 354–355, 366–367, 389–392, 401, 461–464, 468, 472, 478, 482, 488–489, 500 existential quantifier 308, 322, 328 explanation 9, 24–25, 28, 35, 43, 49, 54, 58, 77, 84–85, 89, 94, 119, 136–137, 165, 184, 203–205, 209–212, 218–222, 227–231, 248, 253, 277, 283–284, 295, 298, 302, 305–307, 309–311, 313–316, 319–320, 330, 334, 344, 347, 349, 351, 356, 358, 367–369, 379, 396, 407, 440, 444, 449, 451, 459, 466, 474, 480–481, 483, 485, 488, 491, 493–495, 497, 500 explanatory 4, 24, 26, 32, 44, 49–50, 85, 134, 164–165, 184–185, 187, 190, 196–198, 203–204, 217, 219–221, 281, 284–285, 305–306, 309–310, 313–316, 319–321, 323, 325–327, 329–330, 348–350, 362, 364–366, 368–369, 390, 392, 449, 451, 480, 485, 498 Fairchild, M. 294, 298 Fales, E. 182, 191 family resemblance 77, 84, 382, 496–498 Fara, D. G. 109 Faucher, L. 364 Fausto-Sterling, A. 376 Feldman, F. 150 feminist 11, 375–378, 380–382, 384, 388–389, 391–392, 413, 416–418, 457, 465 Ferejohn, M. T. 28 Feser, E. 415, 418 Fine, C. 376 Fine, K. 3, 6, 9–10, 28, 57, 79–80, 84, 86, 88, 90, 93–95, 107, 111–120, 124, 126–138, 146,

506

Index 158, 170, 177–178, 203, 209, 211, 222, 240–245, 247, 249–251, 253, 261, 271–272, 295, 297, 305, 308, 310–311, 314–315, 319–322, 325–330, 361, 371, 389–390, 393, 396, 442, 457–458, 462, 466, 481 Finkelman, L. 285 Finnigan, B. 411 Finnis, J. 342, 416 Fisk, M. 182 Fitting, M. 90 Foot, P. 410, 414, 419 Forbes, G. 79, 93, 107, 109, 138, 144, 146, 148–150, 152–153, 173–175, 177, 300 form 2, 8, 19–20, 25–27, 30, 32–38, 43, 69–73, 87, 145–146, 165, 171, 186, 218–219, 297–299, 402–405, 407, 415–418, 487, 489–492, 495–497, 499–500 form of life 495–496, 499 formal 9, 25, 34, 36–37, 68, 70–72, 74, 78, 85, 88, 91, 93, 137, 184, 212, 218–219, 240, 314, 403 Foster, J. 184, 230 Foster-Hanson, E. 230 Frankfurt, H. 334–335, 394 Franklin-Hall, L. R. 158, 284, 286 Franz, N. M. 1, 282, 476 Freddoso, A. 182 Frede, M. 404 Fredrickson, G. M. 363–364 French, S. 202–203, 363 Freund, M. A. 149–150 Fritz, P. 254 Frye, M. 388 full essence 7, 200, 312 function 58, 74–77, 162–163, 265–272, 350–351, 378, 394, 401–406, 410–411, 415–416, 419, 461–462, 468 functional 73, 162, 264–266, 270–271, 333, 336, 350, 378, 389, 393, 402, 405–406, 433, 466 functionalist 266, 269 fundamental kind 165, 212 Fuss, D. 391 Gale, R. M. 61–62 Galluzzo, G. 38 Garfield, J. 411 Garson, J. 162 Geach, P. T. 151 Geddes, P. 376 Gelman, S. A. 94, 200, 227–230 gender 3, 11, 94, 227–229, 234–235, 270–272, 276, 286, 354, 356, 361, 371, 375–384, 388–397, 401, 407, 410, 416–419, 450, 455, 457–459, 465, 467–468, 476–477, 484–485 gender identity 382–384, 397 general essence 74, 76, 79, 156, 190 generic essence 7, 136–137, 242, 305, 308, 312–313

generic essentialism 467 generics 226–227, 229–232, 234–235, 314, 370, 395–396 George, R. 169, 171, 195, 363, 416 Ghiselin, M. T. 280–281, 365 Giannotti, J. 195, 202 Gil-White, F. 28 Giladi, P. 364 Gilbert, M. 54 Gill, M. L. 272, 491 Girgis, S. 416 Glasgow, J. 366, 368 Glazier, M. 137–138, 248, 305, 315 Godfrey-Smith, P. 62 Godman, M. 127, 133, 164–165, 212 Goldblatt, R. 95 Goldin, O. 28 Goldman, A. 416 Goodman, C. 411 Goodman, J. 254 Goodrich, E. S. 89, 95 Goodwin, W. 162 Gopnik, A. 231 Gorman, M. 38, 100, 126, 128–130 Goswick, D. 443, 447 Gracia, J. E. 300 Grandjean, V. 152 Grandy, R. E. 149–150 Grant, T. 145, 147–148, 286 Gratvol, N. 310 Grey, J. 50 Grice, H. P. 445, 450 Griffith, A. 55, 62, 271–272, 282, 284, 286, 305, 347–348, 354–355, 358, 368, 371, 379, 390, 394, 397, 413, 418–419, 450, 476, 479, 482, 485 Griffiths, P. E. 161, 282, 365, 371 Grillo, T. 391 Grossman, R. 79 ground 3, 22, 42, 67, 69–72, 77, 79, 94, 109, 119, 127–129, 132, 134–137, 145–146, 181, 183, 187–188, 197, 202, 204–205, 212, 216, 221, 294–299, 305–316, 319, 323, 403, 405–407, 457, 462, 474, 478, 481–485 grounding problem 294–299, 323 Guala, F. 269–270, 272 Guillaumin, C. 363 Gurian, M. 377 The Gurian Institute 377 Hacker, P. M. S. 500 Hacking, I. 3, 191, 348, 352–353, 450 Haefliger, G. 78 Häggqvist, S. 158 Hakkarainen, J. 157 Hale, B. 94, 100, 127, 132–134 Hallett, G. 81 Hanson, W. H. 107, 230

507

Index Hardimon, M. O. 368 Harman, E. 342 Harré, R. 182 Hartmann, N. 76, 78–80 Harwood, J. 363 Haslam, N. 229, 351–352 Haslanger, S. 3, 228, 230, 235, 343, 360 Hauser, C. 211 Havstad, J. C. 162–164 Hawley, K. 144, 158, 164 Haze, R. M. 162 Hazen, A. 109, 162 Heil, J. 191, 204 Held, V. 417 Hendry, R. F. 162–163, 165 Hennig, W. 277–278 Henry, D. P. 37, 78 Hering, J. 67, 78 Hermida, M. 279 Heyes, C. J. 376–378, 381–382 Heyman, G. D. 234 Hickman, L. 63 higher-order logic 7, 242–245, 248 Hildebrand, D. 77, 79 Hilpinen, R. 264, 272 Hindriks, F. 469 Hintikka, J. 91, 109 Hirèche, S. 205 Hirsch, E. 294 Hirschfeld, L. A. 364 historical essence 6, 280, 371, 378, 392, 397 Hobbes, T. 2, 42–44, 49–50 Hochberg, H. 78 Hoffman, J. 191 Hoicka, E. 231 homeostatic property cluster 158, 164–165, 190, 281–284, 286, 348–350, 367–369, 371, 459 Hommen, D. 157–158 hooks, b. 381 Horowitz, M. 419 Horvath, J. 210 Horwitz, A. 356 Howat, A. 429, 499 Hübner, K. 50 Huebner, B. 456 Hughes, C. 172 Hull, D. L. 276, 280–281, 286, 365 Hulswit, M. 57–59, 63 human kinds 158, 348, 352–354, 362, 407 Hume, D. 71–72, 85, 183, 419, 427–428, 480 Humeanism 182–184, 198 Hurka, T. 95 Hursthouse, R. 410 Husserl, E. 67–72, 74–80, 88, 212, 218 Hüttemann, A. 80 hylomorphism 25, 27, 30, 35, 37–38, 186, 188, 297–298, 302, 404–405

hyperintensional 94, 316 identical 1, 77, 107–109, 112, 114, 128, 134, 138, 144–149, 151–152, 176, 186, 212, 241, 244–245, 268, 271, 291, 293, 297, 300–301, 312, 336–337, 339, 343, 490 identity 1, 4–5, 7–8, 77–80, 84–85, 94, 108–109, 111, 128–129, 131–133, 135, 145–149, 151–153, 159–160, 171, 183, 187–188, 195–196, 199–201, 212–213, 216, 226–227, 234, 240, 242–245, 250, 282–283, 285–286, 299–302, 312–313, 327, 330, 333–334, 336–344, 361, 363, 366, 376, 381–384, 389, 391–394, 396–397, 401, 403–404, 406–407, 414–415, 418–419, 429, 432, 434, 439, 441, 450, 485 identity theory 187 immediate essence 7, 247 impossible world 113–116 indexed operator notation 330 indiscernibility 146, 148–149, 153, 293, 295, 336, 343 individual 1, 4–5, 7, 28, 30–32, 34–35, 37–38, 41–42, 46–49, 68–70, 73, 76–79, 105–106, 109–110, 115, 132, 143–153, 156–157, 159, 161, 165, 169, 181, 184–186, 188–190, 209, 212–213, 222, 227–228, 230, 232, 234–235, 247, 251, 262, 267–268, 271–272, 276, 279–284, 286, 291, 296, 299–302, 307, 328, 347, 353–355, 357, 361, 364–366, 368–369, 376–377, 379–384, 388–397, 401–407, 412–413, 416, 418–419, 430, 434, 441, 450, 457–458, 464, 493, 498 individual essence 1, 30–32, 37–38, 47, 68, 76–78, 109, 144–148, 153, 157, 247, 262, 291, 299–302, 347, 389, 392–394 individuation 27, 32, 38, 50, 109, 119–120, 151–153, 189, 195, 220, 269–270, 276, 291, 293, 295, 297, 299–302, 305, 343, 351, 367, 401, 419, 441–442, 446, 448 inevitability 463–464 inference 72–74, 171, 212, 215–217, 221–222, 229, 283, 320, 322, 325, 327–328, 330, 370, 445, 449, 460 Ingarden, R. 67, 78–80 Ingthorsson, R. D. 191, 195–196, 201 Insel, T. 350 intention 5–6, 37, 77, 234, 262, 264–267, 272, 302, 337–339, 344, 416, 441, 448, 451, 495 intentional 157, 264, 269, 272, 335 intentionality 72, 74, 268–269, 272, 389 interactive kinds 352–354, 464 intersectionality 391, 394 intrinsic essence 161–163, 283–284 intrinsic property 160, 185, 281–283, 364, 366, 458–459

508

Index intrinsicality 4, 6, 11, 85–86, 92, 94, 118, 162, 310 intuition 67, 72, 74–76, 80, 88, 107–109, 112, 118, 127, 149, 160, 171, 174–175, 205, 210–212, 214–215, 217–218, 235, 268, 293–294, 306–307, 339–342, 344, 362, 392–393, 410, 439, 450, 468, 483 intuitionistic 249, 253 Irigaray, L. 379 Isasi, R. 412 Ishiguro, H. 500 iterated essence 248–249 Jackson, F. 450 Jackson, J. P. 364 Jacobs, J. D. 191, 194–196, 200 Jago, M. 127, 153, 210, 218, 300 James, S. 411 James, W. 53–55, 59–60, 62–63, 68, 75, 78, 147, 429, 500 Janssen-Lauret, F. 90, 95, 171 Jaswal, V. K. 234 Jaworska, A. 342 Jeffers, C. 368, 392 Jenkins, K. 378–379, 381–382, 384 Jenner, R. A. 281 Jolley, K. 499 Jones, D. 364 Juengst, E. 411–412 justice 2, 19, 22, 138, 171–172, 230, 235, 271, 281, 286, 298, 358, 388–397, 410, 419, 463–464, 467 Kafer, A. 413 Kaiser, M. I. 203 Kamm, F. 419 Kant, I. 72–73, 84–85, 93, 343, 410, 448, 500 Kaplan, D. 429 Kappes, Y. 305 Kasmier, D. 212 Katzav, J. 202, 205 Kaufman, D. 50 Keil, F. C. 228, 231–232 Keinänen, M. 157 Kendler, K. 348–350, 358 Khalidi, M. A. 1, 55, 150, 158, 162, 164–165, 191, 263, 271, 284, 286, 348, 351, 353–354, 367, 456, 460–461 killing 150, 414, 419 Kim, J. 51 Kimmel, M. 469 Kimpton-Nye, S. 196, 204 kind-membership 4, 181, 188, 190, 216–217, 220 King, J. 272 King, P. 38 Kistler, M. 203–205 Kitcher, P. 60, 161, 278, 367

Kitts, D. B. 281 Kitts, J. 281 Kivy, P. 272 Kluge, A. G. 286 Kment, B. 94, 127, 133, 177, 200, 210 Kneale, M. 92 Knobe, J. 227–228, 231, 234 Koppelman, A. 416 Korman, D. Z. 272 Kornblith, H. 272 Korsgaard, C. 343 Koslicki, K. 28, 55, 78, 94–95, 127, 138, 145–146, 158, 191, 232, 241, 249, 253, 261, 266–267, 272, 286, 295, 297–299, 302, 308–309, 316, 371, 404–405, 407, 467, 499 Kotov, R. 350 Kourany, J. 376 Kovacs, D. M. 95, 212, 253, 302, 330 Kriegel, U. 295, 298 Kripke, S. A. 1–2, 5–6, 9, 59, 91–93, 105–111, 119, 138, 158–163, 169–172, 176, 178, 182, 210–213, 216, 218, 220, 226, 232–235, 244, 280, 393, 433–434, 439–440, 451 Kronfeldner, M. 284–285 Künne, W. 79 Laerke, M. 50 Laimann, J. 348, 353–355 lambda 330 Lane, R. 54, 63 Lange, M. 191, 197, 444 Langton, R. 230, 235 LaPorte, J. 159, 161, 278–280 Latour, B. 469 Lawford-Smith, H. 382 Lawler, J. 231 Lawrence, G. 410 laws of metaphysics 133, 221–222, 313–314 laws of nature 70, 157, 159, 164, 183, 186–187, 194–195, 197–200, 203–205, 220, 313–314, 325–326 Lee, P. 416 Leech, J. 90, 94, 126, 131, 178, 330 Legg, C. 54, 56 Lehmkuhl, D. 199 Leibniz, G. W. 44–47, 49–50, 55, 109, 120, 131, 268, 292–295 Lennox, J. G. 286 Leslie, S.-J. 94, 177, 200, 229–230, 234 Leuenberger, S. 317 Levinson, J. 264, 267–268, 271 Levy, S. R. 229 Lewens, T. 287 Lewis, C. I. 90–91 Lewis, D. K. 53, 61, 93–94, 107, 109, 112–114, 116, 120, 149, 160, 194, 197–198, 204–205,

509

Index 291, 293, 297, 334–335, 338, 340, 343–344, 434, 445–446 Lewy, C. 450 Liao, S. 456 Liebesmann, D. 344 Lierse, C. 191, 195, 201–202 Lin, M. 50 Linnebo, Ø. 242, 246, 249–250 Linsky, B. 107, 128, 159 Lipman, M. 297 List, C. 344 Litland, J. E. 8, 95, 330 Livanios, V. 202 Livingstone-Banks, J. 112, 211, 302, 475, 481 Locke, J. 2, 5, 47–49, 71, 95, 185, 211–214, 218–219, 333–334, 337, 339, 391, 415, 429, 438, 445, 450, 457 Loets, A. 80, 305, 393, 400, 412, 414 logic of essence 3, 9, 21, 93–94, 125, 136–137, 240–245, 247–251, 253, 310, 328, 330 logical form 87, 309, 311, 492, 500 logical law 217 logical necessity 86–87, 133, 184, 188, 433 logical possibility 191 logical truth 107, 177, 215, 217, 322, 428, 430, 432, 444 logos 20 Lombrozo, T. 231 Long, R. 501 Look, B. 74, 170, 447 Lopez, K. 416 Lorde, A. 381, 417 Loux, M. J. 28, 182, 191, 404 Love, A. C. 3 Lovett, A. 344 Lowe, E. J. 80, 94–95, 127, 132, 144, 149, 151–152, 156, 158–159, 182, 185, 191, 196, 213–215, 218–219, 261, 300–301, 316, 319, 322–323, 325–328, 330 Luckman, T. 469 Ludwig, D. 278, 284, 286 Macarthur, D. 60 MacBride, F. 435 Macdonald, C. 191 Mach, E. 68, 70, 79 Mácha, J. 79 Machery, E. 364 Mackie, P. 94, 100, 110, 127, 144, 146 Madden, E. H. 182 Magidor, O. 344 Magnus, P. D. 284, 286 Maher, B. 412 Malebranche, N. 376 Malink, M. 78–79, 150, 153, 190, 222, 401, 416, 419, 499–500 Mallon, Ron 158, 271–272, 276, 281

Mallozzi, A. 80, 127, 133, 164–165, 200 Marcus, R. B. 2, 84, 90–91, 107–108, 120, 138 Mares, E. 95, 114 marginalization 389, 393, 396 Markman, E. M. 228, 230, 234 Marmodoro, A. 196, 205, 404 Marques, T. 469 Martin, C. B. 50, 138, 182, 189, 191, 200 Martland, T. R. 63 Mason, R. 271–272, 388, 390–391 material 5–6, 25–26, 30, 32–38, 56, 59, 68, 70–71, 77, 86, 88, 115, 129, 132, 134, 136, 148, 152–153, 169, 173, 183, 204, 211, 213, 248, 264, 268–269, 293, 298–299, 324, 337, 341–342, 365, 384, 402–407, 430, 439, 448, 456 material constitution 152–153, 268, 342 Mates, B. 51 mathematics 73, 80, 89, 219, 249–250, 319, 474 matter 5–6, 8, 20, 25–27, 30, 32–38, 43, 48, 59, 145, 151–152, 171–177, 183, 186, 191, 293, 295, 297–298, 402–406, 416, 451 Maudlin, T. 184, 197 Maurer, A. 38 Maurin, A. 223 Mayden, R. L. 277 Mayr, E. 277, 280, 366 McDaniel, K. 120, 310 McDowell, J. 500 McGinn, C. 170 McKay, T. 174 McKinnon, R. 398 McKitrick, J. 382–383 McMahan, J. 344, 414, 419 mediate essence 7–8, 127, 129–131 Medin, D. L. 227 Meinong, A. 67, 69, 71, 77, 79 Meister, S. 28 mental disorder 347–351, 353, 355, 358 metaphysical explanation 136–137, 212, 305–306, 309, 313–316, 474, 481, 483, 485 metaphysical modality 9, 91, 120, 124, 132, 159, 177–178, 213, 261, 319–320, 330 metaphysical necessity 79, 86, 90, 92–93, 124, 127, 129, 132–137, 159, 169, 184, 198, 209–210, 222, 249, 313, 319–321, 323, 325–330, 433, 443 metaphysical possibility 132, 158, 191, 210, 216, 298, 319, 460 Meylan, A. 80 Meynell, L. 376 Michels, R. 78–80, 132, 138, 204, 499 microstructural essentialism 162–164 microstructure 158–160, 162–164, 213, 282, 311, 449 Mikkola, M. 376, 392 Mill, J. S. 89, 91, 157

510

Index Miller, A. 472 Miller, K. 306 Miller, T.-G. 247 Millett, K. 469 Millikan, R. G. 161, 282 Millum, J. 419 mind-dependence 11, 50, 55, 61, 263, 271, 353, 355, 390, 437, 446, 461, 464, 467–468, 485 mind-independence 4, 11, 30, 50, 59, 75, 157, 212, 263, 271, 348, 355, 390, 437, 439–440, 445–451, 458, 463, 465, 467–468 Minelli, A. 3 Misak, C. 57, 59 Mishler, B. D. 278 modal conception of essence 9–10, 22, 37, 86, 90–91, 93–94, 106, 108–109, 111–113, 115–120, 124–127, 138, 158, 160, 203, 209–211, 216, 298, 302, 306, 309–310, 319–321, 323, 325–330, 393, 433, 442, 483 modal logic 2, 9, 84, 88, 90–91, 93, 106–107, 112–113, 116, 118, 120, 124, 128, 131, 134, 138, 240, 244, 250–251, 429–431, 433–434 modal profile 1, 294–295, 298–299, 319 modal property 110, 116, 169, 183, 294 modality 1, 3, 9–10, 22, 31, 50, 67, 69–72, 77, 79, 86, 90–94, 105, 108–109, 111–113, 119–120, 124, 127, 131–132, 138, 143, 153, 159, 170, 177–178, 190, 204–205, 210, 213, 222, 247, 261, 313, 319–320, 330, 427–434, 442–444, 448–449, 457, 481, 483–484 Modrak, D. K. W. 28 Mohanty, J. N. 212 Molnar, G. 191, 194–195, 199–200 Mondadori, F. 51 Moore, G. E. 84–87, 89, 92, 95, 107, 410 moral responsibility 414–415 morality 53, 270, 415, 419, 485 Morein-Zamir, S. 412 Morris, J. 413 Morrison, S. M. 162 Moschella, M. 416 Moss, S. 28, 344 Mounce, H. 60–61, 499–500 Mulligan, K. 95, 212 Mumford, S. 191, 194, 196–197, 200 Murphey, M. G. 57–58 Murphy, D. 348–351 Nagel, T. 338, 413 Nanamoli, B. 411 Nanay, B. 3 natural kinds 3, 10–11, 28, 55, 59, 63, 78, 91–93, 95, 110, 120, 153, 156–165, 181, 185–191, 203, 205, 209, 219–222, 227, 232–235, 261, 263, 271, 276, 278, 280–286, 296, 302, 305, 330, 347–348, 351–352, 355–358, 367–368,

371, 377, 384, 390, 404–405, 407, 419, 442, 450, 480, 484 natural necessity 70, 188, 326 natural property 366, 368 natural science 2, 70, 88, 319, 325, 352–353, 355, 366, 458 nature 1–2, 4–6, 8, 11, 31, 35–38, 41–44, 46, 48–50, 56, 61–63, 69–71, 73–74, 84–86, 88–89, 95, 111–113, 116, 126–127, 132–133, 135, 161, 170, 183–188, 194–205, 211, 213–215, 217, 220–221, 230, 240–241, 244–247, 249–251, 261–262, 264, 266, 268–269, 284–285, 307–308, 314–316, 322–324, 326–330, 333–334, 341–342, 361, 375–376, 389–391, 393, 395–396, 406, 410–412, 416–419, 434, 457–460, 462–464, 467–468, 474 necessary a posteriori 160, 169, 212–213, 216, 440, 450 necessitism 107, 245 necessity 9–10, 22, 31, 41, 69–74, 79, 84–95, 105–106, 108–112, 116–118, 124–129, 131–138, 144–145, 159–160, 164, 169–170, 172–175, 177–178, 183–184, 197–198, 200–201, 204–205, 209–210, 212–213, 215–216, 245–246, 249, 251, 253, 279, 292, 298–300, 302, 310–315, 319–330, 394, 427–434, 437–445, 450–451, 457–460, 464, 466, 474, 483–485 necessity of identity 109, 160, 216 Needham, P. 161–164 neo-Aristotelian 84, 93, 138, 165, 218, 261–264, 276, 283, 297, 316 neo-Aristotelian metaphysics 316 Neufeld, E. 58, 228–229, 233–234 neutral monism 68, 188 Newlands, S. 50 Newman, G. E. 227–228 Nichols, S. 228 Noddings, N. 417 Nolan, D. 94, 316 nominal definition 1–2, 57, 63, 85, 213, 215–217 nomological necessitarianism 187 nomological necessity 159 non-modal conception of essence 22, 28, 31, 38, 50, 78–79, 93, 95, 111, 113, 120, 124–125, 127–131, 133, 135–138, 143, 153, 158, 165, 169, 183, 195, 200–201, 203–205, 209, 212, 222, 241, 246, 248–249, 253, 298–299, 302, 305, 311, 316, 319–321, 323, 325–330, 389–390, 393, 397, 427, 440, 442, 448, 457, 473, 484 non-substantial universal 325 nonsense 500 Noonan, H. 94, 126, 340, 446 Norlock, K. 417 Normore, C. 50

511

Index Norton, J. 306 Noyes, A. 228, 231 Nozick, R. 335, 342 numerical identity 145, 171, 336–337, 343 Nussbaum, M. 410, 417 Nutting, E. S. 247 O’Conaill, D. 317 O’Leary-Hawthorne, J. 161, 278–279, 282–283 objectual essence 136, 262 Okasha, S. 94, 127, 165, 189, 191, 200 Olson, E. 295, 297, 336, 338, 340 ontological dependence 31, 93, 191, 308–309, 311 ontological independence 405 ontology 9, 23, 68, 74, 109, 184–186, 188–190, 199, 205, 220, 267, 269–270, 305, 338, 348, 352, 357, 404, 407 oppression 3, 235, 366, 388–389, 391–392, 395–397, 412, 418 Ortega y Casset, J. 78 Ortony, A. 227 Oshana, M. 394 ostensive definition 493–495 ousia 20–21, 26, 78 Owen, D. 51 Pagano, E. 469 pan-dispositionalism 187 Paoletti, M. P. 205 Pap, A. 80, 450 Papineau, D. 127, 133, 164–165, 212 Parens, E. 412 Parfit, D. 149, 336–338, 340 Parsons, C. 250 Parsons, T. 107, 435 partial essence 7 Pasnau, R. 38, 50 Passinsky, A. 28, 150, 153, 300, 305, 347 Patterson, R. 28 Paul, L. A. 2, 94, 107, 163, 266, 371 Peacocke, C. 211, 222 Pedroso, M. 172 Peirce, C. S. 53–60, 62–63 Persad, G. 413 persistence 1, 23, 41, 50, 109, 119–120, 138, 153, 204, 276, 282, 285, 291–297, 299–302, 305, 323, 325, 328, 336–337, 343, 347, 352, 393, 401, 403–404, 406, 419, 446–448, 462 persistence conditions 276, 291–297, 300, 323, 325, 336–337, 401, 406, 446–448 person 3–4, 6–7, 20, 69, 76–78, 80, 136, 152–153, 169–172, 213, 226–231, 234–235, 278, 297, 302, 307, 312, 315, 333–344, 352, 354–355, 361–362, 364, 366, 369–370, 379–381, 383, 388–397, 411–417, 419, 445–449, 459, 464, 493–494, 497, 499

personal 77, 138, 333–334, 336–341, 344, 361, 378, 380, 411, 414 personal identity 77, 333–334, 336–341, 344, 361, 414 Pettit, P. 271, 344 Pfänder, A. 67, 80 Phemister, P. 51 phenomenological 67, 69, 71–75, 77, 79, 162, 200, 212, 222, 381 phenomenologist 67–80, 86, 88–89 phenomenology 11, 67, 70–71, 77–78, 80, 95 Philip, K. 95 physical necessity 164, 432–434 picturing 489, 491 Piotrowska, M. 285 Plantinga, A. 93, 100, 107, 109, 138, 143 Plato 2, 19–22, 24, 32, 115, 151, 248, 475 pluralism 165, 268, 272, 352 plurals 229, 246 Pöll, W. 78 Pollock, J. L. 107 populations 277–278, 280, 282, 365–367 positivism 182 possibility 37, 58, 71, 91, 106, 132, 144–148, 157–158, 163–164, 169, 171, 175–177, 190–191, 204, 210, 216, 222, 251, 285, 319, 348, 432, 460, 487–492, 494–496, 499–500 possible 1–2, 27, 30–32, 36–37, 86, 91–93, 105–106, 108, 113–116, 120, 136, 144, 146–152, 157–160, 163–165, 172–178, 181, 183–184, 187, 191, 196–199, 201, 204–205, 233, 249–251, 279–281, 283–284, 301, 438–439, 442–444, 488–491 potential 56, 94, 157, 162, 172, 199, 203, 205, 228, 230–231, 234–235, 245, 249, 279, 282, 284–285, 296, 305, 308, 344, 348–350, 352–354, 357, 416, 443–444, 467 potentiality 9, 35, 203, 205, 342 Pouivet, R. 81 power 24, 26, 45, 145, 157, 183–184, 187, 191, 194, 196, 198–205, 219, 221, 263, 270, 324, 329, 355, 390, 417, 463, 482 pragmatic 53, 55–57, 59, 63, 213, 499 pragmatism 53–55, 57, 59–63, 95, 499 pragmatist 53–55, 57, 59–60, 62–63, 73–74, 182 Prasada, S. 228 predicational essence 7–8, 21, 253, 262–263, 308 Preston, A. 95 Preston, B. 265, 272 Price, H. 53, 55, 218 primitive substance theory 186, 188–189 Prince, S. 28, 339 principle of unity 218–219, 402–404, 407 Prior, A. 28, 115, 204, 344 proof theory 240, 244 property clusters 158, 284, 349, 358, 369, 375, 459

512

Index property dualism 199–201, 205 proposition 43, 70–72, 80, 106, 132–133, 135, 212, 215, 241, 243–244, 248–249, 251, 311, 319–323, 325–327, 330, 396, 445, 488–491, 498–500 propositional 90–91, 134–135, 147, 250, 461 propositions 68–71, 73–74, 77, 79, 92, 132–133, 135, 144, 184, 212, 226, 240–241, 243, 247, 249, 251, 313, 321–322, 328, 330, 458, 474, 487–492, 499–500 Pruss, A. R. 419 Psillos, S. 158, 205 psychiatric kinds 286, 347–353, 355–358, 419, 464, 468 psychological essentialism 58–59, 84, 94, 227, 229, 231 Pust, J. 171 Putnam, H. 1, 53, 59, 63, 91–92, 110, 158–159, 161–164, 182, 211–212, 218, 232–233, 235, 280, 361, 419, 439–440, 447, 450–451 qualitative identity 195, 336, 343 quantified modal logic 90–91, 93, 106, 112, 128, 131, 138, 244, 429–431, 433–434 quantifying in 430 quiddity 30, 195–196, 199–200, 202 quiditas 30 Quine, W. V. O. 2, 53–54, 63, 84, 89–92, 95, 107–108, 111, 120, 169, 178, 427–434, 437–438, 444–445, 450–451 Rabern, B. 134 Rabin, G. O. 134 race 3, 11, 94, 227–229, 270–272, 276, 286, 361–371, 379, 381, 388–395, 397, 401, 407, 410, 450, 455, 464–465, 467–468, 476–477, 485, 493–494 Rachels, J. 414 racial 233, 235, 271, 361–371, 389–391, 456, 464 racial essentialism 271, 361–364, 366–367, 370–371, 464 racial skepticism 366 racialism 363 racism 269, 389, 456, 461 Rashed, M. 357 Raven, M. J. 28, 95, 137, 191, 205, 271–272, 286, 302, 308, 371, 461, 467, 499 Rayo, A. 128, 131, 242 Rea, M. 220, 295, 297, 445 real definition 1–2, 22, 25, 27, 50, 53, 57–58, 79, 93, 133, 137, 209–211, 213–219, 221–222, 261, 306–307, 322, 397, 403, 458, 465–466 realism 54, 63, 109, 149, 157–158, 181–182, 184–185, 189–190, 263, 271, 281, 285, 302, 366, 442, 446, 449, 472–474, 477, 481–485, 498

reduction 67, 79, 94, 132, 137, 186, 249, 253, 306–309, 313, 319, 321, 323, 325, 327, 329, 348, 352, 428 reductionist 272, 350–352 reductive 10, 94, 132, 135, 309, 312–314, 316, 337, 344, 495 reference 6, 11, 23, 25, 33–34, 43, 46, 49, 56, 71, 80, 91–93, 117, 134, 153, 157, 169, 172, 182, 197, 232, 235, 280, 311, 321–322, 324, 327, 334, 343, 364, 367, 378, 432, 456–457, 465, 495, 499 referential 91, 93, 429 Reid, T. 338 Reinach, A. 67–72, 77–80 relational essence 80, 160–161, 371, 383 representation 226–228, 324–325, 328–329, 380, 447, 451, 456, 467, 491, 500 responsibility 343, 411, 413–415, 419 resurrection 2 Reynolds, A. 54, 63 Rhodes, M. 228, 230, 235 Richardson, K. 314–316, 418 Ricketts, T. 491, 500 Riedener, S. 344 Rieppel, O. 282, 286 rigid designation 91, 159–160, 212 Ritchie, K. 253, 285–286, 371, 389 Roberts, J. T. 197, 231 Robertson Ishii, T. 63, 143–144, 153, 200, 249 Robertson, T. 133, 138 Robinson, D. 339–340 Roca-Royes, S. 120, 300 Rogan, M. 392 Rohrbaugh, G. 173, 175–176 Romero, C. 94, 127, 137 Root, M. 366 Rorty, R. 53, 55–61, 63 Rosario, E. 272, 276, 281, 285–286 Rose, D. 228 Rosen, G. 134, 137, 306–307 Rosen, J. 28 Rosenkrantz, G. S. 145, 301 Ross, W. D. 410 Rothbart, M. 229 Rothschild, L. 229 Ruddick, S. 417 rule 22, 54, 56, 72–73, 87, 130, 151, 241, 243–245, 253, 268–272, 311, 322, 328, 344, 363, 377, 380, 394, 407, 437–438, 442, 444, 447, 449–451, 464, 490–492, 494, 496, 498 Russell, B. 68–69, 71, 85–87, 89, 92, 95, 178, 242, 445, 489 S5 9, 90–91, 93, 135, 177–178, 250, 443 Sadler, J. 357 Sahakian, B. 412 Salerno, J. 113–114, 127

513

Index Salmón, N. 172–176 Saul, J. M. 108, 158, 169, 182, 235 Scanlon, T. 335, 342 Schaffer, J. 116, 305, 316, 457 Scheler, M. 67–70, 72–80 Scheman, N. 408 Schilpp, P. A. 95 Schlegel, R. J. 227 Schlick, M. 88–89, 92 Schnieder, B. 128, 295 Schriner, K. 413 Schroer, R. 204 Schwartz, P. 412 Schwartz, S. P. 159 science 2–3, 25, 54, 61, 63, 70–71, 73–75, 80, 87–88, 91, 94–95, 138, 158, 182, 184–189, 191, 202–203, 218, 261–263, 276, 281, 283–286, 305, 316, 319, 325, 347–350, 352–353, 355–357, 362, 366–367, 370, 383, 416, 458, 483 scientific 2–3, 22, 24–25, 58–59, 63, 67, 70, 72–73, 79, 85, 87, 116, 157, 159, 165, 181–185, 187–189, 191, 202–203, 205, 211, 213, 218, 221, 278, 280–281, 283, 285–286, 305, 313, 325–326, 328, 347, 352, 354–356, 358, 367, 439, 456, 495 Scotch, R. 413 Searle, J. R. 92, 268, 271–272, 400, 445 Segal, A. 310 Segelberg, I. 78 Seibt, J. 54 Seifert, J. 78 Seigfried, C. H. 53, 61 self-identification 375, 377, 379–382, 384 semantics 90–91, 93, 106–107, 109, 113, 115, 158, 163, 182, 229, 232–234, 240, 250–253, 314, 442, 445 serious essentialism 319 sex 11, 94, 271, 277, 286, 354, 356, 371, 375–381, 383–384, 389–394, 397, 410, 413, 415–416, 418–419, 450, 457, 459, 465, 467–468, 476–477, 484–485 sex/gender distinction 354, 375–381, 384 sexual orientation 389–394, 397, 410 Sgaravatti, D. 214 Shafer-Landau, R. 215, 218 Shea, William M. 64 Sheehan, L. 357 Shelby, T. 392 Shieh, S. 435 Shoemaker, S. 191, 196, 297, 302, 344 Shook, J. R. 54 Short, T. L. 59, 63 Shumener, E. 343 Sidelle, A. 95, 205, 211, 295, 298, 300 Sider, T. 119, 160, 344, 442, 445 Siderits, M. 411

Siipi, H. 285 Silverman, A. 28 Simpson, G. G. 277 Singer, P. 335, 342, 344 skepticism 3, 10, 21, 48, 55, 59, 63, 90, 95, 108, 111, 138, 211, 316, 366, 382, 384, 416, 427–428, 433–434, 442, 467 Skiles, A. 78, 94, 128, 131, 133, 137, 178, 242–243, 245, 312–313, 315–316, 485 Skull, A. 356 Slater, M. H. 283–284, 286, 367 Sleeper, R. W. 54, 60 Slote, M. 404 Smart, B. 204–205 Smith, B. 78, 80 Smith, D. C. 202 Smith, D. L. 234, 364 Snowden, P. F. 338, 341 Soames, S. 159 Sober, E. 280 social construction 63, 271, 286, 355, 358, 368, 371, 397, 419, 437, 449, 455–457, 459, 461, 463–465, 467–468, 479, 484 social essentialism 3, 465–467 social justice 235, 271, 286, 358, 388–397, 419, 463–464, 467 social kinds 158, 217, 229, 235, 271, 296, 390–391, 393, 401, 407, 418, 457, 459–460, 462, 466–468, 480, 484–485 social position 369, 379, 381, 391, 394, 407 social role 368, 379, 383, 392–394, 396, 407, 477 Sosa, E. 210, 294, 298 sparse property 112–113, 116–117 Specht, K. 81 species 3, 6, 20–24, 27–28, 30–33, 38, 42, 44, 47–50, 68, 71, 88–89, 94–95, 119–120, 144, 153, 157, 161–162, 172, 178, 185–191, 219, 232–234, 276–286, 305, 309, 325, 342, 349, 358, 364–369, 371, 391, 412, 419, 446–447 species essence 30–32, 38, 187, 281, 283–285 species-as-individuals thesis 281–282 Spelman, E. 391, 394, 459 Spencer, J. 173 Spencer, Q. 367 Spiegelberg, H. 78 Spinelli, N. 78, 80 Stalnaker, R. 109, 113 Stanford, P. K. 11, 153, 161 Stebbing, S. 84, 88–89, 95 Stein, E. 67–69, 78–79 Stephanou, Y. 107 Sterelny, K. 365 Sterken, R. 235 Stevens, K. 377 Steward, S. 114–115, 310 Stich, S. P. 366 Stock, K. 382

514

Index Stoljar, N. 228, 272, 281, 285–286 Stone, A. 376, 417, 419, 465 Strawson, G. 150, 191, 204, 450 Strawson, P. F. 300, 445 strong essentialism 458–459, 463, 465, 467 Stroud, B. 450, 500 Stubblefield, A. 469 substance 2, 9, 22–23, 26, 30–38, 42, 44, 46–47, 50, 56, 68, 93, 105, 138, 151–152, 158, 160, 163–164, 181, 185–191, 197, 211–212, 214, 216, 226, 229, 232–233, 296, 380, 393, 402–407, 441, 451, 487–488, 500 substantial universal 158, 325 substratum theory 186, 188 Sullivan, J. 359 Sullivan, M. 302, 305, 309–311 Sundstrom, R. 456 super-explanatory properties 164 superexplanatory 221 Tabb, K. 349, 351–352 Tahko, T. E. 78, 100, 143, 153, 181, 189 Taieb, H. 81 Talbot, M. 412 Talisse, R. 54 Tannenbaum, J. 285, 342, 354, 356, 358 taxa 276–277, 279–282, 286, 349, 371 taxon 276–279, 282–284, 286 taxonomic 277 Taylor Merleau, C. 456 Taylor, C. 376 Taylor, J. H. 392, 395 Taylor, M. 195, 202 Taylor, P. 229 Teitel, T. 9, 94, 134–135, 245 Tekin, S. 359 Templeton, A. R. 365 temporal 67, 70, 75, 144, 198, 276, 291–292, 299–300, 323, 446–447, 500 Tennant, N. 444, 450 Thébault, K. 205 Thomasson, A. 265–267, 269–270 Thompson, J. A. 376 Thompson, M. 419 Thompson, N. 223 Thomson, J. J. 58, 338–339, 344, 376, 400 ti ēn einai 2 Tillman, C. 173, 247 time 1, 23, 41, 149, 151–152, 174–175, 181, 199–202, 213, 227, 265, 267, 280–283, 291–293, 296, 300, 302, 333–338, 347, 463–464 Tirrell, L. 229, 233 Tobia, K. P. 227–228 Tobin, E. 1, 11, 156–158, 162–164, 296 Tooley, M. 187, 197, 414

Torza, A. 78–79, 138, 143, 153, 165, 209, 222, 249, 253, 278, 282, 302, 306, 330, 390, 434, 441, 451, 473, 483 transubstantiation 2, 187, 191 Tremain, S. 413 Trogdon, K. 295, 307 Truth, S. 46, 106, 321, 326, 400, 417 Tsou, J. 356 Tugby, M. 196–197, 201–202, 205 Tuomela, R. 272, 344 Tweedale, M. 182 twin earth 158–159, 163–164, 232, 235, 439 Tworek, C. M. 230 Uebel, T. 86 understanding 2, 10, 22, 43, 47, 54–56, 59, 62–63, 68, 72, 77, 80, 160, 171, 200, 203, 210–211, 213–215, 217, 229, 241, 277, 301, 313–314, 333, 335, 339, 349, 351, 354–356, 361, 370–371, 379, 405, 412, 417, 429, 431, 438–439, 445, 450, 464, 467, 479, 484, 490–493, 495–496 Unger, P. 119, 338, 344 uniessence 401, 407 unity 28, 32, 41, 50, 76–77, 165, 189, 218–219, 306–307, 314, 316, 392, 394, 397, 401–407, 466, 499 universal 20–24, 27, 35, 38, 57–58, 68, 70, 78, 106, 113, 120, 144, 156–158, 181, 184–186, 188–190, 198, 213, 218, 230, 244–245, 281, 300, 314, 325, 364, 377, 381, 391–392, 403–404, 458, 460, 465, 497–498 unjust 3, 59, 411, 417, 463, 467 unreal 263 UPIAS 413 Vaidya, A. 55, 94, 137, 210–211, 214, 220, 286, 305, 368, 371, 380, 394, 443, 450, 499 value 2, 61, 63, 69–72, 74, 77–80, 85–86, 105, 112, 186, 199, 228, 249, 269, 286, 315, 343–344, 347–348, 355–358, 376, 379, 410–413, 415, 417, 419, 429–430, 472–475, 477, 485 van Anders, S. M. 377, 379–381, 384 van Brakel, J. 161 van Cleve, J. 443–444 van Inwagen, P. 172 Van Valen, L. 277 VandeWall, H. 161 Varzi, A. C. 109, 120, 302 Vasilyeva, N. 231 Veatch, H. B. 38 Vermaas, P. 272, 296 Vetter, B. 9, 194, 196–198, 200 Viljanen, V. 50 virtue 19, 410–411, 416–419, 496 virtue theory 410–411, 416

515

Index Vlastos, G. 28 Vogt, L. 137, 250 Wagner, G. P. 282 Wakefield, J. 356 Wallace, J. R. 150 Wallner, M. 55, 94, 137, 210–212, 214–215, 217–218, 220, 286, 305, 368, 371, 380, 394, 443, 450, 499 Walsh, D. 191, 284 Walters, L. 120, 297 Walton, G. M. 234 Wang, J. 109 Ward, T. 50 Warren, J. 445, 450 Wasserman, D. 413 Wasserman, R. 295, 297 Waters, C. K. 3 Waxman, S. R. 234 weak essentialism 458, 460, 463, 467 Weisberg, M. 163 Weiskopf, D. 359 Wendell, S. 413 Werner, J. 94–95, 135, 245 Wertheimer, A. 413 Wetzel, L. 152 White, S. 364 Wierzbicka, A. 234 Wiggins, D. 93, 150–152, 191, 200 Wikforss, A. 158, 233

Wildman, N. 94, 100, 116–117, 127, 135 Wilkins, J. S. 277–278, 286 Williams, B. 344, 410 Williams, D. C. 68, 78–79 Williams, N. E. 161, 195, 197, 199–200 Williams, P. 411 Williamson, T. 107, 134, 177, 211, 245 Wilson, Jack 3 Wilson, Jessica 241, 450 Wilson, R. A. 3, 281–282, 349 Winsor, M. P. 286 Wippel, J. 38 Witt, C. 28, 219, 271, 283, 300, 378 Wittgenstein, L. 53, 63, 71–73, 75–80, 84, 86–89, 92, 95, 382, 487–500 Wittig, M. 380 Wodak, D. 230, 235 Woolgar, S. 469 Woollard, F. 414 Wright, C. 443, 450 Yablo, S. 249–250, 294, 446 Yates, D. 202–204 Young, I. M. 388, 392, 413 Zachar, P. 348–350, 357–358 Zack, N. 366, 466 Zalta, E. N. 94, 107, 128, 130–131 Zimmerman, D. 171, 295, 297 Zylstra, J. 94, 127, 137, 178, 200, 246

516