Kant and Animals 0198859910, 9780198859918

Citation preview

Kant and Animals

Kant and Animals Edited by

J O H N  J.  C A L L A NA N and

LU C Y A L L A I S

1

1 Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © the several contributors 2020 The moral rights of the authors have been asserted First Edition published in 2020 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2020930611 ISBN 978–0–19–885991–8 Printed and bound in Great Britain by Clays Ltd, Elcograf S.p.A. Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

Contents Acknowledgements List of Contributors Notes on Sources and Translations

Introduction: Kant and Other Animals: An Overview Lucy Allais and John J. Callanan

vii ix xi

1

I . T H E O R E T IC A L P H I L O S O P H Y 1. The Comparison of Animals John J. Callanan

19

2. Animals and Objectivity Colin McLear

42

3. What Do Animals See? Intentionality, Objects, and Kantian Nonconceptualism Sacha Golob 4. Kant on Nonhuman Animals and God Ina Goy

66 89

5. Animality in Kant’s Theory of Human Nature David Baumeister

105

6. Kant on Evolution: A Re-evaluation Alix Cohen

123

I I . P R AC T IC A L P H I L O S O P H Y 7. Directionality and Virtuous Ends Arthur Ripstein and Sergio Tenenbaum

139

8. Kant and Moral Responsibility for Animals Helga Varden

157

9. What Do We Owe to Animals? Kant on Non-Intrinsic Value Carol Hay

176

vi

Contents

10. Against the Construction of Animal Ethical Standing Jon Garthoff

191

11. Hope and Despair at the Kantian Chicken Factory: Moral Arguments about Making a Difference Andrew Chignell

213

Bibliography Index

239 257

Acknowledgements First and foremost, we would like to thank our contributors first for their enthusiasm and secondly for their patience. Many of these papers were originally presented at a workshop on Kant and Animals at the University of Witwatersrand in South Africa in 2013, organized by Lucy Allais and John Callanan, with much help from Alnica Visser. During the production process for the volume, Max Edwards provided extremely valuable feedback and help with editing the papers. Finally, Peter Momtchiloff has been an exemplarily encouraging editor and we are hugely grateful for his support and indeed that of the entire OUP team.

List of Contributors Lucy Allais is Professor of Philosophy at the University of the Witwatersrand, Johannesburg and Henry Allison Chair of the History of Philosophy at the University of California, San Diego. She has a book on Kant’s transcendental idealism, Manifest Reality (OUP 2015) and a number of articles on Kant’s theoretical philosophy, as well as papers on forgiveness and some other topics in ethics. David Baumeister  is Assistant Professor of Philosophy at Seton Hill University. His research concerns Kant's anthropology and ethics, environmental philosophy, continental philosophy, and philosophies of race and gender. John J. Callanan is Senior Lecturer in Philosophy at King’s College London. His research focuses upon Kant’s theoretical and practical philosophy as well as various other Early Modern philosophers. Andrew Chignell  is a Laurance  S.  Rockefeller Professor, teaching in Religion and Philosophy at Princeton University. His work to date focuses on Kant and other modern European philosophers, philosophy of religion, the moral psychology of hope and despair, aesthetics, and the ethics of belief. Alix Cohen is Reader in Philosophy at the University of Edinburgh. Her research focuses primarily on Kant, with particular interest in his account of anthropology, epistemology and feeling. Jon Garthoff is an Associate Professor in the Department of Philosophy at the University of Tennessee Knoxville. His research focuses primarily on ethical theory and political philosophy. Sacha Golob is Senior Lecturer in Philosophy at King’s College London. He is the author of Heidegger on Concepts, Normativity, and Freedom (CUP 2014) and the co-editor of The Cambridge History of Moral Philosophy (CUP 2017). Ina Goy is Senior Lecturer in Philosophy at the University of Tubingen. Her research primarily concerns the philosophy of Aristotle and Kant. She is author of the Kants Theorie der Biologie (De Gruyter 2017) and co-editor of the volume Kant’s Theory of Biology (De Gruyter 2014). Carol Hay  is Associate Professor in the Department of Philosophy at the University of Massachusetts Lowell. Her academic work focuses primarily on issues in analytic feminism, liberal social and political philosophy, oppression studies, and Kantian ethics. She is the author of Kantianism, Liberalism, and Feminism: Resisting Oppression (Palgrave 2013). Colin McLear is Associate Professor in the Department of Philosophy at the University of Nebraska-Lincoln. He specialises in early modern philosophy, particularly Kant, and the

x

List of Contributors

philosophy of mind. Representative publications include ‘Kant on Animal Consciousness’ (Philosophers’ Imprint), ‘Two Kinds of Unity in the Critique of Pure Reason’ (Journal of the History of Philosophy), and ‘Kant on Perceptual Content’ (Mind). Arthur Ripstein is Professor of Law and Philosophy at the University of Toronto. He is the author of Private Wrongs (Harvard 2016), Force and Freedom: Kant’s Legal and Political Philosophy (Harvard 2009), and Equality, Responsibility and the Law (Cambridge 1999). Sergio Tenenbaum  is Professor of Philosophy at the University of Toronto. He is the author of Appearances of the Good: An Essay on the Nature of Practical Reason, (CUP 2007) and editor of Desire, Good, and Practical Reason (OUP 2010). Helga Varden is Associate Professor in Philosophy and in Gender Studies at the University of Illinois. Her main research interests are in Kant’s practical philosophy as well as legal, political, feminist philosophy, and the philosophy of sex and love. She is the author of A Kantian Theory of Sexuality (OUP forthcoming).

Notes on Sources and Translations Citations for the Critique of Pure Reason employ the standard A/B format. Other primary texts are cited by the volume and page number in the Academy Edition, Gesammelte Schriften (Berlin: Walter de Gruyter and predecessors, 1900–), and are accompanied either by a short title or an abbreviation (see list below). Unless otherwise indicated, all translations are from the Cambridge Edition of the Works of Immanuel Kant (1992–). AB AC AF AM Anth. AP BL Con. Corr. CPJ CPrR Det. DWL EC EM EV FS G ID IUH JL MD MFNS MH ML1 ML2 MM MMr MV MVo ODR Ped.

Anthropology Busolt (in Kant 2012) Anthropology Collins (in Kant 2012) Anthropology Friedländer (in Kant 2012) Anthropology Mrongovius (in Kant 2012) Anthropology from a Pragmatic Point of View (in Kant 2007) Anthropology Pillau (in Kant 2012) Blomberg Logic (in Kant 1992) Conjectural beginning of human history (in Kant 2007) Correspondence (in Kant 1999) Critique of the Power of Judgment (in Kant 2000) Critique of Practical Reason (in Kant 1996a) Determination of the Concept of the Human Race (in Kant 2007) Dohna-Wundlacken Logic (in Kant 1992) Ethics Collins (in Kant 1997b) Ethics Mrongovius (in Kant 1997b) Ethics Vigilantius (in Kant 1997b) The False Subtlety of the Four Syllogistic Figures (in Kant 2002a) Groundwork of the Metaphysics of Morals (in Kant 1996a) Concerning the Form and Principles of the Sensible and Intelligible World—‘Inaugural Dissertation’ (in Kant 2002a) Idea for a Universal History (in Kant 2007) Jäsche Logic (in Kant 1992) Metaphysik Dohna (in Kant 1997a) Metaphysical Foundations of Natural Science (in Kant 2002b) Metaphysik Herder (in Kant 1997a) Metaphysik L1 (in Kant 1997a) Metaphysik L2 (in Kant 1997a) The Metaphysics of Morals (in Kant 1996a) Metaphysik Mrongovius (in Kant 1997a) Metaphysik Vigilantius (in Kant 1997a) Metaphysik Volckmann (in Kant 1997a) Of the different races of human beings (in Kant 2007) Lectures on Pedagogy (in Kant 2007)

xii

Notes on Sources and Translations

PG Prol. R Rel. RevHerder RevMoscati TP VL WE

Physical Geography (in Kant 2012) Prolegomena to Any Future Metaphysics That Will be Able to Come Forward as Science (in Kant 2002b) Notes and Fragments—‘Reflections’ (in Kant 2005) Religion within the Boundaries of Mere Reason (in Kant 1988) Review of J. G. Herder’s Ideas for the philosophy of the history of humanity. Parts 1 and 2 (in Kant 2007) Review of Moscati (in Kant 2007) On the Use of Teleological Principles in Philosophy (in Kant 2007) Vienna Logic (in Kant 1992) An answer to the question: what in Enlightenment? (in Kant 1996a)

Introduction Kant and Other Animals: An Overview Lucy Allais and John J. Callanan

Non-human animals (henceforth ‘animals’) are not among the many topics Kant deals with systematically and explicitly in his critical philosophy. However, mention of animals occurs throughout his corpus; he touches on them in relation to many of his central concerns, and both the things he says directly about animals as well as what his views commit him to with respect to animals have important and interesting implications for understanding his views.1 This volume brings together, for the first time, a collection of essays discussing Kant’s views on animals and Kantian ideas in relation to animals, across a number of core topics. The volume by no means gives comprehensive cover to the issues raised in relation to animals in Kant’s philosophy, but we hope that the number of issues it touches shows the fruitful nature of what we are confident will be an expanding and rich area of research. Given the centrality of reason in Kant’s philosophy (including his moral philosophy based on reason and his theoretical philosophy concerned with a priori conditions of metaphysics and empirical cognition and with reason’s search for the unconditioned) one might expect him to be more interested in what separates humans from other animals than in these animals themselves. This expectation is often borne out, as he certainly is interested in what he sees as the distinct difference from other animals that follows from humans having reason and understanding. On the other hand, among his most frequent references to animals are those that occur when he is discussing our animal natures or what he refers to as animality in humans, suggesting an interest also in commonality, and he also 1 For a small selection of Kant’s mentions of animals and animality see CPrR 5:61–2; CPrR 5:127n; CPrR 5:160; CPrR5:161–2; CPJ 5:210; CPJ 5:334; CPJ 5:430–3; Rel. 6:26–28; Rel. 6:35; Rel. 6:80n; Rel. 6:195n; MM 6:213; MM 6:216; MM 6:387; MM 6:392; MM 6:418; MM 6:420; MM 6:421; MM 6:426; MM 6:427; MM 6:434–45; MM 6:445; MM 6:456; AC 25:13–15; AC 25:15; AF 25:475–6; AF 25:477; AF 25:516–17; AF 25:579–80; AF 25:607; AF 25:643; AF 25:677–8; AF 25:682; AF 25:682–4; AF 25:694; AP 25:736; AP 25:843; AP 25:844; 25:1170; AM 25:1342–3; AM 25:1360; AM 25:1361; AM 25:1415–16; AM 25:1420; EC 27:272–3; EC 27:379; EC 27:380–1; EC 27:385; EC 27:390–2; EC 27:440–1; EC 27:467; EV 27:492EV 27:494; EV 27:638; EV 27:641–2; EV 27:671–2; ML1 28:266–7; ML1 28:274–7; ML1 28:286–7; ML1 28:295; ML2 28:588; EM 29: 626; MMr 29:772; MMr 29:888; 29:900; MMr 29:911–12; MV 29:1033; MV 29:1038–9. Lucy Allais and John J. Callanan, Introduction In: Kant and Animals. Edited by: John J. Callanan and Lucy Allais, Oxford University Press (2020). © Lucy Allais and John J. Callanan. DOI: 10.1093/oso/9780198859918.003.0001

2

Lucy Allais and John J. Callanan

talks about animals in their own right. Animals are relevant to both Kant’s theoretical and his practical philosophy and thinking about animals touches on central concerns of each of his three Critiques—our understanding of cognition, the nature and basis of moral reasons, and the ontological status of organisms and the role of teleological thinking in understanding nature. We will say something about each of these issues.

Animals in Kant’s Theoretical Philosophy In Kant’s theoretical philosophy, thinking about animals is most obviously relevant to understanding his account of cognition, his account of the nature of life, and his views about the role of teleological thinking in understanding nature. Kant makes many scattered comments about animal cognition. His account of human cognition raises questions for understanding what he is committed to with respect to animal cognition and, in reverse, the things he says about animal cognition may be helpful in making sense of his account of human cognition. Kant attributes to humans basic mental faculties of sensibility, understanding, and desire. He takes understanding to involve concepts (which for Kant are essentially constituents of judgments) and therefore denies that other animals have understanding. However, his account of cognition gives no reason not to attribute sensibility and desire and their corresponding representations to animals, and he frequently does explicitly attribute these to animals. How exactly to understand these attributions though, is greatly in dispute, including the extent to which the sensibility he attributes to animals does or does not involve commonality with sensibility in humans. Issues relating to this question are discussed by three of the papers in this volume (Chapters 1–3); here we give the reader some key examples of discussions about Kant on animals in the secondary literature, a sample of the passages in which Kant talks about the mental lives of animals, as well as the capacities he attributes to them, and sketch some of the debates in the interpretation of Kant to which his account of animal cognition is relevant. As Steve Naragon points out in detail in his important 1990 paper ‘Kant on Descartes and the Brutes’,2 Kant explicitly rejects the Cartesian view of animals as machines that lack a mental life and frequently attributes various features of a mental life to them, including the power of choice, having feeling, expectation, satisfaction, and a sense of agreeableness. In the first Critique (and a variety of other places) Kant talks about an animal power of choice, which he says is sensible (bound up with the faculty of sensibility) and involves determination or necessitation by impulses of sensibility (as opposed to being free, which would involve

2 Naragon (1990); see also McLear (2011).

Introduction

3

independence of such necessitation—A 534/B 562, A 802/B 830). In the Groundwork of the Metaphysics of Morals, he says that non-rational creatures feel sensible impulses (G 4: 460n). In the preface to the Second Critique, emphasizing the importance of granting human reason a capacity for rational cognition, Kant contrasts our capacity to cognize objective necessity with a mere expectation of similar cases, which he attributes to animals, and calls subjective necessity (CPrR 5: 12). The contrast clearly involves attributing a mental life including representations and expectations to animals. Another contrast with animals occurs in Kant’s talking about the role of reason in human well-being: he says that as belonging to the sensible world humans are beings with needs, but that a human ‘is nevertheless not so completely an animal as to be indifferent to all that reason says on its own and to use reason merely as a tool for the satisfaction of his needs as a sensible being’ (CPrR 5: 61–2). This makes it clear that animals have sensible needs and well-being; it also involves stressing commonality with other animals as well as differences. While explaining his notion of reflecting in the Critique of the Power of Judgment, Kant says that reflecting goes on ‘even in animals, though only instinctively, namely not in relation to a concept which is thereby to be attained but rather in relation to some inclination which is thereby to be determined’ (CPJ 20: 211). He talks about gratification as something that is animal in involving bodily sensation (CPJ 5: 334) and in his discussion of different kinds of satisfaction he says that agreeableness is also valid for non-rational animals (CPJ 5: 210). All of these passages make clear that Kant attributes a mental life to animals. Animals are discussed in more specific detail in various notes taken by Kant’s students from his lectures. We quote some passages at more length to give an idea of these discussions. In the Metaphysik Herder Kant says that: Animals are not mere machines or matter, rather they have souls. Animals move themselves. An animal is thus an animated matter, for life is the faculty for determining oneself from an inner principle according to the power of choice. But matter, as matter, has no inner principle of self-activity, no spontaneity to move itself, rather all matter that is animate has an inner principle which is separated from the object of outer sense, and is an object of inner sense; there is in it a separate principle of inner sense. An inner principle of self-activity is just thinking and willing, only thereby can something be moved by inner sense; this is simply a principle for acting according to will and the power of choice. Thus if a matter moves, then it follows that there is in it such a separate principle of selfactivity. But only a being that has cognition is capable of this principle of thinking and willing. Matter can move only by means of such a principle. But such a principle of matter is the soul of matter. Thus: all matter which lives is alive not as matter but rather has a principle of life and is animated. But to the extent matter is animated, to that extent it is ensouled. A principle of life thus underlies animals, and that is the soul. (MH 28: 274–5)

4

Lucy Allais and John J. Callanan

Similarly, in the Metaphysik Mrongovius he says that: We call an animal alive because it has a faculty to alter its own state as a consequence of its own representations. Someone who maintained that in animals the principle of life has no power of representation , but rather that they act only according to general laws of matter, was Descartes, and afterwards also Malebranche, but to think of animals as machines is impossible, because then one would deviate from all analogy with nature, and the proposition: that a human being is itself a machine, is utter foolishness, for we are conscious of our own representations, and all natural science rests on the proposition: that matter can have no representations. (MMr 28: 449–50)

In another passage making clear his attribution of representations of sense to animals, he says: It is believed that animals have certain senses of which we have no concept, and that is entirely possible, e.g., when the wild geese travel in summer to cold lands, they move in the darkest night and yet in the straightest line to the north. Some senses have a large sphere in which they can perceive things. Feeling and taste have the narrowest sphere, but smell one still wider, e.g., the eagle smells carrion from up to a mile away. (MMr 29: 883)

And he says that: All three of these cognitive faculties [namely, the faculties of sense, imagination, and anticipation] can be accompanied by apperception or not. When they are, then they belong only to human beings, when not—then animals also have them. (MMr 29: 884)

Yet while Kant makes these claims about animals, it is not always clear to what extent what he is attributing to them is to be thought of as something they have in common with humans. He talks about understanding animals analogically, which leaves it somewhat disputable exactly what capacities he is attributing to them. In his opening contribution (‘The Comparison of Animals’), John Callanan aims to put Kant’s theory of animal minds in historical context. The debate on the status of animal minds in the Early Modern period was shaped in large part by the opposed positions of Montaigne and Descartes. Montaigne seemed to grant animals rationality, albeit of a lower degree of complexity than the rationality possessed by humans; Descartes, on the other hand, argued that animals must be denied rationality. Descartes argued that since consciousness and rational selfconsciousness were so closely linked, and animals lack the latter, they must lack the former. Descartes thereby reasoned to his notorious conclusion that animals

Introduction

5

must be denied consciousness altogether. Callanan argues that while Kant is in many places keen to stress the shared capacities of humans and animals, he can also be seen to be sensitive to the tension Descartes identified regarding projecting self-conscious perception to animals. Callanan claims that Kant in response adopts what he calls an analogy strategy in order to mediate between the two positions, though Callanan ultimately claims that the position is inherently unstable. An interesting and controversial feature of Kant’s discussion of animal mental lives is that he says that animals have outer sense but lack inner sense (though whether he changes his mind on this is disputed). For Kant, outer sense is the faculty through which we sensibly represent (or are presented with) objects that are distinct from ourselves (objects that are outside us in space), while inner sense is the faculty through which the mind is presented with (intuits) itself or its inner state (A23/B37). In the Metaphysik Herder, amongst other places, he says that: When we imagine beings a priori, then we will notice the difference not by degree, but rather by species; the difference and the comparison must thus rest on our outer and inner sense. Accordingly we can imagine beings that have a faculty of outer sense but forgo the faculty of inner sense, and these are animals. (MH 28: 275–6) Accordingly, animals will have all representations of the outer senses; they will forgo only those representations which rest on inner sense, on the consciousness of oneself, in short, on the concept of the I. Accordingly they will have no understanding and no reason, for all actions of the understanding and of reason are possible only insofar as one is conscious of oneself. (MH 28: 276)

Kant denies animals representations of their own inner lives and of themselves as selves (self-consciousness) while attributing to them representations of objects outside themselves. There is dispute about what follows from this with respect to animals having an inner/mental life at all. Some commentators argue that Kant’s denial of inner sense to animals shows his rejection of an account of perception which takes having an inner life with awareness of one’s own sensations to be primary and perception of a world outside and other than us to be a further step. Allais (2015), for example, emphasizes that Kant takes awareness of objects outside us as primary, and awareness of the self as having an inner life (including as having sensations) as the more sophisticated achievement. This is something which has been drawn on by philosophers who argue that for Kant perception of an external world is something immediate and direct.3

3 Abela (2002), Allais (2015), Collins (1999), McLear (2013, 2014a, 2014b, 2015), Westphal (2004).

6

Lucy Allais and John J. Callanan

An important question highlighted by claims Kant makes about animals is how to understand his use of the word ‘consciousness’. Self-consciousness and the unity of consciousness are among Kant’s most famous preoccupations, being the centre of the transcendental deduction of the categories in the Critique of Pure Reason. However, as Naragon (1990) argues in detail, his discussion of animals may be taken to support thinking that some of the time he uses the word ‘consciousness’ to mean what we would refer to as self-consciousness, rather than in the way the word is more commonly used today. In contexts in which he seems to be attributing to animals representations, sensation, feeling, expectations, imagination, etc., (which we might understand in terms of having a mental life with something that it is like to have it—what is often meant by consciousness in contemporary philosophy), he denies that they have consciousness. It is clear that he denies to animals awareness of the self as having the representations in question, which requires self-consciousness. Naragon argues that he allows animals to have representations but ‘[w]hat he disallows is the brute’s ability to represent to itself some representations that it has, or to bring several or all representations together, or to compare two separate representations’ (1990: 12), all of which seem to involve judging. If paying careful attention to the way Kant uses ‘consciousness’ in talking about animals supports thinking that some of the time he means by ‘consciousness’ what we mean by self-consciousness, this could have far ranging implications for a number of issues in the first Critique, notably the deduction. This question relates to a central way in which animals have recently been invoked in the Kant literature: in the dispute as to whether or not Kantian intuitions are dependent on concepts to be singular representations that give us objects. Intuitions, on Kant’s account, are singular and immediate representations that give us objects, and they are contrasted with concepts, which he claims to be entirely distinct kinds of representations that are mediate and general. The debate about the conceptualism/non-conceptualism of intuition concerns whether Kant takes the representations he calls intuitions to present us with perceptual particulars (individual things outside us that are presented to consciousness) independently of their being brought under concepts, or whether we would have something more like a booming, buzzing confusion of sensation, or perhaps no conscious life at all, without concepts. According to the conceptualist reading of intuition, outer intuitions that present us with distinct things outside and other than us are not the kinds of representations other animals—who lack concepts—could have. In contrast, non-conceptualist readers of intuition allow that we could be presented with perceptual particulars without concepts, and therefore that animals that lack concepts could have intuitions. Non-conceptualists such as Allais and Golob appeal to a reconstruction of what they take Kant’s view to allow him to say about animals as a way to explain their readings of intuition. For example, saying that a non-concept having animal could have outer intuitions is a way of explaining the claim that intuitions do not depend on the application of concepts to play their

Introduction

7

role in giving us objects. Non-conceptualists commentators such as Allais, Golob, and McLear have also appealed to things Kant says about animals in support of their readings; for example, Kant says such things as that ‘animals are acquainted with objects too, but they do not cognize them’ (JL 9: 65), and that the ox sees the gate although he does not see it as a gate (FS 2: 59).4 McLear (2011) gives detailed reasons for thinking that Kant attributes the capacity for perceptual awareness of objects to animals that do not have concepts. Conceptualist accounts of intuition are often based on Kant’s arguing (in the transcendental deduction of the categories) that what he calls synthesis (‘the act of putting different representations together with each other and comprehending their manifoldness in one cognition’ (A77/B103)) is governed by concepts (and in particular, the a priori concepts he calls the categories, e.g., A79/B104–5, B161),5 as well as claims saying that without the application of the categories nothing is  possible as object of experience. Those who interpret intuitions in a nonconceptualist way and argue that intuitions do not depend on the application of concepts to present us with perceptual particulars (distinct objects outside us that are perceptually presented to us) argue that categorially governed synthesis is not needed for us to have intuitions, but rather is something that transforms the intuitions we have, and enables us to cognize them. What Kant’s account implies about animals is therefore relevant to the central question of the role of synthesis in the deduction. It is also relevant to understanding what Kant means by an object of experience or objective experience, something discussed in detail by McLear and Golob in this volume. Claims Kant makes about animals having or not having consciousness are relevant to this debate: if by consciousness he always means having a mental life at all (a mental life that there is something it is like to have), then the passages where he denies that animals have consciousness will count against thinking that animals could have intuitions. Rather, we would need to understand his attribution of representations, feelings, and sensations as something more like unconscious representations or theoretical posits, and not commitments to a mental life with respect to which there is something it is like to have it. On the other hand, if he at least sometimes uses ‘consciousness’ to mean something more like self-conscious awareness that one has a representation of a particular sort, then these claims will be compatible with his allowing that animals perceive a world and have outer intuitions. Thus, we can look at the things Kant says directly about animals as well as to his account of human cognition to understand what he is committed to with respect to animal cognition, and we can also think about what he does and could say about animals to understand his account of human cognition. 4 Allais (2009, 2015, 2016, 2017), Golob (2016a), McLear (2011, 2014a). 5 See Bauer (2010), Ginsborg (2008), Gomes (2014), Griffith (2012), Grüne (2011), Land (2011), Williams (2012).

8

Lucy Allais and John J. Callanan

In his essay (‘Animals and Objectivity’), Colin McLear attempts to make precise the nature of animals’ conscious states according to Kant’s philosophy of mind. A central problem for that task is whether Kant can allow the plausible thought that animals are capable of objective representation of the world. Kant’s Critical apparatus might seem to preclude this and to restrict genuine objective representation to the minds of beings capable of the higher-cognitive activities related to synthesis, apperception, and category-application. McLear distinguishes several different senses of ‘objective’ and claims that Kant has room to attribute a minimal form of objective representation to animals, while restricting a more robust notion to human beings. This theme is continued in Sacha Golob’s contribution (‘What Do Animals See? Intentionality, Objects and Kantian Nonconceptualism’); like McLear, Golob is concerned with what kind of object-perception Kant thinks animals are capable of, but Golob also addresses the closely related questions of whether and how they are capable of representing spatiotemporal particulars, and what kind (if any) of intentional representation they can achieve. Golob too thinks that Kant can attribute to animals a form of intentional perception of spatiotemporal objects and save his philosophy of mind from the implausible impoverishment of animal minds. However, this strategy requires taking a stand in the recent debate about whether Kant’s philosophy of mind is properly of a conceptualist or nonconceptualist kind. Golob claims that the only way to accommodate a plausible picture of animals’ minds is to insist upon the nonconceptualist reading. Animals feature in Kant’s metaphysics in a number of interesting ways and the remaining three essays in the theoretical part of this collection concern other tensions and questions that animals present to Kant’s philosophy. As we saw above in the Metaphysik Herder, and as he repeats in numerous other texts (e.g., 28:680), Kant defines an animal as a being capable of moving its body through its representational states, in particular, its capacity for desire (Begehrungsvermögen). This means that the concept of animality plays an important (negative) role in Kant’s metaphysical account of matter and body in the Metaphysical Foundations of Natural Science, for he there argues that bodies do not, as such, have a capacity to move themselves (4: 544); In other words, bodies themselves are not, as such, animate. The nature of life and of animal life and the empirical principles of explanation we can use for understanding these, are clearly crucial themes in the Critique of the Power of Judgment. Kant’s understanding of animals is therefore relevant to his metaphysics as well as to understanding a central part of his epistemology and the appeal to teleology in his account of biological science. He holds that biological concepts, including the concept of an organism, cannot be explained mechanistically and require teleological thinking. Kant invokes animals, often together with plants, in discussions of our understanding of the world’s structure and the order of nature (A317–18/B374–5), in discussing the question of how far into the intrinsic natures of things our understanding of

Introduction

9

nature penetrates (A 667/B 695), and in discussions of the legitimate regulative use of the idea of a purposive unity of things in theoretical investigations of nature (A688/B716, 20: 217–18). He compares the systematic unity of cognitions to that of an animal body, saying that the whole is articulated and not heaped together, and can grow internally ‘like an animal body, whose growth does not add a limb but rather makes each limb stronger and finer for its end without any alteration of proportion’ (A832–3/B861–2). In ‘Kant on Non-Human Animals and God’, Ina Goy explores the account of animals that appears in Kant’s Critique of the Power of Judgment (1793). She examines the way in which humans and animals alike are held to be subject to both mechanical and physical-teleological laws and suggests that these two levels of explanation are ultimately reconciled for Kant by appeal to the notion of God. Goy argues that the relevant notion of God is ultimately one formed by projection of human characteristics, with the result that the divine being that has been conceptualized as one that has best an indirectly loving relation to animals found in nature. Kant’s universal and a priori moral account of what he calls moral law is often taken by his critics as disregarding, disvaluing, or even hostile to our animal natures. It might come as a surprise then that one of the most common places animality is mentioned in his philosophy is in relation to our own animality and he even claims that this is something we have a duty to preserve.6 In his theoretical philosophy, he talks of our animality in terms of the soul’s interaction with organic matter (A384) and the soul as the ground of our animality (A345/B403). Notably, in the Religion Kant describes as one of the three original predispositions of humans the predisposition ‘to the animality of the human being, as a living being’ (Rel. 6: 26–7). He describes it as a self-love for which reason is not required, which involves a drive for self-preservation, the propagation of the species through the sexual drive and the preservation of offspring, and the social drive to community with other human beings. While Kant holds that there are specific vices that trail the propensity to animality, he holds it to be good and oriented to the good and denies that the ground of the human propensity to evil can be located in the sensuous animal nature of humans (Rel. 6: 35). In his essay (‘Animality in Kant’s Theory of Human Nature’) David Baumeister address the question of animals with regard to the human animal. Readers of the Groundwork might be led to think that Kant sees the animality of human beings— in terms of their being capable of being determined to act by desires, frequently contrary to the demands of rationality—to be something to be overcome. Baumeister argues against this picture, showing how Kant consistently presents human animality as offering an indispensable and positive contribution to human 6 Herman (2018) argues that our understanding of our animality or our animal natures is an inseparable part of our understanding of ourselves as embodied sensible beings, which is part of our understanding of ourselves as moral agents.

10

Lucy Allais and John J. Callanan

existence. Baumeister focuses on three distinct areas: the role of animal nature in Kant’s Religion, in his theory of education, and in his account of historical progress, in order to defend this more nuanced view. In the final essay of the theoretical part of this collection (‘Kant on Evolution: A Re-evaluation’), Alix Cohen looks forward to the question regarding the plausibility of Kant’s theory of natural development and whether it can accommodate the evolution of animal species in a post-Darwinian framework. Cohen argues for the surprising claims that Darwinian evolution would not be strictly incompatible with Kant’s philosophy of biology, and that even Darwinian accounts of the evolution of animal species would, in Kant’s view, still require appeal to teleological reasoning.

Animals in Kant’s Practical Philosophy Some contemporary Kantians have seen the contrast with animals as vital to articulating the central and lasting contributions of Kant’s practical thought, such as the relevance of acting for a reason, acting under the idea of freedom, and the capacity for self-consciousness as interwoven with the human being’s experience of their agency: A lower animal’s attention is fixed on the world. Its perceptions are its beliefs and its desires are its will . . . But we human animals turn our attention on to our perceptions and desires themselves, on to our own mental activities, and we are conscious of them . . . I desire and I find myself with a powerful impulse to act. But I back up and bring that impulse into view and then I have a certain distance. Now the impulse doesn’t dominate me and now I have a problem. Shall I act? Is this desire really a reason to act? (Korsgaard 1996a, 93)

It can seem then that articulating the precise nature of the difference between humans and animals will depend on a precise identification of which cognitive capacities are necessary for the self-consciousness that can accompany practical deliberation and that can hopefully provide access to sources of normative guidance. It is notable, then, that what Kant says about animals in his moral philosophy is the most notorious way in which animals feature in his thought. Most problematically, he seems to say that the primary reason we ought not to mistreat animals is that doing so will harm our own moral characters. This seems to ground our obligations to animals in something about us, rather than their own needs and interests. He says that we do not have direct duties to animals themselves, but rather indirect duties with regard to them, which are really direct duties to ourselves. Kant’s central moral notion is respect, and in the Critique of Practical

Introduction

11

Reason he says respect is always directed to persons never to things, and he includes animals in the category of things. He says that they can awaken in us love, admiration, and amazement, but not respect (CPrR 5: 76). This is counterintuitive to say the least: when someone mistreats an animal, they undoubtedly let themselves down yet they surely let the animal down more. We give here a sample of the passages in which Kant presents these views: As far as reason alone can judge, a human being has duties only to human beings (himself and others), since his duty to any subject is moral constraint by that subject’s will. Hence the constraining (binging) subject must, first, be a person; and this person must, second, be given as an object of experience . . . A human being can therefore have no duty to any beings other than human beings; and if he represents to himself that he has such duties, it is because of an amphiboly in his concepts of reflection and his supposed duty to other beings is only a duty to himself. He is led to this misunderstanding by mistaking his duty with regard to other beings for a duty to those beings. (MM 6: 442)

On the idea of a ‘supposed duty to objects other than persons’, Kant says a ‘propensity to wanton destruction of what is beautiful in nature’ is: opposed to a human being’s duty to himself; for it weakens or uproots that feeling in him which, though not of itself moral, is still a disposition of sensibility that greatly promotes morality or at least prepares the way for it: the disposition, namely, to love something (e.g., beautiful crystal formations, the indescribably beauty of plants) even apart from any intention to us it. (MM 6: 443)

In a similar vein but even more strikingly, he says that: violent and cruel treatment of animals is far more intimately opposed to a human being’s duty to himself, and he has a duty to refrain from this; for it dulls his shared feeling of their suffering and so weakens and gradually uproots a natural predisposition that is very serviceable to morality in one’s relations with other people. (MM 6: 443)

On the one hand, Kant seems to give at least some animal interests moral weight, saying that: The human being is authorized to kill animals quickly (without pain) and to put them to work that does not strain them beyond their capacities (such work as he himself must submit to). But agonizing physical experiments for the sake of mere speculation, when the end could also be achieved without these, are to be abhorred. (MM 6: 443)

12

Lucy Allais and John J. Callanan

But on the other hand, immediately after this he again asserts that concern for these interests is not a duty owed to animals. He says that ‘[e]ven gratitude for the long service of an old horse or dog (just as if they were members of the household) belongs indirectly to a human being’s duty with regard to these animals; considered as a direct duty, however, it is always only a duty of the human being to himself ’ (MM 6: 443). Similarly, students’ notes from his lectures on ethics repeat the thought that we have duties ‘only towards other people; inanimate things are totally subject to our will, and the duties to animals are duties only insofar as they have reference to ourselves’ (EC 27: 413, see also EC 27: 459). Unlike many parts of Kant’s philosophy, there is not a large amount of literature on this topic.7 Much writing on moral concerns to do with animals has not, for understandable reasons, taken Kant as a starting point (the utilitarian starting point of well-being and rights-based talk are both more dominant in moral philosophy concerning animals), and literature on Kant’s ethics has focused more on understanding such notions as respect, moral motivation, maxims, and the relation between the moral law, reason, and freedom. Anyone who thinks that animals matter morally and that their needs and interests ground moral reasons for action will reject the idea that the only reason we have for not being cruel to animals is that this will be bad for our own characters, rather than this being grounded in something about the interests of the animals. So both those who find Kantian ethics plausible and everyone who wants to work out what it is actually committed to needs to work out what to say about this. There are a number of possible directions philosophers have taken this, some of which are explored in this volume. On one extreme, one could simply defend the idea that the ground of our reasons not to mistreat animals is not something about them, but rather the effects this has on us. On the other extreme, we could take the fact that Kant’s philosophy seems to have this implication as showing something fundamentally problematic about his moral philosophy, or with a moral philosophy based on reason. If we were to start with the view that however the moral concerns of animals are best characterized (whether or not, for example, we think the best way to make sense of the nature of our reasons for caring about animals should be understood in terms of animals having rights) they are grounded in the animals themselves, in their interests and needs, we might take Kant’s apparent denial of this as a reductio of his moral theory. Much of the interesting work on Kantian accounts of moral concerns for animals avoids both of these two extremes, though most accounts tend towards one or other end of this spectrum. According to a less extreme version of the first strategy, one insists that Kant’s and Kantian ethics cannot allow for animals to be sources of value themselves, yet claims that this view is not on reflection as 7 A selection of some of the relevant literature here can be found in Broadie and Pybus (1974, 1978), Cholbi (2014), Denis (2000), Kain (2010, 2018), O’Neill (1998); Skidmore (2001), Wood (1998).

Introduction

13

counterintuitive as it might first appear; according to the less extreme version of the second strategy, one accepts that Kant was fundamentally wrong in this insistence, yet claims that there are ways of incorporating concern for animals for their own sake within a Kantian account (rather than taking this to undermine his account of morality). As an example of the first strategy in the existing literature, Lara Denis (2000) explores a Kantian account of duties regarding animals in which she argues that taking seriously as a starting point the view that we cannot have direct duties to animals is compatible with capturing most of the concerns with respect to animals that we would want captured. She argues that taking Kant to hold that ‘the ways we treat animals reflect and affect morally important attitudes and feelings’ (2000: 417) enables us to give a stronger account of moral concerns of others than might be expected. She argues that Kant’s account in fact manages to capture much of what we might want captured with respect to the interests of animals, since it holds that: Tormenting animals for fun is inexcusable. Killing or using animals for food simply because one likes the taste of meet is unjustified. Killing animals for their skins because of the profits of selling leather and fur is wrong when we have other fabrics to keep us warm and dry; and even if we needed leather and fur, we would act wrongly if we captured or killed animals for it in an unnecessarily painful manner . . . Killing or harming animals as part of one’s job (perhaps on a factory-farm or in a slaughterhouse) would be a wrongful means to earning money provided that alternative employment is available. (2000: 415)

And she argues that it is plausible that ‘[c]ausing or allowing pain, damage, or death in an animal—and even treating a deceased animal violently—opposes a class of morally useful emotional dispositions that are part of our animal nature, and systematically damages us in a morally significant way’ (2000: 414–15). Thus, she argues that Kantian duties regarding animals are genuine, well-grounded duties, and capture many of the concerns of animals that we want captured. An objector could argue that even if the latter is true, this is not the primary ground for moral concern for other animals. In response to this kind of worry, Patrick Kain (2010) emphasizes that Kant’s objection is to what mistreating animals expresses about our feelings and dispositions, rather than simply its effects. He argues that to object that Kant’s account makes concern with animals only concern with our own characters misunderstands his account of what is involved in being concerned with the fate of animals, and that a self-respecting person is directly concerned with this fate. Similar points are made by Barbara Herman (2018). She emphasizes Kant’s view that animals are in their nature and their actions analogues to humanity and that our treatment of them should be responsible to this, as well as that loving animals is natural to us, and lacking concern for

14

Lucy Allais and John J. Callanan

what we naturally love is against our better nature (2018: 176). She argues that Kant’s concern is not merely with a contingent claim about treating animals badly being bad for our characters, and that his claim is not that it is only because of a contingent effect on our characters that mistreating animals is wrong. She suggests that to think that he is concerned merely with contingent effects on our characters would be to misunderstand why mistreating animals is bad for our characters and to misunderstand how serious he thinks the damage to our characters is. It is features of animal natures, as well as of our own natures, that explain why mistreating animals harms our characters. Kant says that the harm done to our characters by mistreating animals is that of uprooting a natural disposition. Herman takes him to be saying that it is part of our natural predisposition to take responding to animal suffering as non-discretionary, rather than as something with respect to which we should start by looking for reasons as to why it should count morally. She says that if we took treatment of animals to be discretionary, we would ‘thereby treat our natural unconditional response to suffering as, morally speaking, conditional. We would have replaced a natural response with a question we have to answer: does this suffering matter?’ (Herman 2018: 185). The first three essays in the practical part of this collection concern the first strategy. In the first essay (‘Directionality and Virtuous Ends’), Ripstein and Tenenbaum address the classic objection to Kant’s treatment of animals’ moral status. They distinguish various sources of the objection and argue that the key complaint is duties regarding animals possess on Kant’s account the ‘wrong kind of directionality’, i.e. they are to be explained as fundamentally directed towards ourselves. Ripstein and Tenenbaum claim that when the source of the objection is properly identified, various readings of the objection are available but that none of them are fatal for the Kantian account. Further, they argue that Kant’s claim that we have only indirect duties to animals should be taken to inform our understanding of what he means by a direct duty. Helga Varden argues in her essay (‘Kant and Responsibility for Animals’) that Kant’s position can be similarly defended by distinguishing between duties ‘to’ animals and duties ‘with regard to animals’, allowing that the latter may exist but not the former. Varden maintains that this distinction allows the Kantian to avoid the counterintuitive consequence. Varden’s reading is especially distinctive by virtue of her drawing (like Baumeister) on Kant’s account of human nature in his Religion Within the Boundaries of Mere Reason. It is here, Varden argues, that a richer and more persuasive Kantian account of the source of our responsibilities to animals (and our understanding of what happens when we fail to meet those responsibilities) can be found. In her paper (‘What Do We Owe to Animals? Kant on Non-Intrinsic Value’), Carol Hay argues again in defence of the Kantian view that animals lack intrinsic value. For Hay, however, this metaethical position leaves much of what we would want to say at the first-order level about our responsibilities to animals in place.

Introduction

15

Hay also claims that we can understand a sense in which moral responsibilities to animals are nevertheless fully genuine ones despite animals failing to constitute a source of intrinsic moral value. In fact, Hay claims that a genuine sense of respect for animals is still recoverable even when this Kantian qualification on their moral status is made. As mentioned above, a different strategy is to argue that Kant is simply wrong, in his own terms, about what his moral theory implies for our concerns for animals, and that if we think through the grounds of our moral concerns for humans we will see that Kantian ethics is in fact committed to moral concern for animals for their own sake. In the literature, this option is explored in detail by Christine Korsgaard, in a number of papers and now in her new book, Fellow Creatures (2018) which goes so far as to argue that Kantian ethics should recognize animals as well as humans as ends in themselves. Thus, she develops a Kantian account that rejects the idea of direct duties being only to humans. Korsgaard’s argument is partly based on her constructivist account of moral reasons, according to which value can only be grounded in valuing. Further, she argues that once we understand what value is and what animals are, we will see that both self-conscious, rational creatures and other animals accord themselves value. Since there is nothing more for them to have value than for them to accord themselves value, they have value. She argues (2018, Ch. 2) that it is in an animal’s nature to monitor her own condition and to represent the world in ways that will motivate her to keep her condition good; animals function by making their own well-functioning their end of action or final good. Since, on her view, nothing can be important without being important to someone, she argues that there is no standpoint from which we can coherently ask which creatures (for example, us or other animals) are more important; rather, things are important to creatures. On her view, therefore, people and other animals are the beings to whom things are important and it does not make sense to attribute more importance to people. She argues that to say a creature is an end in itself means that we should accord the creature the kind of value that, as a living creature, she necessarily accords to herself. In contrast, to treat someone as a mere means is to use her to promote your own ends in a way to which she could not possibly consent or for ends that she could not possibly share (2018, Ch. 5). Certain things are good-for creatures and because these creatures are ends in themselves, therefore these things are good. On Korsgaard’s account, the difference between us and other animals is that as self-conscious animals, humans also constantly constitute themselves as agents by determining the unity and coherence of our conception of the world and ourselves as agents. In choosing our actions we are deciding what sort of selves to be (2018, Ch. 3). She disputes the view that Kant thinks of rationality as a valuable property that confers intrinsic value on those who have it, arguing, in contrast, that Kant is committed to thinking that claims about what has value must be established as presuppositions of rational choice and therefore that we must

16

Lucy Allais and John J. Callanan

suppose that rational beings have value in order to engage in practically rational activity (2018, Ch. 8) Whether or not this is the only way Kantian constructivism can be developed, this argument shows that thinking about the implications of the view for other animals can bring out the commitments of an interpretation of Kant. In ‘Against the Construction of Animal Ethical Standing’, Jon Garthoff criticizes what he sees as the limitations of Korsgaard’s brand of ethical constructivism, limitations that he thinks are brought into focus with the case of animals. In particular, Garthoff considers the question of how various types of entity have (or lack) ethical standing, the consideration of what properties or capacities are relevant when asking whether and why a creature’s existence matters. Garthoff criticizes Korsgaard’s account of ethical standing and proposes his own rival account. According to Garthoff, ethical standing comes into being with conscious awareness. This revision amounts to a rejection of some core notions of Kant’s ethical system as traditionally understood, since the basis of ethical standing does not, Garthoff contends, stem from the more demanding condition of rational selflegislation within a community of reflective subjects. Yet such a revision is required, he argues, if we are to have a metaethics that is still Kantian in spirit and which can contend with the question of animals in a plausible way. In the final essay in this collection (‘Hope and the Threshold Chicken: A Kantian Argument Against Purchasing Meat’), Andrew Chignell looks away from the debate regarding the status of animals in Kantian ethics and metaethics, and instead makes a perhaps unexpected use of Kant’s theory of hope and despair for the question of our contemporary responsibilities to animals. The vegan who seeks to abstain from using animals for ‘flesh or fur’ faces a threat of psychological de-motivation when confronted with the empirical facts regarding the relative insignificance of their individual efforts. An argument from Kant’s philosophy of religion can be brought to bear on this issue, Chignell claims. Chignell examines what he calls the ‘moral argument from despair’ to evaluate the prospects for different methods of generating and sustaining hope in the face of moral despair regarding our current and future treatment of animals. In this way, Chignell brings this collection to a fitting forward-looking close, by making a use of Kant’s philosophy to help contend with one of the most crucial ethical issues facing us today.

PART I

THEOR ET IC A L PH ILOS OPH Y

1

The Comparison of Animals John J. Callanan

The concept of animal souls and of higher spirits is only a game of our concepts. (Lectures on Metaphysics, ML1, 28:278)

1.1 Introduction What happens in the mind of a non-human animal when it represents different aspects of its environment?1 Imagine that one such animal—a New Caledonian crow for example—approaches a wall containing two gaps of different sizes.2 Imagine that it pauses momentarily, moving its head from the smaller gap to a larger one, then travels through the larger one. What is a plausible reconstruction of the mental life of the crow as it processed this scene?3 One obvious analysis might be that the crow saw two spaces, saw that the latter one was larger, and for that reason moved to that larger gap. In so doing, the crow would have represented its environment much the way in which we do. Anyone who modelled the crow’s mental life in such a way nevertheless could still vehemently deny what I shall call the continuity thesis. The continuity thesis is the claim that, whatever the variations in their mental lives, animal and human minds manifest no differences in kind but rather exhibit the same general type of mental capacities merely exercised with very different degrees of sophistication. Our representation of the environment is no doubt very different from that of animals, but the advocate of the continuity thesis would claim that animals perform the same very general kinds of mental activity that we do, only at far lower levels of complexity; the denier of the continuity thesis denies that the difference between animals and us can be coherently modelled as a difference in mere complexity. 1 I’ll refer to non-human animals as ‘animal’. As with many of the Early Moderns with whom I am concerned, the intended extension will be the familiar larger land-dwelling vertebrates such as cats, dogs, foxes, horses, cattle, and birds. 2 Corvids provide an excellent example of animals whose mental life appears at once sophisticated enough for their activities to be comparable to human activities while retaining for us an intuitive sense of their radical difference in consciousness, e.g. Angell & Marzluff (2008,  2013); Heinrich (1999). 3 For some recent philosophical literature on this and related topics, see Andrews (2015), Andrews and Beck (2018). John J. Callanan, The Comparison of Animals In: Kant and Animals. Edited by: John J. Callanan and Lucy Allais, Oxford University Press (2020). © John J. Callanan. DOI: 10.1093/oso/9780198859918.003.0002

20

John J. Callanan

When the Early Moderns considered the question of animal minds they did so with relatively simple examples like the one above. One might also think that someone with Kant’s model of the human mind would be well-positioned to explain such animal behaviour. One could attribute to the crow a capacity for sensory input, but also for spatial intuition and perhaps the reproductive imagination, i.e. basic capacities to orient one’s sensory input into coherent spatial arrays, such that it allows for conscious representation of the spatial advantages of the currently represented larger gap over the recently represented smaller one. The crow represents things in its environment in a way that affords it reliable representations of those things, on the basis of which it can then realize a successful plan of action. On various occasions, Kant seems to attribute various parts of the human cognitive apparatus to animals.4 It is clear though that Kant is an ardent denier of the continuity thesis and that he claims that human beings are different in kind from animals by virtue of our ability for self-conscious understanding and the opportunities for normative self-determination that this ability affords.5 Since these features are the ones Kant frequently states with regard to our difference to animals, it might seem plausible to assume that our comparison with animals consists in imagining them as possessing something like human minds minus these higher elements. This view brings with it several puzzles about Kant’s model of the human mind and its relation to some of his argumentative strategies in the First Critique.6 However, I am interested here in those more rare occasions where Kant seems to manifest some hesitancy regarding even modest comparisons of human and animal minds, occasions where he instead gestures towards a more radical discontinuity between human and animal mentality. In a letter to Prince von Beloselsky in 1792, Kant offers a correction to the Prince’s proposed division of mental capacities, claiming instead that one must distinguish the ‘faculty of representation into the mere apprehension of representations, apprehensio bruta without consciousness (that is only for the beast) and the sphere of apperception, i.e. of concepts; the latter constitutes the sphere of understanding in general’.7 In the Dohna-Wundlacken logic lectures we find references to human and animal cognition where the difference is presented equally starkly: {[Consciousness] is required above all for concepts. Due to the lack of consciousness, even animals are not capable of any concept—intuition they do have. 4 For some typical examples, see CPJ 5:464; MVo 28:449; DWL 24:702—for discussion, see McLear (2011), this volume; Golob, this volume. 5 For the denial of the continuity thesis, see some of the references below. For an example of Kant’s account of the relevance of the normative dimension, see CPrR 5: 61–2. 6 Such questions concern the apparent tight conceptual links Kant seems to make between the capacities for spatial or temporal intuition (or both) with concept-use, consciousness, the capacity to represent objects as objects, cognition, and self-consciousness. For some of these issues, see Golob, this volume, Fisher (2017), Naragon (1990); McLear (2011), this volume. 7 Letter to Prince Alexander von Beloselsky, 1792, Corr. 11: 345.

The Comparison of Animals

21

Consciousness is a wholly separate dimension of the faculty of cognition (therefore gradation from animals to man does not occur).}8

Here Kant seems to deny any ‘gradation’ with regard to the class of conscious representations at all. Insofar as there is some nominal continuity—in the shape of a shared capacity for intuition—the capacity possessed by animals might be such that it is not a capacity for conscious representations at all. It is hard to see, however, how such continuity with humans’ capacity for conscious intuition could be coherently articulated. In the metaphysics lectures Kant denies animals inner sense, which he claims entails that they also lack understanding and reason and thus animals ‘are accordingly different not in degree but rather in species’.9 In a crossed-out section of the Anthropology, Kant strikes a similarly radical note concerning theorizing about animal minds: The irrational animal has something similar to what we call representations (because it has effects that are similar to the representations in the human being), but which may perhaps be entirely different—but no cognition of things; for this requires understanding, a faculty of representation with consciousness of action whereby the representations relate to a given object and this relation may be thought.10

Here Kant appears to state that although the behavioural evidence speaks strongly in favour of attributing to the crow the capacity to successfully represent a wall, to see the hole in it that is larger than the other one, etc., one must nevertheless resist such attributions, if they are thought to compare to what we do when we see the same wall and its holes, since in so doing we—and not they—have a ‘cognition of things’. The crow might not have ‘mental representations’ that bear even a family resemblance to the human representation as it is realized in human cognition. The grounds for this caution are stated clearly—Kant has defined ‘cognition’ in such a way that token acts of cognition require a token act of understanding, and where the latter requires a potential second-order mental state whereby one is conscious of the relevant recommended action by the understanding. Since animals entirely lack this more advanced state, they must fail to secure cognition of things, and for that reason might even lack ‘representations’ and ‘consciousness’, at least insofar as those notions are thought of as bearing family resemblances to human representation and consciousness. One might reasonably explain away such textual evidence: these passages are not part of Kant’s main published works nor are they always comments directed to a sophisticated philosophical reader. More importantly, Kant is on the record

8 DWL 24: 702, 440. 9 ML1 28: 276, 87. 10 Anth. 7: 141, 252—editor’s note—see also Fisher (2017).

22

John J. Callanan

in a range of other contexts as opposing the notorious Cartesian position that seemed to deny even simple forms of consciousness to animals.11 Kant is of course known for his Critical period position that human consciousness is marked by the capacity for self-consciousness.12 Thus, when he speaks of denying animals ‘consciousness’, the charitable reading is to take this to be denying them only the specific manifestation of potentially self-conscious representations that we humans possess. While such a reading of the texts would have the merits of avoiding attributing to Kant the absurd Cartesian position, there are grounds to be hesitant about this reconciliation of the texts, or so I will argue. The advantage of opposing Descartes and attributing consciousness to animals is surely that it squares with the empirical evidence, since animals appear to register sensations, feel pain and pleasure, and negotiate their environment successfully to the end of satisfying their various drives. The point of attributing consciousness to animals is to allow for the compelling thought that they might represent things in ways that are comparable to our own ways of doing so, at least with regard to some lower cognitive achievements. This is compatible with denying them our quotidian conscious achievements. For example, while Kant allows that the animal can feel pain, it seems that he denies that it can feel distress or sorrow, because those latter mental events are ones that require a second-order representation of the fact that one is in pain or that one was recently not in pain or that one’s pain is likely to persist, etc., representations he thinks animals cannot possess.13 Presumably, though, this difference is to allow for the intuitive comparison that they do feel pain as we feel pain, and the point of attributing some consciousness to animals is to allow that they nevertheless perform first-order representation in much the way that we do. While Kant might seem sympathetically anti-Cartesian in this respect, I argue that his position is more ambivalent—Kant also maintains the broadly Cartesian claim that the mental life of animals might be so radically different from our own that speaking of their ‘consciousness’ or representational achievements must always be at best hypothetical. In particular I will argue that, since Kant’s notion of human consciousness is one that has rudimentary self-consciousness accompanying cognitive achievements all the way down to the mind’s basic actions, it is singularly difficult for him to articulate a sense of what it is for a creature to have comparable conscious states without some rudimentary form of self-awareness. I will focus upon the activities of animals such as the crow in the example above, where the animal engages in an act of comparing and differentiating a small series of simple sensory inputs such that they convey information about a spatial

11 E.g. ‘[A]nimals are not mere machines or matter, rather they have souls’—ML1 28: 274ff,—see also ML2 28: 594, MVo 28: 449, MD 28: 690, CPJ 5: 374, 246. 12 See A106–7, A127, B131–2, B136, B142. 13 ‘On the philosophers’ medicine of the body’ in Kant (2007), 15: 944, AF 25: 474.

The Comparison of Animals

23

environment. Kant’s model of the human mind demands that the crow must be performing this act of comparison in a way that is very much unlike what we humans do when confronted with the same sensory input. It is so unlike our activity that it is in fact incorrect to say that the crow represents the bigger gap as bigger than the smaller one. Whether this characterization of the animal’s mental life is sufficient to accommodate our intuitions or whether it perpetrates a kind of Cartesian recidivism is open to question. I will present the case for this view though a consideration of the historical context of the animal mind debate.14 I first outline (§2) the issue of animal consciousness as it was presented in Bayle’s Critical Dictionary, a standard eighteenthcentury source text for the philosophical problems that animal minds presented to philosophy and theology. In §3 I outline Locke’s discussion in the Essay concerning the plausible similarities between human and animal cognition. Locke presents a hierarchy of cognitive achievements of increasing complexity, ultimately claiming that while animals are capable of lower-class cognitive operations, such as comparing one particular representation with another, they are incapable of the higher cognitive operation of abstracting general content from those representations. Locke thereby sacrifices comparison as a shared capacity but goes on to identify other higher capacities to support a denial of the continuity thesis. In the following section (§4) I outline how French materialists such as Condillac, La Mettrie, and Helvétius co-opted Locke’s analysis of the activity of comparison and used it as the thin end of the wedge in energetic and controversial support of the continuity thesis. This movement sought to eliminate the difference between humans and animals not by granting animals rationality but by denying it to humans. Once sensation and simple operations like comparison are allowed, they claim, there are grounds to reduce putative ‘higher’ capacities to those former lower ones, thereby eliminating the distinct capacity of reason itself.15 It is in this context that I examine Rousseau’s innovative contribution to the debate in Émile, a contribution that cannot have failed to impress Kant. As part of a polemic against Helvétius, Rousseau argued that the continuity thesis failed even with regard to the basic activity of comparison. Rousseau’s critique hinges on a distinction between representing different contents and representing the difference between those contents. In §5 I argue that the pre-Critical Kant was coterminously drawing similar conclusions, ones that would no doubt have been strongly reinforced by the extraordinary impact of Émile upon the development of his thought.16 In the concluding section (§6) I argue that this influence goes some way to explaining, though perhaps not justifying, what I take to be Kant’s 14 I restrict myself in this paper to standard sources in the debate with which Kant would plausibly have been familiar. 15 For a recent discussion of this theme, see Lloyd (2017). 16 For some discussion of this influence, see Ameriks (2012); Velkley (1989,  2013); Zammito (2012).

24

John J. Callanan

quasi-Cartesian approach to the difference between animal and human minds. In Kant’s view, an animal ‘senses’, ‘imagines’, and perhaps even ‘reasons’, yet only in senses analogous with (and thus not literally the same as) human sensation, imagination, and reason. Therefore Kant’s analogy strategy (as I will call it) can allow that animals ‘compare’ sensory representations while denying that they engage in the same activity that humans engage upon when we compare sensory representations.17 This approach, I claim, is vulnerable to criticisms raised long before by Bayle. Both Descartes and Kant possess a meritorious sensitivity to the ubiquitous dangers of anthropocentricism in the very attempt to model the mental lives of animals. That sensitivity is not exculpatory of their worldviews however; on the contrary, it is arguably dogmatically utilized as a means to the end of rationalizing a pre-eminent position for human beings within the cosmic order.

1.2 Bayle and the Soul of Brutes It is somewhat surprising that Kant never presented a systematic account of the mental capacities of animals. Not only was the question pertinent to his own Critical model of the mind but the issue was already very well established as a theme within European intellectual culture in the seventeenth and eighteenth centuries.18 Part of the reason for the prominence of the topic was surely the enormous influence of Montaigne’s Essays and its claims regarding the affinity between human and animal minds in the ‘Apology for Raymond Sebond’.19 One of the many evocative examples Montaigne uses is that of the fox testing the ice on a frozen river: Consider the fox which Thracians employ when they want to cross the ice of a frozen river; with this end in view they let it loose. Were we to see it stopping at the river’s edge, bringing its ear close to the ice to judge from the noise how near to the surface the current is running, darting forward or pulling back according to this estimate of the thickness or thinness of the ice, would it not be right to conclude that the same reasoning passes through its head as would pass through ours . . . ?20

While reiterating classical debates regarding the capacity of animals to reason, Montaigne’s revisiting of the issue was especially marked by the suspicion he 17 I have explored Kant’s use of analogy in other contexts in Callanan (2008, 2017a). 18 For a very selective sample of some of the enormous literature on this topic, see Beauchamp (1999); Cottingham (1978a); Fudge (2006); Garrett (2006); Manning and Serpell (2012); Matytsin (2016); Melehy (2005); Rosenfield (1941); Smith (2011, 2012). 19 Montaigne ([1580] 2003). 20 Montaigne ([1580] 2003, II: 12, An Apology for Raymond Sebond, 515).

The Comparison of Animals

25

raised with regard to the motives human beings have in considering themselves especially distinct from other animals.21 Human beings have an inherent and ineradicable tendency towards pride, Montaigne claimed—it is our ‘maladie naturelle et originelle’—and this is what makes us susceptible to any narrative that might confirm our sense of our own unique importance.22 Montaigne’s contribution fundamentally and forever after destabilized the frame of the debate in that it raised a suspicion about the sources of our eagerness to deny the continuity thesis. In the Discourse on Method—and no doubt with Montaigne in mind— Descartes had held that ‘after the error of those who deny God . . . there is none that lead weak minds further from the straight path of virtue than that of imagining that the souls of the beasts are of the same nature with ours . . . ’.23 As his own pairing of issues here makes explicit, Descartes’s notorious position on animal minds was not impartially motivated. The narrative later provided by Bayle regarding Descartes’s theological motives runs roughly as follows: Descartes’s end was to secure proof of the possibility of an immortal soul exclusive to human beings; the means to that end was to link the capacity for conscious thought in general with the possession of an immaterial substance; his tight pairing of the two entailed that he was left only with the options of either granting animals immaterial souls on the grounds of their possession of some form of consciousness or of denying them consciousness so as to deny them immaterial and immortal souls, and so he opted for the latter position.24 The characterization of animals as mere ‘beast-machines’—soulless automata lacking in real consciousness—was a point of attack on the Cartesian system from its first presentation.25 The opposition, though passionate, rarely focused upon the argumentative grounds for Descartes’s position, but rather focused upon the absurd conclusion. Few critics though grasped the nettle and addressed the question as to what a contentful theory of a conscious representation in animals might amount to.26 One should ask though just what it might be like for an animal to possess a conscious sensory representation if this is to be understood as a state that does not involve being able to attend to the representation itself. We can speak of the bending of the plant in terms of its representing the location of the sun in aid of photosynthesis or of the rings of the tree as its representing of its own age. On the other hand we can talk about our conscious representations as performing a qualitatively different task, which is that of presenting material to

21 For discussion of the classical tradition, see Adamson and Edwards (2018); Clark (2011); Sorabji (1993). 22 An Apology for Raymond Sebond in Montainge ([1580] 2003: 505, 513). 23 Discourse on Method [1637] Pt VI in Descartes (1985: 141). 24 ‘Rorarius’ in Bayle ([1697] 1991). 25 For discussion, see Cottingham (1978); Hastings (1973); Rosenfield (1941); Wilson (1999). 26 For this point, see Wilson (1999).

26

John J. Callanan

consciousness, of representing as. What might it be though for an animal to possess a capacity for conscious representation without possessing a capacity to take that representation as presenting things to itself? Descartes’s argument then becomes more of a burden-shifting one: if we grant that animal consciousness lacks the self-consciousness that characterizes human consciousness, then it is up to the opponent to say just what they mean by ‘animal consciousness’ if it doesn’t involve an attribution of a capacity for representing as. Interpreting Descartes charitably then would just involve his denial that there is anything like a well-defined sense of what it might be like for animals to possess conscious mental states, and that the beast-machine thesis stands until displaced by such a theory.27 The charitable interpretation does not ameliorate the sense of absurdity regarding Descartes’s conclusions. Bayle’s analysis in the Critical Dictionary became the eighteenth-century touchstone for the animal mind debate and contains a wealth of anecdotal data on animal behaviour to reaffirm the profoundly counterintuitive nature of Descartes’s claim.28 Bayle is not exclusively confronting the Cartesian school but is equally opposed to the Aristotelian tradition of granting an animal a mere ‘sensitive’ soul of a non-rational nature. What both traditions have in common is the denial of rationality or thought to animals. It is the apparent rationality of animal behaviour that Montaigne had stressed requires explanation and the manoeuvres made by both camps to deny this—and by extension deny the continuity thesis—is the focus of Bayle’s critique.29 The general argumentative tenor of those in favour of the continuity thesis is that there is both behavioural—not to mention physiological—continuity between humans and animals. Postulating similar causes, i.e. conscious representation of one’s environment and use of inferential capacities, is a reasonable abductive inference with regard to the similar behaviour we observe in animals and humans alike.30 It is this move that the Aristotelian makes to dismiss the Cartesian position. Bayle seems to grant the point31 but he points out though that once lower mental states are granted to the animal on grounds of explanatory adequacy, higher cognitive capacities must also be granted on those same

27 Cottingham (1978) is widely cited as offering an important corrective to the thought that Descartes denied conscious feelings to animals. Yet Cottingham’s analysis points out only that Descartes latterly attributed states such as ‘fear, hope, and joy’ (557) while denying them ‘thought’. Cottingham himself acknowledges that it is still a tricky question as to just what it means to Descartes when he attributes to an animal joy without thought. It is surely open to Descartes to run the line of argument again—there is no problem in attributing animals ‘feelings’ so long as we acknowledge that what we are attributing is only what physiology can tell us. 28 Especially in the articles ‘Rorarius’ and ‘Pereira’. 29 Bayle’s own position is, as ever, elusive. The view presented does not claim that Bayle endorses the continuity thesis but rather that his trenchant critique is directed at the thought that the denial of the continuity thesis has a rational basis. 30 An Apology for Raymond Sebond in Montaigne ([1580] 2003: 514). 31 ‘No proof is needed with regard to the Cartesians. Everyone knows how difficult it is to explain how pure machines can accomplish what animals do . . . ’ (‘Rorarius’, in Bayle [1697] 1991: 214–15).

The Comparison of Animals

27

grounds. Bayle criticizes the disposition to attribute one cognitive capacity to animals while withholding another, especially where that latter capacity is a characteristically concomitant one: Every Peripatetic who hears that beasts are only automata, or machines, objects immediately that a dog who has been beaten for touching a dish of meat will not touch it again when he sees his master threatening him with a stick. But to show that this phenomenon cannot be explained by the one who introduces it, it is sufficient to say that if this dog’s action is accompanied by knowledge, then the dog must necessarily reason: he must compare the present with the past and draw a conclusion from this. Now is this not definite reasoning? Can we explain this situation by simply supposing a soul that is capable of feeling, but not of reflecting on its actions, but not of recalling past events, but not of comparing two ideas, but not of drawing any conclusions? Look carefully at the examples that have been compiled and are raised against the Cartesians, you will find that they prove too much; for they prove that beasts compare means with ends, and that they prefer on some occasions what is just to what is useful; in a word, that they are guided by the rules of equity and gratitude.32

The obvious response, one with which Bayle was intimately familiar, was that it was in fact eminently plausible to postulate non-rational causes that reproduce behaviour akin to that produced by rational causes. Cartesians and Aristotelians alike would claim that they can accommodate the evidence without the attribution of rationality just because the behaviour can be fully explained by the attributions they do make. For the Aristotelian, the behaviour is completely explicable in terms of the sensitive soul responding in a sophisticated habituated manner. For Descartes, the behaviour is no less explicable once one attributes a sufficiently complex system of mechanistic responses. These responses can mimic human operations exactly when observed from a third-person perspective but are nonetheless entirely different in kind.33 As Bayle notes, the antagonists in this narrative are operating under two constraints. In the first place, they must postulate a cause that is adequate to the effects; secondly though they must postulate a cause that is inadequate to explain the workings of human minds so as not to support the continuity thesis. Bayle suggests that while the Cartesian camp plainly fails to meet the former criterion, he raises the ‘proves too much’ objection to show that the Aristotelian camp fails to meet the second criterion.

32 ‘Rorarius’, Remark B, in Bayle ([1697] 1991: 215). 33 Kant occasionally strikes the same Cartesian note with regard to the necessary commitments for explaining animal behaviour, e.g. ML1, 28: 277. Thanks to Alice Wright for drawing my attention to this and other related passages.

28

John J. Callanan

It is crucial to note that Bayle’s argumentative strategy is not just to appeal to the idea that the attribution of rationality to animals is a warranted explanation of observed behaviour. Rather his target is the ad hoc nature of the moves made by Aristotelian and Cartesian alike to resist this attribution and to claim that the behaviour is fully explicable in terms of the functions of a sensitive soul or a mechanism. The overarching theme of ‘Rorarius’ is the tangles theorists realize by their desperate clinging to the second criterion of the debate, that of retaining at all costs the denial of the continuity thesis. The ‘proves too much’ objection is not just the claim that the attribution of consciousness to animals brings with it an attribution of rationality. It can also be run in the opposite direction, so as to motivate a denial of rationality to humans. The Aristotelian criticism of Descartes suggested that possession of a sensitive soul might be sufficient to explain the manifestation of rational behaviour in animals; Bayle notes that it also threatens to explain away the presence of a rational soul in human beings. If animal rational behaviour can be well accounted for by the attribution of sophisticated responses by a sensitive soul, why can’t human rational behaviour be similarly explained?34 There remains one possible way of negotiating the thicket, which I refer to as the analogy strategy. Here the theorist claims that in one sense animals do possess rationality, but that they possess a mere analogous variant of the rationality we literally possess, one that strictly speaking is not of the same kind as ours. In this way, the theorist can allow that animals ‘sense’ or ‘reason’ or ‘communicate’, but one need not see that as threatening the distinction between automata and conscious minds or the distinction between sensitive and rational souls. The analogy strategy is a means of denying the continuity thesis while accommodating the data that speaks in its favour. Its drawbacks are obvious, however. It is just another blatant piece of ad hoc reasoning—the offering of a distinction without a difference—inviting Montaigne’s original point that such manoeuvres are made exclusively for the advantage of maintaining a sense of human distinctness. The Baylean ‘proves too much’ argument can simply be run again against the analogy strategy: how does one know that the ‘reason’ we possess is not really merely that same lower variant of the more exalted form of reason we would prefer to claim for ourselves?

1.3 Locke on Comparison In ‘Rorarius’, Bayle had asked whether it was plausible that a dog can have some kind of conscious representation ‘but not of comparing two ideas, but not of 34 ‘Rorarius’ in Bayle ([1697] 1991: 215, 232). Bayle intimates that the Cartesian might be committed to a kind of solipsism and should regard their own introspectively accessed rationality as the only securely identified one.

The Comparison of Animals

29

drawing any conclusion’,35 and had praised Locke’s characteristically straightforward attempt to distinguish human and animal cognition in the Essay.36 There Locke presents a range of cognitive achievements of increasing sophistication and asks whether they are likely to be shared by animals. Locke’s judgments are for the most part presented without evidence but are presumably made on the grounds of what he takes to be the likely consensus based on observational data. The first of the cognitive operations (presumably taking the mere receipt of sensory ideas for granted) is the act of comparison: The COMPARING them one with another, in respect of Extent, Degrees, Time, Place, or any other Circumstances, is another operation of the mind about its Ideas . . .37

Locke claims that animals compare representations to a limited extent: How far Brutes partake in this faculty, is not easie [sic] to determine; I imagine they have it not in any great degree . . . I think, Beasts compare not their Ideas, farther than some sensible circumstances annexed to the Objects themselves . . .38

It’s worth noting here that Locke allows that animals do in fact compare their representations one with another, and so do literally partake in the same cognitive operation. The difference concerns the targets compared, in that animals only use that faculty when comparing those things that are immediately spatiotemporally relevant to whatever object the animal is currently sensing. The approach is repeated with regard to the next cognitive operation, that of the composition of ideas (whereby new ideas are formed from the combination of simpler ones): In this also, I suppose, Brutes come far short of Men. For though they take in, and retain together several Combinations of simple Ideas, as possibly the Shape, Smell, and Voice of his Master make up the complex Ideas a Dog has of him . . . yet, I do not think that they do of themselves ever compound them, and make complex Ideas. And perhaps even where we think they have complex Ideas, ‘tis only one simple one that directs them in the knowledge of several things . . .39

35 37 38 39

‘Rorarius’ in Bayle [1697] 1991: 215). 36 ‘Rorarius’ in Bayle ([1697] 1991: 242). Essay, Bk II, Chapter XI, §4, in Locke ([1689] 1998: 157). Essay, Bk II, Chapter XI, §5 in Locke ([1689] 1998: 157–8). Essay, Bk II, Chapter XI, §7 in Locke ([1689] 1998:158).

30

John J. Callanan

Here, Locke employs the standard technique for the denial of the continuity thesis. He allows that animals might take in various ideas and combine them in various ways (presumably through association and what we now think of as classical conditioning). He claims that that this might result in behaviour that is observationally indiscriminable from the behaviour we realize by virtue of possession of an entirely different cognitive state, that of compounding simple ideas into a complex one. Presumably though, Locke is aware that such reasoning risks the appearance of an ad hoc stipulation and so he prefers to point to what he thinks shows ‘a perfect distinction betwixt Man and Brutes’: namely the capacity of the former to generate and possess abstract representations.40 He does not conclude from this however that animals lack rationality: If they are to have any Ideas at all, and are not bare Machins [sic] (as some would have them) we cannot deny them to have some Reason. It seems as evident to me, that they do some of them in certain Instances reason, as that they have sense; but it is only in particular Ideas, just as they receiv’d [sic] them from their Senses.41

Locke therefore follows Descartes in denying the continuity thesis, but does so on different grounds than the denial of rationality. He wants to claim with Montaigne that animals genuinely reason though he qualifies this strongly by claiming that they do so only from particulars and not from the general propositional contents that are the occupants of abstract thoughts.42 In criticizing Locke, Leibniz’s Nouveuax Essais emphasized a derogatory comparison between animal cognition and the hyper-nominalist empiricism associated with some models of medical practice (involving amassing large amounts of micro-observations and withholding general explanatory principles). For Leibniz, the reason why animals cannot have knowledge of necessary truths is because their capacity for representation is limited to discrete temporal occurrences without the capacity to engage in the durational acts of comparison required in order to see the necessary agreement of ideas:

40 Essay, Bk II, Chapter XI, §10 in Locke ([1689] 1998: 159). 41 Essay, Bk II, Chapter XI, §11 in Locke ([1689] 1998: 160). 42 The position that animals cannot reason because they know only particulars, is mocked as a standard piece of Aristotelian dogma by Bayle, though its criticism is extendible to Locke here: ‘Nothing can be more diverting than to see with what authority the Schoolmen endeavor to set limits to the knowledge of beasts. They insist that beasts know only particular and material objects, and that they love only what is useful and pleasant, that they cannot reflect on their sensations and desires, nor infer one thing from another. It would seem that they have searched more successfully into the faculties and the acts of the soul of beasts than the most expert anatomists into the entrails of dogs’ (Bayle [1698] 1991: 221).

The Comparison of Animals

31

[B]easts are sheer empirics and are guided entirely by instances. While men are capable of demonstrative knowledge [science], beasts, so far as one can judge, never manage to form necessary propositions, since the faculty by which they make thought sequences is something lower than the reason that occurs in men. Beasts’ thought sequences are just like those of simple empirics who maintain that what has happened once will happen again in a case which is similar in the respects that they are impressed by, although that does not enable them to judge whether the same reasons are at work.43

Leibniz here is accusing those theorists, especially the ‘empirics’ of the medical tradition, of aspiring to model our proper forms of knowing on that of nonhuman animals. But here the thematic connection of temporal representation and concept-possession is brought into play: it is because animals’ representational capacities are in some sense restricted to ‘instances’ that they cannot even form conceptual or propositional representations, let alone grasp them in a reliably veridical manner. Locke’s capacious conception of reasoning allows for the movement of the mind from one particular representation to another to count as a genuine manifestation of inferential behaviour.44 The move threatens to undermine further the claim that one can identify a ‘perfect distinction’ between animal and human minds. Firstly, the denial of abstract representation to animals is simply asserted rather than argued for; secondly, even if it were granted then Montaigne’s central thesis—that human minds are not distinguished by virtue of the possession of rationality—would be conceded. Leibniz by contrast retains the Cartesian conception of rationality as that which makes scientia possible through the conscious grasping of necessary truths that make understanding possible. On this more restricted notion of reason, Leibniz has no option but to insist with Locke that the apparent inferential behaviour is performed upon representations of particulars only but to insist contra Locke that for that very reason the patterns of animal behaviour are merely analogical with genuinely rational behaviour, a mere ‘shadow of reason’.45

1.4 Helvétius and Rousseau on Comparison and Judgment By the time Kant started to engage with Locke’s philosophical views, many of them had already been taken in a radical direction by the eighteenth-century

43 New Essays, Preface, in Leibniz ([1765] 1997: 50). 44 LoLordo (2012, Ch. 3). 45 New Essays, Bk IV, Ch. XVII, §3, in Leibniz ([1765] 1997: 475).

32

John J. Callanan

French materialists.46 A helpful entry point into that influence is their co-opting of the notion of comparison with regard to judgment. The ‘proves too much’ objection has one further and even more radical application: if one were to grant to the Cartesian that the behaviour of the animal sensitive soul was explicable in terms of a mere mechanical automaton, then given the previous application of the argument, why can’t the behaviour of the human rational soul be explained in terms of a mere mechanical automaton? Just as the apparent consciousness of an animal might be explained away, so too might the apparent distinctiveness of human consciousness if it could now be characterized as a mere mysterious epiphenomenon attached to a human machine.47 What Bayle had proposed as a reductio for Cartesianism became the positive agenda of French materialism. The adoption of Bayle’s objection is explicitly stated in La Mettrie’s MachineMan (1747). La Mettrie characterized the distinction between human and animal as not involving any abrupt transition, and characterized the contingent distinctiveness of human cognition as deriving from its having contingently been the case that human beings have developed language capacities. For La Mettrie, language acquisition and complex perceptual capacities appear to go hand in hand, since he claims that prior to that acquisition, the perceptual phenomenology of human beings was such that ‘[they] saw only forms and colours, without being able to distinguish any of them’.48 Without linguistic or other sign-use one lacks the very capacity for differentiation of particulars and would be like a ‘little child (for then the soul was in its childhood) holding in its hand a certain number of little pieces of straw or wood, who sees them in general in a vague and superficial way without being able to count or differentiate them.”49 Taking imagination as a basic instinctual response accompanying sensation, La Mettrie claims that ‘all the parts of the soul can be properly reduced to imagination alone, which forms them all, and thus that judgement, reason and memory are only parts of the soul which are in no way absolute’.50 In his Traité des Sensations in 1754, Condillac would praise Locke for his empiricist account of the sensory origin of our ideas, but criticize him for allowing for the nominal distinctness of the capacities of the mind to compare and judge through reflection. Locke should have seen the next obvious step, Condillac claims, and is to be scolded for not suspecting that ‘they [the faculties of the soul] could derive their origin from sensation itself ’.51 Condillac has no doubt that ‘[j] udgment, reflection, desires, passions, etc. are only sensation itself which is 46 Kant possibly read Locke’s Essay first of all through an engagement with Leibniz’s Nouveau Essais in 1765. For the reception of Locke by French materialism, see Yolton (1991). I discuss the themes of this section at greater detail in Callanan (2017b). 47 For discussion of the rise of the materialist proposal from these origins, see Matytsin (2016); Rosenfield (1941). 48 La Mettrie ([1747] 1996: 13). 49 La Mettrie ([1747] 1996: 14). 50 La Mettrie ([1747] 1996: 15). 51 Condillac [1754] quoted in O’Neal (1996: 16–17).

The Comparison of Animals

33

transformed differently’.52 A similarly radical sensationist tract came soon after from Helvétius. His 1758 work de l’Esprit begins by putting forward what might be thought of as a clear two-faculty model of cognition, one for ‘receiving the different impressions caused by external objects’ and the other for ‘preserving the impressions caused by these objects’, which Helvétius calls ‘Physical Sensibility’ and ‘Memory’, respectively.53 However, it is clear that the faculty of memory is for Helvétius, at the end of the day, nothing but reproductive, i.e. it functions to ‘preserve’ the contents that have come in through sensibility, the latter being the source of all original representational content. More than this though, Helvétius claims that all the ‘higher’ cognitive faculties, i.e. judgment and inference, are themselves to be thought of as nothing more than variations on the operations of sensibility and ‘that all the operations of the Mind consist in the power we have of perceiving the resemblance and difference, the agreement or disagreement, of various objects among themselves’—in this way ‘every thing is reducible to feeling’.54 The reference here is again to Locke, for whom judgment is ‘the putting Ideas together, or separating them from one another in the mind, when their certain Agreement or Disagreement is not perceived, but presumed to be so’.55 All judgment, Locke seems to say, is the act of seeing when two ideas that we have acquired through sensation ‘agree’ or ‘disagree’.56 In de l’Esprit (1759), Helvétius gives an example of how the sensationist explains such operations, taking the case of the examination of two objects of different lengths: The question being thus properly limited, I shall proceed to examine, if Judgment be not Feeling. When I judge of the magnitude or the colour of objects presented to me, it is evident, that the judgment is formed from the different impressions made by those objects on my senses; and therefore may be said, with the greatest propriety, to be nothing more than a sensation. For I can equally say, I judge, or I feel, that of two objects, the one, which I call a fathom, makes a different impression on me, from that of the other, which I call Foot; also, that the colour which I call Red acts upon my eyes differently from that

52 Condillac, [1754] quoted in O’Neal (1996: 19). See also Condillac, Essay on the Origin of Human Knowledge: “The similarity between animals and us proves that they have a soul; and the difference between us proves that it is inferior to ours” (Condillac [1746] 2001: 39). Condillac differs from Locke in claiming that although animals are conscious they nevertheless do not reason, not on the rationalist grounds that they lack abstract ideas (though they do, he thinks) but because they lack the freedom of thought that he thinks reasoning requires so that ‘their actions, when they appear rational, are merely the effect of an imagination that is not at their command’ (fn, 90). 53 Helvetius ([1759] 1807: 1–2). 54 Helvetius ([1759] 1807: 7). 55 Essay, IV.xiv.4 in Locke ([1689] 1975: 653). For discussion, see Gaukroger (1989); Owen (1999, 2007). 56 Of course it is a more complicated question as to what Locke really thought was involved in the act of judgment—for a discussion of some of the difficulties, see Owen (2007).

34

John J. Callanan which I call Yellow: from whence I conclude that in all parallel cases, Judging is the same with Feeling.57

The notoriety that de l’Esprit earned in France motivated Rousseau to take on Helvétius on exactly these points late in the composition of Émile in 1762.58 For Rousseau the non-reductive difference between sensing and judging can be easily ascertained: To perceive is to sense; to compare is to judge. Judging and sensing are not the same thing. By sensation, objects are presented to me separated, isolated, such as they are in nature. By comparison I move them, I transport them, and, so to speak, I superimpose them on one another in order to pronounce on their difference or their likeness and generally on all their relations. According to me, the distinctive faculty of the active or intelligent being is to be able to give a sense to the word is. I seek in vain in the purely sensitive being for this intelligent force which superimposes and which then pronounces; I am not able to see it in its nature. This passive being will sense each object separately, or it will  even sense the total object formed by the two; but having no force to bend  them back upon one another, it will never compare them, it will not judge them.59

The characteristic feature of the power of judgment is that it is an active manipulation of sense contents—Rousseau speaks of ‘superimposition’, of ‘transportation’, and of the bending back of sensations upon themselves. The grasp of the relations between sensations, he claims, is only made possible through the contribution of the activity of judgment. The argument seems to be that the grasp of relations requires the manipulation of sensations; the power of sensibility itself can only receive and not manipulate sensations; therefore, the grasp of relations cannot be within the purview of the faculty of sensation itself. It’s worth noting a further point here, which is that this argument depends on an acceptance of Helvétius’s atomistic picture of sensations themselves. It is only because sensations are in themselves ‘separated, isolated’ atomistic contents that we can understand that the relations between the atoms cannot themselves be perceived.60 Rousseau takes up Helvétius’s own example of a comparative judgment of length: 57 Helvetius ([1759] 1807: 8–9). 58 (Rousseau ([1762] 1979). For discussion of these themes, see Audidière (2016); Hanley (2012); Smith (1982). 59 Rousseau ([1762] 1979: 270–1). 60 The atomism of sensationism is generally an unargued premise.

The Comparison of Animals

35

To see two objects at once is not to see their relations or to judge their differences. To perceive several objects as separate from one another is not to number them. I can at the same instant have the idea of a large stick and of a small stick without comparing them and without judging that one is smaller than the other, just as I can see my entire hand at once without making the count of my fingers. These comparative ideas, larger and smaller, just like the numerical ideas of one, two, etc. certainly do not belong to the sensations, although my mind produces them only on the occasion of my sensations.61

The argument hinges on the thought that there is a distinction between possessing different representations and possessing a representation of their difference. I might coterminously possess the representations [large stick] and [small stick] without noting their difference and generating [two sticks, one larger than the other]. This latter representation requires a distinct act of comparison, a secondorder mental activity that takes the first-order representations as its targets. Therefore, there is more to judgment than the mere combination and aggregation of sensory contents; there is, rather, added content—in this case the representational content [larger than]—simply not given by sensation and, therefore, contributed by the act of judgment of the human agent. The historical sketch I’ve been presenting suggests that two different questions started to be run together by the middle of the eighteenth century. The question about the continuity between animal and human minds became closely linked with the question of the continuity between the non-rational and rational capacities. Helvétius and Condillac sought to capitalize on Locke’s account of comparison by suggesting that there is no principled reason to postulate conceptual discontinuity between the lower and the higher cognitive faculties at all. Rousseau objected to this reductive sensationism on the grounds that even the apprehension of relational content within human beings’ perceptual phenomenology involves a rudimentary capacity for self-consciousness. Rousseau questions the continuity thesis not on the rationalist’s grounds that animals cannot ‘know whether the same reasons are at work’ in different circumstances; rather he questions it on the same grounds upon which Helvétius and others had argued for it, i.e. as being a plausible hypothesis that explains the phenomenology involved in simple acts of relational representation. For Rousseau, the obvious phenomenological distinction between noting representations and noting the comparison between them blocks the continuity thesis from applying even at the lowest levels of our cognitive lives.

61 Rousseau ([1762] 1979: 271).

36

John J. Callanan

1.5 Kant, the Ox, and the Stable Door Kant seems at first to follow the broadly Aristotelian tradition, granting animals something like ‘sensitive souls’ but not the capacity to act in accordance with a self-conscious and rational faculty of choice that marks the presence of a rational soul.62 Thus when it comes to animal minds he allows them consciousness of some sort, namely whatever consciousness is involved in pure sensation: A representation of sensation simply cannot be invented by us but instead must be given to us by means of the senses. Experience and sensation are distinct as to form and as to degree. As to matter, the two are the same, but in experience there is a form as well, reason. Experience is nothing but reflected sensation, or sensation that is expressed through a judgment. Experiences, namely, are not mere concepts and representations but also judgments [;] e.g., the representations of warmth or of cold are concepts of experience. They are universal characters of things, but we cannot have insight into these merely through the senses, but actually only through judgment. Non-rational animals have no experience, then, but instead only sensations. Anyone who can describe the objects of his experience has experience, for description involves not merely sensation but also a judgment.63

Kant, like Locke, allows that animals ‘compare’ representations. However, unlike Locke, and like Rousseau, Kant claims that this notion of ‘comparison’ is different in kind from the activity that human beings perform: Animals indeed compare representations with one another, but they are not conscious of where the harmony or disharmony between them lies. Therefore they also have no concepts, and also no higher cognitive faculty, because the higher cognitive faculty consists of these. This [faculty] is thus differentiated by apperception from the lower cognitive faculty. As animals, we have the latter in common with them, but the former raises us as thinking beings over animals.64

The point was made explicitly in the False Subtlety essay in 1764—the same year as the publication of Émile—in some well-known passages. Kant was already anticipating the Rousseauian claim about judgment and would have no doubt endorsed what he later found in Émile. Here Kant opposed Meier’s contention that animals can possess a distinct concept:65

62 This wasn’t an unusual position. Kant would have been following Baumgarten, for instance—see his Metaphysics (1757), §792 (‘On the Souls of Brutes’) in Watkins (ed.) (2009: 129). 63 BL 24: 236. For Kant on sensation, see George (1981); Naragon (1990). 64 MMr 29: 888. 65 For a helpful recent discussion of Kant on clarity and distinctness, see Sommerlatte (2016).

The Comparison of Animals

37

The distinctness of a concept does not consist in the fact that that which is a characteristic mark of the thing is clearly represented [vorgestellt], but rather in the fact that it recognized [erkannt] as a characteristic mark of the thing. The door is something which does, it is true, belong to the stall and can serve as a characteristic mark of it. But only the being who forms the judgment: this door belongs to this stable has a distinct concept of the building, and that is certainly beyond the power of animals. (FS 2: 59)66

An animal can possess the representations [door] and [stable] but the complex representation [the door of the stable] requires an additional relational content only provided by acts of judgment. Obviously there is the (by now familiar) question as to how an animal can differentiate the door of the stable from other aspects of its environment so successfully if it nevertheless lacks complex representations. Kant’s response is a version of the analogy strategy: one must differentiate different senses of what it is to differentiate. Just like Rousseau, Kant insists that ‘it is one thing to differentiate things from each other, and quite another to recognize the difference between them’ and the latter ‘cannot occur in the case of animals’.67 For Kant, animals are certainly capable of highly sophisticated representational differentiation, but these must be understood as complex sets of stimulusresponse patterns relating to simple sensory representations: Physically differentiating means being driven to different actions by different representations. The dog differentiates the roast from the loaf, and it does so because the way in which it is affected by the roast is different from the way it is affected by the loaf (for different things cause different sensations); and the sensation caused by the roast are a ground of desire in the dog which differs from the desire caused by the loaf, according to the natural connection which exists between its drives and its representations.68

66 For further discussion of this passage, see Golob, this volume. As the editors of (Newton 2012) note, Kant here may be influenced by Reimarus’s (1760) Allgemeine Betrachtungen Über die Treibe der Thiele—see Kant (1992: 427, note 41). An earlier possible source was perhaps Buffon’s multi-volume Histoire naturelle génerale et particulière, which was published in French from 1753 and which greatly impressed Kant. There Buffon distinguishes animals by connecting together some of the recurring themes I have mentioned in the debate, namely the capacities of comparison, self-awareness and complex temporal representation: ‘Very far from denying anything to animals, I grant them everything with the exception of thought and reflection: they have sensations even to a higher degree than us, they also have the consciousness of their present existence, but not of the past one: they have feelings, but they lack the faculty to compare them’ (Buffon 1753, 4: 41; quoted and translated in Matytsin 2016: 18). Though not published until later, Euler makes similar reflections in his Letters to a German Princess (Letter XCV, 20 January 1761 in Watkins (ed.) (2009: 203)). 67 FS 2: 60–1. 68 FS 2: 60. For a different reading, see Newton (2015).

38

John J. Callanan

The animal differentiates a from b in the sense that there are purely causal laws (‘natural connections’) that determine a stimulus-response pattern that is marked in terms of the differential behavioural outputs. Switching to my original example then, when the crow sees the larger gap in the wall, it does not see the larger gap as larger than the smaller one, rather it has a causally conditioned response that leaves it disposed to move towards gaps of that size and not ones of the smaller size. The crow never ‘holds’ the two representations of the gaps in the wall before its mind nor notes the difference between them. Despite the broadly Aristotelian stance, one can see also see something distinctly Cartesian about Kant’s reasoning here. Descartes’s objection to Montaigne had been that the observable behaviour of animals was adequately accounted for by postulating automata that (we might say) reacted in conformity with the laws of nature but not for the sake of the representation of those laws. Why postulate the capacity to react in recognition of the phenomenal content of what is represented when merely postulating rule-governed reactions to stimuli will adequately explain the behaviour? Kant is similarly here questioning whether we need to attribute to the animal a representation of the distinctness of representation of the stable door when a mere reliable differential reaction to a representation of that door will suffice. It is clear how inadequate Kant’s theory is here. Presumably Kant attributes some kind of phenomenal consciousness to animals so as to avoid the Cartesian claim that they lack pain-experiences etc. To adequately explain the crow’s choosing of the larger hole in the wall over the smaller one, we have to attribute to it one of two general pictures, both of which assume that the crow possesses some kind of phenomenal consciousness. On the first picture, the animal is just responding reliably to causal stimuli and through habituation has learned to be disposed to move towards gaps of certain sizes while refraining from moving towards others. On the other picture, the animal can attend to its phenomenal consciousness to guide its behaviour in aid of the more successful outcome. If Kant intended the first picture then it would seem that the presence of phenomenal consciousness is otiose for the explanation of the behaviour of the crow, since what is guiding its behaviour are the laws of nature that generate its responses, while its phenomenal consciousness is just an intermediary effect of those initial causes of those responses.69 However, if Kant intended the second picture, then it is hard to see what the animal would be doing were it to exploit 69 Kant grants animals a capacity to actively determine their own environment: ‘[w]e call an animal alive because it has a faculty to alter its own state as a consequence of its own representations’ (MVo 28:449.) I would claim that Kant’s meaning here is that external causal stimuli generate a representational response and that representational response is causally connected with a behavioural output. It is in this transitive sense that the animal ‘alters its own states as a consequence of its own representations’ What it does not do is use or exploit its own representations consciously in order to realize this outcome.

The Comparison of Animals

39

the content of its phenomenal consciousness without doing so in exactly the manner Kant denies is occurring, i.e. by attending to the content presented to it within its representations.

1.6 The Shadow of Reason Animals’ behaviour is often ordered, regular, and purposeful. It is natural to infer that the structure of the mentality that we make use of in realizing our ordered, regular, and purposeful behaviour is replicated in other animals also. We are at least strongly inclined to think that their mental lives do not manifest a disordered, irregular, and undirected phenomenal mess. Yet there is an enormous spectrum of possible models of the mind that might account for the orderedness of their mental lives without attributing them the same structure our mental life might possess. A machine-like mechanism is one way of thinking of how that animal mind is ordered, albeit one that risks denying them anything like a mind at all. On the other hand, we would be wise to resist a picture of the ordered mental life too similar to our own—we ought not to think that one pig, upon spotting another in front of it, attests to itself, ‘Lo, a fellow pig on the path!’ The mental life of many animals is instead plausibly modelled somewhere between these two points on the spectrum. Kant’s occasional ambivalence regarding characterizing animal mentality is in one sense meritorious, since one of his motives is epistemic caution in saying more about the mental lives of animals other than that it falls somewhere between these two points. There are also proper philosophical reasons why Kant sometimes appears quasi-Cartesian in his outlook. The critic of the Cartesians accuses them of a self-serving rationalization in making consciousness exclusive to their species. However, the charitable understanding of the Cartesian renders the accusation of anthropocentricism particularly ironic—it is the advocate of the continuity thesis that perpetuates an anthropocentric methodology in presuming that animal consciousness can or should be articulated by comparison with our own. Kant’s model of human cognition invites reflection on the thoroughgoing reflexivity of all human consciousness, including human sensory consciousness. What we know of what it’s like to be conscious per se is saturated with this conception, since our model of consciousness per se is only ever formulated from the human standpoint. When we imagine what it is like for an animal to see gaps in the wall, or a door in a stable, or two sticks of different lengths, we cannot help but imagine what it’s like for an animal to see these things in terms of what it’s like for us to see these things. We try to imagine animals’ mental lives as being like ours but without certain of the features that essentially characterize our mental life. It should be unsurprising then that it is difficult to imagine what it’s like for

40

John J. Callanan

an animal to compare two sticks of different lengths without falling into a model that attributes to them something like a proto-version of our consciousness, whereby we imagine that they attest ‘lo, this stick is longer than that one!’ yet in some non-linguistic and thoroughly non-human manner. Yet it is possible that animals do not do anything even comparable to this. It is with regard to concerns such as these that Kant sometimes follows Leibniz in advocating the analogy strategy. In Metaphysik L1, he states that ‘We can attribute to animals an analogue of reason , which involves connection of representations according to the laws of sensibility, from which the same effects follow as from a connection according to concepts. Animals are accordingly different from human souls not in degree but rather in species . . .’70 So far as Kant can make sense of what it is to possess understanding and reason he demands that they ‘are possible only insofar as one is conscious of oneself ’.71 Since he cannot grant them that higher capacity, the most they can attain is a shadow of reason achieved by virtue of the superior aspects of their sensory apparatuses: But we see from the actions of animals undertakings which we would not be able to bring about other than through understanding and reason. Accordingly sensibility is with us such a state as with animals, except theirs is far advanced over ours. But for this privation we have received compensation through the consciousness of ourselves and through the understanding that follows from it. We are also not at all required to assume reflection in animals, rather we can derive all of this from the formative power. Accordingly we ascribe to these beings a faculty of sensation, reproductive imagination, etc., but all only sensible as a lower faculty, and not connected with consciousness.72

As already mentioned, the analogy strategy’s merits are far from obvious. One way of thinking about Kant’s apparent use of the analogy strategy is that it risks Baylean parody. In the face of the obvious similarity between animal and human physiology and given the need to accommodate this in terms of granting shared cognitive functions, this acquiescence is at once rescinded with the disclaimer that how animals cognize the same environmental patch is of course entirely different from the way in which we humans sense it. Thus, while it is granted that animals ‘compare’, they do so in ways entirely different from the ways in which we compare representations. Considered in this context it is easier to make sense of Kant’s remark in the Anthropology that while animals may represent in the same sense in which humans represent they might also have an entirely different variant of what we call representation. 70 ML1, 28:275–6. The analogy strategy can sometimes seem to be in play with regard to animals’ moral status—in the Collins ethics lectures, Kant says that ‘since animals are an analogue of humanity, we observe duties to mankind when we observe them as analogues to this . . .’ (EC 27: 459). 71 ML1, 28:276. 72 ML1 28: 277.

The Comparison of Animals

41

The beast-machine hypothesis opposed Montaigne’s conception of animals as rational on the grounds that behaviour that mimicked inference and deliberation could be as well explained by an appeal to automatic responses to causal stimuli. The first step of Descartes’s response is that mechanical causes are sufficient to explain behaviour; the second step is that if this so then appeal to conscious states is not necessary. Kant’s model of animal minds attempts a mixed explanation: he claims that animal responses are automatic causal responses but that these responses just happen to be routed through conscious states. Kant does not have an explanation against Descartes as to why conscious states are required to account for animal behaviour.73 Neither though does Kant have an adequate explanation against Montaigne as to why these conscious responses can be known to fail to instantiate self-consciousness. Why isn’t it the case that if animals possess an analogue of our reason, and if our reason requires self-consciousness, then animals possess an analogue of self-consciousness? Similarly, if animals can be granted an analogue of self-consciousness, what grounds are there for insisting that their self-consciousness is really just an analogue of our own rather than a simpler but genuine manifestation of our own?74 Bayle’s point is that denials of the sort made by Descartes, Aristotelians, Leibniz, and now Kant are all simply ad hoc stipulations. Kant therefore grants animals capacities for sensation and imagination, but strictly sensation sans awareness, imagination sans awareness. Yet one cannot help but wonder what the denier of the continuity thesis would accept as falsifying of their position and what observed behaviour could suffice as evidence of the presence of proto-self-awareness of a genuinely comparable kind. Certainly a range of behaviour that resembles rational and self-aware behaviour could be explained away by means of the analogy strategy. The demand that none of animals’ behaviour attests to the presence of self-consciousness appears like re-packaged Cartesian dogma. On such an approach, it is hard to see just what an animal needs to do in order to generate a genuine comparison with humans. The difference between human and animals on this account is no clearer than our ability to distinguish our own rational justifications from our self-serving rationalizations.75 73 This is not to say that Kant lacks any explanation at all as to why animals have mental states. Kant broadly follows Leibniz’s position (originally outlined in the latter’s ‘A New System in the Communication of Substances, as well as the Union Between the Soul and the Body’ in 1695) in characterizing an animal as a spiritual automaton (Leibniz [1695] 1989: 458). The grounds for the distinction between a machine and a spiritual automaton are just the principle of life that determines the latter and not the former. Yet obviously something can be alive yet non-conscious so appeal to a life principle is not itself sufficient to explain why animal’s automatic responses are accompanied with rudimentary consciousness while a plant’s (say) automatic responses to causal stimuli are not. 74 An analogous complaint can be raised against Leibniz’s criterion here: if our reason requires an understanding of necessary truths that carves nature at its joints, why can’t animal behaviour be described in externalist terms as reliable rational responses to just those same truths? 75 I am very grateful to the audience at the Witswatersrand conference on ‘Kant and Animals’ at which some of this content was originally presented. Thanks also to Lucy Allais and Max Edwards for feedback. Special thanks to Alice Wright for very valuable discussion as well as comments on the final draft.

2

Animals and Objectivity Colin McLear

2.1 The Problem of Objectivity Observation of the behaviour of non-rational animals (hereafter ‘animals’) reveals a rich and varied world. While the natural philosophers of the seventeenth and eighteenth centuries were well aware of this fact, there was a tendency to downplay the level of sophistication inherent in animal behaviour, or to the extent that it was acknowledged, to credit it to the amazing powers of nature in yielding mechanisms with such complexity.1 Kant, along with other figures in the German rationalist tradition, is laudable for not falling into the error of construing animals as mere machines.2 Kant is on record in various places as saying that animals have sensory representations of their environment (CPJ 5:464; LM 28:449; cf. An 7:212), that they have intuitions (LM 28:449, 28:594; LL 24:702; OP 21:82), and that they are acquainted with objects though they do not cognize them (JL 9:64–5).3 Without further argument I am simply going to assume that Kant holds, at the least, the position that seems apparent from these texts—viz. that animals enjoy mental representations, and that postulating such representations is necessary for explaining sophisticated animal behaviour. Acknowledging the existence of representation in animals does not commit Kant to thinking that animals possess the faculties of reason or understanding, or that animals possess any power of mental combination beyond that of mere empirical association.4 However, there is a nagging worry for this interpretation. One might readily admit that Kant allows that animals have sensory states, and that these are 1 Descartes and his followers are perhaps the most famous figures here; for discussion, see Cottingham (1978a); Gaukroger (1995: 288); Alanen (2003: 101); Lähteenmäki (2007: 178–9); Naragon (1990). 2 Leibniz’s views on animals are well known, though the details of his views are contested. See e.g. Kulstad (1991) and Hartz and Wilson (2005). Closer and more relevant to Kant is Meier’s work on animals, especially Meier (1749). 3 For further discussion, see Naragon (1990); Allais (2009); McLear (2011). For examples of positions that deny this possibility, see McDowell (1996, Chs. 3 and 6); Ginsborg (2006;  2008); Grüne (2009). 4 Indeed, Kant seems clearly to reject the notion that animals possess the capacity for use of the first person concept, and with it any of the ‘higher’ cognitive faculties of understanding, judgment, or reason. See ML1 (c.1777–80) 28:277; Anthropologie Mrongovius (c.1784/5) 25:1215; Anth.: 7:127. Colin McLear, Animals and Objectivity In: Kant and Animals. Edited by: John J. Callanan and Lucy Allais, Oxford University Press (2020). © Colin McLear. DOI: 10.1093/oso/9780198859918.003.0003

Animals and Objectivity

43

necessary for animals to successfully register, navigate, and survive in their surroundings, and thus that there is a sense in which such states count as ‘representations’. But it seems that one could agree to all this and yet deny that such states deserve the moniker ‘objective’. This distinction between objective representation and sensory registration needs sharpening, but as a first pass we might simply use Kant’s terms and distinguish between representation of an object, and representation merely as modification of the subject (A320/B376–7). What, however, does this distinction come to? Take, for example, Gareth Evans’s discussion of objectivity in his 1980 essay ‘Things Without the Mind’. He says: We can imagine a series of judgements ‘Warm now’, ‘Buzzing now’, made by a subject in response to changes in his sensory state, which have no objective significance at all. But we can imagine a similar series of judgements, prompted by the same changes in the subject’s sensory state, which do have such a significance: ‘now it’s warm’, ‘Now there’s a buzzing sound’—comments upon a changing world. What is involved in this change of significance? (Evans 1985, 249)

Though Evans is here talking about the objectivity of judgment, the same issue seems to arise for perceptual experience since, even at the level of experience, there seems to be a difference between a mere succession of sensory registrations and what we might recognize as objective perceptual experience. While the details of Evans’s theory of the conditions of objective thought may be novel, a similar contrast was expressed by Condillac in 1754.5 He presents us with a thought experiment in which we imagine a statue capable only of olfactory sensory states, which themselves are construed as purely subjective. Like Evans, Condillac asks what might distinguish the statue’s capacity for states such as mere olfactory sensation, from states which present to the subject some aspect of the physical world ‘beyond’ the subject.6 We can thus sharpen our original question a bit further. Why should we think that, on Kant’s view, the mental representations that animals enjoy consist in anything more than those ‘subjective’ ones discussed by Evans and Condillac? Such states may well causally co-vary with the animal’s environment in ways sufficiently complex to explain an animal’s behaviour. If so, then the warrant for attributing representational states to animals would not require that those states were ‘objective’ in any relevant sense.

5 See George (1981) for discussion of Condillac’s sensationism and its purported significance for Kant. 6 In McLear (2011), I distinguished two different ways in which such representations might occur, one phenomenally conscious, one merely in terms of informational access.

44

Colin M c Lear

Given that Kant denies non-rational beings any capacity for judgment or reason, and thus any cognitive faculty beyond that of sensibility, the worry that animals might only enjoy a subjective play of sensations seems a real one. Unfortunately, Kant’s exact views on these matters are difficult to ascertain, not least because his discussion of animal cognition is scattered and unsystematic. For this reason my discussion is somewhat speculative and will examine what Kant could plausibly say given his various commitments regarding the nature of cognition and the cognitive faculties. Despite these difficulties, there are two central reasons for being interested in Kant’s views on the kinds of objective states (if any) attainable by animals. First, thinking about Kant’s views concerning the cognitive lives of animals allows a perspective on the mind, as it were, ‘from below’, whereas in contrast, most of Kant’s time is spent at a position ‘from above’, dealing with the cognitive faculties of reason and the understanding.7 This alternative perspective gives us, among other things, a better sense of what is gained as one moves up the cognitive ladder to rational thought. Second, the interest of Kant’s view of the animal mind does not depend on whether any actual animal really instantiates the psychological capacities assumed by his theory. Exploring the kinds of psychological states that would be available to a creature with only the very limited cognitive capacities Kant allows to animals could provide us with an interesting picture of the limits and underpinnings of objective cognition, and rationality more generally.8 Hence, even if it turns out that the denizens of the animal kingdom are far more cognitively sophisticated than Kant would admit, his view could still offer us a baseline for understanding what is required to have a mind capable of grasping the world in any fashion at all. With these two points in mind, I aim in this chapter to examine more closely the question of whether animals could ever, on Kant’s account, enjoy objective representational states of their environment. I pursue a reply to this question, and also to the ‘nagging worry’ expressed above, by addressing Kant’s discussion of the conditions under which a mental state can be said to enjoy a relation to an object. For example, in the B-deduction, Kant says: The understanding is, to speak generally, the faculty of cognitions. This consists in the determinate relation of given representations to an object. But an object is that in the concept of which the manifold of a given intuition is united. (B137)

7 This is also how Kant sometimes presents his views on the nature of the human vs. the animal mind, e.g ML1 28:277–8. 8 One need only look at the work of contemporary cognitive ethologists to recognize the implausibility of a bright line separating rational (humans) from non-rational animals. The higher primates, whales, dolphins, dogs, and crows, all seem to have higher cognitive functioning than would readily be admitted on Kant’s sparse model of the cognitive functions available to animals. For relevant discussion, see Bermúdez (2003); Lurz (2009; 2011); Andrews (2014).

Animals and Objectivity

45

The common reading of this passage has it that Kant here proposes a necessary condition—viz. possession of a faculty of understanding—for representing any feature of a mind-independent reality.9 If the standard reading of B137 is correct then either Kant is inconsistent in allowing that animals may represent objects, or, despite the textual evidence to the contrary, he never really meant to allow such a possibility in the first place. I argue that we need not accept either of these alternatives. I do this by first differentiating between ways in which a psychological state might count as objective. I then use this framework to locate the sense in which Kant might be conceiving of relation to an object. In brief, my position is that (i) there are different grades or kinds of objectivity; (ii) Kant’s conception of a relation to an object concerns a fairly demanding notion, which is tied up with his conception of an ‘object’, or ‘objectual representation’ as I will tend to call it; and (iii) distinguishing objective from objectual representation allows the possibility for a variety of kinds of objective representational state in animals. The next section articulates various kinds of objectivity that one might deem relevant to the status of the representations of an animal mind. Section 2.3 considers the objection raised above, that ‘relation to an object’ comes only with the presence of intellectual faculties, and thus with the capacity for conceptual representation and judgment. I examine the B137 argument for this position and present an alternative interpretation that emphasizes the importance of distinguishing between acquaintance [Kenntnis] and cognition [Erkenntnis]. Section  2.4 considers, in more detail, the kinds of objective states of which an animal mind might be capable. I sketch a view according to which animal minds enjoy objective states but do not represent ‘objects’ in the sense with which Kant is concerned.

2.2 Some Varieties of Objectivity In calling a representation ‘objective’ there are several things we might mean. It will be helpful for understanding Kant’s own conception of objectivity that we have these other conceptions in hand. One overarching assumption in what follows is that the notion of a representation, understood as a mental event, state, or act, is objective in virtue of some property or properties that it possesses or lacks. A representation thus ‘inherits’ its status as objective/subjective in virtue of possessing or lacking such properties. The different notions of objectivity that I discuss spell out what those properties might be.

9 See Pereboom (2006: 160); cf. Bird (1962: 130–1); Guyer (1987: 11–24); Pereboom (2009). Longuenesse (1998: 20–1) argues that Kant distinguishes between two different relations to an object, both of which are ‘internal’ to representation (i.e. are merely intentional objects) and only one of which is genuinely objective; cf. Okrent (2006: 95–7).

46

Colin M c Lear

Perhaps the most common notion of objectivity concerns what we might construe as the methodological underpinnings of inquiry and judgment.10 Gideon Rosen provides a helpful characterization of this notion: Methodological objectivity, as we may call it, is primarily a feature of inquiries or methods of inquiry, and derivatively of the people who conduct inquiries and the judgments they form as a result. To a first approximation, we call an inquiry ‘objective’ when its trajectory is unaffected in relevant ways by the peculiar biases, preferences, ideological commitments, prejudices, personal loyalties, ambitions, and the like of the people who conduct it. (Rosen, 1994: 283)

A judgment, and by extension, the representations that constitute it, is objective in this sense just in case it is free from bias, preference, or other subjective colourings. While this is, I think, the most common notion of objectivity in use, it is not wholly relevant to our current question. For example, the processes of the visual system of an animal are not going to be distorted by ambition, ideology, or wishful thinking, since these are not the kinds of states of which animals, at least as Kant considers them, are capable. Nevertheless, there is an important sense in which an animal’s cognitive system is skewed by its preferences and biological imperatives of various kinds. For Kant, an animal, unlike a rational being, is never free in the employment of its representations. Its representations are always in the service of biological and contextual imperatives that may lead to unavoidable distortions in the animal’s cognitive relation to the world.11 For this reason, even if animals were capable of some analogue of judgment, they would be unable to achieve methodological objectivity. At work in the notion of methodological objectivity is that of a perspective. One’s methodological objectivity is jeopardized when one allows features of one’s perspective to inform or otherwise colour one’s evaluations of how things are. This notion provides us with the basis for a narrower conception of objectivity, one that is measured simply according to the degree to which the content of a representation reflects the subject’s perspective. We might further understand the notion of a ‘perspective’ here in terms of context-dependence, and contextdependence in terms of the relationship between a token representation’s (i.e. a particular mental state or event) correctness conditions and the context in which that representation occurs. A representation is thus ‘perspectivally subjective’ to the extent that its correctness conditions depend on the context of its occurrence. This is quite clear in the 10 See Feldman (1994) for extensive discussion of this notion. 11 Kant is very clear that in animals lacking spontaneity, instinct and inclination govern (CPJ 5:432; ML 29:949; cf. ML 28:588, 594, 690). In this sense, animals, unlike (rational) humans, are ‘spiritual automatons’ with nothing more than the ‘freedom of a turnspit’. For relevant discussion, see Frierson (2014: 16–17, 172–3).

Animals and Objectivity

47

case of indexical representations such as , , , and so forth.12 The correctness conditions for these representations of the world depend on the context in which they occur and mark a subject’s point of view in that context. Such representations contrast with those like , , , etc.13 If objectivity in representation is something that is attained by moving from the employment of representations whose correctness conditions depend on their context of occurrence to the employment of representations that do not exhibit such context dependence, then one question we can ask is whether animals are capable of enjoying representational states or events whose content is context-independent, at least to some degree.14 We can next articulate a further notion of objectivity, one which prescinds from the issue of context-dependence and instead focuses more generally on the relation between a correctness condition and its subject matter. According to what we might call ‘metaphysical objectivity’, a representation is objective just in case merely being in a state with the relevant content, or enjoying the relevant event with that content, is not itself sufficient for the correctness of that state or event.15 Another notion of objectivity, presupposed even by the metaphysical notion just described, is that of a representation’s having correctness conditions at all. ‘Representational objectivity’ is thus that by which a state or event could be assessed for accuracy or truth, regardless of whether or not its content has a constitutive relation to that which it represents. This is precisely that which differentiates the pure qualitative ‘feel’ of a visual sensation of colour from the notion of the colour as representing the ‘look’ of a thing under certain conditions.16 These four notions of objectivity—methodological, perspectival, metaphysical, and representational—can each be related to a more demanding fifth notion, which I’ll term ‘reflective objectivity’. According to this notion, a representational 12 I use angle brackets to denote the concepts that would normally be expressed by the italicized expression within, e.g., ‘’ is used to denote the concept of substance. 13 For a connection between the notion of perspective and objectivity in the early modern period, see Williams (1978). For an elaboration of Williams’s notion of perspective and its connection to correctness conditions, see Eilan (1997: 239). This is not the only way which Williams’s point has been understood, cf. Korsgaard (2003: 107). 14 Kant does allow that animals have reproductive imagination, so this suggests that they can generate representations independent of their immediate context. But the fact that animals can generate representations independent of a particular context of origin does not mean that the content of those representations is independent of any context. This raises questions, which I can’t fully pursue, concerning how much one should align this perspectival notion of objectivity/subjectivity with the particular/general distinction. Representations that are ‘general’ are typically taken to have their correctness conditions determined in a manner that is independent of any particular context. For relevant discussion of generality in representation, see Burge (2009: 258–72). For present purposes the important distinction is between a representation whose correctness conditions depend on its context (such as or ) and a representation whose correctness is independent of context (such as or ). 15 See Peacocke (2009) for discussion. 16 Whether representational objectivity is the most basic notion of objectivity is not clear. I discuss some reasons for thinking it may not be in §2.4 below.

48

Colin M c Lear

state is objective just in case it meets some condition C, one is in a position to articulate that condition, and one knows (or is in a position to know, tacitly or explicitly) that the representation or judgment meets that condition. This notion of objectivity is by far the most demanding, as it requires the subject to not only meet some condition, but be in a position to reflectively grasp that they so meet this condition.17 Only very sophisticated cognitive beings are going to be able to do this, and certainly, in Kant’s view, the conditions of reflective objectivity cannot be met by non-rational animals. With these five different varieties of objectivity in mind, I next move to Kant’s discussion of objectivity, construed as a ‘relation to an object’, and made possible by virtue of the deployment of conceptual categories by a subject with the capacity for reflective self-consciousness (i.e. Kant’s ‘unity of apperception’).

2.3 Objectivity and Relation to an Object As I noted in §2.1, Kant’s discussion of objective representation in §16 of the B-deduction presents a prima facie challenge to the possibility of objective representation in non-rational beings. Let’s look at the relevant passage in full: The understanding is, to speak generally, the faculty of cognitions. This consists in the determinate relation of given representations to an object. But an object is that in the concept of which the manifold of a given intuition is united. Now, all unification of representations requires the unity of consciousness in the synthesis of the same [i.e. the representations]. It follows that the unity of consciousness alone establishes [ausmacht] the relation of representations to an object, their objective validity, and thus their becoming cognitions. (B137; see also A197/B242–3)

The argument consists of three premises and a conclusion: 1. Cognition consists in the determinate relation of a given representation to an object. 2. An object is that in the concept of which the manifold of a given intuition is united. 3. All unification of representations requires the unity of consciousness in the synthesis of those representations. 4. Therefore, the unity of consciousness alone establishes the relation of representations to an object resulting in a cognition. 17 For the expression of a position along these lines, see Strawson (1966: 107–8).

Animals and Objectivity

49

The validity of the argument has been disputed.18 However, what interests me here is the sense in which Kant construes representation of an object in terms of cognition and the ‘unity’ of intuition. To that end, in what follows I examine premises (1) and (2) and explain what I take Kant to mean by arguing that cognition requires the unity of consciousness in establishing the unity of the ‘manifold of a given intuition’ of an object.19 My central claim is that it is compatible with a state’s lacking ‘relation to an object’, and so not qualifying as a ‘cognition’ in the sense above, that it nevertheless possesses a degree of objectivity. While versions of this argument have been made elsewhere, unclarities and misunderstandings remain.20 I address some of these below.

2.3.1 Premise One: Cognition vs. Acquaintance In the Jäsche Logic, Kant distinguishes different ‘degrees’ of cognition.21 Of particular interest to us is his discussion of the difference between the third and fourth degree or step: The third: to be acquainted with something (noscere), or to represent something in comparison with other things, both as to sameness and as to difference; The fourth: to be acquainted with something with consciousness, i.e. to cognize it (cognoscere). Animals are acquainted with objects too, but they do not cognize them. (JL 9:64-5)22

Kant distinguishes here between two sorts of cognitive connection to an object, where the second builds on the first. He distinguishes cognition from mere ‘acquaintance’ (notitia; Kenntnis), in which the subject is able to differentially discriminate between objects or their parts and compare them with respect to their characteristics. Kant does not specify here what cognition comes to other than

18 See, for example, Allison (2004: 174–6); Pereboom (1995: 20–5); cf. Pereboom (2006: 160); Guyer (1987: 117–18). 19 For discussion, see McLear (2015: Sec. 3). 20 For argument that Kant is concerned with cognition in judgment rather than intuition of a particular, see Hanna (2005); cf. Allais (2009: 392–3, 2011: 104). The notion of ‘cognition’ (Erkenntnis) has itself come under significant scrutiny of late. See Schafer (forthcoming); Willaschek and Watkins (2017a), (2017b) for discussion. 21 This is one of a variety of ‘stepladder’ (Stufenleiter) passages that Kant gives in his logic lectures. There is a very similar one in the Dohna-Wundlacken Logik (c.1792), DWL 24:730–1. There are also several in the earlier Blomberg Logik (c.1771), BL 24:132–3, 134–5, 136. Kant also makes claims in the metaphysics lectures concerning the abilities of animals to compare representations (e.g. Metaphysik Mrongovius (c.1782/3) MMr 29:888). 22 Similar passages are found in other logic lectures. See e.g. Dohna-Wundlacken Logik (c.1792) 24:730–1; Wiener Logik (1780), VL 24:846; Blomberg Logik (c.1771), BL 24:132–3, 134–5, 136).

50

Colin M c Lear

that it is acquaintance with the addition of consciousness. Given that the context of this discussion concerns the importance of clear and distinct consciousness to cognition and logic (9:61–4), it seems reasonable to seek clarification for Kant’s distinction by looking to other discussions he provides of the notion of consciousness and its connection to representation. In the pre-critical work The False Subtlety of the Four Syllogistic Figures Kant argues for a distinction between: logical differentiation, which means to cognize that a thing is A and not B, and is always a negative judgment, and physical differentiation, which means being driven to different actions through [one’s] different representations. (FS 2:60)

Kant argues that, for example, a dog may physically but not logically differentiate (2:60). Moreover, it is, he says, one thing to differentiate (whether physically or logically), but ‘it is something else entirely’ to cognize the difference between things (2:59). This distinction between differentiation and cognition appears to  closely match the distinction between acquaintance and cognition in the Jäsche Logic. In Kant’s critical period, he makes a similar distinction between two different ways of discriminating with respect to similarity and difference and connects these ways with the degree of consciousness possessed by the representation. This distinction occurs in an important footnote to the Paralogisms discussion of the B-edition. Kant states that: Clarity is not, as the logicians say, the consciousness of a representation; for a certain degree of consciousness, which, however, is not sufficient for memory, must be met with even in some obscure representations, because without any consciousness we would make no distinction in the combination of obscure representations; yet we are capable of doing this with the marks of some concepts (such as those of right and equity, or those of a musician who, when improvising, hits many notes at the same time). Rather a representation is clear if the consciousness in it is sufficient for a consciousness of the difference between it and others. To be sure, if this consciousness suffices for a distinction, but not for  a consciousness of the difference, then the representation must still be called obscure. So there are infinitely many degrees of consciousness down to its vanishing. (B414–15n)

Kant makes two important points here. First, he argues that we should not equate consciousness with clarity, as we need to appeal to consciousness to explain an organism’s discriminatory behaviour even if that organism is incapable of having clear representations. Second, Kant says that clarity cannot consist simply in

Animals and Objectivity

51

differential discrimination.23 It must instead be correlated with an awareness of the basis of a correct discrimination. In other words, a being with clear representation has to represent not only that two (or more) things (or properties) are distinct or similar, but what it is about them that is the basis of that distinctness or similarity. This awareness of the ground of a similarity or difference is what I take Kant to intend by his distinction between acquaintance and cognition in the Jäsche Logic, as well as his pre-critical distinction between kinds of differentiation in False Subtlety. It is thus one thing, on Kant’s view, to differentially discriminate. It is another thing to be able to articulate the basis of such discrimination.24 If this is correct, then we can read Kant’s use of ‘cognition’ in premise (1) as claiming that a subject’s awareness of the basis or ground of difference and similarity requires positing things—objects—that are the loci of these differences and similarities. Premises (2)–(3) claim that this sort of awareness requires the unity of apperception, resulting in the conclusion that the unity of apperception is required for awareness of the grounds of identity and difference amongst a subject’s representations (or their contents)—i.e. cognition. Kant’s distinction between acquaintance and cognition might seem implausible, for the reason that it seems to construe acquaintance as involving an ability to brutely discriminate without any awareness of what it is that moves one to do so.25 For example, Kant might seem to be saying that one could discriminate between red and green coloured objects without being aware of their respective colours. However, Kant need not be taken as making such a claim, and is in fact indicating an important fact about ways in which a subject’s sensitivities to incoming information might lead to knowledge. Consider, for example, the famous ‘chicken sexer’, whose ability to sort male from female chicks (a form of reliable differential discrimination) is done without conscious inference, and without the subject’s ability to fully articulate the means by which they make these discriminations, or the features in virtue of which the discriminations

23 This contrasts with, e.g. Wolff, who claims that ‘we are conscious of things when we distinguish them from one another’ (German Metaphysics §729: 454). See also Wunderlich (2005: Ch. 2); Dyck (2011: 45–6); (2014: 33, 106). 24 For related discussion, see Callanan (this volume: §1.5); see also Wunderlich (2005: 141–2); Kitcher (2011: 18–19); Dyck (2011: 47); McLear (2014a: 771–2). 25 Okrent (2006) claims that animals can ‘note that two objects differ, but they can never note or notice how they differ, or the way in which those objects differ’ (104). This way of putting things is, perhaps, slightly misleading. First, animals cannot note that one thing differs from another, because they cannot, according to Kant, enjoy propositional attitudes. So the extent to which an animal responds to identity and difference is not due to its having attitudes to propositions. On this point, see Dummett (1993: Ch. 12) and related discussion in Bermúdez (2003: Ch. 3). Second, Okrent’s discussion might encourage, without addressing, just the kind of notion of a brute discriminatory capacity that I discuss above.

52

Colin M c Lear

hold.26 We need not deny to the sexer some phenomenal awareness of the bases of their discriminations, and we certainly cannot deny their sensitivity to the information upon which they are basing their choices. What chicken sexers lack is any further capacity to integrate that sensitivity (and thus the information to which they are sensitive) in the variety of ways that are constitutive of rational thought and action.27 It thus seems reasonable to say that the sexer’s awareness is of the world, even if they are unable to adequately articulate that in virtue of which they make their verdicts. Thus far I’ve argued that Kant construes ‘relation to an object’ in B137 in terms of cognition of it, where this in turn should be understood as requiring representation of features of an object that are the basis of discriminations of similarity and difference. I’ve suggested that, in contrast to cognition, acquaintance is the awareness of similarity and difference amongst objects without the ability to recognize or articulate the basis on which one is so aware, or in accordance with which one behaves.28 I’ve argued that this need be construed neither as some sort of brute sensitivity to difference, nor as something purely subjective, but rather as an ability that is importantly different from and more basic than the ability to articulate such differences. I suggested that an example of this is the chicken sexer’s ability to sort male from female chicks on the basis of non-inferential, finegrained perceptual discriminations that outrun the subject’s capacity for explicit articulation. Another example of sensitivity to similarity and difference that might also be relevant to Kant’s distinction between cognition and acquaintance is the way in which a subject might be aware of features of an object without being aware of them as properties of the object. For example, a subject might be aware of colour in the sense that colour is used to discriminate an object from others (or an object from its background) without the subject thereby employing a representation of the object’s colour as the relevant feature by which it performs the discrimination. For example, consider the Ishihara test for colour blindness, in which a subject differentiates one object, a numeral (e.g. ‘16’, ‘9’, etc.) consisting of coloured dots, from others and its background.29 The Ishihara test presents us with an example of a subject’s ability to discriminate a numeral from its background and from other figures. If the only available explanation of what is going on in such a case was one according to which the discrimination of the numeral from its 26 See Biederman and Shiffrar (1987). 27 One way of putting this is that a chicken sexer has no ‘access consciousness’, in the sense articulated by Ned Block, to those features by means of which they make their discriminations. See Block (1995); (2008); cf. Roessler (2009), (2011: 113–14). 28 In the terms articulated by Schafer (forthcoming), acquaintance provides a ‘non-trivial standard of material objective correctness’ but only in cognition can the subject represent that standard itself. 29 See Campbell (2006: 32); cf. Campbell (2002: 30–1); Roessler (2009: 1028) for relevant discussion.

Animals and Objectivity

53

background requires the attribution of the property yellow to a particular object in the dot array, then it wouldn’t be clear how Kant could allow such a possibility for animals. But not all discrimination requires attribution/predication or the awareness of a feature as a property—i.e. as the kind of thing that one thing has while another lacks, and thus could be the kind of thing that could be instantiated by more than one object, or by different objects at different times, or by one object at one time but not at another.30 Hence the fact that a feature may be a focus of awareness and a means of discrimination is compatible with the subject’s being unable to identify the feature as one in virtue of which the object is so differentiated, nor even to represent that feature as a property of the object. Kant’s distinction between acquaintance and cognition reflects this fact. One might object here that even if an animal can discriminate between objects on the basis of a non-conceptual representation of their features, if it does not also possess the capacity to articulate the feature of which it is aware as a feature of an object, it cannot be credited with an objective representation. In reply to this, the objection confuses what I have called ‘reflective objectivity’ with any one of the other notions (such as representational or perspectival objectivity). Perhaps there is some reason to tie the capacity for (e.g.) representational objectivity to that of reflective objectivity, but I don’t believe that we see in Kant a reason for thinking this. Kant may well think that cognition requires the capacity for reflective objectivity, but this is not equivalent to the former stronger claim concerning representation generally. Hence, it leaves open the possibility that a subject might have acquaintance with an object without possessing the requisite cognitive capacities for cognizing it. The distinction between cognition and acquaintance also allows us to address an objection to the claim that, for Kant, animals could have intuitions of their environment. Stefanie Grüne (2014b) has argued that a necessary condition of enjoying intuitions, understood as objective sensory representations, is that they be conscious.31 Animals, according to Grüne’s interpretation, do not have conscious states, hence animals cannot have intuitions. Grüne cites three reasons for her position.32 First, she argues that it is not clear that Kant allows animals inner sense (e.g. FS 2:60; LM 28:277), and possession of an inner sense is a necessary condition of having any intuition at all. Second, she argues that Kant’s letter to Herz of 1789 shows that ‘even though Kant [here] allows animals some kind of consciousness of sensible representations, this kind of consciousness is not 30 Another way of putting this is that animal cognition, on Kant’s view, does not satisfy the generality constraint (see Evans 1982: Ch. 4.3). Some recent work suggests that at least certain forms of animal thought or representation do not obey the generality constraint. See Beck (2012); see also Golob (this volume). 31 For a similar objection based primarily on Kant’s pre-critical work, see Leland (2018a, 2018b). 32 See Grüne (2014a: Sec. 2).

54

Colin M c Lear

sufficient for the having of intuitions’.33 Finally, she argues that, even if an animal could be conscious of its sensory representations, they could not display the unity necessary for achieving the status of intuition. I take these points in turn. The first point is easily dealt with. As Grüne herself admits, the relevant texts that seem to deny animals an inner sense are primarily from Kant’s pre-Critical period, at which time he appears to conflate inner sense with apperception. Once this distinction is made in the early 1780s it is not at all obvious that Kant wants to deny inner sense to animals, but rather only the capacity for self-reflection and self-reference as befits possession of the first-person concept.34 Concerning the second point, one should first recognize, as Grüne herself does, that in at least three places Kant allows for the existence of unconscious intuition (7:135; 16:88; 23:19.). Second, it isn’t clear that Kant thinks of consciousness as an all or nothing property of a mental state. As we saw in the discussion above of acquaintance and cognition, Kant believes (at least by the 1787 edition of the Critique) that some degree of consciousness is necessary for any discriminatory responsiveness by a subject to its environment. Given that Kant does think that animals have such discriminatory abilities, that these abilities are best explained by their having representations, and that he explicitly attributes a form of conscious representation to animals—viz. acquaintance—there is nothing about Kant’s account of consciousness per se that would require us to deny that animals have intuition that is to some degree conscious. Finally, concerning the issue of unity, Grüne argues that the objective character of an intuition should be understood in terms its being a kind of intentional state brought about via acts of (conceptually) rule-guided synthesis on non-intentional sensory states.35 Grüne bases this claim in part on the canonical ‘Stufenleiter’ passage in which Kant distinguishes different types of representation (A320/B376–7). Against this, note that the Stufenleiter passage is a passage concerning different kinds of ‘representation’ (Vorstellung). It says nothing of ‘intentionality’ or the concomitant notion of a state’s possessing correctness conditions (more on this issue below). Grüne also cites36 an important note from Kant’s Nachlaß where he says: What is an object? That whose representation is a sum of several predicates belonging to it. The plate is round, warm, made of tin, etc. Warm, round, being made of tin, etc., are not objects, although the warmth, the tin, etc., indeed [are]. An object is that in the representation of which various others can be thought as synthetically combined . . . (R 6350, 18:676)

33 Grüne (2014a: Sec. 2). 35 Grüne (2009: 40).

34 For further discussion, see McLear (2011: secs. 3–4). 36 Grüne (2009: 41).

Animals and Objectivity

55

According to Grüne, relation to an object thus consists in the representation of particular features as ‘unified’ in one subject of inherence. I think it plausible to construe intuition as a complex representation, and that it is equally plausible that representation of a subject of a multiplicity of properties—in effect, a substance—is going to require, for Kant, apperception and synthesis. But I don’t see why complex representation per se must depend on the understanding or the categories (and thus on synthesis). It is true that Kant construes a particular kind of unifying function—viz. ‘combination’ (Verbindung)— as coming about solely via spontaneous mental acts of synthesis. However, this means of generating complexity can only be read in a narrow manner, since Kant does not include association in the category of synthetic acts, and association would seem to allow for the generation of representationally complex states. Nor is it either textually or philosophically obvious that the kind of representational capacities or ‘unifying’ functions needed for the awareness of (e.g.) colour and shape at a particular spatial region require synthesis, so long as we are careful to distinguish such mental states from acts of predication/attribution or the representation of some thing as the occupant of a spatial region and subject of properties. I discuss these considerations further in the next subsection. Finally, there is reason to wonder whether the notion of unification to which Grüne appeals in the generation of intuition is really tied to a faculty of spontaneity, and so to a faculty that animals lack. Grüne seems willing to characterize possession of the primitive capacity for grasping the content of an intuition as a ‘unity’ in ‘wholly non-intellectual terms’.37 She also suggests that the possession of such a capacity should be understood as the cognitive analogue of the possession of the capacity to digest meat.38 But if this is correct, then it is no longer obvious that we’re talking about spontaneity as Kant conceives of it—viz. as a form of free activity. Thus it’s no longer obvious that we’re talking about a capacity that, according to Kant, is lacking in animals and which makes the grasp of higher cognitive unities—such as ‘relation to an object’ in the sense relevant to the argument of the Transcendental Deduction—possible. So much, then, for explanation of the first premise. In B137 relation to an object is tantamount to cognition of it, which is itself understood in terms of the recognition of the basis of similarity and difference in the object (and between it and other things). This does not, however, exhaust Kant’s conception of objectivity, for he goes on, in the second premise, to argue that an object is that ‘in the concept of which a manifold is united’. This is directly related to the last of Grüne’s concerns discussed above. The question is whether it makes sense to think of the sensory experiences of animals as intuitions when such states lack the kinds of ‘unity’ characteristic of the categories. It is to this issue that we now turn. 37 Grüne (2009: 202, note 16).

38 Grüne (2009: 41).

56

Colin M c Lear

2.3.2 Premise Two: The Unity of Intuition Two examples of the kind of cognitive unity that Kant had in mind, and which are especially illustrative of the cognitive limitations facing a subject lacking the categories and the capacity of judgment, are those of and . Without the concept , a cognizer is unable to represent the relational property of inherence, and thus unable to represent a property instance as ‘belonging to’ a subject of inherence. Such a cognizer could represent a particular property instance at a place or time (e.g. redness there now), but not as instanced in some subject which itself exists at that place or time. Relatedly, without the concept a cognizer is limited in the extent to which they can identify or re-identify an object, because they cannot appeal to causal connections between a thing’s properties (or between the thing and its properties) in order to explain that (e.g.) a rose that blooms today is the same as the green vine that grew last week.39 The lack of such categorial concepts in animals is not just a lack of the ability to make particular kinds of judgment. We should read Kant as denying that animals have the capacity for particular kinds of experiences (in our contemporary sense of that term, not just Kant’s technical sense). The reason for this denial is that Kant is clear, especially in the second edition of the first Critique, that imagination plays a role in relating intuitions (or the contents thereof) to one another, and that this is supposed to be governed, in adult rational humans, by a nonjudgmental application of the categories (e.g. A101, A119, B151ff).40 If the above is correct, then the intuitive sensory representations of animals, and thus their perceptual experiences, are going to be limited in various ways related to lacking cognitive capacities marked by the various respective categories. I have here elaborated what I take to be two absolutely central aspects of rational human experience as Kant conceived it—viz. (i) the perceptual awareness of the properties one perceives as inhering in a subject and (ii) the causal connectedness of such a subject, both to itself across time and through change, and to other things and their properties. If an animal lacks the ability to represent inherence relations, or even the causal relations necessary for representing the subject of inherence or its relation to other such entities, is there anything left of the animal’s capacity for objective 39 See Campbell (1994) for a similar point. 40 For one such articulation of how this should be understood, see Land (2015b). It seems to me that Land errs in claiming that it is intuition per se that is generated by such acts, rather than representation (perhaps ‘perception’ (Wahrnehmung)) of a certain kind or complexity. Relatedly, I disagree with Land (2015a) insofar as he there construes so-called ‘nonconceptualist’ positions as restricted to making claims about perceptual judgment rather than intuition or cognition. I don’t see that this is the case, so long as we can distinguish between different levels of complexity in a subject’s representations and degrees or types of consciousness. Nothing in the non-conceptualist position, or in Kant’s writings, seems to prohibit this.

Animals and Objectivity

57

representation? I believe there is. Certainly, the nature of what is experientially given to animals is only going to have the most basic congruence with the experience of an adult rational human. The principles that the animal’s cognitive system will utilize to single out particulars in perception are going to be very basic, and in some ways, quite coarse-grained.41 They will include, for example, spatiotemporal continuity—that a particular object trace a single spatiotemporal path, and spatial cohesion—that an object’s surface and outer boundary be determined by the spatial proximity of its parts.42 Though principles like cohesion and continuity are in some ways coarsegrained, they nevertheless allow for the presentation of spatial particulars rather than mere serial arrays of qualities. For example, in being visually aware of a coloured sphere, an animal could do more than simply have sensations of redness and roundness.43 It could sense the redness and roundness in a particular object, albeit one which is not presented as a subject of properties, but rather simply as a continuous expanse of colour at a specific location. Thus Kant need not be understood as denying that animals can perceptually ‘bind’, both inter- and intra-modally, the various qualities of which they are aware. But the principles by which such binding occurs are not always going to result in perception of particulars in the same sense in which we enjoy perception of particulars, since we can utilize much more sophisticated conceptual capacities, which may well affect our sensory experience (e.g. in the case of abstract art or atonal music).44 There will thus be some inherent indeterminacy with respect to what perceptions an animal actually enjoys. For example, does it perceive the rock and the plant growing from it as one particular or two? When it sees a cup placed on a table does it see two objects go out of existence and a new, complex object, come into existence?45 It is not obvious that there are answers to these questions, or if there are, how we could know them. But there remains a genuine sense in which the perception of animals presents ‘unified’ entities, at least in the sense of entities whose integrity is based on principles like those of continuity and cohesion. 41 See also Golob (2011) for related discussion of the ‘basic measure’ with which intuition provides a cognizing subject. 42 The centrality of principles of continuity and cohesion to our representation of particulars has received a significant amount of empirical attention and confirmation. For representative discussion, see Spelke (1990); Spelke, Vishton, and Von Hofsten (1995); Cheng and Newcombe (2005); Cheries et al. (2008). For further discussion of these issues in the context of Kant’s conception of intuition, see Dunlop (2017). 43 While there are some similarities between the account I propose here and the account of animal representation discussed in Burge (2010a), one significant difference is that Burge conceives of animal representation as have predicative structure while I take Kant to be denying such a possibility. A further issue concerns whether the animal mind should primarily be conceived as a representational mind or as a conscious mind. See Eilan (2011, 2017) for relevant discussion. 44 For further discussion of the issue of binding with regard to intuition, see Allais (2017); for a dissenting view from the one offered here, see Dunlop (2017). 45 See Allais (2009: 406) for a similar point.

58

Colin M c Lear

2.4 Objectivity and Correctness So far I have argued that while animals may be acquainted with objects, they cannot cognize them, where this means that while they can represent features of things so as to reliably discriminate between objects that are similar or different, they cannot represent the bases of their discriminations of similarity and difference. I further claimed that without the concepts of and at their disposal, animals would be unable to engage in predicative acts, and so unable to predicate properties as belonging to objects. Thus, they would be unable to employ anything more than the most basic principles of cognitive unification, such as spatial and temporal continuity and cohesion, in making determinations as to the identity or difference of objects in their environment. What then can be said of the objective status of sensory representation in such animals? According to one prominent account by Lucy Allais, with which I largely agree, animals may nevertheless be said to be perceptually aware of particulars in their environment. Allais puts things this way: the non-human animal (assuming it lacks concepts but has some way of representing space) that perceives its environment represents the world in the sense that it has relational mental states that present it with parts of the world—it does not have an inner display of non-intentional, raw sensations. However, it does not represent an objective world in the sense that it does not represent the world as a law-governed complex that it thinks of at a detached level as existing unperceived/independently of it.46

Allais conceives of the perceptual states of animals as (according to Kant) objective in the sense that they are not ‘of ’, and do not present, mere phenomenal qualitative states of the animal—akin to pains, tickles, or afterimages. Instead, animals are presented in outer sense with spatially located particulars. While I agree with the basic idea expressed here, it leaves open two different ways of construing the issue of objectivity. According to the first, animal states are objective because they attain what I above termed ‘representational’ objectivity—i.e. they have correctness conditions in virtue of representing particular sensory qualities as being present in particular spatial expanses at specific locations in space. This is in contrast with ‘raw’ sensations, which entirely lack correctness conditions. There is no correctness condition integral to having the sensation of, for example, pain— there is just the pain. According to the second way of construing Allais’ point, the sensory experiences of animals are objective in virtue of that which they immediately 46 Allais (2009: 406); cf. McLear (2011; 2015: 98–101).

Animals and Objectivity

59

present to consciousness.47 On this way of construing things, whether or not the sensory states of an animal possess correctness conditions, the key issue is what they immediately present to the consciousness of the animal. In terms then of what is presented, in the subjective case there is only a ‘raw’ sensation, whose nature and existence depends entirely on the subject having it. In the objective case, what is presented is (or includes) a public spatial particular, whose nature and existence is (at least to some degree) independent of the subject enjoying the experience. Call this notion of objectivity ‘presentational’ objectivity. At this point I want to elaborate what I take to be a very tempting way of combining the framework for objectivity I articulated above in §2.2 with the line of thought introduced by Allais. Ultimately though, however attractive the line of thought may be, I argue that we should not give into this temptation, for it depends on attributing to Kant a conception of the representational status of sensory experience that he does not plausibly endorse. According to this tempting line of thought, the best way to reconcile representational with presentational objectivity is by positing that what explains presentational objectivity is precisely that one’s states are at least representationally objective, for (the line of thought goes) it is the possession of correctness conditions that distinguishes purely qualitative phenomenal states from intentional states that are ‘directed at’, and so are able to present, the world. According to this line of thought, in the case of sense perception, it is plausible that it is at least partially in virtue of having spatial form that one’s sensory states take on their representational significance.48 Given the framework for objectivity outlined above, one would then have the resources to explain the objectivity of animal perception as follows. The sensory states of animals, let us assume, in virtue of having spatial form, thereby possess correctness conditions (e.g. representation of shape Ø at location ), and thus have representational objectivity. Since what is represented by the subject in such an experience is itself a public spatial particular, the experiential state also possesses at least some degree of metaphysical objectivity—the correctness conditions that specify the particular do not themselves make it the case that the particular is the way that it is (or possesses related spatial features, such as relative location, motion, etc.). Finally, because Kant regards spatial features as related to the form of intuition, and thus as independent of any specific sense modality by which they may be known, the properties experienced by an animal are not entirely parochial, and thus allow for at least some degree of perspectival objectivity. What animals fail to achieve, and on Kant’s view could never achieve, 47 Allais (2009: 389). 48 The notion that spatial representation is fundamentally basic looms large in the work of Strawson and those influenced by him. See especially Strawson (1959: Chs. 1–2); Strawson (1966, Chs. 1–2); Evans (1982); Peacocke (1992: Ch. 3).

60

Colin M c Lear

is any measure of methodological or reflective objectivity. These come only with the capacity for reflection, judgment, and reason. This picture is attractive in that it allows that there is a difference between ‘raw’ sensation and intuition, that intuition comes about via some kind of a priori imposition of spatial form on sensory matter (here we might speak in contemporary terms of cognitive processing), and that the resulting experience is one which explains the animal’s (potentially very fine-grained and sophisticated) abilities to differentially discriminate between objects and track them over time for, among other things, the purposes of satisfying the animal’s biological needs. As attractive as this picture of perceptual objectivity is, I have doubts about its wholesale applicability to Kant’s conception of animal representation. These doubts stem primarily from the difficulties that come with assuming that Kant construes the objective status of sensory representations in virtue of their possessing correctness conditions, and thus as being representational states in our contemporary sense. Call this assumption concerning the representational nature of sensory experience the Content View.49 This is an especially difficult view to sustain concerning beings lacking spontaneity. Kant’s primary notion of a correctness condition is a truth condition, and judgments, for Kant, are the sole vehicles of truth in virtue of being the product of ‘relating’ representations in one consciousness: The unification of representations in a consciousness is judgment . . . thinking is the same as judging or as relating representations to judgments in general. (Prol. §22 4:304; cf. JL §17 9:101; LL 24:928)

What kinds of representations are related in one consciousness? Kant specifically has concepts in mind here (cf. CJ §35 5:287; B146, B283; JL 9:101; LL 24:928). Judgments consist of concepts that, due to an act of the mind in which they are unified in one consciousness, are brought together to form truth-bearing contents (I leave open how exactly the transcendental unity of apperception accomplishes this). We may contrast the logical relations in which representations stand in an act of objective judgment to the manner in which representations are related in a sensory event or act. In sensory experience representations are related to each other non-logically, and merely as to their form in either space or time. Hence, their logical combination is not given, but rather made. This is, I think, Kant’s point in §15 of the second edition version of the Transcendental Deduction. There Kant says:

49 For discussion, see McLear (2016b).

Animals and Objectivity

61

a manifold’s combination [Verbindung] as such can never come to us through the senses; nor, therefore, can it already be part of what is contained in the pure form of sensible intuition. For this combination is an act of spontaneity by the power of representation; and this power must be called understanding, in order to be distinguished from sensibility. (B129–30)

If the vehicle of truth is judgment and judgment requires (i) the combination of representations (concepts) in one consciousness, and (ii) such combination can never be given in or otherwise accomplished by sensibility, then it seems that sensibility alone could never yield representations that possess a truth value.50 However, even if the above argument is correct, one might nevertheless argue that intuition possesses correctness conditions of its own, i.e. accuracy conditions, which should not be assimilated to the kind of correctness conditions indicative of judgment (i.e. truth conditions).51 Unfortunately, I do not see any positive textual basis for such an assertion. Kant nowhere, to my knowledge, speaks of correctness conditions in any other terms than truth. Moreover, if Kant did endorse such a view one would expect it to play a role in his explanation of perceptual illusion and hallucination, since it is precisely cases of illusion and hallucination that are often thought to demand explanation in terms of a mental state’s possessing correctness conditions. Kant is, however, explicit in denying that perceptual illusion and hallucination be treated in terms of the correctness or incorrectness of sensory representation. Instead Kant endorses a doxastic theory of perceptual error. Error is a product of the relation of the object to the understanding—i.e. in the object as it is judged (A293–4/B350; An §11 7:146).52 One might, at this point, voice the following objection. If intuition does not possess correctness conditions, then it can offer no constraint on determining what object an animal has in mind, for it is only in virtue of such conditions that we can answer the question as to whether an animal has in mind one thing as opposed to another. In particular, the cognitive state would need to offer some material condition for identifying the object of one’s perceptions, and for attributing to it some determinate quality or set of qualities.53 We can see our way to answering this worry if we return to Allais’ distinction between subjective and objective. Recall that presentational objectivity conceives of the subjective and objective aspects of a mental state in terms of what is presented. According to a representationalist reading of Kant, which ascribes to him the Content View, the notion of 50 For further discussion of these issues, see McLear (2016b: Sec. 4.2). 51 See Grüne (2009; 2014a) for an explicit version of this strategy. Land (2015a: Sec. 4; 2015b) also attempts to articulate a non-judgmental basis for the representational content of intuition. 52 Kant is also recorded as asserting this in numerous texts from his logic lectures, cf. LL 24:83, 84, 87, 103, 146, 156, 720, 813, 825, 833. For further discussion of these points, see McLear (2016b). 53 Schafer (forthcoming) argues for just such constraints on cognition.

62

Colin M c Lear

‘what is presented’ is best understood in terms of the state’s possession of a correctness condition. I have presented reasons for thinking this an ill-fitting characterization of Kant’s view. On the alternative I propose, we take the notion of what is presented as exactly that, i.e. as the specification of some element or aspect of the subject’s environment. On this view, individuation of an animal’s mental states in sensory experience, and specifically outer intuition, makes essential appeal to the particulars in its environment that are presented to it.54 Intuition is thus an essentially relational state in which the subject is acquainted with some tract of their current environment. With respect to the notion of a correctness condition, the intuited object presented to the subject is what makes it the case that their states (e.g. in more cognitively sophisticated subjects, their beliefs) have correctness conditions. Thus rather than being correct or incorrect, intuition supplies a standard of correctness for cognition.55 As I’ve indicated, exactly what the standard of correctness is may be difficult to determine in the case of an animal mind, given the coarse-grained character of the principles of individuation operative for the animal. The key point is to keep acquaintance and cognition distinct by distinguishing between that which supplies the standard of correctness—viz. the intuition of the object— and the correctness condition itself—viz. the proposition that the object is thus-and-so. Despite rejecting the Content View in favour of a form of perceptual acquaintance, we can nevertheless maintain much of the framework of objectivity of which the Content View makes use. On the representational way of framing things, I suggested that, despite the cognitive limitations Kant places on animals, they could possess representational, metaphysical, and (to a limited degree) perspectival objectivity. However, even on an interpretation according to which Kant rejects the Content View, we could still see him as accepting that all three of these notions (could) hold for animals. In the case of representational objectivity, it will trivially be true that when an animal (e.g.) perceives a mossy rock, the animal’s state is such as to be correct if and only if there is a rock there to be seen. Since the notion of representational objectivity does not specify what makes it the case that a given state has correctness conditions, nothing stops the Acquaintance View from helping itself to the notion. It is also true that the subject could not be in a state of acquaintance with their environment unless there was something present to them with which they could be so acquainted. But the subject does not entirely generate or otherwise make it  the case that the acquaintance relation obtains. So there is a degree of

54 It may of course appeal to other things as well, such as the subject’s position, and the relevant sense modality. 55 There are obvious issues here regarding hallucination and illusion. I discuss these issues in McLear (2016c). For further discussion of the notion of acquaintance, see McLear (2016b; 2016a).

Animals and Objectivity

63

metaphysical independence between subject and that with which they are acquainted. While there may be a question as to the ultimate metaphysical independence of empirical reality from the intuiting subject, the intuitions of animals are in no way worse off with respect to metaphysical objectivity than those of rational (human) beings. Finally, the acquaintance theory can accommodate perspectival objectivity as follows. The animal’s state of acquaintance is determined in part by their standpoint with respect to the object of perception, as well as the sense modality involved. We can then assess the degree to which the animal’s state achieves perspectival objectivity by looking to the parochial nature of either its sense modalities or standpoint. So, for example, to the extent that an animal shares our forms of intuition, it will achieve some level of objectivity. In sum, the world which comes into view on Kant’s conception of the animal mind is a world of particular qualities bundled or unified according to basic cognitive principles such as spatial continuity and cohesion or proximity. In the sense of ‘object’ with which Kant is most concerned—viz. the object as a persisting real subject, or substance, of properties, with causal powers that put it in community with other such substances—in Kant’s view animals lack any capacity for awareness as of such things. There is thus a relevant sense in which the animal’s world is a world without objects. But, if the discussion above is correct, the contents of the animal mind do not merely consist of subjective sensory states. Animals, on Kant’s view, can enjoy states (or events) that are minimally objective, even if the particulars so presented to them in perception are not the rich particulars of adult rational human experience. Animal intuitions are objective even if not objectual. This way of understanding Kant also has further benefits.56 First, it respects the fact that Kant conceives of humans as radically different from animals, in virtue of their possessing spontaneous faculties of reason and understanding. This difference is one of kind and not merely degree (ML 28:276; cf. LL 24:702). In one set of student lectures, Kant even goes so far as to differentiate between the kinds of imaginative faculty that might be possessed in animals as opposed to humans: All three of these cognitive faculties [i.e. imagination, imagining, and anticipation, which have their source in the reproductive imagination] can be accompanied by apperception or not. When they are, then they belong only to human beings, when not—then animals also have them. We ought, therefore, to have two different names for these, but for this [faculty] there is only one, namely, the reproductive power of imagination. (ML 29:884)

56 Thanks to Thomas Land and Samantha Matherne for discussion concerning the following two points.

64

Colin M c Lear

If what I have argued for above is correct, then Kant can reasonably say that the integration of the spontaneous faculties of reason and understanding radically changes the nature and extent of an animal’s cognitive life (amongst other things by changing the way the faculty of imagination may operate). But we need not understand this addition as one that makes the difference between mere awareness of subjective states and awareness of one’s environment.57 Second, this manner of construing animal cognition presents an important challenge to readings of Kant, which I have elsewhere called ‘Intellectualist’—viz. readings that assume or argue for a dependence of objective states on the synthetic activity of the higher cognitive faculties.58 The import of this challenge does not depend on whether Kant had animals in mind while, for example, composing the Transcendental Deduction, but rather on the fact that Kant explicitly and consistently allows for the possibility of objective cognitive states (in the sense I’ve outlined above) in animals throughout his career. It is difficult to understand how Kant could have even considered this a logical, let alone real, possibility if he held the position that objective representation depends on the possession of a capacity for spontaneity and its concomitant faculties.59 This point holds even if Kant recognizes, as I have argued he does, that animals are importantly different from human beings with respect to their cognitive powers, and that something is importantly lost when we move from rational to non-rational awareness of the world. This ‘something,’ however, is not the world itself. In being unable to recognize the grounds of similarity and difference of which they are aware, animals are blind to structural features of the natural world, the awareness of which is necessary (among other things) for scientific understanding and knowledge. But Kant does not seem to take this blindness as cause for denying them the possibility of being aware of the world überhaupt. In conclusion, if the mental life of animals is not a walled garden of raw sensation, it is not such as to present the world as we experience it either. Instead, according to Kant, animals experience a world that presents itself strictly in relation to the animal’s needs and interests, and is unified only according to coarse cognitive principles such as those of continuity and cohesion. Recognizing Kant’s sensitivity to the subtle differences between degrees of cognitive contact with the world allows us a better appreciation of two of Kant’s most central insights: on the one hand, that our most basic cognitive contact with the world is via sensory 57 I reject the need to construe rational capacities as literally ‘transforming’ the sensibility of rational subjects. For discussion of such views see McDowell (1996, ch.  6); Boyle (2016); Conant (2016); Land (2018). I argue against such ‘transformative’ readings of rationality as applicable to Kant in McLear (n.d.). 58 See McLear (2015). Often such readings posit two extreme opposites, e.g. ‘raw’ sensation or impression vs. full-blown perceptual judgment (e.g. George  1981; Beck  1978; Pereboom  1988; Ginsborg 2006). In contrast, if what I’ve argued above is correct, then a large spectrum of positions is available between such opposites. 59 See Gomes (2014: 14) for the expression of a similar worry.

Animals and Objectivity

65

experience; on the other hand, that there is something importantly special about the capacity for rational thought, which can affect, and even make possible, certain kinds of experiences of the world. While remaining true to these insights in our interpretation of Kant’s thought may be difficult, it need not and should not come at the cost of denying either the complexity or the sophistication of the animal mind. The correct interpretation of Kant is one that allows that animals occupy a genuine, if distinct, rung of the cognitive ladder connecting mind and world.60

60 The ideas on which this chapter is based were first mooted at the Witwatersrand ‘Kant and Animals’ conference. Thanks to all who participated in that. Special thanks also to Lucy Allais, John Callanan, Sacha Golob, Colin Marshall, Tobias Rosefeldt, Karl Schafer, Andrew Stephenson, Clinton Tolley, and audiences at Humboldt Universität and Harvard for helpful comments and discussion.

3

What Do Animals See? Intentionality, Objects, and Kantian Nonconceptualism Sacha Golob

3.1 Three Questions about the Status of Animals within Kant’s Theoretical Philosophy If we are to understand Kant’s theory of experience, in the broadest sense of that term, we need to understand how he thinks about nonrational animals (henceforth ‘animals’). In particular, we need to understand how he sees the differences between animals’ engagement with the world and that of rational agents, such as humans. In this paper, I attempt to contribute to that goal by addressing three related questions: as Kant sees it, can animals intuit spatiotemporal particulars, can animals perceive objects, and can animals have intentional states? I argue, ultimately, that the answers support what has become known as the nonconceptualist reading of Kant.1 Let me begin by explaining why the case of animals is significant for understanding Kant’s theoretical philosophy. There are three reasons. First, the vast majority of work on Kant’s theory of mind and on the transcendental arguments tied to it focuses exclusively on humans—for obvious reasons, given the priorities of the Critique of Pure Reason. But in testing and refining such analyses, animals provide a vital philosophical control case. On the one hand, Kant is explicit that there are certain basic similarities between us and animals: [A]nimals also act in accordance with representations [Vorstellungen] (and are not, as Descartes would have it, machines), and in spite of their specific difference, they are still of the same genus as human beings (as living beings). (CPJ 5:464)

Elsewhere, he states that ‘animals are acquainted with objects’ [kennen auch Gegenstände], and can represent ‘something in comparison with other things [sich etwas in der Vergleichung mit anderen Dingen vorstellen]’ (JL 9:64–5): given 1 The seminal contemporary pieces are Allais (2009) and Hanna (2005).

Sacha Golob, What Do Animals See? Intentionality, Objects, and Kantian Nonconceptualism In: Kant and Animals. Edited by: John J. Callanan and Lucy Allais, Oxford University Press (2020). © Sacha Golob. DOI: 10.1093/oso/9780198859918.003.0004

Intentionality, Objects, and Kantian Nonconceptualism

67

its source, one should be careful in placing too much weight on this remark, but, as we will see, it chimes with passages from elsewhere (FS 2:59; C 11:310–11]). On the other hand, however, Kant clearly believes that there are fundamental differences: for example, animals lack the ‘I think’, and by extension the concepts for which it is a vehicle (Anth. 7:127; A341/B399). Given this combination of views, how should we think about animal experience? If animals lack understanding, in what sense can they have ‘representations’ or ‘be acquainted’ with objects? What might the answers tell us about the links between the Aesthetic and the Analytic, or about Kant’s connections to contemporary representationalism or nonconceptualism? The question of animals thus provides a distinctive angle of approach on core Kantian topics such as the relationship between understanding and sensibility. Second, getting clear on the status of animals is necessary if we are to make sense of many passages that are otherwise simply opaque.2 Some of these obviously deal directly with animals. Take the remark just cited from the Critique of the Power of Judgment: in what sense precisely do animals ‘act in accordance with representations’? But other such texts concern broader issues. As is often noted, for example, Kant appears to align synthesis directly with the understanding: indeed, B130 states bluntly that ‘all combination is an action of the understanding’. If this is taken at face value, the only scope for unconceptualized intuitions would be that allowed by Tolley, namely in those intuitions which neither depend on nor involve any synthesis.3 Yet Kant also grants animals associative powers. As he puts it, ‘if I consider myself as an animal’, representations: [C]ould still carry on their play in an orderly fashion, as connected according to empirical laws of association.4

Perhaps the ‘combination’ of B130 is something more sophisticated than mere association. But then there is no inference from the fact that ‘combination’ is the work of the understanding to Tolley’s conclusion that unconceptualized intuitions must not involve ‘any synthesis at all’.5 In short, to fully understand synthesis in the human case, we need to get clear on its associative, animal counterpart. Third, understanding Kant’s position on animals is a vital part of locating him within the history of philosophy. There are thinkers, such as Hume, who stress explanatory continuity when analysing prima facie similar instances of human and animal behaviour: the Treatise proposes this as a ‘touchstone’ by which one 2 Another such set of passages are the pre-Critical remarks on inner sense (for example, ML1 28:276). McLear provides an extremely helpful discussion of these texts which I will therefore not address here: I agree that the root of the problem is the pre-Critical failure to distinguish inner sense from apperception McLear (2011:9). 3 Tolley (2013: 121–2). 4 C 11:52. 5 Tolley (2013: 122—original emphasis).

68

Sacha Golob

‘may try every system’.6 Clearly, we need to know where Kant stands on this Humean principle. But there are also thinkers who explicitly reject an appeal to the same explanatory apparatus even when animal behaviour closely mimics its human counterpart. Heidegger is, at least in some of his texts, a good example of this; here he is responding rhetorically to Hume’s line of thought:7 But a skilful monkey or dog can also open a door to come in and out? Certainly. The question is whether what it does when it touches and pushes something is to touch a handle, whether what it does is something like opening a door. We talk as if the dog does the same as us; but . . . there is not the slightest criterion to say that it comports itself towards the entity.8

What ‘comportment’ is doesn’t matter here; what I want to highlight is the methodological stance of the passage, the assumption that there is an explanatory ‘abyss’ [Abgrund] between the human and animal cases.9 Where should we locate Kant along this continuum that runs between Hume and Heidegger? I have argued for the systematic importance of Kant’s views on animals; this is not simply a niche area of his thought. Over the last decade, many of the questions highlighted have been treated within the debate over Kantian nonconceptualism. I think that framing is sensible, and I will use it to approach the issues here. ‘Nonconceptualism’ means different things across the various literatures. Within a Kantian context, it refers to a view about the relationship between understanding and sensibility. Specifically, nonconceptualism is the thesis that a subject may possess empirical intuitions of spatiotemporal particulars, even if that subject entirely lacks conceptual capacities and indeed any intellect, as Kant understands that faculty.10 It is clear that the nonconceptualist must further hold that such subjects are capable of perceiving at least some spatiotemporal relations: otherwise every spatiotemporal particular would be perceived in isolation and unrelated to any other; a view of dubious intelligibility and one which clashes with Kant’s emphasis on intuitive relations (A22/B37). I will define ‘conceptualism’ simply as the denial of nonconceptualism. I can now frame the first of three questions central to interpreting Kant on animals:

6 Hume ([1738] 1978:1.3.16.3). Locke is also an important figure here: for an overview of some of the issues, see Jolley (2015: Ch. 3). 7 I say ‘some of his texts’ to avoid the debate surrounding notions such as ‘weltarm’. 8 Heidegger ([1928] 2001: 192). 9 Heidegger ([1949] 1976: 326). 10 This definition follows that used in Allais (2009: 384) and subsequently in the later literature (for example, Gomes (2014: 4–5)). This reference to ‘any intellect’ is intended to explicitly exclude accounts such as Longuenesse’s in which a significant role is played by some pre-conceptual form of the understanding: the nonconceptualist claim concerns subjects who lack not only conceptual abilities but also transcendental apperception (see, for example, Longuenesse (1998: 223)). I am grateful to Colin McLear for highlighting this issue.

Intentionality, Objects, and Kantian Nonconceptualism

69

Intuition: As Kant understands them, do animals possess empirical intuitions of spatiotemporal particulars and at least primitive spatiotemporal relations among them?11 The qualifier ‘as Kant understands them’ implies a combination of exegetical and philosophical considerations: we want to attribute to him a view that is both textually sustainable and intellectually attractive. Intuition is simply the basic nonconceptualist thesis applied to animals; they are, after all, the obvious candidates for the intuiting but nonconceptual subjects posited by the nonconceptualist. The truth of Intuition would thus suffice to validate nonconceptualism. Of course, other issues in the area would remain open—for example, whether adult humans might ever have unconceptualized intuitions—but, given the current context, I am going to focus directly on the animal case.12 The same dispute can also be presented in terms of perception: nonconceptualists hold that ‘the application of concepts is not necessary for our being perceptually presented with outer particulars’ (Allais  2009: 384), while conceptualists contend that at least some concepts ‘have an indispensable role’ in even ‘the mere perceptual presentation of particulars’ (Griffith 2010: 199; similarly, Falkenstein 2006: 141). There are, however, complications in Kant’s use of the terms perceptio, Wahrnehmung, and Perception: while standard contemporary usage employs ‘perception’ to mark intentionality in contrast with mere sensation (for example, Burge (2010a: 7)), Kant often uses these terms to mark conscious states, including sensation, in contrast to those states ‘of which we are not conscious’ (Anth. 7:135; A320/B376; A225/B271). I shall therefore mainly frame matters in terms of intuition, but I will also speak of ‘perception’ understood in the standard modern way, particularly when engaging with contemporary philosophy of mind. When we reflect on animal behaviour, however, it can be hard to see how the conceptualism debate can get off the ground. It is a well-evidenced thesis of empirical science and everyday experience that such organisms adjust their behaviour in line with changing spatial relations: as the mouse moves, the cat adjusts its leap. It is hard to see how animals could survive if they were unable to track, in at least a primitive sense, the spatiotemporal location of objects in relation to their current position: those which bury food require an ability to relocate sites, while grazers need to estimate the distance to the watching predators. There is much fine-grained, species-specific work to be done in explaining how this happens: for example, via use of landmarks, different mapping functions, olfactory clues etc.13 However, translating the evidence to a Kantian framework, it may seem obvious that such animals must have an ability to perceive spatiotemporal 11 I follow Allais in borrowing ‘particulars’ from Strawson as a broader alternative to something like ‘material object’: ‘material objects, people and their shadows are all particulars’ Strawson (1959: 15). 12 I discuss the status of unconceptualized intuitions in humans in Golob (2016b). 13 For a recent survey of the empirical literature, see Dolins and Mitchell (2010).

70

Sacha Golob

particulars and their basic relations (how far away the lion is). As Burge notes, discussing parallel trends in contemporary philosophy of mind, the conceptualist view might seem simply ‘empirically refuted’.14 How should the conceptualist respond to this? One move would simply be to dig one’s heels in exegetically—perhaps Kant just did hold a false or outdated view. Yet we should surely try to do better—especially since so much of the relevant evidence comes from simple observation, rather than any technical achievements of post-Kantian science. Looking at the literature, one finds two more sophisticated paths for the conceptualist to take. One is to say that what is really at stake is intentionality. The exact nature of intentionality will be discussed in detail below, but we can think of it initially in terms of ‘aboutness’: intentional states are ‘about’ or ‘stand for’ something beyond themselves, just as the word ‘Paris’ is not simply a collection of marks or sounds but refers to some entity, a city. Ginsborg, a leading conceptualist, introduces the dispute like this: The debate, as Allais helpfully puts it, is about the possibility of intentional content without concepts.15

Likewise, Hanna defines the argument as one about ‘intentional states’.16 So we have a second question: to keep matters simple, I focus on the visual case, and leave aside smell or sound. Intentionality: As Kant understands them, are the visual experiences of animals intentional states? I use ‘visual experiences’ here broadly and non-technically; it refers to those experiences, whatever they may be, which animals have when light arrives at the eye, assuming their physiology is functioning normally. The other option is to say that what is really at stake is object perception. So, for example, Gomes: The traditional conceptualist interpretation holds that the application of concepts is necessary for the perceptual presentation of empirical objects in intuition. In contrast, the non-conceptualist interpretation of Allais and Hanna holds that intuitions can present us with empirical objects without any application of concepts.17

We can thus frame a third question: Objective: As Kant understands them, are the visual experiences of animals experiences of, or about, objects?

14 Burge (2010a: 23). 16 Hanna (2011: 324).

15 Ginsborg (2008: 68). 17 Gomes (2014: 2).

Intentionality, Objects, and Kantian Nonconceptualism

71

While I have separated them for analytic purposes, Intentionality and Objective are closely linked. This is because one standard way to characterize intentionality is precisely in terms of its object-directedness. Thus Ginsborg glosses the question of nonconceptual intentionality as equivalent to the question of: [W]hether we can have nonconceptual representations which are objectdirected, or which represent objects to us.18

Indeed, Kant himself uses ‘object’ terminology precisely to delimit the difference between mere sensations and intentional states: Now one can to be sure call everything, and every representation, insofar as we are conscious of it, an object [Object]. But it is a question for deeper enquiry what the word ‘object’ ought to signify with respect to appearances when these are viewed not in so far as they are (as representations) objects [Objecte], but only insofar as they stand for an object [Object]. (A189–90/B234–5)

The conceptualist contention would then be that animal experience is to be understood along purely sensory lines: such sensations merely ‘refer to the subject as a modification of its state’ (A320/B376), as opposed to being ‘about’ or ‘intending’ some further thing, in the way in which ‘Paris’ refers beyond itself to that very city. We now have three questions with respect to animal experience; we also have a sharper basis on which to formulate the debate between conceptualism and nonconceptualism. But one can see that there is still a great deal left unclear. First, the key terms, for example, ‘object’, carry multiple non-equivalent meanings within Kant’s work. I completely agree with Longuenesse that the Gegenstand/Objekt distinction is no guide here; Kant simply does not employ it uniformly enough, and I will not track it in what follows.19 But one can equally see the point by considering a passage such as B160, where Kant discusses ‘space, represented as object (as we in fact require it in geometry)’. What is at stake here is a complex abstractive capacity undoubtedly beyond animals and significantly beyond what is in question in the nonconceptualism debate: a being might prima facie have ‘object-directed’ states with respect to material things around it, and lack the ability to reflect on space itself. More generally, there are passages that identify category use as a necessary condition on ‘objects of experience’ (for example, A93/B125). But the relevant notion of objectivity is again unclear: the nonconceptualist can simply argue that ‘objects’ here designates some sophisticated cognitive achievement, outrunning the perception of spatiotemporal

18 Ginsborg (2008: 68).

19 Longuenesse (1998: 70n17).

72

Sacha Golob

particulars.20 Crucially, this allows the nonconceptualist to return a positive answer to Intuition: the fact that animals are unable to represent certain advanced forms of objectivity is perfectly compatible with their being able to intuit empirical particulars and simple relations among them. Such a move finds support in passages such as the following, which disambiguates ‘object talk’ in a way that fits well with nonconceptualism: To make a concept, by means of an intuition, into a cognition of an object, is indeed the work of judgment; but the reference of an intuition to an object in general [die Beziehung der Anschauung auf ein Object überhaupt] is not. (Briefwechsel, 11:310–311).

The suggestion is that, while cognition of objects requires concepts, the capacity for objective reference, and thus presumably intentionality, does not. Second, looking now more broadly, the terms used in our three questions are as contentious as any in philosophy; they do not provide a neutral ground on which to stand. Given the prima facie difference between relational and representational theories of perception, it would be surprising if the choice between them did not affect how we answer Intentionality. Similarly, what counts as ‘experiencing objects’ will vary radically depending on one’s other commitments. Recall Frege’s famous complaint: I must also protest against the generality of Kant’s dictum: without sensibility no object would be given to us. Nought and one are objects which cannot be given to us in sensation.21

Third, the logical relations between the various questions are open to contention. For example, there is the familiar debate over whether a state must be intuitive for it to be objective and intentional (consider A286/B342 or B146). But one might also doubt other inferences across the three terms. Strawson at one point defines ‘objective experience’ as including ‘judgments about what is the case irrespective of the actual occurrence of particular subjective experiences of them’.22 One might think that first personal pain reports have intentional content, for example due to their possible truth or falsity, and even that the state’s qualia supervenes on such content, without thinking of them as objective in this sense. I can now spell out the structure of the article. We have three questions regarding animal experience in play:

20 Allais (2011b: 41).

21 Frege (1884: §89).

22 Strawson (1966: 24).

Intentionality, Objects, and Kantian Nonconceptualism

73

Intuition: As Kant understands them, do animals possess empirical intuitions of spatiotemporal particulars and at least primitive spatiotemporal relations among them? Intentionality: As Kant understands them, are the visual experiences of animals intentional states? Objective: As Kant understands them, are the visual experiences of animals experiences of, or about, objects? One tactic would be to proceed directly, focusing on Intuition. As I see it, however, the main reason the debate has been so inconclusive is the huge variance in how different commentators understand that claim. As noted, some cash it in terms of objects, others intentionality; those terms are themselves in turn deeply ambiguous, thus introducing another layer of confusion. So, my proposal is to approach Intuition via Intentionality and Objective. Specifically, I want to clarify how the last two theses bear on the first one. In §2 I clear the way to address Objective by identifying and setting aside various senses of objecthood, which, while central to Kant’s work, do not speak to the issues at hand; they refer to highly sophisticated senses of objectivity that no one would attribute to animals. In §3, I turn to Intentionality, and discuss the implications of relational and representational views: I argue that framing the question in terms of Intentionality will generally support a positive answer to Intuition. Both §2 and §3 will, of course, raise further questions in the philosophy of mind that I cannot adequately address here—for example, which of the various theories of perception is most attractive. My aim is not to answer those, but rather to map how those debates relate to Intuition and thus to clear away some of the confusions surrounding it. This will allow me in §4 to bring together Objective, Intentionality, and Intuition: I suggest that it is nonconceptualism that offers the best understanding of Kant on animals. Given the importance of that issue, as sketched above, I take this to be a significant point in nonconceptualism’s favour. As a limitation on scope, there are other factors that would need to be discussed to have a full picture of the conceptualism/nonconceptualism issue. One is the assumption that the Transcendental Deduction requires conceptualism if it is to be effective against the sceptic: as Ginsborg and Bowman stress, this is central to their endorsement of conceptualism.23 I have argued elsewhere that this assumption is mistaken, and I will not address that debate here.24 Instead, my goal is more restricted: I will claim that neither objects nor intentionality nor the intuition of particulars poses any problem for the nonconceptualist. On the contrary, insofar as the debate is framed in those terms, it is nonconceptualism which is most attractive. 23 Ginsborg (2008: 70), Bowman (2011: 421).

24 Golob (2016a; 2016b).

74

Sacha Golob

3.2 Objective: Two Initial Models of Objectivity The aim of this section is to start to address Objective. I distinguish two senses in which experience might be an experience of, or about, objects; as above, I concentrate on visual awareness. I argue that both senses are easily accommodated by the standard nonconceptualist tactic of conceding that such ‘objective’ experience outstrips the resources of animals while denying that it is necessary for the perception of spatiotemporal particulars. As a result, the fact that animals lack ‘objective’ experience in this sense poses no threat to Intuition. Ginsborg has suggested that this tactic risks trivializing Kant’s arguments by rendering the transcendental conditions he identified necessary only for certain high level activities; I explain briefly why this worry is misplaced.25 The first notion of objecthood is best approached via one of Kant’s own discussions of animal perception. He begins by confronting an argument of Meier’s in favour of animals being ascribed concepts: An ox’s representation of its stall includes the clear representation of its characteristic mark of having a door; therefore, the ox has a distinct concept of its stall. It is easy to prevent the confusion here. The distinctness of a concept does not consist in the fact that that which is a characteristic mark of the thing is clearly represented, but rather in the fact that it is recognized [erkannt] as a characteristic mark of the thing. (FS 2:59)

I suggest something like the following story about Kant’s position here. The ox has a clear—where that term is understood phenomenologically—visual awareness of some property or ‘mark’ of the stall, namely having a door. This clear representation is the basis for both differential reaction (the ox would behave differently in a stall with no door), and for association (the ox becomes anxious or excited depending on past experiences with doors). The rational agent, however, is distinguished by the ability to recognize this mark, something that can be shared by many stalls and by many non-stalls, as a generic property. One way to express this is to say that we, unlike the ox, see the door ‘as’ a door. This ability to recognize generic properties or marks is, of course, simply the ability to employ concepts: ‘[a]ll our concepts are marks and all thought representation through them’ (R 16:300). Following Kant, we can further analyse concepts in terms of rules, that is, patterns of inference that order and connect our representations (A126; A106). Specifically, to recognize a mark is to recognize a set of inferences as grounded in it; so, to recognize something as exhibiting the mark body is to recognize both a fact about the entity involved and certain implications for how we 25 Ginsborg (2006: 62).

Intentionality, Objects, and Kantian Nonconceptualism

75

must think of it—for example, any body ‘necessitates the representation of extension’ (A106). It is in this sense that the Logic treats marks as both ‘in the thing [Ding]’ and as a ‘partial representation . . . considered as the ground of cognition’ (JL 9:58). Kant’s use of ‘thing’ here is helpful since it avoids a confusing over-repetition of ‘object’, and I follow him in it. To recognize a thing as exhibiting certain marks is thus: (i) To require myself either to attribute further properties to the thing in line with the relevant inferential rules, or to revisit the initial attribution. Mark recognition thus imposes a normative order on experience, preventing it from being ‘haphazard’ (A104). (ii) To possess, if only tacitly, an awareness of some inferences as being putatively grounded in the properties of the ‘thing’, in this case the stall. By extension, it is to possess, again if only tacitly, an awareness of the distinction between such inferences and other ways of combining representations that are not so grounded. In the Transcendental Deduction, Kant expresses these points by contrasting the relations posited in judgment with those posited by associative or ‘reproductive’ imagination. It is in this sense that ‘judgment is nothing other than the way to bring given cognitions to the objective unity of apperception’: it allows me ‘to say that the two representations are combined in the object’ (B141–2).26 (iii) To possess, if only tacitly, an awareness of the fact that insofar as an inference is putatively grounded in properties of the ‘thing’, as opposed to being merely an artefact of my own psychological history, the posited connection should presumptively hold for any other observer, ‘regardless of any difference in the condition of the subject’ (B141–2). Thus: ‘the representation of the manner in which various concepts (as such) belong to a consciousness (in general, not only my own), is judgment’ (R16:633). (iv) To possess, if only tacitly, an awareness of the ‘thing’ as potentially having other generic properties, and an awareness of the mark as a generic property that may potentially be instantiated by other things: as Kant

26 There is a general question as to how one should understand notions like ‘tacit recognition’ in Kant. I take some reliance on them to be near omnipresent: for example, transcendental apperception is standardly taken to imply a self-awareness and self-ascription, which nevertheless falls short of the explicit, thematic judgment that a given piece of content is mine (something that only happens very occasionally). On the Kantian picture, such tacit recognition has systematic consequences (for example, I recognize an obligation to try to maintain consistency among all the representations which are ‘mine’) and underpins its explicit counterpart. I cannot address how exactly this should be spelt out here, but my account can simply rely on whatever is the reader’s preferred model for this general Kantian device. My thanks to Colin McLear for discussion here.

76

Sacha Golob puts it, ‘concepts, as predicates of possible judgments, are related to some representation of a still undetermined object’ (A69/B94). In Evansian terms, an experience characterized by the recognition of marks meets the ‘generality constraint’.

A few comments before proceeding. First, unlike the body/extension example, most of the inferences involved will be synthetic and indeed a posteriori and so contingent (B142). The point of (ii) is that, insofar as one recognizes marks, one is able to represent the fact that such connections, even when contingent, hold in virtue of the thing before you, and not simply because you happen to associate one property with another. Second, while my approach does not require any particular reading of the Prolegomena’s discussion of judgments of perception and experience, it is worth briefly commenting on that, since it is relevant to the questions of accuracy that come up when discussing Intentionality. As I see it, the Prolegeomena treats two issues. One concerns cases that exhibit the syntactic form of judgments and yet where their particular semantics renders the distinctions discussed undrawable. I have in mind here the ‘sugar is sweet’ case: given the assumption that sensations merely ‘refer to the subject as a modification of its state’ (A320/B376), ‘sweetness’, despite compounding with the copula, cannot be taken to attribute a property to the thing. Judgments involving such ‘pseudo-predicates’ are therefore merely ‘logical connections of perceptions’ because their meaning necessarily concerns only ‘myself and that only in my present state of perception; consequently, they are not intended to be valid of the object’ (Prol. 4:298–9) The other issue concerns the transition from judgments that are presumptively objective in the sense defined by (i)–(iv) to judgments that have been found genuinely to have identified such a connection; to reach that level, it must be shown that ‘I and everyone else should always necessarily connect the same perceptions under the same circumstances’ (Prol. 4:299–300). The best illustration of this transitional process, through which a ‘judgment of perception can become a judgment of experience’ is the sun warming the stone (Prol. 4:301). These points can now be summarized; as above; I focus on visual awareness for simplicity’s sake. Definition of Objective1 A visual experience E is objective1 if and only if E at least tacitly represents a spatiotemporal particular P as possessing certain generic properties, represents those properties as standing in inferential relations, represents such inferences as presumptively grounded in facts about P (as opposed, for example, to being merely associative), and thus represents them as presumptively holding for other rational agents encountering P.

Intentionality, Objects, and Kantian Nonconceptualism

77

Definition of Object1 A visual experience E is of an object1 if and only if E is objective1.

It is this notion of objectivity, and a correspondingly defined notion of an object, which Kant has in mind here: If we investigate what new characteristic is given to our representations by the relation to an object, and what is the dignity that they thereby receive, we find that it does nothing beyond making the combination of representations necessary in a certain way, and subjecting them to a rule. (A197/B242)

If we now return to Objectivity, we have an initial disambiguation of it: Objectivity1: As Kant understands them, are the visual experiences of animals experiences of, or about, objects1? The answer is surely not: Kant’s claim is precisely that such objectivity is a function of judgment and conceptualisation, neither of which any commentator thinks animals possess. This is agreed by both conceptualists and nonconceptualists alike. So we can simply set objective1 aside. Here is another way to put the point: the natural nonconceptualist reading of Kant’s ox example is one on which the ox’s perception of the stall is an intuition of an empirical particular, thus validating Intuition. The fact that the ox cannot further represent certain complex connections between the stall’s properties is irrelevant. Of course, the conceptualist might insist that objective1 just is what he or she means by ‘intuition’ or ‘particulars’. But, on those definitions, even Allais would be a conceptualist. So objective1 should be set aside; it does not help in assessing, for better or worse, the nonconceptualist commitment to Intuition. The second sense of objectivity I want to address is linked to the categories. There is, as noted in §1, a widespread belief that the Deduction, as an antiHumean argument, requires that categorical synthesis be a necessary condition on the representation of spatio-temporal particulars. I have argued in detail that this is a mistake.27 I will not, however, treat the Deduction here. Instead, I argue for a conditional claim: if the issue of the Deduction is resolved in a manner compatible with nonconceptualism, then the notion of objectivity associated with the categories can be treated in line with the same nonconceptualist strategy just employed, namely accepting that animals’ representations lack such objectivity but denying that perception of spatiotemporal particulars requires it. The point is best introduced using the example of the Second Analogy. There Kant asks us to consider how, given the necessarily successive nature of

27 Golob (2016a) and Golob (2016b).

78

Sacha Golob

apprehension, we can represent the distinction between successive perceptions and a perception of succession; he claims that this requires us to assume some form of causal order among the events in question (A189/B234; A194/B239). In making this point, he introduces a particular notion of objectivity: If one were to suppose that nothing preceded an occurrence that it must follow in accordance with a rule, then all sequence of perception would be determined solely in apprehension, i.e. merely subjectively, but it would not thereby be objectively determined which of the perceptions must really be the preceding one and which the succeeding one. In this way we would have only a play of representations that would not be related to any object at all. (A194/B239)

Restricting ourselves to this example, we can formulate the preliminary claim: Restricted Definition of Objectivity2 A successively apprehended visual experience E is objective2 if and only if E represents the distinction between successive perception and the perception of succession with respect to a spatiotemporal particular P.

As Kant puts it himself: [O]bjective significance is conferred on our representations only insofar as a certain order in their temporal relation is necessary. (A197/B243)

If we lift the restriction and include cases such as the Axioms where the relevant abilities, while again threatened by the successive nature of apprehension (A162–3/B203–4), are themselves spatial and compositional rather than temporal we get: Definition of Objectivity2 A successively apprehended visual experience E is objective2 if and only if E represents a privileged class of spatiotemporal relations with respect to a spatiotemporal particular P (for example, objective succession and mereological composition). Definition of Object2 A visual experience E is of an object2 if and only if E is objective2.

How should the nonconceptualist think about this second notion of objectivity? Well, given that the Deduction has been set aside, the answer is surely simple: he or she can just grant that animals lack such abilities. The absence of objectivity2

Intentionality, Objects, and Kantian Nonconceptualism

79

implies only that there are some comparatively sophisticated spatiotemporal relations that animals cannot represent. But that is perfectly compatible with the claim that they perceive particulars and primitive spatiotemporal relations, such as distance, between them. To adapt Kant’s ship example, to see a salmon ‘driven downstream’ is, minimally, to successively apprehend a particular, the salmon, in relation to various other particulars: the rocks, the banks, the bushes, etc.: this is what must be in place for the problem that object2 solves to even arise in the first place. Of course, the animal will lack any sophisticated representation of this salmon as a single enduring object, but, as Allais notes, it can represent its identity in a primitive fashion by tracking its path and by responding differentially to it: for example, reacting to the salmon’s movements as it wriggles left and right.28 In other words, Kant’s own example suggests that the absence of objectivity2 is entirely compatible with the ability to perceive particulars and relations such as spatial juxtaposition between them.29 Objectivity2 can thus be set aside: like, objectivity1, it is logically independent of Intuition.30 Of course, we need to know much more about what the nonconceptual perception of the salmon amounts to and why exactly it deserves to be called an ‘intuition of a particular’. But objectivity2 is not going to help address those questions. We can now return to Ginsborg’s worry about trivialisation. There are two fears one might have. On the one hand, nonconceptualism might trivialize the Deduction by making the categories a necessary condition only on something too sophisticated, something which the sceptic would also reject. This worry is misplaced because the categories make possible precisely the abilities that someone like Hume takes for granted, abilities such as event perception.31 On the other hand, nonconceptualism might trivialize the transcendental claims made about the categories themselves. But this is surely not the case; the idea that we need the concept of causality if we are to represent objective succession is a deeply contentious one and remains so independent of whatever one says about animals.32

28 Allais (2009: 405–6). 29 One way to resist this would be to atomize the individual apprehensions to the point where what is perceived at T1 is not ‘salmon in front of rocks’, but simply ‘salmon’. But there would then be no reason to locate the various images in any spatial relation rather than any other: if all I see is salmon then rocks, why assume that the former is in front of the later, not beside it to the left or right? This would apply to the human case too: whatever contribution understanding makes, it does not explain why we perceive something to the left rather than the right. 30 In line with the discussion of objectivity1 the animal will also be unable to see the salmon ‘as’ a salmon, where this means something like ‘recognize the mark salmon in the particular’. 31 For further discussion, see Golob (2016b). 32 Allais makes the same point with respect to her model on which the categories are necessary conditions on empirical concept use Allais (2011b: 47–8). See Golob (2016a) for where I disagree with Allais on categorical necessity.

80

Sacha Golob

3.3 Intentionality: Relationalism, Representationalism, and Animal Experience I have so far identified two notions of objectivity on which the answer to Objective is straightforwardly ‘no’: animals cannot perceive objects in those senses. This is, however, entirely compatible with their perceiving spatiotemporal particulars and relations in some weaker sense: for example, seeing the salmon against various backdrops (we’ll return to what exactly this would amount to in §4). I now want to turn to Intentionality; again, I’ll use a visual case. Intentionality: As Kant understands them, are the visual experiences of animals intentional states? To answer this, I need to say a little about the two approaches that dominate the debate on perceptual intentionality: relationalism and representationalism.33 We can begin with the following rough characterisation: Representationalism The explanatorily fundamental characterisation of perceptual experience is given in terms of representational contents that determine accuracy conditions for that experience. Relationalism The explanatorily fundamental characterisation of perceptual experience is given in terms of a non-representational relation between the subject and the perceived objects.

In a full discussion, one would need to treat positions that use elements of both: McDowell’s or Schellenberg’s for example.34 But my focus here is on the links between the larger debate and Kant. Allais, in defending nonconceptualism, has argued that Kant’s own sympathies lay with relationalism.35 In response, Gomes suggests that there need be no tension between conceptualism and at least moderate versions of relationalism. I remain neutral on both those points. My claim instead will be that, whichever of relationalism or representationalism one favours, the answer to Intentionality is likely to be either straightforwardly positive or at least ‘non-prejudicial’. I introduce the notion of a ‘non-prejudicial’ answer because many relationalists are reluctant to talk in terms of ‘intentionality’ themselves: this means they cannot give a positive or negative answer to 33 One could equally make these points using alternative taxonomies—for example, ‘Fregean or Russellian’. I have gone for the option above in order to provide broader coverage: many ‘Russellian’ views are really representationalist positions with object-dependent senses. 34 McDowell (2013); Schellenberg (2015). 35 Allais (2011a: 380).

Intentionality, Objects, and Kantian Nonconceptualism

81

Intentionality. However, a relationalism on which the explanatorily fundamental characterisation of animals’ perception is the same as that of humans will be said to be ‘non-prejudicial’ to nonconceptualism. This is because, while it does not return a direct answer to Intentionality, it supports the broader nonconceptualist case by aligning humans and animals: insofar as the former have empirical intuitions of spatiotemporal particulars, so should the latter. Suppose one endorses representationalism. What distinguishes, say, sensations from intentional or object-directed experiences is then the fact that the latter represent the world; as Kant puts it, such states ‘stand for an object’, they point to something beyond themselves (A189–90/B234–5). The representationalist cashes this in terms of contents with accuracy conditions: the content of the relevant experiences represents some state of affairs and is said to be accurate or inaccurate depending on whether that state of affairs obtains.36 Within this framework, I want to make two points regarding animals. First, it is standardly assumed that one of the chief advantages of representationalism is that it allows easy treatment of hallucinations and illusions.37 This is because the representationalist can simply treat these as misrepresentations; the relationalist has a harder time accommodating states where, although I experience X as F, there is no X that is F and thus no obvious candidate for the relata. What I want to stress is that it is hard to conceive of an attractive representationalism that did not emphasize its privileged ability to handle, say, optical illusions in terms of inaccurate contents. There is well-documented empirical evidence that animals too are susceptible to such illusions. The Müller-Lyre, for example, has been shown to affect the grey parrot (Psittacus erithacus); other species, for example, bamboo sharks (Chiloscyllium griseum), are affected by Kanisza squares.38 It is hard to see how a representationalist could maintain that one needs to posit representational content to deal with such cases at the human level, and yet not do the same in the animal case. But if that is true, then the Kantian representationalist must concede, assuming the principle of charity, that animals have intentional states. While animals as Kant sees them certainly lack the ability to make judgments, this need not present a problem for the representationalist. The most direct strategy is simply to argue that a state’s being a judgment is sufficient but not necessary for its having accuracy conditions; as Crane has emphasized, for example, a picture might be accurate or inaccurate even while there are good reasons for thinking that the 36 As Tye puts it, ‘any state with accuracy conditions has representational content’ (Tye 2009: 253). 37 I sympathize with Brewer when he describes this as the ‘primary motivation’ for representationalism Brewer (2011: 59). Similarly, Smith divides his The Problem of Perception into two sections entitled simply ‘The Argument from Illusion’ and ‘The Argument from Hallucination’ Smith (2002). 38 Fuss, Bleckmann, and Schluessel (2014); Pepperberg, Vicinay, and Cavanagh (2008).

82

Sacha Golob

way it represents the world is not propositional.39 In a Kantian context, one might, therefore, naturally construe animals as forming three dimensional egocentrically orientated images of the world, images which can then be associated either with each other or with non-intentional contents such as sensations. The images’ full representational structure could be given by appeal to something like Peacocke’s scenario content.40 In short, (i) there are plausible candidates for the contents of Kantian animals’ representational states, and (ii) the distinctive dialectic with respect to illusion and hallucination that is one of the core motivations for representationalism actively requires that the theory be applied to both human and animal cases. Of course, there is a great deal of textual work to be done to cash this: for example, in defending the proposed non-judgmental content bearers given Kant’s claim that ‘error is a burden only to the understanding’ (A293–4/B350, Anth. 7:146). My own preferred candidate would be to link them to the imagination: this is the faculty of intuition precisely when the object does not exist (Anth. 7:153), as is the case in misrepresentation. Imagination’s notoriously ambiguous place within Kant’s architectonic could also explain his apparent confinement of content to the understanding (compare the standard strategies for dealing with the apparent disappearance of the imagination from the B Deduction). This is not the place to undertake that exegetical work, however; what I want to do is rather map the basic dialectical lines available.41 What we have established is a conditional claim: if one were to adopt a representationalist approach, there is a strong philosophical motivation for returning a positive response to Intentionality. Insofar as Intentionality provides a natural way of cashing Intuition, this supports a positive answer to Intuition—and that supports nonconceptualism. Suppose next that one endorses relationalism. The issue of truth value immediately becomes otiose since, as Brewer puts it: The intuitive idea is that, in perceptual experience, a person is simply presented with the actual constituents of the physical world themselves. Error, strictly speaking, given how the world actually is, is never an essential feature of experience itself.42

Illusion and hallucination, meanwhile, become more complex. I agree with Siegel, for example, that negative naïve realist characterisations of hallucination face problems when transferred to the animal case.43 But this is ultimately an artefact of the general difficulty relationalism faces over hallucination, and something 39 Crane (2009). 40 Peacocke (1992a). 41 For highly sophisticated treatments of some of the textual issues in play here, see Stephenson (2016) and McLear (2016). 42 Brewer (2006: 5). 43 Siegel (2008).

Intentionality, Objects, and Kantian Nonconceptualism

83

that any relationalist must come to terms with. When one looks beyond hallucination, relationalism dovetails with Intuition because of the comparatively thin conditions typically imposed on the relation or ‘openness to the world’ that grounds the story. As Smith puts it, commenting on the dominant form of relationalism: Naïve realism draws its strength from the apparent simplicity of perceptual consciousness. You open your eyes and objects are simply present to you visually. The shutters go up, as it were, and the world is simply there.44

If one feels the pull of this rationale, it would surely equally apply to animals. The relationalist story is typically developed by introducing notions like the ‘perspective from which something is seen’ and salient similarities between that entity and other objects, but these notions, usually cashed in causal or evolutionary terms, need present no problems for the animal case.45 In sum, if one endorses relationalism and is prepared to bite the bullet on hallucination generally, the pressure will be towards a parity between the human and animal cases at the explanatorily fundamental level; this is precisely the spirit of the account as captured by Smith. There is one move that would run counter to this dynamic: a form of relationalism on which conceptual capacities are necessary for the relation to be established.46 How exactly this should be dealt with depends in part on the details—in particular whether it is a representationalism with object-dependent contents, or whether it is a genuine relationalism eschewing any accuracy conditions at the perceptual level. This is not the place to assess the philosophical potential for such a theory. Rather, as with Objectivity, my aim is to try to clarify the overall topography of the debate: we can now see that glossing Intuition in terms of Intentionality will support the former, unless one defends a very specific sub-form of relationalism. Relationalism is thus likely to support what I called a ‘nonprejudicial’ verdict on Intentionality, one that supports Intuition and thus, nonconceptualism.

3.4 Intuition: Spatial Awareness and Intuitive Particulars With the preceding material in place, I can now look more clearly at Intuition itself. McLear has suggested that the conceptualist’s best option is to construe animal consciousness as follows:

44 Smith (2002:43). 45 Brewer (2011:118–19). 46 One natural candidate would be McDowell’s recent work (McDowell, 2013).

84

Sacha Golob [B]eings lacking concepts nevertheless possess a form of experiential consciousness. However, this form of consciousness is extremely primitive, lacking any object-directed nature. All such conscious states are thus purely subjective forms of awareness. They cannot be instances of an awareness of physical particulars or their properties . . . on this view, all sensory presentation is limited to the subject’s own states.47

This proposal—a good one—cashes ‘object-directed’ in something like the following terms: Definition of Objectivity3 A visual experience E is objective3 if and only if E represents a distinction between spatiotemporal particulars and the mental states of the subject of that experience. Definition of Object3 A visual experience E is of an object3 if and only if E is objective3.

In the absence of objectivity3, as Husserl observes: [S]ensations mean nothing, they do not count as indications of the properties of objects, their complexity does not point to the objects themselves. They are simply lived through.48

Objective3 is in many respects a more attractive notion than objective1 or objective2 in terms of which to frame the debate. While the latter two are naturally identified with capacities going beyond an ability to intuit particulars, objective3 is a more plausible gloss on the line between sensations and genuine intuitions. Hence the remark from Kant cited above: Now one can to be sure call everything, and every representation, insofar as we are conscious of it, an object [Object]. But it is a question for deeper enquiry what the word ‘object’ ought to signify with respect to appearances when these are viewed not in so far as they are (as representations) objects [Objecte], but only insofar as they stand for an object [Object]. (A189–90/B234–5)

I will now argue that when Intuition is glossed via objectivity3, Kant’s animals do indeed perceive objects. By extension, they do indeed intuit spatiotemporal particulars, insofar as that is glossed in terms of such objectivity.

47 McLear (2011: 3)—McLear himself argues for a nonconceptualist view. 48 Husserl (1984: 80).

Intentionality, Objects, and Kantian Nonconceptualism

85

One familiar Kantian question is whether spatial content is necessary for objectivity: that will depend, for example, on how on one reads Kant’s claim that thought alone would be ‘without any object’ (B146). But what is important here is that it seems very plausible that spatiality is sufficient for objectivity in the sense of objectivity3.49 More specifically, the claim is that a three dimensional egocentrically orientated awareness of space within which something is seen as more or less distant is sufficient to sustain a distinction between spatiotemporal particulars and the subject’s own states, such as sensations. Smith provides a neat formulation of the idea: Perception concerns the ‘external world’. The suggestion is that this is, in essential part, because perceptual experience presents ‘external’ objects as literally external—to our bodies. A bodily sensation such as a headache is experienced as in your head; it is not perceived as an object with your head. When, by contrast, you look at your hand, although the object seen is not spatially separated from you (since it is a part of you), it is, nevertheless, spatially separate from the eye with which (and from where) you see it.50

Kant believes that self-consciousness and consciousness of objects1 or objects2 stand in a biconditional relation. But the present question is whether there might also be a weaker version of this biconditional, applicable to animals and based on objectivity3. On the side of the self, the claim is not that the animals have the representation (Anth. 7:127). Rather, it is that animals experience space in egocentric terms. O’Brien offers a helpful syntactic formulation of the contrast: Egocentric contents are . . . given by monadic notions such as ‘to the right’ and ‘up ahead’ in contrast to first-personal contents that are given by relational notions such ‘to the right of me’, ‘in front of me’.51

Likewise, on the side of the object, the claim is not that animals have concepts such as external world or sensation. Rather, the proposal is that the way in which entities are given to them as spatially arrayed is sufficiently distinct from the way in which sensations are given to them that it constitutes a distinctly perceptual or intuitive, as opposed to sensory, mode of experience, one which is well described as ‘object-directed’. For example, insofar as the animal encounters something as arrayed within such a space, it is given only from a single perspective, a perspective which changes as the object gradually unfolds in line with the animal’s movements and 49 Allais, drawing on Campbell and Smith, makes a similar point Allais (2009: 413). I want to press it further by using some of the resources of phenomenology. 50 Smith (2002: 134). 51 O’Brien (2007: 106).

86

Sacha Golob

motor dispositions. In contrast, sensation is non-perspectival—while the dog experiences the pain in its foot, as opposed to its leg, there is no angle from which it does so. In short, to borrow a formulation from Husserl, there is a distinctive and phenomenologically articulable mode of givenness that allows one to legitimately ascribe the object3 distinction to animals, even though, of course, they cannot articulate it. By extension, animals will intuit particulars, and not mere sensations, insofar as Intuition is glossed not in terms of objecivity1 or objectivity2, but objectivity3. One way to develop this proposal is by comparison with Strawson, who employs several non-equivalent concepts of objectivity. Some have been dealt with above. For example, he identifies objectivity with an ability to recognize a distinct temporal order within which objects, as opposed to our perceptions of them, stand.52 This is not plausibly attributed by animals: there is no temporal parallel to egocentric space that would allow for a corresponding mode of givenness. Yet that is because, as Strawson is well aware, this is simply the Analogies’ version of objectivity2, something that no nonconceptualist would attribute to animals in any case. What is more important is rather a second definition Strawson offers: objective experience is ‘experience of objects that are distinct from the experience of them’.53 It is trivially true—operating throughout as within an empirical realism—that the experience of animals is usually of such objects. But Strawson’s point concerns rather the subjects’ ability to represent that fact, and so to avoid the solipsism where a creature ‘simply has no use for the distinction between himself and what is not himself ’.54 As the conceptualist sees it there are only two options here: either one lacks this distinction, or one has a conceptual awareness of it that one can at least potentially articulate. My appeal to a ‘mode of givenness’ is intended to offer a third alternative, and one which thus secures objectivity3. At this point, the conceptualist will likely protest: ‘All this shows’, he or she might retort, ‘is that if the animals have experience with spatial phenomenology then they experience them as objective3—but it is precisely the antecedent that I deny’. This might be because the conceptualist denies that animals have any phenomenological consciousness or, less severely, because he or she allows them some phenomenological consciousness but denies that it is spatial.55 But what we can now see is the very high price one must pay for this view. First, I have shown that neither objectivity1 nor objectivity2, nor a general representationalism nor most relationalisms, present any problem for the nonconceptualist. This radically reduces the possible independent reasons for denying that 52 Strawson (1966: 98). 53 Strawson (1966: 24). 54 Strawson (1959: 65). 55 I take McLear (2011) to have provided compelling textual arguments against the former view and in what follows I will focus on the latter.

Intentionality, Objects, and Kantian Nonconceptualism

87

animals perceive spatiotemporal particulars and their relations. Of course, as we saw, one might contest the word ‘particulars’—perhaps for some speakers that simply means objects1 or objects2. But to make this move is to concede the key nonconceptualist claim that animals have intentional experiences of parts of the world, given as external to them and as standing in distance and other relations. Whether one wishes to call such parts ‘particulars’ is a purely terminological matter. Second, the conceptualist can offer only an improbably baroque alternative. To see this, consider even the less severe version of the view, on which animals are allowed some form of consciousness but denied a spatial phenomenology. The conceptualist must surely concede that the behaviour of animals does in fact track spatial relations such as depth and distance, and even primitive temporal relations: a dog can be trained to react to two flags only when they are raised simultaneously. If the conceptualist nevertheless wishes to deny that animals experience entities as arrayed in a three dimensional egocentric space, he or she must posit some set of sensations in which things are not given as at a certain distance, and yet which systematically change as distance relations change. In effect, the animal would not directly experience spatiality, only some kind of systematically correlated sensational proxy. How could this be cashed? Suppose a predator sees a fish under the surface of the water to the left and grabs it. On my account, the explanation is simple: a physical object, the fish, is phenomenologically manifest to the predator as lying a certain distance from it. Of course, the predator lacks the ability to articulate this, just as it lacks the concept fish. But it nevertheless intuits that very object and intuits it as external to itself, a certain distance away. The conceptualist cannot grant this, since it implies both objectivity3 and the basic nonconceptualist contention that intuitions are independent of the understanding. So instead she must claim that the predator is aware of some non-spatial, presumably qualitative, sensational correlates. But what could these be? Perhaps when there is something to the left, the animal experiences a sensation of a particular colour? Yet we have good reason to think, from the structure of the eye and empirical testing, that at least some of the predators involved lack colour vision. Perhaps then they experience some kind of light/dark or hot/cold sensations as they get closer to their prey? But that seems too crude: many of the predators can distinguish minute differences in range and angle—are we to believe that they have a similarly finegrained awareness of degrees of brightness or heat, even though they have no evolutionary need for such, and no correspondingly specialized sense organs? We should surely refrain from committing Kant to such an unpromising programme if we can possibly avoid it. What, from a biological perspective, could explain the reliance on such a convoluted and roundabout method—would it not be far simpler to posit that, sharing as they do much of our perceptual apparatus, animals also directly experience the world in spatial, and thus objective3, terms?

right 2020. OUP Oxford. rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law.

88

Sacha Golob

I am not claiming that Kant himself had a fully worked out story as to how we should understand ideas like a ‘mode of givenness’; he has little to say about the role of embodied motor dispositions in encountering objects. I am rather claiming that such a view would chime with his overall project—the Aesthetic elucidates various further conditions on encountering spatial relations—and, I think, be an attractive supplement to it. Bringing this all together, we can now gloss the sense in which animals do indeed intuit spatio-temporal particulars: these are intuitions of particulars, as opposed to mere sensations, because they are objective3. How exactly should we construe the phenomenology? Well, here I think we must be guided by the animal’s behaviour. If, for example, it is capable of differential reactions to minutely different species of fish, we should assume that its intuition includes all the relevant visual details for distinguishing those species. In short, animal visual experience is neither a ‘blooming, buzzing confusion’ nor is it indistinct or crude: it is of intuitive particulars presented at a level of visual detail which often far outstrips our own capacities.56

56 I would like to thank all participants at the Witwatersrand ‘Kant and Animals’ conference where this material was first presented for their extremely helpful comments and suggestions. I am particularly indebted to Colin McLear and to an anonymous referee for their detailed and insightful comments on an earlier draft.

EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 9/14/2020 5:03 PM via COLUMBIA UNIVERSITY - MAIN

4

Kant on Nonhuman Animals and God Ina Goy

4.1 Introduction In this paper, I analyse Kant’s account of the nature of nonhuman and human animals1 in the Critique of the Power of Judgement (hereafter, CPJ), in order to show nonhuman and human animals relate to humans’ belief in God. I will argue that Kant is committed to the following six claims, whereby the first three of these six claims concern the human, the second three the divine point of view. First, from the perspective of the human power of judgment nonhuman and human animals are mechanical objects since they undergo motions and changes that fall under mechanical powers and laws. Second, nonhuman and human animals are organized beings, since their mechanical motions and changes are directed toward the fulfilment of natural purposes, and the manifold mechanical motions and changes that fall under mechanical laws can be subordinated to physical-teleological powers and laws. Third, nonhuman and human animals are not only organized beings with natural purposes and, as such, relative ends, but are, in external relations, means to the fulfilment of humans’ moral purposes, and means to the fulfilment of the highest moral purpose: the highest good as their final end.2 If nonhuman and human animals can be explained by three kinds of powers and laws from the human point of view, how do these mechanical, physicalteleological and moral-teleological powers and laws relate to each other? Kant is committed to the claim, fourth, that the mechanical and physical-teleological 1 When Kant speaks about animals he means—to name examples that Kant himself mentions in the CPJ—birds (CPJ 5:360.11), cattle (CPJ 5:368.2, CPJ 5:378.6, CPJ 5:379.19), sheep (CPJ 5:368.2, CPJ 5:379.19), horses (CPJ 5:368.3, 20; CPJ 5:378.7), oxen, asses, swine (CPJ 5:368.21), dogs (CPJ 5:378.7), beavers (CPJ 5:464.20), camels (CPJ 5:368.4, 5:378.6), wolves, tigers, lions (CPJ 5:368.6), reindeer (CPJ 5:369.5), maggots (CPJ 5:411.34), polyps (CPJ 5:419.3), vermin (CPJ 5:379.22), and mosquitoes (CPJ 5:379.25–6). A first group of what Kant calls ‘animals’ consists in those nonhuman animals that we also identify as animals today. In addition, in line with a long philosophical tradition, Kant characterizes the human being as animal rationale (CPrR 5:61.33, CPrR 5:160.25, CPrR 5:162.17–20). But the human being is an ‘animal endowed with reason’ (MM 6:456.28) and only its natural, nonrational part counts as ‘animality [Thierheit]’ (CPJ 5:430.7, CPJ 5:432.9, CPJ 5:433.27). 2 Among these three kind of powers and laws, in the case of nonhuman animals, interpreters have often discussed the mechanical and physical-teleological powers and laws, and the natural purposes of nonhuman animals, so McLaughlin (1989), Ginsborg (2001 and 2004), Zuckert (2007), and van den Berg (2014), but they have ignored their external relations to the natural and moral purposes of human animals. Ina Goy, Kant on Nonhuman Animals and God In: Kant and Animals. Edited by: John J. Callanan and Lucy Allais, Oxford University Press (2020). © Ina Goy. DOI: 10.1093/oso/9780198859918.003.0005

90

Ina Goy

powers and laws do not conflict since they can be thought of as originally united in the regulative idea of an intelligent and intentional God, whose theoretical consciousness accounts for the unity of the mechanical and physical-teleological powers and laws and whose creation accounts for the unifiability of the mechanical and physical-teleological powers and laws in the human consciousness and the human experience of the world. But how do the physical-teleological and moral-teleological powers and laws relate to each other? Kant is committed to hold, fifth, that these do not conflict either, since they, too, can be thought of as originally united in the regulative idea of a theoretical intelligent and practical wise God, whose theoretical practical consciousness accounts for the unity of the physical and moral-teleological powers and laws, and whose creation accounts for the unifiability of the physical and moral-teleological powers and laws in the human consciousness and in the human experience of the world. Finally, Kant is committed to the claim that the regulative idea of a God who accounts for the unity of the natural powers and laws is an aspect of the regulative idea of the God who accounts for the unity of the natural and moral powers and laws; God, according to Kant’s position, is one. A significant feature of Kant’s account of nonhuman and human animals is, thus, that he tries to develop a consistent account of the mechanical, physicalteleological, and moral-teleological powers and laws that explain all animals, and that he guarantees the coherence and harmony of these powers and laws with the regulative idea of God and God’s creation. Kant’s main idea is that the three kinds of natural and moral powers and laws are inconsistent themselves, and can be thought of in harmony only if God creates the human consciousness and the world as it appears to it (CPJ 5:175.37–176.15, CPJ 5:450.17–30).3 In light of these claims one could say that in his account of nonhuman and human animals Kant defends a certain form of teleological holism, which raises a number of difficulties. A first difficulty is that Kant organizes nonhuman and human animals into hierarchies, in which human animals are superior to nonhuman animals due to their faculty of reason and the related capacity to set and select rational (natural and moral) ends. I will argue that Kant’s account implies an unhealthy anthropocentrism and an Enlightenment prejudice in the form of the overestimation of reason as a faculty belonging exclusively to human consciousness. A second difficulty, I claim, is that the Kantian model of God also lacks, for instance, one of the main characteristics of the Christian conception of God: the universal divine love, a power that unifies and embraces all beings, including nonhuman and human animals and their orders. I will now turn to each of the six claims separately. 3 Most of the traditional interpreters have, I claim, either ignored this regulative theological framework of Kant’s account of nonhuman and human animals, such as McLaughlin (1989), Quarfood (2004), van den Berg (2014), and Ginsborg (2014), or have outlined only the divine and its function in Kant’s account, but have not presented it in greater detail, such as Guyer (2005), Zuckert (2007), Breitenbach (2009), and Förster (2011).

Kant on Nonhuman Animals and God

91

4.2 Animal orders from the human point of view 4.2.1 Animals belong to the mechanical order of things Let us begin with the first of the six claims that Kant is committed to. From the perspective of the human power of judgment, nonhuman and human animals are mechanical ‘beings’ since they undergo motions and changes that fall under mechanical powers and laws. Textual evidence for this position can be found in central passages of §§64 and 65 of the CPJ, where Kant compares the generation of nonhuman and human animals as organized beings with the production of mechanical objects, and says that an organized being is ‘not a mere machine, for that has only motive power, while the organized being possesses in itself formative power . . . which cannot be explained through the capacity for movement alone (that is mechanism)’ (CPJ 5:374.21–6). In these lines Kant says two things. First, that essential parts of nonhuman and human animals can, and even must, be explained as machines by means of mechanical powers and laws.4 Second, that organized beings are in some way more than machines, since organized beings cannot be explained by mechanical laws alone.5 I will discuss the first claim now, and will come back to the second statement in the next section of the paper. Kant never sufficiently defines the terms ‘moving powers’, ‘mechanism’, and ‘mechanical laws’ in the CPJ. As a consequence, the particular meaning of these terms is controversial among interpreters. But it is relatively uncontroversial that there are two important sources of the meaning of these concepts beyond the CPJ: the First Critique and the Metaphysical Foundations of Natural Science.6 In the first of these sources, Kant’s general metaphysics or ‘ontology’ (A846/B874) in the First Critique, Kant develops an account of transcendental laws of nature, which contains the central mechanical law or law of efficient causality: ‘[a]ll alterations occur in accordance with the law of the connection of cause and effect’ (A189/B232). Kant points to this mechanical law of efficient causation in the central §65 of the Third Critique where he describes it as a causal series that only proceeds on the part of the effects. Mechanical efficient causality consists in a 4 In the immediate context of this passage (CPJ 5:374.9–26) Kant gives further explanations of the term ‘machine’. He uses the example of a watch and claims that a machine is an object that cannot produce and reproduce itself, in whole or in part, but is produced by outer causes. 5 In his comparison of organized beings and machines, Kant uses the machine example of a clock. This comparison can be considered a critical response to Descartes’s mechanical reductionism; in his Discourse on the Method (1637, in: 1985: 141), Descartes equates animals with clock mechanisms. Animals, Descartes writes, ‘have no intelligence at all’, and ‘it is nature which acts in them according to the disposition of their organs. In the same way a clock, consisting only of wheels and springs, can count the hours and measure time’. A similar equation of a clock mechanism with a tree can be found in the Principles of Philosophy (1644, in: 1985: 288–9). 6 See McFarland (1970: 1, 14–16), Ginsborg (2001: 238–40 and 2004: 40, 43), McLaughlin (1989: 137–46), van den Berg (2014: 53–87), Quarfood (2004: 196–205, esp. 197–8), Zuckert (2007: 101–8), and Breitenbach (2009: 37–59). For an extended discussion of these analyses, see Goy (2017: 194–7).

92

Ina Goy

series of causes and effects ‘that is always descending’, and in which things ‘which as effects presuppose others as their causes, cannot conversely be the causes of these at the same time’ (CPJ 5:372.20–3). Kant contrasts this mechanical law of efficient causation with final causes. The second source of the meaning of the concepts ‘mechanism’ and ‘mechanical laws’ is Kant’s special ‘metaphysics of corporeal nature’, alternatively called ‘rational physics’ or ‘rational physiology’ (A846/B874). In the Metaphysical Foundations of Natural Science Kant develops another multifaceted account of mechanisms and mechanical laws, which covers phoronomical, dynamical, mechanical (in a narrower sense), and phenomenological aspects of mechanism. If we take these two sources of the meaning of the concepts ‘moving powers’, ‘mechanism’, and ‘mechanical laws’ seriously, what do humans know when they know that animals fall under mechanical powers and laws? Based on the account of mechanism in Kant’s general metaphysics, they know that the causal behaviour of animals follows the law of efficient causation (and the other transcendental powers and laws of nature). A natural product or one of its parts is something that follows something else according to a rule. States of motion and rest, and their changes, are effected by causes prior in time. Based on the account of mechanism in Kant’s special metaphysics, according to phoronomical laws humans know that the matter of animals and human beings as natural beings moves in simple or complex motions (MFNS 4:483.8–9, MFNS 4:486.31–3). Dynamical laws let them understand that animals constitute and preserve their place in space by means of attractive and repulsive moving powers (MFNS 4:498.17, 21). By means of mechanical laws (in a narrower sense) they know that animals and human beings as natural beings are either moved or in the state of rest. They can change their state of rest into a state of motion, their state of motion into a state of rest, or their state of motion into another state of motion if external causes bring about changes in their existing states (MFNS 4:541.28–30, MFNS 4:543.16–20, MFNS 4:544.32–3). These external causes can either be other material corporeal objects or external causes that do not belong to the system of physics, for instance the formative power as a final natural cause or the causality of freedom as a final moral cause. But in view of the Third Critique’s account of mechanism, mechanical laws, and moving powers, the accounts of mechanism outlined previously seem to be insufficient. Now the term ‘mechanism’ in the CPJ concerns the empirical (mechanical) manifold of nature (CPJ 5:179.16–186.21), which Kant tries to subordinate to higher teleological orders and laws, rather than the a priori universal mechanical powers and laws of organized beings. The mechanical powers and laws that Kant discusses in the Third Critique are not only the highly abstract laws discussed previously, but are rather their concrete and manifold empirical applications.7 7 Here I agree with Breitenbach (2009: 56).

Kant on Nonhuman Animals and God

93

4.2.2 Animals belong to the physical-teleological order of things Let us consider a second claim to which Kant is committed. From the perspective of the human power of judgment, nonhuman and human animals are organized beings, since their mechanical motions and changes are directed toward the fulfilment of natural purposes. The manifold mechanical motions and changes that fall under mechanical powers and laws are subordinate to (as I call them) physicalteleological powers and laws.8 Similar to the case of mechanism and mechanical powers and laws, Kant rarely mentions concrete examples of physical-teleological powers and laws; but some textual evidence for both can be found in §65 again, and in §70. In §65 Kant describes a physical-teleological power, when he says that an organized being is ‘not a mere machine, for that has only motive power, while the organized being possesses in itself formative power [bildende Kraft], and indeed one that it communicates to the matter, which does not have it (it organizes the latter); thus it has a self-propagating formative power, which cannot be explained through the capacity for movement alone (that is mechanism)’ (CPJ 5:374.21–6). The formative power [bildende Kraft] in nonhuman and human animals is different from mechanical powers of motion (e.g. the attraction and repulsion of matter), since, other than moving powers, the formative power is directed towards the realization of the natural end. The formative power is part of final causal processes in nonhuman and human animals, since it organizes and directs the motion and change that mechanical powers and laws bring about towards a particular natural end.9 In §70, Kant formulates the general form of a physical-teleological law in the context of his discussion of an apparent antinomy (CPJ 5:386.4–10) between mechanical and physical-teleological powers and laws.10 In this context, Kant expresses doubts that nonhuman and human animals can be explained by mechanical powers and laws alone: ‘Some products of nature’, Kant notes, ‘cannot be judged as possible according to merely mechanical laws (judging them requires an entirely different law of causality, namely that of final causes)’ (CPJ 5:387.6–9). Short as this passage is, it gives its readers three significant features of physicalteleological laws. Firstly, they are regulative maxims of the reflecting power of judgment; secondly, they are final causal powers and laws; thirdly, they go beyond mechanical explanations. What exactly does this mean? Physical-teleological laws are hypothetical laws in the form of imperatives that let humans judge all mechanical characteristics of nonhuman and human 8 I call these powers and laws ‘physical-teleological powers and laws’ because their teleological features are related to the sphere of physics (nature), as the teleological features of moral-teleological powers and laws are related to the sphere of morality (freedom). 9 I provide a highly detailed analysis of sentence CPJ 5:374.21–6 in Goy (2014a: 52–6). 10 This antinomy is similar to the third and fourth antinomies in Kant’s First Critique and the antinomy of practical reason in Kant’s Second Critique.

94

Ina Goy

animals as if they were brought about to fulfil a natural purpose; i.e. as if they were brought about to fulfil what it means for this human or nonhuman animal to be this human or nonhuman animal; or to judge all mechanical characteristics of a part of a human or nonhuman animal as if they were brought about to fulfil what it means for this part of a human or nonhuman animal to be this part of a human or nonhuman animal. For instance, physical-teleological laws let humans consider all mechanical characteristics of a nonhuman animal like a bird such as if they were brought about to fulfil the bird’s natural purpose (its birdhood); or to judge a wing of a bird such as if its mechanical characteristics were there to fulfil what it means for this wing of a bird to be this wing of a bird. To put this differently: physical-teleological laws let humans judge all mechanical explicable characteristics of a wing such as if they were there ‘for the sake of flying’. Analogous in the case of a human animal: they let humans judge all mechanical explicable characteristics of an eye such as if they were there ‘for the sake of seeing’. In a physical-teleological law, mechanical explicable empirical characteristics are referred to the idea of a natural purpose, whereby the concept of a natural purpose establishes a uniform and coherent relation among the mechanical explicable empirical characteristics that is not already contained in the mechanical explanations. Physical-teleological powers and laws have an empirical a priori structure, since they contain the concept of natural purpose, e.g. ‘for flying’, ‘for seeing’. Natural purposes are empirical concepts, since flying and seeing as purposes can be observed. But the unity of the mechanical characteristics that are brought under these concepts, e.g. the unity of all the mechanical characteristics of a wing that serve the function of flying, is thought with strict necessity. Since there is no necessary unity in the empirical realm, Kant argues that the necessary unity of characteristics that are brought under the concept of a natural purpose cannot be derived from experience—it is a product of an a priori activity of human reason. Kant hints at the empirical a priori character of physical-teleological laws in the first ‘Introduction’ of the Third Critique, when he says that the teleological judgment ‘is grounded on a principle a priori . . . although in such judgments we discover the end of nature solely through experience’ (CPJ 20:239.27–30). Ideas of natural purposes as part of physical-teleological laws are a priori ideas of reason, insofar as they instruct humans to search for the purposive unity among the empirical manifold mechanical characteristics of nature. But they are empirical concepts insofar as they can be known solely through experience. Ideas of natural purposes as part of physical-teleological laws thus force humans to search for an a priori necessary unity under empirical concepts, for instance, under the empirical concept ‘for flying’, ‘for seeing’. According to Kant, empirical mechanical laws and the mechanical characteristics of human and nonhuman animals that they describe can be subordinate as

Kant on Nonhuman Animals and God

95

means to the ends of physical-teleological laws and the natural purposes that they contain. Thus physical-teleological powers and laws have an explanatory power that mechanical powers and laws do not have.11 Physical-teleological laws introduce the causality of final causes into the observation of nature. The natural purpose in physical-teleological laws is considered the final cause of a certain teleological form of nonhuman and human animals. According to Kant’s account of mechanism, the motion of matter as an effect is directed towards something; however, this direction is contingent in the sense that it is dependent upon the direction of the impacting power and could be turned into another direction under the influence of an alternative power. Kant’s second law of mechanics, for instance, says that ‘[e]very change in matter has an external cause’ and that ‘[e]very body persists in its state of rest or motion, in the same direction, and with the same speed, if it is not compelled by an external cause to leave this state’ (MFNS 4:543.16–20). According to a mechanical law, humans know that a nonhuman or human animal is in a state of rest or motion. If it is in motion, it moves into a certain direction when an external cause pushes it into this direction. If it is pushed, it moves into the direction of the pushing power; if it is pressed, it moves into the direction of the pressing power, and so on. It moves with a certain speed when an external cause moves it with a certain speed. But the direction of the impacting power is contingent, it depends as an effect on the direction of another impacting power, or alternatively on another impacting power, or another, and so ad infinitum. Within the realm of empirical mechanical explanation, it is impossible to determine the source of the necessary directedness of mechanical motions and changes towards a unified end. There is no ultimate necessity for this direction of a motion or change, or an ultimate reason why several motions of nonhuman and human animals move into the same direction. Natural purposes as final causes explain the necessary directedness, the intentionality of motions and changes in nonhuman and human animals, insofar as they are directed towards a particular end, namely the fulfilment of the natural purpose. But let us pause again for critical considerations. Kant assumes certain hierarchies among natural laws when he suggests that nonhuman and human animals have characteristics that can be explained in accordance with mechanical and physical-teleological powers and laws. One of these hierarchies is that mechanical characteristics are subordinate to physical-teleological powers and laws. Another hierarchy exists among physical-teleological powers and laws themselves. Kant claims that not all natural purposes are of the same rank: there are subordinate 11 See, for the discussion of the debate of the mechanical inexplicability of animals, Ginsborg (2001 and 2004), van den Berg (2014: 54, 82–4, 89), Zuckert (2007: 108–19), Breitenbach (2009: 61–6), and Goy (2017: 216–23).

96

Ina Goy

and superordinate natural purposes, since certain natural purposes of human animals predominate over the natural purposes of nonhuman animals. Human animals are the ‘titular lord[s] of nature’ (CPJ 5:431.5) because the natural purposes of human animals outmatch the natural purposes of nonhuman animals in two respects: human animals strive for happiness and for culture. Happiness as a natural purpose is the ‘idea of a state’ of physical (bodily) satisfaction to which human animals ‘would make [their] instincts adequate’ if it were possible (CPJ 5:430.6–10). Kant seems to think that only humans can have the idea that the arrangement of all their instincts and physical mechanisms serves their physical fulfilment as a whole, whereas nonhuman animals simply follow their instincts and physical mechanisms without a reflection on their purpose (even if they have one) and the resulting happiness. But Kant does not make this argument very strong, since he admits that this idea of happiness is an ‘unstable concept’, which humans ‘arbitrarily set for [themselves]’, determine in ‘many ways’ and with ‘frequent changes’ (CPJ 5:430.10–16), since their nature ‘is not of the sort to call a halt anywhere in possession and enjoyment and to be satisfied’ (CPJ 5:430.22–3). A stronger notion of a superior natural purpose that only human animals have is what Kant calls ‘culture’ and defines as the ‘production of the aptitude of a rational being for any ends in general’ (CPJ 5:431.28–30). Kant distinguishes between two kinds of cultivation. The ‘culture of skill is . . . the foremost subjective condition of aptitude for the promotion of ends’ in general, which is not sufficient to determine the will in its choice of ends. The ‘culture of training’, in contrast, is the promotion of ‘the will in the determination and choice of its end, which . . . is essential for an aptitude for ends’. The latter is still a negative capacity, which ‘consists in the liberation of the will from the despotism of desires’. It makes human animals capable of choosing ends of (practical) reason (CPJ 5:431.35–432.12), but does not already include the capacity to choose them. The culture of training, Kant claims, is the ‘ultimate end’ (CPJ 5:431.13) within the realm of ends of nature, even if it is not the ‘final end’ (CPJ 5:431.15) of nature.

4.2.3 Animals belong to the moral-teleological order of things Let us consider a third claim to which Kant is committed. From the perspective of the human power of judgment, nonhuman and human animals are not only natural purposes and, as such, relative ends in themselves, but they serve in internal and external relations as means to the realization of the moral kingdom of ends as their final, absolute end. Physical-teleological powers and laws can be subordinate as means to the end(s) of the moral-teleological power and law. This third claim concerns the perspective of the individual human animal that falls under physical- and moral-teleological powers and laws. It also concerns the community

Kant on Nonhuman Animals and God

97

of nonhuman and human animals and their relations to each other and to the whole that Kant calls a ‘kingdom of ends’. Even though many scholars recognize that animals belong to mechanical and physical-teleological orders in the Third Critique, the majority of interpreters, for instance, Zuckert (2007) and Ginsborg (2001 and 2004), underestimate or even ignore the moral dimension of animal orders. But the problem is pressing for nonhuman animals and their interactions with other animals, since nonhuman animals belong to an order according to which other beings can externally impose demands on them. It is pressing also for human animals, since they have to deal with the sometimes incompatible aims of natural- and moral-teleological orders in themselves as individuals and in human society. What are moral-teleological laws according to Kant, and in which sense can natural purposes and physical-teleological powers and laws become subordinate to moral-teleological powers and laws? Kant’s practical law in its general formulation is familiar to many readers. It is an imperative that obligates human animals: to ‘act only in accordance with that maxim through which you can at the same time will that it become a universal law’ (G 4:421.7–8), or, in an alternative formulation, so to act ‘that the maxim of your will could always hold at the same time as a principle in a giving of universal law’ (CPrR 5:30.38–9). For the mediation of Kant’s thought at this point, it is decisive to see that Kant’s account of animals in the CPJ does not appeal to the practical law in its general formulation, but to its teleological variants, since with regard to moral issues, not the moral but the moral-teleological perspective is the primary concern of the CPJ. One of the teleological formulas of the practical law is a hypothetical imperative that summons human animals so to ‘act that [they] use humanity, whether in [their] own person or in the person of any other, always at the same time as an end, never merely as a means’ (G 4:429.10–12). This moral-teleological law attributes an absolute worth to human animals as final ends (persons), but not to nonhuman animals. Human, and only human, animals can and must subordinate natural purposes to moral purposes, and treat the latter as absolute ends. Another version of the teleological formula of the practical law summons human animals to consider all beings such as if they were part of a kingdom of ends that human animals bring about by means of their moral actions; the ‘kingdom of ends’ is the idea of ‘a whole of all ends in systematic connection’ (G 4:433.21–4). Kant provides extensive argumentative support for the end-directed hierarchy of natural and moral purposes and for the superiority of moral purposes over natural purposes. He argues that the rational (noumenal) part of the human animal is the final end of nature because of the rational (noumenal) part of the human animal it ‘cannot be further asked why (quem in finem) it exists’ (CPJ 5:435.25–7); it is that end ‘which needs no other as the condition of its possibility’ (CPJ 5:434.7–8). The reason for this is that the causality of freedom is ‘teleological, i.e. aimed at ends’ and the law in accordance with which it has to determine ends

98

Ina Goy

is ‘unconditioned and independent of natural conditions but yet necessary in itself ’ (CPJ 5:435.15–19). Furthermore, only the human animal as a rational (noumenal) being ‘forms a concept of ends for himself ’ and can ‘by means of his reason’ make ‘a system of ends out of an aggregate of purposively formed things’ (CPJ 5:426.35–427.3). While, as I have said earlier, physical-teleological laws are maxims of the reflecting power of judgment in which the empirical manifold of nature is subordinated to the a priori empirical concept of a natural purpose, the moral-teleological law, as is visible now, is a maxim in which the still empirical conditioned concepts of natural purposes get subordinated to the pure rational, a priori concept of a moral purpose. The moral purpose is no empirical concept, since it concerns the purely rational aspect of human being(s) only: an individual human being that chooses to act in accordance with and for the sake of its rational (noumenal) nature.

4.3 Animal orders from God’s point of view So far we have considered three kinds of powers and laws that humans use when they judge nonhuman and human animals: mechanical, physical-teleological, and moral-teleological powers and laws. But how can it be that the human power of judgment uses three kinds of powers and laws to judge animals? Does this not result in quarrels between the three kinds of powers and laws? With these considerations we reach a new dimension in Kant’s account of nonhuman and human animals, and a point of view other than the human one: the regulative idea of God’s point of view. Nonhuman and human animals enable humans to believe in the regulative idea of God and God’s creation since God represents the original unity of the different kinds of powers and laws and God’s creation accounts for the possibility of their unification in nature and their unifiability in the human judgment of nature. While it is possible to form a hierarchy, there is no strict unity among the three kinds of powers and laws in human consciousness. The problem of the lack of unity cannot be resolved from the human point of view. In order to account for the compatibility and unifiability of the three kinds of powers and laws in the point of view of humans and in nature as it appears to human judgment, Kant assumes the regulative idea of a point of view of God, in which the powers and laws are one (CPJ 5:176.10; CPJ 5:412.24, §§85–6).

4.3.1 The unity of animal orders from God’s theoretical point of view A fourth claim that Kant is committed to is that nonhuman and human animals allow humans to believe in God for theoretical reasons, since humans use two

Kant on Nonhuman Animals and God

99

different kinds of theoretical (mechanical und physical-teleological) natural powers and laws when they judge nonhuman and human animals. These powers and laws could conflict if there were no original unity among them. Kant argues in particular in §§74–8 and in §85 of the CPJ, that mechanical and physicalteleological powers and laws can be thought of as originally united in the regulative idea of God’s understanding—more precisely, God’s theoretical (CPJ 5:440.5; CPJ 5:441.16, 34), intentional and intelligent (CPJ 5:438.6, 14–15; CPJ 5:440.22; CPJ 5:444.13) consciousness; and, since God is a creator, these powers and laws do not conflict—neither in the human consciousness nor in human experience of nature, both of which are created. In the §§70–1 Kant gives a detailed account of the possibility of an apparent conflict between the theoretical laws of nature in human consciousness and human experience of nature. What is the nature of this apparent conflict? When humans consider the empirical manifold of the characteristics of nonhuman and human animals and want to bring them, by means of more general powers and laws, under a superordinate unity of higher-order powers and laws, they can achieve this superordinate unity either by the subordination of the empirical manifold to more general mechanical powers and laws or by the subordination of the empirical manifold to physical-teleological powers and laws. The difference between both kinds of superordinate mechanical and physical-teleological natural powers and laws consists in the discursive element: a superordinate mechanical law contains a concept of understanding; a superordinate physical-teleological law contains a concept of reason, the idea of a purpose. To give an example: when humans observe several concrete pressing powers and the resulting mechanical motions in nonhuman and human animals, they can judge, on the one hand, that whenever several kinds of pressing powers and laws affect an animal, the resulting mechanical motions turn into the directions into which they are pushed by the pressing powers and laws. In this way they find a superordinate mechanical explanation of the pressing powers and laws in animals that explain the uniform structure in all pressing powers and laws and the resulting mechanical motions. Or, on the other hand, they could judge that the concrete pressing powers and laws that affect the animal and cause the corresponding mechanical motions help to fulfil the purpose of the human or nonhuman animal, namely for instance the purpose to become this man or this bird, or the purpose to fulfil this function of a leg or this function of a wing, etc. In §78 of the CPJ, Kant claims that, from the human point of view, the apparent conflict between both kinds of higher-order powers and laws can be resolved, since the mechanical powers and laws can be subordinate to the superordinate physical-teleological powers and laws (but not the opposite). Both kinds of superordinate explanations form a hierarchy but no unity in human judgment, in which superordinate physical-teleological contain superordinate mechanical explanations (CPJ 5:414.9, 15, 27; CPJ 5:415.11, 22). Unlike how it is in humans’

100

Ina Goy

limited and incomplete theoretical consciousness, Kant claims, in God’s unlimited and complete theoretical consciousness natural mechanical and physicalteleological powers and laws form a strict unity. The nature of animals as it is represented in God’s consciousness is not divided into different natural powers and laws; it is one.12 How can this be? The divine understanding, Kant claims, is not a discursive but an exclusively intuitive, archetypical understanding (CPJ 5:402.2; CPJ 5:406.24–5, 33; CPJ 5:407.21, 32), one that exists even prior to the possibility of diverse laws, since the diversity of the laws depends on their discursive elements.13 Kant describes the divine theoretical understanding as an ‘intellectus archetypus’ (CPJ 5:408.19, see CPJ 5:407.32), i.e. the faculty of the ‘complete spontaneity of intuition’ (CPJ 5:406.21–2). In the divine intuitive understanding, everything that can be represented at all is represented by means of intuitions (instead of different concepts). For this reason, the original unity of the theoretical orders of nature in God’s consciousness is different from the mere unifiability of the theoretical orders of nature in the human consciousness. The awareness of this difference lets human beings believe in an intelligent and intuitive God as the creator of the theoretical (natural) orders of nonhuman and human animals. Only the possibility of an original unity of the natural orders in God can guarantee that human experiences of the theoretical orders of nature are consistent since they are derived from the original theoretical divine ground of nature in which the powers and laws of nature are one: it is God’s creation that brings about the human consciousness as much as nature (nonhuman and human animals) and their theoretical orders (CPJ 5:399.37–400.6, CPJ 5:410.7–11).14

4.3.2 The unity of animal orders from God’s theoretical and practical point of view Let us now consider the fifth of the six claims to which Kant is committed. Nonhuman and human animals enable humans to believe in God for theoretical and practical reasons, since humans must make use of two different theoretical physical-teleological and practical moral-teleological powers and laws when they 12 In the background of this regulative, theoretical idea of God Kant discusses the classical physicotheological argument for the existence of God to give prove of the concept of God that he needs to defend the unity of natural powers and laws; see Wood (1970: 171–6 and 1978: 130–45), McFarland (1970: 1–2), and Goy (2014b: 203–20). 13 There is a significant amount of new literature exploring Kant’s conception of God’s consciousness, the intuitive understanding, as opposed to man’s consciousness, the discursive understanding. Most original and influential in this debate is Förster’s research (both 2002a, 2002b, 2008; 2011: 147–60, 253–76); but see also Frank and Zanetti (2001: 1292–303), Quarfood (2004: 160–208 and 2014), and Goy (2015). 14 See my detailed interpretation of the antinomy of teleological judgment in Goy (2015).

Kant on Nonhuman Animals and God

101

judge nonhuman and human animals as relative ends and as means and ends in internal and external relations in which they are subordinate to absolute moral ends. These powers and laws could conflict if there were no original unity among them. Kant argues that physical-teleological and moral-teleological powers and laws can be thought of as originally united in the regulative idea of God’s theoretical and practical consciousness, and for this reason do not necessarily conflict, neither in our human consciousness nor in the natural and moral spheres of the world, since both are created. In the second ‘Introduction’ to the CPJ, Kant describes the possible conflict between physical-teleological and moral-teleological laws and indicates the solution to this problem in the divine unity of both kinds of powers and laws: Now although there is an incalculable gulf fixed between the domain of the concept of nature, as the sensible, and the domain of the concept of freedom, as the supersensible, so that from the former to the latter . . . no transition is possible, just as if there were so many different worlds, the first of which can have no influence on the second: yet the latter should have an influence on the former, namely the concept of freedom should make the end that is imposed by its laws real in the sensible world; and nature must consequently also be able to be conceived in such a way that the lawfulness of its form is at least in agreement with the possibility of the ends that are to be realized in it in accordance with the laws of freedom.—Thus there must still be a ground of the unity . . . [that] makes possible the transition from the manner of thinking in accordance with the principles of the one to that in accordance with the principles of the other (CPJ 5:175.36–176.15).

There are two main types of unification problems at this point: the unification of natural and moral powers and laws within the human animal, and the unification of natural and moral laws in external relationships between nonhuman and human animals.15 Let us consider examples of both. First, imagine a human who intends to perform a moral action. It might be that the human needs his or her physical skills, such as running, in order to reach a friend who is in danger and need of help. In this situation, the human has to find a connection and order between two different kinds of natural and moral powers and laws in order to subordinate the

15 Nonhuman animals play no ineliminable role in inducing religious belief in Kant’s account, neither as natural beings, nor as beings whose nature can serve the moral purposes of humans. If nonhuman animals did not exist, humans would still have reasons to believe in the regulative notion of a theoretical practical God since it would allow them to assume that the mechanisms of the motions of their matter serve natural purposes, and that their natural purposes serve their moral purposes. In this paper, I argue just for the claim that animals can be part of the nature that human beings consider and that is referred to in their formation of their belief in God.

102

Ina Goy

natural to the moral powers and laws (running fast in order to help the friend). If there were no possibility of connecting both kinds of powers and laws, the human animal would not be able to perform a moral action. A more complicated case appears when the human’s physical needs, which fall under mechanical and physical teleological powers and laws (hunger, concern for health), contradict his or her moral duties (to run fast in order to help a friend in danger). In this situation the human has to find a connection between two different kinds of natural and moral powers and laws that impose contradictory demands on them. If the human wants to act morally, he or she must not only subordinate the mechanical and physical-teleological powers and laws (hunger, concern for health) to the moral-teleological power and law (to run in order to help a friend in danger), but must also postpone the latter in order to perform the moral action first. If there were no possibility to connect and arrange both kinds of powers and laws, the human would not be able to perform a moral action. Second, let us imagine situations in which a human animal wants to perform a moral action that requires talents and facilities of nonhuman animals that fall under mechanical and physical-teleological natural powers and laws. For instance, if a human animal wants to find the traces of a murderer and needs the help of a nonhuman animal, say, a sniffer dog, or if a human animal must take care a comrade who got lost in ski accident in an avalanche and needs the help of nonhuman animals and artificial objects, say, two dogs and a sled. In these cases, the human animal who wants to follow moral teleological powers and laws will use nonhuman animals and artificial objects, insofar as they fall under mechanical and physical-teleological powers and laws, and will subordinate the mechanical and physical-teleological powers and laws to the moral-teleological laws in an external means-end relationship. If there were no possibility to connect both kinds of natural and moral powers and laws, the human animal would not be able to perform a moral action. We could construe further cases in which the moral purposes of the human community require a subordination of physical-teleological and mechanical to moral-teleological laws. From the human point of view, a hierarchy and subordination of powers and laws suffices to unify both kinds of powers and laws, even though humans cannot unite both kinds of powers and laws. What does the regulative idea of God and God’s creation add in this case, and how does it help to overcome the problem of an apparent conflict and lacking unity between both kinds of natural and moral powers and laws? Kant develops the answer to these questions in §86. In this section, Kant describes the regulative idea of a theoretical intelligent, intentional and practical wise (CPJ 5:441.3–4, CPJ 5:442.3–4) God, who is the ground of the original unity of the theoretical orders of nature and the practical order of morality. Analogous to the unity of the theoretical mechanical and physical-teleological natural powers and laws, Kant employs a regulative idea of God whose consciousness contains the unity of the physical- and moral-teleological powers and laws.

Kant on Nonhuman Animals and God

103

As in the previous consideration, the original unity of the theoretical and practical orders of nature and morality in God’s consciousness is different from the mere unifiability and unification of the theoretical and practical orders of nature and morality in human consciousness. This difference lets humans believe in an intelligent, intuitive, and practical wise God as the creator of the theoretical and practical orders of nonhuman and human animals. Only the possibility of an original unity of the natural and moral orders in God can guarantee that human beings’ experiences of the theoretical and practical orders of nature and morality are consistent since they are derived from the divine ground of nature and morality in which the powers and laws of nature and morality are one.

4.3.3 The unity of God’s point of view Finally, we have to consider our sixth claim. I will keep this point short. The purposive existence of nonhuman and human animals, embedded in a purposive creation, enables humans to believe in the regulative idea of one and only one God. The God who represents the original unity among the natural powers and laws is identical with the God who represents the original unity among the natural and moral powers and laws. To create unity is the task of one and the same God. To put this differently: the regulative idea of God’s theoretical consciousness is just an aspect of the idea of God’s theoretical practical consciousness. Even though the extension of the former would suffice to account for the unity of the natural powers and laws (mechanism and physical teleology) it would not suffice to account for the natural and moral laws (natural and moral teleology). What kind of unity is the unity of God’s point of view? The unity of natural and moral powers and laws in God’s consciousness is an absolute unity; it is the human regulative idea of a supersensible ground, which is creative and causally active (‘Grund der Einheit’ CPJ 5:176.10, CPJ 5:180.18–181.2, CPJ 5:395.13–21). It is not the right question to ask Kant what this ground adds if the natural and moral powers and laws are unifiable and have relative ‘unity’ already in the human consciousness, because for Kant the latter would not be consistent and logical coherent at all if they were not created. According to Kant, God’s consciousness is uncreated and perfect; it represents the absolute unity of the faculties of the mind and the absolute unity of the natural and moral powers and laws. The human consciousness, in contrast, is created and less perfect; the faculties of the human mind are distinct—they bring about distinct natural and moral powers and laws, and with them the need of the consistency and relative unifiability of these powers and laws. With this analysis of Kant’s account of the notion of the role of nonhuman animals and of the human animal’s belief in God, a limitation of that conception can be made more clear. Although Kant connects the regulative idea of God and

104

Ina Goy

God’s theoretical intelligence and practical wisdom with classical biblical divine attributes such as omniscience, omnipotence, omnibenevolence, justice, eternity, omnipresence, etc. (CPJ 5:444.15–28), Kant never explicitly refers to a (regulative version of a) Christian model of God. The Kantian model of God lacks, for instance, one of the main characteristics of the Christian conception of God: the universal divine love, a power that unifies and embraces all beings, including nonhuman and human animals and their orders. According to Kant neither is the regulative idea of the unity of natural and moral powers and laws in God’s point of view connected to the Christian idea of the universal love of God, nor are human animals committed to love nonhuman and human animals to achieve a unification of natural and moral laws from man’s point of view. Kant’s regulative idea of God is not that of a lover whose love makes all animals equal and one, but that of a divine craftsman and king (CPJ 5:444.12–27), an ordering, organizing God who creates nonhuman and human animals and lets them subordinate mechanical orders to physical-teleological, and physical-teleological to moralteleological orders. Kant’s anthropocentricism and Enlightenment rationalism entail that the notion of God that is formed is one that must love only human animals in a genuine way—since they possess a rational nature—but that must at best only love nonhuman animals indirectly, if at all.

5

Animality in Kant’s Theory of Human Nature David Baumeister

5.1 Introduction It is well known that Kant posits a series of categorical distinctions between humans and non-human animals. The oft-quoted opening lines of the first section of the Anthropology from a Pragmatic Point of View summarize his position well: ‘The fact that the human being can have the “I” in his representations raises him infinitely above all other living beings on earth. Because of this he is a person . . . i.e. through rank and dignity an entirely different being from things, such as irrational animals, with which one can do as one likes’ (Anth 7: 127).1 A less remarked upon, but arguably equally significant, feature of Kant’s position is that, despite their differences, both humans and non-human animals are animals, and share a common animality or Tierheit. Whereas on this view irrational animals are things, it does not follow from this that because the human is not a thing he or she is thereby not an animal. Indeed, on Kant’s view of human nature, animality—human animality—provides an indispensable basis for the development of more properly human capacities and characteristics—such as technical skill, personality, and humanity itself. Hence, while certain essential features differentiate humans from other animals (above all the possession of reason), it remains for Kant a fact of human nature that humans cannot jettison their animality or cease to be animals; although the human’s capacity for self-consciousness ‘raises him infinitely above’ the other animals, the human nevertheless ‘retains something of the beast in him, which he cannot overcome’ (EC 27: 441). Kant wrote no treatise on human animality. Instead, we find piecemeal discussions scattered throughout his writings, including each of the Critiques, numerous lecture courses, and most of the texts concerning anthropology, religion, and history. This essay considers a range of these sources in order to provide an account of the role of animality in Kant’s theory of human nature. I set aside both the task of examining Kant’s account of animality as a general biological category 1 All italics in passages from Kant’s texts cited in this essay appear in the original.

David Baumeister, Animality in Kant’s Theory of Human Nature In: Kant and Animals. Edited by: John J. Callanan and Lucy Allais, Oxford University Press (2020). © David Baumeister. DOI: 10.1093/oso/9780198859918.003.0006

106

David Baumeister

applying to non-human animals in addition to humans,2 as well as the question of how exactly the account of human animality examined here sits with Kant’s claims regarding the moral status (or lack of status) of non-human animals.3 Limiting myself to those texts most directly involved in Kant’s project of delimiting human nature, I connect and interpret Kant’s most illustrative remarks on human animality, revealing how animality, though often in tension with the aims of morality in the human being, remains an irremediable component of human nature, contributing in its own way to the well-being and goodness of human life. The three sections that follow proceed systematically, first presenting Kant’s view of animality’s contribution to human nature in general and then specifying how this contribution plays out at the level of human nature’s individual and species-level development: 1) I begin by examining Kant’s concise exposition, in the Religion within the Boundaries of Mere Reason, of animality as an original and natural human predisposition. This is Kant’s most programmatic treatment of the concept. In the Religion, Kant situates animality among a typology of predispositions, including also the predispositions to humanity and personality. Locating this typology alongside the related typology of predispositions offered in the Anthropology, and relating it to Kant’s discussion of natural predispositions elsewhere, I show how human animality in the Kantian picture is both basic and good, only resulting in evil when improperly directed by injudicious maxims. 2) I next explore the place of animality in Kant’s theory of education. Paradigmatically in his lectures on pedagogy, Kant proposes a model of childhood education that encourages the reasoned, harmonious development of each of the human’s natural predispositions, including animality. While the technical, pragmatic, and moral predispositions require a practical education involving the progressive use of reason, the predisposition to animality requires a physical education involving the use of discipline to prevent the formation of bad habits. Rather than simply repressing animality, however, the sort of discipline Kant recommends is intended to guide the human’s animal nature in a way consistent with the aims of the other, more narrowly human predispositions. Kant’s educational model, therefore, highlights the contribution that animality, in tandem with other predispositions, makes to individual human development. 2 For a general discussion of Kant’s view of the nature of animals, see Kain (2010: 214–19); for a comparative analysis of Kant’s view in relation to eighteenth-century German philosophy, see Ingensiep (1996). For a comprehensive treatment of Kant’s philosophy of biology, including some, albeit limited, attention to the concept of animality, see the essays in Goy and Watkins (eds.) (2014). 3 For a general discussion of Kant’s view of the moral status of non-human animals, see Wilson (2012).

Animality in Kant ’ s Theory of Human Nature

107

3) Finally, I consider the place of animality in Kant’s account of the development of the human species as a whole. Similar to its important but curtailed role in childhood education, animality plays a decisive part in the teleological unfolding of human nature in history. And though the initial emergence and progress of humanity must, for Kant, take place against the background of the human’s animal nature, the ultimate goal of history, I show, is a reconciliation of human animality with the more properly human predispositions. As in the educational model, the vices and ills that characterize human history emerge not out of animality itself, but from the conflict of the other predispositions with animality. Kant’s view of true historical progress thus seeks to accommodate, rather than eliminate, the animality within human nature. I conclude that, for Kant, human animality is no mere obstacle to human development and moral perfection. It is, rather, an indispensable feature of human nature and a necessary contributor to human well-being. Recognizing this indispensability and necessity should occasion us to revisit our understanding of the relation between human and non-human animals in Kant’s philosophy more generally.

5.2 Animality as an Original Human Predisposition to the Good Although animality is invoked in a wide range of Kant’s texts, the concept is only given something like an explicit definition in Kant’s major work on religion, Religion Within the Boundaries of Mere Reason. Near the beginning of this work, animality appears as one of several predispositions that constitute the overall ‘original predisposition to [the]4 good in human nature’ (Rel. 6: 26). Kant enumerates three such predispositions: ‘the animality of the human being, as a living being’, ‘the humanity in [the human being], as a living and at the same time rational being’, and, finally, of the ‘personality [in the human being], as a rational and at the same time responsible being’ (Rel. 6: 26). That animality concerns the living dimension of the human being, while humanity concerns both this living dimension and a distinct rational dimension, provides a first indication that animality and humanity do not, in their original form, conflict with one another. Rather, humanity entails animality—the humanity of the human requires that the human be a living being. By contrast, in this formulation the predisposition to 4 Kant’s title for the section in questions reads, ‘Von der ursprünglichen Anlage zum Guten in der menschlichen Natur’ (Rel. 6: 26). While George di Giovanni, in the Cambridge translation cited above, renders ‘zum Guten’ as ‘to good,’ ‘to the good’ is both more true to the German article structure and more consistent with Kant’s usage of ‘zum Guten’ soon thereafter (at Rel. 6: 28, for instance), which di Giovanni in these latter instances himself translates as ‘the good.’

108

David Baumeister

personality, on which basis the human becomes morally responsible, does not entail animality in the same way. When Kant, in other texts, such as his pedagogy lectures or the ‘Conjectural Beginning of Human History’, writes of the conflict emergent in human nature between animality and humanity, it is precisely humanity on this latter, moral register that he has in mind. The typology of predispositions in Religion provides an initial sense of why this is so: the moral dimension of human nature is seen as categorically distinct from the animal dimension of that nature. Crucially, however, both dimensions remain natural, and the human, on Kant’s view, is naturally predisposed to both. This shared natural providence, along with the fact that at least one other properly human predisposition explicitly entails animality, resurfaces in Kant’s argument, which I consider below, that the human’s animal and moral destinies will ultimately harmonize, rather than conflict, with one another. Kant devotes only a handful of lines to defining each of the original predispositions in Religion, spending the fewest on the predisposition to animality. He writes the following: The predisposition to animality in the human being may be brought under the general title of physical or merely mechanical self-love, i.e. a love for which reason is not required. It is threefold: first, for self-preservation; second, for the propagation of the species, through the sexual drive, and for the preservation of the offspring thereby begotten through breeding; third, for community with other human beings, i.e. the social drive.—On these three can be grafted all sorts of vices (which, however, do not of themselves issue from this predisposition as a root). They can be named vices of the savagery of nature, and, at their greatest deviation from the natural ends, are called the bestial vices of gluttony, lust, and wild lawlessness (in relation to other human beings). (Rel. 6: 26–7)

That human animality consists in the human’s relating to itself as a physical being is also evidenced in Kant’s theory of education, which approaches the human’s animal nature in terms of ‘physical education’. The first aspect of this self-love or self-relation—self-preservation—appears as the child’s natural ability to determine its posture or physical form—an ability that, for Kant, must be cultivated with physical education. The second aspect of this self-relation features in Kant’s account of human history, in that the animality of the human being is seen to provide the impetus for the physical continuation of the human species generation after generation, allowing for history to unfold. That human animality is on this account held to be responsible for human community—providing as it does the ‘social drive’ [der Trieb zur Gesellschaft]— must be reconciled with other moments in Kant’s writings where the achievement of sociality is held to be a function of human reason, as here, in Kant’s 1771,

Animality in Kant ’ s Theory of Human Nature

109

anonymously published review of the Italian anatomist Pietro Moscati’s Of the Corporeal Essential Differences Between the Structure of Animals and Humans5: [T]he first foresight of nature was that the human being as an animal be preserved for himself and his kind; and for that the position which is most suited to his internal build, the situation of the foetus and the preservation in dangers is the four-footed one; but . . . there also has been placed in him a germ of reason through which, if the latter develops, he is destined for society, and by means of which he assumes permanently the most suitable position for society, viz., the two-footed one. Thereby he gains, on one side, infinitely much over the animals, but he also has to live with the discomforts which result for him from the fact that he has raised his head so proudly above his old comrades. (RezMoscati 2: 425)

It may be that Kant is employing different senses of society [Gesellschaft] in Religion and the 1771 review. More plausibly, society, on Kant’s account, may admit of being achieved by different means. The predisposition to animality allows the human to enter into society with other humans as a physical being, while reason allows it to do the same as a moral being. Indeed, this is consistent with Kant’s division of education into physical and practical components in the pedagogy lectures. The coherence of Kant’s account does not require that human society have a single source in human nature, especially if animality and reason are both held to be natural ‘germs’ of the human. Kant’s claim that numerous vices can be ‘grafted’ [gepfropft] onto each aspect of animality, while the predisposition to animality itself is nevertheless not the ‘root’ of such vices, evinces the duplicitous character animality assumes throughout Kant’s theory of human nature. Given that, on Kant’s account, the predisposition to animality in the human being fosters self-preservation, propagation, and physical community, it is understandable that many of the physical vices that afflict human beings are understood as extensions of this animal nature. Overindulgence in either eating, sexual activity, or ‘lawless behaviour’ extend otherwise nonvicious potentialities within human animality in this way. This is why, in his pedagogical theory, Kant recommends certain forms of discipline to prevent these vices from ‘grafting’ themselves onto our animal nature. For, as Kant is quick to point out, these vices represent a ‘deviation’ from ‘natural ends’ [Abweichung vom Naturzweck], and so do not reflect human animality itself, but rather its corruption or exploitation. As Allen Wood remarks, on Kant’s view, ‘[o]ur sensuous or animal nature is innocent, and in itself even something good’ (Wood 2012: 117). 5 For a reading of the development of Kant’s relationship to the science of anatomy, see Olson (2016).

110

David Baumeister

Despite the ‘innocence’ of human animality vis-à-vis the vices that may be grafted onto it, the predispositions to animality, humanity, and personality are, in Religion, enumerated progressively, one after the other. Hence it must be specified whether Kant holds them to be of different values. That is, does Kant hold the predispositions to humanity and personality to be somehow fundamentally better than the predisposition to animality? The answer, it would seem, is no. Kant elaborates: ‘All these predispositions in the human being are not only (negatively) good (they do not resist the moral law) but they are also predispositions to the good (they demand compliance with it). They are original, for they belong to the possibility of human nature’ (Rel. 6: 28). From this, several basic points become clear: 1) each predisposition is equally good, being equally original, and equally directed to the good, 2) the good must be understood teleologically, in that the predispositions are each directed to it, and 3) a predisposition’s being directed to the good entails that predisposition’s demanding compliance with the moral law. What does it mean, though, for a predisposition to demand compliance with the moral law? In this case, it appears that this cannot amount to that predisposition’s being moral, as Kant claims that the predisposition to animality ‘does not have reason at its root at all’ (Rel. 6: 28). This, along with the fact that, for Kant, morality requires reason,6 and, moreover, that the predisposition to animality is distinct from the predisposition to personality, means that human animality is not itself good in the moral sense. Even so, human animality has some relation to the good even in the moral sense, for it is directed to such a good. Hence, even though human animality is categorically distinct from human morality on Kant’s account,7 it can nonetheless, insofar as it ‘demands our compliance’ with the moral law, be said to contribute to morality, or, as Holly Wilson puts it, to ‘encourage moral action’ (Wilson 2001: 454). The specific ways in which animality, for Kant, contributes to morality, as well as to the flourishing of humanity more generally, will be explored in greater detail in the following sections. Comparing the set of predispositions provided in Religion with the distinct three-part typology of human predispositions Kant offers in the Anthropology provides a further means of appreciating the role of human animality as it is relationally defined in the former. Here neither animality nor the good appear in the typology. Rather, the three predispositions Kant identifies serve to set the human apart from the Earth’s other living inhabitants:

6 In Korsgaard’s words, ‘a Kantian does not believe in the split between rationality and morality’ (Korsgaard 1996: 395, n. 40). 7 I examine Kant’s separation of human animality and human morality, and explore the consequences of this separation for our theorization of present-day environmental crises, in Baumeister (2019).

Animality in Kant ’ s Theory of Human Nature

111

Among the living inhabitants of the earth the human being is markedly distinguished from all other living beings by his technical predisposition for manipulating things (mechanically joined with consciousness), by his pragmatic predisposition (to use other human beings skillfully for his purposes), and by the moral predisposition in his being (to treat himself and others according to the principle of freedom under laws). And any one of these three levels can by itself alone already distinguish the human being characteristically as opposed to the other inhabitants of the earth. (Anth. 7: 322)

While one might read the absence of the predisposition to animality in this account as revealing of how, on Kant’s view, animality is less essential to human nature than the technical, pragmatic, or moral predispositions, the omission of animality here in fact reflects the particular scope of the Anthropology. While Religion, which on the whole concerns the problem of evil in the world and in human life, seeks to specify every source of goodness in human nature, and so includes animality in its typology of human predispositions, the Anthropology is concerned with delimiting what is proper and specific to human beings and human beings alone, and so excludes animality from its typology; as Wilson explains, ‘The predisposition to animality is an aspect that other living beings on earth share, and, therefore, it is not fitting to include it in his account in the Anthropology’ (Wilson  2001: 451). Of course, human animality is referred to many times in the Anthropology, though always by way either of contrasting it with the more properly human predispositions or of specifying how it can factor, in better or worse ways, into human development. It is also significant that the specific typology of technical, pragmatic, and moral human predispositions appears only in the Anthropology from a Pragmatic Point of View and not in any of the extant transcripts of the many anthropology lecture courses that Kant offered between 1772 and his retirement from teaching in 1796, and which supplied the lion’s share of the raw material for the 1798 Anthropology book. The typology of predispositions that appears in Religion, written in the early 1790s, would thus seem to be a more viable precursor for the typology found in the Anthropology than any of the periodically diverse anthropology lectures. The typology in Religion might therefore be called the ‘original’ typology of ‘original human predispositions’. Though the typologies that appear in the Anthropology and Religion are distinct in terms of both scope and terminology, on inspection it is clear that they address the same basic set of predispositions, which in the final analysis totals four. Echoing Wilson’s reasoning in sorting the predispositions Kant names into a set of four (Wilson 2001: 451), we can conclude that, for Kant, human nature contains: 1) the predisposition to animality (which, being not exclusively human, falls outside the scope of the Anthropology), 2) the technical predisposition (which, in having little to do with good and evil falls outside the scope of Religion),

112

David Baumeister

3) the pragmatic predisposition (which, providing the basis for the human’s interaction with other human beings at a rational and prudential but not fully moral level, is in Religion called the predisposition to humanity), and 4) the moral predisposition (which, in so far as the human becomes a person when it becomes capable of acting morally, is in Religion called the predisposition to personality). Though the notion of advancing a tripartite typology of predispositions appears to be an innovation of the 1790s, Kant had used the concept of predisposition amply before, investing great promise in the pursuit and articulation of human predispositions. The Pillau transcript of Kant’s anthropology lecture course from 1777–8 states, ‘[f]rom various circumstances . . . we can discover certain predispositions from time to time and infer from them what nature’s goal for humanity is’ (AP 25: 839). A reflection of his broader teleological conception of nature, Kant took each human predisposition to have been implanted or impressed into human nature by nature writ large. That the human’s predisposition to animality would be implanted by nature is to be expected. That the more properly human predispositions—including the moral predisposition—also have a natural origin is more striking, given that Kant is commonly read to condone a moral disregard of nature and natural beings.8 The common natural providence of every predisposition within human nature helps us understand how, as in the typology found in Religion, animality fosters the good in human nature as much as any other predisposition. Such commonality also begins to explain how, owing to their being co-packaged within human nature, the predisposition to animality, on the one hand, and those to humanity, personality, or morality, on the other hand, can and do come into conflict—and hence must then be brought into accord. As I explore next, Kant’s vision of the task of education, as well as of historical progress, relies upon and further instantiates his view of the common natural ground among the predispositions.

5.3 Focus on the Philosophy of Education (Individual-Level Development) In his pedagogy lectures (posthumously published as Über Pädagogik), Kant declares that education [Erziehung] is ‘the greatest and most difficult problem that can be given to the human being’ (Ped. 9: 446). Education is so difficult because it requires that the human ‘develop his predispositions toward the good’ (Ped. 9: 446), which, in so far as multiple, potentially conflicting predispositions are involved, is no simple task. The goal of education, for Kant, is to develop the 8 Both Kant’s critics and defenders have made much of this area of Kant’s work, and the question of whether Kant is or is not ‘friendly to nature’ remains an open one in the literature. For representative criticisms of Kant on this point, see Singer (1975: 244ff) and Regan (2003: 77–82).

Animality in Kant ’ s Theory of Human Nature

113

predispositions of human nature so that the good at which these predispositions aim is not just a potential for the human being, but comes to be realized in a properly human life. ‘Many germs lie within humanity,’ Kant explains, ‘and now it is our business to develop the natural predispositions proportionally and to unfold humanity from its germs and to make it happen that the human being reaches his vocation’ (Ped. 9: 445). Bringing forth good in the world on a global scale can be called the task of the progress of human history, wherein, for Kant, human nature as a whole is carried further towards perfection. Bringing forth good in the world at the scale of the life of an individual is the proper task of education, and reflects the bearing of human nature in a single human life. That Kant across many texts repeats his call to both tasks attests to the broad compass of his theory of human nature, which has at its centre the accounts of original human predispositions discussed above. Animality plays a central role in Kant’s theory of education, being as it is one of the natural predispositions that the human must develop in order to bring good into the world. It bears repeating, then, that animality, as with the other predispositions, is, in Kant’s theory, itself a source of good, and that a proper education can develop the good out of human animality: Good education is exactly that from which all the good in the world arises. The germs which lie in the human being must only be developed further and further. For one does not find grounds of evil in the natural predispositions of the human being. The only cause of evil is this, that nature is not brought under rules. In the human being lie only germs for the good. (Ped. 9: 448)

With animality, however, and to a greater extent than with the other predispositions, it is possible for education to go astray. If the human’s animal nature is not ‘brought under rules’ through a good education, then evil can indeed result. The vices that can be grafted onto the human’s animal nature threaten in childhood to become bad habits that may last well into adulthood, imperilling or preventing the proper development of human morality as they are sustained. Animality must therefore be closely attended to during childhood development, and a sizable portion of Kant’s educational theory (including, arguably, the entire section of Über Pädagogik on ‘physical education’) is devoted to explaining just how it might be properly so attended. While the technical, pragmatic, and moral predispositions9 are each best addressed through practical education, which engages the activity of the child’s

9 In Pädagogik, Kant uses a third set of terms to mark the product of each of the exclusively human predispositions. He states, ‘Practical education includes 1) skill, 2) worldly prudence, and 3) morality’ (Ped. 9: 486). Upon examination, these turn out to be effectively identical to the technical, pragmatic, and moral predispositions outlined in the Anthropology.

114

David Baumeister

reason and appeals to maxims, human animality must be addressed through physical education, which involves the measured use of discipline [Disziplin] or training [Zucht]. Accurately defining what Kantian discipline is, as well as what it is not, is critical to understanding the place of human animality in his theory of education and in his account of individual human nature. For while Kant’s pedagogy lectures offer no sustained exposition of the concept of animality, Kant’s account of discipline, and of physical education more generally, casts light on human animality in the negative, as it were. The form Kant recommends discipline take reveals the contours of animality in the human being. Early in the lectures, Kant states, ‘Discipline or training changes animal nature into human nature’ (Ped. 9: 441). The formulation is succinct, but on Kant’s own terms must be qualified. Human animality is never eliminated from human nature, even in discipline, and so animal nature can be said to be ‘changed into’ human nature only in the sense that discipline fosters the development of the exclusively human predispositions by carefully tempering human animality. Discipline does indeed change human animality, but only in the sense of preventing it from being developed in certain directions. As Robert Louden remarks: ‘Changing animal into human nature by no means entails eradicating or demolishing instincts, desires, and inclinations—a radical change of this sort would result in beings who were no longer human. Rather, it refers to the ability (an ability which itself is one of humanity’s germs or natural predispositions) to control them through the exercise of reason’ (Louden 2011: 139). Discipline reflects the human’s ability to control and temper its animality through the use of reason. Even though disciplining the animality of a child does not rely upon the child’s own reason for the discipline to be successful (for in discipline the child itself is only passive), the implementation of discipline, which must be a carefully measured act taking into account the full register of human nature, involves reason. Discipline can be said to be reasonable in so far as it aims to prevent, via targeted responses, the formation of bad habits in the individual. Kant takes a harsh view of habit in the pedagogy lectures, stating: ‘The more habits someone has, the less he is free and independent. It is the same with the human being as with all other animals: they always retain a certain propensity for that to which they were accustomed early. The child must therefore be prevented from getting accustomed to anything; it must not be allowed to develop any habits’ (Ped.: 9: 463). Kant provides several examples of the sorts of bad habits he has in mind. One is the habit of expecting others to respond to one’s crying or complaining. Many parents, Kant observes, respond to the cries of their children all too readily, playing with or feeding them on demand. ‘But this,’ Kant explains, ‘only makes them cry more often. If, on the other hand, one does not care about their cries, they finally stop. For no creature enjoys a futile task. If children are accustomed to having all of their whims fulfilled, then afterwards the breaking of the will comes too late. If

Animality in Kant ’ s Theory of Human Nature

115

one lets them cry, they will grow weary of it by themselves’ (Ped.: 9: 460). Another example is that of a child’s learning to walk, which some parents seek to bring about artificially with the aid of ‘leading-strings and go carts’—contrivances that, Kant believes, instil bad habits in the child. For Kant, children ‘do not learn to walk as steadily with this means of help as when they learn it by themselves. It is best to let them crawl about on the ground, until they gradually begin to walk by themselves’ (Ped. 9: 461). The bad habits of expecting others to respond to one’s cries and walking with artificial aids represent counter-purposive developments of the child’s animality. With them, basic expressions of the child’s drive to selfpreservation, which could be satisfied and developed both naturally and reasonably, are developed in ways that, if left unchecked, would prevent the child from properly developing its moral predisposition later on. The danger inherent in bad habits, then, lies not simply in their disrupting the natural development of the child as a living being, but, what is more irksome for Kant, in their hampering the development of the other, exclusively human, predispositions. Hence, Kant declares, ‘To discipline means to seek to prevent animality from doing damage to humanity, both in the individual and in society. Discipline is therefore merely the taming [Bezähmung] of savagery [Wildheit]’ (Ped.: 9: 449). Savagery has a specific meaning in Kant’s educational theory, being simply ‘independence from laws’ (Ped. 9: 442). Discipline amounts to the ‘taming’ of savagery in so far as discipline brings lawfulness to bear on the human’s otherwise lawless animal nature: ‘Through discipline the human being is submitted to the laws of humanity and is first made to feel their constraint’ (Ped. 9: 442). Introducing the ‘laws of humanity’ during a child’s education sets limits or constraints on the development of the child’s animality. Of course, the child’s animality must still be developed, as must each of their natural predispositions. But this development cannot be taken in just any direction, for this would in many cases lead to the instillation of bad habits that might conflict with the child’s other predispositions, thereby ‘damaging its humanity’. Discipline is hence in each case implemented with the aims of the child’s predispositions in mind. Kant specifies, however, that the means of this implementation, while placing constraints on human animality, do not supplement or fundamentally augment the child’s animal nature. The child’s animality, for Kant, should not be treated artificially, for this would violate the integrity of animality as a natural predisposition. Hence, Kant states, ‘In general it should be observed that the first stage of education [physical education or discipline] must be merely negative, i.e., one should not add some new provision to that of nature, but merely leave nature undisturbed’ (Ped. 9: 459). Kant’s view of physical education is quite nuanced. It involves introducing constraints upon human animality—‘taming’ the ‘savagery’ of animality in its lawless state—while also ‘leaving undisturbed’ that which is natural in the predisposition.

116

David Baumeister

An apt illustration of this balance is Kant’s analogy between the proper education of a child and the proper growth of a tree. Referencing the botched education of princes in his time, who he claims generally do not meet with enough resistance in childhood, Kant remarks, ‘But a tree which stands alone in the field grows crooked and spreads its branches wide. By contrast, a tree which stands in the middle of the forest grows straight towards the sun and air above it, because the trees next to it offer opposition’ (Ped. 9: 448). Just as the tree developing within the constraints of the forest grows higher, straighter, and better positioned to benefit from the sun and the air, the child educated within the constraints placed upon it by discipline is able to develop its natural predispositions to a greater and fuller extent. At least in principle, Kant’s approach to the disciplining of human animality eschews the repression or denial of animality in favour of a reasoned and controlled development or preservation of animality in light of the aims of humanity, and the aims of nature that underlie each of the human’s predispositions. Another way of getting at the sense in which, for Kant, human animality is and must be preserved in education is to consider Kant’s view of the place of animal instincts or impulses in human nature. Throughout its childhood, as indeed throughout its adult life, the human being is, for Kant, affected by various instincts or impulses that stem from its animality. Such instincts, depending on the circumstances of an individual’s physical being and environment, engender certain inclinations or habits, and can be more or less virtuous (as when shielding one’s child from sudden danger) or vicious (as when our instinct for self-preservation is followed to the point of gluttony). The human’s animal instincts can indeed provide the vehicle with which the human might run astray from the course set for it by the other predispositions. Unchecked by reason, Kant imagines, these animal instincts can even imperil the human’s life. Discipline thereby keeps the human on track to develop its humanity in light of the fact that some of its animal instincts may lead it in contrary directions. Kant explains: ‘Discipline prevents the human being from deviating by means of his animal impulses from his destiny: humanity. Discipline, for example, must restrain him so that he does not wildly and thoughtlessly put himself in danger’ (Ped. 9: 442). It is plain from this why Kant regards physical education as a sort of maintenance [Verpflegung], which, Kant remarks, ‘the human being has in common with animals’ (Ped. 9: 455). To maintain the human being is to provide it with what is necessary for the continuation and preservation of its physical being. Such maintenance takes the form of discipline when the human must be protected from itself, and thereby prevented from deviating from the natural course whereby its natural predispositions—animality, humanity, and morality alike— will each come to be developed. Though the human’s animal instincts must be met with discipline, it is critical to note that the presence of animal instincts in the human can never be extinguished. Hence, in his lecture course on rational theology from the mid-1780s,

Animality in Kant ’ s Theory of Human Nature

117

Kant remarks that ‘at the same time the human has and must have, if he should continue as a human, many instincts that belong to animality’ (V-Phil-Th/Pölitz 28: 1078, my translation).10 That is, animal instincts in themselves, and the animal nature that they reflect, are not denied or rejected on Kant’s model of education. As Wood explains, ‘Instincts, and the inclinations based directly on them, are in themselves innocent and are capable of being involved in evil only insofar as we incorporate them as incentives into a freely chosen maxim . . . But then it is this choice, and not its instinctive source, that is good or evil’ (Wood  2010: 151). Indeed, not only are animal instincts not a source of evil in themselves, but they can and often do contribute to the proper development of human life. Discipline, and the physical component of Kant’s educational theory more broadly, keeps human animality on a proper course by preventing the human’s animal instincts from engendering bad habits or inclinations (the basis of evil in an individual’s life), which, being mis-directions or mis-extensions of human animality, have their source in the misuse of human reason, not in the nature of human animality itself.

5.4 Focus on the Philosophy of History (Species-Level Development) Kant’s philosophy of history is consistent and continuous with his theory of education. Both are concerned with human nature and the routes though which human nature can be properly or improperly developed. Both are deeply teleological, and have at their centre the Kantian account of original human predispositions. While education fosters the development of these predispositions in an  individual’s life, history tracks their development across the broad intergenerational unfolding of the human species. Though Kant’s account of the purposive development of the exclusively human predispositions within and across history has received relatively robust scholarly attention,11 his view of the historical development of human animality remains virtually unexplored. As discussed above, the development of the human’s predispositions results, for Kant, in the proliferation of goodness in the world. The development of these predispositions in history entails the perfection of humanity, and an increase in the overall well-being of the human species. Education, though an individuallevel enterprise, provides an entry point to the species-level enterprise of the

10 The original passage reads: ‘der Mensch zugleich viele Instinkte, die zur Thierheit gehören, hat und haben muß, wenn er als Mensch fortdauern soll.’ The Philosophische Religionslehre nach Pölitz cannot be dated precisely, but various pieces of evidence suggest they were given between 1783 and 1786. For discussion, see the editor’s introduction to the Cambridge edition of these lectures (there titled ‘Lectures on the Philosophical Doctrine of Religion’), Kant 2001: 337–8. 11 See, for instance, the essays in Rorty and Schmidt (2012).

118

David Baumeister

perfection of humanity, for Kant’s estimation of the potential for education to benefit the entire human species is very high: Perhaps education will get better and better and each generation will move one step closer to the perfection of humanity; for behind education lies the great secret of the perfection of human nature. Henceforth this may happen. Because now for the first time we are beginning to judge rightly and understand clearly what actually belongs to a good education. It is delightful to imagine that human nature will be developed better and better by means of education, and that the latter can be brought into a form appropriate for humanity. This opens to us the prospect of a future happier human species. (Ped. 9: 444)

Kant’s use of the future tense is crucial. Paralleling the famous distinction from his popular 1784 essay ‘An Answer to the Question: What is Enlightenment?’ that the present is not an ‘enlightened age’, but rather an ‘age of enlightenment’, Kant here indicates that the perfection of human nature in history—the secret of which lies behind education—is a perfection to come or a perfection in the making, but not a perfection that humans presently enjoy (WE 8: 40). A closer look at the role of animality in Kant’s account of human history helps explain why the perfection of the species must be delayed. For just as the conflict between the individual human’s animal and moral natures presents a major challenge for education, historical progress is complicated by the ineradicable presence of animality in the nature of the human species; yet crucially, as at the individual level, the resulting obstruction to proper historical development does not stem from human animality itself, but from reason’s difficulty in appropriately constraining, and reconciling itself with, the human’s animal nature. In the 1784 essay ‘Idea for a Universal History with a Cosmopolitan Aim’, Kant contrasts the development of natural predispositions in human individuals, which tend to be ‘confused and irregular’, with the development of the same in the human species, which ‘can be recognized as a steadily progressing though slow development’ (IUH 8: 17). The first of nine propositions Kant advances in the essay states that ‘All natural predispositions of a creature are determined sometime to develop themselves completely and purposively’, which, given the typology of predispositions Kant lays out in Religion, indicates that the development of human animality, being the development of a natural predisposition of the human creature, is not excluded from the steady unfolding of history (IUH 8: 18). Indeed, in so far as Kant in Religion identifies one of three aspects of the predisposition to animality in human nature to be the source of human sexuality, it is plain how history—an inter-generational enterprise—requires the development of human animality. As Susan Meld Shell puts it, for Kant, the goal of human history ‘is the simultaneous realization of our animal and human perfections, the two natural ends whose historical and sexual divergence keeps human

Animality in Kant ’ s Theory of Human Nature

119

development in motion’ (Shell  2003: 213). Animality, by engendering human sexuality and ensuring the continuation of the species across generations, is central to human historical development, for without it the movement of this development would come to a halt. At the same time, animality’s centrality to human history is continually displaced by human reason, which Kant takes to be categorically distinct from the instinctual legacy of the human’s animal nature. The third proposition of Kant’s essay reads, ‘Nature has willed that the human being should produce everything that goes beyond the mechanical arrangement of his animal existence entirely out of himself, and participate in no other happiness or perfection than that which he has procured for himself free from instinct through his own reason’ (IUH 8: 19). To say that the human should, as a matter of natural will, only participate in that happiness or perfection that comes about through the use of human reason indicates that animality, no matter how central it may be, can never yield human perfection on its own. Even though human animality aims at the good, it alone is an insufficient means of achieving this good; the human must in some sense ‘go beyond’ its ‘animal existence’ in order to achieve this. However, the inherent difficulty of this call to self-sufficiency in reason—and what makes the progress of human history so slow, albeit steady—is that the human’s animal existence can never fully be ‘got beyond’, even though this is the direction in which, for Kant, human reason tends. Human history, on Kant’s view, is a project of species-level self-overcoming—of human reason, in history, overcoming human animality. But this project is vexed, even paradoxical, in structure. For animality, which is to be overcome, at the same time remains necessary for (all-too-human) history to move forward. In the Kantian picture, the human’s predisposition to animality threatens and troubles human history from within. More concretely, human animality undermines the establishment of a proper civil society. The fifth proposition states, ‘The greatest problem for the human species, to which nature compels him, is the achievement of a civil society universally administering right’ (IUH 8: 22). Achieving this would represent the full development of the properly human predispositions, and would bring the promise of human history to fruition. The supreme value of the achievement is matched by its supreme difficulty and unattainability, and, for Kant, human animality lies at the root of the problem. ‘The difficulty which the mere idea of this problem lays before our eyes is this,’ Kant explains in the sixth proposition, ‘the human being is an animal which, when it lives among others of its species, has need of a master’ (IUH 8: 23). This need arises, Kant continues, because the human’s ‘selfish animal inclination still misleads him into excepting himself from [the law of civil society] where he may’ (IUH 8: 23). That is, the human is the sort of animal whose animality puts it in the position of needing to be led and ruled by some force above it, which ensures that the human does not ‘selfishly’ obey its animality alone, but

120

David Baumeister

rather obeys those laws that form the basis of civil society. This need encounters a regressive dilemma, however, when humans actually begin to look for the master they need. Kant goes on: ‘But where will he get this master? Nowhere else but from the human species. But then this master is exactly as much an animal who has need of a master’ (IUH 8: 23). Not only does human animality introduce the difficulty with achieving the aim of a perfect civil society in history—it also sustains the difficulty into perpetuity. Any progress the human species may make towards this goal must therefore be slow, for animality presents complications at all levels of human social organization. Kant’s perspective on human animality in the 1784 essay on history appears more scathing than the accounts found in either Religion or the pedagogy lectures. Even so, Kant’s frustration with the inhibition to historical progress presented by human animality never leads him to doubt or deny the significance of animality to the nature of the human species. Despite the difficulties it engenders, animality does and must remain present in human history, for not only does it provide the impetus for the movement of this history through sexual reproduction, it is also a basic and inescapable facet of human nature. As such, it forms part of the natural basis of human history, and even of the establishment of civil society, which in other respects is a human-made or artificial endeavour. As Kant remarks in the Anthropology, ‘In a civil constitution, which is the highest degree of artificial improvement of the human species’ good predisposition to the final end of its destiny, animality still manifests itself earlier and, at bottom, more powerfully than pure humanity’ (Anth. 7: 327). Though he is discouraged by the challenges entailed by its tension with human reason and human morality, Kant never denies the power or priority of human animality in the human’s specieslevel development. Whereas in his theory of education, Kant had elaborated the means by which the individual’s animality could be met with discipline, he is less specific about precisely how the animality of the species might be properly reconciled with human reason. It is, however, clear that here, as in his picture of individual development, Kant does not place blame on animality itself. It is rather the conflict between animality and reason (the wellspring of morality) that is to blame. Kant is explicit about this in the ‘Conjectural Beginning of Human History’. Imagining the emergence of the human being and of human history from an otherwise entirely non-human, non-rational natural background, Kant suggests that ‘Before reason awoke, there was neither command nor prohibition and hence no transgression; but when reason began its business and, weak as it is, got into a scuffle [Gemenge] with animality in its whole strength, then there had to arise ills and, what is worse, with more cultivated reason, vices, which were entirely alien to the condition of ignorance and hence of innocence’ (Con. 8: 115). The double-sided character of human animality plays out in the development of the species just as it does in that of the individual. Though the persistence and necessity of animality

Animality in Kant ’ s Theory of Human Nature

121

in human history hampers the project of achieving the ends of history, the ills and vices symptomatic of this project’s failure are not the fault of animality, which is itself something innocent. Günter Zöller coins the phrase ‘rationally enhanced animality’ to describe how, within what Zöller calls Kant’s ‘political anthropology’, the animality of the human species is actively reconciled with reason through the unfolding of human history along an ultimately moral and political trajectory.12 Zöller writes: But unlike in his moral philosophy, Kant’s concern in his anthropology is not with the norms and form of . . . moral willing and acting but with the arduous path of the human species through rationally enhanced animality toward morally conditioned rationality. That path leads not through the heart of the individual moral agent but through history and employs not the inner constraint of conscience, moral respect, or moral feeling but the outer means of socially organized constraint. Accordingly, the domain of socially mediated and historically manifest human self-perfection is the political. (Zöller 2011: 156)

On Zöller’s reading, Kant’s ‘political anthropology’ involves the idea that, through human history, human nature can be perfected, not in spite of the challenges posed by human animality, but through a certain ‘rational enhancement’ of the human’s animal nature. This enhancement makes possible the interminable process of ‘human self-governance’ referred to in Kant’s account of the human’s everpresent need of a master (Zöller  2011: 156). The phrase ‘rationally enhanced animality’ nicely summarizes, perhaps more succinctly than any single idiom from Kant’s own text, the spirit and substance of Kant’s commitments regarding the centrality of animality to the progress of human history, and the development of human nature more generally. For Kant is sometimes guilty of drawing more attention to what, from one angle, may be seen as the ‘misfortune’ of the human species in being subject to a set of predispositions that often come into conflict with one another, rather than to what, from a different angle, can just as well be seen as the characteristic form of life that emerges out of the nest of tensions internal to human nature.

5.5 Conclusion In recent years, a thriving scholarly literature has emerged around Kant’s positions on the metaphysical, cognitive, and moral status of non-human animals. Largely neglected by this literature, however, is Kant’s robust, if fragmented, 12 For a recent, system-spanning treatment of the synergy between Kant’s anthropology and political theory, see Huseyinzadegan (2019).

122

David Baumeister

account of the animality that human and non-human animals have in common. As this essay has endeavoured to show, in his treatment of animality as a component of human nature in the Religion, and in his elaboration of this conception in his philosophies of education and history, Kant frames the relation between humans and other animals at a level below or prior to categorical differences. Far from denying the animality of human beings, Kant recognizes the necessary and co-constitutive part played by animality within the developmental trajectories of human nature. While Kant’s account remains, from a contemporary perspective, problematic in multiple ways, it nonetheless offers distinctive, and uniquely Kantian, resources for rethinking the relationship of humans to other animals, and of human beings to themselves.

6

Kant on Evolution: A Re-evaluation Alix Cohen

6.1 Introduction Kant’s notorious remark about the impossibility of there ever being a Newton of a  blade of grass has often been interpreted as a misguided pre-emptive strike against Darwin and evolutionary theories in general:1 It would be absurd for humans even to make such an attempt or to hope that there may yet arise a Newton who could make comprehensible even the generation of a blade of grass according to natural laws that no intention has ordered; rather, we must absolutely deny this insight to human beings. (CPJ 5:400)

My aim in this chapter is to re-evaluate this claim in the context of Kant’s account of organic generation and argue that, contrary to what is usually thought, it does leave room for the possibility of evolution. After spelling out the constraints Kant identifies for any scientific study of the history of living beings, I examine his theory of generation and draw its implications for biological heredity, species diversity, and the role played by environmental factors in organic development. On this basis, I conclude that first, evolution is a possible albeit far-fetched hypothesis for Kant; second, Darwin’s theory of natural selection would have turned a far-fetched possibility into a plausible candidate; and third, it would not have disposed of the need for teleology. This is why, I argue, Darwin could never have been a Newton of a blade of grass.

6.2 The distinction between history of nature, system of nature, and natural history Kant begins his account of the scientific study of the natural world and its changes by noting the necessity of dissipating the equivocation of the notion of ‘history of

1 See, for instance, Lovejoy (1968), Cornell (1986), Ruse (2006), and O’Shea (2012). Alix Cohen, Kant on Evolution: A Re-evaluation In: Kant and Animals. Edited by: John J. Callanan and Lucy Allais, Oxford University Press (2020). © Alix Cohen. DOI: 10.1093/oso/9780198859918.003.0007

124

Alix Cohen

nature’.2 To this effect, he distinguishes between different kinds of histories, depending on their object of study and their methodology. Some are truly historical in the sense that they approach the natural world diachronically (e.g. natural history (Naturgeschichte) and the system of nature (Systeme der Natur)); others approach it synchronically, by examining the state of the natural world at a given point (e.g. the description of nature (Naturbeschreibung)). While these scientific projects differ significantly, they are not only intrinsically related but depend upon each other for their possibility. For there is a sense in which the description of nature is the basis of any diachronic study of it: [N]atural history would only consist in tracing back, as far as the analogy permits, the connection between certain present-day conditions of the things in nature and their causes in earlier times according to laws of efficient causality, which we do not make up but derive from the powers of nature as it presents itself to us now. (TP 8:161–2)

It is only from the current state of the natural world, which we can experience, that we can hope to discover its previous states and the causal connections at work between those states, which have made its current state possible. The description of nature, the least ambitious project perhaps, grounds them all in the sense that it is their epistemic condition of possibility: [W]hich came first, history or geography? The latter is a prerequisite for the former, because events necessarily take place with reference to something. History is a continuous progression, but things, too, change, and give an entirely different geography at particular times. Geography is thus the foundation [of history]. (PG 9:163)

Despite the epistemic contribution of geography, Kant underlines the limitations of historical studies of nature. As such, they have not reached the status of a proper science: ‘while one (the description of nature) appears as a science with all the splendour of a great system, the other (natural history) can only point to fragments or shaky hypotheses’ (TP 8:162). It is difficult to know with certainty nature’s history in remote times: ‘it has to be guessed, more through experiments than by accurate testimony’ (PG 9:162). One type of ‘guesswork’ is the analogical study of the shapes of natural beings. Through comparative anatomy, it aims to group them into various classes, obtained by trial, and thereby form a system of nature. On the basis of the resemblance between different kinds of natural objects, it organizes them into a logical system by recording ‘[t]he agreement of so many 2 ‘I demand nothing new thereby but merely the careful separation of one business from the other, since they are entirely heterogeneous’ (TP 8:162).

Kant on Evolution: A Re-evaluation

125

genera of animals in a certain common schema, which seems to lie at the basis not only of their skeletal structure but also of the arrangement of their other parts’ (CPJ 5:418). However, insofar as this classification is based on the mere ‘similarity of form’ between natural beings (PG 9:160), it has the appearance of a system but it does not really unify its parts within a common natural whole. Yet it is only by showing the real filiation between natural beings that we can bring unity to their apparent diversity and thereby hope to ground a true historical study of nature: The school division concerns classes, which divide the animals according to resemblances, the natural division concerns phyla, which divide the animals according to relationships in terms of generation. The former provides a school system for memory; the latter provides a natural system for the understanding. The first only aims at bringing creatures under titles; the second aims at bringing them under laws. (ODR 2:429)

The classes of a school division such as Linnaeus’ do not track a real continuity between living beings belonging to the same class. They do not tell us anything more about the nature of things than logical continuity, that is to say the conceptual relations of similarity and difference.3 Such a system of nature does not inform us about the real possibility of things, only their conceivability. There is in this sense an artificial character to the system of nature that the history of nature does not share: Natural history, which we still lack almost entirely, would teach us about the changes in the shape of the earth, likewise that of its creatures (plants and animals) that they have undergone through natural migrations and the resultant subspecies from the prototype of the phyletic species. It would presumably trace a great many of seemingly different kinds to races of the same species and would transform the school system of the description of nature, which is now so extensive, into a physical system for the understanding. (ODR 2:434)

The difference between natural history and school system of nature is a difference between the conceptual and the real, a conceived unity and a natural unity. What is at stake is thus a different sense of ‘connection’ between living beings. One is logical: it is based on the principle of resemblance. The other is natural: it is based on filiation, which, in the case of living beings, is defined by the common law of natural reproduction:

3 ‘Division of knowledge according to concepts is logical [ . . .] we obtain a system of nature (systema naturae), as for example that of Linnaeus’ (PG 9:159–60).

126

Alix Cohen

The natural division into species and kinds in the animal kingdom is grounded on the common law of propagation, and the unity of the species is nothing other than the unity of the generative power that is universally valid for a certain manifoldness of animals. For this reason, Buffon’s rule, that animals which produce fertile young with one another (whatever difference in shape there may be) still belong to one and the same physical species must properly be regarded only as the definition of a natural species of animals in general in contrast to all school species of the latter. (ODR 2:429)

By defining species in terms of the capacity to reproduce and have fertile offspring, Kant can account for organic unity within a species. Species-groupings thereby acquire a historical reality that logical classes lack. However, to make sense of this notion of species, we need to turn to Kant’s account of organic generation. For to understand the possibility of a biological line of descent, he needs to explain organic development in a way that secures the inheritance of biological characteristics within the same species.

6.3 Kant’s epigenetic account of natural predispositions On my reading, Kant’s position on organic generation is a middle ground between preformation and epigenesis, by which I mean that it has both an epigenetic and a preformationist component.4 The epigenetic dimension of Kant’s account consists in, negatively, leaving aside the question of life’s beginnings, and limiting itself to the claim that an organism can only be conceived as the product of another organism: it ‘begins all physical explanation of these formations with organized matter’ (CPJ 5:424) and does not try to account for the possibility of an original form of organization.5 Positively, it characterizes nature as something that is productive and has a teleological element (Bildungstrieb, ‘formative impulse’). In this sense, the decisive contribution of epigenesis to the debates on organic generation is to acknowledge a primitive organization and, accordingly, subordinate mechanical principles to teleological principles: ‘our judging of them must always be subordinated to a teleological principle as well’ (CPJ 5:417). Kant’s endorsement of epigenesis is supplemented with a strong preformationist component, as appears most clearly in this passage: 4 Kant’s position is what I have called an ‘epigenesis of natural predispositions’ (Cohen (2006)). It is unique since, as Sloan notes, most forms of preformationist theories were opposed to the thesis of epigenesis (Sloan (2002): 233). For an account of the scientific context of the time and the debates on generation in particular, see Lenoir (1982), Sloan (2002), Zammito (2002), and Lagier (2004). 5 As Kant notes, Blumenbach ‘rightly declares it to be contrary to reason that raw matter should originally have formed itself in accordance with mechanical laws’ (CPJ 5:424). Kant’s relationship with Blumenbach has been the subject of numerous divergent interpretations. See, for instance, Lenoir (1982), Sloan (2002, Section 3), and Zammito (2003).

Kant on Evolution: A Re-evaluation

127

[T]he productive capacity of the progenitor is still preformed in accordance with the internally purposive predispositions that were imparted to its stock, and thus the specific form was preformed virtualiter. (CPJ 5:423)

Kant restricts epigenesis by positing the existence of natural predispositions that are dynamic and purposive ordering principles inherent in the species’ original stock. By playing the role of structures that limit the evolution of the species, they account for the fact that it cannot transform and its characteristics are predetermined:6 I myself derive all organization from organic beings (through generation) and all later forms (of this kind of natural things) from laws of the gradual development of original predispositions [ursprünglichen Anlagen], which were to be found in the organization of its phylum. (TP 8:179)

There are structuring powers, acting upon specific pre-existent germs (Anlagen), that underlie all developments in a given species: an organized being ‘incorporate[s] nothing into its generative power that does not belong to one of the undeveloped original predispositions of such a system of ends’ (CPJ 5:420).7 In other words, crucially, these predispositions predetermine the development of natural species by providing intrinsic purposive structures that guide their organic development. Thus, at variance with purely epigenetic theories of generation, Kant’s theory entails that biological inheritance can only be caused by natural predispositions present in the original stock: ‘outer things can well be occasioning causes but not producing ones of what is inherited necessarily and regenerates’ (ODR 2:435).8 The seeds of all organic changes had to be present from the start since their transmission is the only way of securing the inheritance of species-characteristics. As a result, Kant’s theory of the epigenesis of natural predispositions provides the biological ground for his account of the development of natural species by uniting the diversity of living beings organically and thus historically. Without a  preformationist component that allows for natural predispositions to be 6 See also Sloan (2002: 246–7) and Zammito (2003: 88–93). 7 Kant offers a distinction between germs [Keime] and predispositions [Anlagen], although it is not particularly compelling: ‘The grounds of a determinate unfolding which are lying in the nature of an organic body (plant or animal) are called germs, if this unfolding concerns particular parts; if, however, it concerns only the size or the relation of the parts to one another, then I call them natural predispositions’ (ODR 2:434). In this paper, I treat them as equivalent since the distinction spelt out here is not directly relevant to my argument. 8 See also ‘heredity, even only the contingent one, which does not always succeed, [can never] be the effect of another cause than that of the germs and predispositions lying in the species itself ’ (Det. 8:97). As Lovejoy notes, Kant is ‘a vigorous opponent of the supposition that acquired characters can be inherited and an unqualified partisan of the doctrine of the continuity and unmodifiability of the germ-plasm’ (Lovejoy, 1968: 183).

128

Alix Cohen

developed and then transmitted, permanent lineages within the various species cannot be secured. Yet without an epigenetic component that allows some seeds rather than others to be actualized, the diversity within these species cannot be accounted for. However, there are different kinds of biological diversity within the same species, for some traits are invariably inherited and others are not. Thus, within each species, two types of differentiation can be made: one according to which the members of a species who possess characters that are invariably inherited belong to the same race; and another according to which the members of a species who possess characters that are only partially hereditary constitute varieties. Heredity is thus the key to the distinction between different kinds of biological groups, which Kant calls race, strain, variety, and sort (see Table 6.1): Among the subspecies, i.e. the hereditary differences of the animals which belong to a single phylum, those which persistently preserve themselves in all transplantings (transpositions to other regions) over prolonged generations among themselves and which also always beget half-breed young in the mixing with other variations of the same phylum are called races. Those which persistently preserve the distinctive character of their variation in all transplantings and thus regenerate, but do not necessarily beget half-breeds in the mixing with others are called strains. Those which regenerate often but not persistently are called varieties. Conversely, that variation which produces with others halfbreeds but which extinguishes gradually through transplantings is called a special sort. (Det. 2:430)

On my account, it is because of the preformationist component of Kant’s theory of generation that there must have been an original single stock containing the seeds of all races within the same species. This stock, in turn, guarantees that all races belong to the same species since the biological lineage is preserved by inheritance. On the other hand, it is because of the epigenetic component of his theory of generation that some seeds rather than others can be actualized depending on the environment, which explains racial and varietal diversity within the same species. Through the conjunction of species monogenesis and racial epigenesis, Table 6.1 Different kinds of biological groupings Biological groupings

Persistence of characters through transplanting

Persistence of characters through breeding

Race Strain Variety Sort

unfailingly hereditary persistent characters partially hereditary characters gradual extinction of characters

unfailing half-breed contingent half-breed contingent mix of breed no half-breed

Kant on Evolution: A Re-evaluation

129

Kant’s account of the history of living beings consists of generations of species confronting various kinds of environment and slowly producing a diversity of races from a unique strain in which an infinity of germs was originally located. Only if one were to describe the events of the whole of nature as it has been through all time, then and only then would one write a real so-called natural history. If, for example, one were to consider how the various breeds of dogs descended from one line, and what changes have befallen them through all time as a result of differences in country, climate, reproduction, etc., then this would constitute a natural history of dogs. Such a history could be compiled for every single part of nature, for instance, on plants and so forth. (PG 9:162)

To have a better understanding of Kant’s account of the relationship between biological heredity and environmental conditions, let’s examine the case of human races, which Kant defines as the result of the adaptation of the human species to various environments through the development of different skin colours.9

6.4 Heredity, diversity, and environmental conditions: the case of skin-colour In line with Kant’s preformationism, the seeds of all human races were present from the start in the species’ original stock. And in line with his epigeneticism, the appropriate seeds were first actualized to serve a purpose that arose from environmental circumstances, and then transmitted to the offspring. Insofar as skin colour appears to be the only character that is invariably dependent on the two parents, Kant identifies it as the biological criterion for distinguishing between races. As he notes, traces of the colours of a ‘Negro’ and a ‘white’ who breed both unfailingly appear in the offspring, whereas the complexions of a brunette and a blonde who breed do not.10 On this basis, his classification distinguishes between ‘the whites, the yellow Indians, the Negroes, and the copper-red Americans’ (Det. 8:93).11 However, the choice of skin colour as the criterion for 9 As is now well known, Kant’s discussions of human races are full of stereotyping, prejudice, and bigotry. In this section, I am solely concerned with their biological definition. For a nuanced and insightful analysis of the relationship between his views on race and the rest of his philosophy, see Frierson (2013: 104–7). See also Eze (1995) and Larrimore (1999). 10 ‘[B]londes and brunettes are not different races of whites, because a blond man can have entirely blond children with a brunette woman, even though each of these subspecies is preserved throughout extended generations in all transplantings. For this reason, they are strains of whites’ (ODR 2:430]). 11 As Kant writes, ‘I think one is only compelled to assume four races of the human species in order to be able to derive from these all the easily distinguishable and self-perpetuating differences. They are 1) the race of the whites, 2) the Negro race, 3) the Hunnish (Mongolian or Kalmuckian) race, 4) the Hindu or Hundustani race’ (ODR 2:432]). For an account of the incongruity between Kant’s two definitions of the four races, see Zammito (2006: 41–3).

130

Alix Cohen

distinguishing between races is only partly based on the fact that it appears to be the only character that is invariably inherited from both parents. It is also based on what Kant believes to be its essential feature, namely its purposiveness: The reason why this character [skin colour] is an appropriate basis for a class distinction [ . . . ] is that the expulsion of wastes by means of sweating is the most important bit of concern exercised by nature insofar as the creature—which is affected quite differently by exposure to all sorts of different climates—is supposed to be preserved with the least amount of recourse to artificial means. (PG 8:93)

Skin colour fulfils a crucial function in the survival of the human species: it allows its adaptation to different climates and different environmental conditions by regulating the constitution of the blood and allowing the expulsion of waste through sweating.12 In this sense, skin colours are purposive: they are pre-adapted to a specific type of climatic environment, and they are all present, latently, as seeds, in the original human stock. [T]hose of their descendants [of the first human couple] in which the entire original predisposition for all future subspecies was still unseparated were fit for all climates (in potentia), such that the germ that would make them suitable to the region of the earth in which they or their early descendants were to find themselves could develop in that place. (TP 8:173)

It is not that the human species has to adjust itself to fit the environment in which it lives; rather, the human species can adjust to different environments because it possesses a variety of seeds pre-adapted to all environments. Certain seeds contained in the original stock are actualized in accordance with the requirements of the environment human beings find themselves in, and the actualization of these seeds is precisely what constitutes a race. The racial characters thereby produced are then transmitted to the offspring, thus guaranteeing permanent racial lineages.

12 This appears most clearly in Kant’s speculations about the physical basis of blackness, where he appeals to iron particles in 1777 and to phlogiston in 1785. For instance, ‘Now with respect to the peculiarity of a race, this purposive character can be demonstrated nowhere so clearly as in the Negro race; yet the example taken from the latter alone also entitles us at least to conjecture the same of the remaining ones, according to the analogy. For one knows now that the human blood becomes black (as can be seen at the underside of a blood cake) merely by being overloaded with phlogiston. Now already the strong odor of the Negroes, which cannot be helped through any cleanliness, gives cause for conjecturing that their skin removes much phlogiston from the blood and that nature must have organized this skin so that the blood could dephlogistize itself in them through the skin in a far greater measure than happens in us, where that is for the most part the task of the lungs’ (Det. 8:103); see also ODR 2:440.

Kant on Evolution: A Re-evaluation

131

The human being was destined for all climates and for every soil; consequently, various germs and natural predispositions had to lie ready in him to be on occasion either unfolded or restrained, so that he would become suited to his place in the world and over the course of the generations would appear to be as it were native to and made for that place. (ODR, 2:435)13

The human species adapts differently to different conditions, and the best adaptation for each type of environment survives and becomes dominant, thereby giving the impression that the race was designed for this place. Of course, it is not strictly speaking ‘native’ to this place since it is the result of an evolution of the species. But the appearance of design isn’t just an impression on Kant’s account since the fit between a particular race and its environment is due to the actualization of an original predisposition whose purpose was the survival of the species in these types of conditions.14 It is in this sense that Kant places teleology at the centre of his biological account of species development. Now, as is now well-known, Kant is famously ambivalent with regards to the role of teleology in science.15 Without getting into the detail of his account here, I believe that two claims are relevant to my discussion.16 On the one hand, he believes that we should always think of organisms as being mechanically possible and go as far as possible in our mechanical explanation of them: ‘It is thus rational, indeed meritorious, to pursue the mechanism of nature, for the sake of an explanation of the products of nature, as far as can plausibly be done’ (CPJ 5:418). Kant is and remains committed to the claim that all objects of experience can be accounted for in mechanical terms.17 On the other hand, he notes that we cannot help but rely on teleology as a principle for reflective judgment when confronting organisms: ‘it is merely a consequence of the particular constitution of our understanding that we represent products of nature as possible only in accordance with another kind of causality than that of the natural laws of matter’ (CPJ 5:408). Due to their reflective nature, teleological judgments are hypothetical modes of  explanation that cannot attain the level of objectivity required by natural science.18 But in spite of that, we cannot do without them. Organisms have a ‘special character’ (CPJ 5:369): they stand out in the natural world in virtue of their 13 See also Det. 8:98–9. 14 Beyond the case of human skin colour, Kant’s model of organic generation is intended to account for the diversity of characteristics within the same animal species in different countries: ‘what is peculiar to each country, e.g. the animals. The local ones need not be noted, however, unless they are different in other places. Thus the nightingales do not sing nearly so loudly in Italy as they do in northern regions. On desert islands, dogs do not bark at all’ (PG 9:164–5). I focus on the human case since it is the clearest discussion of it in Kant’s works. 15 Contrast, for instance, McLaughlin (1990) with Quarfood (2004), and Breitenbach (2008). 16 I have defended these claims in Cohen (2004) and (forthcoming). 17 Teufel (2013) is particularly enlightening on this point. 18 As Kant notes, ‘positing ends of nature in its products provides no information at all about the origination and the inner possibility of these forms, although it is that with which theoretical natural science is properly concerned’ (CPJ 5:417).

132

Alix Cohen

self-organization, and we cannot grasp this unique feature without using the concept of purposiveness.19 Thus, when talking about organisms qua self-organized beings, we cannot do without teleology, and this applies to talking about their evolution as well their functioning. As a result, no matter how far we are able to go back in our mechanical explanations of living beings and their natural history, we cannot do away with teleology.

6.5 Different kinds of evolutionary impossibilities On the basis of Kant’s account of evolution as I have interpreted it, I believe that we should distinguish between different kinds of evolutionary impossibilities for Kant (see Table 6.2). Accounting for ‘the first origin of the plants and animals’ is a ‘science for gods’ (TP 8:162). Whatever this origin consists in, human beings cannot possibly know it. This epistemic closure differs from the absurd hypothesis that explains ‘the generation of an organized being through the mechanism of crude, unorganized matter’—what Kant calls ‘generatio equivoca’ (CPJ:419). What is not absurd, however, although it is often misunderstood as such, is the idea that all creatures originally display less purposive forms that evolve to fit better with their environment. Kant actually changed his mind on this point, since he originally deemed these evolutionary theories ‘so monstrous that reason recoils before them’ (Rev. Herder 8:54, 1785). The thought that ‘one species would have arisen from the other and all from a single original species or perhaps from a single procreative maternal womb’ (Rev. Herder 8:54) eventually becomes a mere ‘daring adventure of reason’ in 1790 (CPJ 5:419, 1790).20 This shift is crucial for

Table 6.2 Different kinds of evolutionary impossibilities Evolutionary hypothesis

Epistemic status

First origin of the organic From inert matter to life (generatio equivoca) Common biological ancestry (generatio heteronyma)

Science for gods Absurd hypothesis Monstrous (1785) / Daring adventure of reason (1792)

19 See, for instance, Kant’s claim that organisms ‘first provide objective reality for the concept of an end that is not a practical end but an end of nature, and thereby provide natural science with the basis for a teleology’ (CPJ 5:376). My treatment of this issue is of course far too brief, but I have defended this view in more detail in Cohen (forthcoming). 20 As Lovejoy noted, Kant ‘no longer condemns transformism on a priori grounds’ (Lovejoy, 1968: 203–4). See also Huneman (2006: 15–6) for a description of Kant’s shift.

Kant on Evolution: A Re-evaluation

133

the purpose of my argument, since the evolutionary ‘adventure’ of ‘generatio heteronyma’ is now a possibility for Kant: . . . [I]nsofar as something organic would be generated out of something else that is also organic, even though there would be a specific difference between these kinds of beings, e.g., as when certain aquatic animals are gradually transformed into amphibians and these, after some generations, into land animals. A priori, in the judgment of mere reason, there is no contradiction in this. (CPJ 5:419 - note)

Thus, it is not contrary to reason to think that some species could transform into other species.21 Instead of adopting a transformationist hypothesis however, Kant endorses what he calls ‘generatio homonyma’, i.e. the theory that species only produce organisms of the same kind.22 If I am correct that generatio heteronyma, the hypothesis of species transformation, is compatible with his general account of organic generation, what stops Kant from adopting it? I believe that it is the lack of evidence of its occurrence. For as he writes, ‘generatio heteronyma, so far as our experiential knowledge of nature goes, is nowhere to be found . . . experience gives no example of it’ (CPJ 5:420). It is on this basis that it remains a farfetched albeit possible hypothesis rather than a conceptual monstrosity for Kant. As he notes, what experience does give numerous examples of, however, is the production of an organism by another organism of the same species (rather than either from another species, as per generatio heteronyma, or from inert matter, as per generatio equivoca). Thus the conjunction of generatio univoca and generatio homonyma seems more plausible to him. But as I will show in the next section, the empirical evidence that there have been different varieties of the same species that have evolved into different species (e.g. extinct species and varieties as descendants of these species, as shown by the evidence that supports Darwin’s theory of natural selection) would have turned a far-fetched possibility into a plausible candidate. To support this claim, I will confront Kant’s account with the basic tenets of Darwinian evolution to determine whether a Kantian account of evolution is possible. I will focus on the two issues that seem to me essential: adaptation, and natural selection.23 21 Recall that, in Section 6.5, I have shown in what sense Kant’s original account in 1775 (Of the Different Races of Human Beings) and 1785 (Determination of the Concept of a Human Race) can be interpreted as circumscribing the supposed contradiction intrinsic to the idea of generation heteronyma. 22 ‘all generation that we know is generatio homonyma [ . . . ] and produces a product that is in its organization itself homogeneous with that which has generated it’ (CPJ: 5:420). 23 Other issues, such as the inheritance of acquired characteristics or the formation of new species, are not directly relevant to my discussion. While they are important for the consistency of Darwin’s account, they are less so in the context of its contrast with Kant’s.

134

Alix Cohen

6.6 Kant vs. Darwin Adaptation: Preformation vs. natural selection. Where Kant and Darwin seem to differ the most is that Kant believes that adaptation cannot be a result of chance: Chance or the universal mechanical laws could not produce such agreements [between racial characteristics and their natural environment]. Therefore we must consider such occasional unfoldings as preformed. (ODR 2:435)

As shown in Section  3, due to his commitment to preformation, Kant cannot comprehend what mechanical process could lead to a fit between a particular race and its environment. On his account, the appearance of design isn’t just an impression because it cannot be explained away; it can only be explained by the actualization of an original predisposition whose very purpose was the adaptation of the species to its environment (i.e. by introducing a teleological element within the species). But first, as already hinted at, the subspecies’ adaptation is itself the result of an evolution triggered by the specific features of the environment. Thus the end result of this evolution, the race, is not native to this particular place. Second, and more importantly for my purposes, Kant’s account does not commit him to the claim that the fit cannot be accounted for in any other, non-teleological way. What he is missing is a mechanical principle that can explain the adaptation of the organism to its environment. Without it, he cannot dispose of the need for preformation in the form of natural predispositions. But crucially, his account is compatible with natural selection à la Darwin. For, as shown in Section 3, although Kant may not be able to explain how, or even that, it is possible, he is committed to the possibility of a mechanical explanation of nature in general, and of organic generation in particular; a possibility that Darwin fulfils with his account of adaptation as the result of natural selection. Natural selection as a teleological explanation. On my interpretation of Kant’s account, even Darwinian natural selection would not dispose of the need for teleology. For, when Kant talks about organisms’ ‘purposive predisposition to selfpreservation’ (CPJ 5:420), he means that the actualization of a biological trait is explained by the fact that it favours the survival of the organism and thus the species. What this suggests is that the very concept of survival defines organisms and their evolution in teleological terms, as natural purposes. Thus, insofar as our explanations of nature cannot but rely on teleology in this way, even if this reliance is reflective or heuristic rather than constitutive, they cannot be purely mechanical. If this claim is correct, crucially, it also applies to Darwin’s account, since the very form of explanation provided by natural selection is teleological in this way: it accounts for the selection of a trait by the overall function it fulfils in

Kant on Evolution: A Re-evaluation

135

the survival of the species.24 In this sense, even Darwinian natural selection understood as a mechanical explanation of the possibility of evolution does not dispose of the need for teleology. Of course, many believe that Darwin is the Newton of a blade of grass, for he makes comprehensible the evolution of species according to natural laws that no intention has ordered.25 However, by contrast with this widely shared view, I believe that strictly speaking, Darwin isn’t the Newton of evolution, and that, moreover, there can never be one on Kant’s account.26 For an authentic Newton of evolution would in effect be a physicist rather than a biologist. She would account for evolution at a molecular level, by providing entirely mechanical explanations of organisms and their evolution qua bundles of cells, just as she would of any other physical object. In this sense, she would collapse the distinction between organic and non-organic nature and thereby dissolve biology into a branch of physics. This is why strictly speaking: It would be absurd for humans even to make such an attempt or to hope that there may yet arise a Newton who could make comprehensible even the generation of a blade of grass according to natural laws that no intention has ordered; rather, we must absolutely deny this insight to human beings. (CPJ 5:400)

However, while there can never be a Newton of a blade of grass for Kant, there can be archaeologists of nature (what we could call natural historians or modern evolutionary biologists) who study the evolution of organisms qua living beings by adopting teleology as an epistemic principle, albeit a heuristic one. On my account, Darwin is one of them.27 For no matter how far his account of natural selection allows him to go back: ultimately he [the archaeologist of nature] must attribute to this universal mother an organization purposively aimed at all these creatures, for otherwise the possibility of the purposive form of the products of the animal and vegetable kingdoms cannot be conceived at all. (CPJ 5:419–20)

Therefore, although evolution is a possible albeit far-fetched hypothesis for Kant, I have argued that first, Darwin’s theory of natural selection would have turned it into a plausible candidate, and second, it would not have disposed of the need for teleology. This is why, on my interpretation of Kant, Darwin could never have been a Newton of a blade of grass. 24 See, for instance, Lennox: ‘Selection explanations are inherently teleological, in the sense that a value consequence (Darwin most often sues the term “advantage”) of a trait explains its increase, or presence, in a population’ (Lennox, 1993: 410). 25 See, for instance, Shanahan (2004: 283–4). 26 Contrast with Schuster (2011), esp. 9. 27 For a slightly different take on this claim, see Cornell (1986: 408–9).

PART II

PR AC T IC A L PHILO S OPHY

7

Directionality and Virtuous Ends Arthur Ripstein and Sergio Tenenbaum

7.1 Introduction Kant’s arguments about the moral status of animals may well be the most frequently criticized aspect of his frequently criticized ethics.1 His view that all our duties regarding non-human animals are duties to ourselves is widely thought to capture neither the content of these duties nor their ground. These objections have a variety of different sources, but we will focus on one in particular: a particular conception of what it is for a duty to run from one person to another, for duties to be, as it is sometimes said ‘directional’. Our aim in this chapter is to articulate and defend an account of Kant’s understanding of the directionality of duty, and to deploy it to explain and defend his notorious claim that our duties regarding animals are duties to ourselves. More generally, we seek to explain the relation between the content of a duty and its directionality. We argue that the nub of the standard objections to Kant’s view on animals is not that the duties are owed to the animals (the absence of the supposed right kind of directionality), but that they are owed to ourselves (the presence of the supposed wrong kind of directionality). We then distinguish between three kinds of problems to which the supposed wrong kind of directionality is supposed to give rise: wrong content (Kant’s account cannot justify duties we all agree we have regarding animals), instrumentality (Kant’s account makes our concern for animals instrumental to the end of developing a virtuous disposition), and contingency (Kant’s account makes our duties to our animals contingent on the obtaining of certain empirical facts). We then show that ‘wrong content’ and ‘instrumentality’ can be easily set aside, once we properly distinguish the directionality and the content of the duty. ‘Contingency’, on the other hand, seems to present a more serious challenge. However, we argue that ‘contingency’ rests on a misunderstanding of how a general principle such as the moral law applies to finitely rational agents like us. 1 Even philosophers who have drawn inspiration from Kant’s work to provide a defence of the ethical treatment of animals have rejected Kant’s own views in this matter as presented in the ‘Amphiboly in Moral Concepts of Reflection’ (MM 6: 442–4). See, for instance, Christine Korsgaard (2005, 2018) and Wood (1998). For some notable exceptions, see Denis (2000) and Kain (2010).

Arthur Ripstein and Sergio Tenenbaum, Directionality and Virtuous Ends In: Kant and Animals. Edited by: John J. Callanan and Lucy Allais, Oxford University Press (2020). © Arthur Ripstein and Sergio Tenenbaum. DOI: 10.1093/oso/9780198859918.003.0008

140

Arthur Ripstein and Sergio Tenenbaum

7.2 Kant and Ordinary Morality Kant insists that his moral philosophy does not introduce any new or surprising principle; it rather answers a philosophical question about how morality is possible. He does not suppose that most moral questions are particularly difficult, even if doing what morality requires is. His aim instead is to explain the distinctive type of interest that we take in morality, and the way in which it can be a rational constraint on conduct. As we go through his controversial, indeed notorious, views about animals, we will identify philosophical misconceptions that make some of those views seem forced and implausible. Before doing so, however, we want to draw attention to the ways in which what Kant proposes about our indirect duties to animals coheres well with familiar moral ideas about those duties. There are, no doubt, people who oppose those familiar ideas, but much of that opposition can be traced to different interpretations of the indirect duty, rather than to the supposition that we owe directed duties to animals. We therefore begin with several examples. Bill van der Zalm was minister of social services in the Canadian province of British Columbia. He cut social services dramatically. A cartoon in the Victoria Times-Colonist depicted him gleefully pulling the wings off flies. Van der Zalm sued (unsuccessfully) for defamation, drawing even more attention to his conduct than the cartoon had. His suit failed because the court did not accept his characterization of the cartoon as making a false statement. Nobody viewing it would come to the conclusion that the cartoonist was claiming that he actually pulled the wings off flies. Instead, they noted that pulling wings off flies is a familiar trope for cruelty. But why would this be so? One possibility is that we are concerned with how bad it is for the flies to be subjected to this treatment. Perhaps some people think in this way, but many people who would happily kill flies, swat them, and spray toxic chemicals that kill them slowly would regard pulling wings off them as emblematic of cruelty. This is not, we propose, because of the suffering of the flies, but rather because of the disposition that such conduct manifests. But our concern about the disposition it manifests is not, as it were, purely dispositional. It is not that most people think that pulling the wings off flies makes someone more likely to cause harm to other human beings or other animals. There’s something wrong with the person who would do that, even if we could be perfectly secure in our confidence that, although it could no longer be said of such a person, ‘he wouldn't hurt a fly’, he wouldn’t hurt anything other than a fly. Kant’s discussion of moral duties with respect to animals is accompanied by examples of the mistreatment of inanimate nature. This assimilation of mistreatment of animals to wanton destruction of nature is not an unusual or idiosyncratic view; the person who would destroy the beautiful crystal formation displays a character that is defective in the same way, even if that person wouldn’t even

Directionality and Virtuous Ends

141

hurt a fly. These examples suggest that there is a special kind of defective character that is on display in the case of mistreatment of animals and wanton destruction of nature.2 To show this distinctive defect is not, without more support, sufficient to show that it is the only, or most significant, moral dimension of human-animal interaction. If flies and crystal formations do not suffer, perhaps a set of principles apply to them that is different from those that apply to sentient animals. Even here, however, animals seem to figure as objects of moral concern in a different way than do persons. Consider, for example, the proposal to cull some of a herd of caribou for the sake of the herd—perhaps the natural predators have been wiped out, and the herd is in danger of overgrazing its range. We are not wildlife zoologists, and so take no position over whether this is in fact a prudent idea, or good for the herd or ecosystem. We mention it here only to draw attention to the sense in which it is a moral possibility in a way in which proposals to cull a population of human beings for the good of the group is entirely outside the scope of moral acceptability. When we care about the caribou, we care about them as a group (unlike fascists, in this instance we care about a group of which we are not members). The same frame of thought is available for other sentient beings, such as flocks of birds, schools of fish, and so on. Neither of these examples shows that there are not, or could not be, directed duties as well; they show only that at least some of our familiar ways of thinking about animals do not include directed duties, but may include indirect duties. In an Op-ed in the New York Times,3 Jeff McMahan suggested that the world would be better if there were no carnivorous animals. He was not proposing to eliminate them, but merely making an axiological judgment: given that being killed for food is a bad thing for an animal, it would be better if there were no animals that did this to other animals. McMahan’s argument drew a lot of objections, from a variety of quarters, including those who worried that without predators the world would be quickly overrun with herbivores. A more interesting response came from a different direction: animals, this objection went, have natures, and the diversity of animal life is a wondrous thing, each animal having its own nature, even if it is bad for some animals that other animals have the natures that they do. What is most striking about this debate is that both sides have some moral coherence to them, but neither side thinks of animals as having rights. The axiological argument focuses exclusively on vulnerability to pain, and says nothing about duties (McMahan neither thinks that he provides an argument for eliminating 2 The same point applies even to artefacts; a point made vividly in the Pixar film Toy Story, in which the main villain is a boy named Sid who mutilates and destroys toys. The toys in the film are alive and conscious, but none of the humans is aware of that fact, so Sid’s viciousness is apparent independently of it. 3 Jeff McMahan ‘The Meat Eaters’, The New York Times, 19 September 2010, available at http:// opinionator.blogs.nytimes.com/2010/09/19/the-meat-eaters/?_r=0. Accessed 25 August 2015.

142

Arthur Ripstein and Sergio Tenenbaum

carnivorous animals, nor that those animals wrong their prey). The argument from biodiversity finds value—action-guiding moral value—in the existence of animals but not in a way that turns on rights. These examples do not show that Kant is correct, but they do show that ordinary moral thought is more hospitable to his approach than might have been thought.

7.3 Some Preliminary Distinctions: Directed Duties v. Mandatory Ends The issue of directionality is often confused with the issue of what my end is. One way to put the complaint is to say that on the Kantian view, we care about our own perfection rather than about the welfare of the animals. The style of objection is easily recognized; it belongs in the same family as the complaint, for instance, that even if utilitarian or social practice-based accounts of promising can deliver the result that prohibits routine promise breaking, they would be doing it for the wrong reasons: I should not break a promise because I owe it to the promisee to keep it, not because of its possible marginal effects on the institution of promising. A familiar objection to Kantian ethics more generally turns on the same strategy. Schopenhauer argued that Kant’s ethics is ultimately a form of egoism,4 since the final object of moral attention is the consistency of the agent’s own will with its proper principle. More narrowly, generations of commentators have suggested that the Kantian attention to the form of one’s maxim and disdain for sympathy seems to give preference to the person who visits a sick friend begrudgingly over a more directly loving one,5 and so, effaces the moral attention to the friend of its fundamental role. But the question of improper ends and directionality are different questions; they are fundamentally different ways in which things can go wrong morally. Kant’s most developed treatment of the direction of duties comes in his Doctrine of Right, the first part of the Metaphysics of Morals. He argues that the moral concept of right: . . . has to do, first, only with the external and indeed practical relation of one person to another, insofar as their actions, as deeds, can have (direct or indirect) influence on each other. But, second, it does not signify the relation of one’s choice to the mere wish (hence also to the mere need) of the other, as in actions of beneficence of callousness, but only a relation to the other’s choice. Third, in 4 Schopenhauer (1840/2000, § 7). 5 Schiller is probably the first to press this objection. See Friedrich Schiller and Johann Wolfgang von Goethe, Xenien, translated by Paton (1947: 48). Kant replies to it directly in Religion (Rel. 6: 23n), but his reply is often ignored by contemporary commentators who make the same point—see Stocker’s well-known article (1976).

Directionality and Virtuous Ends

143

this reciprocal relation of choice no account at all is taken of the matter of choice, that is, of the end each has in mind with the object he wants . . . (MM 6: 230)

Directed duties of right do not focus on ends at all, they rather restrict the ways in which free beings can use means—centrally their bodies and property—to set and pursue their own ends. In the Doctrine of Virtue, Kant introduces directional duties of virtue, including the duty of love to other human beings (comprised by beneficence, gratitude, and sympathy) and duties of respect, which are specified by their correlate vices of arrogance, defamation, and ridicule. Directed duties of virtue differ from directed duties of right in that the former, but not the latter, presuppose an end. In order to understand Kant’s distinctive position about the moral significance of animals, we must proceed in two stages. First, we will explain relational duties that are correlative to rights. We will then turn to relational duties of virtue. In characterizing relational duties of right as abstracting from all ends, here as elsewhere, Kant is not purporting to invent a new moral principle, but rather to draw our attention to what we already understand about morality. An action can be directed towards someone without implying any kind of concern towards that person. In the case duties of right, the duty to, say, a party to a contract is owed to the party. So they are directed to another agent. But this, of course, does not imply any concern for the party to whom the duty is owed. If you repair my refrigerator, I need to pay you a sum of money, but I do not need to do so out of any sort of concern for you. Indeed, in the normal course of events, I will enter into the arrangement under which I have to pay you because of some other purpose of mine, such as preventing my food from spoiling. You are not my end, and your enrichment is not my end. That is not to say that in hiring you I treat you as ‘a mere means’. I do not, but only because our transaction is consensual; for the sake of receiving the payment, you use your means (your skills and equipment) to repair my refrigerator. Even with my duty to avoid using or damaging your property without your authorization, I do not need to care about the fact that it is yours; I act in perfect conformity with right if I resent the fact that you have a nicer car than I do, so long as I do not interfere with it. Conversely, I may be so infatuated with your beautiful vehicle that I get carried away in my attempts to avoid interfering with it in any way, so much so that I inadvertently damage it. I have wronged you, even though my end was avoidance of wrong to you. So, too, if I am vigilant about avoiding crossing into your property, but end up, disoriented, on it anyway. Again, you have a right to be beyond reproach that prohibits me from defaming you, but it does not demand that I care about your standing in the eyes of others. That is why I can make all manner of damaging statements, provided that they are true, and under traditional juridical understandings (still honoured everywhere on the planet outside the United

144

Arthur Ripstein and Sergio Tenenbaum

States) I can go out of my way to avoid saying anything false about you, but if, in spite of my diligence, I say something false and damaging, I thereby wrong you, because I interfere with your right to be beyond reproach. Duties of right do not require the adoption of any particular end (as opposed, for instance, to compliance with duties of virtue such as gratitude). Not only do relational duties of right not depend on ends; the priority that Kant assigns to right over virtue, according to which non-rightful means cannot be used in pursuit of morally mandatory ends, could only be intelligible if relational duties can be comprehended without any reference to any end whatsoever. This feature of right is the hallmark of Kant’s account, but is easy to miss because of the tendency in so much recent philosophical writing to imagine that the basic form of a right is a rule protecting against some type of harm. This way of thinking about rights—which has its ultimate roots not in Kant or indeed any idea about right, but rather in the utilitarianism of Sidgwick and Mill—treats them as ways of protecting particularly urgent interests. On this view, the point of the right is to protect the interest by placing some person or persons under a directed duty to perform acts conducive to the protection (or advancing) of the interest. On such an approach, the directionality of the duty is secondary to the interest it seeks to protect. If rights are viewed as instruments in this way, it does indeed appear puzzling that animals would not have them also, if we suppose animals to have pressing interests. But that is not the way in which Kant analyses the directionality of duties. Kant’s approach to directionality has considerable advantages. Most notably, it explains what is wrong with what might be called moral circumvention—interfering with a right so as to see to it that the rightholder benefits, as in cases of paternalism. More significantly, a characterization of familiar directional duties in terms of the interests that they serve faces a conceptual difficulty: although the relevant interest is sometimes put in terms of well-being or, at other times, autonomy, on closer inspection, the most familiar relational duties protect interests that are impossible to describe except by reference to the concept of a right that they were supposed to explain. Consider the right to be free of unwelcome caresses, or of unauthorized use of your property. As the words ‘unwelcome’ and ‘unauthorized’ indicate, these are not interests in being free of a certain state of affairs; the interests instead are in being free of certain types of violations of appropriate interpersonal relations. That is, the only thing that can set back the interest in question is the violation of the right. This is not a case in which an interest is particularly important to a person’s life, and others are placed under duties to protect or promote that interest through actions likely to conduce to it. There is no conceptual space between the interest and the norm of conduct that protects it. These are familiar markers of ordinary moral thought about rights, markers that writers have struggled to accommodate in interest-based accounts. But the view has the

Directionality and Virtuous Ends

145

clear advantage of making each of these things not an outlier standing in need of special pleading, but instead a constitutive and structuring feature of the very concept of a right. This way of thinking about rights has an immediate implication for non-human animals. In the ‘Division in Accordance with the Relation of the Subject Imposing Obligation to the Subject Put under Obligation’ at the end of the Introduction to the Doctrine of Right (MM 6: 241), Kant distinguishes between the relation in terms of right of human beings towards beings that have rights as well as duties from other possible sets of relations. Beings with duties but no rights would be slaves, a situation that is factually possible but morally impossible. Beings lacking reason, who cannot be bound and so, too, cannot bind other beings (Kant’s conception of animals), have neither rights nor duties. Those who suggest that animals have rights must slot them into what Kant regards as the fourth category, which is also vacant, that of God, a Being that has only rights but no duties.6 There is a familiar objection to so-called ‘will theories’ of rights, which might be thought relevant here. The nub of the objection is that by focusing on how things stand between beings capable of asserting and waiving claims against each other, those theories are unable to account for the rights of children or even rights of a person who is asleep or comatose. Whatever the fate of the will theory of rights (with its emphasis on the possibility of waiving as constitutive of a right), the objection fails to engage with the Kantian account. For those examples start with the thought that we have already answered the question ‘who is a person?’ in familiar ways, that is by identifying each person with his or her body, just as Kant proposes. That there are human bodies that are temporarily incapacitated (by childhood or sleep) does not make them stop being persons and so does not make them cease to be bearers of rights. There are also human bodies that are more severely, perhaps permanently, incapacitated. They are still bearers of rights because we include them in the category of persons, since the only way in which we can individuate embodied rational beings is by keeping track of human bodies. Thus we appoint guardians for such persons, charged with protecting their personality, even if we fear that it will never manifest itself.7 This is not an instance of recognizing rights of beings that are incapable of choice; given that they have a human body, we have the same grounds to impute the power of choice to them that we have to impute to other human beings, despite the fact that there is no immediate empirical manifestation of that power. Most significantly, it is not a case of imputing rights to a being because of an interest that can be expressed in terms of anything other than concepts of right. 6 We are grateful to Jacob Weinrib for drawing our attention to this structure. 7 For a detailed discussion of the grounds to attribute personality to all human beings (and to no other animals), see Kain (2009, 2010).

146

Arthur Ripstein and Sergio Tenenbaum

This robustly relational conception of rights also explains Kant’s juridical interpretation of the traditional theological idea that the Earth was given to human beings in common. Shorn of its biblical origins, such a view might be thought to draw an arbitrary distinction between humans and other animals inhabiting the Earth. Kant sees things differently, however: he suggests that prior to the appropriation of land as property, human beings were in what he calls ‘disjunctive’ possession of the Earth’s surface. That is, each person is entitled to be wherever he or she happens to be, and does no wrong by occupying space; conversely, one person wrongs another by displacing that other from the space he or she happens to occupy at that time. Thus the idea of possession in common is subordinated to the idea of reciprocity. Each person is entitled to be wherever another person is not. The advantage of this way of thinking about it is clear: the Earth is not ‘given’ to human beings in order that they may meet their needs (leaving other needy beings out of the picture). Instead, each person is restricted only by the rights of others.

7.4 Directionality and Ends Duties of virtue are fundamentally different. As Kant explains in the introduction to the Doctrine of Virtue, they are always duties to adopt an end. Outward conformity does not qualify as virtuous conduct, because the virtue consists in the end for which you act, rather than the action you thereby perform. Outward conformity is beside the point. The duties of love and respect are always directed to other specific individual human beings, because the only way in which love or respect for others can be an end is if it is directed to some particular person. In the case of duties of love, another person can only be your end in his or her particularity. You do not owe humanity in general an undifferentiated duty to love them (at least not in a pathological sense). Instead, the concrete way in which you make another person your end depends upon the relationship in which you already stand with that person. This structure is completely familiar in the case of friendship: the duties that you owe to your friends are owed to them in particular. You may also have a further duty to make friends, but this is not owed to everyone; it is instead a duty owed to yourself. In case of duties of respect, the humanity of another person only constrains your conduct in its particularity: in interacting with others, you always interact with someone in particular, and must make that person’s humanity your end. That is why duties of respect are negative, characterized in terms of their opposed vices of arrogance, defamation, and ridicule, each of which is always directed at someone in particular. On Kant’s understanding, the arrogant person wants to be superior to others—particular others; defamation is finding fault, which is by its nature always particular, and ridicule is always ridicule of some particular person.

Directionality and Virtuous Ends

147

Even though duties of virtue are duties to adopt an end, the directionality of the duty is distinct from its end. For just in the case of right, directionality of virtues of justice does not pertain to the object of the will, but to how my will (rather than my conduct) is constrained by the will of others. An end, according to Kant, is ‘an object of choice (of a rational being), through the representation of which, choice is determined to an action to bring this object about’ (MM 6:381). Of course, the object represented could be a human being, or more specifically, the state of a human being. My end might be the well-being of a friend or happiness of other agents in general; yet this says nothing about the directionality of the action. I have a duty to myself not to lie to you even if it the lie is harmless or for an otherwise good end; in this case, both you and I are represented as the object of my will (a relation in which I express the contents of my thoughts to you). But the duty in this case is a duty to myself (MM 6:429-30), as it is a constraint that my own will imposes on itself. In general, the directionality of the duties of virtue depends on how the universalization of my maxim is constrained by the will of others. In the general case of beneficence, my need for the help of others in the unavoidable pursuit of my own happiness can only be universalized as mutual constraint of all our wills by the wills of each other. If you are to make the object of my choice also the object of your choice, I must constrain my choice so that the object of your choice can also be its object. But the directionality is determined by the source of constraint (your rational will), not by the nature of the resulting object of choice. So, for instance, in the specific case of gratitude, gratitude owed due to your past action might require that I take an interest in the welfare of your children. It would be wrong to say that my gratitude for you requires that I take an interest in the welfare of your children only as a means to something else, like your happiness. It would be strange, to say the least, if my interest in the welfare of your children was only sensitive to how happy you would be as the result of my actions on behalf of your children; gratitude requires that I care for your children for their own sake. Yet the duty is a duty to you, as the object of my choice here is constrained by your will. Since non-human animals do not have a rational will, they cannot constrain the possibility of our maxims having the form of universality. Notice that Kant’s distinction between directionality and the object of one’s choice (one’s end) is also confirmed by ordinary morality. Kant says that in pursuing the happiness of someone else ‘I can benefit him only in accordance with his conception of happiness’ (MM 6:454). Even I think your life will be much improved by listening to concerts instead of playing video games, I cannot discharge duties of beneficence by dragging you to the theatre hall or destroying your game console. My choice must be constrained by your will and thus by the specific objects of your choice, not by a general desire for your well-being. In the case of the non-human animals, the opposite is true; the relevant ends are fully determined by my general desire

148

Arthur Ripstein and Sergio Tenenbaum

for their well-being. But just as in the case of my concern for the well-being of the children of my benefactor, the fact that the duty is not to the non-human animals does not imply that the well-being of the animals is not the immediate object of choice; here too, we care for the well-being of the animal for its own sake. For the same reason, animals cannot stand in anything more than weak analogues of relations of friendship and love. Because they do not set their own ends, they cannot stand in relations of mutual adoptions of ends characteristic of friendship between human beings. Another example that Kant gives in the amphiboly section is our duties regarding inanimate objects—in particular, beautiful nature. Kant argues that we have a duty to refrain from the ‘wanton destruction of nature’, a duty not to uproot a disposition to love things ‘even apart from any intention to use it’ (MM 6: 443). This emphasis on the absence of intention to use plainly marks the subject matter of the disposition: it is a duty to adopt an end. In characterizing this as a duty to adopt an end, Kant is presupposing his more general conception of purposive action. In the Introduction to the Metaphysics of Morals he distinguishes between choice and mere wish. Choice necessarily involves the taking up of means. Only if the faculty of desire is ‘joined with one’s consciousness of the ability to bring about its object by one’s action is it called choice; if it is not joined with this consciousness its action is called a wish’ (MM 6:213). Secondarily, it is also the duty to develop the inclinations to have the preservation of beauty in nature as an end. The duty to develop those inclinations is indirect because it is a way of having the end in question; developing inclinations is the means you use to have that end. If Jane refrains from such a destruction of nature to avoid fines from the Department of Forestry, her actions, as far as we have described them, cannot be the expression of any action done from duty. In this way, she’s no different from Kant’s honest shopkeeper example: they lack moral worth due to the fact that the wrong end is being pursued. As Kant makes clear in his discussion of the shopkeeper, that does not mean that either the shopkeeper or Jane does the wrong thing; Kant passes over ‘all actions that are already recognized as contrary to duty’ (G 4:397). Importantly, they also could not be expressions of the relevant virtuous disposition given that they do not involve an immediate liking of nature. Notice that this is true even if having enough money is also useful to the ends of morality. The pursuit of one’s own happiness is an indirect duty, and this duty certainly implies that one should adopt the end of saving money. For most of us, avoiding hefty fines is certainly necessary for this end. But that is not the sense in which the duty is indirect. The virtuous disposition is not an inclination that happens to generate actions of preservation of nature that are useful to the development of a virtuous character; the virtuous disposition is a disposition to develop a particular inclination that has the preservation of nature as its end.

Directionality and Virtuous Ends

149

In the Groundwork, another well-known character acts in accordance with duty but not from duty. The sympathetic person has an ‘inner satisfaction’ to pursue the welfare of others; the sympathetic person has towards other people’s happiness the same ‘immediate inclination’ we all have towards our own current well-being. The sympathetic person, we know, does not pursue the welfare of others from duty, at least when this natural inclination to pursue the well-being of others determines the action. Thus, such actions lack, a fortiori, the proper directionality as they are not pursued from our practical cognition of our duties to others. However, as Kant points out, beneficence towards someone will give rise to love, as ‘an aptitude to the inclination of beneficence in general’ (MM 4: 402). Moreover, Kant recognizes in the duty of friendship a moral ideal; thus the cultivation of sympathetic feelings towards one’s friends is morally required. A virtuous person will thus have sympathetic feelings much like those of the Groundwork character, but as she acts from these feelings, she acts from inclinations that were themselves determined by virtuous agency and thus her act has a (possibly indirect) moral worth that cannot be found in the sympathetic person. These acts then have the proper directionality, not by the way that the virtuous person does, but the sympathetic person does not, care for the recipients of her beneficent acts. Both care for the recipient, but the virtuous person cares about the recipient as a person; the virtue consists in making the other an end, rather than just the object that will satisfy the agent’s inclination. That is, the proper directionality comes from the determination, in the case of the virtuous agent, of her sensibility by the cognition of moral law. More particularly, her benevolent inclinations are the effect of an agency guided by the moral law and by the practical cognition of the humanity of other agents. Of course, the virtuous agent who has developed her friendships and cultivated neighbourly love is not the only agent whose actions of helping others have moral worth. The Groundwork sympathetic character is contrasted with his possible later self, when his mind has been ‘overclouded by his own grief, which extinguished all sympathy with the fate of others’ (G 4: 398). If in this case he ‘nevertheless tears himself out of his deadly insensibility’ and acts ‘from duty’, then, as we all know, his actions have for the first time ‘genuine moral worth’ (G 4: 398). Let us say that the actions of the benevolently inclined virtuous agent have indirect moral worth, while those of the unsympathetic agent have direct moral worth. It is important to note that ‘indirect’ and ‘direct’ here do not distinguish between lower and higher grades of moral worth, but between more and less direct effects of the consciousness of the moral law. Helping others can have direct or indirect moral worth, and those who have no benevolent inclinations can still help others from duty. This is, at least in some sense, not true in the case of animals. An agent can act from the virtuously cultivated inclination to prevent

150

Arthur Ripstein and Sergio Tenenbaum

animal suffering (and other similar inclinations), but there seems to be no room for the corresponding agent who, perhaps overclouded by grief of losing her own pet, would tear herself out of her indifference for the fate of her neighbour’s cat and help him down a tree from duty. Kant seems to be committed to the view that if she were to act in this way, she would be acting on a misconception of what duty requires; she would have fallen prey to an amphiboly of concepts of reflection, and her actions would not have any moral worth. Before we address this issue, we should note that at least some of Kant’s critics are already in a bind. Kant’s treatment of the sympathetic agent is notoriously controversial; many find it counterintuitive not to accord the actions of the sympathetic agent more moral worth than the actions of the person who acts from duty; actions from duty are supposed to be ‘alienated’ or ‘repugnant’. However, in Kant’s picture, the motivational structure of the virtuous agent is, in the relevant respects, like the one of the sympathetic agent. In her pursuit of the welfare of non-human animals, she does not act in the absence of, let alone contrary to, inclination. She is moved by her properly cultivated predispositions to care for the pain and pleasure of the brutes. So, surprisingly, the usual critic of Kant should think that his treatment of animals is the highlight of his work—the one issue in which Kant seems to put moral worth where it belongs. Of course, we don’t point out this fact as a way to sell some portion of Kant’s views to nonKantians; rather, this surprising conclusion should show that there is nothing obviously counterintuitive that follows directly from the claim that duties regarding animals are not duties to animals. But, of course, it is imprecise to say that a Kantian needs to acquiesce on the uncaring action of the agent saddened by the loss of her pet. Cultivating inclinations or preventing their destruction is nothing other than acting in the way determined by such inclinations. Inclinations just are habitual desires. Someone who acts cruelly to non-human animals is thereby destroying the inclination to be kind to the brutes. The view does imply that the correct way to describe the reason to perform this action in the case of the lack of inclination is that one should not allow one’s cold insensitivity to cats to develop further (or one should cultivate whatever care one still has for the brutes). But is it counterintuitive to say that the agent helps the cat because she understands that she should not be insensitive to the plight of cats? It is interesting to compare what Hume has to say about the agent that acts from the motive of duty: When any virtuous motive or principle is common in human nature, a person who feels his heart devoid of that principle, may hate himself upon that account, and may perform the action without the motive, from a certain sense of duty, in order to acquire by practice the virtuous principle. (Hume 1738/1978: 3.3.1)

Of course, Kant must say that Hume is completely wrong in his general understanding about what it is to act from duty. But he does describe a recognizable

Directionality and Virtuous Ends

151

form of the attempt to cultivate a virtuous inclination, when that inclination is lacking: a form of second best for those who do not have the appropriate virtuous motivation. Although Hume’s general view cannot capture the way in which a benevolent action done from duty expresses respects for the humanity of the recipient of help, its more specific version correctly describes the motivation of the agent who recognizes she fails to have virtuous inclinations regarding animals. Kant does seem to have a ‘harsher’ view of our relations to animals in some of his writings; one might suspect that we are hiding under the carpet the extent to which Kant would allow us to treat animals in whichever way would advance whichever end we might have. Kant himself seems very close to saying exactly that in Conjectural Beginnings: [After making use of a sheep’s skin, humans] became aware for the first time of a prerogative that he had by his nature over all animals, which he no longer regarded as his fellow creatures (als seine Mitgenossen an der Schöpfung), but rather as means and instruments given to his will for the attainment of his discretionary ends. (Con. 8: 114)

This passage suggests that Kant seems to have a much less kind view of our treatment of animals than we’ve been claiming. After all, he is in fact saying that they are means and instruments (Mittel und Werkzeuge), and there seems to be no limitation to the discretionary ends that animals can be put to use for. But what exactly do such passages show? First, it would be surprising that Conjectural Beginnings should be in such stark contradiction with writings, such as The Metaphysics of Morals, in which Kant so explicitly restricts the range of permissible actions in our relationship with animals; and indeed, read correctly there is really no conflict. Conjectural Beginnings claims that humans distinguish themselves from the animals as they realize that the latter can be treated merely as means. That animals can be treated merely as means should come as no surprise; human beings must be treated as ends in themselves in virtue of their humanity; in fact, the claim that animals can only be treated as means is inseparable from the claim that we have no duties to them—we owe duties to those whom we must treat as ends. But the fact that animals must be treated as means does not contradict in any way the claim that a virtuous person has the welfare of the animals as her end. It is worth here looking at what the Formula of Humanity requires from us. As we said above, we do not fail in our duties to others when we hire their services without being particularly concerned with their well-being. For the same reason, we also do not treat another merely as means if we, for instance, hire a cab without any thoughts about the well-being of the cab driver. And just as in the case of complying with duties of right, the fact that you have someone’s happiness as your end in no way guarantees that you are not treating her as means. If I drag someone into my car without her consent and drive her to the baseball game because I am confident that it will be great fun for her once she’s there, I treat her as means

152

Arthur Ripstein and Sergio Tenenbaum

no matter how correct I am in my confidence that she will enjoy it and how careful I was not to cause her any pain or harm. Although it is beyond the scope of this paper to put forward an interpretation of the Formula of Humanity, there is no question that the prohibition of treating other as means is largely independent of the end I am pursuing; although there are ends that I could not pursue without treating you as means (the infliction on pain for its own sake, for instance), there are many ways in which I can interact with you that are not particularly caring but that do not amount to treating you merely as means. But more importantly, I can easily act in the pursuit of your happiness in a way that treats you merely as means to my end of pursuing your well-being. The requirement expressed by the Formula of Humanity is a requirement to engage the rational will of others in a certain way (in a nutshell, the requirement to regard their will as a condition of goodness), and a corresponding prohibition to act in ways that are incompatible with this kind of engagement. All of these constraints on using others as mere means turn on the thought that humanity is an end it itself. By contrast, it is simply unintelligible to treat nonrational nature as an end in itself. The distinctive status of humanity is not an open-ended permission to subordinate everything else, including animals, to human desires; it is rather that there is a mode of interaction that is available among human beings (as fellow citizens of the kingdom of ends) that is simply unavailable with regard to other creatures. You cannot have a duty to engage the rational will of a being that lacks one; but treating the brutes as means is compatible with having their well-being as our end. In fact, when I take my dog into the car without (pointlessly) trying to secure her consent because I am correct in my confidence that she will enjoy the baseball game (or at least the discarded hotdogs!), I clearly do something admirable, unlike the case in which I drag my friend. The kind person who takes the thorn from the lion’s paw unconcerned by the lion’s uncomprehending stare is obviously treating (rather bravely) the lion as means to her end of helping the lion; had she done the same to a rational agent without securing his consent, she would be assaulting him. But interaction with a lion is neither consensual nor nonconsensual on the lion’s part. The virtuous disposition must care for the welfare of non-human animals without the thought of any further end, but it can only secure the object of its concern by treating these same animals as means.

7.5 Contingency If we are right so far, then the Kantian virtuous agent who has developed the proper inclinations will have the well-being of animals as her end; she will not be taking care of Tibbles in order to develop her own moral character. Rather, the development of her moral character has ensured that she cares for Tibbles for his

Directionality and Virtuous Ends

153

own sake. However, one might complain that there are still some seriously objectionable aspects to Kant’s understanding of our duties regarding animals. After all, on Kant’s view, it seems a matter of contingency that animal cruelty is not permissible; were human nature a bit different, it might have turned out that our dealings with animals were irrelevant to the development of our moral character, and Kant seems to be committed to the view that if this were the case, we would have no duties regarding animals. The charge of contingency can be sharpened with a comparison. We have a duty to take care of our children. There is certainly a similar duty not to interfere with and to develop one’s disposition to love one’s children. There is no doubt that we find in human nature an inclination to love one’s own children, and it would be hard to understand the content of our duties to and regarding our children in abstraction from these inclinations. We could imagine a possible world in which such inclinations are absent from human nature, but instead, adult human beings have benevolent feelings that are indifferent towards all children; the ties of affection in such an imagined world would be like the ones that we would ideally find in a Platonic city. Such adults would perhaps organize child rearing in the manner suggested by Plato’s Republic; perhaps in such a world specific groups of professional caregivers would be in charge of specific life stages of the community’s children, and would develop inclinations of love towards whoever was in their charge that day, but, when their shift ended, feel nothing. Although parents in such a world would not particularly love their children, there is nothing morally objectionable in such a world. However, in such a world there would be, arguably, no duty to cultivate feelings of parental love or emotional dispositions towards one’s own children. By contrast, a world in which children are left to fend for themselves is not one that the ‘judgment of impartial reason’ could deem as good. However we try to imagine other ways in which our duties to others can be fulfilled, they must be compatible with the promotion of our children’s happiness and the development of their rational faculties. But it seems hard to separate this constraint on acceptable specifications of duties from the fact that we owe these duties to the children or at least that the happiness of others is an end that is also a duty. Given that neither of these things is true in the case of animals, wouldn’t it be possible to imagine a configuration of human sentiments such that our duties to ourselves demand nothing regarding animals? Couldn’t there be colonies of rational beings on Saturn, whose moral perfection was completely disconnected from any concern for animal welfare? The contingency objection says that this very possibility shows that Kantian ethics does not give animal welfare its due. Our duties to animals is simply the by-product of empirical features of our nature; the very possibility that there are morally acceptable worlds in which animals are cruelly treated, or that their welfare receives no consideration, shows that there is something wrong with this conception of our duties to animals.

154

Arthur Ripstein and Sergio Tenenbaum

The cultivation of inclinations is part of the development of a virtuous character. A divine will would have no inclinations and thus would act from the moral law on each occasion. A holy will might have inclinations, but they would be themselves determined by the moral law and never conflict with it; the pursuit of the moral law encounters no resistance. Now if the contingency objection is heard as saying that the duties of a divine being might not involve caring for animals in the ways we find appropriate and morally correct, it is certainly unanswerable. We know nothing about how the moral law applies to a divine will; whenever Kant derives a duty, he appeals to some limitation of human agency—‘someone feels sick of life because of a series of troubles’, ‘another finds himself urged by the need to borrow money’, ‘many cases could occur in which one would need the love and sympathy of others’—and although Kant does not say this explicitly in the case of developing one’s talents, obviously the derivation depends on the fact that we find ourselves in need of these various talents; they are given to us ‘for all sorts of possible purposes’; a being with intellectual intuition has no need of any other talent. We do not know whether a divine will would help others, and one does not imagine that she would incur many debts (not to mention that self-destruction is out of the question for a necessary being and that her talents are already fully developed); the notion of a maxim does not even apply to an infinite will (CPrR 5: 79). We can also imagine other forms of finitely holy wills for which we know very little about what kinds of duties they have. But if the ‘contingency’ objection needs to appeal to these facts in order to explain the relevant sense in which our duties to non-human animals are merely contingent, it is not particularly worrisome. The fact that we do not know how the moral law would guide radically different beings merely shows a limitation of our moral imagination. For the contingency objection to have any force, it has to show at least that it is compatible with human nature, understood as that of a finitely rational being with sensibility (or even a sensibility that has the same general characteristics as ours) that the cultivation of virtues will involve, or at least permit, the cruel treatment of animals (or other obviously impermissible treatment of animals), but this possibility can be at least partly ruled out. Although our will could not be brought to conform to the ideal of a holy will in this life, we have a duty to approximate such a holy will, to become more virtuous and to progressively conform our inclinations to the commands of duty. Our inclinations, qua habitual desires, are part, of course, of our sensible nature. Desires are effective representations, and some of these representations involve concepts of the understanding; they are representations whose origins are in experience. The representation brings about its object, and, of course, the object in question is an object of experience (unlike pure willing that modifies only itself). But inclinations cannot be guided directly by representation of the moral law; they must be desires formed in response to sensible representations.

Directionality and Virtuous Ends

155

We can now see how the objection that Kant’s view of our duties to animals is based on a dubious empirical claim can be avoided. The duty to cultivate inclinations that conform to our duty to others is a duty to develop those aspects of human nature that respond to the ‘analogue’ of actions from duty; ideally the actions done from duty will not find in our sensible desires any obstacles. Our sensibility cannot respond to virtue, freedom, or any ideas of reason. But it can respond to the sensible aspects of the highest good: pleasure and the absence of pain in beings that have an animal nature. Natural dispositions whose ends are analogous to those of the moral law (in that they also forego the pursuit of one’s well-being in the pursuit of the welfare of another creature) and whose development, by its very nature, conform to the ends that the moral law requires us to adopt, are dispositions that we must cultivate. They constitute at least one specific form in which we can approximate the ideal of holy will. There thus are empirical assumptions that are required to make the case that we have an obligation to cultivate the relevant inclinations, but they are much weaker than claims about the conduct of butchers, farmers, or lab workers. The empirical assumption is that we have the capacity to develop a general, natural disposition for sympathetic pain or pleasure, but we do not need to await the result of psychological research to accept the correctness of this assumption. If we act under the idea of freedom, we conceive of our own character as subject to our choice, and so must suppose ourselves to be able to develop a sensibility that is compatible with the demands of morality. Of course, this does not rule out the possibility that empirical research would find out that cruelty to animals is correlated with a greater likelihood that people will help others. We might find out that slaughterhouse workers tend to be nicer to children or that those who are indifferent to the fate of their pets are particularly kind to strangers. But if any of these putative facts could be shown to obtain, they would remain irrelevant to our conclusion. The virtuous inclinations in questions are specifications of what counts as a virtuous disposition for the kind of finitely rational agents that we are; they are not instrumental means by which we cause ourselves to acquire the dispositions that are in fact independently virtuous. Someone who used these empirical generalizations in deciding their behaviour would be engaged in self-manipulation rather than in proper moral agency. As we said earlier, Kant’s view cannot rule out the possibility that there are beings whose sensibility does not include the general capacity for sympathetic pain or pleasure of any sort, beings that have a sensibility, and the ability to represent the sensibility of other sensible beings but who are entirely unmoved by that representation. Such beings would not be under a duty to treat non-human animals in any particular way. In fact, we know very little about what duties they would be under until we know more about their sensibility. But is the possibility of such a being enough to underwrite a plausible version of the contingency objection? It is far from clear that it is. It is hard for us to imagine that virtue can

156

Arthur Ripstein and Sergio Tenenbaum

be compatible in any possible world with treating animals badly. Of course, insofar as we are virtuous, given our ends, we cannot but contemplate such possible worlds, or any possible worlds in which animals are mistreated with sadness. But to say that we cannot rule out the possibility that such a virtuous existence is possible is not to say that we can have any understanding of what it would be like; we have no idea what the duties are of beings that are so different from us. Of course, the threat to Kant’s view is not so much the bare logical possibility of beings that do not have duties regarding non-human animals, but that such a possibility might illustrate that we have missed the true grounds of our duties to animals. However, we hope that the rest of the paper has done enough to assuage these worries.8

8 This paper was presented at the Kant on Animals Conference, Kruger Park, South Africa, July 2013. We are very grateful to the helpful comments from the participants on this occasion.

8

Kant and Moral Responsibility for Animals Helga Varden

8.1 Introduction Working out a Kantian theory of moral responsibility for animals1 requires the untying of two philosophical and interpretative knots: i) how to interpret Kant’s claim in the important ‘episodic’ section of the Doctrine of Virtue that we do not have duties ‘to’ animals, since such duties are only ‘with regard to’ animals and ‘directly to’ ourselves; and ii) how to explain why animals don’t have rights, while human beings who (currently or permanently) don’t have sufficient reason for moral responsibility do have rights.2 At the heart of the problem lies the philosophical challenge of whether a Kantian account of moral responsibility for animals can take the animals themselves into account in the right way, that is, without utilizing arguments that (wrongly) presuppose that non-human animals are moral agents or can be morally responsible for their actions. The task of untying these two knots is the aim of this essay. I start by defending Kant’s general claim that our duties regarding animals are not ‘to’ them, but rather ‘with regard to’ them. Relations between human and other animals are not relations between two kinds of moral beings, since animals are not capable of the kind of consciousness, in particular the reflective self-consciousness that moral 1 For ease of prose, I mostly use the words ‘humans’ and ‘animals’ throughout this paper. Only when it seems useful do I use language that explicitly draws attention to the obvious, assumed presumption (given the structure of my account), namely that humans are a kind of animal; those times I use the contrast human and non-human animals. 2 A special thanks to Lucy Allais, who not only decided that I would write this paper, but whose engagement with the ideas from the beginning has been invaluable. Huge thanks also to: Ingrid Albrecht, Sarah Broadie, Rachel Bryant, John Callanan, Lara Denis, Katerina Deligiorgi, Dan Hooley, Arthur Melnick, Eric J. Miller, Barbara Sattler, David Sussman, Sergio Tenenbaum, Shelley Weinberg, Ekow  N.  Yankah, and the audiences at related talks given at: the ‘Kant on Animals’ conference, University of Witwatersrand/Kruger Park; The Philosophy Club, University of St. Andrews; New Voices in Legal Theory workshop, University of St. Andrews; Department of Philosophy, University of Toronto; Brady public lecture series, Northwestern University. Finally, thanks to the Department of Philosophy and the Center for Advanced Studies at the University of Illinois at Urbana-Champaign; the University of St. Andrew’s Centre for Ethics, Philosophy and Public Affairs; and the Brady Scholars Program in Ethics and Civic Life at Northwestern University for having funded this research project. The mistakes that have survived despite all these efforts are, of course, mine. Helga Varden, Kant and Moral Responsibility for Animals In: Kant and Animals. Edited by: John J. Callanan and Lucy Allais, Oxford University Press (2020). © Helga Varden. DOI: 10.1093/oso/9780198859918.003.0009

158

Helga Varden

being requires. Instead, the more complex human-animal relations are affectionate, social relations to which humans can and sometimes should relate morally. This is why these moral duties are ‘direct’ only to ourselves; we are the only ones in the relations that can be morally responsible for them. Correspondingly, respect (in Kant’s precise sense of the word) is a distinctly reflective, moral emotion internal to the specific normative orientation we have only to beings that can be truly free (reflectively self-conscious), and so, to human beings; morality and respect single out how we regard ourselves and other human beings capable of freedom. As we will see, Kant has a consistent and not counter-intuitive account of why we have the moral attitudes we do about animals and of how it is that these attitudes have the appearance of being about the animals themselves. In addition, Kant can explain why it doesn’t follow from this that these moral attitudes arise from attributing moral rights to the animals themselves. Central to this interpretation is Kant’s Religion, as it contains an account of human nature that adequately explains why we have the positive attitudes we do towards animals; similarly, the Religion contains an account of human evil that explains why we consider other occasions of attitudes towards animals as genuine examples of moral failure. Hence, the Religion is central to understanding why Kant’s approach to animals is neither internally inconsistent nor counterintuitive.

8.2 Duties with Regard to Animals The main difference between human and non-human animals, Kant argues, is that only humans are capable of free choice since they alone have the reflective self-consciousness that makes free choices possible. And since morality is exactly the realm of free choices, only human animals can enter into moral relations. It follows, Kant argues in the famous ‘episodic’ section of the Doctrine of Virtue, that only a human being can have moral obligations, and consequently a human being’s moral duties when interacting with animals will always be direct in relation to oneself and indirect with regard to them. The aim of this section is to clarify this argument of Kant’s. In one of his very latest major publications on moral philosophy, The Metaphysics of Morals, Kant makes these arguments about human and animal choices in the following way: an animal choice is ‘determined only by inclination (sensible impulse, stimulus)’, whereas a human choice ‘can indeed be affected, but not determined by impulses;’ in fact, as humans mature and become capable of moral responsibility, their choices ‘can be determined by pure reason;’ their choices are then truly free (MM 6: 213f). I have an ability to step back and think about what I want to do regardless of how strong my current inclinations are (negative freedom) and I have an ability to do something just because it is the right thing to do (positive freedom) (MM 6: 213). Moreover, as always, this

Kant and Moral Responsibility for Animals

159

stepping back to evaluate the morality of one’s action (negative freedom), for Kant, involves ‘subjecting the maxim of my action to the condition of its qualifying as universal law’ (MM 6: 214). Essential to Kant’s take on these things is a fundamental conviction that no other animals we know of (on our planet, at least) can engage in the complex conscious way of engaging with the world characteristic of human being. To start, other animals can perform actions—do things and choose to do something rather than something else—but as they engage with the world, they cannot think in the way consciously using abstract concepts involves (as they think associatively and without abstract concepts).3 Correspondingly, in the practical realm, a non-human animal cannot act on what Kant calls a maxim, namely a ‘rule that the agent himself makes his principle on subjective grounds’ (MM 6: 224, cf. G 4: 401n). In light of these differences between human and animal choices it is possible to understand why Kant proposes that believing we have moral duties directly to  animals is an ‘amphiboly’ or ambiguity ‘in the moral concepts of reflection’ (MM 6: 442). What Kant means, I believe, is that in our ordinary ways of talking, we are ambiguous about how to understand our duties in relation to non-human beings—as directly ‘to’ them or as ‘with regard to’ them. So, when we say something like we have duties ‘to’ animals, we know what people mean even if most people also think that there’s something a little strange, off, or not quite right about these ways of talking. According to Kant, these ways of talking make sense to us because they rest on this ambiguity concerning how we talk about duties, namely how we customarily do not draw the distinction between having duties to and having duties with regard to. If we speak precisely, however, then, Kant continues, we should draw this distinction to capture how non-human animals are incapable of morality—or ‘of obligation (active or passive)’—which is why they cannot enter into moral relations (MM 6: 442, cf. 443). Consequently, strictly speaking, the correct way of describing these duties is to say that human beings have duties with regard to animals, but that the morality of these duties—the moral ought—is directed to the humans themselves. Again, this does not mean that we do not have moral obligations with regard to animals, but only that the direct moral obligation is to ourselves—not to the animals. In sum, animals have the power to decide (arbitrium) but it is determined by an ‘is’ as comparative (associative) evaluations, whereas humans’ power to decide involves comparative (abstract conceptual) evaluations that are subject to an ought, and so presuppose transcendental freedom.

3 I take myself to be following Allais’s analysis of animals in her (2015), which is of course neither to claim that my interpretation of her is correct nor to deny that there may be several Kantian and non-Kantian ways of making these kinds of arguments regarding the differences between human and animal consciousness. For an alternative, though in my view not incompatible, way of bringing out Kant’s account of animal consciousness and choice, see Patrick Kain’s excellent (2010).

160

Helga Varden

8.3 Animals’ Social Natures After having drawn this distinction between duties ‘to’ and duties ‘with regard to’ in the ‘episodic’ section of the Doctrine of Virtue, Kant seemingly continues by arguing that this direct moral duty to ourselves concerns our duty not to destroy our ability to be responsive to animal suffering. More specifically, by acting cruelly towards animals, a person ‘dulls his shared feeling of their suffering and so weakens and gradually uproots a natural predisposition that is very serviceable to morality in one’s relations with other people’ (MM 6: 443). In other words, it seems that the reason why we shouldn’t act cruelly towards non-human animals is because doing so undermines our ability to act morally towards other human beings. The reason we should be nice to animals therefore appears purely instrumental in nature: being nice to animals is good for us (human beings) in our interactions with each other. Kant’s approach appears thoroughly anthropocentric in a problematic, instrumentalist sense.4 The main objection is to the specific, instrumental way in which the account is anthropocentric: the account does not appear to get the concern for beautiful nature and non-human animals into focus in the right kind of way. The reason we shouldn’t do these things is not just that it undermines human morality, since the badness of what we’re doing goes beyond this. In short, the worry is that Kant’s conclusion (that we should care for, and not torture or mistreat, animals) is the right one, but the reason he gives is the wrong one; we should care about them because they matter for their own sake. I believe this line of objection is wrongheaded and rests on a mistaken interpretation of what Kant is saying about why we should care for animals. It’s true that his account of these moral duties is anthropocentric in a certain sense, but not in the problematic instrumental sense the above interpretation assumes. Animals do have value in themselves, but we easily confuse the way we value them with moral valuing.5 Analogous to how there is (as mentioned above) an amphiboly with regard to how we talk about duties and animals (duties ‘to’ and 4 Kant seemingly proceeds in this text by listing several examples of how humans should and should not treat animals because of how such interactions have the potential of being destructive of, or undermining of, our capacity for morality: e.g. (MM 6: 443, cf. TP 8: 294, CL 27: 458–9). For more on these apparent troubles, see Timmermann (2005). 5 Although considerations of space prohibit me from going into details here, note that this discussion ties in with the discussion in the Kant literature on the value of happiness in human lives. Kant also describes our moral duty to happiness as ‘indirect’ in nature (GW 4: 399), which leads many Kantians to conclude that happiness (for Kant, at least) also is something we should be concerned about because it makes it subjectively easier for us to be moral. Of course, I don’t disagree that happiness makes it subjectively easier for us to be moral; rather, as I argue elsewhere (Varden 2020), concerns rooted in happiness can also not be thought of as being subject to our control or characterized as involving respect, but that does not mean that concerns of happiness are not also valuable in themselves. This is why Kant argues that the aim of a good human life is not to rid ourselves of concerns grounded in our animalistic and social natures (our happiness); rather, he argues, the ‘highest good possible in the world . . . consists in the union and harmony of . . . human morality . . . and human happiness.’ (TP 8: 279)

Kant and Moral Responsibility for Animals

161

duties ‘with regard to’), there is an amphiboly with regard to how we talk about respect and animals. Strictly speaking, we do not respect animals—as respect is a reflective feeling we have internal to a relation between two self-reflective beings—but we do love, care for, and even admire animals. As we shall see in Section 8.5, we do not need to change the structure of Kant’s account to explain why we do not have a moral right to torture animals or subject animals in our care to torturous conditions. Rather we need to add to Kant’s ‘episodic’ section arguments we find elsewhere in his works, in particular his account of human nature in the Religion, as they suffice to explain why we have the moral attitudes we do towards animals, without it being the case that those considerations are themselves the grounds of moral rights for animals.

8.4 Kant’s Account of Human Nature Kant’s complete practical philosophy is threefold. It consists of: (i) an analysis of freedom (comprised of two distinct lenses: that of right,6 or justice, and that of virtue, or ethics), which is restricted to moral (and, so, human) beings; (ii) a normative account of human nature, some elements of which are teleological and social and some of which are viewed as shared with non-human animals; and (iii) an empirical, scientific analysis of everything appearing spatiotemporally (including de facto inclinations). To understand how Kant analyses something like animals and our proper relationship to them, therefore, we must incorporate Kant’s normative writings on human nature, including as they concern our embodied, social (and so affectionate) being, many aspects of which we share with non-human animals. Since this account is crucial to understand why Kant’s take on our duties with regard to animals is not anthropocentric in the problematic, instrumental sense encountered above, sketching the relevant features of it is the main purpose of this section. And to do this, I draw mainly upon Kant’s account of human nature in the Religion. In the Religion, Kant proposes that there is a threefold ‘original predisposition to good in human nature’, namely one comprising the predispositions to ‘animality’, to ‘humanity’, and to ‘personality’ (R 6: 26). Concerning the first, the predisposition to animality, Kant clarifies that it can be characterized as ‘physical or merely mechanical self-love, i.e. a love for which reason is not required’ as well as being (also) threefold: first, it includes the natural drives for ‘self-preservation;’ second, ‘the propagation of the species, through the sexual drive’, including caretaking of one’s offspring; and third, ‘community with other human beings, i.e. the social drive’ (R 6: 26). Concerning the second, the predisposition to humanity, 6 Note that for reasons of space, I do not fully engage the question of our legal duties with regard to animals in this paper.

162

Helga Varden

Kant says that it can be characterized as ‘self-love which is physical and yet involves comparison (for which reason is required) . . . Out of this self-love originates the inclination to gain worth in the opinion of others, originally, of course, merely equal worth’ (R 6: 27), and later adds that the reason in question here is ‘indeed practical, but . . . subservient to other incentives’ (R 6: 28). I will call this type of love ‘reciprocal self-love’ for ease of reference. Also the type of self-love corresponding to the third predisposition (to personality) is not given a name by Kant in the Religion, but in the second Critique he calls it ‘rational self-love’ (CPrR 5: 73). Here in the Religion Kant simply describes this predisposition as ‘the susceptibility to respect for the moral law as of itself a sufficient incentive to the power of choice. This susceptibility to simple respect for the moral law within us would thus be the moral feeling . . . [as the] incentive of the power of choice’ (R 6: 27). After briefly describing these three predispositions to good in human nature, Kant says: All these predispositions in the human being are not only (negatively) good (they do not resist the moral law) but they are also predispositions to the good (they demand compliance with it). They are original, for they belong to the possibility of human nature. The human being can indeed use the first two inappropriately, but cannot eradicate either of the two. (R 6: 28)

Although we are of course conscious in all we do, there are some aspects of us as emotional, embodied beings whose origins must be thought of as belonging to our sensible nature; they concern natural, unreflective ways in which we love or care for ourselves and each other. Those predispositions we have by means of which we care for ourselves as animals include both things relating to our drive to survive, as individuals and as a species, as well as a basic social drive. These drives can be described as ‘mechanical’ in that, again, although we are of course conscious of everything we do and as we develop, transform, and integrate them into our lives by means of our capacity for abstract thought, these ways of being do not, as such, necessarily involve or require reason; they are unreflective, affective ways in which we (and other animals) are naturally oriented by means of natural and social drives. They are ‘original’ and ‘necessary’, as we saw Kant argues, as well as ‘good’ for us, as the embodied beings we are. They are original and necessary for the human being because they are part of what makes the human being—in its fullest sense—possible. And they are good both in the negative sense that they do not resist morality’s demands and in the positive sense that they push us towards complying with morality’s demands. Insofar as we humans have good, healthy upbringings, therefore, we develop a basic comfortableness with both the physical and social world, a background assumed as fundamentally unproblematic as we learn to set and pursue ends of

Kant and Moral Responsibility for Animals

163

our own (including together with others) responsibly in it. In addition, I believe it is most plausible to locate aspects of our own personal life as grounded at this animalistic level, that is, in the particular persons who are our loved ones: in our families, in our friends, and, as Kant himself says, for some, in one’s (social) animals (such as one’s horses and dogs). With these people and these highly social, affectionate animals with whom we identify, we share a grounding ‘us’: we make each other feel at home in the world.7 The second natural predisposition (to humanity) is the predisposition that makes us susceptible to how others perceive us, or to being affirmed by others as equally important or as ‘worthy’ of attention, and vice versa. This type of social predisposition requires reason in that it requires awareness of being seen or valued by others as well as awareness of the importance for others that we see and value them (reciprocally) as equals. As Kant emphasizes, if these comparisons are emotionally healthy they involve comparing oneself with another person or group as having equal worth, and even the competition it draws us towards is good for us—it propels us forward as social and cultural beings. All of these first-level and second-level non-moral relations (those enabled by our predispositions to animality and humanity) constitute part of the sense of self each of us has; they are part of what makes us the particular human beings we are, although Kant believes that the second one requires a lower-level use of reason and hence will not be available to beings incapable of this.8 They also comprise the material that morality sometimes restricts, which brings us to the third predisposition, namely ‘personality’, which enables us to be moved to act merely because it is the right thing to do; it enables us to be motivated by pure reason when we act. Notice that on this account, there is nothing wrong with acting as motivated by sensibility, such as out of affection, in most circumstances; in fact, doing so simply reveals that we are emotionally healthy human beings. Most of the time, what I do for my loved ones, for example, I do just because I love them—and there is nothing about morality that hinders me in this; in fact, morality affirms it. As I am letting those I love be emotionally oriented towards me and take part in my

7 One way to bring out these points is by invoking Simone de Beauvoir’s analysis in The Second Sex, developed in part by exploring writings in the psychoanalytic tradition, of our experiences as prerational beings, including as newborn babies, as infants, and as small children (de Beauvoir 2011). I attend to some of the other similarities between Kant and de Beauvoir in Varden (2017, 2020). 8 It seems consistent to argue that those animals who pass the mirror test—such as elephants and dolphins—may display more complex behaviours and emotions than those that do not because they are aware of being seen by others. For example, maybe these animals are susceptible to emotions that this awareness enables us to have, such as jealousy, envy, and shame. I presume what they might not be able to do is to display either moral correction of these behaviours or moralized versions of such emotions. This would fit with how Kant says of the elephant in one of his lecture notes that it is ‘an Analogon of Morality’ (quoted in Kain, 2010: 218).

164

Helga Varden

life in this way, I’m oriented in the same way towards them. It is therefore not just that the morality ‘permits this’, but it affirms it: personal love for another human being is to love them in this affectionate, affirming way, to be so directed towards them. And of course, this is what we most want from our loved ones, and what we get when things go well. Morality only comes in when I’m getting worried that I might be a little off in what I’m doing—when my conscience alerts me that perhaps something is not quite right. Leaving controversies and various possible problems this argument may have to the side, what does it entail for how we interpret what Kant says about our indirect duties to animals in the Doctrine of Virtue; how does this explain why the troublesome instrumentalist, anthropocentric reading above is not quite right? The simple answer is of course not to deny that developing these natural predispositions is instrumentally useful to morality—after all, morality is supported by them since having developed the susceptibilities these natural predispositions in the right ways makes it much easier to be moral. The emotionally healthy, morally mature human being moves easily between unreflective (say, someone playing affectionately with her dog) and reflective (thinking about how wonderful this interaction is) ways of being; indeed, for such a person the suggestion that she is playing with her dog in order to improve her morality will immediately strike her as weird or foreign. Of course, she will affirm, upon reflection, that it is the case that such play is good for her (emotionally), but this is not why she is doing it; she’s doing it because she loves, and loves playing with, her dog. And Kant is not denying this. Rather, Kant’s main point is rather that duty—the moral ought— tracks freedom or the ability we have to be morally responsible in virtue of acting on our reason; when playing with her dog, the human being is the only one capable of moral responsibility for the interaction, including knowing that perhaps today, although she doesn’t really feel like it, she ought to play with her dog (because, for example, doing so is important for the dog). The duties regarding animals are therefore self-referential (anthropocentric) in that they track what we must do in response to our own reason when interacting with animals. Let me put this last point from two different directions, first from the point of view of affection: insofar as our natural dispositions are healthy, we will not want treat animals badly, let alone cruelly. We just will not want to do this, since it goes against what we naturally are affectionate or kindly disposed towards (given the natural predispositions we have). When we do act in bad ways, however, that is the time that morality kicks in for us, and we feel an obligation to stop and think about what we are doing (moral feeling). The source of this obligation is our own reason. It is our reason that makes it possible for us to be aware that perhaps we can no longer simply trust our sensibilities and it is our susceptibility to be moved by our reason that makes it possible for us to act otherwise than our current inclinations direct. And, so, our direct moral duties are to ourselves, as morally responsible beings, namely as embodied sentient rational beings who can and

Kant and Moral Responsibility for Animals

165

must (have an obligation to) act on and interact with the world in accordance with our moral, reflective standards, including insofar as these actions and interactions engage embodied beings which do not appear to have such capacities for moral responsibility. Centrally, we do not experience respect in our interactions with animals. Respect, strictly speaking, is a moral, and so reflective emotion. It is an emotion internal to relations between reflectively self-conscious beings. One way to get at this may be to point to how social beings, such as cats and dogs, do not comfortably meet our direct gaze, or seek moments where we ‘rest’ in or understand each other by looking each other in the eyes. Humans, on the other hand, cannot only look each other in the eyes to affirm our affection and appreciation for one another, but can look each other in the eyes to affirm our respect for each other. And it is our capacity for reflective self-consciousness that enables us to do this last thing; I can (and do) seek another’s eyes not only when I want to show my expectation of moral responsibility to that person, but also to seek or to affirm my respect from or for them (CPrR 5: 76). Still, that we do not experience the moral feeling of respect with regard to animals does not mean that we don’t value them; it’s just that the specific moral attitude we have towards human beings is absent in our relations to non-human animals (since they are not capable of reflective selfconsciousness). And there’s nothing wrong about that; there are other, important kinds of valuing than moral valuing in a good life. Although it is true that doing torturous things to animals is also to damage ourselves (to ‘uproot or weaken the natural predispositions in us’) and we have a direct duty to ourselves not to do this, we also have a direct duty to ourselves to act in relation to the animate world in a morally responsible way. To illustrate by means of Kant’s own examples, to kill animals quickly (if one kills animals), not to subject animals to agonizing experiments for the sake of speculation, to show gratitude to old horses and dogs for long service (as members of the household), and so forth are ways in which we relate morally responsibly to animals. Since, however, the directedness of the moral call is to ourselves (both the calls not to damage ourselves by how we act in nature and with regard to assume a moral perspective with regard to non-human nature), the duties with regard to nonhuman nature are indirect in nature. It’s not only the conclusion that is right, but also the reasoning behind it: moral duties are always directed at moral agents, including when the duties in question are with regard to non-human nature or non-moral beings. And, again, although it seems incorrect to describe relations themselves with non-human beings by means of moral concepts, or concepts tracking rational beings, such as rights and respect, to act appropriately in our relationships with non-human animals often involves holding attitudes such as admiration, concern, care, and affectionate love. If the above is correct, then we can also see why children and all of us, insofar as our capacities for reason are temporarily or permanently damaged, do have

166

Helga Varden

rights, even though non-human animals do not have rights. Our care for other humans is fundamentally informed by all three considerations in our aim to treat the one cared for in line with all three at all times and with the aim that the person can restore or realize as much as possible of all three predispositions, hopefully all the way to full moral responsibility (again).9 Let me restate this last point by addressing a common worry in the literature on animals, namely that the only way to give all human beings proper moral considerations is by also granting the same consideration to at least higher cognitive, social non-human animals. For example, some argue that attributing rights only to humans is mistaken because any morally salient feature that only humans exhibit—such as practical reason—is not exhibited by all humans, and since the (allegedly) morally salient feature that all humans share—such as sentience (Singer)10 or being the subject of a life (Regan)11—is also exhibited by many nonhuman animals. Hence, if it is having this feature that grounds moral consideration, then not only humans end up with moral consideration. In addition to taking on these kinds of objections, to make my case, the position will also need to be able to respond properly to the opposite extreme type of position found in the literature, namely those who argue that it is impossible to justify any legal protection of animals beyond what is necessary to protect human health. This extreme is exemplified by many (especially right-wing) libertarians, who characteristically deny not only any notion of animal rights, but also any notion of special, legal rights—so-called care rights—for human beings incapable exercising rights. Consequently, not only do animals end up without rights or without any special legal protection beyond what is necessary to protect human beings’ health, but so do human beings whose incapacitation is so severe as to render them unable to utilize or develop their capacities for practical reason to the extent needed to exercise their rights. As one might suspect at this point, I am convinced by neither of these extreme types of accounts nor by the way in which the anti-speciesist intuitions are used to support the idea that animals have rights or moral standing equal with at least some humans. Something goes wrong in both lines of reasoning and the supplied intuitions show less than what they are taken to do here. But what is it that has gone wrong? To start, a problematic, underlying assumption informing any nonpractical reason account is that if we grant animals the kind of social, conscious nature I have outlined in this first part of the paper, then we have thereby given good reason to think that they have moral rights. But this presumption is false. The account of the social nature we share with animals is not the basis we should use to analyse human beings’ moral rights and duties to care. Rather, our social, conscious nature is concerned with certain unreflective, yet normative preconditions 9 For Kant’s legal analysis of legal guardianship, see Varden (2012). 11 Regan (1983).

10 Singer (1975).

Kant and Moral Responsibility for Animals

167

for animal and human choice, as explained above. The social account of animals also in the case of human relations is not what gives human being moral rights, including rights to care. Instead, this social account is part of a fuller account of human being that captures, for example, the place of affect and loving relations of particular others—for human and non-human animals—and it is part of the explanation of why human being have direct, moral duties to themselves and indirect duties to non-human beings not to do these things. Hence, this account of our social being is a part of what a moral theory must accommodate in the right way; it is part of a full practical account of normativity, but it is not itself a moral account or an account of freedom—not for human beings and not for other animals. This also applies to the case of saving pets over strangers, though I believe that it may be plausible to apply the idea of viewing one’s relationship to animals as possibly involving existential, sometimes religious sentiments as relevant to a complete analysis here. To start, the account of animals’ social nature outlined above can explain why some might feel psychologically compelled to save their pets first, namely because of the close loving relation that exists between them and the specific role the pet plays in these persons’ life. In light of this, we might think that our moral analysis must make space for this: perhaps we would and should withhold our judgment here and let this be one of those issues that each person must decide for oneself, on the basis of what one’s actions reveal one can do and can live with. But I don’t find this quite convincing unless, on this analysis, the major reason we remain silent on this issue is because there is something tragic about the scenario that gives us the reason to remain silent, namely that a pet has such a stabilizing role to play in a particular person’s life. After all, making the choice of saving one’s pet includes having to deal with meeting (or thinking about) the stranger’s relations, including in light of the fact that the devastating, existential effect of losing one’s relation invokes a different category of loss and grief than losing one’s pets. It seems to me, however, that the better analysis may be along the abovementioned, existential (sometimes religious) lines. To get at this idea, the following analogy might be useful: even if one doesn’t understand it first-personally, one does recognize and respect that some people will not engage in violence regardless of what happens to them or their loved ones; they just will not do it. On my analysis above, the moving ground (reason) here may be considered existential in nature; for these, this has become a question profoundly about how they need to go about life to find it meaningful and good. I believe one’s relations to animals may be similar in the following sense: some people cannot bring themselves to place any kind of life above any other, and so they cannot imagine—in the sense of honestly, clearly feel it to be true about themselves—that what they will do in emergency situations of these kinds will always favour saving humans over nonhumans. It’s just not how they need to be in the world to be who they are and affirm the world as good in this profound way; they would rather risk it all than

168

Helga Varden

abandon this part of themselves. And maybe there’s something emotionally healthy about not wanting to reflect upon, endlessly, various hypothetical scenarios of such extreme emergency situations and instead just be content with the fact that these deep, unreflective existential, emotional orientations are what will ultimately be revealed in what we actually do in such situations. Insofar as we are emotionally grounded in a good way, it may be wiser to trust that we will know what to do—what this is such that we can live with it—if we ever have to face such horrible situations. Regardless of how we go on this issue, notice that because no citizen (as a private citizen) can be under a legal obligation to save strangers in emergency situations, let alone risk their lives to do so, we might want in any case to be silent about this as a matter of law (in the sense of what private persons may be legally required to do). Nevertheless, and for reasons that will become clearer below, public officials will still have to save humans first. The reason is simply that as a representative of the public authority (public official) one is entrusted to act on behalf of the citizens (and other residents on the territory), including by protecting and saving them when possible. And since the issue of how we analyse these suitably described emergency situations does not justify universal moral nonspeciesism and animal rights, acting as entrusted in this way necessarily involves saving humans first. Moreover, because these situations are so complicated and involve extremely difficult situations that public officials are under an obligation to act in—that’s their job—we do not want to wait here to specify rules helping public officials in these situations, and we want training to prepare the officials for them and procedures (including possible sessions with expert psychologists) to help them live with what they have had to do afterwards.12 And, of course, regardless, the individuals involved will have to live with the decisions they make in these emergency situations, and the price of getting it wrong—of living life with having made grave mistakes in these regards—is probably harder for good human beings than what any state punishment could possibly be. This might plausibly be one reason why private individuals don’t get punished for what they do, whereas the only punishment for public officials failing (such as losing courage) typically is that they lose their job, meaning that they cannot be entrusted any longer to hold the position of being such a public official. Also, notice that regardless of our conclusions regarding saving pets and animals, it plainly seems to be a mistake to treat our relations to non-animal being and incapacitated human beings alike. Any human being who needs others to take care of them—obviously all of us insofar as we are children, mentally disabled, mentally sick, or severely physically ill—finds themselves in situations of

12 I’m grateful for many discussions with Ekow N. Yankah on this point.

Kant and Moral Responsibility for Animals

169

inequality with their particular caregiver(s). When the person for whom we care is someone we love, we strive to provide loving, caring surroundings, that is conditions that are not only affectionate, intimate, playful conditions but objectively caring, safe, and respectful. Insofar as the care-receiver is capable of affectionate love and of some moral responsibility, she will do her best to do the same to the caregiver—be affectionately loving, caring, and respectful—but because of the dependency involved, the relation and the extent to which the people involved can do these things is not reciprocal, but asymmetrical. Ultimately, moreover, the caregiver’s aim is always to provide conditions consistent with the person cared for regaining or developing her own free agency insofar as possible. The fact that a particular person is currently or permanently incapacitated does not mean that she is no longer the same kind of being as the rest of humanity; indeed, characterizing someone as impaired or incapacitated, for example, is already to say that he or she is of a particular kind, but that her capacities are not operating, realized, or realizable so as to enable full functioning, at least not right now. Let me point out two important consequences of this last point: first, the tragedy of a terrible accident that significantly destroys one’s mental or physical capacities and so brings someone down to an actual functioning-level similar to that of a less complex animal is not that it’s tragic to be such a different animal, but it’s tragic for a human being to suddenly find themselves with such reduced abilities to exercise their capacities. Refusing to recognize, acknowledge, or appreciate the tragedy for someone who finds themselves or their loved ones in such a situation is plainly extraordinarily insensitive (revealing significant moral immaturity and/or interpersonal clumsiness). The only worse mistake, on this Kantian analysis, is to think (or say) that a human being after the accident (or any other human being that is incapacitated temporarily or permanently in any way) has lesser moral worth than a non-incapacitated human being, that somehow the accident could take away from this person her or his ‘dignity (an absolute inner worth) by which [they exact] . . . respect for [them]self from all other rational beings in the world’ (MM 6: 435). Second, if it is the case that someone is not able to feel sympathy (feeling pain at another sentient being’s suffering)—a person who is partially or absolutely emotionally flat in this regard—then there is nothing natural already there in support of what one’s practical reason demands of oneself. This kind of social disability, the lack of at least part of one of the natural predispositions (to animality) entails neither that this person should not be treated with respect nor that this person cannot act in accordance with and as motivated by their practical reason. The standard also for such a socially disabled person is the moral standard, which is available to them due to their ability engage the world as a reflective, selfconscious being. It’s simply that it’s much harder for such a socially disabled person to do what is right, since they always have to use (only, in severe cases) their capacities for self-reflective reason to figure out how to act appropriately in

170

Helga Varden

various situations; their affective and social emotional life is not of much or any (again, depending on how severe the disability is) assistance in this regard, and so they much more easily get it wrong and end up in trouble (is perceived as rude, insensitive, or very clumsy) in their social interactions. And correspondingly, of course, this is also why a parent of a child with such a social disability is responsible for teaching the child how to manage social interactions that are naturally difficult for them to master. It is also why everyone who does not struggle with such disabilities is morally required to take the time it takes, when necessary and asked, to explain the social-emotional dynamics that someone’s social disability makes it very hard to gain access to first-personally. Finally, given the analysis above, including how it maintains that we can have affectionate, loving relations with other animals, we might ask whether (ethically) we ought not to kill and eat animals at all, whether we may be stopped from killing and eating them as such, and whether killing and eating them should even be illegal. For example, we may ask with Christine Korsgaard: although non-human animals do not consider killing wrong, isn’t it the case that we should, since we can uphold a higher, non-predatory standard?13 In light of what I’ve argued above, killing or not killing non-human animals or eating or not eating nonhuman meat as such is not immoral. In addition, my suggestion is that this is one of those questions that appear better understood as a deep existential question. It is a question the answer to which depends on one’s basic, existential openness or sensibility towards the world as a whole and as a good place. As history and contemporary differences between people’s related, existential sensibilities show us, for some, killing an animal is incompatible with feeling oneself as participating in a world that is fundamentally a good place, one that fills one with profound peace, awe, and wonder. For others, we are part of a world where animals kill and eat each other, and being an animal in this sense is not, as such, experienced as existentially troublesome. Given the position I have been defending here, neither kind of people or persons is more or less ethical than the other; they are simply profoundly different in their related existential (sometimes as explicated through religious doctrines) sensibilities. Rather, what all can agree on, as a matter of morality, is that animals should be treated well. That is, we must treat all animals well in their lived lives (as the beings they are), and insofar as possible, and if one kills them, this should be done as quickly and painlessly as possible; it should be less painful and scary to be killed by a human animal than by a nonhuman animal.

13 Correspondingly, and as we well known, Korsgaard argues (against Kant) that we ought not to kill nor eat animals (Korsgaard  2014). The main differences between my analysis here and that of Korsgaard concern not only these philosophical points, but also the way in which I use Kant’s account of human nature in my analysis.

Kant and Moral Responsibility for Animals

171

8.5 Torturing Animals At this point, it may be useful to return to a question that I have only briefly mentioned above, namely why do we ever get any of this wrong? That is to say, if we have the threefold natural predisposition to good—to animality, to humanity, and to personality—why do we ever do anything wrong? If these predispositions to good really are good (pushing us towards morality and being affirmable by our reason) and original (constitutive of the human kind), how could or would we ever want to do anything wrong? Why, that is, does our pure reason (rational selflove) ever need to infringe or restrict the natural, social forms of self-love (enabled by the predispositions to animality and humanity)? Although this is not the place to engage this large topic fully, in short, the reason is the fact of a ‘human being’s power of choice as a moral being’ in relation to our various desires (R 6: 31). That is, as human beings, we can choose in relation to our desires and hence we may, and will inevitably, sometimes choose badly in relation to which inclinations (habitual desires) we develop. Kant calls the leaning we have towards being tempted to develop bad habits through our choices a propensity to evil, and describes this propensity as: the subjective ground of the possibility of an inclination (habitual desire) . . . insofar as this possibility is contingent for humanity in general. This is distinguished from a predisposition in that a propensity can indeed by innate yet may be represented as not being such: it can rather be thought of (if it is good) as acquired, or (if evil) as brought by the human being upon himself. (R 6: 29)

Inclinations as such, therefore, are not the problem. The problem is our ability to choose and develop bad ones: because we can choose ends (we can step back from any inclinations and set and pursue ends of our own), we are necessarily tempted to choose to do what we ought not to do—to choose just because we can or because doing so gives us the most pleasure in the moment or that it will increase our sensation of pleasure. We are, that is to say, tempted to exercise our freedom in choosing ends that are not truly good for us and which are not affirmable upon reflection as reconcilable with respect for ourselves and one another (the moral law). Indeed, we are tempted to do whatever we want to do where this is viewed as choosing for choice’s sake (the thrill involved in being able to choose whatever we like, or power) or those ends that give us the most intense pleasures and senses of ourselves (in the short or long term). These temptations are always present for us as embodied (natural, social) beings, who can choose freely in the sense of setting and pursuing ends of our own in an open-ended kind of way. We can choose to act for or against the moral law since we decide which maxims to act on; we do not simply act as our practical

172

Helga Varden

rational will directs (R 6: 31). Kant calls this fact about us as beings capable of making bad, including patterned bad, choices ‘the propensity to evil in human nature’ (R 6: 29). The propensity to evil concerns the fact that we can and do feel the pull of choosing otherwise than our practical rational will directs; we can choose in other than morally justifiable ways. Obviously, what we might choose, what is tempting, is still determined by what we can feel as pleasant, namely inclinations available to us as natural, social beings capable of setting ends of our own, but they are now chosen for their own sake—in attempts at maximizing them—rather than as and insofar as they are morally unproblematic or affirmable upon reflection. Kant continues by proposing that the propensity to evil comes in three degrees, or grades: First, it is the general weakness of the human heart in complying with the adopted maxims, or the frailty of human nature; second, the propensity to adulterate moral incentives with immoral ones (even when it is done with good intention, and under maxims of the good), i.e. impurity; third, the propensity to adopt evil maxims, i.e. the depravity of human nature, or of the human heart . . . [This] can also be called the perversity . . . of the human heart since it reverses the ethical order as regards the incentives of a free power of choice. (6: 30)

There are, in other words, three grades to such wrongdoing: first, there are those situations where we know what is right, but we don’t do it (the ‘frailty of human nature’). To use Kant’s examples: we know we shouldn’t destroy beautiful crystals or not respond to suffering (whether from human or non-human animals), but sometimes we don’t act or respond appropriately nevertheless. Or to illustrate with regard to the first two natural predispositions (to animality and humanity), respectively: we know we should not, for example, eat or drink too much, but we might do it anyway, and I know I should fight any jealous feelings I have (since they’re groundless), but instead I yield to them. Second, sometimes we act with an ‘impure’ heart, meaning that we do what is right (we don’t destroy nature and we keep tending to suffering and pain), but we do so with immoral incentives, such as wanting to look good in the eyes of others rather than, ultimately, because it is the good and/or right thing to do.14 Acting in this way is worse than frailty because it involves a patterned instability in how we value; it is generally hard for us to be moved to act in appropriate ways, for the right reasons. Third, we can start acting on evil maxims in self-deceptive ways, that is, we start doing wrong in

14 This does not entail that we cannot, for example, act out of affectionate love but must always aim to act out of duty. Rather, when so acting is appropriate, then, as explained in earlier sections of this paper, the actions are good and so have value (and would be affirmed as such upon reflection), and they neither involve ‘adulterating’ moral maxims nor do they problematically lack moral worth.

Kant and Moral Responsibility for Animals

173

the name of the good. Hence, whatever gives us or will give us the strongest inclinations and senses of power is what we do, but we describe what we do by means of moral language, such as ‘giving people what they deserve’ or as doing heroic things when we dehumanize others. This is the worst kind of wrongdoing, and doing so is to be at war with one’s own ability to act morally responsibly, or should be understood as attacking one’s own personality or what Kant also calls realizing a ‘depraved’ heart or acting with a ‘perverted’ heart. Let me explain this last point, about the depraved or perverted heart, from a somewhat different direction, namely in a way that brings out the importance of making the full analysis of destructive or violent treatment of inanimate or animate nature attentive to both the ways our natural predispositions are non-moral, yet normative in nature as well as the crucial role our parents and families and loved ones play with regard to developing and sustaining these in the natural, emotionally healthy, or good ways. Kant says of the vices associated with depraved hearts that ‘the vices that are grafted upon this inclination [to gain worth in the opinion of others] . . . can be named vices of culture, and in their extreme degree of malignancy (where they are simply the idea of a maximum of evil that surpasses humanity), e.g. in envy, ingratitude, joy in others’ misfortunes, etc., they are called diabolical vices’. (6: 27) Obviously Kant’s point here cannot be that one has a depraved heart if one is a little envious, feels insufficient gratitude, or sometimes takes joy in somebody else’s misfortunes—after all, Kant is neither after the weakness nor the impurity of the human heart. What he is after is the depravity or the perversity of the human heart, our worst sides. My suggestion, rather, is that the depravity or perversity of the human heart occurs when these twisted ways in which we can relate to how others view us (envy, ingratitude, joy in others’ misfortunes) fundamentally ground how we go about our lives and yet it is done in the name of the good. Those who develop depraved hearts will take real pleasure, in the sense of feeling really alive and empowered, only if others are made to suffer around them and they will feel unimportant or neglected if those around them are not oriented towards them at all times. Such people will be aggressive and destructive, and not until they realize that their descriptions of their actions are deeply self-deceptive—and so they are actually acting in destructive and aggressive ways—can they change. In addition, of course, some people turn this destructiveness directly upon themselves, either by self-mutilation or by seeking people who violently affirm their own sense of worthlessness.15 In all cases, the fundamental problem is that such people lack an emotionally healthy sense of self, the kind of fundamentally stable self that is

15 A lack of self can manifest itself non-violently in simply seeking to live through others, or violently, in various types of aggressive self-absorption – see Tom Hill’s ‘Servility and Self-Respect’, in Hill (1991: 4–19). I have applied this type of analysis in my paper on the terrorist attacks in Norway in Varden (2014).

174

Helga Varden

related to the second, comparative natural predispositions to good—to humanity; although because of the way in which the two predispositions (to animality and humanity) are connected, such self- and other destruction and aggression reflect that the sense of self enabled by the first predisposition (animality) is unstable or lacking too—and hence such (self-)aggression further undermines or destabilizes it. Still, regardless of which kind of wrongdoing we are liable to, because the root of evil is our frail nature due to the coexistence of inclinations and the fact of free choices, and not our rational will, ‘it must . . . be possible to overcome . . . evil’ (6: 37). Because the root of evil is not our rational will, but a corruption of our ability to act in accordance with it—a corruption that we have brought upon ourselves through our capacity to choose ends of our own—we can always overcome our liability to do wrong and, instead, do the right thing and heal our troubled, self- and other-destructive selves; we are morally responsible for our bad actions and for working on these aspects of our characters. Moreover, if we pay attention to the role the people we love and care for enabling us to develop and sustain these first two natural predispositions to good, then it follows that these kinds of unhealthy or unstable selves are typically related to having been subjected to abuse, including damaging neglect. If one has been so subjected, whether or not one is able to develop a fundamentally healthy or stable self, how hard the route to healing is depends upon to what extent one has learned to protect oneself against this and yet remained open to the world and other, non-abusive people in the right kinds of way. On this position, one clear sign that one is doing very poorly is, I believe, if one experiences oneself as being thrilled, happy, or excited when one is doing bad things to others, let alone if one is starting to use moral language to describe these actions; such normative facts, on this analysis, indicate that things are heading in the ‘perverted heart’ direction. Instead of taking pleasure in the right kinds of things—things reason will affirm upon reflection—not only does one fail to seek good pleasures and let reason guide one when one is worried, but one uses one’s ability to think about things so as to pursue and increase these kinds of pleasures, including by describing them in moralized ways. Whether one does these bad things to people one loves or to strangers, then, what is lacking is a basic, emotionally healthy sense of self and what one is doing is not only undermining their healthy sense of self, but also attacking or further destroying oneself. Moreover, when one does this to nature—whether as a way of going about aspect(s) of one’s life or in particular actions—one is violating one’s own duties to oneself (by acting in self-damaging ways) and one is failing in one’s indirect duties to non-human animate and inanimate nature, namely by treating them inconsistently with how one’s morality requires one to live in relation to the natural world. And insofar as we share the first preconditions to good with other social animals, like we seem to do in degrees depending on how socially and cognitively complex other animals are, the more torturing them has the same effect:

Kant and Moral Responsibility for Animals

175

it fundamentally disorients them and robs them of their trust in the world as a good world in which they confidently live and play. This is also why the problem with treating animals who share in our social natures is not only to inflict pain on them, but to torture them, to make them suffer. Finally, it seems that the reasons someone may be tempted to do any of this are two: on the one hand, because not tending to the pain may be in one’s self-interest, such as business interest, in which case he or she is not taking pleasure in it (but trying to avoid observing it or learning to numb the natural sympathy one has to others’ suffering) or because she gets a heightened sense of self from doing it. Such a heightened sense of self is possible because humans have the second predisposition—to humanity—and hence can get a sense of satisfaction, a sense of power and of being alive, by terrifying another social being with a power of choice. And it seems plausible that it is because many such troubled souls so often do not feel satisfied with the terror beings without reflective self-consciousness can experience by such treatment that they so often also abuse human beings.

9

What Do We Owe to Animals? Kant on Non-Intrinsic Value Carol Hay

9.1 Introduction In this chapter, I defend the ostensibly morally distasteful claim that animals, along with all other non-rational beings, have only non-intrinsic value. Things that are intrinsically valuable are valuable in themselves, while things that are non-intrinsically valuable are valuable only because someone actually values them. With Kant, I argue that intrinsic value is necessarily connected to the rational ability people have to value things. Because animals do not have this ability, they cannot have intrinsic value. This means that if animals are to have any value at all, their value must be non-intrinsic. But I argue that, despite their seemingly secondrate moral status, we can construct a surprisingly robust Kantian account of what we owe to animals. What Kant says about what we owe animals is notoriously unsatisfactory: he is usually interpreted as arguing that the only reason we have to avoid cruelty to animals is that such behaviour is likely to harden our characters, making us more likely to behave cruelly towards people (who are the only beings that warrant our genuine ethical regard). This plainly unpalatable story has led most philosophers who have any inclination to take animals’ interests seriously to despair of the possibility of using a Kantian framework to make sense of what we owe to animals. According to the account I defend, we can affirm the basic Kantian story about the loci and sources of both intrinsic and non-intrinsic value without being committed to a variety of morally problematic conclusions about animals. On this picture, we can still say that animals matter morally, that their interests must be taken into account, that they are moral patients or subjects, and that they deserve genuine moral consideration or regard. The deflationist interpretation of Kantianism that I defend here tells a Kantian story at the metaethical level of explaining the nature and source of value, a Kantian story at the normative level of explaining how our behaviour is constrained with respect to intrinsically valuable beings, and a pluralist story at the normative level of explaining how our behaviour is constrained with respect to non-intrinsically valuable beings and Carol Hay, What Do We Owe to Animals? Kant on Non-Intrinsic Value In: Kant and Animals. Edited by: John J. Callanan and Lucy Allais, Oxford University Press (2020). © Carol Hay. DOI: 10.1093/oso/9780198859918.003.0010

What Do We Owe to Animals? Kant on Non-Intrinsic Value

177

things. This interpretation remains neutral on the well-worn metaethical debates about whether Kant is best understood as a constructivist or a realist, instead prioritizing a practical discussion of how the recognition that a thing is nonintrinsically valuable can and should affect our judgements about what behaviour is morally permissible with regard to it.

9. 2 Kant’s Views on Intrinsic and Non-Intrinsic Value Let us begin by laying out an interpretation of Kant’s arguments about how our rational nature gives us humans our distinctive moral value, turning then to consider how Kant distinguishes our value from the value of animals. Kant establishes the value of our rational nature by first having us consider the value of more ordinary objects in the world. When an ordinary thing like a table or a chair is valuable, this is because someone—a being capable of valuing things—actually values that thing (G 4:428). This sort of value is non-intrinsic: if no one valued the thing in question, it would have no value. And because for Kant the relation between value and reason is such that we can usually have a reason to perform an action only if the goal of that action is valuable to us, non-intrinsically valuable things can give us reasons for action only if we happen to value them (G 4: 428). This is what I will call Kant’s anthropogenic argument: humans are the source of all non-intrinsic value because value is generated or conferred exclusively by humans. By contrast, if anything is to give us reasons for action that are independent of what we contingently value—that is, if we can have reason to do something other than if we just happen to want to do it—Kant says there must be something with intrinsic value to provide these reasons (G 4: 428). Whatever this is, it must be valuable for its own sake. In Kant’s words, it must be an end in itself (G 4: 428–9). And Kant asserts that there is something that has this sort of value: the rational capacity to make decisions about what is valuable (G 4: 428). And so, in virtue of our rational capacity to make value judgements, we have this sort of value. Our intrinsic value means that, unlike the non-intrinsically valuable things we value contingently, we have value independently of whether anyone happens to value us (G 4:429). This is what I will call Kant’s anthropocentric argument: humans are the only locus of intrinsic value because if this kind of value is to exist at all it must exist in the beings responsible for all other (non-intrinsic) value. Kant is very clear about how our behaviour is constrained when it comes to beings with intrinsic value. Anything that has intrinsic value must be respected accordingly, Kant argues. Kant attributes to us a very special kind of intrinsic value—what he calls dignity, or absolute worth—in virtue of our rational nature (G 4: 435). Because we are an end in itself we have dignity, Kant argues, while all other ends have only price (MM 6: 462). What has dignity is of unconditional

178

Carol Hay

worth (it is valuable even if no one happens to value it) and incomparable worth (it is literally priceless). And something that is an end in itself may neither be regarded nor treated as a mere means to some other end; this would be to regard or treat it as if its value were non-intrinsic rather than intrinsic. Unlike mere things, we must, in virtue of our rational nature, be respected as ends in ourselves (CPrR 5:131–2). This respect requires an appropriate attitude of reverence towards anyone who possesses rational nature (CPrR 5:161). It also imposes concrete constraints on how we may behave towards such beings. This respect rules out, for example, the permissibility of killing us, or injuring us in ways that would prevent the future exercise of our rational nature. It also requires that others never coerce or manipulate us: forcing or tricking us would be using our rational nature as a means to someone else’s ends. The only way for others permissibly to obtain our cooperation is for them to let us decide matters for ourselves by appealing to our reason. We should therefore be free to set and pursue our own ends and not have ends imposed upon us against our will. But Kant is much less clear about how our behaviour is constrained when it comes to things with non-intrinsic value. Given his central focus on the moral importance of rational nature, he is, ultimately, just not that interested in these sorts of questions. In fact, Kant himself did not think that questions of how nonintrinsically valuable things are to be treated are properly thought of as questions of morality at all. Kant scholars argue that he is committed to what has been called a ‘personification principle’, according to which ‘humanity or rational nature has a moral claim on us only in the person of a being who actually possesses it’ (Wood 1998: 193). This personification principle sets limits on the sorts of beings that can be moral patients; it tells us who or what can have moral duties owed directly towards them. Because Kant connects morality inextricably to rationality, he insists that only rational agents can be moral patients (MM 6: 442). For this reason, he thinks we can have direct moral duties only towards humans; if we have moral duties towards non-intrinsically valuable things these duties will only ever be indirect.

9.3 Kant’s Views on Animals The very little that Kant says explicitly about what he thinks we owe to animals seems to bear this out. He argues that the only reason we have to avoid treating animals cruelly is to avoid becoming the sort of person who is more likely to treat people cruelly: With regard to the animate but nonrational part of creation, violent and cruel treatment of animals is far more intimately opposed to a human being’s duty to himself, and he has a duty to refrain from this; for it dulls his shared

What Do We Owe to Animals? Kant on Non-Intrinsic Value

179

feeling of their suffering and so weakens and gradually uproots a natural predisposition that is very serviceable to morality in one’s relations with other men. (MM 6:443) If a man has his dog shot, because it can no longer earn a living for him, he is by no means in breach of any duty to the dog, since the latter is incapable of judgement, but he thereby damages the kindly and human qualities in himself, which he ought to exercise in virtue of his duties to mankind. Lest he extinguish such qualities, he must already practice a similar kindliness towards animals; for a person who already displays cruelty to animals is also no less hardened towards men. We can already know the human heart, even in regard to animals . . . In England, no butcher, surgeon or doctor serves on the twelve-man jury, because they are already inured to death. (EC 27:459-260)

We do not have direct duties towards animals, Kant thinks, because they are nonrational beings, and nonrational beings are simply not the sorts of things we can have direct duties towards. Our duties to animals are thus more correctly described as duties ‘with regard to’ animals; they are merely indirect duties that follow from the direct duties we have to ourselves to promote our moral perfection by cultivating the right kind of disposition (EC 27:459). From here, many critics understandably go on to assume that this approach commits Kant to thinking that we owe animals no genuine ethical regard.1 They characterize Kant as thinking that the only real reason we have to avoid harming animals is a reason that, at bottom, has to do with a selfish concern to protect our own moral characters. This, they argue, is at odds with the basic moral intuition that the reason it is wrong to harm an animal is not because it wrongs us, but because it wrongs the animal. We are not the subject of moral harm when we make an animal suffer; the animal is. Animals are capable of making legitimate moral claims upon us, they insist, and these claims are what ground our duties toward them. Kant’s insistence on thinking otherwise gets the story of why animals are valuable and how and why we must respond to that value insultingly wrong. These critics charge that Kant’s views are unjustified in both their anthropogentrism and their anthropocentrism. Humans are neither the only source of non-intrinsic value, nor the only locus of intrinsic value, the argument goes, and Kant’s account goes wrong on both fronts. Thus, almost every philosopher who is concerned to take animals’ interests seriously rejects Kant’s account. I want to consider the possibility that this rejection of Kant is perhaps too quick. To begin with, we must pay attention to those places where Kant insists that the indirect duties we owe to animals do not permit us to treat animals as

1 See e.g. Fieldhouse (2004), Regan (2001: 199–204), Singer (1975: 211, 256), Skidmore (2001), Tucker and MacDonald (2004).

180

Carol Hay

mere instruments for our wanton use or plunder.2 Kant has a number of substantive views about how animals may be permissibly treated: he argues that animals may not be destroyed nor misused wantonly or cruelly, and while he does think they may be sold, used for labour, and killed to be used as food, he argues that their labour must be non-excessive and their killing painless. He argues that we ought not to work our animals any harder than we ourselves are willing to work, and that we owe them the same gratitude for their service that we would owe to any other member of our household. He even goes so far as to suggest that we should put a grub back on a leaf after we are finished looking at it, rather than destroying it for no reason. The human being is authorized to kill animals quickly (without pain) and to put them to work that does not strain them beyond their capacities (such work as he himself must submit to). But agonizing physical experiences for the sake of mere speculation, when the end could also be achieved without these, are to be abhorred.—Even gratitude for the long service of an old horse or dog (just as if they were members of the household) belongs indirectly to a human being’s duty with regard to these animals; considered as a direct duty, however, it is always only a duty of the human being to himself. (MM 6:443) If a dog . . . has served his master long and faithfully, that is an analogue of merit; hence I must reward it, and once the dog can serve no longer, must look after him to the end, for I thereby cultivate my duty to humanity, as I am called upon to do . . . Leibniz put the grub he had been observing back on the tree with its leaf, least he should be guilty of doing any harm to it. It upsets a man to destroy such a creature for no reason, and this tenderness is subsequently transferred to man. (EC 27:459)

When he argues that our behaviour towards animals ought to be constrained by the indirect duties we have toward them, Kant does not mean this to be a trivial protection. Indirect duties, after all, are duties nevertheless: they bind us, and constrain our behaviour, just as strongly as direct duties. And while Kant’s explanation for why we are constrained in how we may permissibly treat animals may strike many of his critics as strange, we should recognize that Kant is fairly uncontroversial (and perhaps even more friendly to animals’ interests than many) in how he thinks we are constrained. With this amenability to animals’ interests in mind, let us move forward now to consider a new Kantian account of what we owe to animals.

2 Other Kantians who pursue this strategy include Denis (2000), Korsgaard (2018), O’Neill (1996).

What Do We Owe to Animals? Kant on Non-Intrinsic Value

181

9.4 A Deflationary Interpretation of Kant’s Metaphysics of Value The Kantian interpretation I defend here can be thought of as one voice in a rather small chorus of those who defend the possibility of using Kant to discuss what we owe to animals. This interpretation agrees with Kant in arguing that humans are the only source of non-intrinsic value because value is generated or conferred exclusively by humans. It also agrees with him in arguing that humans are the only locus of intrinsic value because if this kind of value is to exist at all it must exist in the beings responsible for all other value. Where Kant goes wrong, I argue, is in insisting that animals’ non-intrinsic value means we can have only indirect moral duties towards them. Instead, the regard we owe to animals in virtue of their nonintrinsic value should be seen as a genuine moral regard, because non-intrinsic value is perfectly capable of constraining our morally permissible behaviour. Value is non-mysterious on this deflationary interpretation of the Kantian picture: it exists simply because there are beings capable of valuing things. The virtue of understanding the nature of value in this way has to do with both explanatory power and simplicity. This story can make sense of both the source and the locus of value while avoiding commitment to any number of lofty or problematic or controversial metaphysical assumptions. There is, as I like to say, no pixie dust required here to explain why and how things are valuable. This interpretation is deeply consonant with a naturalistic understanding of the world, but, for those so inclined, is also agnostically open to more metaphysically robust worldviews. With the ethical question of what we owe to animals first and foremost in front of us, questions about the nature and source of value become interesting primarily in virtue of their practical, not metaphysical, upshots. So the motivation to endorse this Kantian interpretation is less a desire to affirm a deflationary metaphysics as such than a desire to avoid what are, from a practical point of view, mostly unnecessary metaphysical questions. In what follows, then, we will focus on value primarily in terms of how its recognition constrains our behaviour. What it means for something to be valuable, in practical terms, is that the actions and attitudes we may permissibly have in its regard are constrained in some way. Accounts that interpret Kant as a constructivist focus on the passages where he emphasizes the autonomy of reason. In these passages, Kant describes reason as a ‘self-legislative activity’, and argues that the moral law binds us because we give it to ourselves autonomously, rather than it being imposed upon us heteronomously from an outside source (G 4: 431). Accounts that interpret Kant as a realist focus on the passages where he describes our awareness of the moral law as a fact of reason. In these passages, Kant says we know the moral law as a ‘fact’ and feel its pull as a sort of reverence (CPrR 5: 47). Constructivists worry that realist accounts sacrifice the ability of the moral law to motivate us independently of our inclinations to be so motivated; realists worry that constructivist accounts flirt with a

182

Carol Hay

pernicious form of relativism. The deflationary interpretation of Kant I am putting forward is intentionally neutral on these debates, emphasizing instead that, whether we endorse a constructivist account of value or a realist one, we ultimately end up with an account where our morally permissible behaviour is constrained with respect to valuable things. This deflationary interpretation affirms almost the entire Kantian story about what value is and where it comes from. The only place Kant goes wrong here is when he argues that animals’ non-intrinsic value means we can only ever have indirect moral duties towards them. In other words, Kant’s commitment to the personification principle is mistaken.3 In what follows, I will make a case for thinking that we can have direct moral duties towards things with non-intrinsic value. Animals, I will argue, are moral patients. They can have morally relevant actions performed on them. They can make moral claims on us. They can constrain our permissible behaviour in ways such that if we fail to respect these constraints we have done something genuinely immoral. We can affirm all these sentiments, I will show, and still say that animals have non-intrinsic rather than intrinsic value.

9.5 The Moral Significance of Non-Intrinsic Value In virtue of what are animals non-intrinsically valuable? Quite simply, animals have value in this Kantian picture simply because we value them. As we have seen, something has non-intrinsic value when someone who is capable of valuing things actually values it. That we do value animals is abundantly clear from even a cursory glance at our practices and attitudes regarding them. We shamelessly spoil our pets, often treating them like full-fledged members of our families. We create wildlife preserves for animals, preserve them in zoos, and mourn the loss of those species that are endangered or extinct. We incorporate them into virtually all of our cultural narratives: they are present in everything from cave paintings to children’s stories to masterpieces of art and cinema and literature. We love them as companions, enjoy them as entertainment, and take their help as workers. And, despite its ubiquity (and our frequent hypocrisy in its face), we hate animal suffering. We cannot help but baulk at cruel treatment towards animals—we react with rage towards people who abuse their pets, we turn our eyes from the 3 Other Kantians have also rejected Kant’s personification principle, but they draw different conclusions about how this rejection ramifies throughout the rest of the Kantian view towards animals. In short, most argue that Kant is wrong to restrict direct moral duties to things with humanity, but they affirm that direct moral duties only ever apply to things with intrinsic value. Their common strategy, then, is to establish that animals have intrinsic value. See e.g. Callicott (2002), Cholbi (2014): Korsgaard (2018). Implausibly, Leonard Nelson goes one step further, arguing that animals are full moral persons—Leonard Nelson (1956: 97–8, 136–44). Gregory Bock argues, just as implausibly, that if we have direct duties to animals then we must have them to plants as well—see Bock (2014).

What Do We Owe to Animals? Kant on Non-Intrinsic Value

183

torturous practices of industrialized agriculture, and when we are forced to face animal suffering directly (as slaughterhouse workers are, for example) we exhibit significant psychological and emotional strain. To be sure, the value we accord to animals is only ever non-intrinsic. But it is value nevertheless. How, then, should the recognition that animals have non-intrinsic value constrain our behaviour? It might seem that this recognition should require a certain kind of respect from us. But most Kantians consider respect to be the appropriate attitude to have only towards human beings, because our rational nature gives us what Kant calls dignity. As we saw above, this respect requires that we have a certain attitude of reverence towards people, and that we recognize and respond appropriately to people’s rational ability to set and pursue ends. But because animals do not have rational nature they cannot have dignity; thus the respect we owe to them will be quite different than the respect we owe to people. Specifically, because animals do not set and pursue ends according to reason the way people do, it is inappropriate to try to understand the respect we owe them in terms of recognizing or responding appropriately to their goals or projects. Some philosophers do try to argue that (at least many) animals have something like rational agency.4 They argue that sentient animals are capable of forming preferences—a cow prefers one patch of grass over another, for example—and this process is, for moral purposes, the same as the process by which we humans set and pursue our ends. There is obviously a difference between what is going on when a cow instinctively eats grass and what is going on when a plant grows towards the sun, but the former seems much closer to the latter than what is going on when a human prefers peas to carrots. It is simply not clear that animal consciousness is sophisticated enough to warrant attributing animals rational agency: whatever a dog is doing when it prefers one bowl of kibble over another is simply not what Kant meant when he described what rational agents are doing when we value things. The relevant sense in which humans’ capacities differ from animals’ capacities has to do with our ability to set and pursue ends according to reason (G 4: 412, MM 6: 392). Humans are autonomous; this means that we are free, in the sense that we are capable of acting according to motives other than instinct (G 4: 446). It also means that we give the laws of morality to ourselves (G 4: 431). We are capable of thinking about which ends to set and about how we should act to achieve these ends. (G 4: 414–15). We are capable of doing things because we think we should do them, not merely because we want to do them (G 4: 413, 4: 416). We are capable of articulating to ourselves the reasons for our actions. We are capable of choosing between multiple courses of action. We are capable of

4 See e.g. Cavalieri (2001), Heeger and Brom (2001), Korsgaard (2018), Regan (2001), Sebo (2004), Watson (1979), Wilson (2011), Wood (1998).

184

Carol Hay

anticipating future consequences, adopting long-term goals, resisting short-term temptations, and adopting ends for which we have no sensuous inclinations. Obviously, these various capacities come in degrees, and it is probably the case that some of the ‘higher’ animals (primates and dolphins come to mind) have enough of them that they should count as autonomous rational agents. But the overwhelming majority of animals do not have an autonomous will, nor the capacity to give reasons, all of which is required for valuing things in the Kantian sense. Of course, we shouldn’t be dogmatic here: if it turns out that some animals have sufficiently well-developed rational capacities then we should be perfectly happy to grant that they have autonomous agency. This would be to grant that these higher animals have intrinsic value, and thus to grant that the moral obligations we have to them are very different from our obligations to other animals. But we should still insist on a morally important distinction between those animals (including us) that have robust rational capacities and those that do not. There is of course the remaining question of how and where to draw the morally relevant line here. For both moral and practical purposes, we should probably be fairly promiscuous with this judgement—better, after all, to err on the side of granting rights where we should not than to infringe on rights that do exist. Still, even given this, the vast majority of animals would not qualify as autonomous and so the question of their different moral status remains. What do we owe to animals, then? How is our behaviour constrained in their regard? While the respect we owe most animals is not the same respect we owe humans, it is important to recognize that it is respect nevertheless. Respecting animals in virtue of their non-intrinsic value requires that we have a certain attitude of reverence towards them. We should appreciate the beauty, the harmony, the elegance, the intricacy, and the complexity, of animals and their relations with each other and with us. We must also appreciate that, despite their lack of rationality, animals’ nature is closer to ours than many like to admit. In particular, animals share with us a capacity to experience pleasure and pain, and they react to these experiences in the same ways we do. As a matter of consistency, then, just as we think pleasurable experiences are good and painful experiences are bad for us, we must think pleasure is good and pain is bad for any being capable of experiencing these things. So, most important for our purposes here, the respect we owe to animals means that we must recognize and respond appropriately to the fact that animals are capable of suffering.5 In practical terms, this means we must avoid causing unnecessary animal suffering. The only clear exceptions to this injunction are those cases where following it

5 Richard Dean and Christine Korsgaard argue for a similar conclusion about the relevant moral similarities between animal and human pain, but neither endorse anything like the Kantian framework I endorse here, particularly the pluralist account of non-intrinsic value I defend below—see Dean (1996: 188), and Korsgaard (2018).

What Do We Owe to Animals? Kant on Non-Intrinsic Value

185

would endanger our rational nature. This is because the intrinsic value of our rational nature trumps the non-intrinsic value of animals. Practically, this means that if the only alternative to making an animal suffer is my own death (or perhaps malnutrition severe enough to damage my rational capacities), then it is morally permissible for me to cause animal suffering or death. This also means that some amount of scientific research using animal subjects is morally permissible, but only in those cases where we have no other practical alternative and have very good reason to believe will result in scientific advances that will save human lives (or perhaps protect our rational natures). What this injunction against causing unnecessary animal suffering does not permit, however, are most of the practices involving animals that many people currently consider to be acceptable. It does not permit eating animals in cases where we are not at risk of starvation or malnutrition; this rules out the permissibility of virtually all meat-eating in wealthy countries such as ours, where all essential nutrients are readily available from plant-based sources.6 It does not permit raising animals in deplorable conditions on factory farms for their meat or milk or eggs. It does not permit raising or killing animals to wear their fur or skin for the purposes of comfort or fashion. It does not permit using animals as research subjects to test the safety of trivialities like cosmetics. It does not permit redundant scientific research, nor research that could be performed on alternative non-sentient subjects. It does not permit hunting for unnecessary food nor sport (but it might permit hunting to control wildlife stocks). In all of these cases, our happiness—the balance of our pain versus our pleasure—must be balanced against animals’ happiness. The pleasure we get from eating meat, or wearing fur, or wearing cosmetics tested on animals, must be weighed against the pain that animals experience as a result of these practices. In virtually every one of these cases of unnecessary animal suffering, the pleasure we gain is greatly outweighed by the animals’ pain. Of course, if it turned out that the amount of pleasure we gained from these practices did outweigh the amount of animal pain these practices caused, these practices would be morally permissible. But no one can seriously think that the gustatory pleasures from eating a steak outweigh the anguish a cow experiences in a confined feeding operation, nor that the narcissistic pleasures from a new shade of lipstick can outweigh the torture a rabbit experiences in a lethal dose test. Hypothetical possibilities aside, our pleasure does not actually outweigh animals’ pain for any of the animal-torture practices that are currently considered by many to be acceptable. This injunction against causing unnecessary animal suffering thus imposes a great number of restrictions on which ways of treating animals are morally acceptable.

6 In fact, it is widely accepted amongst health and nutrition experts that a vegetarian diet is actually healthier than a non-vegetarian diet. See e.g. Keya, Daveya, and Applebya (1999), Sabaté (2003).

186

Carol Hay

These moral restrictions on our behaviour with respect to animals are serious, nontrivial, and are owed directly to the animals themselves. Animals’ non-intrinsic value might ultimately derive from our ability to confer value upon the world, but once they possess this value we owe them a great deal in virtue of it. Consistency demands that we recognize that animals share with us an interest in not suffering unnecessarily, and that this interest in not suffering is just as morally relevant for animals as it is for people. Animals and people might have very different kinds of value, but at least with respect to the injunction to avoid causing unnecessary suffering, this is a difference that makes very little difference.

9.6 A Pluralist Account of Non-Intrinsic Value It has probably struck the reader by now that the story just given about how our behaviour is constrained with respect to animals is not a new one. It is, in fact, basically just the standard utilitarian line. Utilitarians argue that sentient beings have morally relevant interests in virtue of their capacity to experience pain and pleasure, and that the morally correct thing to do in any situation is to maximize the pleasure, and minimize the pain, experienced by all sentient beings. Given that I explicitly set out to articulate a Kantian account of what we owe to animals, this affection for utilitarian reasoning might seem strange. But this strangeness, I submit, actually opens up the space to think about Kantian ethics in a radically new way. As we have seen, Kant has rather a lot to say about how our behaviour is constrained with respect to beings with intrinsic value, and almost nothing to say about how our behaviour is constrained with respect to things with non-intrinsic value. Instead of viewing this relative silence as reason to abandon the Kantian framework, we can instead view it as an opportunity to fill in this lacuna with an  independent (but compatible) account of how we ought to respond to nonintrinsically valuable things. Furthermore, we need not look for one story about how different non-intrinsically valuable things constrain our behaviour. We can, that is, articulate a pluralist account of non-intrinsic value. The utilitarian framework just discussed provides a clear and straightforward explanation of how we should treat these particular things (i.e. animals) that are non-intrinsically valuable. We should recognize that animals’ ability to suffer gives them an interest in not suffering, and we should respond morally to this recognition by causing as little animal suffering as possible. But, as has been wellestablished, utilitarianism is not terribly well suited for solving other ethical problems; it is particularly deficient when it comes to accommodating basic moral intuitions about justice and rights, for example. Virtually every moral framework is generally regarded as better suited for dealing with some moral issues and less good at dealing with others. I propose that

What Do We Owe to Animals? Kant on Non-Intrinsic Value

187

we should respond to this generally agreed-upon phenomenon by treating the various moral theories available to us as tools in our philosophical toolbox. This would mean that we could appeal to Kantianism to explain how we should treat intrinsically valuable things like us humans, and, as long as a theory did not contradict the fundamental tenets of Kantianism, look elsewhere to explain how we should treat non-intrinsically valuable things. So, for example, we could appeal to an account of property rights, possibly supplemented with a background justification of the state and an account of the importance of the rule of law, to explain why it would be wrong to damage or destroy a non-intrinsically valuable physical possession like our neighbour’s car. Similarly, we could appeal to an aesthetic or cultural account of the value of artifacts with historical, cultural, or aesthetic significance to explain why it would be wrong to damage or destroy a non-intrinsically valuable piece of art like the Mona Lisa. More controversially (because talk of abortion is always controversial), we could appeal to an account like Mary Anne Warren’s well-known defence of the permissibility of abortion to explain why we should treat infants differently than foetuses. (Warren argues, on roughly Kantian grounds, that abortion is permissible because foetuses do not have rational agency, but that infanticide is immoral because our behaviour is constrained with respect to how we permissibly may treat infants not because they have intrinsic value, but because they have a nonintrinsic value that does not permit their wanton disregard).7 And, as I have done here, we could appeal to a utilitarian account that prescribes minimizing all unnecessary suffering to explain why it would be wrong to torture animals. To summarize: the picture being proposed here is one that is Kantian at the metaethical level of determining what value is and where it comes from, Kantian at the normative level of determining what is owed to intrinsically valuable beings, and pluralist at the normative level of determining what is owed to different non-intrinsically valuable beings and things. Such a picture is intentionally neutral in the metaethical debates about whether to interpret Kant as a constructivist or a realist, preferring instead to emphasize that at the normative level, the recognition that something is valuable brings with it a recognition that one’s behaviour is constrained in its regard.

9.7 Potential Objection: ‘Merely’ Non-Intrinsic Value In this last section, I consider a number of versions of what I take to be the most pressing objection to the account just provided. This objection, roughly, claims that it is arrogant or insulting to assert that animals are valuable only because we happen to value them, because this implies that we are free to use (or abuse) them 7 Warren (1973).

right 2020. OUP Oxford. rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law.

188

Carol Hay

however we like. If we are the source of animals’ value, then if we do not actually value them then they have no value and we can permissibly treat them in any way we see fit. Our behaviour is only constrained with respect to animals if we want it to be, the thought might be, and this means it is not really constrained at all. Now, Kant himself does often use the language of ‘mere means’ to contrast the permissible ways we may treat non-intrinsically valuable things from the permissible ways we may treat intrinsically valuable things. Non-intrinsically valuable things may be used as mere means to our ends, he says. It is not hard to see why a critic might respond to this by claiming that Kant’s language of ‘merely’ suggests an insulting sort of second-class status for animals, one that fails to capture their true importance. This sort of critic is right to be worried here. After all, many things that are merely non-intrinsically valuable are simply not that important. Post-its, fabric softener, drive-through windows: all these things are merely nonintrinsically valuable. If we mean to put animals in the same category as these useful but trivial things, there would be serious cause for concern here. There are a number of negative or derogatory connotations one might have in mind when saying that something has merely non-intrinsic value. First, one might mean that because that thing may permissibly be used as a means, whatever ends the tool might itself have would be subordinate to the user’s ends. This would involve neglecting or denying the ends the tool might have for itself or asserting that its ends can under no circumstances trump the ends of its user. Second, one might mean that a non-intrinsically valuable thing is monetarily quantifiable. When something is for sale, this stands in stark contrast to Kant’s assertion that what is intrinsically valuable is beyond all price. Relatedly, this could mean that its value is quantifiable solely in terms of its usefulness, crassly understood in terms of commercial value. Lastly, one might mean that something having non-intrinsic value forecloses the possibility of it warranting legitimate ethical regard. If this is what we mean by ‘non-intrinsically valuable’, there is serious cause for concern. All of these problematic uses of the term could, for example, be appealed to if one wanted to justify the morally repugnant practice of factory farming. If animals on these farms are thought of as merely non-intrinsically valuable in these senses, they could legitimately be conceived of as ‘production units’, their natural instincts could permissibly be thwarted, their genetics could permissibly be tampered with via selective breeding, their offspring could permissibly be separated from them prematurely, they could permissibly be treated as the sorts of things whose welfare or suffering may be ignored unless it affects their ‘productivity’, and so on. What is most valuable about this objection is that it has us pay close attention to what we usually think is so bad about a thing’s value being only non-intrinsic. The negative connotation here is that we usually think of non-intrinsically valuable things as permissibly fungible, or replaceable, or violable. We might think that, because of their relative unimportance, we are allowed to use them, or mistreat them, or destroy them, however we like. But notice that this relative

EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 9/14/2020 5:04 PM via COLUMBIA UNIVERSITY - MAIN

What Do We Owe to Animals? Kant on Non-Intrinsic Value

189

unimportance is not true of many non-intrinsically valuable things. Many nonintrinsically valuable things are incredibly important; we are not allowed to misuse, nor mistreat, nor destroy many of them. The Mona Lisa is neither fungible nor replaceable, for example. Something I own is not permissibly violable by you. It simply does not follow from animals having non-intrinsic rather than intrinsic value that they are unimportant. And it does not follow that we are not constrained in how we may treat them. To say that animals have non-intrinsic value is simply to invoke Kant’s particular story about what value is and where it comes from. Nothing about the importance of animals, nor the ways our behaviour is constrained in their regard, necessarily follows from this. A related version of this same objection takes issue with the anthropogenic and anthropocentric Kantian stories about the source and locus of value. According to this version of the objection, there is something deeply troubling about the claim that everything in the universe that is valuable is valuable simply because we humans happen to value it. It follows from Kant’s view that if humans (or other rational agents capable of valuing things) did not exist, then animals and the natural world would have no value whatsoever. To say that the natural world would be valueless without us, again, seems deeply disrespectful of the natural world. It also seems drastically and arrogantly to inflate the importance of us humans. But this objection is premised on a mistaken understanding of the Kantian worldview, and is motivated only by an equivocation on what is meant by the word ‘valueless’. To say that something is ‘valueless’ can mean that thing is valueneutral (not the sort of thing to which the concept of value is appropriately applied) or it can mean that thing is worthless (not the sort of thing that we need to care about, or that can constrain our behaviour, or that matters morally—to be contrasted with those things that do merit this special treatment). The former definition is what the anthropogenic Kantian means when she argues that humans are the source of all value; the latter definition is what is appealed to by the proponent of this objection. The mistake this critic makes when insisting that a counterfactual world without humans would still have value is that they are tacitly smuggling themselves in as a valuer. To say that there would be no value in nature without humans does not mean in this world with humans we are free to do with nature whatever we like (torture the animals in it, alter its ecosystems irreparably, impose monocultures to replace biodiversity, etc.). Rather, it means that in a world without humans, nature is not valuable, it just is. A final version of this objection hinges on the argument given above for why animals are non-intrinsically valuable. Animals are valuable, we saw, simply because we value them. This claim was bolstered by looking at some of the many different practices we engage in that evidence the ways we care about animals. A critic, however, could just as easily point to the very many practices we engage in that seem to evidence that we do not really care about animals very much at all. We eat animals’ flesh, and we usually torture them a great deal before we do so.

190

Carol Hay

We neglect and abandon our pets. We destroy animals’ natural habitats and deplete their natural numbers. We poison them as pests. We use them in torturous scientific experiments, most of which are completely unnecessary and fail to advance scientific knowledge in the slightest. And so on. So, this critic will claim, since we are the source of animals’ value, if we do not actually value animals then they have no value and we can permissibly treat them however we like. Insofar as these inhumane practices are evidence of our failure to value animals, they would also actually function as justifications for animal torture. One could be exonerated from torturing animals simply by insisting that he or she did not care about them. The only response available here is to insist that despite these horrible practices we really do value animals. That we are inconsistent and hypocritical in how we respond to animals’ value does not mean we do not value them; it means we are inconsistent and hypocritical. Furthermore, it cannot be emphasized strongly enough that just because morality is a human construct this does not mean that moral conclusions are up to us. The laws of morality impose objective constraints on our behaviour; we cannot escape morality’s demands simply by changing our minds about what we find valuable. We are what give things value, but we cannot necessarily take it away. Once something is valuable, it is not up to any one of us (or even any large group of us) to revoke this value.8 And so, because we do care about animals (however inconsistently or hypocritically we respond to this value), animals really are valuable. An analogy: it seems uncontroversial that the Mona Lisa would continue to be valuable even if we all stopped caring about it, because it has a particular historical and cultural importance (and probably also because of its particular aesthetic qualities). Another way to put the same point would be to say that we would be wrong or mistaken to stop caring about the Mona Lisa, given its particular value. Analogously, animals would be valuable even if we stopped caring about them, given their particular value. Animals’ value derives from the fundamental roles they play in our lives and from the fact that we have more in common with them than we differ from them. Our rational nature might be what gives us our intrinsic value, but (and this is something that Kant was probably mistaken about, or at least did not grasp the full importance of) it alone does not make up what it means to be human. We are rational, to be sure, but we are also animals. In virtue of our animal nature—i.e. our bodies—we experience both pain and pleasure. And the other sentient animals share this with us. This is why, for basically utilitarian reasons, animal suffering is morally relevant. And, because of their importance, this suffering would remain valuable even if we decided to no longer care about animals. 8 Of course, some things of relatively trivial non-intrinsic value can lose their value if we change our minds about them (e.g. the acid-washed jeans I loved when I was a teenager really are worthless now) but this is not the case for all non-intrinsically valuable things, particularly not those with historical, cultural, or aesthetic significance.

10

Against the Construction of Animal Ethical Standing Jon Garthoff

10.1 Introduction One way to understand large-scale divides in contemporary ethical theory is in terms of differing loci of attention. Broadly Neo-Aristotelian theories tend to emphasize the importance for ethics of a form of biological life, broadly consequentialist theories tend to emphasize sub-rational psychological capacities like sensation or preference, and broadly neo-Kantian theories tend to emphasize the  specifically reflective, critical reason of mature humans. These respective emphases often support different answers to the question of which entities have capacities such that what happens to them is of ethical significance, which is to say these respective emphases commonly support different answers to the question of which beings have ethical standing. These approaches are of course not blind to differences among beings who merit ethical attention. Theories influenced by Aristotle emphasize that many animals have not only biological but also psychological capacities, and that humans furthermore typically have rational capacities; they also commonly conjoin these observations with the further claim that the greatest goods available for an entity must involve the functioning of its highest capacities.1 Consequentialists may note in a Millian spirit that there are differences not only in quantity but also in kind among the goods available to entities with different psychological capacities.2 Kantians may emphasize that even if sub-rational capacities (in their view) strictly speaking fail to constitute a locus of ethical significance, concern for nonhuman animals emerges from a proper account of respect owed to mature humans.3 Notwithstanding these complexities, it is fair to say that to date there 1 Exemplars of Neo-Aristotelianism include Foot (2001), Kraut (2007), and Thompson (2008). 2 In the context of animal ethics, Singer (1975) stands out among consequentialist theorists. For John Stuart Mill’s famous case for qualitative differences among pleasures, see Ch. 2 of Mill (1861). The context of Mill’s argument is of course criticism of the utilitarian theory of Jeremy Bentham (1789), which was itself a path-breaking contribution to animal ethics. 3 Wood (1998, 1999, 2008) expounds a view of this sort.

Jon Garthoff, Against the Construction of Animal Ethical Standing In: Kant and Animals. Edited by: John J. Callanan and Lucy Allais, Oxford University Press (2020). © Jon Garthoff. DOI: 10.1093/oso/9780198859918.003.0011

192

Jon Garthoff

has been a tendency in each approach to privilege the ethical significance of a single capacity. I think that this tendency is best resisted. While we should aspire to the systematicity the above approaches aim to provide, it is important to take seriously that different animal capacities figure in ethics in importantly distinct ways. We should resist the idea that there is a unique locus of ethical standing or status, and we should also resist the idea that all entities who have ethical standing have it equally.4 Such ‘standing egalitarianism’, as I call it, figures prominently in the pioneering work on animal ethics of Peter Singer and Tom Regan.5 I think this general position is fairly obviously mistaken, and that the default in discussions of animal ethics should be strongly against it. The common understanding which forms part of the default starting point for ethical theory includes (i) that mature humans have ethical standing, (ii) that at least some non-human animals have ethical standing, and (iii) that not all non-human animals with ethical standing have an ethical standing equal to that of a mature human. These three positions jointly entail that there are at least two non-equivalent kinds of ethical standing. In a series of papers written during the past fifteen years, which culminated in her 2018 book Fellow Creatures, Christine Korsgaard articulates an ingenious strategy for accommodating the rejection of standing egalitarianism within a fairly orthodox Kantian ethical theory.6 This strategy (which I lay out more fully in the next section) claims that some non-human animals possess ethical standing in virtue of their own capacities, but that these capacities constitute a locus of ethical standing only in virtue of the activity of mature humans. The ingenuity of this strategy consists in part in denying standing egalitarianism without forfeiting what she takes to be the centrepiece of Kant’s signal idea of the kingdom of ends: that the existence of reasons, and so the intelligibility of the subject matter of ethics, results from mutual self-legislation of these reasons by entities able to think about them.7 The kingdom of ends is a landmark expression of the spirit of the Enlightenment, a bold attempt (among other things) to capture in ethical theory the emphases on equality, mutuality, and autonomous self-legislation of the emerging modern political world. The appeal of standing egalitarianism for modern ethical theorists derives in large part from the deep appeal, to modern ears, of the idea of a fundamental equality of standing. Korsgaard resists that sort of appeal to equality, and in my view is right to do so. But in so doing she seeks to preserve, by analogy with Enlightenment political ideals, an emphasis on mutual

4 I present a much fuller argument for this claim, with an emphasis on its significance for the development of broadly Kantian ethical theories, in Garthoff (2010c). 5 See Singer (1975) and Regan (1983). 6 See especially Korsgaard (2005, 2012, 2013, 2014, 2018). 7 G 4:431–439 articulates the idea of the kingdom of ends. Kant elaborates the idea in later work, including MM 6:354–355. As Wood (1999) notes at pp. 376–77, a predecessor idea is presented at A808/B836, which is Kant’s first statement of the moral law.

Against the Construction of Animal Ethical Standing

193

and autonomous self-legislation in the most fundamental level of her theory: all and only those beings who are subject to moral norms, in her view, are ultimate sources of reasons. Korsgaard’s recent work on animals is a major contribution not only to animal ethics but to ethical theory more generally. In this essay, I articulate reasons for rejecting the view that she presents in that work, however, notwithstanding its considerable appeal. The concerns I articulate in this essay build on the reasons for rejecting standing egalitarianism. Standing egalitarianism should be rejected because, as Korsgaard and I both maintain, the capacities in virtue of which entities have ethical standing—or as we might say more colloquially, the capacities in virtue of which they matter—are not the same as the capacities in virtue of which mature humans have their distinctive sort of dignity. I think the best candidate criterion of ethical standing is conscious awareness, Korsgaard appears to think the best candidate is consciousness plus a capacity for aversion or attraction, and a third plausible candidate is biological life. The dignity of humans, by contrast, is grounded in their possession of reflective and critical reason, or ‘humanity’ in Kant’s sense.8 Below I present a further, analogous criticism of Korsgaard’s efforts to preserve mutual self-legislation in her most fundamental account of reasons-response. The capacity in virtue of which an entity responds to reasons is judgment, which consists paradigmatically in propositional thought and inference. But this is not the same as the capacity in virtue of which mature humans have their distinctive mode of reasons-response, moral obligation; possession of moral obligations requires not merely judgment but also specifically reflective and critical reason. If this is correct, then ethical theory must move farther than Korsgaard has done from the idea of the kingdom of ends, notwithstanding that idea’s tremendous appeal as a quintessential expression of the spirit of the Enlightenment. Just as the appealing Enlightenment model of equal citizenship fails to provide inequalities of moral status needed for a complete ethical theory, so does the appealing Enlightenment model of mutual and autonomous authorship of norms fail to provide an accurate account of the constitutive conditions of reasons-response. The structure of the discussion is as follows. In the next section, I sketch Korsgaard’s account of animal ethical standing. In Sections 3–4, I articulate concerns about Korsgaard’s constructivist account of what is good for individuals, according to which something can be genuinely of value for an entity only if that entity can enjoy or appreciate it. More specifically those sections articulate a series of distinctions that complicate Korsgaard’s theoretical ambitions. In Sections 5–6, I raise concerns about Korsgaard’s constructivist account of reasons. More specifically, I show how Korsgaard’s account of reasons relies on an implausible—as 8 G 4:426–37, articulates and elaborates his distinctive notions of humanity and dignity. These notions are developed further in Rel.: 26f, also throughout MM.

194

Jon Garthoff

well as non-Kantian—denial that belief constitutively responds to reasons. If we instead accept the Kantian view that belief as such involves response to reasons, I  argue, then we must deny that all reasons arise from the activity of critical reason. This is because the standard view of animal cognition, which Korsgaard and I both accept, maintains that some non-critically rational animals have beliefs. The seventh and final section concludes the essay by briefly explaining why Korsgaard’s naturalism provides no independent support for her view. The upshot of this discussion will also include the outlines of a new position to rival Korsgaard’s. According to this view, value came into the world with the development of the first functionally organized beings; values that matter (which I label things of ‘worth’) came into the world with the first conscious beings; reasons were first possessed by the first judging beings; and moral obligations were first possessed by the first critically rational beings. There are also distinct kinds of moral status that correspond to each of these kinds of being (though in the case of merely functionally organized beings, it is the null moral status). According to this approach, it is crucial not to flatten the domain of ethics by conflating any of these tiers or theorizing on the assumption that any of them may be reduced to the others. My suggestion is that contemporary theorists who aim to preserve moral overridingness and other crucial features of Kant’s ethical theory—also those who aim to synthesize attractive features of the ethical theories of Kant and Aristotle—would be better served by following this approach than by adopting Korsgaard’s.

10.2 Korsgaard’s Account of Animal Ethical Standing Korsgaard’s general strategy for accepting mutual self-legislation of all reasons without endorsing ethical standing egalitarianism is to maintain that there are forms of ethical standing a being can have even though this standing is conferred by another being. Her view is that critically rational animals—humans—are the ultimate source of all reasons.9 As Korsgaard is careful to point out, however, this view does not entail that the ethical standing of critically rational entities is the only type of ethical standing, nor does it entail that the good of critically rational entities is the only good of genuine worth. In this section, I lay out the features of Korsgaard’s view that enable her to defend mutual self-legislation of reasons without entailing standing egalitarianism. This is an important feature of her position, since—as was observed in the introduction—the common understanding grounding ethical theory should include a strong default against standing egalitarianism.

9 Korsgaard (1996) explicates this view most fully; but see also Korsgaard (1983, 1986a, 1986b).

Against the Construction of Animal Ethical Standing

195

Though Korsgaard claims that the flourishing of plants is not of genuine worth as such, she allows there is an ordinary sense in which it is appropriate to speak of the good of plants.10 In her view, it is a mistake to attribute a final good to plants, however, where this is understood as ‘the end or aim of all our strivings . . . a state of affairs that is desirable or valuable or worth achieving for its own sake’.11 But on her view only beings with a capacity for conscious experience, whose environment may ‘strike her as attractive or aversive’, have a final good.12 The mere fact that an entity has a final good in this sense does not entail it has reasons to preserve or promote that final good. Indeed, in Korsgaard’s view even animals with consciousness and a capacity for belief lack reasons to preserve or promote their final good. She articulates this feature of her position by marking a distinction between ‘intelligence’ and ‘reason’.13 Animals with more sophisticated cognitive capacities acquire knowledge of the world, and they use that knowledge to act more effectively on their natural instincts. But in Korsgaard’s view this more sophisticated cognition is merely a type of intelligence; it is not a type of reasons-response. Only critical reasoners, who are equipped with a conceptual capacity to represent reasons as such, are in Korsgaard’s accounting well understood as acting for reasons. ‘A rational animal,’ Korsgaard writes, ‘is (at least sometimes) aware of the grounds of her beliefs and actions.’14 She later comments that ‘a rational being makes an assessment of the whole principle . . . of doing this act for the sake of this end on this kind of occasion’.15 Korsgaard is explicit that her reasoning proceeds from an a conception of what it is to act for a reason to the conclusion that no animal which lacks critical reason acts for reasons at all, a point I return to in Sections 5–6. Korsgaard’s writings on animals bring out how her approach to ethical theory is doubly constructivist. Her position is constructivist in the most familiar sense because it grounds genuine worth in the psychological capacity of individuals to enjoy or appreciate their good; I refer to this the ‘first stage’ of her proposed construction of value. In Korsgaard’s view, this is why all and only beings with a capacity for aversion or attraction have a good of genuine worth, and why, in particular, plants do not have a good of genuine worth. Korsgaard’s view is also constructivist in its contention that all reasons to promote or otherwise respond appropriately to genuine values are constructed by the 10 Korsgaard (2018: 19–20). 11 Korsgaard (2018: 17); and Korsgaard (2014). I criticize this use of ‘final good’ below. 12 Korsgaard (2018), 20. 13 Korsgaard (2018: 39–43). See also Korsgaard (2014). 14 Korsgaard (2018: 39); the parentheses are Korsgaard’s. 15 Korsgaard (2018: 119). In the unpublished lectures on which Fellow Creatures was based Korsgaard wrote that, in contrast with rational human animals, ‘non-human animals don’t do more than follow their strongest impulse, even in cases where they are choosing the means to ends or choosing between two options’. See Korsgaard (unpublished), Lecture 2, 35. It is unclear to me whether the elimination of this characterization reflects a change in Korsgaard’s view.

196

Jon Garthoff

operation of specifically critical reason; I refer to this as the ‘second stage’ of Korsgaard’s proposed constructivism. This view of Korsgaard’s entails that on Earth only human beings are subject to evaluation in terms of rationality. On her view, the class of beings subject to these rational norms is the same as the class of beings who are subject to moral norms. I deny that view, and in what follows I raise doubts about both stages of Korsgaard’s double-constructivism. But for now I simply observe that they are distinct, also that each could be endorsed without endorsing the other. As I understand it, Korsgaard’s position has three principal motivations, two of which she articulates explicitly in recent work. One is what she calls the ‘essentially subjective’ character of value—her view that a genuine value for an individual must be able to be experienced as positively valenced by that individual.16 A second motivation, not emphasized in recent work but implicitly present there, is to articulate an account of value that coheres with her general strategy for vindicating the overridingness of moral obligation: namely, to claim that nothing has genuine worth antecedent to a rational order constituted by the activity of critically rational beings. The third is her contention that a view must be ‘naturalistic’ in a way that at least some rival positions fail to be.17 I discuss each of these motivations for Korsgaard’s position more fully in the discussion below. Korsgaard’s recent writings on animal ethics are a major contribution both to our understanding of the moral status of animals and to our understanding of the available options for extending and developing Kantian moral theory. In my judgment, they are the most important contribution to the literature on animal ethics during that period. In the remainder of this essay, however, I articulate a number of considerations that tell against it. My aim here is not to refute Korsgaard’s position, since even to attempt that would demand a fuller discussion than a single essay can accommodate. My aim is rather to articulate the motivation and spirit of serious complications for both the view she advances and the argumentative strategy she pursues. This is an important exercise even apart from the merits of Korsgaard’s view, moreover, for it points the way toward a novel approach to both animal and human ethics, an approach itself worthy of much fuller development on another occasion.

10.3 Distinguishing ‘Good For’ and ‘Good To’ Let me first criticize Korsgaard’s use of ‘final good’ and ‘good for’, including especially her claim there is a normal use of these terms according to which they apply only to beings with minds. This concern is terminological, and does not as such 16 Korsgaard (2018: 20).

17 Korsgaard (2018: 167–9).

Against the Construction of Animal Ethical Standing

197

constitute a substantive objection to her view. Nevertheless, it is worth discussing briefly, as it clarifies substantive concerns to come. Korsgaard contends that the ordinary phrase ‘good for’ is ambiguous between what she calls ‘functional’ and ‘final’ senses.18 Korsgaard acknowledges we speak non-metaphorically of events being good or bad for all functionally organized entities, as in: ‘I hope it rains, since more water would be good for the oak tree in the front yard’. This is what she terms the functional use of ‘good for’.19 While Korsgaard does not acknowledge it, we also use ‘ought to’ in connection with all living things, as in: ‘The oak tree in the front yard ought to have deeper roots, it is not taking in enough water’. These uses of ‘good for’ and ‘ought to’ do not entail the existence of reasons. Oak trees and other organisms lacking the capacity of judgment cannot themselves have reasons, since they have no capacity to cognize reasons. But more importantly and generally: to say that flourishing is good for a non-conscious organism need involve no commitment to anyone under any circumstances having a reason to do what is good for the organism as such.20 Korsgaard claims there is another sense of ‘good for’, which she associates with the psychological attitudes of caring or striving, and which her technical term ‘final good’ is meant to capture. This is why she claims that only beings with minds have a final good. This restriction on the use of ‘final good’ is at variance with standard philosophical use, however, because this term is strongly associated with the Neo-Aristotelian tradition of contemporary ethical theory, within which it applies to (at least) all living organisms. Indeed Neo-Aristotelians typically intend their philosophical use to be the ordinary use, and I think they are correct in thinking that (in this respect at least) it is.21 Korsgaard’s supposed ambiguity in ‘good for’ is in my view an artefact of her own theoretical commitments, not a reflection of standard ordinary or technical uses. Accordingly, I recommend against her use of ‘final good’. Better is to use an alternative term to pick out those goods whose existence entails reasons for us to act, and the ordinary term ‘worth’ may be well suited for this use. In addition to following both ordinary and philosophical use, this also enables us to distinguish ‘good for’ more clearly from ‘good to’. By ‘good to’ I do not mean this term’s use in ‘she was good to her cat when she gave it the food’, but rather its use in ‘the food 18 Korsgaard (2014, 2018). She introduces the functional sense by calling it ‘motherly’, as she associates it with advice about health stereotypically given by mothers to their children. 19 See Kraut (2007), especially Ch. 2. I think there are licit, non-metaphorical uses of ‘good for’ for all functionally organized entities, though that claim is not crucial for present purposes. Korsgaard (2014) concurs, which is why she labels one sense of ‘good for’ the functional sense. 20 Neo-Aristotelians sometimes claim there always are such reasons, but that is a substantive position. See Kraut (2007) for an example of this commitment. 21 Thus Foot (2001) and Thompson (2008) refer to their claims as ‘grammatical’, and Kraut (2007) makes extensive appeals to ordinary linguistic uses of ‘good’ and ‘good for’. Rosalind Hursthouse (1999) merits mention here as well. Plato may have thought that all teleology is psychological, and so may have thought that all final goods involve literal striving. It is doubtful that Aristotle held that position, however, and Neo-Aristotelians definitely do not hold it.

198

Jon Garthoff

she gave looked good to her cat’. Seeming good is itself an important phenomenon, and one of the many values of Korsgaard’s discussions is her distinctive way of calling attention to that phenomenon. I believe Korsgaard is mistaken to think that everything that is genuinely good for a being depends on what seems good to that being. But even if she is correct in thinking this, it is a substantive claim that is more clearly understood when ‘good for’ and ‘final good’ are distinguished from ‘good to’.

10.4 Distinguishing Animacy, Perception, and Consciousness Korsgaard works with an Aristotelian conception of plants, according to which they are marked by a functional capacity of nutrition. She also works with an Aristotelian conception of animals, according to which animals are marked by the possession of three further capacities. The three most important capacities animals characteristically possess but plants do not are perception, consciousness, and animacy, the last of which is the capacity functionally to move oneself (inasmuch as the term ‘animal’ is used in contrast to ‘person’ or ‘human’, animals are also marked by lacking critical and reflective reason). As I explain in this section, however, these three capacities that the Aristotelian conception associates with animals are distinct.22 While on Earth all organisms with minds are animate, the converse is not true. There are animate earthly organisms, such as hydras, that lack perception and consciousness, and so lack minds altogether. There are also very likely earthly organisms, including some arthropods, which are animate and perceiving but lack consciousness. Korsgaard is aware of these distinctions, but she does not attend to them at length; she instead bundles these three animal capacities together. Similarly, Korsgaard makes use of the important distinction between these animal capacities on one hand and the capacity of nutrition on the other, mainly in an attempt to explain why she thinks that animals but not plants have a genuine final good. But when explicating the ethical status of animals she does not distinguish roles played by each of these capacities. I think her view appears more plausible than it is as a result of its failure to distinguish carefully among these capacities. Accordingly, I begin this section by first laying out these distinctions clearly, and then I proceed to articulate concerns about how paying careful attention to  these distinctions complicates, and thereby undermines the appeal of, Korsgaard’s view.

22 Also, as may perhaps go without saying, the modern notion of animal is phylogenetic and does not track possession of any of these capacities. In this sense of the term there are inanimate animals.

Against the Construction of Animal Ethical Standing

199

All living entities are functionally organized. Indeed what it is to be alive, in the typical biological sense of ‘life’, is to have a functional capacity for nutrition.23 Nutrition is roughly the capacity to take in objects from the environment and metabolize them: to extract and process the chemicals needed for survival, maintenance, growth, and maturation. Some living entities are also animate, which is to say they have a functional capacity to move themselves.24 All animate living entities have a functional capacity for sensory registration, a capacity to carry information that is also functionally related to animacy.25 There are various modes of sensory registration, and these correspond to various environmental features that an organism may need to move toward or away from, including light, heat, chemical concentrations, and magnetic fields. Animacy and sensory registration are not psychological capacities.26 In some organisms with these capacities, their contribution to the organism’s overall functioning is exhausted by their contribution to its biological functioning, such as its nutritional or reproductive functioning. There are at least two general biological functions: nutritive flourishing and (Darwinian) fitness.27 The former is (roughly) doing well as an entire organism, while the latter is (again roughly) surviving-toreproduction. These two general functions are independent, in that neither is well understood as wholly in the service of the other. It may be that all biological functions serve one or both of these two general biological functions; or perhaps there are other independent general biological functions. In any case, the functional capacities of animacy and sensory registration, where these occur in organisms that lack psychological capacities, are plausibly understood to function only as contributors to biological functions such as these. In addition to animacy and sensory registration, consciously perceiving animals also have psychological capacities. One is perception. Perception is representational, since its operations involve functional attribution of and reference to reality.28 It is also sensory, for it is constitutively mediated by capacities for sensory registration; visual perception, for example, is constitutively mediated by

23 Aristotle, Nicomachean Ethics 1.7 1097b20–1098a5. Aristotle’s conception of life is more ordinary than, and in my view superior to, Kant’s; Kant, oddly, defines life (leben) in psychological terms, as the ‘capacity to act in accordance with representations’—see MM 6:211. 24 For more on animacy (which he terms ‘primitive agency’), see Burge (2010a: 326–41). Animacy occurs in very primitive organisms, likely even in single-cell organisms; for further development of that point see van Houten (2000). 25 For more on the idea of sensory information registration, see Burge (2010a: 315–19). 26 For an excellent summary of empirical work supporting this claim, see Burge (2010a: 342–66). 27 For more on the idea of biological function, see Burge (2010a: 292–308). For a helpful earlier discussion see also Wright (1973). For a sophisticated (but I think ultimately unsuccessful) denial of the claim that there are at least two general biological functions, see FitzPatrick (2000). 28 See Burge (2010: 379–96). There are sophisticated accounts of perception according to which it is not constitutively representational; see, for example, Martin (2002) and Kalderon (2011). The default position is that perception is representational, however, and is in my view given decisive defense by Burge in Origins of Objectivity.

200

Jon Garthoff

non-psychological capacities to sense light.29 Furthermore, perception is objective in that it constitutively involves representation as of entities which are external to the perceiver’s mind.30 Perception evolved, of course, because it serves biological functions like fitness; in some circumstances having a capacity to form a representation as of a distal object conduces to an organism’s survival-to-reproduction more than merely having a capacity for sensory registration. This helps to explain how perceptual capacities arose in the natural world. But once evolved, the proper functioning of a perceptual capacity is not exhausted by its contribution to the biological functioning of the organisms that have it. In addition to serving biological functions, perceptual capacities also have their own specifically psychological function of representing accurately. This supports functional assessment of perceptual capacities in terms of the accuracy of the perceptual representations that they yield. It also supports functional assessments of perceptual capacities in terms of further norms that are derived from this basic norm, such as excellence or reliability in representing accurately.31 Consciously perceiving animals also have the capacity for consciousness. Consciousness is less well understood than perception and is accordingly more difficult to define. It is a capacity for awareness, but that is not a highly illuminating gloss, as it uses a virtual synonym of the idea to be explained. Perhaps an inkling of what consciousness consists in is found in characterizing it as awareness subjectively presented. It is the immediate inside of awareness, presentation of the internal state of the mind, as contrasted with representation as of objects external to the mind.32 Phenomenology is the qualitative character of consciousness, ‘what it is like’ in Thomas Nagel’s famous phrase.33 Consciousness and phenomenology are intimately related capacities, probably mutually necessary.34 They are not, however, identical. Conscious awareness may be possible in the absence of phenomenology; awareness of an occurrent thought [that 2+2=4], for example, may have no associated qualitative character.35 Sensation is the capacity for felt phenomenology.36 This is a more specific capacity than the general capacity for phenomenology, since conscious states may include qualities—appearing red, perhaps—which, unlike pains and pleasures, are not well understood as feelings. Since consciousness is less well understood than perception, it is more challenging to say what functions it serves, or even whether it is constituted by serving any function at all.37 Some conscious states contribute to biological function, 29 See Burge (2010a: 376–8). 30 For more on this idea, see Burge (2010a: 396–416). 31 This theme runs throughout Burge (2010a) but is most prominent in Ch. 8. 32 See Burge (2007a, 2007b). 33 See Nagel (1974). 34 See Burge (2007a, 2007b). 35 Block (1995, 2005) first called attention to the distinction between phenomenal consciousness and consciousness of psychological states readily accessible for cognition. 36 Allen (2004) and Ch. 5 of Varner (2012) contain fine summaries of empirical evidence bearing on the question of which animals are sensate. 37 See Burge (2007b: 401–3).

Against the Construction of Animal Ethical Standing

201

and this is important, since it would otherwise be (even) more mysterious how consciousness arose. There are plausible hypotheses about how consciousness contributes to biological function; pain may contribute to fitness, for example, by causing organisms to avoid noxious features of their environment.38 Despite uncertainty about how consciousness functions, it is clear that it is as such psychological (this is clearer even than the analogous claim about perception). In consciously perceiving animals, furthermore, the mind to which perceptual representations are attributed is identical to the mind to which conscious states are attributed. This unity of the subject to which we attribute representational states and conscious states provides a hint of what unifies psychology as a subject matter. Equipped with these distinctions, let us consider the first stage of Korsgaard’s double-constructivism: the stage where it is claimed that a valuing being confers worth on its own final good by exercising its capacity to enjoy or appreciate that good. The distinctions of this section enable us to raise a question Korsgaard herself does not explicitly address. When we speak of the final good of an animal, do we refer to its functioning well as a living being, as an animate being, as a perceiver, or as a conscious being (assuming that this last makes sense)? Korsgaard’s view appears to be: all of the above. She notes a general correspondence between animal mental states and biological functions like survival and reproduction, and claims ‘an animal functions, in part, by making her own well-functioning, the things that are good or bad for her in the functional sense, an end of action, a thing to go for, a final good’.39 Her ultimate conclusions involve claims to the effect that different organisms have different forms of life incorporating their modes of functioning. Indeed, she even argues that claims of value across this boundary are intelligible only within, hence entirely relative to, each form of species-life.40 There are a number of concerns with these suggestions. First is the fact, which Korsgaard does not observe, that there are multiple biological functions. Does an animal’s attitudes confer final value on its fitness, its nutritional flourishing, or both? These can come apart, as when the death of an organism or the deterioration of its systems conduces to the survival of offspring or other kin.41 Since the 38 Issues here are, however, complex; Allen (2004) articulates some healthy skepticism about the standard explanations of how and why sensation contributes to fitness. 39 Korsgaard (2018: 21). See also Korsgaard (2014). In her earliest work on animals, Korsgaard claims that as humans we must value our nature as valuing beings, but she entertains and rejects the parallel claim that as humans we must value our nature as nutritive beings (Korsgaard 2005: 105–6). This suggests, against Korsgaard’s apparent later position, that nutrition figures in the final good of an animal’s life only insofar as the animal has attitudes towards aspects or constituents of this function. 40 Korsgaard (2018: 53–74). This idea resonates with the views of Foot (2001) and Thompson (2008). 41 For extensive discussion of functions in the context of evolutionary biology, see FitzPatrick (2000). As FitzPatrick argues, if there is only one fundamental biological function, it is fitness; but that claim is of no use to Korsgaard, since an animal’s good, inasmuch as we should care about it, is manifestly not the same thing as its fitness.

202

Jon Garthoff

animal does not have mental states about fitness or nutrition as such, the contents of its mental states do not answer the question. Second is the possibility, which Korsgaard does not highlight, that the mind of an animal is a thing distinct from the animal itself. To the extent that our various functions are unified by a species-relative form of life, this is true at the level of biology. That is: functional psychological capacities, perception included, in part serve biological functions like nutrition and fitness. Their functioning is not, however, exhausted by their role in serving these biological functions. We can and do assess perceptual systems for accuracy independent from whether perceiving accurately conduces to fitness or nutrition.42 This point is clearer still for more sophisticated representational functions like belief, and it carries over also to the domain of (at least some) conscious capacities. Part of the purpose of removing a hand (or paw) from a stove is to avoid bodily damage to serve fitness or nutritive flourishing. But averting bodily damage can also contribute to psychological function. At the level of psychology, fitness need not be functional at all, and nutritional function may be only a constitutive means for continuing to function psychologically. Korsgaard does not say why she denies that psychological flourishing alone is what is good for animals with minds. Nor does she say how she thinks that these functions are unified within the forms of life peculiar to various species. Against Korsgaard, it is unclear why an account of specifically psychological function needs to take a species-specific form. And more strongly: it is unclear how a putative account of animal psychology could be comprehensive if this account is species-specific at its fundamental level. The idea of a species is not a psychological idea. Psychological theories do not exclusively relativize their accounts of the capacities they discuss (such as perception and consciousness) to the proper functioning of these capacities in the forms of life of particular species. A substantive account of psychological flourishing, using these and other categories instantiated across species, is accordingly already under development. I see no reason to think this account will fail to render intelligible comparisons of psychological flourishing across species.43 A third complication concerns what Korsgaard calls the ‘essential subjectivity’ of value. Her remarks in this connection are on the trail of a significant phenomenon, in my view, even for those who reject her constructivism. This is an extremely challenging topic, one I do not attempt to address fully here. But I would observe briefly that the domain of consciousness is distinctive in that our

42 Burge (2010a: 292–308) emphasizes this point. Part of the context of Burge’s discussion is criticism of the opposed view of Millikan (1984, 1989, 1990). 43 This point is underscored by the fact that an animal’s mind could perhaps continue to exist even if it had a different biological substrate, and perhaps even if it had a non-biological substrate.

Against the Construction of Animal Ethical Standing

203

primary access to it is through its subjective presentation.44 Korsgaard emphasizes conscious capacities in her account of value—I endorse this emphasis, even though I recommend developing it differently—and it is perhaps no surprise this emphasis comes with an emphasis on subjectivity in the nature of value. The brief observation I make here, however, is that the sense in which consciousness is essentially subjective is not readily marshalled to defend constructivism in ethics. The main point I emphasize here is that the subjective nature of consciousness provides little support for the thought that the function of a conscious judgment is greatly informed by the fact that it is presented in part subjectively. This is most clear with non-phenomenal conscious states, such as occurrent awareness [that 2+2=4]. Here epistemic relations among representations appropriately dominate our interest—including especially this judgment’s appropriate role in inferences— and these epistemic relations do not mainly concern whether the representation is conscious. This suggests by analogy that we should hesitate before projecting the subjective character of the most primitive proto-evaluative attitudes up through all the more sophisticated evaluative representations animals possess. When we judge that something is good, or is to-be-done, we use conceptual categories originating in subjectively presented intimations of objects or behaviours as desirable. Goodness and ‘to-be-doneness’ are, in this etiological respect, akin to appearances and feels.45 Appearances and feels retain this essentially subjective character when they occur in organisms with cognition or critical reason, however, while goodness and to-be-doneness do not. In thinking that an object or behaviour is good, one need not have any subjective presentation of the object or behaviour as good. Nor does the correctness of the thought appear to depend on subjective features of the thinker, or of anyone else. This contrasts with appearances and feels. To think that an object appears tasty or that a behaviour feels pleasurable is to have a thought whose correctness-conditions concern essentially subjective features of the world. The default should therefore be against accounts of ethics as subjectively constructed or projected. We can acknowledge that the origins of genuine worth are subjective—presented to the first valuing beings directly inside their minds, rather than represented by them as an external object—without thinking that worth is itself somehow constructed by the (possibly idiosyncratic) contents of those beings’ subjective mental states. The distinctions of this section complicate Korsgaard’s account of animal ethical status by giving rise to questions she does not explicitly address. One such 44 See Burge (2007a: 417–18). Note that the relevant notion of subjectivity here is roughly internal to the mind of the subject, by contrast with the notion of objectivity as external to the mind of the subject. This is not the distinction, emphasized in Nagel (1986), between first- and third-personal forms of representation. For more on the distinction between subjective and objective see Burge (2010a: 46–54, 396–416). 45 The phrase ‘to-be-done’ was made famous by Mackie (1977).

204

Jon Garthoff

question is: since animacy, perception, and consciousness can come apart, which of these is necessary or sufficient for having a life of genuine worth? A second question is: what is the relationship between a form of life and the distinct biological functions of fitness and nutrition? A third is: how can a form of life reconcile conflicts between flourishing biologically and flourishing psychologically? And a fourth is: even allowing both that consciousness is an essentially subjective phenomenon and that genuine worth originates in consciousness, why think that worth is itself subjective, projected, or constructed? Once the distinctions of this and the previous section are more carefully attended to, and the question of whether and how each distinct capacity helps to enable a life of genuine worth is considered, these outstanding complications for Korsgaard’s constructivism come more clearly into view.

10.5 Distinguishing Inferential Judgment from Conscious Perception In this section, I distinguish the capacity of judgment from the less sophisticated capacity of conscious perception, and in the following section, I distinguish it from the more sophisticated capacity of critical reason. More clearly distinguishing the category of animals with inferential judgment, as I explain in the remainder of the chapter, throws the mutual self-legislation of reasons into doubt. This in turns casts doubt on Korsgaard’s constructivist account of ethics, and with it her strategy for denying ethical standing egalitarianism by claiming that distinct ethical statuses are all conferred by the activity of critically rational persons. Judgment is a representational capacity, and it is one which is importantly distinct from mere perception. While it is not clear whether there are genuine capacities for judgment of a non-propositional character, I here take propositional thought as the paradigm of judgment, since it is the only capacity for judgment of which we have much understanding.46 Belief and preference are paradigm capacities for propositional thought. They are attitudes toward contents of propositional form, which can be true or false.47 They are inferential in that a thinker may move from one belief or preference to another such state on the basis of reasons, as when one infers (hence judges, and characteristically comes to believe) [that p q] from a prior belief [that p].

46 If there are non-propositional judgments, they likely are constituted by a sophisticated form of imagistic cognition. For discussion of the boundary between propositional thought and less sophisticated capacities of representation, see Camp (2007) and Burge (2010a: 537–51). 47 Contemporary discussions of propositional thought descend from Frege; see in particular Frege (1918), though he uses the category much earlier. Propositional attitudes themselves need not be truth-apt, but the content toward which an attitude is taken must be a proposition, and so must be truth-apt.

Against the Construction of Animal Ethical Standing

205

In animals with judgment, perceptions too may perhaps have a propositional form, and so may constitute the basis for an inference. But this is not a general feature of perceptual representation. Perceptual states may instead have a subpropositional form, with contents akin to [large body ahead] or [danger ahead]. These representational contents are not true or false. They are subject to functional assessment instead in terms of accuracy: the former is accurate just in case  there is a large body ahead, the latter just in case there is a danger ahead. These contents are also subject to assessment in terms of functional relationships to other capacities and states – the memories and behaviours they often produce, for example – but these assessments do not pertain as such to truth or to truth-preservation. I suspect Korsgaard denies any non-human animal has a capacity for judgment, but the point is not entirely clear to me, since she attributes judgmententailing states like beliefs to some non-human animals. In any case her remarks about the nature of practical agency in animals with beliefs or other propositional attitudes are open to criticism. Korsgaard resists describing animals as responding to reasons, even when she describes these same animals as having beliefs. She sees the difference between the relatively primitive mental life of a spider or frog and the relatively sophisticated mental life of a chimpanzee as ranged on a scale of intelligence, rather than as subject to important threshold differences, at least insofar as any threshold differences concern capacities to respond to reasons.48 The primary concern with this account of the more sophisticated non-human animals is that we need to understand what it is to attribute propositional attitudes to them without thereby attributing to them a capacity for reasons-response. This concern may be summarized as follows: having beliefs entails a capacity to draw inferences from a belief—so too for other propositional attitudes—and a capacity to draw inferences from a belief (or other propositional thought) in turn entails a capacity to respond to reasons. Korsgaard associates reasons-response with the capacity for explicit justification, which involves representing reasons, and so she associates reasons-response with capacities to evaluate explicitly the rationality of intentions, beliefs, inferences, and the like.49 Following Tyler Burge I have called this complex of psychological capacities ‘critical reason’.50 In the unpublished lectures on which Fellow Creatures is based, Korsgaard writes: ‘Non-human animals may have beliefs and may arrive at those beliefs under the influence of evidence, but it is a further step

48 In the unpublished lectures on which Korsgaard (2018) is based, she writes that a chimpanzee’s choice about what to do is ‘made for him by the strength of his affective states’. She further explicitly likens this to Hobbes’s account of psychology, which uses the idiom of fluid dynamics. See Korsgaard (unpublished), Lecture 2, 6–16, 21–2, and 29–31; and see Hobbes (1651), especially Chapter VI. It is unclear to me whether Korsgaard continues to endorse these claims in Fellow Creatures. 49 Korsgaard (2018: 38–44). 50 See Burge (2013), especially 166–86. See also Kitcher (2005).

206

Jon Garthoff

to be the sort of animal that can ask oneself whether the evidence really justifies the belief and can adjust one’s conclusions accordingly.’51 That is certainly correct, so far as it goes. It is worth dwelling, however, on what it is to arrive at a belief ‘under the influence of evidence’. If these animals (a chimpanzee is a fine candidate) arrive at their beliefs under the influence of evidence, then they arrive at these beliefs in response to reasons. To think otherwise would treat the evidence as if it stood in a merely functional and causal relationship to the belief, rather than in a rational relationship. But if its influence were merely functional and causal, then it would not be treated as evidence by the animal. Perception is a representational capacity that is functional and causal, but does not entail a capacity for inference or judgment; indeed, precisely in virtue of that feature, perceptions can be arrived at properly without treating anything as evidence. Indeed, some animals (here Korsgaard’s frog and spider are good candidates) form perceptions and inaccurate perceptual representations without ever treating anything as evidence. We do of course speak of functional and causal ‘influence’ as well. Perhaps Korsgaard’s use of ‘under the influence of evidence’ in the passage above should be understood instead as a non-reasons-responsive process, a case of (mere) functional causation of a representation. On this interpretation, however, it is unclear why the source of the influence is properly characterized as evidence, and it is also unclear why the psychological state so functionally produced is properly characterized as a belief. Perhaps the most important way that belief differs from perception as a representational capacity is that it is constituted in part by a capacity for judgment.52 There is no special difficulty understanding how a frog’s mind moves from one perceptual representation to another, or from a perceptual representation to a memory or a behaviour, simply in virtue of the frog’s causal relationships with its environment, the causal and functional relationships within the frog’s body, and the (possibly algorithmic) functional connections within the frog’s mind. If all the frog’s representational capacities are exhausted by operations like these, however, then it is for that reason doubtful we should attribute a capacity for judgment or reasons-response to the animal. When we attribute beliefs to a chimpanzee, by contrast, we use an idea that we do not fully understand in merely functional and causal terms. Korsgaard might reject this characterization of belief and other propositional attitudes. I will not here attempt to provide a conclusive case that no purely functional and causal account of these attitudes can succeed. It is no small cost to commit to such an explication of the minds of chimpanzees, however, when we

51 Korsgaard (unpublished), Lecture 2, 21–2. 52 This concern about the intelligibility of propositional attitudes without a reasons-responsive capacity for judgment is bolstered, in my view, by reflection on our practices of training and punishment. For further development of this point see Garthoff (forthcoming).

Against the Construction of Animal Ethical Standing

207

have so little understanding of how this explication could succeed. And regardless of which capacities chimpanzees and other non-human animals have, it is important to bear in mind that there is no compelling reason to think attributing a capacity for reasons-responsive judgment to an animal must entail also attributing to it a capacity to reflect on reasons or to engage in explicit justification.53 Korsgaard is correct to note that those capacities occur only in critically reasoning animals, also that humans may be the only such animals on Earth. If it is claimed further that no psychological capacity could count as reasons-response without involving reflection on reasons, that claim requires a much more extensive explication and defence.54 In many respects, this section’s concern about Korsgaard’s account of acting for a reason parallels Burge’s criticisms of certain neo-Kantian theories of mind. Burge charges these theories, in my view correctly, with overintellectualizing the phenomenon of belief, since they characteristically require capacities to think about reasons as a necessary condition for having a capacity for belief.55 I suggest here, in an analogous spirit, that Korsgaard overintellectualizes acting for reasons by incorporating into her account of this phenomenon a requirement that the agent itself represent reasons. There is an irony here, since a concern not to underintellectualize judgment—by understanding belief as a merely functional state—is part of what led many Neo-Kantian theorists to understate the cognitive capacities of the more psychologically sophisticated animals. These Neo-Kantian theorists would characteristically hold, against Korsgaard, that the capacity for belief brings with it a capacity to act in response to reasons. Even if these theorists overintellectualize belief, I see no good reason to deny their claim that any entity with a capacity for belief also has a capacity for practical reasons-response. In  fact, this position is deeply embedded in Kant’s philosophy of mind, more so  than the view—even if he held it, which is controversial—that specifically reflective and critical reason is necessary for judgment. Thus Korsgaard’s ascription of beliefs to non-human animals while denying that these animals are assessable in terms of rationality, even if it were correct, would not be well understood as a Kantian view. To summarize this section’s concern: Korsgaard faces a dilemma. She must either deny that belief as such is a reasons-responsive mental state or deny that 53 We do not need to attribute a capacity to think about functions or causes to a frog in order to attribute to it a perceptual capacity whose successful operations depend on functional and causal relationships. Thus there must be a specific ground for claiming that a capacity to think about reasons must be attributed to an animal (such as a chimpanzee) if we are to attribute to it a capacity (such as belief) whose success depends on rational relationships. 54 Korsgaard (2008a, 2008b, 2008c) addresses these issues indirectly, by calling into question the intelligibility of acting for reasons not represented as principles. If the points in the main text here have force, they can also be leveraged back against the main conclusions of those essays. 55 See Burge (2010a: 154–210). Burge’s principal neo-Kantian targets are Strawson (1959) and Evans (1982); he also criticizes Davidson (1982) in a similar spirit.

208

Jon Garthoff

any animal lacking critical reason has beliefs. The first option issues an enormous philosophical promissory note, given the numerous failed attempts during the twentieth century to explicate belief as a non-rational state; and the first option is anyway peculiar for a Kantian, since the view that belief is constitutively responsive to reasons is both more clearly held by Kant and more central to Kant’s understanding of mind than is the view that belief constitutively involves representation of reasons. The second option is at variance with ordinary practice and is also in my view perilously close to empirical refutation. If we deny that belief constitutively involves representation of reasons, however, then we no longer face this dilemma. If it is true both (i) that belief constitutively involves responsiveness to reasons and (ii) that belief does not constitutively involve representation of reasons, then it cannot be, as Korsgaard’s constructivism demands, that all reasons are constructed through the activity of critical reason.

10.6 Distinguishing Inferential Judgment from Critical Reason If the observations of the previous section are correct, then the capacity for judgment that is necessary for belief and the other propositional attitudes is constitutively a capacity to respond to reasons. As was previously noted, however, a capacity for judgment or propositional thought does not entail a capacity to represent reasons as reasons. That further representational capacity helps enable critical assessment of the quality of a judgment or inference. The ability to represent reasons as such is likely not sufficient for possession of critical reason—other necessary capacities may include a relatively sophisticated self-conception and a long memory—but it lies close to the heart of what critical reason is. As I mentioned in my explication of her view near the beginning of the chapter, I suspect Korsgaard denies that animal propositional attitudes entail a capacity for reasons-response in part due to her long-standing commitment to a distinctive strategy for vindicating the overridingness of moral obligations. According to this account, moral obligations override other considerations because non-moral considerations do not even constitute reasons except when they manifest in the psychology of individuals with moral obligations, namely critically rational individuals. Non-moral considerations accordingly do not gain status as reasons that favour one action rather than another except insofar as they are already vindicated as compatible with morality. This is how Korsgaard spells out and defends Kant’s famous commitment to the ‘priority of right’.56 If this account of the overridingness of moral considerations were correct, it 56 See CPrR, 5: 61–3. Rawls (1971: 31–2) popularized the phrase ‘priority of right’; see also Rawls (1988). See Korsgaard (1996a, 1996b, 1996c, 1996d) for her account of moral overridingness.

Against the Construction of Animal Ethical Standing

209

would provide an especially compelling defence of the primacy of morality in practical thinking. If the considerations of the preceding section have force, however, then Korsgaard’s strategy for vindicating the overridingness of moral obligations cannot be sustained. If (as is highly likely) a chimpanzee acts and thinks for reasons but cannot represent reasons as such, then a chimpanzee acts and thinks for reasons which are never vetted by critical reason and which are not constituted by any act of autonomous self-legislation. Fortunately for those of us who draw inspiration from broadly Kantian accounts of moral obligation, there are other strategies available for defending moral overridingness. While I cannot provide a full account here of what I take to be the best such defence, a brief sketch is perhaps helpful. The general dependence of reasons to do what is good for a being on reasons to have concern for the being itself—in other terms, the general dependence of the worth of individual well-being on the worth of individuals themselves—alone suffices to vindicate the exceptionless overridingness of moral obligations, regardless of the representational capacities of the individual in question.57 This general dependence can rule out the all-things-considered rationality of acting for the sake of an individual’s well-being at the expense of the worth of the individual itself. This strategy dovetails with the discussion of value in Section 4. Genuine worth came into the world, on the understanding I propose, with the coming to be of the first conscious being. It did not, however, come into the world in virtue of this being’s perceptions or intimations of value. Rather it came into the world with the being itself, as the first being that mattered of itself. Since it mattered, this being’s well-being also mattered. But since this was true regardless of the presentational or representational capacities or states (apart from consciousness itself) the being possessed, the existence of genuine worth did not depend on these specific mental capacities or on the idiosyncratic content of any of these states. On this view there remain substantive issues about what this being’s well-being consisted in; I have not made significant progress on those issues here.58 They can be addressed on their own merits, however, regardless of how exactly this being’s mental states related to biological goods like fitness or nutrition or to any species-relative form of life in which it participated. The strategy I suggest for defending the overridingness of moral obligation is compatible with Korsgaard’s view that moral obligations are constituted only by the activity of specifically critical reason.59 It simply denies all reasons are constituted this way, allowing instead that there are reasons constituted by non-critical

57 For further development of this strategy see Garthoff (2015). Velleman (1999) contains resources helpful for elaborating how such a strategy proceeds. 58 For discussion of well-being see Garthoff (2010a). 59 This theme runs throughout her work, but Korsgaard (2009) develops it most fully.

210

Jon Garthoff

judgment. In my view, however, this is anyway a more plausible constitutivist hypothesis. Responding to reasons is one thing; representing reasons, or reflecting on them, is another.60 Thus, as I have attempted to explain in previous work, we can acknowledge the indispensability of a robust understanding of moral obligation to ethical theory without endorsing either stage of Korsgaard’s constructivism.61 In fact, as I have explained in this and the previous section, careful consideration of the psychology of the more sophisticated non-human animals tells strongly against the second stage of her constructivist project. Accordingly, those who share Kant’s view that moral obligations override other practical considerations should pursue other strategies for defending this position.

10.7 Conclusion In the preceding, I discussed two of the three principal motivations for Korsgaard’s account of the value of animals and their lives: what she calls the ‘essential subjectivity’ of value, and the fit between that account and her independently articulated explication of the overridingness of moral obligation. I conclude with a few observations about the third main motivation I identified for Korsgaard’s account, namely her conviction that this account enables a naturalistic understanding of the practical realm in general. The argumentative force of this conviction flows in part from a certain conception of what rival positions maintain. Throughout her career, Korsgaard has often presented Moorean non-naturalism as a stalking horse to her own view, where juxtaposition with G. E. Moore’s view is intended to help explain relative virtues of her own position.62 Korsgaard’s view that final value is partly constituted by the operations of consciousness, for example, is in her recent work on animals contrasted with two rival views of the role of consciousness. One rival holds that consciousness merely makes possible something—pleasure, perhaps—whose good is intelligible independently from the well-functioning of organisms. The other rival view holds that consciousness merely makes possible awareness of the organism’s biological well-functioning, which is independently intelligible as valuable.63 I concur with Korsgaard’s claim that her position is superior to both these alternatives. Where I dissent is from the thought that these are the most important rival positions to consider. 60 There is a parallel between this point and Watson’s famous (and apt) criticism of Frankfurt’s account of freedom. See Frankfurt (1971) and Watson (1975). 61 See especially Garthoff (2015). 62 Korsgaard discusses Moorean non-naturalism as a stalking horse in many places, but the theme is most prominent in Korsgaard (1996a). See Moore (1903) for the fullest statement of his non-naturalism. The point in the text emulates, with respect to the role of Moorean non-naturalism in ethics, Burge’s observation that Cartesian dualism is the ‘bogeyman’ of the philosophy of mind—see Burge (2010a: 297). 63 Korsgaard (unpublished), Lecture 1, 29–34.

Against the Construction of Animal Ethical Standing

211

I think against all these views that consciousness is crucial in the foundations of ethics because being conscious is constitutively both necessary and sufficient for mattering as such. This view is far more ecumenical than Korsgaard’s view about what the well-being of a conscious being consists in. It is compatible, for example, with the view that pleasure is as such good for any conscious being. It is also compatible with the view that psychological flourishing is as such good for any conscious being, regardless of whether or how this psychological flourishing relates to a species-specific form of life. It is furthermore compatible with naturalism, so long as both consciousness itself and the constituents of the well-being of all conscious individuals—such as pleasure and psychological flourishing—occur naturally. But there is no good reason to deny this, since we anyway should be committed to the emergence of consciousness in the natural world. The same point holds for other capacities crucial for developing a full ethical theory, including judgment and critical reason. This view concurs with Korsgaard that ‘good for’ must be understood in relation to the nature of the individual whose good is in question, and perhaps also that genuine worth (unlike representation) has subjective origins. The view dissents in insisting that theories of value should aspire to overcoming this obstacle to understanding reasons and values as objectively real. The view concurs with Korsgaard that moral obligations override other practical considerations. The view dissents by explaining this in terms of the value of beings with a capacity (consciousness), rather than as the product or result of a necessary exercise of a capacity (mutual self-legislation using critical reason). These are only beginning statements of a rival position. It is best if various views that share these or nearby motivations are articulated, developed, criticized, and compared, and ethical theorists owe Korsgaard a debt of gratitude for blazing a trail in this direction. But however these alternative views are developed, I suspect the best ones will involve greater departures from the idea of the kingdom of ends, which models ethics on Enlightenment political ideals. This is perhaps to be expected, however, for we now have available intellectual resources Kant did not foresee: mature teleological sciences. Modern biology and psychology use teleological notions indispensably, as unreduced elements of their frequently impressive scientific explanations. Modern physics and cosmology do not. We should not shy from using teleological concepts in ethical theory from fear of non-naturalism or envy of physics. Naturalism is neither physicalism nor excision of teleology. Appropriately understood, naturalism is integration into systematic empirical explanations.64 The scientific vindication of teleology does not alone support the claim that anything matters, or that reasons exist. Korsgaard has done as much as anyone to direct our attention to the need for explanations beyond biological

64 For allied observations about naturalism, see Burge (2010a: 296–8).

212

Jon Garthoff

teleology for how there could be genuine worth or reasons. But consciousness and some types of psychological teleology are not biological teleology; they enjoy a much stronger prima facie case for ethical significance. To claim an action is to-be-done is to claim a well-functioning reasons-responsive agent would do it. Such a claim is no more unscientific, and no more non-naturalistic, than the claim that an animal’s reproduction functionally contributes to its fitness or the claim that an animal’s seeing functionally contributes to the accuracy of its representations. Thus we need not advert fundamentally to contingent or idiosyncratic contents of psychological states in order to ground ethics in psychological capacities naturalistically understood. Neither need we eschew realism for constructivism simply because biological teleology does not support ethics.65 Value realism is grounded in the natural occurrence of functional organization; worth realism is grounded in the natural occurrence of consciousness; ethical realism is grounded in the natural teleology of judgment; and moral realism is grounded in the natural teleology of specifically critical reason. In a dispute between these realisms and Kantian constructivism, all parties deny Moorean non-naturalism. Broadly Kantian or Aristotelian ethical theories can and should proceed without the empiricist’s fear of that spectre. In particular these modestly rationalist approaches to ethical theory can take animal psychological capacities on their own terms, and can assess which of them give rise to reasons or genuine worth, without first detouring through the exercise of the specifically critical capacities of mature humans.

65 For a classic discussion of this theme, see Nagel (1978).

11

Hope and Despair at the Kantian Chicken Factory Moral Arguments about Making a Difference Andrew Chignell

11.1 Kant and Kantians on animals Kant famously—notoriously—claims that we have no direct duties to non-human animals. Like plants, animals1 are ‘things’ with only ‘relative worth’, he says, and so it is permissible to treat them (their bodies, secretions, and eggs) as mere means to our ends of feeding, clothing, and entertaining ourselves. The impermissible actions regarding animals are those that disrespect their owners or desensitize us to the suffering of the sentient creatures who do morally count—namely, the rational ones. In short: for Kant, our direct duties are to persons; any duties to other animals are indirect and go via our duties to persons.2 Many contemporary philosophers—including many Kantians—regard Kant’s arguments about animals (like his arguments about gender,3 race,4 masturbation,5 homosexuality,6 and political revolution7) as tied to his cultural context in a way that makes them (in letter if not in spirit) obsolete and sometimes repugnant.8 A few contemporary philosophers, most prominently Tom Regan and Christine Korsgaard, have developed alternative approaches that are broadly Kantian in flavour and yet come to quite different conclusions about our obligations to animals. They are broadly Kantian because they cite the kinds of considerations that Kant himself uses in establishing the dignity of human beings. They are not narrowly Kantian because they conclude that we do have direct duties to animals, and that we may not treat them merely as means to our ends. Regan (1983) does this by arguing that animals are, like us, sentient ‘subjects of a life’ (this is a technical term for him) and that this is evidence that they have an 1 2 4 7 8

In what follows I will use ‘animals’ to refer to non-human animals. G 4:427–8; Con. 8:110ff; MM 6:241, 443–4; EC 27:459, 710. 3 See Varden (2017). See Allais (2016). 5 See Kielkopf (1997). 6 See Altman (2010). See Flikschuh (2018). But for a positive assessment of Kant’s views about animals, see Wilson (2008).

Andrew Chignell, Hope and Despair at the Kantian Chicken Factory: Moral Arguments about Making a Difference In: Kant and Animals. Edited by: John J. Callanan and Lucy Allais, Oxford University Press (2020). © Andrew Chignell. DOI: 10.1093/oso/9780198859918.003.0012

214

Andrew Chignell

irreducible ‘inherent value’ that confers moral standing and welfare rights. Regan’s view has been influential (often playing the role of the ‘Kantian’ or ‘deontological’ approach in animal ethics anthologies), but critics have wondered how to understand the metaphysics and epistemology of the obscure conception of value involved. Korsgaard (2018) tries to avoid this concern by developing an explicitly ‘naturalistic’ Kantian-constructivist approach. She starts with the Aristotelian point that animals have ‘tethered goods’—goods for them, just as we do. One of the main goods for any sentient animal is to live in safe, comfortable environs; another is to have a full life—one that is not radically curtailed, even painlessly. Obviously we humans take, and require others to take, what is good for us as a defeasible basis for making moral claims on ourselves and others. Korsgaard regards such taking and requiring as conferring ‘absolute’ value on our tethered goods, and as ‘claiming standing’ for ourselves as ends rather than mere means (2018: 139). But there is no non-arbitrary basis, she says, for refusing to take the tethered goods of other sentient creatures into account in this way, even if they are not able to do so themselves. Thus when we consider the reason why Kant thinks we must claim the standing of ‘end-in-itself ’ for ourselves, we will see that we must claim that standing for the other animals as well. (2018: 130)

I have examined Korsgaard’s innovative argument in detail elsewhere (Chignell, forthcoming 2020), but one major problem involves the following premise (this is my paraphrase rather than a quotation): Parity: If there is no difference between us and the other animals with respect to either the having of tethered goods or the importance of those goods, and if we treat our tethered goods as goods absolutely (thereby making them good absolutely), then we ought to treat other animals’ tethered goods as good absolutely (thereby making them good absolutely).

Korsgaard argues over the course of several chapters that the two conjuncts of the antecedent are satisfied, and then spends the central chapter of the book arguing for the conditional as a whole. But an opponent could plausibly suggest that the reason we take our tethered goods to be part of the absolute good is not because they are tethered goods simpliciter, but because they are our tethered goods. In other words, the opponent could plausibly suggest that we treat human goods (of a full life, health, safety, capacities, opportunity, etc.) as part of the absolute good because, well, they are goods for us. Interestingly, Korsgaard herself seems to articulate the premise this way in places: ‘I have no other reason for taking my end to be good absolutely, than the fact that it is good for me’ (2018: 144).

Hope and Despair at the Kantian Chicken Factory

215

Instead of Parity, then, it looks like we have Parity*: If there is no difference between us and the other animals with respect to either the having of tethered goods or the importance of those goods, and if we treat our tethered goods as goods absolutely because they are our tethered goods (thereby making them good absolutely), then we ought to treat other animals’ tethered goods as good absolutely (thereby making them good absolutely).

Parity arguments are tricky, but now this looks invalid: the italicized phrase in Parity* adds something to the antecedent that makes the conditional as whole implausible. Genuine parity would require a consequent like ‘. . . then the other animals ought to treat their tethered goods as goods absolutely because they are their tethered goods (thereby making them good absolutely)’ But that, unfortunately, is not something that our fellow creatures can do. In what follows, my goal is not to rehearse Kant’s narrow views about our duties to animals; nor is it to examine further these broadly Kantian efforts to improve on Kant’s ideas.9 Instead, I propose to draw on another part of Kant’s philosophy altogether—namely, his moral psychology of hope and despair—to develop a different line of thought about our relationship to animal products. For these purposes I will simply start by assuming that there are sound arguments (of one of these broadly Kantian varieties, perhaps, or of a more consequentialist variety) for the claim that: Don’t Farm: It is morally wrong to raise, kill, and harvest animals for the purpose of selling their bodies, secretions, and eggs if there are readily available alternatives for us to eat and wear.

Now consider an average North American consumer of middling means—call him Oppy—who likes meat and cheese and omelettes and leather. He would purchase these products if he could, but he believes that Don’t Farm is true, and he assumes, like many of us, that Don’t Farm implies: Don’t Purchase: It is morally wrong to purchase the bodies, secretions, or eggs of farmed animals if there are readily available alternatives for us to eat and wear.

Recently, however, Oppy has confronted some sobering empirical facts that seem to problematize this natural move from Don’t Farm to Don’t Purchase. The sobering empirical facts, roughly put, are that the sheer size and complexity of the

9 For more of the latter, see Varden (this volume).

216

Andrew Chignell

industrial system that brings 99% of the animal products to our refrigerators and closets—and, in particular, the oversupply and buffers in many of its supply chains—make it exceedingly unlikely that an individual’s choice not to purchase will have any effect at all on the system, whether positive or negative. Despite his unwavering commitment to Don’t Farm, then, Oppy has become an opportunist: he thinks it is okay on occasion to privately enjoy and wear some of his favourite farm animal products, since (as long as he is purchasing from the industrial system) doing so almost certainly doesn’t make a difference.10 Michael Martin (1976) and R.G. Frey (1985)—and much more recently Mark Bryant Budolfson (2016, 2019) and Eliot Michaelson (2016a)—have argued that difference-making concerns like these are pressing for consequentialist advocates of Don’t Purchase. This is because act consequentialism (‘consequentialism’ hereafter) endorses the following principle: Causal Consequences Principle (CCP): Action A is morally wrong only if the causal consequences of A are worse than they would have been if some readily available alternative to A had been performed.11

‘Worse’ here is traditionally construed in a welfarist way that includes all sentient animals.12 So given CCP, if a private13 purchase of a chicken sandwich that is wrongfully produced almost certainly doesn’t cause any additional loss in welfare, and clearly does lead to the goods of pleasure and nourishment, then it cannot be morally wrong. (It might even be morally required!) This sets up a trilemma for the consequentialist: give up CCP, give up welfarism, or give up Don’t Purchase and get comfortable with opportunism. The whole problematic is now referred to under the label of the inefficacy (or futility or causal impotence) objection. Michaelson (2016b) and Budolfson (unpublished) have argued that the inefficacy objection also makes things difficult for some non-consequentialist efforts to move from Don’t Farm to Don’t Purchase. For instance, it makes things difficult for views on which only acts that involve or lead to the infringement or violation of rights count as morally wrong. That’s because the sobering empirical facts indicate that a choice to purchase the products of a gigantic, insensitive system is

10 For the term ‘opportunistic carnivore’, see Almeida/Bernstein (2000). For more discussion, see (Chignell 2016). 11 Compare Almeida/Bernstein (2000: 206) who state this as a biconditional. I use a necessary condition here, since some consequentialisms will have different ways of filling out the account of moral wrongness. See Sinnott-Armstrong (2015). 12 Some welfarist views restrict the morally relevant outcomes to human animals. Joseph Raz calls this ‘the humanistic principle’: ‘the explanation and justification of the goodness or badness of anything derives ultimately from its contribution, actual or possible, to human life and its quality’ (1986: 194). Here I am working with a welfarist picture that is not speciesist in this way. 13 The privacy condition is meant to deflect concerns about the expressive significance of one’s purchasing behaviour and the influence it might have on the behaviour of others.

Hope and Despair at the Kantian Chicken Factory

217

extremely unlikely to infringe or violate any rights. And most non-consequentialists of this variety will allow, in other contexts, that it is permissible to perform actions that are extremely unlikely to infringe or violate rights. Driving to the store, for instance, may end up in an accident that puts other people in a hospital; provided this result is extremely unlikely, however, it is still permissible to drive to the store. Other varieties of non-consequentialism (certain forms of virtue theory, symbolic or expressive value theory, divine command theory, and even Parfit’s neoKantianism14) are not as obviously vulnerable to the inefficacy objection, however. The decision privately to purchase the products of a morally wrongful system may be vicious, symbolically unacceptable, expressively noxious, divinely proscribed, or categorically irrational even if it has no other negative outcomes.15 In what follows, I will look a bit further at the sobering empirical facts and how they problematize the move from Don’t Farm to Don’t Purchase (Section 11.2). In Section 11.3, though, I want to turn our attention away from Oppy in order to focus on another character: Hope. As we will see, Hope is someone who likes animal products just as much as Oppy does but takes herself to have good reasons for thinking that Don’t Purchase is true. These reasons might be grounded in one of the non-consequentialist theories just mentioned, and in any case do not hang on whether her action makes a causal difference to welfare outcomes. All the same, Hope often—and quite understandably—finds her efforts to adhere to Don’t Purchase threatened by her awareness of the sobering empirical facts. And she’s not alone in this: for all but the most impressively stoical among us, the awareness that our boycotting efforts almost certainly make no causal difference can pose a serious psychological threat to our ongoing moral resolve, even if our reasons for boycotting do not depend on the boycott making a causal difference. Here is another way to put the point. The usual version of the inefficacy problem is moral-conceptual: it starts with the intuitive idea that purchasing a certain product is wrong, and then shows that various ethical theories imply the opposite in circumstances where individual choices are almost certainly inefficacious. The version of the problem that I develop in Section 11.3, by contrast, is moralpsychological: it, too, starts with the conviction that purchasing is wrong, and then points out that people who both recognize this and are still inclined to do it will often be psychologically demoralized under conditions of perceived inefficacy such that they are unable to resist.

14 See Parfit’s reformulations of the Categorical Imperative in response to the ‘problem of imperceptible differences’ in his 2011 (341). 15 I call these non-consequentialist theories, but the relevant factors can obviously be ‘consequentialized’ in a non-welfarist axiology such that the increases in exercises of virtue, symbolic value, obedience to God, practical consistency, etc. become the consequences with respect to which we ought to seek the optimific result. Thanks to Daniel Rubio and Ryan Darr for conversation here. See Dreier (1993).

218

Andrew Chignell

Interestingly, Kant has something much more useful to say about this second set of concerns than he does about our direct obligations to animals. Or at least that’s what I will suggest below. In Section 11.4, I reconstruct one version of his famous ‘moral proof ’ of God’s existence and the future life of the soul, one that he developed in the 1790’s towards the end of his career (just as his hopes regarding the French Revolution were giving way to despair). I call it the ‘moral-psychological proof ’ in order to highlight the fact that—unlike the better-known versions of the moral proof—it relies on empirical premises about our tendencies to hope and despair, and the concomitant effects on our psychological resolve. Like the earlier and more famous versions of the moral proof, the conclusion of the moralpsychological version is not that God and the future life exist, but rather that we are morally justified in having faith (the German term is ‘Glaube’16) that God and the future life exist. After making this detour through Kant’s theistic moral psychology, my ultimate goal (in Section 11.5) is to see whether a secular analogue of the proof applies to people in contemporary industrial contexts who, like Hope, have trouble sustaining their resolve in the face of the sobering empirical facts.17

11.2 Inefficacy at KCF Inefficacy concerns arise in many industrial contexts (leather, coffee, fabrics, energy, etc.), but I propose to focus on the chicken system in a world very much like our own: one in which 66 billion chickens are processed each year (this means that around 10,000 birds will have been harvested while you were reading this very sentence).18 Let’s also suppose that key parts of the massive supply systems in this world are ‘lumpy’ rather than linear. In other words, in these parts of the chain, supply responds to demand in large lots or ‘lumps’ rather than in one-byone adjustments.19 This means that there will be demand thresholds—upper and lower—that must be crossed in order to trigger any change in supply at all. It also 16 Kant uses this term in a variety of contexts, and there is no good English translation of it. In other work I have used the Germanic-looking term ‘Belief ’ (see 2007b); here, however, I’ll use ‘faith’, since that is standard terminology in discussions of Kant’s moral proof of the rationality of theism. 17 I have tried to set up Oppy’s situation more precisely here than I did in my (2016) and (2018), but some of the argumentation is the same. I am grateful to the editors of both volumes for letting me incorporate revisions of that material in what follows. 18 That was the statistic in our world for 2017, according to the Food and Animal Organization of the United Nations. Data can be found here: http://www.fao.org/faostat/en/#data/QL, selected for World + (Total), Producing Animals/Slaughtered, Meat, Chicken > List, 2017. The number has certainly gone up since then. 19 See Michaelson (2016a) for a discussion of the extent to which the inefficacy objection hangs on contingent facts about the actual food supply. For doubts, partly from an economist’s point of view, about how much insensitivity there is in the actual world’s system, see Halteman and McMullen (2018). But even if systems in the actual world are not insensitive in this way, they certainly could be. More importantly for present purposes, they could reasonably be perceived to be that way by would-be moral agents.

Hope and Despair at the Kantian Chicken Factory

219

means that the system will be insensitive to slight changes in demand, and thus that an individual’s occasional choice privately to purchase a small poultry product is very unlikely to have an effect on the conditions of the animals and workers in the system, or the environment and wildlife affected by it. Now let’s consider a massive global fast food company in this nearby world— one that is run by people who share the narrow Kantian view about our direct duties to animals (viz. that we have none). This company sources chicken from vast industrial operations that carefully comply with human health and worker regulations (like good Kantians) but uses intensive factory techniques on the birds (dark crowded indoor sheds, beak-clippers, battery cages, automated throat-cutters, scalding tanks, etc.) in order to maximize yield. We can call this company the Kantian Chicken Factory: ‘KCF’ for short. Let’s suppose that in this world (which, again, is very much like our own) the regional KCF has a policy of ordering 1,000,000 chickens every month from its supplier in order to meet an average monthly demand just shy of 20 million chicken sandwiches. (I am estimating that a fast-food chicken patty contains around 1/20th of the edible parts of a chicken.20) More specifically, the policy says that if the demand in any given month is between 19,900,000 and 20 million sandwiches, then KCF won’t change its usual order: their supply chain managers are prepared to waste (or donate, when they can) that much oversupply in a month if they think the average demand will continue to be around 20 million sandwiches.21 That’s because, given how cheap the government-subsidized chicken fed on government-subsidized corn is in this nearby world, and how expensive it would be to introduce mechanisms that would allow them to track demand more closely, and the unpredictability of product recalls, the KCF executives figure it is worse to face a shortage in a given month than it is to throw out some extra supply. Moreover, average monthly demand in this region has been within these two thresholds every month for the past ten years. Still, the policy is not entirely insensitive to market changes. If the number of sandwich orders in a given month falls below the 19,900,000 threshold, this will trigger a ‘lump’ reduction in their order for the following month: they will order 995,000 chickens

20 The average broiler chicken weighs around five pounds. The average chicken sandwich is a quarter-pound or less. So that’s around twenty sandwiches per bird. But a good portion of the chicken isn’t edible, and a lot of fat is lost in the cooking process. On the other hand, a chicken sandwich patty is not entirely made of chicken. The Canadian Broadcasting Company recently sponsored efforts by DNA researcher Matt Harnden (of the Trent University Wildlife Forensic DNA Laboratory) to analyse the contents of various fast food chicken sandwiches. They ranged between having 43% chicken DNA content (Subway) and 89% chicken DNA content (Wendy’s). The rest was a mixture of soy and other additives. https://www.cbc.ca/news/business/marketplace-chicken-fast-food-1.3993967. So it seems like 1/20 is a decent estimate. 21 KCF is a fictional company, but a real-world company with a similar name processes around 2.5 million chickens per day (i.e. 850 million chickens per year). So although the numbers I’m using here may seem large, they are probably much too small.

220

Andrew Chignell

instead of 1 million from the supplier. They have similar thresholds every 100,000 sandwiches below that. So now, finally, consider Oppy, who is morally opposed to farming chicken generally, and thus finds the industrial chicken system morally atrocious. He has walked into his local food court for a private lunch, and is trying to decide whether to purchase one of the delicious KCF spicy mesquite chicken sandwiches. Given our stipulations, if the number of other chicken sandwich orders during the month ends up being more than 19,899,999 but less than 20 million—as it has for the past decade—then Oppy’s decision won’t make any causal difference to the number of actual chickens harvested. For if he abstains, they will just dispose of or donate the patty that he would have purchased, and they won’t change their monthly order from the supplier.22 Likewise if the total number is below 18,999,999 but above 17,999,999, then his decision also won’t make a difference. And so on. It is only if Oppy happens to be at the food court during a month in which the number of other sandwich orders, not counting his, is precisely 18,999,999 (or 17,999,999, or 16,999,999, etc.) that his order will have any effect at all. But if he is on such a threshold, and if he orders something other than a chicken sandwich, then his action will trigger a big lump reduction and KCF will order 5,000 fewer chickens from the regional supplier next month.23 The situation here is a difficult one for Oppy given his commitment to Don’t Farm and his taste for spicy mesquite. But here we can offer him a standard reply from the literature on expected utility. For even if the supply chain is lumpy like this, a consumer can never know when he is at a pivot or threshold point, and so he should just equate the expected effect of his single choice with the average effect of all the choices between thresholds. We have already stipulated that every ‘lump’ order is for 5,000 chickens, and that these are only triggered every 100,000 sandwich purchases. So the average effect of each purchase is 1/100,000 x 5,000 = 1/20. In other words, the effect of Oppy’s individual sandwich purchase is precisely what one would predict, given the amount of chicken in a patty—namely, 1/20th of a chicken. It’s then an empirical question whether the thought of occasionally saving 1/20th of a chicken (or of saving one chicken every 20 times he chooses the spicy Beyond patty) is sufficient to motivate Oppy’s abstention. Many philosophers who offer a reply like this argue that it should be.24 22 I’m setting aside the fact that Oppy might order a salad instead, and thereby signal increased demand for plants instead of animals. 23 This is essentially a large-scale version of the ‘Three-in-a-Boat’ game (named after a J.K. Jerome short story): If only one of us rows, the boat goes in a circle; if the two of you row, then my rowing won’t make a difference. So I should only bother rowing if exactly one of you is going to join me. Bovens (2015) argues that the ‘tragedy of the commons’ is best modelled as this kind of game rather than a Prisoner’s Dilemma. Thanks to Bovens for helpful discussion here. 24 For versions of this argument, see Singer (1980), Norcross (2004), Kagan (2011), and Norcross (draft). Despite some similarities, this chicken sandwich scenario poses problems for both rationality and resolve that are somewhat different from real-world voting situations. When voting for president,

Hope and Despair at the Kantian Chicken Factory

221

The expected utility reply would make sense in a system that was perfectly or at least highly efficient. But the reply is threatened by the sobering empirical facts about KCF’s tolerance for oversupply and the decade-long trend in the region. Remember, KCF has received between 19,900,000 and 20 million chicken sandwich orders every month for ten years now, and so it is extremely likely that this will happen again this month. They also have a standing policy of throwing out or donating any oversupply between thresholds. This means there is a tolerance for up to 99,999 sandwiches’ worth of oversupply in the system—a buffer that keeps them from ever falling short of demand. But it also means that an individual decision to purchase or abstain in a given month is likely to be drawing from that buffer, rather than moving the system along to the next threshold. This in turn suggests that it cannot simply have the ‘average effect’ of a purchase between thresholds. In a system with buffers, we have to consider not just what it takes to get from one threshold to another, but also the size of the buffer and the regional trend. Some people may find it implausible that a real-world analogue of KCF would tolerate a buffer as large as 99,999 patties per month.25 But even if that’s correct and the buffer is smaller, it is almost certainly not the only one. In many industrial systems, even in those striving for vertical integration, there will be a series of buffers of this sort—a series of points along the chain where waste is tolerated or absorbed. KCF’s suppliers, for instance, will presumably also have some sort of lumpy threshold policy vis-à-vis the farmers who raise the chickens, and be prepared to absorb some decrease in demand before reducing their order. Similarly, the farmer gets the chicks from an incubator facility, which itself will have a threshold policy that tolerates a certain amount of oversupply. Moreover, in a market as big as the chicken market, the suppliers almost certainly have other customers who will be able to pick up the slack in a given period, and so may not tell the farmer to produce 5,000 fewer chickens after all, even if KCF occasionally crosses the relevant threshold; likewise at the level of farmer and incubator, and so on: it’s a network of buffers all the way down. If this description is coherent, then it looks like there is not a 1/20 chance but rather a much smaller chance that Oppy, standing at the other-worldly food court that day, is at a point in the industrial supply chain that will make his choice

for instance, one might not make a difference to the outcome but still be able to signal support, or increase the ‘manifest normative mandate’ of the winner (see Guerrero (2010). Something similar cannot be said about someone who abstains from a chicken sandwich in a lumpy supply system where her purchase would simply be taken from the buffer. For this reason I think assimilating inefficacy cases to voting cases (in the way that, e.g. Norcross (draft) and Nefsky (2017) do) can be misleading. 25 Again, for more discussion of the situation in the actual world, see Halteman and McMullen (2018).

222

Andrew Chignell

efficacious. If we add in other variables26 and supply-chain ‘noise’, then the chance that he will buy a threshold sandwich becomes, in Budolfson’s words, ‘infinitesimal’ (2016: 208). Calculating the expected effect for each customer over the course of the month would be difficult, given all these factors, but it is clearly nowhere near as high as 1/20th of a chicken.27 This is an interesting thought-experiment in itself. But our actual world also seems regrettably similar to Oppy’s. If that’s right, then the sobering empirical facts problematize the intuitive move from Don’t Farm to Don’t Purchase in our world too. Rabbi Yanklowitz exemplifies this intuitive move in The Jewish Vegan when he states that ‘it is not hard to buy a leather-free belt to spare animals from unnecessary suffering’ (2015: 27). Forty years earlier, Peter Singer likewise suggested that: [b]ecoming a vegetarian is not merely a symbolic gesture . . . Becoming a vegetarian is a highly practical and effective step one can take toward ending both the killing of non-human animals and the infliction of suffering upon them. (1975: 168–9).

Most of us share Yanklowitz’s and Singer’s sense that we have strong moral reasons to boycott products of wrongful activity on the part of others, i.e. that there is a way to defend not just Don’t Farm but also Don’t Purchase. However, if what morally matters is making a causal difference to the welfare of the creatures involved, then in our world—as in Oppy’s—lots of individual purchases of industrial animal products may be neither here nor there.28

26 For instance: cultural trends can make the demand for certain cuts of poultry go up (turkey breast at Thanksgiving, chicken wings during years when the local football team does well, etc.). But animals come in wholes, and so if wings are in high demand this year, a decision not to eat a chicken sandwich (which is not made of wings) is even less likely to have any effect on how many chickens are produced and processed, since retailers are demanding as many wings as possible. Thanks to conversations with Cornell food scientist and poultry expert Joe Regenstein here; see also Parcell and Pierce (2000). 27 Halteman and McMullen (2018) acknowledge these facts about buffers in the real-world poultry supply, but then note that the consumer will ‘know little to nothing about the distance to a threshold at their particular retailer, and even less further down the supply chain’. This is true, but in the noisy, buffered market situation described here, a consumer’s ignorance of precisely where she is relative to the next threshold seems moot—she only needs to recognize that the chance of being on any threshold is extremely small. Someone like Oppy can easily come to know that the vastly most likely outcome, given the buffer, the noise, and the trend, is that his decision to abstain will increase that month’s wasted oversupply by precisely one patty. 28 Note that the sobering empirical facts also suggest that the choice to purchase small, familyfarmed products is much more likely to make a difference. So if someone like Oppy thinks that Don’t Farm holds across the board, he should only indulge his opportunism with products from the industrial system.

Hope and Despair at the Kantian Chicken Factory

223

11.3 Difference-making and Despair The reasoning that motivates the negative conclusion we just reached (and the reasoning that Oppy exploits when he goes opportunist) is underwritten by something like this thought: Badness as a Reason to Abstain: The badness of a production practice is a reason to abstain from purchasing its products only if abstaining has a non-negligible chance of making a positive difference with respect to the badness of that practice.

The more general principle here is: Badness as a Reason to Act: The badness of an outcome O is a reason to perform action A only if A has a non-negligible chance of making a positive difference with respect to O.

In keeping with CCP, we are also construing what it is to ‘make a difference’ in a causal way: Difference-Making, Causal: An action A makes a difference with respect to outcome O just in case performing A causally leads to a change in O.

It would be worth thinking more about whether consequentialist ethical theories can rebut the conceptual version of the inefficacy objection or rule out opportunistic purchasing without giving up one or more of their core principles. But I don’t propose to go further into that debate here.29 Instead, I want to turn to the psychological version of the inefficacy problem, and the broadly Kantian argument from the moral need to avoid demoralization and despair. So let’s now consider Hope, and suppose that she has been convinced by one of the many arguments in the literature that seek to bridge the gap from Don’t Farm to Don’t Purchase without appealing to causal difference-making. One family of such arguments invokes the symbolic value of ‘standing with the good’ and avoiding ‘moral taint’ by not consuming the products of a morally objectionable practice (see Hill 1979; Appiah 1986; Boey 2016). Another invokes a theory of ‘complicity’ or a ‘don’t benefit from wrongness’ principle that prohibits purchasing even if it makes no causal difference (Martin 2016; McPherson 2016). Another family of arguments appeals to virtue-theoretic or psychological considerations that don’t require any causal difference-making (Nobis 2002; Halteman/Halteman 29 I attempted some of that in Chignell (2016). See Nefsky (2018) for an overview of such efforts.

224

Andrew Chignell

Zwart 2016; Bramble 2016). Yet another emphasizes that an individual can be a joint cause of bad outcomes even without making a causal difference, and that the badness of being a joint cause can sometimes be a reason to abstain (Parfit 1984; Harman 2016; Albertzart 2019). Difference-making concerns are not so easily dispatched, however. Even if Hope accepts Don’t Purchase on one of these other grounds, the sobering empirical facts according to which her food choices with respect to the industrial system—even over an entire lifetime—are unlikely to make a significant difference can still have a profound psychological effect on her moral resolve. In other words, recognition of the sobering facts might lead to a very natural kind of demoralization: (D1) General discouragement and dejection in the face of the very long odds of making a significant positive difference with respect to the outcomes in question.

Often this leads to another and more serious kind of de-moralization: (D2) Loss of psychological resolve required to do what the agent still takes herself to have moral reason to do.

Let’s call the state of being doubly demoralized in this way despair. When we fall into despair, we are liable to give up boycotts and revert to purchasing whatever is most convenient, tasty, and affordable—despite any ongoing moral qualms.30 Kant was acutely aware of our psychological vulnerability to perceived inefficacy in the face of large-scale evil and injustice. In one of his most lyrical passages in third Critique, he considers the psychology of an atheistic but ‘righteous man (like Spinoza)’ who does not demand any advantage for himself from his conformity to the moral law, whether in this world or another; rather, he would simply and unselfishly bring about the good to which that holy law directs all his powers.

30 This is clearly anecdotal and conjectural. To make the case we would need to look at sociological studies of consumer and activist behaviour in response to perceptions of inefficacy. Interestingly, the Meat Institute itself cites recent CNN and USDA survey data showing that of the approximately 5 percent of Americans who claim to be vegetarians, around 65 percent will confess to having eaten meat products in the past 24 hours. The Meat Institute’s conclusion? ‘Bottom line: meat is amazing— and irresistible’. (North American Meat Institute, n.d., 2) https://www.meatinstitute.org/index. php?ht=d/sp/i/101931/pid/101931, accessed 1 November 2019.

Hope and Despair at the Kantian Chicken Factory

225

Such a person’s resolve, Kant says, is still liable to be sapped by the perception that his own moral strivings, as well as those of others, are inefficacious against largescale structural evil and injustice: But his strivings (Bestreben) have limits . . . Deceit, violence, and envy always surround him, even though he is himself honest, peaceable, and benevolent. The other righteous people that he encounters at times will, in spite of all their worthiness to be happy, nevertheless be subject by nature, which pays no respect to that, to all the evils (Übeln) of poverty, illnesses, and untimely death, just like all the other animals on earth. It will always remain so until one wide grave engulfs them all together (whether honest or dishonest, here it makes no difference) and hurls them, the very ones who were capable of believing that they were the final purpose (Endzweck) of all creation, back into the abyss of the purposeless chaos of matter (Schlund des zwecklosen Chaos der Materie) from which they all were drawn. (CPJ 5:452)

In the end, when faced with the abyss in this way, the righteous Spinoza has two options: either he will ‘certainly have to give up his end [i.e. of being righteous] as impossible’ or ‘he will have to accept (annehmen) the existence of a moral author of the world (Welturheber), i.e. of God, from a practical point of view’.31 I read ‘impossible’ here psychologically: even someone as righteous as Spinoza simply cannot maintain his resolve without some source of hope that, ultimately, justice will prevail.32 These reflections on demoralization suggest that, in addition to the sobering empirical facts about the industrial food system, there is another empirical question that we need to consider in connection with inefficacy. It is an empiricalpsychological question about which sorts of background commitments or mindsets do better by way of helping ordinary agents avoid despair in the face of perceived inefficacy. Kant clearly thought that someone who accepts the existence of a benevolent deity working behind the scenes to bend the arc of history towards justice—presumably also recording and rewarding our efforts—will have more resources to sustain those efforts, despite long-term setbacks and perceived inefficacy. Kant is not alone in this: supersensible commitments are clearly effective in both providing motivation and sustaining resolve, and many ‘moral arguments’ along these lines can be found throughout the nineteenth-century Pragmatist, Idealist, and Existentialist traditions. But Kant is the éminence grise: ‘What may I hope?’, he says in the Critique, is one of the three questions that motivate his 31 ‘Accept’ here is ‘annehmen’, the verb Kant often associates with the noun ‘Glaube’ (faith). 32 Camus (1942) famously rejects this: he considers the same predicament and says that we must both accept the demands of the moral law and embrace the absurdity of a world in which justice never prevails: ‘One must imagine Sisyphus happy’.

226

Andrew Chignell

entire philosophy. So in the next section I propose to look more closely at the structure of his moral-psychological proof. After that we can consider whether some analogous forms of reasoning can help with psychological inefficacy problems in contemporary consumer contexts.

11.4 Moral Arguments for Theoretical Conclusions Kant’s project in Religion Within the Bounds of Reason Alone (1793) is to discover the rational essence of religion—the doctrines and practices that can survive examination by the tribunal of universal rationality. The result is called ‘moral religion’, ‘rational religion’, or ‘rational faith’ (Vernunftglaube) in order to distinguish it from creedal or enthusiastic forms based in special revelation or alleged mystical experience. At one point, he offers the following as the rational essence of the doctrine of providence: Each must, on the contrary, so conduct himself as if everything depended on him. Only on this condition may he hope that a higher wisdom will provide the fulfilment of his well-intentioned effort. (Rel. 6:101, my emphasis)

The passage displays what might be called a consequence-dependent moral psychology. Kant is famously not a consequentialist in ethics generally: ‘The fulfillment of duty consists in the form of the earnest will, not in the mediating causes of success’ (CPJ 5:451). Nor is he a divine command theorist: ‘Morality . . . needs neither the idea of another being above him in order to recognize his duty nor as an incentive anything other than the law itself in order to observe it’ (Rel. 6:3). But Kant also recognizes that it is often psychologically crucial to be able to hope that our actions will have their intended results, even if by way of a supersensible mechanism (‘a higher wisdom’). This fits with a general pattern of sensitivity to embodied, empirical psychology in Kant’s ethical thought—i.e. sensitivity to the fact that it is hard to keep soldiering on in the moral life without the hope that it will do some good. The best state for human beings is not (as the Epicureans say) one of mere desiresatisfaction, but neither does it consist (as the Stoics suggest) simply in duty and virtue alone. Rather, the best state—the ‘highest good’—is a state in which happiness is perfectly proportioned to moral worth. But happiness itself is at least partly bodily: it involves the satisfaction of our inclinations, many of which arise from our sensible nature. And in the Critique he says that ‘all hoping aims at happiness’ (A805/B833). So despite the traditional picture of Kant as a rule-obsessed Prussian bachelor, he was in fact keenly aware that we are embodied ‘creatures of need’—animals for whom happiness consists largely in bodily and emotional welfare (CPrR 5:61). Although he does not endorse the satisfaction of our inclinations willy-nilly (cf. his infamous remarks about ‘defiling oneself by lust’), he

Hope and Despair at the Kantian Chicken Factory

227

consistently characterizes happiness as the proper satisfaction of the ‘sum total’ of them (MM 6:424–6; G 4:399, 418). Behind his theory of the highest good, then, is Kant’s recognition that prolonged experience of injustice—of a world-history in which the wicked prosper and virtue seems to make no difference—can chip away at our resolve. In the Religion passage quoted above (from 6:101), he acknowledges that one of the things we hope for—and thus one of the things that make us happy, in the context of ethical and political action—is having our ‘well-intentioned efforts’ fulfilled. For example: suppose the moral law demands that we try to help the disadvantaged by donating money to various causes; for Kant, this will be true regardless of actual outcomes. Still, he recognizes that most of us also have a ‘need’ for our altruism to be effective33: it helps us retain not our rational motive but our psychological resolve if we can believe (or at least reasonably hope) that the needy are genuinely benefiting from our efforts. So even in a Kantian context, a morally good person can reasonably care a lot about the goodness of the consequences of her willing, and not just about the goodness of her will. This is a complicated piece of reasoning, so it is worth looking at it in stepwise fashion:34 (1) I ought to do what is morally right. [Independent argument] (2) For me, it would be demoralizing in the (D1) sense (i.e. it would lead to discouragement and dejection) not to be able to have substantial hope that there is a moral order by which a just arrangement (i.e. a ‘moral world’) will come about, for then I would have to regard it as certain that the entire history of the world will not be good on the whole, no matter what I do. [Empirical premise, concept of ‘hope’] (3) Such demoralization has an enervating effect on my resolve, and is thus demoralizing in the second, (D2) sense: I will almost certainly no longer perform actions that I take to be morally good or required. [Empirical premise] (4) Despair of this sort is seriously morally undesirable. [From (1)–(3)]

33 The ‘effective altruism’ movement introduces itself as follows: ‘Most of us want to make a difference. We see suffering, injustice, and death, and are moved to do something about them. But working out what that ‘something’ is, let alone actually doing it, is a difficult problem. It would be easy to be disheartened by the challenge’ (https://www.effectivealtruism.org/articles/introduction-to-effectivealtruism, accessed 1 November 2019). 34 Robert M. Adams sketched an empirical-psychological way of reconstructing Kant’s proof in his (1979). My presentation here is influenced by his, though also different in many key respects. In the first Critique, Kant presents an early version of the moral proof according to which hope for happiness seems to be part of the incentive for acting rightly. In the second Critique, he provides the more canonical articulation of the proof that starts not with (1) above but with the more controversial claim that we ought to will the highest good, and this involves, as a sort of presupposition, adopting moral faith in the existence whatever is required to make the highest good really possible (i.e. God and the afterlife). For articulations of that form of the proof, see Wood (1970) and Willaschek (2016). Very little attention has been devoted to the moral-psychological argument from despair in the third Critique and Religion, although see Fugate (2015) and Ebels-Duggan (2016).

228

Andrew Chignell

(5) Therefore, there is serious moral advantage for me in being able to have substantial hope that there is a moral world order. [From (2)–(4)] (6) Substantial hope that p requires the positive belief or faith that p is really, practically possible. [Theoretical premise] (7) Therefore, there is serious moral advantage for me in being able to believe or have faith that a moral world order is really, practically possible. [From (5)–(6)] (8) If a being or state of affairs is really practically possible, then there must be something in the actual world that can account for that fact. [Concept of ‘real practical possibility’] (9) The actual existence of God provides the only adequate account of the real, practical possibility of a moral world order. [Theoretical premise] (10) Therefore, there is serious moral advantage, for me at least, in being able to believe or have faith that God exists. [From (7)–(9)] (11) There are no good epistemic reasons either for or against the existence of God. [Results of examination of natural theology in Transcendental Dialectic] (12) Rational belief requires good epistemic reasons. [Kant’s evidentialism about belief 35] (13) Therefore, belief in God’s existence or non-existence is irrational. [From (11)–(12)] (14) Rational faith (Vernunftglaube) does not require good epistemic reasons; it can instead be based on good moral or pragmatic reasons. [Conceptual truth36] (15) Therefore, faith (though not belief) that God exists is prima facie morally (though not epistemically) justified, for me at least. [From (10), (13), (14)] The ‘for me at least’ sounds worrisome, but Kant is explicit: ‘I must not even say ‘It is morally certain that there is a God’, etc., but rather ‘I am morally certain’ etc’. (A829/B857, original emphasis). Still, the proof does not count as an exercise in mere wishful thinking or self-deception, provided a few conditions are met: • The justification is moral rather than epistemic • The justification is defeasible • The result is not the attitude that we would call ‘belief ’ and Kant would call logical ‘conviction’ (Überzeugung). The result, rather, is faith (Glaube). 35 Kant is a conceptual evidentialist about what we would call ‘belief ’—the kind of holding-for-true (Fürwahrhalten) that can count, if true and justified, as knowledge (Wissen) (Kant calls it ‘conviction’ (Überzeugung), as do many contemporary German epistemologists). See Chignell 2007a. 36 Rational faith, for Kant, is a voluntary state of holding-for-true (Fürwahrhalten) that, for nonepistemic reasons, a subject uses to guide deliberation, action, and assertion in certain contexts. See Chignell (2007b).

Hope and Despair at the Kantian Chicken Factory

229

• The faith in question is about a state of affairs that is evidentially ambiguous (in this case theoretically undecidable; see discussion below) Let’s take a closer look at a couple of the key premises. Regarding (6): everyone agrees that there is some sort of modal constraint on hope, but it is clear that superficial hopes often involve merely taking possibility for granted (see Chignell 2014). Deep, life-structuring hopes that p, on the other hand, arguably do require moderately clear conceptions of how, given the things and powers that exist in the world, p could turn out to be true.37 It would be extremely hard to hope in a serious way that there will be perpetual peace—to really structure one’s life around this hope, for example—without having at least some positive conception of how we denizens of the actual world could really get there from here. This is what I mean by ‘really, practically possible’.38 (9) is obviously another lynchpin; a friend of the argument would have to rule out other accounts of the real, practical possibility of a moral world: dialectical historical processes, political revolutions, a karmic system that ensures that justice will be done, liberal democratic institutions, and so on (compare Adams 1979). She would also have to explain why the full-blown classical deity, with all its omni-properties, is required to provide the relevant explanatory ground. Kant himself makes some efforts in this direction, but many readers have been unconvinced (see Michalson 1999). (11)–(14) are also crucial: the existence-claim here must be theoretically undecidable in order for the faith to be rational. Large swaths of the Critique of Pure Reason are dedicated to undermining traditional efforts to prove or even render probable the existence of God on either demonstrative or empirical grounds. Kant also rejects all atheistic arguments, including the empirical argument from evil. So, in the famous phrase, all knowledge-claims about God’s existence—theistic and atheistic—are ‘denied’ in order to ‘make room’ for the kind of moral faith that overcomes demoralization. In the Religion passage quoted at the beginning of this section, Kant is not discussing a moral world order generally but rather hope for specific outcomes of specific actions. He clearly thinks that this hope, too, justifies faith in the existence of a supersensible mechanism: a providential ‘higher wisdom’ that makes the fulfilment of our well-intentioned efforts possible. Returning to the contemporary inefficacy problem: it seems clear that people who already have such a supersensible commitment can just focus their hopes around that. They can focus, in other words, on the possibility that this mechanism arranges things such that some of  their choices make a significant positive difference—both unobservable 37 For the claim that there is a distinction between ‘superficial’ hope and ‘substantial hope’, and that the latter has extra conditions on it, see Pettit (2004) and McGeer (2004). 38 Compare Willaschek on ‘practical possibility’ (2016).

230

Andrew Chignell

and  observable. Their faith allows them reasonably to hope that their efforts matter—that they are recorded somehow, even if they seem to be inefficacious. And that hope would presumably sustain psychological resolve to keep doing what they regard as required. But is a version of this argument against demoralization available in secular contexts as well?

11.5 A Moral-Psychological Argument for Hope Given that the industrial poultry system is deeply insensitive to slight changes in demand, any attempt on Hope’s part to inculcate full-blown belief that my abstention is going to make a significant positive difference would be a miserable exercise in self-deception. Faith in that proposition also looks more like a Kierkegaardian leap than sweet Kantian reason. As we have seen, Kant anticipates James and others in the moderate pragmatist tradition in rejecting the idea that a moral argument can support a conclusion whose negation we have strong epistemic reason to believe. With this in mind, let’s consider a moral argument that is based, like Kant’s own ‘proof ’, in a claim about what’s required for the substantial hope that sustains resolve: (1*) Don’t Purchase: It is morally wrong to purchase the bodies, secretions, or eggs of farmed animals if there are readily available alternatives to eat and wear. [From an independent argument or moral intuition] (2*) It would be demoralizing in the (D1) sense (i.e. it would lead to discouragement and dejection) for me not to be able to have substantial hope that my abstention from animal products over time will make a significant positive difference with respect to animal welfare. [Empirical premise] (3*) Such demoralization has an enervating effect on my resolve, and is thus de-moralizing in the second, (D2) sense: I will almost certainly fail in my efforts to abstain over time. [Empirical premise] (4*) Despair of this sort is seriously morally undesirable. [rom (1*)–(3*)] (5*) Therefore, there is serious moral advantage for me in being able to have substantial hope that my abstention will make a significant positive difference with respect to animal welfare. [From (2*)–(4*)] So far, so good: this looks like a simple argument for the practical rationality of hope that by abstaining I will somehow make a significant difference to the system in question. Note that ‘significant’ is left vague precisely because the amount of difference-making required to avoid despair will clearly differ from person to person.

Hope and Despair at the Kantian Chicken Factory

231

But what are the conditions on having this sort of hope? We saw earlier that (6*) Substantial hope that p requires the positive belief or faith that p is really, practically possible. [Theoretical premise] But then from (5*) and (6*) we can infer: (7*) Therefore, there is serious moral advantage for me in being able to believe or have faith that it is really, practically possible that my abstention will make a significant positive difference. People of great willpower may find it psychologically possible to leave things there. Even if they dearly miss the taste of deep-fried spicy mesquite chicken, they will abstain in the stubborn hope that this behaviour, over time, will somehow make a significant positive difference. Despite the incredibly long odds (remember those 2000 birds per second), they are able to maintain belief or faith that it is really possible—and that is enough to preserve their resolve. But Kant thought such moral saints are rare: even someone as righteous as Spinoza might despair in the absence of some sense of how justice might prevail. For the rest of us, clinging to the brute possibility that we just might make a difference won’t be enough for long-term resistance, especially when the products in question are so convenient, tasty, and cheap. This is one reason why Kant’s moralpsychological argument proceeds from a claim about what is really possible to a claim (in (8) and (9)) about the actual ground or explanation of that possibility: we need to have a sense, not just that it is really possible, but how it is so. What is it about the actual world that makes the hoped-for outcome really, practically possible? Well, as we saw at the end of the last section, one idea is just this: (9*) The actual existence of God provides the only adequate account of the real, practical possibility that my abstention will make a significant positive difference. [Theoretical premise] If this were correct, then the rest of the argument could run like the previous one and conclude with full-blown moral theism. Only such faith, this version of the argument might say, makes it psychologically possible to retain the hope that my abstaining (and/or my abstinence over time) will make a significant positive difference to a morally objectionable system. So that hope and that faith, together, allow me to avoid despair, and are prima facie morally justified as a result. The problem here is obvious: (9) was debatable, but (9*) is just patently false. Perhaps there is some plausibility in the idea that the existence of a supreme being is the only adequate account of the real, practical possibility of a perfectly just

232

Andrew Chignell

world order (the highest good). But in the case of the more localized hope that my individual actions will make a significant positive difference with respect to the industrial chicken system, (9*) itself is hopeless. We already saw back in Section 11.2 that there are naturalistic ways to account for the bare possibility here, even in a massive and lumpy supply system. It’s just barely possible, for instance, that this is the month in which everyone else will purchase exactly 18,999,999 sandwiches, and thus that Oppy is indeed going to stand on one of those thresholds when he walks into the food court. If that were the case, then a choice on the threshold to abstain will make a significant positive difference (5,000 chickens saved!). Given the availability of naturalistic ways of accounting for the real possibility that my abstinence makes a difference, then, there can be nothing like a moral proof that God, the ‘universe’, karma, or fate is arranging for individual efficacy. Again, if someone already has a supersensible commitment that can do the job, then it makes sense for him to fix his hopes by appealing to a premise like (9**) The actual existence of supersensible mechanism X provides an adequate account of the real possibility that my abstinence will make a significant positive difference.

But for someone who does not already have the supersensible commitment, there is no compelling pressure to adopt one. No compelling pressure. There might be a little pressure, though, depending on how we think about the naturalistic options and the psychological constraints on hope. I just said that someone like Hope who is threatened by demoralization can try to focus on the ‘infinitesimal’ possibility of her own pivotality—she can give that possibility a prominent place in her mind every time her colleagues drag her to the bar on ‘Shotz-n-Wingz Nite’, and this way allow the idea of 5000 chickens saved to sustain her resolve.39 But that’s not going to be easy: given the sobering empirical facts, the chance that she is at a threshold point there in the bar is vanishingly small, and it may be hard for Hope to keep from sliding into despair (especially after a few shotz). Indeed, according to the account of hope that I favour, the key difference between hope and despair has to do with where the subject is disposed to place her mental attention or ‘focus’. I might strongly desire something and regard it as possible, but if I keep focusing on it as massively improbable, I am in despair. If instead I am disposed to focus on it ‘under the aspect of its possibility’—i.e. as having at least a chance of occurring—then I am

39 For more on how ‘Shotz-n-Wingz Nite’ can destroy a person’s resolve, see Halteman and Halteman Zwart (2016: 131). Those authors claim that such events occur regularly at a place called ‘Baloneez’, but Barnhill et al. (2015: 171) reference a sister establishment called ‘Jimmy’s You-Hack-itYourself BBQ’.

Hope and Despair at the Kantian Chicken Factory

233

hoping.40 But other things equal, the more unlikely a desired outcome is, the harder it is to stay mindfully focused on it as possible, rather than fixating on the overwhelming odds against. So while hoping for pivotality is an option for staving off despair, most of us will find it a hard psychological row to hoe. In light of this, it would be useful to find some other naturalistic scenarios whereby Hope’s abstinence might make a significant positive difference for animals. A candidate that comes to mind is this: Hope’s decision might somehow be connected to the decisions of numerous other people such that, if she abstains, then a significant number of other people will or will be likely to abstain (where a ‘significant’ number is precisely what’s required for there to be a ‘significant’ difference made with respect to animals—keeping in mind that what counts as ‘significant’ might be different across different subjects). This scenario divides into two: one causal, and one evidential. I’ll discuss each in turn. The causal version of the scenario would invoke this premise: (9***) The existence of a causal connection between my decision to abstain and the decisions of a significant number of other people to do the same provides an adequate account of the real, practical possibility that my abstinence will make a significant positive difference.

This seems true: such a causal connection would provide an adequate account. Following the logic of the moral-psychological argument, we would then arrive at: (10*) Therefore, there is serious moral advantage, for me at least, in being able to believe or have faith that such an interpersonal causal connection exists. [From (7)–(9***)]

But could such belief or faith be rational? That is, could it ever be rational to believe or have faith in the following? Interpersonal Causal Connection: If I choose to abstain, my action will cause a significant number of other people to abstain, and if I choose to purchase, my action will cause a significant number of other people to purchase.

Causal Connection looks like a non-starter for people who aren’t celebrities, dictators, or top-level ‘influencers’. There is no reason to think that there is any such connection between one’s person’s actions and the actions of a significant number

40 See Chignell (draft). Note that I am not claiming that hope just is the desire and the presumption of possibility plus a disposition to focus on the outcome as possible. But I do hold that substantial hope is at least accompanied by these states.

234

Andrew Chignell

of others.41 A quick look at Hope’s number of Twitter followers provides decisive reason to think that there is not such a connection in her case. But then both belief and faith are ruled out on rational grounds (again, Kant is the forerunner of James rather than Kierkegaard here—a live option for moral faith has to be evidentially ambiguous). There is a weaker version of the scenario, however, that looks more promising. Consider: (9****) The existence of a strong evidential connection between my decision to abstain and the decisions of a significant number of other people to do the same provides an adequate account of the real, practical possibility that my abstinence will make a significant positive difference.

(9****) seems true, and so by the logic of Kant’s moral-psychological argument, it would seem to give us prima facie moral justification for either belief or faith that: Interpersonal Evidential Connection: If I choose to abstain, that is strong evidence that a significant number of other people are likely to abstain, and if I choose to purchase, that is strong evidence that a significant number of other people are likely to purchase.

Is this a principle for which we have any grounds one way or the other? Here I think things are less obvious than they were with Causal Connection. It seems reasonable for Hope to think that a significant number of people could be motivated by the same reasons against purchasing animal products that she is (that might be a part of what it is to take them to be reasons). But that’s different from thinking that if she chooses to abstain, a significant number of people will be so motivated, or are likely to be so motivated. Given that most of these decisions will be made in distant places and times (and given that she is not a top-level influencer), it is simply unclear whether there is any evidence for Evidential Connection. Indeed: even if, over the short- to medium-term, she acquires evidence that other people around her are not following suit, her choice might still be strong evidence that over the long haul a significant number of people (here or elsewhere) will

41 Obviously Hope could try to establish such causal connections (and thereby shore up her resolve) by signalling her choices publicly when she can, attempting to influence others, becoming part of a broader cultural movement, and so on. And obviously the question of whether her actions have made any causal difference must be considered over the longer haul. It’s not clear how much this will help in private one-off moments under conditions of perceived inefficacy, however. Thanks to Elizabeth Harman and Renée Jorgensen Bolinger for discussion here. See also Lawford-Smith (2015).

Hope and Despair at the Kantian Chicken Factory

235

come around and see the power of her reasons.42 This means that in most epistemic contexts, anyway, belief either way would be out of rational bounds. But here is where Kant’s famous idea about denying belief to make room for faith comes in. When there is no clear evidence one way or the other, we might still have moral reason for adopting propositions in this other non-doxastic way. So if supersensible mechanisms aren’t a live option for Hope, and if her own causal pivotality is too improbable to sustain her focus, she might still be defeasibly morally justified in taking Evidential Connection on faith. That is: she might be defeasibly justified in holding that a significant number of people (somewhere, sometime) will be similar enough to her to act on the same reasons in a broad range of cases. Thus, if she chooses to abstain now, she will have strong evidence that there will be a significant improvement in farm animal welfare. Faith like that would presumably suffice to stave off Hope’s despair. However, if she think like this up to the moment of choice, but then defects and purchases (perhaps believing that all those other people will still act on the reasons that she has), she thereby loses her evidence that others will abstain. That is why it is crucial that Evidential Connection says that she only gets the relevant evidence if she actually abstains, and not just if she appreciates the reasons for abstaining.43 There is obviously more to be said about this last scenario, but here I simply want to note two of its most intriguing features. First, the reasoning is broadly Kantian in a manner that goes beyond the mere use of moral arguments. Hope relies on the idea that (other things equal) if she has a good moral reason to do something, then it is also good moral reason for others to do likewise. This is an implication of the universalizability of moral reasons—a theme that we typically associate with Kant. I have a good reason to act on a certain maxim only if it is a good reason for everyone else to do likewise in saliently similar circumstances. But the universalization goes the other way, too: if it is a good reason for me to act in such-and-such a way, then it is a good reason for everyone else in similar circumstances to do likewise. When Hope ‘thinks for herself ’ about these matters, she also takes herself to be ‘thinking from the point of view of others’—a maxim that Kant thinks of as common sense (CPJ 5:294). And so her moral faith in Evidential Connection invokes a broadly Kantian idea about the universalizability of reasons. Second, and even more intriguingly, if faith in Evidential Connection is what sustains the hope to make a difference, then we seem to have arrived in a very roundabout way at a key principle of evidential decision theory (EDT). EDT says

42 There are many dissimilarities between the cases, but just by analogy think of someone in late 17th century South Carolina who decides, on moral grounds, to free the slaves that he inherited from his family. Those reasons were good, and in order to sustain his resolve he might also have taken on faith that his choice was strong evidence that others would (at some point) do the same. But it took a long while. 43 Thanks to Victoria McGeer for discussion here.

236

Andrew Chignell

(roughly) that an agent ought to perform actions that are such that, if she performs them, then the chances of the desired outcome are maximized, regardless of whether there is any causal connection between her actions and the outcome. It is sometimes characterized as the choice to be guided by ‘auspiciousness over efficacy’; it can also be articulated as a theory of difference-making: Difference-Making, Evidential: An action A makes a difference with respect to outcome O just in case performing A provides evidence that a change in O will occur.

According to this principle, no causal connection is required for an action to make a difference. That is why EDT is thought to support taking the one box in the Newcomb problem.44 Even if there is no causal connection between your choice and the action of a near-perfect predictor of your choice (since the prediction has already been made), there is an evidential connection: what you do provides strong evidence regarding what the predictor did. So on the evidential conception of difference-making, your action does ‘make a difference’ and you should take just the one box (thus containing a million dollars). On the causal conception of difference-making, your action can’t make a difference and so you should be safe and take both boxes (thereby acquiring a thousand dollars).45 On the scenario we have just been considering, Hope abstains for independent reasons (involving e.g. symbolic value worries or concerns about some sort of complicity), but she also has a psychological need to hope that her action makes a difference, especially over time. If she takes Evidential Connection on as an item of moral faith, with Difference-Making, Evidential in the background, then she can retain faith that her action makes a difference even while recognizing that the sobering empirical facts strongly suggest that her action doesn’t make a causal difference.46 44 Originally stated in Nozick 1969. 45 Jon Elster cites empirical evidence for the claim that people use ‘diagnostic thinking’ to move from the assumption that they are ‘fairly typical members’ of a reference group to the conclusion that others ‘will tend to act like me’. Elster sees no flaw in this sort of reasoning unless explicitly causal claims are made—i.e. unless people start thinking that ‘my action will bring it about’ that others do the same. But the EDT conception of difference-making avoids that sort of ‘interpersonal magic’. See Elster (1985: 142–5). Thanks are owed here to John Pittard, Ewan Kingston (who pointed me to Elster), Philip Pettit, Richard Bradley, and Kian Mintz-Woo, as well as to the unknown author of a paper I recently refereed in which a connection between one-boxing in Newcomb and an evidentialist account of difference-making was made. Mintz-Woo and Bradley point out to me that this interpersonal ‘evidentialist’ approach to consumption and voting cannot rationally be extended to Prisoner’s Dilemma cases, however, since defecting is always the best thing to do in the latter, even on EDT. This is presumably why followers of EDT have resisted Lewis’s claim (in 1979) that ‘Newcomb’s Problem is a Prisoner’s Dilemma’. See also Pittard (2018). 46 If this is correct, then the present broadly Kantian response to the psychological inefficacy problem has the added benefit of providing prima facie moral justification for faith in a principle that

Hope and Despair at the Kantian Chicken Factory

237

11.6 Conclusion Recent discussions of the inefficacy problem focus primarily on the conceptual side of the issue—that is, on how inefficacy poses a challenge to traditional consequentialism. In the bulk of this paper, I focused instead on the psychological side of the problem—that is, on how the perception of our almost certain inefficacy can lead to the kind of despair that, in turn, threatens to undermine moral resolve. I suggested that although Kant’s own narrow view regarding how we ought to treat animals has little appeal, his moral-psychological argument against despair can be applied to justify various resolve-sustaining strategies. These include focusing in hope on the fact that: A. An extremely unlikely threshold event can cause at least one of my abstaining actions to make a significant positive difference to outcomes.

Alternative strategies involve adopting defeasible moral faith in one (or more) of the following claims: B. A supersensible mechanism exists and can causally connect my abstinence to some significant positive difference in outcomes. C.  My choosing to abstain is strong evidence that there will be a significant positive difference in outcomes, even if there is no causal connection between the two.

Although it has a structure that is analogous to Kant’s moral-psychological argument for the existence of God, only one of the options here involves faith in something supersensible. A final, Kant-scholarly point: the structure of the moral-psychological argument I’ve been considering here reveals an underappreciated way in which the practical has ‘primacy’ for Kant. The argument says that if we morally ought to act a certain way, and we are threatened by resolve-sapping despair, then we are prima facie morally justified in seeking strategies that will sustain our hope and thus our commitment to the ought in question. But although the moral commitment and the concomitant hope have primacy, Kant is not an advocate of irrational leaps. Theoretical reason does kick in at some point and require a coherent account of how the hoped-for scenario could really, practically come about. That in turn makes us prima facie morally justified in having faith in

supports one-boxing. There is something right about the prosperity gospel after all: faith can indeed make you rich . . .

238

Andrew Chignell

whatever can adequately explain that practical possibility, as long as it is evidentially ambiguous. The faith in question can even involve a ‘theoretical’ issue: an existenceclaim, for instance, or a principle in decision theory.47

47 For feedback on these ideas as they developed, I am grateful to the editors of this volume and to Anne Barnhill, Renée Jorgensen Bolinger, Luc Bovens, Mark Budolfson, Gabriel Citron, Silvia De Toffoli, Tyler Doggett, Gabriele Gava, Matthew Halteman, Jakob Huber, Tania Lombrozo, Victoria McGeer, Kian Mintz-Woo, Philip Pettit, Michael Milona, Leigh Vicens, and Marcus Willaschek. I am particularly indebted to McGeer for lengthy written comments. I also thank audiences at Humboldt University in Berlin, Goethe University in Frankfurt, the Hungarian Academy of Sciences in Budapest, Georgetown University, Princeton University, and the unforgettable ‘Kant and Animals’ conference sponsored by University of Witswatersrand in the Krüger Animal Park, South Africa.

Bibliography Abela, P. (2002) Kant’s Empirical Realism. Oxford: Oxford University Press. Adams, Robert  M. (1979) ‘Moral Arguments for Theistic Belief ’, in C.  F.  Delaney (ed.), Rationality and Religious Belief, Notre Dame: Notre Dame Press: 116–40. Adams, Robert M. (1999) Finite and Infinite Goods. New York: Oxford University Press. Adamson, P. and Edwards, G. F. (2018) Oxford Philosophical Concepts: Animals. Oxford: Oxford University Press. Alanen, Lilli (2003) Descartes’s Concept of Mind. Cambridge, MA: Harvard University Press. Albertzart, M. (2019) ‘A Kantian Solution to the Problem of Imperceptible Differences’, European Journal of Philosophy, online first. Allais, L. (2009) ‘Kant, Non-Conceptual Content and the Representation of Space’, Journal of the History of Philosophy, 47: 383–413. Allais, L. (2011a) ‘Idealism Enough: Response to Roche’, Kantian Review, 16: 375–98. Allais, L. (2011b) ‘Transcendental Idealism and the Transcendental Deduction’, in D. Schulting and J. Verburgt (eds.) Kant’s Idealism. London: Springer: 91–108. Allais, L. (2015) Manifest Reality: Kant’s Idealism and His Realism. Oxford: Oxford University Press. Allais, L. (2016) ‘Kant’s Racism’, Philosophical Papers, 45 (1–2): 1–36. Allais, L. (2017) ‘Synthesis and Binding’, in A. Gomes and A. Stephenson (eds.) Kant and the Philosophy of Mind, Oxford: Oxford University Press: 25–45. Allen, C. (2004) ‘Animal Pain’, Noûs, 38 (4): 617–43. Allison, H.  E. (2004) Kant’s Transcendental Idealism: Revised and Enlarged. New Haven: Yale University Press. Almeida, M. and Berstein, M. (2000) ‘Opportunistic Carnivorism’, Journal of Applied Philosophy, 17 (2): 205–11. Altman, M. (2010) ‘Kant on Sex and Marriage: The Implications for the Same-Sex Marriage Debate’, Kant-Studien, 101 (3): 309–30. Ameriks, K. (2012) ‘Kant, Human Nature, and History after Rousseau’, in S. M. Shell and R.  L.  Velkley (eds.) Kant’s Observations and Remarks: A Critical Guide, Cambridge: Cambridge University Press: 247–65. Andrews, K. (2014) The Animal Mind: An Introduction to the Philosophy of Animal Cognition. London: Routledge. Andrews, K. and Beck, J. (2017) The Routledge Handbook of Philosophy of Animal Minds. London: Routledge. Angell, J. M. and Marzluff, T. (2007) In the Company of Crows and Ravens. New Haven, CT and London: Yale University Press. Appiah, K. A. (1986) ‘Racism and Moral Pollution’, Philosophical Forum, 18 (2): 185–202. Aristotle (2000) Nicomachean Ethics (trans. and ed. Roger Crisp), Cambridge, England: Cambridge University Press. Audidière, Sophie (2016) ‘Why Do Helvétius’s Writings Matter? Rousseau’s Notes Sur De L’esprit’, British Journal for the History of Philosophy, June: 1–19.

240

Bibliography

Barnhill, A., Budolfson, M. B., and Doggett, T. (eds.) (2016) Food, Ethics, and Society. New York: Oxford University Press. Barnhill, A., Budolfson, M.  B., and Doggett, T. (eds.) (2018) Oxford Handbook of Food Ethics. New York: Oxford University Press. Bauer, N. (2010) ‘Kant’s Subjective Deduction’, British Journal for the History of Philosophy, 18 (3): 433–60. Baumeister, D. (2019) ‘Kant, Chakrabarty, and the Crises of the Anthropocene’, Environmental Ethics, 41 (1): 53–67. Bayle, P. (1991) Historical and Critical Dictionary (Selections). Indianapolis: Hackett. Beauchamp, T. L. (1999) ‘Hume on the Nonhuman Animal’, The Journal of Medicine and Philosophy: A Forum for Bioethics and Philosophy of Medicine, 24 (4): 322–35. Beck, J. (2012) ‘Do Animals Engage in Conceptual Thought?’ Philosophy Compass, 7 (3): 218–29. Beck, L. W. (1978) ‘Did the Sage of Königsberg Have No Dreams?’, In Essays on Kant and Hume. New Haven: Yale University Press: 38–60. Bentham, J. (1789) An Introduction to the Principles of Morals and Legislation. Public domain material available at: www.earlymoderntexts.com/assets/pdfs/bentham1780.pdf. Berg, Hein van den (2014) Kant on Proper Science. Dordrecht: Springer. Bermúdez, J. L. (2003) Thinking Without Words. Oxford: Oxford University Press. Biederman, I. and Shiffrar, M. M. (1987) ‘Sexing Day-Old Chicks: A Case Study and Expert Systems Analysis of a Difficult Perceptual-Learning Task’, Journal of Experimental Psychology: Learning, Memory, and Cognition, 13 (4): 640–5. Bird, G. (1962) Kant’s Theory of Knowledge. New York: Humainties Press. Bloch, E. (1921) Thomas Müntzer als Theologe der Revolution. Frankfurt: Suhrkamp. Bloch, E. (1959) Das Prinzip Hoffnung. Frankfurt: Suhrkamp. Block, N. (1995) ‘On a Confusion about a Function of Consciousness’, Behavioral and Brain Sciences, 18 (2): 227–47. Block, N. (2005) ‘Two Neural Correlates of Consciousness’, Trends in Cognitive Sciences, 9 (2): 46–52. Block, N. (2008) ‘Consciousness and Cognitive Access’, Proceedings of the Aristotelian Society, 108: 289–317. Bock, G. (2014) ‘Korsgaard and Non-Sentient Life’, Between the Species, 17 (1): 74–84. Boey, G. (2016) ‘Eating Animals and Personal Guilt: The Individualization of Responsibility for Factory Farming’, in A. Barnhill, M. B. Budolfson, and T. Doggett (eds.), Food, Ethics, and Society. New York: Oxford University Press: 22–9. Bovens, L. (2015) ‘The Tragedy of the Commons as a Voting Game’, in M. Peterson (ed.), The Prisoner’s Dilemma, New York: Cambridge University Press: 156–76. Bowman, B. (2011) ‘A Conceptualist Reply to Hanna’s Kantian Non-Conceptualism’, International Journal of Philosophical Studies, 19: 417–46. Boyle, M. (2016) ‘Additive Theories of Rationality: A Critique’, European Journal of Philosophy, 24 (3): 527–55. Bramble, B. (2015) ‘The Case against Meat’, in B. Bramble and R. Fischer (eds.), The Moral Complexities of Eating Meat. New York: Oxford University Press: 135–52. Bramble, B. and Fischer, R. (eds.) (2015) The Moral Complexities of Eating Meat. New York: Oxford University Press. Breitenbach, A. (2008) ‘Two Views on Nature: A Solution to Kant’s Antinomy of Mechanism and Teleology’, British Journal for the History of Philosophy, 16: 351–69. Breitenbach, A. (2009) Die Analogie von Vernunft und Natur. Berlin/New York: De Gruyter. Brewer, B. (2006) ‘Perception and Content’, European Journal of Philosophy, 14: 165–81.

Bibliography

241

Brewer, B. (2011) Perception and Its Objects. Oxford University Press. Broadie, A. and Pybus, E. M. (1974) ‘Kant’s Treatment of Animals’, Philosophy, 49 (190): 375–83. Broadie, A. and Pybus, E. M. (1978) ‘Kant and the Maltreatment of Animals’, Philosophy, 53 (206): 560–1. Budolfson, M. B. (2016) ‘The Inefficacy Objection to Utilitarian Theories of the Ethics of the Marketplace’, in A.  Barnhill, M.  B.  Budolfson, and T.  Doggett (eds.), Food, Ethics, and Society. New York: Oxford University Press: 204–10. Budolfson, M. B. (2019) ‘The Inefficacy Objection to Consequentialism, and the Problem with the Expected Consequences Response’, Philosophical Studies, 176 (7): 1711–24. Budolfson, M. B. (unpublished) ‘The Inefficacy Objection to Deontology: What it is, Why it is Important, and How to Respond to It.’ Burge, T. (2007a) ‘Two Kinds of Consciousness’, in Foundations of Mind, Oxford, England: Clarendon Press: 383–91. Burge, T. (2007b) ‘Reflections of Two Kinds of Consciousness’, in Foundations of Mind, Oxford: England: Clarendon Press: 392–419. Burge, T. (2009) ‘Five Theses on de Re States and Attitudes’, in J. Almog and P. Leonardi (eds.) The Philosophy of David Kaplan, New York: Oxford University Press: 246–316. Burge, T. (2010a) Origins of Objectivity. Oxford: Oxford University Press. Burge, T. (2010b) ‘Steps Toward Origins of Propositional Thought’, Disputatio, 4 (29): 39–67. Burge, T. (2013) ‘Self and Self-Understanding’, in Cognition through Understanding. Oxford, England: Oxford University Press: 140–226. Callanan, J. J. (2008) ‘Kant on Analogy’, British Journal for the History of Philosophy, 16 (4): 747–72. Callanan, J.  J. (2017a) ‘The Ideal of Reason’, in J.  R.  O’Shea (ed.) Kant’s Critique of Pure Reason: A Critical Guide, Cambridge: Cambridge University Press: 243–58. Callanan, J. J. (2017b) ‘Kant on the Spontaneous Power of the Mind’, British Journal for the History of Philosophy, 25 (3): 565–88. Callicott, J.  B. (2002) ‘The Pragmatic Power and Promise of Theoretical Environmental Ethics: Forging a New Discourse’, Environmental Values, 11 (1): 3–25. Camp, E. (2007) ‘Thinking with Maps’, Philosophical Perspectives, 21: 145–82. Campbell, J. (1994) ‘Objects and Objectivity’, in C. Peacocke (ed.) Objectivity, Simulation and the Unity of Consciousness, Oxford: Oxford University Press: 3–20. Campbell, J. (2002) Reference and Consciousness. Oxford: Oxford University Press. Campbell, J. (2006) ‘Manipulating Colours: Pounding an Almond’, in T.  Gendler and J. Hawthorne (eds.) Perceptual Experience, Oxford: Oxford University Press: 31–48. Campbell, J. (2007) ‘What’s the Role of Spatial Awareness in Visual Perception of Objects?’, Mind & Language, 22 (5): 548–62. Camus, A. (1955) The Myth of Sisyphus. London: Hamish Hamilton. Carruthers, P. (2009) ‘Invertebrate Concepts Confront the Generality Constraint (and Win)’, in R.  W.  Lurz (ed.), The Philosophy of Animal Minds, Cambridge: Cambridge University Press: 89–107. Cavalieri, P. (2001) The Animal Question: Why Nonhuman Animals Deserve Human Rights. New York: Oxford University Press, 2001. Cheng, K. and Newcombe, N.  S. (2005) ‘Is There a Geometric Module for Spatial Orientation? Squaring Theory and Evidence’, Psychonomic Bulletin & Review, 12 (1): 1–23. Cheries, E. W., Mitroff, S. R., Wynn, K., and Scholl, B. J. (2008) ‘Cohesion as a Constraint on Object Persistence in Infancy’, Developmental Science, 11 (3): 427–32.

242

Bibliography

Chignell, A. (2007a) ‘Kant’s Concepts of Justification’, Nous, 41 (1): 33–63. Chignell, A. (2007b) ‘Belief in Kant’, Philosophical Review, 116 (3): 323–60. Chignell, A. (2014) ‘Rational Hope, Possibility, and Divine Action’, in G. E. Michalson (ed.) Religion within the Bounds of Mere Reason: A Critical Guide, New York: Cambridge University Press: 98–117. Chignell, A. (2016) ‘Can We Really Vote with our Forks? Opportunism and the Threshold Chicken’, in A. Chignell, T. Cuneo, and M. Halteman (eds.), Philosophy Comes to Dinner, New York: Routledge: 182–202. Chignell, A. (2018) ‘Religious Dietary Practices and Secular Food Ethics; or, How to Hope that your Food Choices Make a Difference Even When you Reasonably Believe that they Don’t’, in Barnhill, A., Budolfson, M. B. and Doggett, T. (eds.) (2018) Oxford Handbook of Food Ethics. New York: Oxford University Press: 287-312. Chignell, A. (forthcoming 2020) ‘The Formula of Sentient Animality: Critical Review of Korsgaard, Fellow Creatures’, Mind. Chignell, A. (draft) ‘The Focus Theory of Hope’. Chignell, A., Cuneo, T., and Halteman, M. (eds.) (2016) Philosophy Comes to Dinner. New York: Routledge. Cholbi, M. (2014) ‘A Direct Kantian Duty to Animals’, The Southern Journal of Philosophy, 52 (3): 338–58. Clark, S. R. L. (2011) ‘Animals in Classical and Late Antique Philosophy’, in T. Beauchamp and R. Frey, Oxford Handbook of Animal Ethics, Oxford: Oxford University Press: 35–60. Cohen, A. (2004) ‘Kant’s Antinomy of Reflective Judgment: A Re-evaluation’, Teorema, 23 (3): 183–97. Cohen, A. (2006) ‘Kant on Epigenesis, Monogenesis and Human Nature: The Biological Premises of Anthropology’, Studies in History and Philosophy of the Biological and Biomedical Sciences, 37 (4): 675–93. Cohen, A. (forthcoming) ‘Biology’, in M. Timmons and S. Baiasu (eds.), The Kantian Mind, London: Routledge. Collins, Arthur (1999) Possible Experience. Berkeley and Los Angeles: University of California Press. Conant, J. (2016) ‘Why Kant Is Not a Kantian’, Philosophical Topics, 44 (1): 75–125. Condillac, E.  B. (2001) Essay on the Origin of Human Knowledge (ed. H.  Aarsleff). Cambridge: Cambridge University Press. Cornell, J.  F. (1986) ‘Newton of the Grassblade: Darwin and the Problem of Organic Teleology’, Isis, 77: 405–21. Cottingham, J. (1978a) ‘A Brute to the Brutes?: Descartes’ Treatment of Animals’, Philosophy, 53 (206): 551–9. Cottingham, J. (1978b) ‘Descartes on “Thought”’, The Philosophical Quarterly, 28 (112): 208–14. Crane, T. (2009) ‘Is Perception a Propositional Attitude?’, Philosophical Quarterly, 59: 452–69. Cuneo, T. (2016) ‘Conscientious Omnivorism’, in in A.  Chignell, T.  Cuneo, and M. Halteman (eds.), Philosophy Comes to Dinner, New York: Routledge: 21–38. Darwall, S. (2009) The Second Person Standpoint. Cambridge, MA: Harvard University Press. Davidson, D. (1982) ‘Rational Animals’, Dialectica, 36 (4): 317–27. Dean, R. (1996) The Value of Humanity in Kant’s Moral Theory. Oxford: Clarendon Press. de Beauvoir, S. (2011) The Second Sex (trans. C. Borde and S. Malovany-Chevallier). New York: Vintage Books.

Bibliography

243

Denis, L. (2000) ‘Kant’s Conception of Duties regarding Animals: Reconstruction and Reconsideration’, History of Philosophy Quarterly, 17: 405–23. Descartes, R. (1985) The Philosophical Writings of Descartes Volume I. (eds. J Cottingham, D. Murdoch and R. Stoothoff. Cambridge: Cambridge University Press. Doggett, T. (2016) ‘Introduction’, in A. Barnhill, M. B. Budolfson, and T. Doggett (eds.), Food, Ethics, and Society. New York: Oxford University Press. Dolins, F. L. and Mitchell, R. W. (2010) Spatial Cognition, Spatial Perception: Mapping the Self and Space. Cambridge: Cambridge University Press. Dreier, J. (1993) ‘Structures of Normative Theories’, The Monist, 76 (1): 22–40. Dummett, M. (1993) The Seas of Language. Oxford: Oxford University Press. Dunlop, K. (2017) ‘Understanding Non-Conceptual Representation of Objects: Empirical Models of Sensibility’s Operation’, in A. Gomes and A. Stephenson (eds.), Kant and the Philosophy of Mind, Oxford: Oxford University Press: 46–66. Dyck, C.  W. (2011) ‘A Wolff in Kant’s Clothing: Christian Wolff ’s Influence on Kant’s Accounts of Consciousness, Self-Consciousness, and Psychology’, Philosophy Compass, 6 (1): 44–53. Ebels-Duggan, K. (2015) ‘The Right, the Good, and the Threat of Despair: (Kantian) Ethics and the Need for Hope in God’, in J.  Kvanvig (ed.), Oxford Studies in Philosophy of Religion, vol 7. New York: Oxford University Press: 81–109. Eilan, N. (1997) ‘Objectivity and the Perspective of Consciousness’, European Journal of Philosophy, 5 (3): 235–50. Eilan, N. (2011) ‘Experiential Objectivity’, in N. Eilan, J. Roessler and H. Lerman (eds.), Perception, Causation and Objectivity, Oxford: Oxford University Press: 51–68. Eilan, N. (2017) ‘Perceptual Objectivity and Consciousness: A Relational Response to Burge’s Challenge’, Topoi, 36 (2): 287–98. Elster, J. (1985) ‘Rationality, Morality, and Collective Action’, Ethics, 96 (1): 136–55. Evans, G. (1982) The Varieties of Reference. Oxford, England: Clarendon Press. Evans, G. (1985) ‘Things without the Mind: A Commentary upon Chapter Two of Strawson’s Individuals’, in Collected Papers, Oxford: Clarendon Press: 249–90. Eze, E.  C. (1995) ‘The Color of Reason: The Idea of “Race” in Kant’s Anthropology’, in K.  M.  Faull (ed.), Anthropology and the German Enlightenment: Perspectives on Humanity, London, Toronto: Associated University Press: 200–41. Falkenstein, L. (2006) ‘Kant’s Transcendental Aesthetic’ in G. Bird (ed.), A Companion to Kant. Blackwell: Oxford: 140–53. Feldman, H. L. (1994) ‘Objectivity in Legal Judgment’, Michigan Law Review, 92 (5): 1187. Fieldhouse, H. (2004) ‘The Failure of the Kantian Theory of Indirect Duties to Animals’, Animal Liberation Philosophy and Policy Journal, 2: 45–53. Fisher, N. (2017) ‘Kant on Animal Minds’, Ergo, 4: 441–62. FitzPatrick, W. J. (2000) Teleology and the Norms of Nature. New York, New York: Garland Publishing. Fitzpatrick, W. J. (2005) ‘The Practical Turn in Ethical Theory’, Ethics, 115 (4): 651–91. Flikschuh, K. (2008) ‘Reason, Right, and Revolution: Kant and Locke’, Philosophy and Public Affairs’, 36 (4): 375–404. Flikschuh, K. (2018) What is Orientation in Global Thinking? New York: Cambridge University Press. Floros, J. (2009) ‘Getting Real about our Modern Food System’, Food Technology Magazine, ePerspective. https://foodtecheperspective.wordpress.com/2009/09/01/getting-realabout-our-modern-food-system, accessed 5 January 2018.

244

Bibliography

Food and Agriculture Organization of the United Nations (2006) Livestock’s Long Shadow: Environmental Issues and Options. http://www.fao.org/docrep/010/a0701e/a0701e00. HTM, accessed 2 January 2018. Foot, P. (2001) Natural Goodness. Oxford, England: Clarendon Press, 2001. Förster, E. (2002a) ‘Die Bedeutung von §§76, 77 der Kritik der Urteilskraft für die Entwicklung der nachkantischen Philosophie’, Zeitschrift für philosophische Forschung, 56 (2): 170–90. Förster, E. (2002b) ‘Die Bedeutung der §§76, 77 der Kritik der Urteilskraft für die Entwicklung der nachkantischen Philosophie (Teil II)’, Zeitschrift für philosophische Forschung, 56 (3): 321–45. Förster, E. (2008) ‘Von der Eigentümlichkeit unseres Verstandes in Ansehung der Urteilskraft (§§74–78)’ in O. Höffe (ed.), Immanuel Kant. Kritik der Urteilskraft, Berlin: Akademie Verlag: 259–74. Förster, E. (2011) Die 25 Jahre der Philosophie. Frankfurt/M.: Klostermann. Frank, M. and Zanetti, V. (2001) Schriften zur Ästhetik und Naturphilosophie. Frankfurt/M.: Suhrkamp, vol. 3. Frankfurt, H. (1971) ‘Freedom of the Will and the Concept of a Person’, Journal of Philosophy, 68 (1): 5–20. Freegan.info (2008) http://freegan.info, accessed 12 January 2017. Frege, G. (1884) Die Grundlagen Der Arithmetik. Breslau: Koebner. Frege, G. (1918/1956) ‘The Thought’ (trans. P. T. Geach), Mind, 65 (259): 289–311. Frey, R. G. (1985) Rights, Killing, and Suffering: Moral Vegetarianism and Applied Ethics. Oxford: Blackwell. Frierson, P. (2013) Kant’s Questions: What is the Human Being? London: Routledge. Frierson, P R. (2014) Kant’s Empirical Psychology. Cambridge: Cambridge University Press. Fudge, E. (2006) Brutal Reasoning: Animals, Rationality, and Humanity in Early Modern England. Ithaca: Cornell University Press. Fugate, C. (2014) ‘The Highest Good and Kant’s Proof(s) of God’s Existence’, History of Philosophy Quarterly, 31 (2): 137–58. Fuss, T., Bleckmann, H., and Schluessel, V. (2014) ‘The Brain Creates Illusions Not Just for Us: sharks (Chiloscyllium griseum) can “see the magic” as well’, Frontiers in Neural Circuits, 8: DOI: 10.3389/fncir.2014.00024. Garrett, A. (2006) ‘Human Nature’, in K.  Haakonssen (ed.) The Cambridge History of Eighteenth-Century Philosophy, Cambridge: Cambridge University Press: (1): 160–233. Garrett, A. (2011) ‘Animals and Ethics in the History of Modern Philosophy’, in Beauchamp, T.  L. and Frey, R.  G. (eds.) Oxford Handbook on Animal Ethics, Oxford: Oxford University Press: 61–90. Garthoff, J. (2010a) ‘Structuring Ends’, Philosophia, 38 (4): 691–713. Garthoff, J. (2010b) ‘Mimicking Korsgaard’, in P. Muchnik (ed.) Rethinking Kant: Volume II. Newcastle upon Tyne, England: Cambridge Scholars Publishing: 122–56. Garthoff, J. (2010c) ‘Meriting Concern and Meriting Respect’, Journal of Ethics and Social Philosophy, 5 (2): 1–28. Garthoff, J. (2015) ‘The Priority and Posteriority of Right’, Theoria, 81 (3): 1–27. Garthoff, Jon (forthcoming) ‘Animal Punishment’, Philosophical Papers. Gaukroger, S. (1989) Cartesian Logic: An Essay on Descartes’s Conception of Inference. Oxford: Oxford University Press. Gaukroger, S. (1995) Descartes: An Intellectual Biography. Oxford: Oxford University Press. Gaut, B. (1997) ‘The Structure of Practical Reason’, in G. Cullity and B. Gaut (eds.), Ethics and Practical Reason, Oxford: Clarendon Press: 161–88. George, R. (1981) ‘Kant’s Sensationism’, Synthese, 47 (2): 229–55.

Bibliography

245

Gibbard, A. (1999) ‘Morality as Consistency in Living’, Ethics, 110 (1): 140–64. Ginsborg, H. (2001) ‘Kant on Understanding Organisms as Natural Purposes’ in E. Watkins (ed.) Kant and the Sciences. Oxford: Oxford University Press: 231–58. Ginsborg, H. (2004) ‘Two Kinds of Mechanical Inexplicability in Aristotle and Kant’, Journal of the History of Biology, 42: 33–65. Ginsborg, H. (2006) ‘Kant and the Problem of Experience’, Philosophical Topics, 34 (1&2): 59–106. Ginsborg, H. (2008) ‘Was Kant a Nonconceptualist?’ Philosophical Studies, 137 (1): 65–77. Ginsborg, H. (2011) ‘Perception, Generality and Reasons’ in A. Reisner and A. SteglichPetersen (eds.) Reasons for Belief. Cambridge: Cambridge University Press: 131–57. Ginsborg, H. (2014) ‘Oughts without Intentions: A Kantian Approach to Biological Function’ in I. Goy and E. Watkins (eds.) Kant’s Theory of Biology. Berlin and New York: De Gruyter: 259–74. Golob, S. (2011) ‘Kant on Intentionality, Magnitude, and the Unity of Perception’, European Journal of Philosophy, 22 (4): 505–28. Golob, S. (2016a) ‘Kant as Both Conceptualist and Nonconceptualist’, Kantian Review, 21 (3): 367–91. Golob, S. (2016b) ‘Why the Transcendental Deduction Is Compatible with Nonconceptualism’, in D. Schulting (ed.), Kant and Nonconceptualism, London: Palgrave Macmillan: 27–52. Gomes, A. (2014) ‘Kant on Perception: Naïve Realism, Non-Conceptualism and the B-Deduction’, The Philosophical Quarterly, 64 (254): 1–19. Goodman, C. (2014) ‘Ethics in Indian and Tibetan Buddhism’, in Stanford Encyclopedia of Philosophy, edited by Ed Zalta. http://plato.stanford.edu/archives/fall2014/entries/ethicsindian-buddhism/, accessed 30 December 2017. Goy, I. (2014a) ‘Epigenetic Theories. Caspar Friedrich Wolff and Immanuel Kant’, in I. Goy and E. Watkins (eds.) (2014) Kant’s Theory of Biology. Berlin: Walter de Gruyter: 45–60. Goy, I. (2014b) ‘Kant’s Theory of Biology and the Argument from Design’, in I. Goy and E. Watkins (eds.) (2014) Kant’s Theory of Biology. Berlin: Walter de Gruyter: 203–20. Goy, I. (2015) ‘The Antinomy of Teleological Judgment’, Studi Kantiani, 28: 65–87. Goy, I. (2016) Kants Theorie der Biologie. Ein Kommentar. Eine Lesart. Eine historische Einordnung. Kant-Studien-Ergänzungshefte xx, Berlin/New York: De Gruyter. Goy, I. and E. Watkins (eds.) (2014) Kant’s Theory of Biology, Berlin/New York: De Gruyter. Griffith, A. (2010) ‘Perception and the Categories’, European Journal of Philosophy, 20: 193–222. Griffith, A. (2012) ‘Perception and the Categories: A Conceptualist Reading of Kant’s Critique of Pure Reason’, European Journal of Philosophy, 20: 193–222. Grüne, S. (2009) Blinde Anschauung. Frankfurt am Main: Vittorio Klostermann. Grüne, S. (2011) ‘Is there a Gap in Kant’s B Deduction?’, International Journal of Philosophical Studies, 19 (3): 465–90. Grüne, S. (2014a) ‘Précis of Blinde Anschauung’, Critique. https://virtualcritique.wordpress. com/2014/08/19/precis-of-blinde-anschauung-2. Grüne, S. (2014b) ‘Reply to Colin McLear’, Critique. https://virtualcritique.wordpress. com/2014/08/20/reply-to-colin-mclear. Goy, I. (2017) Blinde Anschauung Kants Theorie der Biologie. Ein Kommentar. Eine Lesart. Eine historische Einordnung. Berlin/New York: De Gruyter. Guerrero, A. (2010) ‘The Paradox of Voting and the Ethics of Political Representation’, Philosophy and Public Affairs, 38 (3): 272–306. Guyer, P. (1987) Kant and the Claims of Knowledge. Cambridge: Cambridge University Press.

246

Bibliography

Guyer, P. (2005) Kant’s System of Nature and Freedom. Oxford: Oxford University Press. Halteman, M. and McMullen, S. (2018) ‘Against Causal Inefficacy Objections: The Real Economic Impact of Individual Consumer Choices on Animal Agriculture’, Food Ethics, online first: https://doi.org/10.1007/s41055-018-00030-4. Halteman, M. and Halteman Zwart, M. (2016) ‘Philosophy as Therapy for Recovering (Unrestrained) Omnivores’, in A.  Chignell, T.  Cuneo, and M.  Halteman (eds.), Philosophy Comes to Dinner, New York: Routledge: 129–48. Hanley, R. P. (2012) ‘Rousseau’s Virtue Epistemology’, Journal of the History of Philosophy, 50 (2): 239–63. Hanna, R. (2005) ‘Kant and Nonconceptual Content’, European Journal of Philosophy, 13: 247–90. Hanna, R. (2011) ‘Beyond the Myth of the Myth’, International Journal of Philosophical Studies, 19: 323–98. Harman, E. (2016) ‘Eating Meat as a Morally Permissible Moral Mistake’, in A. Chignell, T. Cuneo, and M. Halteman (eds.), Philosophy Comes to Dinner, New York: Routledge: 215–31. Hartz, G. A. and Wilson, C. (2005) ‘Ideas and Animals: The Hard Problem of Leibnizian Metaphysics’, Studia Leibnitiana, 1–19. Hastings, H. (1973) Man and Beast in French Thought of the Eighteenth Century. New York: Johnson Reprint Corp. Heeger, R. and Brom, G. (2001) ‘Intrinsic Value and Direct Duties: From Animal Ethics Towards Environmental Ethics?’, Journal of Agricultural and Environmental Ethics, 14 (2): 241–52. Heidegger, M. (1949/1976) Wegmarken. Frankfurt: Klostermann. Heidegger, M. (1928/2001) Einleitung zu ‘Was ist Metaphysik?’ in Wegmarken. Frankfurt a.M.:Klostermann. Frankfurt: Klostermann. Heinrich, B. (1999) Mind of a Raven: An Investigation into the Mind of the Raven. New York: Cliff Street Books. Helvetius, C. A. (1807) De l’Espirit; or Essays on the Mind and Its Several Faculties. London: M. Jones. Herman, B. (2018) ‘We Are Not Alone: A Place for Animals in Kant’s Ethics’, in E. Watkins (ed.), Kant on Persons and Agency, Cambridge: Cambridge University Press: 174–91. Hill, T. E. (1979) ‘Symbolic Protest and Calculated Silence’, Philosophy and Public Affairs, 9 (1): 83–102. Hill, T.  E. (1991) ‘Servility and Self-Respect’, in Autonomy and Self-Respect, New York: Cambridge University Press: 4–19. Hobbes, Thomas (1651) Leviathan. Public domain material available at http://www.oregonstate.edu/instruct/phl302/texts/hobbes/leviathan-contents.html. Hume, D. (1978) A Treatise of Human Nature. Oxford: Clarendon Press. Huneman, P. (2006) ‘Naturalising Purpose: From Comparative Anatomy to the “Adventure of Reason”’, Studies in History and Philosophy of Biology and the Biomedical Sciences, 37 (4): 649–74. Hursthouse, Rosalind (1999) On Virtue Ethics. Oxford, England: Oxford University Press. Huseyinzadegan, D. (2019) Kant’s Nonideal Theory of Politics. Evanston: Northwestern University Press. Husserl, E. (1900–1901/1984) Logische Untersuchungen. The Hague: Nijhoff. Huxford, G. (2019) ‘Kant’s Journey on Evil’, in A. Chignell (ed.) Evil: A History. New York: Oxford University Press: 315–21.

Bibliography

247

Ingensiep, H.  W. (1996) ‘Tierseele und tierethische Argumentationen in der Deutschen philosophischen Literatur des 18 Jahrhunderts’, NTM: Internationale Zeitschrift für Geschichte und Ethik der Naturwissenschaften, Technik und Medizin, 4 (1): 103–18. Jamieson, D. (2014) Reason in a Dark Time: Why the Struggle against Climate Change Failed and What it Means for our Future. New York: Oxford University Press. Jennings, C. D. (2015) ‘Attention and Perceptual Organization’, Philosophical Studies, 172 (5): 1265–78. Johnston, M. (2006) ‘Better Than Mere Knowledge? The Function of Sensory Awareness’, in, T. Gendler and J. Hawthorne (eds.), Perceptual Experience, Oxford: Clarendon Press: 260–90. Jolley, N. (2015) Locke’s Touchy Subjects. Oxford University Press. Kagan, S. (2011) ‘Do I Make a Difference?’, Philosophy and Public Affairs, 39 (2): 105–41. Kain, P. (2009) ‘Kant’s Defense of Human Moral Status’, Journal of the History of Philosophy, 47: 59–102. Kain, P. (2010) ‘Duties Regarding Animals’, in L. Denis (ed.), Kant’s Metaphysics of Morals: A Critical Guide, Cambridge: Cambridge University Press: 210–33. Kain, P. (2018) ‘Kant on Animals’ in P. Adamson & G. F. Edwards (eds.), Animals: A History (Oxford Philosophical Concepts), New York: Oxford University Press: 211–32. Kalderon, M. (2011) ‘Color Illusion’, Noûs, 45 (4): 751–75. Kant, Immanuel (1781/7) Kritik der reinen Vernunft, 2nd edition. In vol. 4 of Kant 1902–. Kant, Immanuel (1785) Grundlegung zur Metaphysik der Sitten. In vol. 5 of Kant 1902–. Kant, Immanuel (1786) Mutmaßlicher Anfang der Menschengeschichte. In vol. 8 of Kant 1902–. Kant, Immanuel (1790) Kritik der Urteilskraft. In vol. 5 of Kant 1902–. Kant, Immanuel (1793) Religion innerhalb der Grenzen der bloßen Vernunft. In vol. 6 of Kant 1902–. Kant, Immanuel (1797) Metaphysik der Sitten. In vol. 6 of Kant 1902–. Kant, I. (1902–) Immanuel Kants Schriften: Ausgabe der königlich preussischen Akademie der Wissenschaften, Erich Adickes et al. (eds.), Berlin: Walter de Gruyter. Kant, I. (1956) Critique of Practical Reason. Trans. Lewis Beck White. New York, New York: Macmillan. Kant, I. (1992) Lectures on Logic, M.  J.  Young (ed./trans.), Cambridge: Cambridge University Press. Kant, Immanuel (1996a) Practical Philosophy, A.  W.  Wood (ed.), M.  J.  Gregor (trans.), Cambridge: Cambridge University Press. Kant, I. (1996b) Religion and Rational Theology. Trans. and Eds. Allen  W.  Wood and Georgi Di Giovanni. Cambridge, England: Cambridge University Press. Kant, I. (1996c) Critique of Pure Reason, W. Pluhar (ed./trans.), Indianapolis: Hackett. Kant, I. (1997a) Lectures on Ethics. Edited and translated by Peter Heath. New York: Cambridge University Press. Kant, I. (1997b) Lectures on Metaphysics, K.  Ameriks and S.  Naragon (eds./trans.), Cambridge: Cambridge University Press. Kant, I. (1998) Critique of Pure Reason, P.  Guyer and A.  W.  Wood (eds./trans.), Cambridge:Cambridge University Press. Kant, I. (1999) Correspondence, A. Zweig (ed./trans.), Cambridge: Cambridge University Press. Kant, I. (2000) Critique of the Power of Judgment, P. Guyer (ed.), P. Guyer and E. Matthews (trans.), Cambridge: Cambridge University Press.

248

Bibliography

Kant, I. (2001) ‘Lectures on the Philosophical Doctrine of Religion’, in A. Wood and G. Di Giovanni (eds.), Religion and Rational Theology, Cambridge: Cambridge University Press: 335–452. Kant, I. (2002a) Theoretical Philosophy: 1755–1770, D.  Walford and R.  Meerbote (eds./ trans.), Cambridge: Cambridge University Press. Kant, I. (2002b) Theoretical Philosophy After 1781, H.  Allison and P.  Heath (eds.), G.  Hatfield, M.  Friedman, H.  Allison, and P.  Heath (trans.), Cambridge: Cambridge University Press. Kant, I. (2005) Notes and Fragments, P. Guyer (ed.), C. Bowman, P. Guyer, and F. Rauscher (trans.), Cambridge: Cambridge University Press. Kant, I. (2007) Anthropology, History, and Education, R. B. Louden and G. Zöller (eds.), M.  J.  Gregor, P.  Guyer, R.  B.  Louden, H.  Wilson, A.  W.  Wood, G.  Zöller, A.  Zweig (trans.), Cambridge: Cambridge University Press. Kant, I. (2012) Lectures on Anthropology, A. W. Wood and R. B. Louden (eds.), R. R. Clewis, R.  B.  Louden, G.  F.  Munzel, A.  W.  Wood (trans.), Cambridge: Cambridge University Press. Keya, T. J., Daveya, G. K., and Applebya, P. N. (1999) ‘Health benefits of a vegetarian diet’ Proceedings of the Nutrition Society, 58: 271–5. Kielkopf, C. (1997) ‘Masturbation: A Kantian Condemnation’, Philosophia, 25: 223–46. Kitcher, P. (2005) ‘Two Normative Roles for Self-Consciousness’, in H.  S.  Terrace and J.  Metcalfe (eds.), The Missing Link in Cognition, Oxford, England: Oxford University Press: 174–87. Kitcher, P. (2011) Kant’s Thinker. New York: Oxford University Press. Korsgaard, C. M. (1996a) The Sources of Normativity. Cambridge: Cambridge University Press. Korsgaard, C. M. (1996b) ‘Two Distinctions in Goodness’, in Creating the Kingdom of Ends. Cambridge, England: Cambridge University Press: 249–74. Korsgaard, C. M. (1996c) ‘Kant’s Formula of Humanity’, in Creating the Kingdom of Ends. Cambridge, England: Cambridge University Press: 106–32. Korsgaard, C.  M. (1996d) ‘Aristotle and Kant on the Source of Value’, in Creating the Kingdom of Ends, Cambridge, England: Cambridge University Press: 225–48. Korsgaard, C.  M. (2003) ‘Realism and Constructivism in Twentieth-Century Moral Philosophy’, Journal of Philosophical Research, 28 (Supplement): 99–122. Korsgaard, C. M. (2005) ‘Fellow Creatures’, The Tanner Lectures on Human Values, Volume 25. Ed. G. B. Peterson. Salt Lake City: University of Utah Press: 77–110. Korsgaard, C. (2008a) ‘The Normativity of Instrumental Reason’, in The Constitution of Agency. Oxford, England: Oxford University Press: 27–68. Korsgaard, C.  M. (2008b) ‘The Myth of Egoism’, in The Constitution of Agency. Oxford, England: Oxford University Press: 69–99. Korsgaard, C.  M. (2008c) ‘Acting for a Reason’, in The Constitution of Agency. Oxford, England: Oxford University Press: 207–29. Korsgaard, C. M. (2009) Self-Constitution. Oxford, England: Oxford University Press. Korsgaard, C.  M. (2011) ‘Interacting with Animals: A Kantian Account’, in, in T.  Beauchamp and R.  Frey, Oxford Handbook of Animal Ethics, Oxford: Oxford University Press: 91–118. Korsgaard, C. M. (2012) ‘A Kantian Case for Animal Rights’, in M. Michael, D. Kühne, and J. Hänni (eds.), Animal Law - Tier and Rect: Developments and Perspectives in the 21st Century, Zurich: Dike Verlag: 3–27. Korsgaard, C.  M. (2013) ‘Kantian Ethics, Animals, and the Law’, The Oxford Journal of Legal Studies, 33 (4): 1–20.

Bibliography

249

Korsgaard, C. M. (2014) ‘On Having a Good’, Philosophy, 89 (3): 405–29. Korsgaard, C. M. (2018) Fellow Creatures: Our Obligations to the Other Animals. New York: Oxford University Press. Korsgaard, C. M. (unpublished) ‘Moral Animals’, Work in progress available at http://www. people.fas.harvard.edu/~korsgaar/Essays.htm. Kraut, R. (2007) What Is Good and Why. Cambridge, Massachusetts: Harvard University Press. Kuhn, M. (2001) Kant: A Biography. New York: Cambridge University Press. Kulstad, M. (1991) ‘Leibniz, Animals, and Apperception’, in Leibniz on Apperception, Consciousness, and Reflection. Philosophia Verlag Gmbh: 16–52. Kutz, C. (2000) Complicity. New York: Cambridge University Press. Lagier, R. (2004) Les races humaines selon Kant. Paris: Presses Universitaires de France. Lähteenmäki, V. (2007) ‘Orders of Consciousness and Forms of Reflexivity in Descartes’, in S. Heinämaa, V. Lähteenmäki, and P. Remes (eds.), Consciousness: From Perception to Reflection in the History of Philosophy, Dordrecht: Springer: 177–201. Lamb, M. (2016) ‘Aquinas and the Virtues of Hope’, Journal of Religious Ethics, 44 (2): 300–32. Land, T. (2011) ‘Kantian Conceptualism’, in G.  Abel and J.  Conant (eds.), Rethinking Epistemology De Gruyter: 197–239. Land, T. (2015a) ‘Nonconceptualist Readings of Kant and the Transcendental Deduction’, Kantian Review, 20 (1): 25–51. Land, T. (2015b) ‘No Other Use Than in Judgment?: Kant on Concepts and Sensible Synthesis’, Journal of the History of Philosophy, 53 (3): 461–84. Land, T. (2018) ‘Conceptualism and the Objection from Animals’, in V.  L.  Waibel and M.  Ruffing (eds.), Natur und Freiheit: Akten des XII. Internationalen Kant-Kongresses, Berlin: De Gruyter: 1269–76. Larrimore, M. (1999) ‘Sublime Waste: Kant on the Destiny of the ‘Races’’, Canadian Journal of Philosophy, Supplementary vol. 25: 99–125. Lawford-Smith, H. (2015) ‘Unethical consumption and obligations to signal’’ Ethics & International Affairs 29/3 (2015), pp. 315-330. Laycock, H. (1972) ‘Some Questions of Ontology’, The Philosophical Review, 81 (1): 3–42. Leclerc, C d. B. (1753) Histoire Naturelle Génerale et Particulière. Paris: Imprimerie royale. Leibniz, G. W. (1765/1997) New Essays on Human Understanding. Edited by P. Remnant and J. Bennett. New York: Cambridge University Press. Leibniz, G. W. (1989) Philosophical Papers and Letters. Edited by L. E. Loemker. Dordrecht: Kluwer. Leland, P. R. (2018a) ‘Kant on Consciousness in Animals’, Studi Kantiani, 31: 75–107. Leland, P.  R. (2018b) ‘Unconscious Representations in Kant’s Early Writings’, Kantian Review, 23 (2): 257–84. Lennox, J. G. (1993) ‘Darwin was a teleologist’, Biology and Philosophy, 8 (4): 409–21. Lenoir, T. (1982) The Strategy of Life: Teleology and Mechanics in Nineteenth-century German Biology. Chicago: University of Chicago Press. Lewis, D. (1979) ‘Newcomb’s Problem is a Prisoner’s Dilemma’, Philosophy & Public Affairs, 8: 235–40. Lloyd, H. M. (2017) ‘The French Enlightenment Attempts to Create a Philosophy without Reason: The Case of Diderot and the Effect of Helvétius’, Intellectual History Review, November: 1–22. Locke, J. (1975) An Essay Concerning Human Understanding. Edited by P.  H.  Nidditch. Oxford: Clarendon Press. LoLordo, A. (2012) Locke’s Moral Man. Oxford University Press.

250

Bibliography

Longuenesse, B. (1998) Kant and the Capacity to Judge. Princeton: Princeton University Press. Louden, R. (2011) Kant’s Human Being: Essays on his Theory of Human Nature. Oxford: Oxford University Press. Lovejoy, A.  O. (1968) ‘Kant and Evolution’, in B.  Glass, O.  Temkin and W.  L.  Strauss Jr (eds.), Forerunners of Darwin 1745–1859, Baltimore: Johns Hopkins University Press: 173–206. Lurz, R.  W. (ed.) (2009) The Philosophy of Animal Minds. Cambridge: Cambridge University Press. Lurz, R.  W. (2011) Mindreading Animals: The Debate over What Animals Know About Other Minds. Cambridge, MA: MIT Press. Mackie, J. L. (1977) Ethics. New York: Viking Press. Manning, A. and Serpell, J. (eds.) (2012) Animals and Human Society: Changing Perspectives. London: Routledge. Martin, A. (2015) ‘Factory Farming and Consumer Complicity’, in A. Chignell, T. Cuneo, and M. Halteman (eds.), Philosophy Comes to Dinner, New York: Routledge. Martin, M. (1976) ‘A Critique of Moral Vegetarianism’, Reason Papers, 3: 13–43. Martin, M.  G.  F. (2002) ‘The Transparency of Experience’, Mind and Language, 17 (4): 376–425. Marzluff, J. M. and Angell, T. (2005) In the Company of Crows and Ravens. Yale University Press. Marzluff, J.  M. and Angell, T. (2013) Gifts of the Crow: How Perception, Emotion, and Thought Allow Smart Birds to Behave like Humans. New York: Atria Paperback. Matytsin, A. (2016) ‘Of Beasts and Men: Debate about Animal Souls in EighteenthCentury France’, Eighteenth-Century Thought, 6: 1–32. McDowell, J. (1996) Mind and World: With a New Introduction. Cambridge, MA: Harvard University Press. McDowell, J. (2013) ‘Perceptual Experience: Both Relational and Contentful’, European Journal of Philosophy, 21: 144–57. McFarland, J. (1970) Kant’s Concept of Teleology. Edinburgh: University of Edinburgh Press. McGeer, V. (2004) ‘The Art of Good Hope’, Annals of the American Academy of Political and Social Science, 592: 100–27. McLaughlin, P. (1989) Die Antinomie der Urteilskraft. Bonn: Bouvier. Translated Kant’s Critique of Teleology in Biological Explanation. Antinomy and Teleology. Lampeter: Edwin Mellon Press. McLaughlin, P. (1990) Kant’s Critique of Teleology in Biological Explanation. Antinomy and Teleology. Lampeter: Edwin Mellon Press. McLear, C. (2011) ‘Kant on Animal Consciousness’, Philosophers’ Imprint, 11 (15): 1–16. McLear, C. (2013) Essays on Kant on Perception and Cognition, dissertation, Cornell. McLear, C. (2014a) ‘The Kantian (Non)-Conceptualism Debate’, Philosophy Compass, 9 (11): 769–90. McLear, C. (2014b) ‘Comments on Stefanie Grüne’s Blinde Anschauung’, Critique https:// virtualcritique.wordpress.com/2014/08/19/colin-mclear-on-stefanie-grunes-blindeanschauung. McLear, C. (2015) ‘Two Kinds of Unity in the Critique of Pure Reason’, Journal of the History of Philosophy, 53 (1): 79–110. McLear, C. (2016a) ‘Getting Acquainted with Kant’, in D.  Schulting (ed.), Kantian Nonconceptualism, London: Palgrave-Macmillan: 171–97. McLear, C. (2016b) ‘Kant on Perceptual Content’, Mind, 125 (497): 95–144.

Bibliography

251

McLear, C. (2016c) ‘Review of Robert Brandom, From Empiricism to Expressivism’, Ethics, 126 (3): 808–16. McLear, C. (n.d.) Rationality: What Difference Does It Make? Manuscript. McPherson, T. (2016) ‘Why I Am a Vegan (and You Should Be One Too)’, in A. Chignell, T. Cuneo, and M. Halteman (eds.), Philosophy Comes to Dinner, New York: Routledge: 73–91. Mehta, Hemant (2013) ‘Pastor Mark Driscoll: Christians Don’t Need to Care About the Environment because Jesus is Coming Back for Us’, Patheos. http://www.patheos.com/ blogs/friendlyatheist/2013/05/04/pastor-mark-driscoll-christians-dont-need-to-careabout-the-environment-because-jesus-is-coming-back-for-us, accessed 15 January 2018. Meier, G.  F. (1749) Versuch Eines Neuen Lehrgebäudes von Den Seelen Der Tiere. Hemmerde. Melehy, H. (2005) ‘Silencing the Animals: Montaigne, Descartes, and the Hyperbole of Reason’, Symplokē, 13 (1/2): 263–82. Mettrie, J. D. (1996) Machine Man and Other Writings. Cambridge: Cambridge University Press. Michaelson, E. (2016a) ‘Act Consequentialism and Inefficacy’, in A.  Barnhill, M.  B.  Budolfson, and T.  Doggett (eds.), Food, Ethics, and Society. New York: Oxford University Press: 210–14. Michaelson, Elliot (2016b) ‘A Kantian Response to Futility Worries?’, in A.  Barnhill, M.  B.  Budolfson, and T.  Doggett (eds.), Food, Ethics, and Society. New York: Oxford University Press: 215–18. Michalson, G. E. (1999) Kant and the Problem of God. New York: Blackwell. Michalson, G. E. (ed.) (2014) Religion within the Bounds of Mere Reason: A Critical Guide. New York: Cambridge University Press. Mill, J. S. (1861) Utilitarianism. Public domain material available at: www.utilitarianism. com/mill1.htm. Millikan, R.  G. (1984) Language, Thought, and Other Biological Categories. Cambridge, Massachusetts: Massachusetts Institute of Technology Press. Millikan, R. G. (1989) ‘Biosemantics’, The Journal of Philosophy, 86 (6): 281–97. Millikan, R. G. (1990) ‘Truth Rules, Hoverflies, and the Kripke-Wittgenstein Paradox’, The Philosophical Review, 99 (3): 323–53. Mittleman, A. (2009) Hope in a Democratic Age. New York: Oxford University Press. Montaigne, M. de (2003) Essays (trans. M. A. Screech). London: Penguin Books. Moore, G. E. (1903) Principia Ethica. Public domain material available at: fair-use.org/g-emoore/principia-ethica. Nagel, T. (1974) ‘What Is It Like to Be a Bat?’ The Philosophical Review, 83 (4): 435–50. Nagel, T. (1978) ‘Ethics without Biology’, in G.  S.  Stent (ed.), Morality as a Biological Phenomenon, Berlin, West Germany, Dahlem Konferenzen: 221–32. Nagel, T. (1986) The View from Nowhere. Oxford, England: Oxford University Press. Naragon, S. (1990) ‘Kant on Descartes and the Brutes’, Kant-Studien, 81 (1): 1–23. Nefsky, J. (2017) ‘How You Can Help Without Making a Difference’, Philosophical Studies, 174: 2743–67. Nefsky, J. (2018) ‘Consumer Choice and Collective Impact’, in Barnhill, A. Budolfson, M. B. and Doggett, T. (eds.) (2018) Oxford Handbook of Food Ethics. New York: Oxford University Press: 267–86. Nobis, N. (2002) ‘Vegetarianism and Virtue: Does Consequentialism Demand Too Little?’, Social Theory and Practice, 28 (1): 135–56. Norcross, A. (2004) ‘Puppies, Pigs, and People: Eating Meat and Marginal Cases’, Philosophical Perspectives, 18: 229–45.

252

Bibliography

Norcross, A. (draft) ‘More Mistakes in Moral Mathematics’. Nelson, L. (1956) System of Ethics. Trans. Norbert Guterman. New Haven: Yale University Press. Nozick, R. (1969) ‘Newcomb’s Problem and Two Principles of Choice’, in N. Rescher (ed.), Essays in Honor of Carl Hempel. Dordrecht: Reidel: 440–72. Nozick, R. (1974) Anarchy State and Utopia. New York: Basic Books. O’Brien, L. (2007) Self-Knowing Agents. Oxford: Oxford University Press. O’Neal, J.  C. (1996) The Authority of Experience: Sensationist Theory in the French Enlightenment. University Park, Pa.: Pennsylvania State University Press. O’Neill, O. (1998) ‘Kant on Duties Regarding Nonrational Nature’, Aristotelian Society Supplementary Volume 72 (1): 211–28. O’Shea, James (2012) ‘Kant and Dennett on the Epistemic Status of Teleological Principles’, in F.  O’Rourke (ed.), Human Destinies. Philosophical Essays in Memory of Gerald Hanratty, University of Notre Dame Press): 252–70. Okrent, M. (2006) ‘Acquaintance and Cognition’, in R. Kukla (ed.), Aesthetics and Cognition in Kant’s Critical Philosophy, Cambridge: Cambridge University Press: 85–108. Olson, Michael J. (2016) ‘Kant on Anatomy and the Status of the Life Sciences’, Studies in History and Philosophy of Science, 58: 77–84. Owen, D. (1999) Hume’s Reason. Oxford: Oxford University Press. Owen, D. (2007) ‘Locke on Judgment’, in L. Newman (ed.), The Cambridge Companion to Locke’s ‘Essay Concerning Human Understanding’. Cambridge: Cambridge University Press: 406–35. Parcell, J. and Pierce, V. (2000) ‘Factors Affecting Wholesale Chicken Prices’, Journal of Agricultural and Applied Economics, 32 (3): 471–8. Parfit, D. (1984) Reasons and Persons. Oxford: Oxford University Press. Parfit, D. (2011) On What Matters. Oxford: Oxford University Press. Paton, H. J. (1947) The Categorical Imperative: A Study in Kant’s Moral Philosophy. London: Hutchison. Peacocke, C. (1992a) A Study of Concepts. MIT Press. Peacocke, C. (1992b) ‘Sense and Justification’, Mind, 101 (404): 793–816. Peacocke, C. (2008) ‘Sensational Properties: Theses to Accept and Theses to Reject’, Revue Internationale de Philosophie, 1: 7–24. Peacocke, C. (2009) ‘Objectivity’, Mind, 118 (471): 739–69. Pepperberg, I., Vicinay, J., and Cavanagh, P. (2008) ‘Processing of the Müller-Lyer Illusion by a Grey Parrot (Psittacus Erithacus)’, Perception, 37: 765–81. Pereboom, D. (1988) ‘Kant on Intentionality’, Synthese, 77 (3): 321–52. Pereboom, D. (1995) ‘Self-Understanding in Kant’s Transcendental Deduction’, Synthese, 103 (1): 1–42. Pereboom, D. (2006) ‘Kant’s Metaphysical and Transcendental Deductions’, in G.  Bird (ed.), A Companion to Kant, Blackwell Publishing: 154–68. Pereboom, D. (2009) ‘Kant’s Transcendental Arguments’, Stanford Encyclopedia of Philosophy. (Spring 2019 Edition), Edward N. Zalta (ed.), URL = . Pettit, P. (2004) ‘Hope and Its Place in Mind’, Annals of the American Academy of Political and Social Science, 592: 152–65. Pinkert, F. (2015) ‘What If I Cannot Make a Difference (and Know It)?’, Ethics, 125: 971–98. Pittard, J. (2018) ‘Evil and God’s Toxin Puzzle’, Nous, 52 (1): 88–108. Quarfood, M. (2004) Transcendental Idealism and the Organism. Stockholm: Almquist & Wiksell.

Bibliography

253

Quarfood, M. (2014) ‘The Antinomy of Teleological Judgment: What It Is and How It Is Solved’, in I. Goy and E. Watkins (eds.), Kant’s Theory of Biology, Berlin/New York: De Gruyter: 167–84. Raftopoulos, A. and Zeimbekis, J. eds. (2015) The Cognitive Penetrability of Perception: New Philosophical Perspectives. Oxford: Oxford University Press. Rawls, J. (1971) A Theory of Justice. Cambridge, Mass.: Harvard University Press. Rawls, J. (1988) ‘The Priority of Right and Ideas of the Good’, in S. Freeman (ed.) Collected Papers, Cambridge, Massachusetts: Harvard University Press: 449–72. Raz, J. (1986) The Morality of Freedom. Oxford: Clarendon Press. Regan, T. (1983) The Case for Animal Rights. Berkeley, California: University of California Press. Regan, T. (2001) The Animal Rights Debate. Lanham: Rowman & Littlefield. Regan, T. (2003) Animal Rights, Human Wrongs. Lanham, MD: Rowman & Littlefield. Roessler, J. (2009) ‘Perceptual Experience and Perceptual Knowledge’, Mind, 118 (472): 1013–41. Roessler, J., Hemdat L., and Naomi, E. (eds.) (2011) Perception, Causation, and Objectivity. Oxford: Oxford University Press. Rorty, A.  O. and Schmidt, J. (eds.) (2012) Kant’s Idea for a Universal History with a Cosmopolitan Aim: A Critical Guide. Cambridge: Cambridge University Press. Rosefeldt, T. (2003) ‘Kant’s Self: Real Entity and Logical Identity’, in H.  Glock (ed.) Strawson and Kant, Oxford: Oxford University Press, 141–54. Rosen, G. (1994) ‘Objectivity and Modern Idealism: What is the Question?’, in M. Michaelis and J.  Hawthorne (eds.), Philosophy in Mind: The Place of Philosophy in the Study of Mind, Dordrecht: Springer, 277–319. Rosenfield, L.  C. (1941) From Beast-Machine to Man-Machine Animal Soul in French Letters From Descartes to La Mettrie. Oxford: Oxford University Press. Rousseau, J-J. (1979) Emile: Or, On Education. Trans. by Allan Bloom. New York: Basic Books. Ruse, M. (2006) ‘Kant and Evolution’, in J.  E.  H.  Smith (ed.), The Problem of Animal Generation in Early Modern Philosophy, Cambridge: Cambridge University Press: 402–16. Sabaté, J. (2003) ‘The contribution of vegetarian diets to health and disease: a paradigm shift?’, American Journal of Clinical Nutrition, 78: 502S–507S. Schafer, K. (forthcoming) ‘Kant’s Conception of Cognition and Our Knowledge of Thingsin-Themselves’, in K.  Schafer and N.  Stang (eds.), The Sensible and Intelligible Worlds, Oxford University Press. Schellenberg, S. (2015) ‘The Relational and Representational Character of Perceptual Experience’, in B.  Brogaard (ed.), Does Perception Have Content?, Oxford: Oxford University Press: 199–219. Schopenhauer, Arthur (1840/2000) On the Basis of Morality (trans. E.  F.  J.  Payne). Indianapolis: Hackett.Schuster, P. (2011) ‘Is there a Newton of the blade of grass? The Complex Relation Between Mathematics, Physics, and Biology’, Complexity, 16 (6): 5–9. Sebo, J. (2004) ‘A Critique of the Kantian Theory of Indirect Moral Duties to Animals’, Animal Liberation Philosophy and Policy Journal, 2: 54–72. Shanahan, T. (2004) The Evolution of Darwinism: Selection, Adaptation and Progress in Evolutionary Biology. Cambridge: Cambridge University Press. Shell, S. M. (2003) ‘Kant’s ‘True Economy of Human Nature’: Rousseau, Count Verri, and the Problem of Happiness,’ in B. Jacobs and P. Kain (eds.), Essays on Kant’s Anthropology, Cambridge: Cambridge University Press: 194–229.

254

Bibliography

Siegel, S. (2008) ‘The Epistemic Conception of Hallucination’, in A.  Haddock and F. Macpherson (eds.), Disjunctivism. Oxford: Oxford University Press: 205–24. Siegel, Susanna (2017) The Rationality of Perception. New York: Oxford University Press. Singer, Peter (1975) Animal Liberation. New York: Random House. Singer, Peter (1980) ‘Utilitarianism and Vegetarianism’, Philosophy and Public Affairs, 9 (4): 325–37. Sinnott-Armstrong, W. (2015) ‘Consequentialism’, in Stanford Encyclopedia of Philosophy (Winter 2015 Edition), ed. by Edward Zalta. https://plato.stanford.edu/archives/ win2015/entries/consequentialism, accessed 5 January 2018. Skidmore, J. (2001) ‘Duties to Animals: The Failure of Kant’s Moral Theory’, Journal of Value Inquiry, 35: 541–59. Sloan, P. R. (2002) ‘Preforming the Categories: Eighteenth Century Generation Theory and the Biological Roots of Kant’s A Priori’, Journal of the History of Philosophy, 40 (2): 229–53. Smith, A. D. (2002) The Problem of Perception. Harvard University Press. Smith, D. W. (1982) Helvetius: A Study in Persecution. Westport, Conn.: Greenwood Press. Smith, J. E. H. (2011) Divine Machines: Leibniz and the Sciences of Life. Princeton University Press. Smith, J.  E.  H. (2012) The Problem of Animal Generation in Early Modern Philosophy. Cambridge: Cambridge University Press. Solnit, Rebecca (2004) Hope in the Dark. New York: Nation Books. Sommerlatte, Curtis (2016) ‘Empirical Cognition in the Transcendental Deduction: Kant’s Starting Point and His Humean Problem’, Kantian Review, 21 (3): 437–63. Sorabji, Richard (1993) Animal Minds and Human Morals: The Origins of the Western Debate. Vol. 45. Cornell University Press. Speaks, J. (2005) ‘Is There a Problem About Nonconceptual Content?’, Philosophical Review, 114: 359–98. Spelke, E. S. (1990) ‘Principles of Object Perception’, Cognitive Science, 14 (1): 29–56. Spelke, E. S., Vishton, P., and Von Hofsten, C. (1995) ‘Object Perception, Object-Directed Action, and Physical Knowledge in Infancy’, in M.  Gazzaniga (ed.), The Cognitive Neurosciences, Cambridge, MA: The MIT Press: 165–79. Spiekermann, K. (2014) ‘Small Impacts and Imperceptible Effects: Causing Harm with Others’, Midwest Studies in Philosophy, 38: 75–90. Stephenson, A. (2016) ‘Kant on the object-dependence of intuition and hallucination’, Philosophical Quarterly, 65: 486–508. Stocker, M. (1976) ‘The Schizophrenia of Modern Moral Theories’, Journal of Philosophy, 73, 453–66. Strawson, P. F. (1959) Individuals. London: Methuen. Strawson, P. F. (1966) The Bounds of Sense. London: Routledge. Teufel, T. (2013) ‘The Impossibility of a ‘Newton of the Blade of Grass’ in Kant’s Teleology’, in J. Smith, O. Nachtomy (eds.), The Life Sciences in Early Modern Philosophy, Oxford: Oxford University Press): 47–61. Thompson, M. (2008) Life and Action. Cambridge, Massachusetts: Harvard University Press. Timmermann, J. (2005) ‘When the Tail Wags the Dog: Animal Welfare and Indirect Duties in Kantian Ethics’, Kantian Review, 10: 128–49. Tolley, C. (2013) ‘The Non-Conceptuality of the Content of Intuitions: A New Approach’, Kantian Review, 18: 107–36.

Bibliography

255

Tye, M. (2009) ‘Representationalist Theories of Consciousness’, in A.  Beckermann, B.  McLaughlin, and S.  Walter (eds.), The Oxford Handbook of Philosophy of Mind. Oxford University Press. Tucker, B. and MacDonald, C. (2004) ‘Beastly Contractarianism? A Contractarian Analysis of the Possibility of Animal Rights’, Essays in Philosophy, 5: 2 article 31. Treisman, A.  M. and Gelade, G. (1980) ‘A Feature-Integration Theory of Attention’. Cognitive Psychology, 12 (1): 97–136. Van Cleve, J. (1999) Problems from Kant. Oxford: Oxford University Press. van Houten, Judith (2000) ‘Chemoreception in Microorganisms’, in T. E. Finger, W. L. Silver and D. Restrepo (eds.), The Neurobiology of Taste and Smell, New York, New York: John Wiley & Sons: 11–40. Varden, H. (2012) ‘A Kantian Critique of the Care Tradition: Family Law and Systemic Justice’, Kantian Review, 17 (2): 327–56. Varden, H. (2014) ‘The Terrorist Attacks in Norway, July 22nd 2011— Some Kantian Reflections’, Norsk Filosofisk Tidsskrift/Norwegian Journal of Philosophy, 49 (3–4): 236–59. Varden, H. (2017) ‘Kant and Women’, Pacific Philosophical Quarterly, 98 (4): 653–94. Varden, H. (2020) Sex, Love, and Gender: A Kantian Theory. Oxford: Oxford University Press. Varner, G. (2012) Personhood, Ethics, and Animal Cognition. Oxford, England: Oxford University Press. Velkley, R. L. (1989) Freedom and the End of Reason: On the Moral Foundation of Kant’s Critical Philosophy. University of Chicago Press. Velkley, R. L. (2013) ‘Transcending Nature, Unifying Reason: On Kant’s Debt to Rousseau’, in O.  Sensen, Kant on Moral Autonomy, Cambridge: Cambirdge University Press: 89–106. Velleman, D. (1999) ‘A Right of Self-Termination?’ Ethics, 109 (3): 606–28. Warren, M.  A. (1973) ‘On the Moral and Legal Status of Abortion’, The Monist, 57(1): 43–61. Watkins, E. (2009) Kant’s Critique of Pure Reason: Background Source Materials. Cambridge: Cambridge University Press. Watkins, E. and Willaschek, M. (2017a) ‘Kant’s Account of Cognition’, Journal of the History of Philosophy, 55 (1): 83–112. Watkins, E. and Willaschek, M. (2017b) ‘Kant on Cognition and Knowledge’, Synthese, 1–19. Watson, G. (1975) ‘Free Agency’, Journal of Philosophy, 72 (8): 205–20. Watson, R. (1979) ‘Self-Consciousness and the Rights of Nonhuman Animals and Nature’, Environmental Ethics, 1 (2): 99–129. Westphal, K. (2004) Kant’s Transcendental Proof of Realism. Cambridge: Cambridge University Press. Willaschek, M. (2016) ‘Must We Believe in the Realizability of our Ends? On a Premise of Kant’s Argument for the Postulates of Pure Practical Reason’, in T.  Höwing (ed.), The Highest Good in Kant’s Philosophy. Berlin: De Gruyter: 223–44. Williams, B. (1978) Descartes: The Project of Pure Enquiry. London: Routledge. Williams, B. (1981) ‘Persons, Character, and Morality’, in Moral Luck, Cambridge University Press: 1–19. Williams, J.  J. (2012) ‘How Conceptually Guided are Kantian Intuitions?’, History of Philosophy Quarterly, 29 (1): 57–78.

256

Bibliography

Wilson, H. L. (2001) ‘Kant’s Views on Human Animality’, in V. Gerhardt, R.-P. Horstmann, and R.  Schumacher (eds.), Kant und die Berliner Aufklärung: Akten des IX. Internationalen Kant-Kongress, Band 4, Berlin: Walter de Gruyter: 450–7. Wilson, H. L. (2008) ‘The Green Kant: Kant’s Treatment of Animals’, in L. P. Pojman and P. Pojman (eds.), Environmental Ethics: Readings in Theory and Application, 6th edition, Boston: Wadsworth: 87–96. Wilson, H. L. (2011) ‘Kant’s Treatment of Animals’, in P. Pojman (ed.) Food Ethics, Boston: Wadsworth: 12–20. Wilson, H. L. (2012) ‘The Green Kant: Kant’s Treatment of Animals’, in L. P. Pojman and P. Pojman (eds.), Environmental Ethics: Readings in Theory and Application, 6th edition, Boston: Wadsworth: 62–70. Wilson, M.  D. (1999) ‘Animal Ideas’ in Ideas and Mechanism, Essays on Early Modern Philosophy. Princeton University Press: 495–512. Wood, A. (1970) Kant’s Moral Religion. Ithaca: Cornell University Press. Wood, A. (1978) Kant’s Rational Theology. Cornell: Cornell University Press. Wood, A. (1998) ‘Kant on Duties Regarding Nonrational Nature I’, Proceedings of the Aristotelian Society Suppl. Vol. 72 (1): 189–210. Wood, A. (1999) Kant’s Ethical Thought. Cambridge: Cambridge University Press, 1999. Wood, Allen (2008) Kantian Ethics. Cambridge, England: Cambridge University Press. Wood, A. (2010) ‘Kant and the Intelligibility of Evil’, in S. Anderson-Gold and P. Muchnik (eds.), Kant’s Anatomy of Evil, Cambridge: Cambridge University Press: 144–72. Wood, A. (2012) ‘Kant’s Fourth Proposition: The Unsocial Sociability of Human Nature’, in A. O. Rorty and J. Schmidt (eds.), Kant’s Idea for a Universal History with a Cosmopolitan Aim: A Critical Guide. Cambridge: Cambridge University Press: 112–28. Wright, L. (1973) ‘Functions’, The Philosophical Review, 82 (2): 139–68. Wunderlich, F. (2005) Kant Und Die Bewußtseinstheorien Des 18. Jahrhunderts. Berlin: Walter de Gruyter. Yanklowitz, S. (2015) The Jewish Vegan. New York: The Shamayim V’Aretz Institute. Yolton, J. W. (1991) Locke and French Materialism. Vol. 42. Oxford University Press. Zammito, J.  H. (2002) Kant, Herder, and the Birth of Anthropology. Chicago, London: University of Chicago Press. Zammito, J. H. (2003) ‘This Inscrutable Principle of an Original Organization’: Epigenesis and ‘Looseness of Fit’ in Kant’s Philosophy of Science’, Studies in History and Philosophy of Science, 34: 73–109. Zammito, J.  H. (2006) ‘Policing Polygeneticism in Germany, 1775: (Kames,) Kant, and Blumenbach’, in S. Eigen and M. Larrimore (eds.), The German Invention of Race, New York: SUNY: 35–54. Zammito, J.  H. (2012) ‘The Pursuit of Science as Decadence in Kant’s Remarks in ‘Observations on the Feeling of the Beautiful and Sublime’, in S.  Meld Shell and R.  Velkley (eds.) Kant’s Observations and Remarks: A Critical Guide, Cambridge: Cambridge University Press: 234–46. Zöller, G. (2011) ‘Kant’s Political Anthropology’, Kant Jahrbuch, 3: 131–62. Zuckert, R. (2007) Kant on Beauty and Biology. An Interpretation of the ‘Critique of Judgment’. Cambridge: Cambridge University Press.

Index Acquaintance 45, 49–54, 62–3, 223 Allais, Lucy 6–7, 58–9 Analogy 23–4, 40, 155 A posteriori 76 A priori 92, 94, 98 Apperception 4, 20, 36, 51, 54–5, 63, 75 Apprehension 20, 35, 77–8 Aristotle 194 Aristotelian 26–8, 191–2, 197–8, 212 Autonomy 144–5, 181–4, 192–3, 209 Baumeister, David 9–10, 14 Bayle, Pierre 23–4, 26–8 Bowman, Brady 73 Brewer, Bill 82 Budolfson, Bryant 216–17 Burge, Tyler 69–70, 205–7 Callanan, John 4–5 Categories 6–7, 48, 55–6, 77, 79, 203 Causality 79, 91–5, 97–8, 131–2 Chignell, Andrew 16 Children 32, 114–16, 145, 147, 153, 165–6, 168–9 Cognition 1–2, 7, 20–1, 39–40, 44–5, 48–51, 53, 62, 72, 74–5, 149 Cohen, Alix 10 Combination 35, 42, 55, 60–1, 67, 77 Comparison 28–9, 32–3, 35 Concepts 2, 6–7, 11, 20–1, 58, 69, 72, 74–5, 84–5, 154, 158–9, 165 Conceptualism 6–7 Condillac, Étienne 23, 32–3, 43 Consciousness 6–7, 20–2, 25, 209, 211 Constructivism 15, 176–7, 181–2, 187, 193–6, 204, 214 Crane, Tim 81–2 Darwin 10, 123, 133–5, 199 Deduction See Transcendental Deduction Desire 2, 8–10, 16, 148, 171 Despair 16, 224–5 Denis, Lara 13 Descartes, René 4–5, 22–6, 38, 41 Dignity 105, 169, 177–8, 183, 193, 213

Distinctness 36–7 Duties 149–51 direct 10–11, 14–15, 157, 159, 165, 178, 213 directionality of 139 indirect 10–12, 14, 179–81, 213 Education 106, 112–13, 117–18 End-in-itself 15, 177 Enlightenment 90, 103–4, 192–3, 211–12 Epigenesis 126–30 Evans, Gareth 43 Evil 171–4 Evolution 10, 132–4 Freedom 10, 92, 97–8, 101, 111, 155, 157–9, 161, 166–7, 171 Frege, Gottlob 72 Frey, R.G. 216 Geography 124–5 Garthoff, Jon 16 Ginsborg, Hannah 70–1, 73, 79, 97 God 9, 89, 98–100, 102–4, 218, 228, 231–2 Golob, Sacha 6–8 Goy, Ina 9 Grüne, Stephanie 53–4 Hanna, Robert 70 Happiness 96, 185, 226–7 Hay, Carol 14–15 Helvétius, Claude Adrien 23, 32–4 Herman, Barbara 13–14 History 117, 119 Hope 15, 226–30 Holy Will 154–5 Hume, David 150–1 Husserl, Edmund 84–6 Imagination 4, 20, 23–4, 32, 40–1, 56, 63–4, 75, 82 Inner sense 5, 21, 54 Intentionality 8, 54, 69–70, 80–1, 95 Intuition 6–7, 20–1, 42, 44, 48–9, 53–5, 59–62, 69, 72, 77, 82, 88, 100, 154

258

Index

James, William 233–4 Judgment 32–4, 37, 204–5, 207 Kain, Patrick 13–14 Kierkegaard, Søren 233–4 Korsgaard, Christine 10, 15–16, 170, 198, 201–14 La Mettrie, Julian Offray de 23, 32 Language 32 Laws 89–90, 92–7 Leibniz, Gottfried Wilhelm 30–1 Linnaeus, Carl 125 Locke, John 23, 28–9 Louden, Robert 114 Martin, Michael 216 Matter 3–4, 8–9, 15–16, 60, 92–5, 126, 131–3 McDowell, John 80–1 McLear, Colin 6–8, 83–4 McMahan, Jeff 141–2 Mechanism 92–3 Michaelson, Eliot 216–17 Mill, James Stuart 144, 191–2 Montaigne, Michel de 4–5, 24–5, 38, 41 Moore, G.E. 210, 212 Morality 110 Naragon, Steve 2–3, 6 Nature 4, 8–9, 11, 38–9, 42, 91–4, 96, 98–101, 103, 108, 112–13, 116, 119, 123–5, 129, 131–2, 134–5, 140–1, 148, 172–3, 189 Newton, Isaac 123, 135 Nonconceptualism 6–7, 68, 70 O’Brien, Lucy 85 Object 7, 11, 21, 54, 71 Objectivity 8, 43, 45–8, 76–8, 84 Obligation(s) 145, 155, 159, 164–5, 168, 196, 208–10 Outer sense 5 Paralogisms 50 Parfit, Derek 217 Peacocke, Christopher 81–2 Perception 56–7, 59, 69–70, 72, 74, 77–9, 86, 198–200, 204 Plants 8–9, 11, 125, 129, 132–3, 195, 198, 213 Plato 153 Perspective 46–7 Predisposition(s) 9, 106, 108, 127–8, 161–3, 169–71 Preformation 126, 129–30

Race 129–30 Reason/Rationality 26, 31, 42, 195, 205–6 Regan, Tom 166, 192, 213–14 Regulative 8–9, 89–90, 93, 98–104 Respect 10–11, 121, 143, 146, 157–8, 160–2, 165, 169, 177–8, 183–4, 191–2 Rights 166–7 Ripstein, Arthur 14 Rousseau, Jean-Jacques 23, 34–5 Schellenberg, Susanna 80–1 Schopenhauer, Arthur 142 Self-consciousness 6–7, 10, 15–16, 21–2, 25–6 Sensation 32–3, 36, 42–3 Sidgwick 144 Siegel, Susanna 82–3 Singer, Peter 166, 192 Smith, A.D. 82–3, 85 Soul 36 Space/Spatial 20, 22–3, 55–60, 63, 69–70, 78–9, 85–8 Spinoza, Bruch 224–5, 231 Spontaneity 3, 55, 60–1, 64, 100 Strawson, P.F. 86 Substance 25, 55–6, 63 Synthesis 7, 48, 54–5, 67 Teleology 8–9, 93–6, 131–2, 134–5 Tenenbaum, Sergio 14 Tolley, Clinton 67 Transcendental Deduction 73, 77 Unconditioned 1–2, 97–8 Understanding 2, 5, 21, 40, 42, 44–5, 48, 61, 63, 66–8, 98–100 Universalization 235 Value 14–15, 196, 202–3, 209, 213–14 Intrinsic 176 Non-Intrinsic 177–8, 181, 189–90 Van der Zalm, Bill 140 Varden, Helga 14 Virtue 149 Warren, Mary Anne 187 Wilson, Holly 110–12 Wood, Allen 109, 116–17 Yanklowitz, Rabbi Shmuly 222 Zöller, Günter 121 Zuckert, Rachel 97