Maimonides on God and Duns Scotus on Logic and Metaphysics [Unabridged] 144387843X, 9781443878432

Citation preview

Maimonides on God and Duns Scotus on Logic and Metaphysics

Also available in the series: The Immateriality of the Human Mind, the Semantics of Analogy, and the Conceivability of God Volume 1: Proceedings of the Society for Medieval Logic and Metaphysics Categories, and What Is Beyond Volume 2: Proceedings of the Society for Medieval Logic and Metaphysics Knowledge, Mental Language, and Free Will Volume 3: Proceedings of the Society for Medieval Logic and Metaphysics Mental Representation Volume 4: Proceedings of the Society for Medieval Logic and Metaphysics Universal Representation, and the Ontology of Individuation Volume 5: Proceedings of the Society for Medieval Logic and Metaphysics Medieval Skepticism, and the Claim to Metaphysical Knowledge Volume 6: Proceedings of the Society for Medieval Logic and Metaphysics Medieval Metaphysics; or Is It "Just Semantics"? Volume 7: Proceedings of the Society for Medieval Logic and Metaphysics After God, with Reason Alone—Saikat Guha Commemorative Volume Volume 8: Proceedings of the Society for Medieval Logic and Metaphysics The Demonic Temptations of Medieval Nominalism Volume 9: Proceedings of the Society for Medieval Logic and Metaphysics Skepticism, Causality and Skepticism about Causality Volume 10: Proceedings of the Society for Medieval Logic and Metaphysics Metaphysical Themes, Medieval and Modern Volume 11: Proceedings of the Society for Medieval Logic and Metaphysics

Maimonides on God and Duns Scotus on Logic and Metaphysics Volume 12: Proceedings of the Society for Medieval Logic and Metaphysics

Maimonides on God and Duns Scotus on Logic and Metaphysics: (Volume 12: Proceedings of the Society for Medieval Logic and Metaphysics) Edited by

Gyula Klima and Alexander W. Hall

Maimonides on God and Duns Scotus on Logic and Metaphysics (Volume 12: Proceedings of the Society for Medieval Logic and Metaphysics) Edited by Gyula Klima and Alexander W. Hall This book first published 2015 Cambridge Scholars Publishing Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2015 by Gyula Klima, Alexander W. Hall and contributors All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-4438-7843-X ISBN (13): 978-1-4438-7843-2

TABLE OF CONTENTS

Introduction ............................................................................................... vii Alex Hall Part 1: Maimonides on God Creation and the Argument from Particularity ............................................ 3 Kenneth Seeskin Maimonides, Aquinas, and Particularity: Comments on Seeskin’s “Creation and the Argument from Particularity” ....................................... 17 Stephen R. Ogden The Sixth Method Doesn’t Work, But the Third Way Does?: Moses Maimonides’ Assessment of Two Proofs Based on Possibility and Necessity ............................................................................................. 23 Jamie Anne Spiering Conceiving Creation: Response to Spiering and Seeskin .......................... 45 Daniel Davies Part 2: Duns Scotus on Logic and Metaphysics On the Authenticity of Scotus’s Logical Works ........................................ 55 Edward Buckner Scotus on the Species of Quality ............................................................... 85 Lloyd A. Newton Logic, Ontology, and the Psychology of Universals in Duns Scotus ...... 101 Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio The Semantic Unity of the Analogous Concept according to John Capreolus .................................................................................... 133 Domenic D’Ettore

vi

Table of Contents

Can We Speak About That Which Is Not? Actualism and Possibilism in Analytic Philosophy and Scholasticism............................................... 155 Lukáš Novák Appendix ................................................................................................. 189 Contributors ............................................................................................. 191

INTRODUCTION ALEX HALL

The Proceedings of the Society for Medieval Logic and Metaphysics (PSMLM) collects original materials presented at sessions sponsored by the Society for Medieval Logic and Metaphysics (SMLM). SMLM was founded in 2000 by Gyula Klima (Director), Joshua Hochschild, Jack Zupko and Jeffrey Brower, in order to recover the profound metaphysical insights of medieval thinkers for our own philosophical thought. The Society currently has over two hundred members on five continents. Alex Hall took up the position of Assistant Director and Secretary in 2011, with secretarial duties passing to Timothy Kearns in 2014. The Society’s maiden publication appeared online in 2001 and the decade that followed saw the release of eight more online volumes. In 2011, PSMLM transitioned to print and republished volumes 1-8 as separately titled editions. Sharp-eyed readers of these volumes will note the replacement of our lions (lamentably copyrighted for commercial use), who guarded the integrity of the body of an intellectual tradition thought to be dead, with the phoenixes that mark this print rebirth. Volumes 9 and 10 appeared in a dual print/online format, with Volume 11 PSMLM switched to print only. Friends of the lions will be happy to note that they remain at their post, protecting the first ten volumes of the PSMLM at http://faculty. fordham.edu/klima/SMLM/, where interested readers can also keep up with SMLM activities and projects. The 2014 PSMLM (the twelfth in the series) comes in two parts. Part 1, Maimonides on God, presents SMLM papers read at a satellite session of the American Catholic Philosophical Association, hosted by The Catholic University of America in Washington D.C. SMLM has sponsored such sessions at the ACPA since 2001. Beginning in 2011, SMLM has likewise gathered annually at the International Congress on Medieval Studies at Western Michigan University. Part 2, The Logic and Semantics of John Duns Scotus, draws from our ICMS panel, featuring in addition several papers collected through an open call. Volumes 13 and 14 (forthcoming in

viii

Introduction

2017) will feature papers on medieval hylomorphism and mereology and medieval theories of philosophy of mind, respectively.

Part 1 – Maimonides on God Moses Maimonides and Thomas Aquinas agree that we lack any valid argument for creation de novo and yet they refuse to cede its possibility to Aristotelian thinkers who draw on familiar laws of nature to deny the notion’s coherence. In response, Maimonides and Aquinas charge the Aristotelians with an illicit extrapolation. Who is to say what laws might or might not obtain during such a singularity event? Hence it seems that the question of creation de novo must necessarily remain open and the faithful may then rest content. Or not. As Kenneth Seeskin notes in Creation and the Argument from Particularity, Maimonides and Aquinas may part company on this point, for, whereas Aquinas will allow that creation de novo must be accepted “on faith alone” (Summa Theologica (ST) 1.46.2), Seeskin suspects that Maimonides recommends authority only for those who “cannot follow the technical arguments for creation” that Maimonides will present. Accordingly, Seeskin asks why Maimonides invokes rational argument in this matter whereas Aquinas does not consistently adopt this approach. As Seeskin gathers the materials that will constitute his answer, he traces alongside his discussion of Maimonides and Aquinas reflections concerning contemporary science, philosophy and theology that comprise in themselves an argument for the persistent (or rather, invariably systemic) persuasive force of intelligent design theory in its various forms. To make the case for creation de novo, Maimonides adapts the Islamic KalƗm tradition of rationalist theology. Dating to eighth-century Iraq, KalƗm comprises roughly a millennium of theological and philosophical literature in Arabic, inspired in its late, post eleventh-century period by neoplatonized Aristotelian philosophy. According to Maimonides, Mutakallimnjn (practitioners of KalƗm) uniformly rely on what has come to be called the Argument from Particularity, which holds that the fact that things could have been otherwise than they in deed are betrays the hand of an intelligent designer (Guide of the Perplexed (GP 1.71)). Maimonides is critical of Mutakallimnjn reliance on imagination to establish the principle that things might always be other than they are (that we can imagine this certainly does not make it the case) and the occasionalist threat to methodological naturalism to which this attitude gives rise. Nevertheless, twelfth-century Ptolemaic and Aristotelian systems were admittedly

Maimonides on God and Duns Scotus on Logic and Metaphysics

ix

subject to “grave incongruities and perversities” (GP 2.22), such that Maimonides despairs of astronomy attaining the rank of a science (GP 2.24). For this reason, Maimonides discerns in the motions of the heavenly bodies the work of an intelligent designer, allowing that, should science provide a principled explanation for these motions, the doctrine of eternity would carry the day, necessitating a reinterpretation of the Torah. Seeskin suggests that Maimonides’ optimism relative to Aquinas as regards this extension of Mutakallimnjn proof strategy develops out of Maimonides’ comparative willingness to subject dogma to trends in scientific opinion and an epistemic privilege that Aquinas (but not Maimonides) wants to assign to faith over reason. Stephen Ogden’s response to Seeskin, Maimonides, Aquinas, and Particularity, commends Seeskin’s development of the Argument from Particularity, which allows Seeskin to make the case that, as long as science recognizes arbitrary constants, such as the precise value of the force of gravity, the strategy remains effective, while taking issue with the aforementioned contrasts that Seeskin draws between the two thinkers. At GP 1.74, Maimonides sketches seven of the Mutakallimnjn’s arguments for creation de novo (and, hence, for the existence of God), several of which deploy the particularization strategy. He offers a mix of criticism and praise for these arguments, contrasting them with those of the neoplatonized Aristotelian philosophers who contend that the world is eternal (GP 1.71) and adduce this conclusion in their attempt to demonstrate the existence of God (GP 2.1). As noted, however, Maimonides holds that reason cannot decide in this matter (GP 1.71) and hence neither the Mutakallimnjn nor the Aristotelian proofs, at least taken in isolation from one another, are demonstrative. Nevertheless, this provides a disjunctive strategy as the universe is either created or not. Provided that we have good proof that God exists on both accounts, then, whether the universe is or is not created, we have evidence for the existence of the creator (GP 1.71). It is in this spirit that Maimonides offers a philosophical proof for the existence of God, which he attributes to Aristotle, based on premises that allow for (and may even suppose) the eternity of the universe, in the opening chapter of Book Two of the Guide. And yet this philosophical proof (Maimonides’ third way in GP 2.1 – not to be confused with Aquinas’ own, rather similar, Third Way in ST 1.2.3c) very much resembles a Mutakallimnjn argument (the sixth method) that Maimonides criticizes (though perhaps not decisively) at GP

x

Introduction

1.74, as both draw on the inference that the way things in fact are is not the way that they must be, betraying the hand of an intelligent designer. Jamie Spiering’s The Sixth Method Doesn’t Work, But the Third Way Does? seeks to account for Maimonides’ rejection of the former given his endorsement of the latter (or rather, what appears to be Maimonides’ estimation in each case, given the question of his esotericism). Spiering argues that, despite surface similarities, the proofs are crucially different, particularly inasmuch as the sixth method appears question begging, relies on the principle that what is imaginable is thereby possible and culminates in a rather vague notion of some (possibly corporeal) place-holder type of deity. Exposing these and other differences allows Spiering to shed light on the fundamental nature of the project of proving God’s existence, affording her the opportunity to evaluate certain contemporary proof strategies. Between one extreme that contends that Maimonides’ KalƗm and philosophical proofs are sound and its opposite, Daniel Davies notes that Spiering’s admittedly naïve endorsement of Maimonides’ apparent literal meaning profitably strikes out on a new path (Conceiving Creation: Response to Spiering and Seeskin). Davies speaks to the strengths of Spiering’s treatment relative to so-called sophisticated interpretations that advance esoteric readings to discern Maimonides’ true intent. Davies’ discussion takes up what he sees as the root of Maimonides’ criticism of the KalƗm proof strategy, namely, that it relies too much on imagination and hence issues in an extreme occasionalist teaching that denies any natural causality. But if this is the case, why, as we have seen, does Maimonides employ a modified version of this strategy drawn from the motions of the planets at GP 2.19-24? Here Davies’ discussion shifts to Seeskin’s and Ogden’s accountings on this point and urges that Maimonides advances a particularization argument that is consistent with empirical evidence and only looks beyond explanation through natural causality when the phenomena seem to require it. This attitude is in keeping with Maimonides’ own theological positions that God creates through will and that violations of natural law occur in the form of miracles, positions which should lead one to expect phenomena that cannot be accounted for within a nomic, Aristotelian framework.

Maimonides on God and Duns Scotus on Logic and Metaphysics

xi

Part 2 – Duns Scotus on Logic and Metaphysics John Duns Scotus began to lecture on the Sentences of Peter of Lombard at Oxford in 1298 and in the decade that preceded his death around the age of 42 in 1308, Scotus’s rich metaphysics, with its blend of neoplatonic and Aristotelian elements, produced a variety of conceptual instruments that were influential in carving out the fourteenth-century philosophical agenda. Scotus’s philosophical innovations are likewise apparent in thinkers such as Francisco Suárez, David Hume, Immanuel Kant, Charles Sanders Peirce and Martin Heidegger and contemporary philosophers and philosophical movements remain fruitfully engaged with his work.1 The array of Scotus’s developments includes a modal semantics of compossibility; the notion of an extra-sensory, intellectual, intuitive cognition of existence; the formal distinction between realities separable only in thought (as the generic and specific aspects of a concrete, extramental particular); ‘thisness’ or ‘haecceity’, the individuating principle by which one instance of a common nature such as humanity is distinct from another; a distinction, as regards the Aristotelian paradigm, between experiential and unqualified scientific knowledge; as well as the notion of the univocity of transcendentals and consequent belief that metaphysics is the science of natural theology. More than is evident in his later works, Scotus’s earliest writings (logical works, which likely date from 1290-1295)2 are influenced by the modistae (modists) or speculative grammarians, active in the thirteenth and early fourteenth centuries, who posited an ontological parallelism between modes of signification, understanding and being. Again, the early Scotus 1 For Kant and various contemporary projects inspired by a Scotist metaphysics, see Ludger Honnefelder, “Metaphysics as a Discipline: From the ‘Transcendental Philosophy of the Ancients’ to Kant’s Notion of Transcendental Philosophy,” In Russell Friedman and Lauge Nielsen, eds., The Medieval Heritage in Early Modern Metaphysics and Modal Theory, 1400-1700 (Kluwer, 2003); and Nathan Strunk, “Is the Doctrine of Transcendentals Viable Today?” . On David Hume, see Eileen Serene, “Demonstrative Science,” In Norman Kretzmann, Anthony Kenny and Jan Pinborg, eds., The Cambridge History of Later Medieval Philosophy: From the Rediscovery of Aristotle to the Disintegration of Scholasticism, 1100-1600. Cambridge: Cambridge University Press, 1982), 496-518. 2 See Edward Buckner and Jack Zupko, trans., Duns Scotus on Time and Existence: The Questions on Aristotle’s “De Interpretatione,” (The Catholic University of America Press, 2014).

xii

Introduction

belongs to the English tradition of the latter half of the thirteenth century that sets logic apart from metaphysics and physics, limiting the considerations of the former to various relations that hold between concepts of things (concepts of first intention), concepts of concepts (concepts of second intention) and terms, as opposed to the study of real links between real things, taken up in the latter. A feature of this tradition, evident in Scotus’s early works, is the belief that for the logician the notion ‘being’ is equivocal, whereas for the metaphysician it is analogous, a position at odds with Scotus’s aforementioned, mature consideration of our concepts of being and the other transcendentals.3 Yet despite such dissimilarities, Scotus’s early writings nevertheless foreshadow some of the positions that grow out of them (such as the notion of haecceity and the formal distinction) and are consequently of interest to those who would better understand the development and nature of Scotus’s thought. However, this effort has been hampered by the disarray in which Scotus’s writings were left owing to his untimely death. As regards Scotus’s logical works in particular (the parva logicalia or little logical works), recent scholarship has whittled the number that might actually have been written by Scotus down from seven to four, Scotus’s commentaries on the Isagoge (Qpor), Categories (QCat), De interpretatione (QPer I, QPer II) and De sophisticis elenchis (QSE). Owing to doctrinal inconsistencies pertaining to the aforementioned theory of univocity, the authenticity of QCat has been the subject of debate and, as the author of this commentary is unquestionably the author of QPor and QPer I and II, and, moreover, as the authenticity of QSE is dubitable, we may well ask whether Scotus wrote anything on logic at all. Edward Buckner’s “On the Authenticity of Scotus’s Logical Works” takes up this question. Drawing on Buckner and

3 See Giorgio Pini, “Univocity in Scouts’ Quaestiones super Metaphysicam: The Solution to a Riddle, in Medioevo 30, 2005, 69-110. Pini notes that it is because of his immersion in the English tradition that the early Scotus rejects “any inference concerning how things are from the mode of signifying of a term. Specifically, we cannot make any inference from the way in which the term ‘being’ signifies to the way things are in the world. In this respect, Scotus is an enemy of a logicosemantic investigation of reality. His belonging to the English tradition of distinguishing between a logical and a metaphysical approach to analogy clearly separates him from the Paris modist tradition, according to which there is some parallelism between modes of signifying, modes of understanding and modes of being” (86). How to resolve this tension between a rejection of modism on the part of Scotus, on the one hand, and the extent to which modist doctrine influences his thought (see below in “On the Authenticity of Scotus’s Logical Works”), on the other, seems to be unclear.

Maimonides on God and Duns Scotus on Logic and Metaphysics

xiii

Zupko’s recently published translation of QPer I and II,4 Buckner contends that “there is overwhelming evidence that three works – the Questions on Porphyry, on the Categories, and on the Perihermenias (both versions) are authentic. About the Questions on the Sophistical Refutations there is still doubt.” In “Scotus on the Species of Qualities” we turn to a discussion of QCat in particular, wherein Lloyd Newton presents Scotus’s attempt to render a principled account as to why Aristotle divides the category of quality into four species in Categories 8. The immediate division of a genus is into two species by means of a specific difference, as the genus animal divides into rational and irrational. Why then does Aristotle divide quality into habits and dispositions; abilities and inabilities; affections and affective qualities and, lastly, form and figure? Commentators appear split over whether an explanation is in order, with Simplicius, Aquinas and Scotus in agreement that such is the case. Yet, contends Newton, Scotus’s attempt “avoids the problems” that confront the accounts of Aquinas and Simplicius, thereby providing “a more coherent account of quality than had previously been offered.” With “Logic, Ontology, and the Psychology of Universals in Duns Scotus,” Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio shift the focus of this volume from logic proper to the intersection of logic and metaphysics with respect to Scotus’s doctrine of universals. Scotus is in the moderate realist tradition as regards universals, meaning that he accepts the Aristotelian assertion that there exist immanent principles or universals whereby an entity is fixed with respect to some natural kind, e.g., the principle of humanity. By contrast, Ockhamist nominalism would characterize these universals as affections of the intellect that are not universal except in their signification, i.e., in signifying many things that do not themselves share in any general nature (Summa logicae I.14). In what sense the universals of the moderate realist might be said to exist, how accurate are our notions of these principles and in what manner said notions are formed was a subject of heated debate in the medieval universities. By Scotus’s time it was generally accepted that universals are present to the human intellect by means of concepts of second intention (generic and specific concepts of concepts) that are derived from reflection on concepts of first intention (i.e., concepts of individual entities) and it is in this context that he develops his mature understanding of the univocity 4

Buckner and Zupko (2014).

xiv

Introduction

of the concept of being and other, transcendental notions used to describe any entity inasmuch as it exists.5 Yet modern commentators are (as Scotus’s immediate disciples were) divided over the correct interpretation of Scotus on the matter of univocal concepts and tend to bifurcate between moderate realist and near-nominalist accountings that lend more or less ontological heft to the univocal concept of a transcendental attribute, respectively.6 Accordingly, Scotus’s understanding of the nature of universals is of interest on several fronts; historically, as the immediate predecessor of Ockham’s nominalism; theologically, as a vehicle better to understand Scotus’s assertion that certain concepts are univocal to God and creatures; and, on its own account, as a sophisticated, moderate realist semantics. Gonzalez-Ayesta and Gonzalez-Ginocchio therefore attempt to shed light on Scotus’s understanding of universals and the process of abstraction through a study of: (1) the various senses of the term ‘universal’ deployed across Scotus’s writings that seeks to clarify the difference between Scotus’s logical and metaphysical universal, (2) the mode of being of the extramental universal and (3) the complete universal, viz., the concept of first intention that is predicable of many. Returning to the consideration of univocal and analogous concepts, Domenic D’Ettore examines the response provoked amongst Thomists by Scotus’s thesis that God and creatures are conceived univocally in our apprehension of the transcendentals in “The Semantic Unity of the Analogous Concept According to John Capreolus.” Scotus’s notion of univocity developed in reaction to Henry of Ghent’s theory of analogy, wherein concepts of God and creatures are analogous because, though they resemble one another, they are in fact distinct. For Scotus, this similarity demands an overlap of semantic content, such as the notion of being that mediates between (and thus for Scotus is univocal to) finite and uncreated being. Early Thomists saw here a threat to Aquinas’s belief that concepts of God and creatures are analogous in a manner that explicitly does not 5

The notion of transcendental concepts emerges from the philosophical tradition in the thought of Philip the Chancellor (d. 1236), who pondered in what sense the term ‘good’ can meaningfully be applied to both God and creatures (N. Wicki, ed., Philippi Cancellarii Summa de bono (Berne: Francke, 1985)), but the heart of the idea derives from Plato’s middle period theory of forms, which posits the extramental existence of eternal, immutable forms or ideas by means of participation in which an entity is a particular type of thing (as a triangle is triangular inasmuch as it approximates triangularity). 6 See Dumont, Stephen (1992). ‘Transcendental Being: Scotus and Scotists’. Topoi 11/2: 135-48.

Maimonides on God and Duns Scotus on Logic and Metaphysics

xv

hang on univocal semantic content and hence denied any common ratio or signification between the concept of a creaturely analog of a divine attribute and the corresponding concept whereby we grasp that attribute as an attribute of the divine essence. Capreolus, by contrast, admits an overlap of semantic content between the different notions whereby we conceive God on the one hand and creatures on the other, whereas he denies that the overlap is univocal, because he believes that the conjunction of concepts requisite to forming a complex notion of God (e.g., infinite and goodness) alters the semantic content of its elements such that they can thereafter no longer pick out the creatures from reflection on which they were drawn. Our final chapter, “Can We Speak about That Which Is Not?” by Lukáš Novák, places Scotus squarely into a dialog with twentieth-century analytic philosophers over the ontological status of what is merely possible. Metaphysical possibilists hold that in some irreducible sense merely possible entities must factor into our inventory of things that are, whereas metaphysical actualists deny this. Both parties subscribe to what Novák terms the Principle of Reference: (PR) It is impossible to refer to that which is not.

For the metaphysical actualists in the Frege-Russellian tradition, PR entails a semantic actualism wherein we can’t really talk about merely possible entities; hence when fans of Tolkien and Martin discuss Smaug and his counterparts out of Essos, semantic actualists are forced to construe their discourse in some manner that does not involve reference to non-existents. By contrast, Alexius Meinong and others (such as Henry of Ghent and Scotus’s immediate disciple Francis of Meyronnes) are driven by semantic possibilism, which accommodates our pre-theoretical intuition that we do speak of merely possible entities, to adopt metaphysical possibilism in order to supply the subject of this discourse. Scotus sidesteps the dilemma by embracing both semantic possibilism and metaphysical actualism, leaving him free to discuss utterly non-existent entities, grounding this account on his understanding that intelligibility is granted to non-existents “in virtue of their being conceived, prior to creation, by the absolute divine intellect.”

PART 1: MAIMONIDES ON GOD

CREATION AND THE ARGUMENT FROM PARTICULARITY KENNETH SEESKIN

Outwardly Maimonides and Aquinas take similar positions with respect to the creation of the world de novo.1 Aristotle and his medieval followers denied that such a creation is possible. Both Maimonides and Aquinas respond by saying that the arguments against creation are not persuasive because they rest on an unsupportable assumption: that the causes and principles that explain the world as it is at present can also explain its genesis or creation. At present, change from one condition, say rest, to another, motion, requires an agent that is already in motion and can actualize the potential of the subject to move. As applied to creation, this assumption would require us to say that the first motion must receive a push from something prior to it; in effect that there must be something in motion before the first motion. Because this is absurd, the idea of a first motion or creation de novo is incoherent. Both thinkers respond by saying that there is no reason to suppose that creation is analogous to change. As noted above, change involves the transition from one state of existence to another; by contrast, creation involves the coming into existence of something that did not exist before. At Guide of the Perplexed 2.17, Maimonides argues that even in nature as we now observe it, the birth of a creature often follows a different pattern from that which obtains in maturity. This amounts to saying that extrapolation backwards from what we know about something now to what happened at the moment it came to be is fraught with peril. We can 1

I distinguish creation de novo, which holds that there is a first moment in time (i.e. that the age of the universe is finite) from creation ex nihilo, which holds that no material cause or underlying substratum was involved in creation. For an excellent discussion of this distinction, see William Dunphy, “Maimonides and Aquinas on Creation: A Critique of their Historians,” in Graceful Reason, edited by Lloyd Gerson. Toronto: Pontifical Institute of Medieval Studies, 1983, 361-79.

4

Creation and the Argument from Particularity

see this in modern terms by recognizing that according to the prevailing view of cosmologists, extrapolation backwards leads to a singularity, which is to say a point of infinite temperature and pressure from which the universe that we observe originated. By definition, a singularity is unique. Why should we believe that the laws of physics which apply under finite conditions and assume already existing structures of space and time also apply at the moment of the Big Bang?2 To return to the medievals, once the analogy between creation and change is questioned, the traditional argument against a creation de novo loses its force, giving us reason to think that such a creation is at least possible. Both thinkers also agree that while creation de novo is possible, one cannot demonstrate that it is true. But it is here that Maimonides and Aquinas part company, for while Aquinas goes on to say (ST 1.46.2) that the lack of a demonstration means that creation de novo must be accepted “on faith alone,” Maimonides takes a somewhat different path. I say somewhat because at one point (GP 2.16), he says that because the issue of creation and eternity cannot be solved by demonstration, one should accept creation on prophetic authority. This seems to commit Maimonides to the same position as Aquinas. I believe however that the context reveals that Maimonides’ remark is qualified: one should accept creation on prophetic authority if one cannot follow the technical arguments for creation that he is about to present. While Maimonides’ arguments do not constitute a demonstration in the strict sense of the term, he tells us that they prevail over the opposition (ibid.), and at another point, that they comes as close to being a 2

Cf. Alan Guth, The Inflationary Universe, Reading, Mass.: Addison-Wesley, 1997, 86-7: “If one continues the extrapolation backwards in time, one comes to a point of infinite density, infinite pressure, and infinite temperature -- the instant of the big bang explosion itself, the time that in the laconic language of cosmologists is usually called “t = 0.” It is also frequently called a singularity, a mathematical word that refers to the infinite values of the density, pressure, and temperature. It is often said – in both popular-level books and in textbooks – that this singularity marks the beginning of the universe, the beginning of time itself. Perhaps this is so, but any honest cosmologist would admit that our knowledge here is very shaky. The extrapolation to arbitrarily high temperatures takes us far beyond the physics that we understand, so there is no good reason to trust it. The true history of the universe, going back to “t = 0,” remains a mystery that we are probably still far from unraveling.”

Kenneth Seeskin

5

demonstration as anything could (GP 2.19). I take this to mean that while we cannot be certain of creation de novo, based on the available evidence, Maimonides thought it is very likely that it is true. Though Aquinas cites Maimonides’ argument for creation de novo in the De Potentia Dei (3.17) and in connection with providence in his commentary on Job, it is noteworthy that he does not employ it in the Summa Theologica, where, to repeat, creation de novo is to be accepted on faith alone. This paper raises the question of why Maimonides invoked it while Aquinas did not.

1. Maimonides’ argument for creation de novo is derived from the Mutakallimnjn and has come to be known as the Argument from Particularity.3 Briefly stated, the argument asks: “Why this rather than that?” If the scientific account of the universe is correct, then there ought to be a sufficient reason for everything we observe telling us why it cannot be otherwise. According to Aristotle, we know a thing most fully when we know what it is, why it is what it is, and why it cannot be otherwise than what it is. If this is true, why does Venus move this way and Mercury that way? Why is the orbit of the sun tilted 23 ½ degrees to the plane of the earth rather than, say, 24 ½ degrees? Why does this quadrant of the night sky contain a large number of stars while that quadrant contains comparatively few? The demand for a sufficient reason could be met in either of two ways: (1) by giving the efficient cause of the phenomenon in question or (2) by showing what purpose it serves. Suppose we ask what purpose is served by having Venus and Mercury move in opposite directions? Unless one can give a convincing explanation, proponents of the argument conclude that the only way we can understand their movement is to say that God willed it so. According to the argument’s most famous proponent Alghazali, the will just is a faculty of deciding between similar or 3

For historical background to the argument, see Harry Wolfson, The Philosophy of the KalƗm (Cambridge: Harvard University Press, 1976), 434-52, who traces it to the Asharite theologian al-Juwayni. Its best-known version is that of Alghazali, The Incoherence of the Philosophers, translated by Michael E. Marmura (Provo, Utah: Brigham Young University Press, 1997), 12-27. For a modern discussion of the consequences of the argument, see Seymour Feldman, “’In the Beginning God Created’: A Philosophical Midrash,” in God and Creation, edited by David Burrell and Bernard McGinn (Notre Dame: University of Notre Dame Press, 1990), 3-26.

6

Creation and the Argument from Particularity

apparently indistinguishable alternatives.4 As the argument goes on to say, God made any number of decisions in creating the universe that could have gone the other way. If God exercises freedom of choice, then there is every reason to think that God could decide to create the world at a certain point, which is exactly what creation de novo asserts. At Guide 1.73, Maimonides criticizes the argument for two reasons. His first objection is the degree to which it relies on the imagination. According to the Mutakallimnjn, anything that can be imagined is possible. This assumption led them to ask why the sun is circular rather than triangular, why the earth is not completely under water, or why a flea cannot be as large as an elephant. Maimonides objects that the imagination is a poor guide to possibility so that instead of asking what we can imagine, we should be asking what we can think. That the earth should be completely under water is a manifest absurdity given the principles of Aristotelian physics, so that while it might be possible to imagine such a situation, to ask why it is not so establishes nothing. His second objection involves the extremes to which the argument was often taken, the most obvious being its association with metaphysical atomism. According to that view, every atom is qualitatively indistinguishable from every other. Why, then, does this atom occupy one region of space and another atom occupy another? Because no scientific reason could be given, the answer was again referred to the will of God. The effect of this argument was to say that God’s will is responsible for the placement of every atom and is therefore the sole cause behind every natural phenomenon. In Maimonides’ words (GP 1.73): “They [the Mutakallimnjn] assert that when a man moves a pen, it is not the man who moves it; for the motion occurring in the pen is an accident created by God in the pen.”5 Putting both of these objections together, we can say that unless severe limits are placed on its application, the Argument from Particularity runs the risk of undermining all scientific reasoning about the universe. Although it is clear why Maimonides would not accept this argument as a demonstration of creation, he felt free to invoke it given the lesser 4

Incoherence, 22-23. All quotations from the Guide of the Perplexed are taken from the Pines translation, Chicago: University of Chicago Press, 1963.

5

Kenneth Seeskin

7

standards of rigor he introduces in his discussion of creation in Book Two of the Guide. If the argument leads to silly results when the imagination was allowed to run wild, it raises serious questions when applied to the motion of the planets. It is well known that by the twelfth century, medieval astronomy had reached an impasse. According to the standard view, the universe consisted of nine primary spheres accounting for the motion of the fixed stars, the planets observable by the naked eye, the sun, and the moon. Each sphere was set in motion by a heavenly intelligence focused on the eternal perfection of God. In such a scheme, common sense dictated that the outer spheres should transfer motion to the inner spheres so all the spheres would rotate in the same direction with the outer spheres moving faster. But experience proves that this is not the case. As Maimonides observes (GP 2.19): We see that in case of some spheres, the swifter of motion is above the slower; that in the case of others, the slower of motion is above the swifter; and that, again in another case, the motions of the spheres are of equal velocity though one be above the other. There are also other very grave matters if regarded from the point of view that these things are as they are in virtue of necessity.

In addition to the different velocities of spheres, there is also the question of their different directions. Why does it often appear that one sphere moves in the opposite direction of the one directly above or below it? To account for this phenomenon, Aristotle introduced the notion of a secondary sphere. The need for such spheres can be explained if we recognize that the motion of every planet is unique to it and that from our perspective on earth, the planets often seem to reverse their direction. So a host of secondary spheres would be needed to explain why Saturn’s motion is very different from that of the fixed stars. But the spheres introduced to explain Saturn’s motion would all have to be reversed when we get to Jupiter, at which point we would have to introduce new spheres to explain its motion and reverse them when we get to Mars.6 In all 55 spheres were needed to make even a rough approximation to what we observe with the naked eye. Reverse motion is all the more difficult to 6

Metaphysics 1073b38-1074a5. For further comment, see Edward Grant, Planets, Stars, and Orbs: The Medieval Cosmos, 1200-1687 (Cambridge: Cambridge University Press, 1994), 275-77 as well as D. R. Dicks, Early Greek Astronomy to Aristotle (Ithaca: Cornell University Press, 1970), 199-201.

8

Creation and the Argument from Particularity

explain given the belief that there is no space or vacuum between one sphere and another. In effect, they are like the layers of an onion. Why, then, do the various layers seem to move at different speeds and in opposite directions for no good reason? Finally, there is the question of why the fixed stars do not exhibit retrograde motion, the planets do, but the sun and moon, which according to Aristotle are closest to the earth, also do not. From the standpoint of empirical accuracy, Ptolemy’s theory was more accurate than Aristotle’s. But Ptolemy’s accuracy could only be maintained at the cost of introducing epicycles and eccentric orbits, both of which contradicted Aristotle’s account of natural motion by introducing an orbit whose center of rotation is outside the center of the earth. To this Maimonides objected. “How,” he asks (GP II.24, 326), “can one imagine a rolling motion in the heavens or a motion around a center that is not immobile?” The answer is that one cannot, from which it follows that epicycles and eccentrics are (GP 2.24, 322): “outside the bounds of reasoning and opposed to all that has been made clear in natural science.”7 What do you do when a theory with great predictive power flies in the face of a basic scientific principle? Maimonides calls this “true perplexity” and maintains (GP 2.24) that astronomy will never resolve it and attain the status of a true science. Citing Psalm 115.16 (“The heavens are the heavens of the Lord, but the earth hath He given to the sons of man”), he argues that only God knows the nature, substance, motions, and causes of the heavenly bodies and that they are too far away and too high in place or rank for us to agree on assumptions from which conclusions can be drawn. In the same chapter, he goes so far as to say that the purpose of astronomy is not to uncover the truth of the phenomena it investigates but merely to provide a mathematical model from which accurate predictions can be made. In his words: For his [the astronomer’s] purpose is not to tell us in which way the spheres truly are, but to posit an astronomical system in which it would be possible for the motions to be circular and uniform and to correspond to 7

For the historical background to Maimonides’ rejection of Ptolemy, see A. I. Sabra, “The Andalusian Revolt against Ptolemaic Astronomy,” in Everett Mendelson, ed., Transformation and Tradition in the Sciences (Cambridge: Cambridge University Press, 1984), 133-53. Note that Aquinas rejects epicycles for the same reason. See Commentary on the Metaphysics, Vol. 2, 12.10, 904.

Kenneth Seeskin

9

what is apprehended through sight, regardless of whether or not things are thus in fact.

This is another way of saying that all we can expect from astronomy is a mathematical model that saves the phenomena and preserves the theory of natural motion. With his usual respect for Aristotle, Maimonides argues (GP 2.22) that he himself realized that his account of the heavens was weak and makes reference to “grave incongruities and perversities that manifestly and clearly appear as such to all the nations.” As Maimonides recognizes, there is always the possibility that someone will come up with a deep level explanation of phenomena that seemed puzzling to him (2.24). We should keep in mind however that he says this in a context that assumes that at least part of the Aristotelian theory is true. In Maimonides’ defense, it is worth pointing out that no one ever succeeded in doing this. I should point out that while Copernicus’ theory caused a great stir when it was published, it did not provoke an official condemnation from the Church because an introduction not written by Copernicus himself was added that said that the theory was only a mathematical model. Part of what caused the Church to condemn Galileo was his belief that Copernicanism was more than a mathematical model but a statement of truth. We are now in a position to understand why Maimonides thought that the world we inhabit does not present itself as a system governed by strict causal necessity but as a system in which, for all we know, some things could have been different from what they are. If things could have been different, then God must have exercised free choice in creating them the way he did. If God exercised free choice, then while creation de novo has not been demonstrated, it appears to be the most reasonable option available to us. Critics of the argument objected that just because astronomy cannot come up with a satisfactory explanation of planetary motion at present, it does not follow that it will never be able to do so. In fact, some went so far as to say that Maimonides’ argument is so weak it is hard to believe he actually held it.8 To the degree that the critics’ objection relies on what science 8

See for example Moses of Narbonne, Commentary on the Guide of the Perplexed, ed. J. Goldenthal (Vienna, n.p., 1852), 2.19. It is worth noting that many Jewish

10

Creation and the Argument from Particularity

might be able to do in the future, it is impossible to refute. We saw however that while Maimonides himself recognized this possibility, he was skeptical that it would ever be realized. When push comes to shove, all we can do is formulate our conclusions on the basis of the evidence currently available to us. In Maimonides’ opinion, that evidence suggests that astronomy will never be able to give a proper explanation of planetary orbits.

2. While it is true that a proper explanation was eventually found, it went well beyond anything Maimonides or his contemporaries could have imagined. Even so, we may ask whether our predicament is all that different from theirs. Suppose – what may well be false – that one could subsume all observable phenomenon under a single set of scientific laws. Even on this scenario, questions similar to those raised by the Mutakallimnjn would continue to arise because one can always ask why one set of laws obtains rather than another. To take an obvious example, I can understand the ebb and flow of the tides by subsuming the relevant facts about the earth and the moon under Newton’s law of gravitation. The problem is that as we attain higher levels of generality, we will eventually reach something so general that it will be difficult, if not impossible, to subsume it under anything else. If there is nothing more general to which thinkers became enamored of Averroes after Maimonides’ death; Moses Narboni is just one example. For a modern version of Narboni’s criticism, see Isaac Husik, A History of Medieval Jewish Philosophy (1916; rpt. New York: Antheneum), 275. There is a centuries old debate on whether Maimonides’ defense of creation and contingency is to be taken at face value or whether it is written to disguise his real position, which is an eternal world governed by necessity. For a modern defense of the latter view, see Warren Zev Harvey, “A Third Approach to Maimonides’ Cosmology Prophetology Puzzle.” Harvard Theological Review 74 (1981): 287301. Needless to say, if Harvey is right, Maimonides’ position would be much closer to Spinoza’s. For an extended defense of my reading of Maimonides, see Seeskin, Maimonides on the Origin of the World, New York: Cambridge University Press, 2005. For approaches similar to my own, see Harry Wolfson, Studies in the History and Philosophy of Religion, I. Twersky and G. H. Williams eds. (Cambridge: Harvard University Press, 1973), 207-21, Arthur Hyman, “Maimonides on Creation and Emanation,” Studies in Medieval Philosophy, J. F. Whipple, ed. (Washington, D.C.: Catholic University of America Press, 1988), 4561, and Seymour Feldman, “Abravanel on Maimonides’ Critique of the KalƗm,” Maimonidean Studies 1 (1990): 5-25.

Kenneth Seeskin

11

we can appeal, we will be left with a simple statement of fact. When that happens, the question “Why does the force of gravity take this value rather than that?” will make sense but will have no readily available answer. Let us therefore extend our example and ask why the force of gravity – or the dark energy that repels it – takes the precise value it does. The question is apt because there is no apparent contradiction in imagining that it might be greater or lesser than it is. The problem is that as the idea of an expanding universe began to take hold, it became clear that the conditions needed to produce a universe like ours had to be highly specific to allow for the formation of stars and thus be compatible with the formation of life. In 2002 Stephen Hawking wrote:9 If the rate of expansion one second after the big bang had been smaller by even one part in a hundred thousand million million, the universe would have recollapsed before it ever reached its present size. On the other hand, if the expansion rate at one second had been larger by the same amount, the universe would have expanded so much that it would be effectively empty now.

Even if the rate of expansion were set to exactly the right point, other conditions would have to be met for human life to be possible. In 2010 Hawking spoke more generally when he wrote:10 Most of the fundamental constants in our theories appear fine-tuned in the sense that if they were altered by only modest amounts, the universe would be qualitatively different, and in many cases unsuitable for the development of life . . . The emergence of the complex structures capable of supporting intelligent observers seems to be very fragile . . . Were it not for a series of startling coincidences in the precise details of physical law, it seems, humans and similar life-forms would never have come into being.

It is clear that “finely tuned” is just another name for particularity and that the “startling coincidences” that Hawking mentions are of an order of magnitude most people can scarcely imagine. 9

Stephen Hawking, The Theory of Everything. Beverly Hill, Ca.: The New Millennium Press, 2004, 104. 10 Stephen Hawking, The Grand Design (New York: Bantam Books, 2010), pp. 160-61.

12

Creation and the Argument from Particularity

Like the medievals, we face the question of how far to press the need for a scientific explanation. Are there deep level explanations for the masses of elementary particles and the strength of cosmic forces or is it that a benevolent God fashioned the universe with the intention of allowing life to develop? There are always some who hold out for the possibility of a deep explanation. In an autobiographical statement published in 1949, Einstein expressed the faith that that there are no arbitrary constants but only strongly determined laws from which one can derive completely determined constants.11 As Maimonides said with respect to medieval astronomy, it is possible that some day someone will find these kinds of laws, but to date, no one has. Again from Hawking: “Not only does God definitely play dice, but He sometimes confuses us by throwing them where they can't be seen.” As long as arbitrary constants are with us, the Argument from Particularity has force. To be sure, the existence of such constants does not constitute a demonstration that God exists, that God exercises freedom of choice, or that the universe might have been different from what we now observe. But, to follow in Maimonides’ footsteps, such constants give us reason to think that all of these things might be true.

3. That brings us to the question of why Aquinas did not avail himself of this argument. It does not contradict anything he said and may even have lent force to what he said. One could hold, for example, that creation de novo is both an article of faith and that it is a reasonable conclusion based on the scientific evidence at our disposal. I suggest there are two ways of answering this question. To understand the first, let us return to Maimonides. Although he did not think science would ever have a satisfactory explanation for planetary orbits, he did more than admit that in principle it might. After concluding his defense of creation de novo, he goes on to say (GP 2.25), that if it could provide such an explanation, and the case for causal necessity carried the day, he would have no choice but to reinterpret the Torah in light of the doctrine of eternity. His reasoning is simple. The Torah is a vehicle of truth. As such, our interpretation of it can never contradict well-established scientific principles. As he says in his 11 Albert Einstein, 1949. Philosopher-Scientist, Vol. 1, edited by Paul Arthur Schilpp. New York: Harper & Row: 1949, 63.

Kenneth Seeskin

13

Letter on Astrology: “A man should never cast his reason behind him, for the eyes are set in front, not in back.” In short, Maimonides subjected his reading of the Torah to the prevailing trends of scientific opinion. Although medieval science lacked adequate answers to the questions he asked, experience shows, however, that no sooner does science put old questions to rest that new ones begin to arise. At the beginning of the twentieth century, no less a figure than Einstein was committed to the idea of an eternal world. As the Big Bang theory began to take hold, new questions of the sort raised by Hawking came to the fore giving the Argument from Particularity a second life. Yet even now, the question of creation versus eternity remains open. Even if our universe has a temporal beginning, it may be the case that other universes preceded it and that the chain of previous universes stretches infinitely into the past. My first suggestion is that Aquinas did not want to make Christian dogma subject to the prevailing trends of scientific opinion. If something is a dogma, it has to be put out of reach of anything science might or might not come up with. For that reason, he does not employ the Argument from Particularity and is content merely to show that creation de novo is logically possible. The second suggestion has to do with epistemology. If demonstration cannot establish creation de novo, Aquinas had the option of saying that one should accept it on faith. Although Maimonides allows one to accept creation de novo on prophetic authority if one cannot follow the arguments pro and contra, there is no room in his philosophy for a form of awareness that perfects or supersedes knowledge. On the contrary, in the “Parable of the Palace,” which begins the rhetorical climax of the Guide, those who have achieved mastery of the sciences and are engaged in speculation are described as closer to God than those who accept religious doctrines on traditional authority (GP 3.51). Earlier on (GP 1. 32), he characterizes his position this way: For if you stay your progress because of a dubious point; if you do not deceive yourself into believing that there is a demonstration with regard to matters that have not been demonstrated; if you do not hasten to reject and categorically to pronounce false any assertions whose contradictories have not been demonstrated; if, finally you do not aspire to apprehend that

Creation and the Argument from Particularity

14

which you are unable to apprehend – you will have achieved human perfection and attained the rank of Rabbi Akiba . . .

Simply put: if you recognize the limits of reason and stay within them, you will achieve the same status as one of Israel’s greatest sages. So far from a recommendation to turn to faith, this sounds like a recommendation to proportion one’s belief to the weight of the evidence – recognizing that for some questions the evidence will never be decisive. For Aquinas, it is otherwise. As he says (ST 1.12.13): “We have a more perfect knowledge of God by grace than by natural reason.” We can understand this better by recognizing that the faith Aquinas is talking about involves assent to something to which the intellect would not assent if left to its own devices. On matters that natural reason can decide on its own such as the existence of God, faith is offered to those who do not have sufficient learning to follow the arguments. But on matters that natural reason cannot decide on its own such as creation de novo, it is offered to everyone. Faith, then, is a gift. As we learn from Ephesians 2:8 (cited at ST 2.2.6.1): “For by grace you have been saved through faith, and this is not your own doing; it is the gift of God . . .” It could be said, therefore, that faith perfects or surpasses knowledge in the same way that grace perfects or surpasses nature. Once it has been shown that creation de novo is logically possible, appeal to scientific evidence is not needed to generate conviction and if anything may interfere with it.

4. My purpose in writing this paper is not to argue that one thinker is right and the other wrong. Maimonides leaves us with less than total conviction on creation de novo. This puts him in what some would consider an embarrassing position. Because belief in creation is one of the foundations of the Law (GP 2.25), it follows that according to Maimonides, one of the foundations of the Law can only be known with presumptive certainty. I remarked earlier that from the time the Guide was published, there were scholars who questioned whether he was committed to creation de novo at all.12 Though Aquinas’ commitment to creation de novo is total, he has to invoke what he himself characterizes as a supernatural form of cognition to explain it. 12

See n. 9 above.

Kenneth Seeskin

15

Needless to say, Maimonides also believed in revelation, and praises the prophetic tradition for achieving greater clarity on creation than the philosophic. But if we were to pursue his understanding of prophecy, we would find that it is the perfection of the human intellect rather than a condition that allows one to supersede it. Strictly speaking, prophecy is an emanation from God to the rational faculty of the prophet, and then from the rational faculty to the imaginative (GP 2.36). At bottom, the prophet is a philosopher who, in addition to being able to reason abstractly, is also able to express his ideas in the form of parables or metaphors. The upshot is that even prophets have to deal with the limitations of the human intellect and have no way of getting beyond them. If this leads to a degree of uncertainty with respect to important doctrines, Maimonides’ answer is: So be it. In the Introduction to the Guide, he compares our epistemological situation to a dark night in which flashes of light appear momentarily and then recede. While prophets may be able to receive some of these flashes and communicate them in a way that we can understand, they cannot offer the kind of conviction that Aquinas sought in faith.

MAIMONIDES, AQUINAS, AND PARTICULARITY: COMMENTS ON SEESKIN’S “CREATION AND THE ARGUMENT FROM PARTICULARITY” STEPHEN R. OGDEN

I thank Professor Seeskin for this engaging paper and for the honor of interacting with his work. The paper offers lucid comparison between the three Abrahamic faiths (with special attention to Maimonides and Aquinas) and also timely reference to contemporary issues in science, philosophy, and theology. Specifically, Seeskin provides an intriguing examination of the argument from particularity for the creation of the world de novo, Maimonides’s use of it, and Aquinas’s apparent decision to refrain from employing it. Two major questions linger regarding this main thesis of the paper. First, why does Aquinas not embrace the particularity argument (at least not in major works after De Potentia Dei)? This question from Seeskin is of some interest, and it is not easy to know why Aquinas does not seem as sanguine about its prospects as Maimonides. I will venture one guess below. But even more pressing is the question, why does Maimonides embrace the particularity argument? Unfortunately, Seeskin does not explicitly answer this question in the paper (other than to show – quite compellingly – how the argument does have real bite).1 This is the pressing debate on interpreting Maimonides which Seeskin mentions at a couple of points in his paper. Here is how I would frame the “embarrassment.” Does 1

In a way, I think that is Seeskin’s ultimate answer in his more thorough and quite enlightening Maimonides on the Origin of the World (Cambridge: Cambridge University Press, 2005); see esp. pp. 3 and 185.

18

Maimonides, Aquinas, and Particularity

Maimonides really think that we should be in the habit of accepting “speculative” arguments rather than prophecy if we are presented with an option on an open question such as this? By the end of the paper, Maimonides is presented as a straightforward evidentialist: one should proportion one’s belief to the evidence. So why trust in merely dialectical arguments at all? If it is acceptable to do so, what is wrong with the merely dialectical arguments of the Mutakallimnjn quite generally (as Maimonides usually seems fairly dismissive of their methodology and erroneous “proofs”)?2 Of course, if the particularity argument truly “outweighs” or “preponderates” (tarjƯ‫)ۊ‬3 in favor of creation de novo, in what sense can Maimonides still call this an open question at all? Here is my own brief thought on the matter, and perhaps Seeskin would agree.4 We cannot escape the fact that for Maimonides, there is no apodictic demonstration (burhƗn) of either the eternity of the world or of creation.5 That is the sense in which this is an open question – no demonstration for either side, doubts attaching to each. Although the particularity argument does not seem to be in any different or more certain category than those of the Mutakallimnjn – the same term dalƯl is used for both6 – still it seems to have two benefits of which the arguments of the Mutakallimnjn cannot boast. First, Maimonides presents his argument, as he says, “without having to take upon myself what the Mutakallimnjn have undertaken – to abolish the nature of that which exists and to adopt atomism, the opinion according to which accidents are perpetually being 2

Guide, II.16, p. 293 (M 34a, A 316): “I approve of nothing in those proofs [of the Mutakallimnjn] and … I do not deceive myself by designating methods productive of errors as demonstrations….” N.B. All references to Maimonides’s Guide of the Perplexed give first the page number in the Pines translation (Chicago: University of Chicago Press, 1963), followed by that in the Munk-Joel edition (M) (DalƗlat al-‫ۉ‬ƗҴirƯn, Jerusalem: Janovitch, 1929), and then the Atay Arabic transliteration (A) (DalƗlat al-‫ۉ‬ƗҴirƯn, Cairo: Maktaba al-ThaqƗfa al-DƯnƯya, 1980). All English translations are those of Pines unless otherwise stated. 3 Ibid., (M 34b, A 316). Seeskin cites Pines’s translation (p. 294) which uses the term “prevail.” 4 I think my answer here is largely compatible with Seeskin’s considered exposition in Maimonides on the Origin of the World, op. cit.; see esp. pp. 3-4 and chs. 5 and 7. While above I emphasize the role of the particularity argument in casting doubt, Seeskin also at times describes its role in more positive terms, as providing “evidence” (albeit inconclusive evidence) for creation. 5 II.22, p. 320 (M 50a, A 343). 6 II.16, pp. 293-294 (M 34a-b, A 316).

Stephen R. Ogden

19

created, and all their principles….”7 A good argument from particularity can be advanced without needing to accept atomism or an occasionalistic doctrine of a particularizing God at every point in space and time. Seeskin sufficiently displays how this works for Maimonides who rejects atomism while acknowledging our lack of explanation at the level of astronomy. Second, although the particularity argument does not demonstrate creation, it does attach an insuperable doubt to the eternity of the world thesis. This is the sense in which the particularity argument can be seen as outweighing the doubts that attach to Maimonides’s own position, the position of faith.8 After explaining in II.22 the process of how to subscribe to one of two beliefs even when no demonstration is possible (viz., through choosing the one with the least doubts), he clarifies at the beginning of II.23 that in certain cases “a single doubt is more powerful than a thousand other doubts.”9 I think he means here the preponderating doubt raised by the one particularity argument. While I suspect that Maimonides and Aquinas may be closer here than Seeskin supposes, I do think he has highlighted some relevant differences in their epistemological frameworks. One guess as to why Aquinas does not put as much stock in the argument from particularity is that perhaps he is generally more reticent to rely on what might be shaky arguments, especially when addressing such a firmly held “scientific opinion” as the eternity of the world (more on this below). What he does say on this question is that: It is useful to consider this [i.e., the fact that, albeit indemonstrable, creation de novo is believed by faith], so that some, presuming to demonstrate what is by faith, may not put forward non-necessary reasons, which would give fodder for unbelievers to laugh, thinking that we believe because of such reasons those things which are actually held by faith.10

Maimonides himself similarly admonishes the reader of the Guide concerning the speculative arguments of the Mutakallimnjn: 7

II.19, p. 303 (M 40a, A 325). II.16, p. 294 (M 34b, A 316): “…just as a certain disgrace attaches to us because of the belief in the creation in time, an even greater disgrace attaches to the belief in eternity.” Cf. II.22, p. 320 (M 50a-b, A 343). 9 II.23, p. 321 (M 50b, A 343). 10 Aquinas, Summa Theologiae (Rome: Leonine ed., 1888), I.46.2, resp., p. 481; my translation. Cf. Summa Contra Gentiles, II.38. 8

20

Maimonides, Aquinas, and Particularity If a man claims that he sets out to demonstrate a certain point by means of sophistical arguments (bi-mughƗla‫ܒ‬Ɨt), he does not, in my opinion, strengthen assent to the point he intends to prove, but rather weakens it and opens the way for attacks against it . . . It is preferable that a point for which there is no demonstration remain a problem or that one of the two contradictory propositions simply be accepted.”11

Still, we should note that, although Aquinas says one should hold to creation de novo by faith alone in the Summa Theologiae, in a similar article of the Summa Contra Gentiles, he allows that various proofs for creation do “have probability,” and later grants that a particular consideration of the divine will is “more efficacious” for proving (ostendum) that God created de novo.12 Now we come to doubts I have about Professor Seeskin’s own answer to the question about Aquinas and his interpretation of Maimonides. Seeskin’s first response regarding Aquinas is that “he did not want to make Christian dogma subject to the prevailing trends of scientific opinion.” What were these trends for the likes of Maimonides and Aquinas? I myself am wary of reading this contemporary category back into the medieval milieu. “Scientific opinion” (including natural science and cosmology) was simply regarded as a part of philosophy. Therefore, I don’t think Maimonides can be as laissez-faire as Seeskin presents him (more on that below), nor can Aquinas be as insular. Both thinkers sense the urgency of (and thus complete) a refutation of the opinion that the world’s eternity is demonstrable. In turn, they both want to show that the position of faith is not impossible. For Aquinas as much as Maimonides, “scientific opinion” had to be grappled with precisely because the truths of faith cannot afford to ignore it. As I said above in my own take on Aquinas, Seeskin’s second suggestion regarding their differing epistemologies seems more likely. However, I want to point out that any talk of faith as “higher than” or “superseding” knowledge in Aquinas needs to be nuanced. In the article Seeskin cites (ST I.12.13), Aquinas clarifies that faith as cognitio is deficient compared with true scientia, because we have no vision in the former. Faith is only 11

Maimonides, Guide, II.16, p. 293 (M 34a, A 316); Arabic inserted. Aquinas, Summa Contra Gentiles (Rome: Leonine ed., 1918), II.38, pp. 355 and 356.

12

Stephen R. Ogden

21

“higher” because of its object.13 With this caveat, I agree with Seeskin’s assessment that Aquinas and Maimonides probably approach religious knowledge differently. And perhaps this is part of the reason why Aquinas fails to mention the particularity argument as performing some greater probative role for the believer than the teachings of revelation. That brings me to my final worries over Seeskin’s interpretations of Maimonides. It is true that Maimonides seems to think the particularity argument can make some rational headway in support of creation such that it at least provides an additional supplement to what is known on the basis of prophecy. Still, I cannot find him in II.16 (or elsewhere) stating that we should accept creation from prophecy only if we cannot follow the speculative argument. On the contrary, Maimonides urges: …since this question is open – I mean the pre-eternity of the world or its creation de novo – [the answer] should be accepted on the basis of prophecy which makes things clear that are not in the power of philosophical speculation (quwa al-naܲar) to achieve.14

Furthermore, even when discussing the argument from particularity and its role of casting doubt on the eternity of the world, he seems to say that the latter is also to be rejected because of the harm it brings to faith, all while affirming again that creation is to be accepted on the basis of prophecy: …the opinion favoring the eternity of the world is the one that raises more doubts and is more harmful for the belief that ought to be held with regard to the deity. And this, in addition to the fact that the world’s being produced in time is the opinion of Abraham our father and our prophet Moses, may peace be on both of them.15

The particularity argument, then, has a role to play in casting doubt on the eternity of the world, but it seems to serve only as a supplement to a decision made additionally on the basis of revelation. It is not presented as the first line of defense.

13

Cf. Aquinas, Summa Theologiae, II-II.4.8. Maimonides, Guide, II.16 (M 34b, A 316); my translation. If anything, Pines’s translation (p. 294) only makes this point more forcefully, but does so by inserting “without proof,” which is not in the text. 15 II.22, p. 320 (M 50a-b, A 343). 14

22

Maimonides, Aquinas, and Particularity

Finally, Seeskin argues that Maimonides (especially in II.25) is open to “causal necessity” and the possibility that the eternity of the world might be proven,16 in which case he would simply interpret Scripture allegorically. However, it seems to me that though Maimonides is thus open to a Platonic Timaeus schema of eternity, he categorically denies that he could ever accept the “Aristotelian” causal necessity view. The latter “destroys the Law in its principle and necessarily gives the lie to every miracle, and reduces to inanity all the hopes and threats that the Law has held out….”17 Maimonides summarizes the gravity of this question at the end of the chapter: “…if the philosophers would succeed in demonstrating eternity as Aristotle understands it, the Law as a whole would become void, and a shift to other opinions would take place. I have thus explained to you that everything is bound up with this problem.”18 The shift of opinions envisioned is not simply a new interpretation of the Law, but rather an abandonment of the Law as such—which is precisely why Maimonides attaches such importance to this issue in his Guide.19

16

This specific claim seems undermined by Maimonides’s confidence in the lack of such a demonstration in II.23, p. 322 (M 51a, A 344). 17 II.25, p. 328 (M 55a, A 350). 18 Ibid., p. 330 (M 55b, A 352). 19 Indeed, second only to monotheism (God’s oneness, unity, and uniqueness), creation de novo is considered by Maimonides to be one of the primary bases or pillars of the Law – see, e.g., GP II.13, p. 282 (M 28a, A 306). Seeskin himself emphasizes this point in Maimonides on the Origin of the World, op. cit., p. 7, and lists numerous other places where Maimonides calls it such.

THE SIXTH METHOD DOESN’T WORK, BUT THE THIRD WAY DOES?: MOSES MAIMONIDES’ ASSESSMENT OF TWO PROOFS BASED ON POSSIBILITY AND NECESSITY JAMIE ANNE SPIERING

Introduction In The Guide of the Perplexed, Moses Maimonides summarizes various proofs for God’s existence, contrasting the methods used by the KalƗm thinkers, on the one hand, and the philosophers, on the other. One of the KalƗm proofs runs like this: The world is possible: it could exist, or it could not exist, and both alternatives are equally valid. Since, however, it does exist, we need something to explain why there is existence, rather than non-existence. What tipped the balance? It can only be God who gave preference to existence over non-existence. Moses harshly criticizes this proof. He says that the author leads us into error and that the proof is based on “very pronounced vain fantasy.” Later, however, when Maimonides is summarizing philosophical proofs, he presents the reader with another argument involving possibility, claiming that all things other than God are only possible beings, and that a necessary being is required in order for those possible beings to exist. This proof, unlike the earlier argument from possibility, calls forth high praise from Maimonides: “This is a demonstration concerning which there can be no doubt, no refutation, and no dispute, except on the part of one who is ignorant of the method of demonstration.”1 In this paper, I will discuss precisely why Maimonides considers the philosophical demonstration (which is quite similar to 1

Moses Maimonides, The Guide of the Perplexed, II.1, trans. Schlomo Pines (Chicago: University of Chicago Press, 1963) 248.

24

The Sixth Method Doesn’t Work, But the Third Way Does?

Thomas Aquinas’s Third Way) so superior to its cousin. Both proofs argue that God must exist to provide an answer to the question, “Why is there something rather than nothing?” Why, then, are they treated so differently? I will argue that the surface reasons behind Maimonides’ preference for the longer demonstration point to deeper, very crucial differences between the proofs. Maimonides’ comparison sheds light on the fundamental nature of the project of proving God’s existence, and his criticisms may also be helpful in evaluating some recent “new proofs” for the existence of God. I will mention some of these applications of Moses’ ideas briefly at the end of this paper.

A. A Few Words About Reading the “Guide” To begin, I should say a few words regarding the challenge of writing about The Guide of the Perplexed. Maimonides specifically warns his reader to expect contradictions in his work;2 he assures us that he has chosen his wording and organized his passages in such a way as to partially obscure his meaning, only revealing it in flashes;3 he writes on purpose in order to be impenetrable to all but the most searching and careful student. Moreover, he begs his reader, in God’s name, not to explain or interpret him as offering any kind of original or distinctive teaching.4 Who would presume that he could follow the real thread of Maimonides’ teaching? And even if he could, how could he ignore his teacher’s plea not to interpret him? There are further reasons why it is rash for me, in particular, to write about Maimonides: he wrote for Jews, and I am a Christian; he made careful, subtle use of Arabic and Hebrew terms, and I have no knowledge of either language. Lastly, I was trained as a Thomist, and so I can hardly avoid reading through a lens that Maimonides had no way of anticipating, and, worse yet, I habitually expect a relatively clear and open style. It seems there is no way for me (or really, for anyone) to avoid being a naïve and foolish reader of Maimonides.

2

Guide I, introduction, 17-20. “My purpose is that the truths be glimpsed and then again be concealed.” Ibid., 6-7. 4 “I adjure – by God, may he be exalted! – every reader of this Treatise of mine not to comment on a single world of it and not to explain to another anything in it save that which has been explained and commented upon the words of the famous Sages of our Law who preceded me.” Ibid., 15. 3

Jamie Anne Spiering

25

Yet Maimonides himself provides some comfort for anyone undertaking the foolhardy project of discussing his text. First, he sets us an example in rashness: the project of commenting on Scripture is, as he sees it, terrifically presumptuous and he hardly dares to speak of the concealed things of God. Yet, as Maimonides explains, it is God who desires us to come to know him better – seeking for truth about him is service to heaven. If that means interpreting scripture, Maimonides is willing to take this step. Similarly, if this means trying to interpret Maimonides, we should take this step as well. A second source of comfort is that Maimonides seems to anticipate and plan for naïve readers, such as myself. He writes that “every beginner will derive benefit from some of the chapters of this treatise, though he lacks even an inkling of what is involved in speculation.”5 He also gives a piece of advice to his readers that is especially helpful, I think, for the naïve: All into whose hands it falls should consider it well; and if it slakes his thirst, though it be on only one point from among the many that are abstruse, he should thank God and be content with what he has understood…. If anything in it, according to his way of thinking, appears in some way harmful, he should interpret it, even in a far-fetched way, in order to pass a favorable judgment.6

Maimonides asks for serious and careful reading, but he also encourages the reader to make use of the Guide as an offering from a friend. Anything that seems harmful should be given a kindly reading; only what is found good should be kept. Most importantly, what we cannot understand in the treatise should not stop us from thanking God for what we can understand. I will take this passage as my guide in reading his treatise.

B. Introduction to the Two Proofs The parallel proofs this paper will discuss come from the middle section of the Guide, in which Maimonides turns his attention to proofs for the existence of God. Through careful summary, he places before the reader a fundamental contrast between the proofs used by the Mutakallimnjn7 and

5

Ibid., 16. Ibid., 15. 7 Those who want to know exactly which thinkers are referred to as “Mutakallimnjn” or who are curious about how accurately Moses represented them, 6

26

The Sixth Method Doesn’t Work, But the Third Way Does?

those used by the Aristotelian philosophers. The Mutakallimnjn, he says, pin all their hopes on demonstrating that the world has been created in time – they then conclude, based on this, that God exists.8 The philosophers accept the eternity of the world, and use premises drawn from the assumption that nature is stable and unchanging to prove God’s existence, unity, and incorporeality.9 Maimonides does not withhold his own opinion of these arguments from the reader: he says that the proofs of the Mutakallimnjn are weak10 and the proofs of the philosophers are perfect demonstrations.11 Two of the eleven proofs Maimonides discusses are parallel in the sense that they both use the idea of possibility: the sixth method of the Mutakallimnjn and the third speculation of the philosophers. The sixth method is fairly easy to summarize: the world is not a necessary being – otherwise, it would be God. As it is, the world is merely possible: we need something, then, to explain why a possible world really does exist. That something is God. What’s so bad about this proof? Maimonides’ criticism has three elements, but I’ll save them to discuss later. One thing I should add right now, however, is that an addendum is probably required for this proof in order to answer a possible objection. Why can’t we say that the God whose existence is demonstrated in this proof might also either exist or not exist, and thus we will require yet another cause to explain the first one? The answer can only be that, to avoid an infinite regress, we posit that God is a

would do well to begin with Michael Schwarz, “Who Were Maimonides’ Mutakallimnjn?” (Maimonidean Studies 2 (Jan. 1992) 159-209). 8 Guide I.71, 179-180. 9 Ibid., 181, 183. 10 “Every one of their premises, with few exceptions, is contradicted by what is perceived of the nature of that which exists.” (Ibid.,183) “They have not demonstrated the creation of the world in time and have destroyed for us the demonstrations of the existence and oneness of the deity and of the negation of His corporeality.” (Guide I.76, 230-231.) 11 “Through this method [the method of the philosophers] the demonstration becomes valid and perfect certainty is obtained with regard to those three things: I mean the existence of the deity, His oneness, and His not being a body.” (Ibid., 181.)

Jamie Anne Spiering

27

being whose existence is necessary rather than possible. Maimonides both suggests and grants this addendum to proponents of the sixth method.12 Let’s move on, now, to consider the third “philosophic speculation,” which also uses the idea of possibility and concludes to a necessary being.13 This proof is considerably more complex than the Mutakallimnjn’s method and makes greater demands on its audience. (The only consolation I have to offer is that those who are acquainted with Thomas Aquinas’s third way will find themselves on very familiar ground.) The first premise in this proof is that things exist, or, as Maimonides puts it, “There is no doubt that there are existent things; these are the existent things apprehended by the senses.”14 Maimonides then asserts a threefold disjunction: Either a)

no beings are subject to generation and corruption

b)

all beings are subject to generation and corruption,

c)

or some are and some are not.

It is obvious, Maimonides says, that (a) is not the case, since we see plenty of beings that can begin and end. For instance, I have observed my houseplants die and new weeds arise. What about (b), though? Could all beings be things that begin and end? Maimonides puts this option out of the running by using the following argumentation: If every existent falls under generation and corruption, then all the existents and every one of them have a possibility of undergoing corruption. Now it is indubitable, as you know, that what is possible with regard to a species must necessarily come about. Thus it follows necessarily that they, I mean all existents, will necessarily undergo 12

Guide I.76, 230. Although the proof as presented here has some common elements with arguments made by Avicenna and Averroes, it contains important differences from both, as Josef Stern points out. [Stern, “Maimonides’ Demonstrations: Principles and Practice,” Medieval Philosophy and Theology 10 (2001): 47-84, at 64-77.] It seems that the argument as presented here was structured carefully by Maimonides himself, despite its use of elements from both Aristotle and his commentators. 14 Guide II.1, 247. 13

28

The Sixth Method Doesn’t Work, But the Third Way Does? corruption. Now if all of them have undergone corruption, it would be impossible that anything exists, for there would remain no one who would bring anything into existence. Hence it follows necessarily that there would be no existent thing at all. 15

If every being could die or disintegrate, Maimonides says, there comes a point when this possibility becomes a reality: a point at which everything has ended. If this situation ever comes about, nothing more could happen, since all possible causes of events are gone. This total non-being would thus preclude any state in which things really do exist. However, lucky for us, we are not in such a state, and it is obvious that things do exist. As Maimonides says, “We perceive things that are existent. In fact we ourselves are existent.” Thus, according to the reasoning of this proof, it cannot be true that every real thing is corruptible because this would result in a state of non-being completely at odds with reality. At this point in the argument, then, we must grant that there are some beings that are subject to generation and corruption (as is obvious) and some that are not, which was originally not so obvious. “There must be a certain existent that is not subject to generation and corruption,” which has in it “no possibility at all; rather, its existence is necessary, not possible.”16 Like Thomas Aquinas after him, though, Maimonides adds a second part to this proof. He is not content solely with proving the existence of some being not subject to generation and corruption, because this being could still have its necessity caused by another – that is, it could be preserved from beginnings and endings by some other being. There are two kinds of necessary being, Maimonides says: ones that are necessary because of what they are and ones that are necessary through a cause: With reference to this existent’s being necessary of existence, there are two possibilities: this may be either in respect to its own essence or in respect to the cause of this existent. In the latter case, its existence and nonexistence are possible in respect to its own essence, but necessary in respect to its cause.17

By the principles of philosophy, though, Maimonides says, a being whose necessity is caused – one preserved from the realm of coming to be and 15

Ibid. Ibid. 17 Ibid., 248. 16

Jamie Anne Spiering

29

passing away by something outside its own essence – would still depend on a being whose necessity belonged to it by its essence. Thus: There can be no doubt that there is an existent that is necessary of existence in respect to its own essence. For without it, there would be no existent at all, neither one that is subject to generation and corruption, nor one that is not subject to them.18

(I should probably note, for those familiar with Thomas, that Maimonides does not have an infinite regress argument as part of the proof here). Some of the premises in this proof can be challenged – most of us, indeed, are familiar with common criticisms of the Third Way even if we haven’t read up on this Maimonidean proof in particular. My interest, however, is not in defending this proof against all comers, but in discussing why Maimonides thinks it is so much better than the other argument based on possibility and necessity. For he does think it is better: he gives this proof a ringing endorsement, saying that there can be “no doubt, no refutation, and no dispute”19 about this argument.20 Yet, both the sixth method and the third speculation seem to make the same basic claim that God must exist in order to explain why there is something rather than nothing. Both of them involve the reader in the act of imagining a state of absolute non-being or nothingness; both point out that such a state does not obtain, as revealed by sense perception. And for both, the reason why such a state did not occur must be that God really exists. What’s more, if we accept the addendum to the sixth method, both proofs give us a God who is “a necessary being” and the cause of the reality of all the possible things there are. What is it then, that, for Maimonides, makes the third speculation so different and so superior? There is much to be learned by considering Maimonides’ criticism of the sixth method and the parallels that can, and cannot, be drawn between these two texts. In what follows, I will discuss first Maimonides’ specific criticism of the sixth method and then his general critique of all the proofs offered by the Mutakallimnjn, exploring also whether the third speculation is subject to the same points of critique.

18

Ibid. Ibid. 20 But is he sincere in this praise? I’ll discuss that below. 19

30

The Sixth Method Doesn’t Work, But the Third Way Does?

C. Maimonides’ specific criticism of the Sixth Method and the superiority of the Third Speculation Why isn’t the sixth method a good proof? Maimonides’ explicit criticism has three basic elements: the proof fails because it is circular, because it is involved in the difficulties of a particularization argument, and because it cannot be used to demonstrate that God is incorporeal. To begin his criticism, Maimonides points out that the author of this proof is not doing a good job meeting his opponents, who would not necessarily agree that the world’s existence needs an explanation. Indeed, for the philosophers, the world had no limit in the past, will have no limit in the future, and has not come into existence after non-existence.21 In fact, the author of this proof is assuming what he is trying to prove, namely, that the world did not exist before God created it. If the world is, as the philosophers think, eternal, there is no question of it having been somehow preferred or caused to exist over some other alternative. A second line of criticism is based on the fact that this proof is an argument of particularization. Roughly speaking, a particularization argument claims that God must exist in order to answer the question of why things are this way rather than that way, when both are equally possible. The sixth method, instead of saying that God must exist to explain why some flowers are red and others are yellow, or why the sun is round instead of triangular, simply says that God must exist to explain why the world is there instead of not there. According to Maimonides, though, there is a weak spot in all particularization arguments: How does one support the premise that things could be otherwise? The Mutakallimnjn say that what is “intellectually admissible,” is possible, that is, if we can think something otherwise, it can be otherwise. Thus an elephant could be the size of a flea, for instance, or the sun could be triangular instead of round – but, on the other hand, since we

21

“The question – who has given the preponderance to its existence over its nonexistence? – can be asked only after it has been recognized that this existent has come to existence after non-existence. For this is a subject concerning which there is disagreement.” Guide I.74, 220.

Jamie Anne Spiering

31

cannot conceive of a round square, that is not a genuine possibility.22 But Maimonides has a basic problem with this argument from conceivability to possibility: what these thinkers mean by “intellectually admissible” might be merely what is imaginable. It may well be that a flea-sized elephant is a biological impossibility. When the Mutakallimnjn accept it as possible, then, could this be merely because they have no difficulty imagining a little tiny elephant? And could someone who knows more about anatomy and metabolism have privileged knowledge, arguing that a miniature elephant is not “intellectually admissible” or possible? If the proponents of the sixth method have no way of proving that the nonexistence of the world is rationally possible as opposed to possible in imagination only, they are exposed to the charge that they are proving the existence of God based on their own fantasies. As Maimonides points out, imagination is a poor foundation for reasoning: as a power of the soul, it is quite limited, since it is unable to conceive of universals and has no necessary connection to reality.23 He gives several examples to illustrate how our imaginations might tell us one thing about possibility, and yet reason or the evidence of our senses tells us the opposite. When we imagine that the world is round, for instance, we can’t help but worry that the people on the bottom would fall off. It seems impossible – to the imagination – that they could stay on. Yet they do. Reason’s “possibility report” and imagination’s “possibility report” conflict. Maimonides’ second example is a geometric one: the case of asymptotic lines. How could a line and a curve (such as a hyperbola) get closer, forever, yet never meet? Imagination says they can’t, but mathematical demonstration says 22 Readers of early modern philosophy will be reminded of similar claims: for Descartes, whatever he can conceive “clearly and distinctly” is said to be possible: “I know that everything which I clearly and distinctly understand is capable of being created by God so as to correspond exactly with my understanding of it.” Rene Descartes, Sixth Meditation. [In The Philosophical Writings of Descartes, vol. 2, trans. John Cottingham, Robert Stoothoff, and Dugald Murdoch. (Cambridge: Cambridge University Press, 1984.) 54]. For Hume, whatever is conceivable is possible: “Tis an established maxim in metaphysics, that whatever the mind clearly conceives includes the idea of possible existence, or in other words, nothing we imagine is absolutely impossible. We can form the idea of a golden mountain, and from thence conclude that such a mountain may actually exist. We can form no idea of a mountain without a valley, and therefore regard it as impossible.” David Hume, A Treatise of Human Nature (London: Penguin Books, 1969), 81. 23 Guide I.73, 209.

32

The Sixth Method Doesn’t Work, But the Third Way Does?

they can.24 Thus, just because we can imagine the world not existing gives us no grounds at all for saying that its non-existence is a genuine possibility, suitable as the foundation of further reasoning. Even if the claim that the world could exist or not exist is not based on imagination, how can one prove that it isn’t? Maimonides can’t see a way.25 One man’s “intellectual admissibility” is another man’s “vain fantasy.” The last reason for Maimonides’ rejection of the sixth method is that (like the other proofs of the Mutakallimnjn) it does not provide a solid basis for arguing that God is simple, single, and immaterial. After all, there is no reason why the cause of particularization must be a single spiritual being. Even worse, in fact: the reasoning used in the addendum to this proof can be used to prove that God is a body. Existence, which requires a cause in other things, does not require a cause in God, because otherwise we’d have an infinite regress. But by this reasoning, shape, size, color, and other bodily qualities that usually require a cause of particularization (pigment, a certain arrangement of parts, etc.) might not require one in God, and so he could necessarily have such qualities and be a body.26 For Moses this proof fails to follow the laws of Aristotelian logic: its premises are not true and its conclusion does not follow. The premise that the world is possible in the sense that it could not exist is not admitted by the philosophers; nor can it be demonstrated, nor is it self-evident. It is true that we can imagine the world not existing. However, if the premise is supported only by imagination, the conclusion doesn’t follow by logical necessity, but only by imaginary necessity. If I imagine a baby elephant who can fly, I will be led to imagine him dropping unpleasant evidence on people below, but this doesn’t prove that falling elephant patties are real. Similarly, if I imagine the world as the kind of thing that requires a cause, I will perforce imagine a cause, but that does not mean that there really is 24

Ibid., 210. Guide I.73, 211. See also a later passage: “Would that I knew whether this gate is open and licit, so that everyone can claim and assert with regard to any notion whatever that he conceives: This is possible; whereas someone else says: No this is impossible because of the nature of the matter. Or is there something that shuts and blocks this gate so that a man can assert decisively that such and such a thing is impossible because of its nature? Should this be verified and examined with the help of the imaginative faculty or with the intellect? By what can one differentiate between that which is imagined and what is cognized by the intellect?” III.15, 460. 26 Guide I.76, 230. 25

Jamie Anne Spiering

33

such a first cause. Moreover, even if we could somehow be sure of the premise that the world is possible and thus genuinely requires the existence of a cause, that cause whose existence would then be demonstrated might not be God in any real sense: the necessary being of the sixth method might well be material, shaped like a triangle, and have green stripes. While some might call that thing God, for any believer (Moslem, Christian, or Jew) to worship such a being or call it the deity would be idolatry. How does third philosophic speculation match up to these critiques? It is certainly not an easy or uncontroversial proof. Moreover, Maimonides does not present it as his own, but as that of the philosophers. He’s not necessarily committed to it, and, in fact, he comments that detailing such proofs is not the purpose of the Guide.27 However, he does praise it, and very strongly. Above all, after this proof has been used to demonstrate the incorporeality of the necessary being, he says that “It is he who is the deity, may His name be sublime.” This is an affirmation he pointedly withholds from the KalƗm proofs. If he is serious here, Maimonides claims that the necessary being demonstrated by the third proof is the God that he worships. How did the third proof achieve this level of superiority? The first problem that Maimonides had with the sixth method was that its author assumes the kind of world you need a God to create, then proves that there must be a God who created it. The third speculation avoids this difficulty. Whatever its shortcomings, this proof does not assert the most convenient world from a theist’s point of view – namely, a world that has come into existence after non-existence. Instead, the first part of the proof is usually thought to assume an eternal world (to give enough time for the state of total non-being to occur) which is the tougher case. The second problem that Maimonides had with the sixth method was that its claim that the world might not exist is unsupported. The third method avoids such an unproven assertion of possibility. It only claims that some things are possible (that is, subject to generation and corruption) a fact which is confirmed by observation in a way that claims about the whole world cannot be.

27

Guide II.2, 253.

34

The Sixth Method Doesn’t Work, But the Third Way Does?

Moses’ third problem with the sixth method was that it did not allow further proof of God’s incorporeality. The third way, however, does provide a direct proof that the necessary being is immaterial. The necessary being exists (necessarily) because of what it is, and thus can have no other cause of its existence. But material and composite things have causes besides their essences: the matter and form that make them up, not to mention their quantity, shape, and position. The argument Moses presents draws from several premises taken from natural philosophy, which I give in the footnotes: It follows necessarily that the existence of everything that is necessary of existence with respect to its own essence can have no cause, as has been set forth in the twentieth premise;28 and that in anything that is necessary of existence there cannot be a multiplicity of notions, as has been mentioned in the 21st premise.29 Hence it follows necessarily that, as has been set forth in the twenty-second premise,30 it is not a body or a force in a body.

This, above all, is why Maimonides seems to regard the third way as a demonstration for the existence of God, saying, after he has finished the proof of immateriality, that “It is he who is the deity, may His name be sublime.” Thus, initially, the third way seems to be superior simply because it avoids the egregious errors of the sixth method: it does not make a theist assumption the basis for a proof for God’s existence, nor does it make an unproven assertion of the possibility; finally, it can be used to demonstrate an incorporeal God. A little speculation reveals further, implicit reasons for Moses’ harsh words about the KalƗm proof. The errors in this proof encourage very 28 “Everything that is necessarily existent in respect to its own essence has no cause for its existence in any way whatever or under any condition.” Guide II, prologue, 238. 29 “Everything that is composed of two notions has necessarily that composition as the cause of its existence as it really is, and consequently is not necessarily existent in respect to its own essence, for it exists in virtue of the existence of its two parts and of their composition.” Ibid. 30 “Every body is necessarily composed of two things and is necessarily accompanied by accidents. The two things constituting it are its matter and its form; and the accidents accompanying it are quantity, shape, and position.” Ibid.

Jamie Anne Spiering

35

damaging tendencies in the life of the mind. For those who believe in God, reading a circular argument for something that they already hold as true can weaken their sense of logic, their sense of what has been demonstrated, what is merely plausible, and what is merely insisted upon. For non-believers, reading a circular proof makes believers look silly. Moreover, the believer who grants himself the premise he wants forms a very bad habit – instead of being open to reality, he insists that reality be the way he wants it. Those who are willing to be guided by imagination alone form a similar bad habit – as Maimonides points out, if we had all been guided by imagination alone, we’d still believe the earth was flat.31 Finally, we cannot underestimate the blasphemy that Maimonides sees as a very real possibility – proving that something exists, and calling it God when it is not, puts the believer in a horrible position.

D. Interlude, with doubts: is the third way really better? Unfortunately, almost as soon as we create a clear picture of the superiority of the third proof in these respects, doubts arise to blur it. Given that Moses warns us that he will occasionally say things he doesn’t mean, can we be sure that he really means his words of praise for the proof or his claim that what the third speculation proves is the deity? There are reasons for careful readers to suspect that the third proof might have just as many problems as the sixth method. A number of serious criticisms can be made about the logical moves in both parts of the proof. I will mention only one here: a claim can be made that the first part of the proof is circular. In order to prove that there are some necessary beings, the author of the proof has to prove that not all beings are contingent. But, in order to prove that the all-contingent world leads to a state of present non-being, the usual defense involves an infinite time, thus assuming a necessary being in order to prove that there is a necessary being.32 Thus, the third way, it might be said, is just as circular as the sixth method – it’s just disguised a little better.

31

Maimonides also thinks that following the imagination is morally dangerous: “Every deficiency of reason or character is due to the imagination or consequent upon its action.” Guide II.12, 280. 32 This argument is made by Josef Stern, in “Maimonides’ Demonstrations: Principles and Practice,” 68.

36

The Sixth Method Doesn’t Work, But the Third Way Does?

Just as seriously, the third way might not prove the existence of the God that people of faith worship any more than the sixth method does. After all, Maimonides argues in the first part of the Guide that all names are said of God equivocally. If so, then any proof that concludes that God exists, using any middle term, must commit the fallacy of equivocation.33 In Maimonides’ writing, there is no doctrine of analogy to soften the fact that no term is said of God in the same way it is said of us. If what we mean by “necessary being” in any part of the proof is said only equivocally of the God we worship (“the Deity, may his name be sublime”) then the third speculation doesn’t prove the existence of God at all. Given these criticisms, it is natural to question Maimonides’ sincerity in praising the third proof. Perhaps he expects the careful reader to see that the third philosophic speculation merely provides us with a more complicated version of the same three mistakes made by the KalƗm argument. Perhaps, rather than making the point that the philosophers have good proofs for the existence of God, he is making the point that nobody can really solve the difficulties we encounter in such a project. Is this the interpretation we should take? I don’t believe so, and will explain why in the next section.

E. Further superiority of the third speculation I believe an argument can be made for the sincerity of Maimonides’s words in praise of the third way. The third “philosophic speculation” is a genuine advance over the KalƗm arguments, not just a more complicated proof or one with a different set of assumptions about the world, and I believe Maimonides wanted us to see it as such. This is what I will argue, at least, in the following section. As I have said already, I cannot defend this argument against all comers. I am not undertaking, here, to prove that the demonstration is a perfect one; only that its superiority to the sixth method is absolute. 33 The problem can be explained clearly by the use of an example. If I argue that “The dog in the picture is an animal, and animals breathe, therefore, the dog in the picture breathes,” my argument fails because “animal” is used equivocally in my premises. Similarly, the author of this proof may argue that a necessary being exists, but if he then claims that that being is God, he may be making a leap too far. The syllogism: “A necessary being exists, God is the necessary being, therefore, God exists,” will fail if “necessary being” is said of God in a different sense in which it is proved by the argument.

Jamie Anne Spiering

37

Maimonides makes a number of general remarks about the superiority of the philosophical proofs that are helpful in this context. One thing he frequently says is that the philosophers base their arguments on “sense perception” and on “the nature of existence” while the Mutakallimnjn abolish the nature of existence and do not pay attention to what can be observed in the world.34 This praise of Aristotelian empiricism seems, at first, as if it wouldn’t really apply to the third proof, which is not thought to be particularly empirical. True, it makes use of the evidence of sense perception to assert that things exist, but so does the sixth method. However, the third speculation is really more empirical in one important way. Both proofs must use the possibility of the world to argue for the necessity of God, but only the third speculation appeals to our observation that things begin and end, are born and die. It appeals to the reality of observable substantial change, probably the most fundamental precept in Aristotelian natural philosophy. For Maimonides, this is no small matter: premises are more demonstrative if they are “primary and immediate” rather than being derived from other demonstrations, and this is exactly the kind of premise that “some things in the world are born and die” is. It rests on immediate perception of the world. The other thing that Maimonides frequently notes as a point of superiority for the philosophical proofs is that they are demonstrations,35 whereas the KalƗm proofs are only called “methods.” (It is obvious that he is assuming the Aristotelian criteria for demonstration, as they are explained in the Posterior Analytics.) In the most strict sense, of course, no proof for God’s existence can be a demonstration, because a demonstration works through the cause as the middle term, and God has no cause. However, as Josef Stern has argued,36 Maimonides was quite aware of what Thomists call demonstration quia and the third speculation (like the other philosophic proofs) is a demonstration of this kind, from effect to cause. This kind of argument uses the effects as a middle term, because the effect is at least the cause of our knowledge. And, in fact, both the sixth method and the 34

“All the first Mutakallimnjn…did not conform in their premises to the appearance of that which exists.” Guide I.71, 178; “Every one of their premises, with few exceptions, is contradicted by what is perceived of the nature of that which exists.” Ibid., 183. “These philosophic proofs…are derived from the nature of existence.” Ibid., 182. 35 “This speculation has led by means of a demonstration…” Guide II.1, 246; “This is a demonstration concerning which there can be no doubt…” Ibid.,248; 36 Stern, 55-64.

38

The Sixth Method Doesn’t Work, But the Third Way Does?

third proof work this way, using the world as an effect to conclude to the existence of God as the cause. However, there is an important difference between them. To put it as simply as possible, the author of the sixth method asserts that God exists as a hypothesis that would explain a certain effect. The author views a phenomenon (the real existence of a world which is only possible) which he cannot explain, and “God” is his name for whatever mysterious X tips the balance from possibility to reality. For the reader of such a proof, God is, basically, a theoretical entity like phlogiston, or like the atom. The hypothetical God of the sixth method may be real, but the method used to prove his existence means that he is a theory that may need to be refined, or, possibly, scrapped. The third way, however, does not merely produce God, like a rabbit from a hat, in order to explain the unexplainable. Instead, God’s existence is affirmed as a necessary prerequisite for the world we sense. If the sixth method proposes God as a theory, more or less like the first step in the hypothetico-deductive method of modern science, the third way follows the strict reductio ad absurdum of Euclidean geometry. This is the method that proves things we cannot imagine and would perhaps never have imagined or theorized: that there are asymptotic lines,37 for instance, or that no straight line will fit between a curve and its tangent. Rather than being the only hypothesis that completely explains the phenomena, God’s existence for the reader of the third proof is considered the only condition under which phenomena could be the way we see them. It’s important not to underestimate this distinction. To say, “Were it not for God’s existence, I would have no explanation of true premises,” as the sixth method does, is very different from saying “Were it not for God’s existence, true premises would be false,” as the third speculation does. To summarize the previous points a bit, the third way does not demonstrate that God exists through his cause – that would be impossible – but it does work through the causes of our knowledge, using both sense perception and the kind of reasoning that is used to establish things we can’t imagine, rather than the tentative reasoning we use to assert that theoretical (or imaginary) entities are real. Because of this, in the eyes of 37

Apollonius, Conics, book II, propositions 1, 13, and 14. [Trans. R. Catesby Taliaferro (Santa Fe, New Mexico: Green Lion Press, 2000)]. Maimonides himself cites this proof as an example of a proven unimaginable.

Jamie Anne Spiering

39

someone committed to Aristotelian logic and natural science, its strength as a demonstration is clear, as is its inherent superiority to the sixth method. As yet, though, I have said nothing about the most serious doubt regarding the third proof: is the necessary being whose existence it asserts really God himself? How can Maimonides, with his strict doctrine of divine names, look at this proof and not see the fallacy of equivocation? I would suggest that the third speculation is designed to acknowledge the key elements in Maimonides’ doctrine about predicates said of God. These elements are as follows: a) Names are always said of us and of God equivocally.38 b) Names said positively of God ought to be interpreted as a negation of the opposite.39 (e.g. good = not bad) c) We increase in our knowledge of God by increasing the negations we work through.40

Why do I say that the third way acknowledges these teachings? To begin with, the author considers whether things can be generated and corrupted; this possibility is denied of some “necessary being” that the first part proves must exist, so that “there must be a certain existent that is not subject to generation and corruption.” Thus, our concept of this necessary being is formed by a negation of possibility. In the second part, however, a new type of possibility is introduced: things are called possible in this sense in that, even if they have no beginning and end, they have their being caused by another. This kind of possibility is likewise negated of the being that the second part proves to exist. Thus, at the end of the proof, when God is said to be a necessary being, the sense of necessity has been reached by way of negation: a reader understands “necessary” to mean “not possible” – and, in fact, not possible in two senses, reached progressively. This is precisely the kind of thinking that Maimonides recommends; we grow in our understanding of God by repeated negation. In addition, the proof acknowledges the equivocity of the names of God, at least in the sense that it makes sure that the reader understands that 38

Guide I.56, 131. Guide I.58, 134-135 40 Guide I.59, 138-139. 39

40

The Sixth Method Doesn’t Work, But the Third Way Does?

“necessary,” in the sense in which it is said of God, can be said of no other being. If necessity is said at all of anything other than God, it is said equivocally. Getting a sense of the mental journey that the two parts of the third way encourage us to undertake gives us a final way of expressing the genuine superiority of the third proof. Both the sixth method and the third way prove the existence of God as a necessary being, but exactly what is the “necessary being”? It is fairly easy to see that “necessary being” means a lot more to the reader of the third way than it does to someone who thinks through the KalƗm proof. The sixth method demonstrates a “necessary being” without ever making its audience think about what a necessary being is. For all we know, it could have necessary purple polka dots and be necessarily 17 feet high. The third speculation won’t allow a reader to get away with such sloppy thinking. It requires us to think about things that come to be and pass away, and we know, if we stop to think about it, that purple fades. If something were to be eternally purple, it would need something to keep it that way. And 17 foot tall things are composed. Even if, for some reason, they never die or corrode, they depend on the parts that make them up. Only the third way makes its audience think through all this, realizing that a necessary being is far outside both experience and imagination. Going through the work of the proof carries the reader to where he needs to be in terms of understanding what necessary existence means. The sixth method is easy, but if Maimonides is right, that charm is not only deceitful but dangerous. Someone who embraces it not only thinks that he has proved God when he hasn’t; he makes no commitment to the work of realizing what he’s actually thinking about when he thinks about God. There is no point in claiming that there are no difficulties involved in the third philosophic speculation as it is presented by Maimonides. And certainly Maimonides could see some of them, if not all of them. But I think there are good reasons to see why the philosophic sense is not just a view from a different camp but a genuine improvement. It may not be the best way to prove God’s existence, but it’s the right way to do so: in its appeal to sense perception, it is more demonstrative; in its appeal to the reductio ad absurdum, it helps us avoid the pitfalls of mere theorizing; in its consideration of coming to be and passing away, it encourages the reader to think about what a necessary being actually means.

Jamie Anne Spiering

41

Conclusion, and Postscript: Suggested questions to ask about a “new proof” Thinking through the comparison Maimonides sets up between these proofs gives us key principles that guide and illumine the task of proving God’s existence. Everyone agrees that, if we are going to prove God’s existence, we have to start with the world, and reason from effect to cause. Maimonides, however, points out what is perhaps the most serious question to be asked about any proof for God’s existence: does it lead its hearers to the conclusion that God must be if the most obvious things in our lives are, or does it simply toss God at the audience, like a bone, in order to feed a ravenous clamoring for causes that it first uses imagination to create? Again, it has been admitted for a long time that an argument can probably be put together based on the contingency of the world. However, Maimonides points out that it is important to be careful about the foundation for any assertion of contingency. The fact that we can imagine the world as non-existent, or as uglier or more limited than it is, does not give us license to claim that it truly could be that way. Believers, of course, make this claim on faith. But such claims can have no place in a proof for the existence of God. We can’t think without imagination, but that’s different from letting our imagination do our thinking for us; being able to imagine or conceive or even describe a different world is not at all relevant to the world we have here. On the flip side, it has seemed to many, over the years, that if we can prove the existence of a necessary being, we can prove the existence of God. But Maimonides reminds us that we are going to have to grasp necessity through negation of possibility – we have to think through what necessity might mean, not just assert a being who could fill in the gaps in our knowledge, labeling that gap-filling ability his “necessity.” One final issue is perennial for the task of proving God’s existence: Why be critical to people who are on our side? If someone has produced a proof for God’s existence, why should a believer go out of his way to point out the flaws? Maimonides can give us the answer to that as well: at best, such proofs can make believers look bad if they are presented seriously as the reason we believe. At worst, though, such proofs are nothing but a way of playing to imagination, losing sight of reason, and they can lead to the blasphemy which is a very real possibility for the

42

The Sixth Method Doesn’t Work, But the Third Way Does?

careless speculator. If Maimonides is right, producing an argument like the 6th method feeds the three things that are worst for humankind: living by imagination, trying to make reality the way we want it to be, and refusing to see God except in our own image. The third way, however, does the opposite: it uses our openness to reality, encourages us to make imagination serve reason, and requires us to see that God is not what we see in ourselves or in the things of this world. Maimonides’s discussion of the relative merits of KalƗm and philosophical proofs might have timely as well as perennial value. I would like to close with a sample of how his ideas might be used to evaluate recent theistic arguments: it’s not the main purpose of this paper, just something to think about as a practical application of the ideas discussed here. I have in front of me a popular book called New Proofs for the Existence of God,41 which our department has been encouraged to use. In the section called “New Proofs From Physics” the author summarizes the following argument, made, he says, by a large number of recent thinkers: 1. The values of the universal constants controlling the interrelationship among space, time, and energy in the universe must fall within a very narrow, closed range in order to allow any life form to develop. 2. But the possible values that these universal constants could have had that would have disallowed any life form from developing are astronomically higher (falling within a virtually open range.) 3. Therefore, the odds against an anthropic condition occurring are astronomically high, making a life form (or universal condition allowing a life form) exceedingly improbable. This makes it highly, highly unlikely that the conditions for life in the universe occurred by pure chance, which begs for an explanation (cause) – physical or metaphysical.42

Although the author presents this as an argument from design, it’s not difficult to see that this proof, fleshed out as it is with equations, data, and talk of probability, has the familiar bones of the sixth method. Why does the world (as we know it) exist rather than not exist, when the possibility 41

Robert J. Spitzer, New Proofs for the Existence of God, (Grand Rapids, MI: William B. Eerdmans Publishing Company, 2010). 42 Ibid., 50.

Jamie Anne Spiering

43

of its being otherwise is “virtually open”? It can only be God! Based on Maimonides’ thinking, however, there are questions that should be put to anyone who pursues this line of argument. On the simple side, on what do these thinkers base their assertion that constants such as the gravitational constant could be other than they are? Can they prove the contingency they assert? Or are they just imagining a world with different natural laws? I certainly do not doubt that there is reasoning behind such claims, but, at least in this book, the author presents us with none43 – and as long as he does not, the argument is wide open for critique. Then there are the more searching questions. Do those who offer such proofs acknowledge that the God they posit here is purely a hypothesis, no more real than phlogiston? Are readers aware of this? Are they aware of the many long chains of reasoning that support each premise, fundamentally different in character from the observations of change in things that surround us? And finally, and most seriously, do those who write such proofs take seriously their responsibility to lead the reader to a sense of what the word God actually means? How imaginary is the God their readers are left with? The more serious among the “new” theists, I am sure, are doing what they can to answer these questions; I think all philosophers of faith would do well to begin asking them and seriously considering the answers we get before offering our students these newer, physics-based proofs. In this way, Maimonides’ criticism of the KalƗm arguments can have practical value as well as theoretical interest.

43

The author says that “the equations of physics describe empirically discovered consistent interactions and interrelationships among energy, space, and time within a universe. Hence, the equations of physics (unlike mathematics) could be other than they are.” Ibid., 52. I find it difficult to take this very seriously. Of course, the descriptions could be otherwise if reality were otherwise. But who can say whether reality can be otherwise?

CONCEIVING CREATION: RESPONSE TO SPIERING AND SEESKIN DANIEL DAVIES

In her paper on two proofs for God’s existence in Maimonides’ Guide, Spiering addresses a venerable question asking what the difference is between two arguments that sound very similar. As she mentions, Isaac Abravanel argued that both proofs are potentially sound and ought to be accepted, even though the KalƗm argument requires tweaking. At the opposite extreme, Moses Narboni thought that neither argument succeeds. Moreover, he seemed to indicate that Maimonides’ attitude toward the sixth KalƗmic argument is meant to hint at a covert rejection of the third philosophical argument and ultimately, it seems, to hint at Maimonides’ hidden acceptance of the world’s eternity.1 Spiering ploughs a furrow different from both of these and follows what she calls a naïve reading which endorses Maimonides’ explicit meaning. I would like to begin this response with a general comment supporting naïve readers of the Guide. I take a naïve reader to be one who assumes that Maimonides adheres to those theological doctrines that he openly professes.2 The sophisticated reader, by contrast, is one who is able to discern Maimonides’ true belief, which often opposes the naïve theological doctrines and is only indicated by hints, if at all. There are a 1

Der Commentar des Rabbi Moses Narbonensis edited by Jacob Goldenthal (Vienna: 1852): 19b. 2 There are a great variety of ways to read Maimonides, far more than this thumbnail sketch can cover, so it is of necessity simplified. For more detailed analysis see Aviezer Ravitzky, who explains that the popular opposition between the naïve and the philosophical readers is judged on the basis of whether one can perceive that Maimonides’ “heart and his lips were not at one.” “The Secrets of the Guide to the Perplexed: between the thirteenth and the twentieth centuries,” in Isadore Twersky (ed.) Studies in Maimonides (Cambridge Mass.: Harvard University Press, 1990): 177.

46

Conceiving Creation: Response to Spiering and Seeskin

variety of ways in which the secret teaching is said to be detected, one of which involves identifying positions that Maimonides adopts that do not cohere with one another. In such cases, it is often claimed, Maimonides must adhere to one of the two positions but, in the interest of consistency, not to both of them. A naïve reader is forced, according to the more advanced interpretation, to hold that Maimonides accepted inconsistent positions, either because he was unaware of the inconsistency or because it did not bother him. This is a slight on the naïve reading, I think, and Spiering’s engaging with the subject can help lend it greater credence. The naïve reader is simply one who attempts to understand Maimonides’ words without presupposing that a particular position would be superior to another. She presumes that the Guide is worth listening to and learning from. Maimonides’ text and our contemplation of it is therefore the criterion for our explanations; our limited understanding cannot be the measure of what Maimonides truly believed.3 My own assumption is that Maimonides would have been concerned to be consistent, but if his statements do not appear, at first sight, to cohere, that does not automatically allow us to conclude that they do not in fact cohere or that Maimonides believed them not to.4 There might be some aspect of the argument that we have not yet grasped. The naïve interpretation, if it strives to reconcile apparently disparate opinions, has the potential to be more philosophically fruitful than the more sophisticated view that refuses to ask whether the opinions can be understood in more subtle ways. It also has the potential to clarify Maimonides’ arguments by identifying logical gaps or rooting out presuppositions that are not immediately obvious, especially to a current reader. This helps us to understand the limitations in Maimonides’ arguments and judge whether they are apparent or real. Given such an assessment, there is no need to apologise for approaching the Guide as a Christian, nor as a trained Thomist. On the contrary, such training would be appreciated by many of Maimonides’ readers, particularly those who identify with Spiering’s own naïve approach of taking the Guide at face value.

3

The naive reading might be equated with what Moshe Halbertal terms a “conservative” view, and the sophisticated with a “radical”. See Maimonides: Life and Works (Princeton University Press, 2014): 312–321. 4 Frank Griffel makes this point about medieval philosophers more generally in AlGhazƗlƯ’s Philosophical Theology (New York: Oxford University Press, 2009): 286.

Daniel Davies

47

When addressing the two arguments for God’s existence, Spiering homes in on the major reason for the different attitudes that Maimonides displays toward the KalƗm proof, on the one hand, and the philosophical proof, on the other, which is the way in which they use imagination and reason in their arguments and the order in which they arrange them. This is also a central part of his overarching critique of the KalƗm generally. The philosophers’ proof, as it appears in the Guide, may strike some readers as strange. Tamar Rudavsky explains that Maimonides seems to be “caught between two poles: that of Averroes, who argues that metaphysics alone is not sufficient to prove the existence of God, and that of Avicenna, who feels he has in fact offered such a metaphysical proof.”5 Avicenna wanted to prove God’s existence from metaphysical premises alone, whereas Averroes used premises based on the sphere’s eternal motion. Maimonides’ argument includes a version of Avicenna’s proof as its second part but also contains a prior section built on empirical premises.6 It is not easy to see how the two parts fit together. Spiering compellingly interprets the proof in a unified way, making sense of the need for physical premises in the first part, and arguing that they reflect the necessity of reasoning from empirically verified premises. But while Maimonides bases scientific claims on premises drawn from sense experience or from primary intelligibles, the KalƗm uses premises that are based on imagination and are false. One of these involves the claim that the world is the kind of thing that could be nonexistent because its nonexistence can be imagined. This raises a question about Maimonides’ presentation of the KalƗm argument. Can a genuine creation ex nihilo be imagined? The imagination is able to combine substances and accidents perceived by the senses and depict objects such as a flea sized elephant. In doing so, it combines different categories, substance and size, in a fictional way. One of the reasons it is able to do so is because there is no opposition between the two parts of the combination. It is normal for a substance to be of a certain size, even though elephants are not the size of fleas. The kind of impossibility that is created by the imagination seems to differ from the 5

Tamar Rudavsky, Maimonides (Chichester: Wiley-Blackwell, 2011): 53. There is some scholarly disagreement over whether the Avicennan proof is itself entirely “ontological” or inclusive of “cosmological” elements. See Toby Mayer, “Ibn SƯnƗ’s ‘BurhƗn al-ৡiddƯqƯn’,” Journal of Islamic Studies 12 (2001): 18–21.

6

48

Conceiving Creation: Response to Spiering and Seeskin

intellectually inadmissible round square. The tiny elephant would be more like imagining a giant football the size of the sun, or a match between miniature players taking place in a plant pot. And whereas the tiny elephant and giant football can be imagined, because a substance could indeed be small as a flea or large as a star, the round square cannot because both are shapes. They are in the same category and to affirm one would involve denying the other since to be a circle means to be something different from a square; it would involve the claim that a circle is not a circle. The KalƗm agree that such a thing would be impossible as it cannot be depicted in the imagination. These examples are different from those that could be compared to creation, however, as imagining creation would involve a change, specifically from nonbeing to being, rather than a combination of diverse attributes into a single substance. Instead, it would be closer to a generation of one thing from another thing. For example, the imagination could depict a rose bush growing from an apple seed. Although impossible, this is a change from one thing that can be imagined, because of its matter, to another that can also be so imagined. If absolute nothingness genuinely involves no matter at all, and the imagination deals only with material things, absolute nothingness cannot be imagined. So when Spiering states that “we can imagine the world not existing,” what sort of nonexistence are we imagining? Is it the case that what we would be imagining would be something else in place of the world, such as a void space? So, since the imagination is only able to depict material realities, it is unable to judge whether the world could possibly not exist, and therefore the possibility of an absolute beginning, because it is unable to form a picture of absolute rather than relative nonexistence. I believe that Maimonides makes this point in a chapter that Spiering rightly draws attention to, III:15. On this reading, he says that neither the imagination nor the intellect is capable of judging creation ex nihilo to be possible or impossible, because neither can properly depict it. While the philosophers are right to object to the KalƗm accounts of what is possible, which are based on imagination, they are mistaken to liken creation ex nihilo to a square circle. If it is indeed impossible, it cannot be shown to be impossible in the same way. Whereas both square and circle can be both conceived and imagined independently, but not in combination, “nothing” cannot be conceived or imagined at all. So Maimonides states that he does

Daniel Davies

49

not have a criterion for judging whether creation ex nihilo is possible. Neither imagination nor intellect will do the job.7 If Maimonides is right to say that the imagination cannot free itself from matter,8 every generation that the imagination depicts must be from a prior existent matter. Since they say that only what is imaginable is possible, the sort of creation that the KalƗm are able to posit must be a creation from relative nonbeing. So when criticising the KalƗm reliance on imagination, the examples that Maimonides uses to illustrate false images that it generates do not seem to correspond to a creation from absolutely nothing. If the KalƗm judge creation to be possible, the creation that they posit must be imaginable. But in this case, I contend, they are not really imagining creation ex nihilo but, rather, a creation from relative nonbeing. The reliance on imagination, therefore, might also account for the equivocation which is the basis of the second criticism that Spiering explains, that the KalƗm argument fails to meet the philosophers on their own terms. If the imagination views creation as a generation of one thing that did not previously exist out of a prior existing thing, the possibility used in this proof indicates the potentiality inherent in all matter. The philosophers would not agree that the entire world is possible in this sense, however, but only in the sense that it is not self-sufficient and is therefore caused. It does not seem far-fetched, however, to say that all contingent things come to be and pass away. The philosophers’ proof seems also to rely on the idea that every possible thing will, at some time, not exist, so there seems to be a connection between the two ideas, even if not a per se connection.9 This raises a further question about the major premise of the KalƗm proof, which states that the world is contingent because otherwise it would be God. God’s existence as the necessary being is granted and, for the argument as a whole to work, it is accepted that whatever is possible begins to exist at a certain time. But is it legitimate for the KalƗm 7

I have argued for this reading more extensively in “Between Philosophers and Theologians: Maimonides’ Response to Avicenna’s Infinite World,” Proceedings of the Seventh CISMOR Conference on Jewish Studies (2013): 84–103. 8 E.g., Guide I:73, 209. 9 For Maimonides’ use of this idea see Charles Manekin, “Problems of ‘Plenitude’ in Maimonides and Gersonides,” in Ruth Link-Salinger (ed.) A Straight Path: Studies in Medieval Philosophy and Culture (Washington, D.C.: CUA Press, 1988): 183–194.

50

Conceiving Creation: Response to Spiering and Seeskin

argument to begin by attributing possibility to the world as a whole? As Spiering points out, one of the differences between the philosophers’ proof and the KalƗm approach is that the philosophers begin with premises verified through sense experience. Perhaps the KalƗm argument treats possibility as a property of the entire cosmos, whereas the philosophical proof requires that it be predicated only of each individual thing in the cosmos. The latter claim can be supported inductively by drawing on sense experience, but the former cannot. Does the KalƗm proof therefore rely on a fallacy of composition? It may be illegitimate to argue that because all individual parts of the whole are contingent, the aggregate shares contingency in the same way. For example, if all bricks in a wall are red, I can conclude that the wall as a whole is red. But from the fact that all bricks in a house are cuboid, I cannot draw the conclusion that the house itself is cuboid. If this is right, the equivocation might relate not only to different senses of possible (potential or contingent) but also to what it is said of (the whole of created existence or all individual objects). So the imagination is at the root of all criticisms that Spiering identifies in Maimonides’ treatment of the argument. Its circularity leads to an argument that can only be a creation from relative nonbeing, it ignores the empirical science of the time by asserting premises that allow the imagination to judge what is possible, and it leads to belief in an anthropomorphic God because the imagination cannot free itself from matter.10 Spiering also mentions that the KalƗm argument is subject to all the criticisms that apply to arguments from particularity generally. This comment raises the question of what these criticisms are, and why they do not also apply to Maimonides’ own argument, which is explained in Seeskin’s paper. This question is, of course, related to Ogden’s question about why Maimonides would use such a proof. It seems to me that there is a fundamental difference between the way in which Maimonides explains the KalƗm proof for particularity and his own proof. The KalƗm proof is based on the claim that there is no reason for anything to be one way rather than another. For example, there is no reason that a camel would generate another camel rather than an ant or superman. According to the scientific system that Maimonides attributes to the KalƗm, there is no difference between the parts making up one object and those that 10

It is worth noting that, despite this, Maimonides reports that the KalƗm agree that God cannot become a body, 1:73, 211.

Daniel Davies

51

constitute another object. It is therefore necessary for them to be created anew at every moment with the form that they have. Should God so wish, God could instantaneously transform a camel into Superman, and that this does not happen can be put down to the fact that God does not generally do such things. The consequence is that natural causality is denied and science does not reveal any truth about the natures being investigated. Spiering asks how it is possible to support the position that things could be otherwise than they are. The KalƗm are able to do so because, in Maimonides’ terms, they “abolish the nature of that which exists” by denying causal power to any created being.11 Again, Spiering points out that scientific negligence is a major charge that Maimonides levels against the KalƗm. A particularisation argument such as this cannot be used as a foundation for our scientific understanding of the universe. It destroys science at the root since, in the KalƗm view, there is no way to conclude anything about the natures being investigated. They are all simply fixed by God however God should wish them to be. But if there is a particularisation argument that is consistent with empirical evidence, it can support the claim that we do not understand everything and limit science without destroying it. It draws attention to the limitations of human intellect and ultimately places God at the root of science rather than in opposition. By contrast with the KalƗm argument, Maimonides’ particularisation argument does not begin from denying natural processes altogether: he affirms causality but uses particularisation to argue for limitation on our scientific knowledge. Asking why there is one actual state of affairs, or set of natural laws, rather than another seems different from asking why there is an apparent disorder that is scientifically scandalous. In response to the question why Maimonides would use a particularisation argument, one could say that it accords with the theological positions he explicitly supports, which are founded on the claim that God creates through will. If will directs knowledge toward an end, an object that is willed must also be known, so there is a connection between these doctrines and God’s knowledge.12 Maimonides’ particularity argument therefore includes the 11

Guide II:19, 303. See also Guide III:21, 485, in which Maimonides expresses the belief that God knows particulars and that such a belief, which, like creation, opposes the opinions of the philosophers, cannot be demonstrated.

12

52

Conceiving Creation: Response to Spiering and Seeskin

claim that God knows the particulars of the superlunar world and is therefore able to arrange in such a way as God’s will requires. So rather than setting off a process, which then proceeds along the lines of natural necessity, God must be able to create states of affairs that are not intelligible when judged from the perspective that we understand to be natural: the superlunar realm is not subject to natural necessity. As Seeskin explains, Maimonides connects the idea that God creates through will with creation ex nihilo and de novo. He also ties these concepts to other doctrines, such as belief that miracles can happen.13 Miracles indicate that God’s activity in the sublunar world is carried out through God’s willing, and therefore knowing, particulars, as they do not follow the universal, natural laws. As Tzvi Langermann has argued, belief in miracles indicates further the limitations of human understanding, as miracles are individual events that are not subsumed by scientific knowledge.14 This assessment returns us to Spiering’s closing remarks in which she reminds us of the need to pay attention to what we do not or cannot know. Acknowledging as much is crucial if we are to avoid making God an intelligible explanation resembling other causes, as Spiering shows that the KalƗm proof does.

13

For Maimonides’ view of divine will, and its following from divine wisdom, see Charles Manekin, “Divine Will in Maimonides’ Later Writings,” Maimonidean Studies 5 (2008): 189–221. 14 Langermann, Y. Tzvi, “Maimonides and miracles: The growth of a (dis)belief,” Jewish History 18 (2004): 147–172.

PART 2: DUNS SCOTUS ON LOGIC AND METAPHYSICS

ON THE AUTHENTICITY OF SCOTUS’S LOGICAL WORKS EDWARD BUCKNER

Traditionally, a total of seven works on logic and semantics were attributed to John Duns Scotus: six question-commentaries on the book of Porphyry, on the Categories, the Perihermenias (in two versions), the Sophistical Refutations, the Prior Analytics, the Posterior Analytics, and a work on the ‘Modes of Signification’ (De modis significandi). They were supposed to have been written early in Scotus’ career during his stay in the Oxford Franciscan studium, perhaps between 1290 and 1295.1 It has long been uncertain whether Scotus wrote any of these works. In the early modern period, Jacobus Naveros found inconsistencies between some of these works and Scotus’s commentary on the Sentences, leading him to doubt whether Scotus had written any logical works at all.2 More recently, Richter3 has argued that owing to the lack of citations by contemporaries of Scotus, and the late date of most of the manuscripts attributing the logical works to him, it is doubtful whether any are by him. In 1922 the author of De modis significandi was shown by Martin Grabmann to be Thomas of Erfurt, a fourteenth-century modist logician. The authenticity of Questions on the Prior Analytics4 was called into doubt in the 1930s.5 Its true author remains unknown. The two books on the Posterior Analytics (In librum primum et secundum Posteriorum Analyticorum Aristotelis Quaestiones) are also thought to be inauthentic,6 1

See Vos 2006: 124. See Ashworth 1987. 3 Richter, 1988 p.16, cited in Vos 2006, p.126. 4 In librum primum et secundum Priorum Analyticorum Aristotelis Quaestiones, Opera Omnia II (Paris: Vivès 1891), Q.8, pp. 98-101; see also Gál 1977, pp. 7071, and Zupko 1994-97. 5 Gölz 1934, 56. 6 See In librum primum et secundum Posteriorum Analyticorum Aristotelis Quaestiones Opera Omnia II (Paris: Vivès 1891), pp. 199–347. 2

56

On the Authenticity of Scotus’s Logical Works

probably written by John of Cornwall, who studied at Oxford in about 1300 and who taught theology at Paris from 1310 to 1315.7 There is also controversy over Scotus’s Questions on the Sophistical Refutations,8 which A. C. S. McDermott and Allan Bäck have argued were also written by John of Cornwall,9 although both extant manuscripts of the Questions on the Sophistical Refutations, unlike those of the other two commentaries, explicitly ascribe them to Scotus. Of the remaining three works – the Questions on Porphyry’s Isagoge, the Questions on the Categories, and both sets of Questions on the Perihermenias – there have been concerns on internal grounds that the author one of them (the Questions on the Categories) was not Scotus. Since there is strong internal evidence that all three are by the same author, did Scotus write any works on logic? In what follows, I will re-examine the question of authorship by an appraisal of the existing evidence, both external and internal, including recent work on the comparatively obscure Questions on the Perihermenias. I shall argue that the external evidence for authorship is weaker than has been previously thought, but that the internal evidence is considerably stronger. I conclude that there is overwhelming evidence that three works – the Questions on Porphyry, on the Categories, and on the Perihermenias (both versions) are authentic. About the Questions on the Sophistical Refutations there is still doubt.

7

See Minges 1930, VI, and Smeets 1942, 8–9. In libros Elenchorum Quaestiones, Opera Omnia II (Paris: Vivès 1891), pp. 180; now in the critical edition: B. Ioannis Duns Scoti Opera Philosophica II (St. Bonaventure, NY-Washington, DC: The Franciscan Institute and The Catholic University of America, 2004), pp. 255-730. 9 McDermott 1972, p. 274; Bäck 1996, pp. 205-7, 274-78. 8

Edward Buckner

57

Abbreviations QPor QCat QPer I QPer II QMet QSE

Questions on Porphyry Questions on the Categories Questions on the Perihermenias Opus I Questions on the Perihermenias Opus II (II.1 = book 1, II.2 = book 2) Questions on the Metaphysics. (If not otherwise indicated, all references will be to QMet IV.1) Questions on Sophistical Refutations

External Evidence External evidence includes how the text was produced, including scribal identification, contemporary references or allusions to the work. There are three lines of external evidence commonly given for the authenticity of Scotus’s logical works. (1) The manuscript tradition. According to Andrews (2004), eight of the thirteen manuscripts of QPer I and five of the six manuscripts of QPer II attribute the works to Scotus. However, most of these manuscripts date from the 15th century. The earliest, Ms Bruxelles Bibl. Royale 2908, is tentatively dated around 1325, and is one of only two that contain QSE. Of the three Ars Vetus works, the only one that is complete is QPor. QCat is missing questions after Q12 n.5, and does not contain QPer at all. Consequently, this cannot rule out an early misattribution. Because of Scotus’s early and unexpected death, few of Scotus’s works had any official circulation during his lifetime. Probably as a result, works by other authors were mistakenly attributed to him, in a period where manuscripts were often circulated without the author’s name, and where it was not uncommon for an editor, on finding a work whose teaching agreed with or resembled that of a Master, to attribute it to him. This could conceivably have happened between Scotus’s death in 1308 and the earliest dating of the Brussels manuscript. (2) A citing by Adam Wodeham. There are two early citings of the questions in the Lectura Secunda of Adam Wodeham, dated to about

58

On the Authenticity of Scotus’s Logical Works

1330.10 Again, this does not rule out an early misattribution, or the possibility that the text that Wodeham was referring to was not our text. (3) Summaries by Antonius Andreas. Antonius was a follower of Scotus who promotes his work extensively. He was born in Catalonia (then called Aragon) in about 1280.11 He studied in Paris from about 1300 until at least 1307, and is supposed to have been a student of Scotus while there. Antonius promoted Scotus’s work, unsuited to teaching on account of its notoriously obscure language and style, by publishing summaries, or more accurately, redactions of Scotus’s text, omitting passages or whole sections that were prolix, or confusing, or obscure, occasionally omitting parts that seemed inconsistent with what he perceived as the ‘true’ Scotistic doctrine, often changing the section order by including some of the main arguments in the reply, in the style of Thomas Aquinas. Redactions of the Questions on Porphyry, the Categories and the Perihermenias are included in Antonius’ book on the Ars Vetus,12 and we can possibly date these back to 1312, following I. Vázquez Janeiro’s discovery in 1981 of a reference to Antonius’ logical work (either the quaestiones ordinariae or the commentary on the Ars Vetus) in a manuscript dated 1312. On the assumption that the reference is to the Ars Vetus commentary, this places the existence of these works within four years of Scotus’s death at the earliest, and at the very latest, the late 1320s. For some time, the existence of these redactions carried a great weight, probably too great a weight, in support of the authenticity of the logical works. For example, Andrews claims13 that they indicate how old the tradition is, “since Antonius is one of the earliest and, on balance, one of the most faithful followers of Scotus”. To deal with Antonius comprehensively would be a separate work of great complexity, but it is sufficient to note that the redactions prove at most how old the tradition is, 10

Adam Wodeham, (ed. Wood and Gál 1990), I d.1 q.14, p. 187: “Item, hoc idem, scilicet quod res extra est extremum propositionis, licet non ut extra, tenet ipse [Scotus] super I Perihermenias, quaestione 2”; ibid. I d.22 qu. Unica, p. 283: “Aliter respondet Scotus, quaestione 2 Super primum Hermenias… ”; ibid. I d.22 qu. Unica, p. 287: “Sic enim exponit Scotus illas auctoritates, quaestione 2 super I Perihermenias” (I d. 22 qu. unica, ed. Wood-Gal, 287). 11 Gensler 1998 p.2. 12 Venice 1508. 13 Andrews 2004, p.32.

Edward Buckner

59

and that the redacted works existed at that time. Their existence does not prove they were by Scotus, nor that Antonius faithfully redacted them. The summaries never indicate which parts are by Scotus, and which are by Antonius or others. At the end of QCat14 he says that he is following the words of Scotus, but then he says a similar thing15 in his closing remarks at the end of his Expositio of Aristotle’s Metaphysics, which is little more than a paraphrase of Aquinas’s commentary on the Metaphysics, adjusted where necessary to conform to Scotistic principles16. We do not know that the same kind of misappropriation may not have happened with the Ars Vetus works. The existence of the redactions at such an early date does not of itself prove that they are redactions of Scotus’s work.

Internal Evidence Internal evidence includes similarity of ideas, stylistic features, references to the author’s own work, reference to opponents’ work, and doctrinal consistency. Modern critical studies assign more weight to external evidence and physical production of the text than to internal evidence. There is persuasive internal evidence that the same person wrote the three Ars Vetus questions (QPor, QCat, QPer). For example, in QPer I n.7, the author says that a ‘science’ is said to be of many things, of which none is the principal subject,17 and refers to question 3 of QCat, where he says the same thing in almost exactly the same words.18 In QCat 42 “Whether in a negative proposition the subject stands for a being or a non-being indifferently”19 he argues that in every affirmative or negative proposition, 14 Haec de dictis magistri fratris Ioannis Duns, natione Scoti, sedentis super cathedram magistralem, ut potui, colligens, in unum compilavi (“These are from the sayings of master brother John Duns, of the Scottish nation, occupying the master’s chair, as far as I have been able to collect them into one [work]”), Pamplona, Biblioteca de la Catedral, ms. 6, fol. 87rb, cited in P. Sagues Azcona 1968 3-19, esp. p.4. 15 He says you will frequently find Scotus’s words in his work, “just as the words of the scriptures are found handed down by him” Vives 7 p. 600. 16 Antonius’s commentary on the Metaphysics is “nothing more than a revision of Aquinas' work carried out according to Scotistic criteria,” according to Pini 1995, p. 380. 17 De multis dicitur scientia esse, quorum non quodlibet est principale subiectum. 18 Nam circa multa laboratur in scientia, non tamen propter hoc quodlibet illorum est principale subiectum in scientia. 19 Utrum in negativa subiectum stet pro ente et pro non-ente indifferenter.

60

On the Authenticity of Scotus’s Logical Works

the subject supposits for its ‘per se supposita’, and says he will perhaps say something about this in [his] book on the Perihermenias. As promised he deals with the subject comprehensively in QPer I q.11. There are references in QCat back to QPor, such as when he mentions something he wrote in his book on Porphyry.20 Such cross-references, together with the fact that the three works are always found together, strongly suggest they are by a single author. But the question is whether this author was Scotus. Also, there are no explicit cross-references that connect the three Ars Vetus works with QSE, and the manuscript tradition for QSE is also different, as noted above. Connecting the Ars Vetus works to the mature ‘theological’ works is problematic for a number of reasons. There are pronounced stylistic differences between the early and the later works. Scotus’s mature style is magisterial and confident, his inventive terminology almost a parody of the scholastic style (“The second conclusion about the first effective thing is this, that the simply first effective thing cannot be caused … The proof is that it is an in-effectible independent effective thing”). The structure and division of the theological works is a complex web of nested syllogisms, digressions, corollaries and lemmas, stretching over many pages. By contrast, the style of the logical works is simple, almost approachable, and the structure is the standard scholastic Quaestio format of main arguments for, arguments against, resolution and replies to the main questions. The difference between the early ‘Aristotelian’ works and the later theological work for which Scotus became famous is the difference between a village church and a gothic cathedral. More generally (although perhaps unsurprisingly, given the difference of subject matter), there is a lack of interest in theology in the Questions, whereas theology infuses Scotus’ mature work. God (‘the divine’) is mentioned only once in the Questions and that in a question which is a closely worded imitation of Thomas Aquinas’ discussion of truth and falsity in his commentary on the Perihermenias (Book I, lect. 3). Likewise, there are only two references to God in QCat, and none in QPor. More problematically, the logical works differ doctrinally from the later theological works by Scotus on two substantial points. First, Scotus posits a complete equivocation among being in the different categories in QCat, 20

“De hoc dictum est in libro Porphyrii” QCat 3.13.

Edward Buckner

61

whereas the mature Scotus was a proponent of the univocity of being. Second, in the discussion of the sea-battle problem in QPer (Opus II.1, qq.7-9) there is no suggestion of the ‘synchronic’ conception of modality, a highly original thesis allowing for alternative possibilities at a given time, which Scotus advanced in his Lectura.21 In summary, there is an apparent lack of internal evidence for the authenticity of the logical works, and apparently strong evidence against.

Univocity of Being The most serious challenge on internal grounds is the inconsistency between Scotus’s early position on univocity and his mature position. In the fourth question of QCat, he argues that ‘being’ is an equivocal term for a logician because no single reckoning (ratio) belongs to it, but rather several, one for each of the ten Aristotelian categories. This reflects the view commonly held by scholastic theologians, namely that if ‘being’ were univocal, then it would be a kind of genus of which the categories were ten separate species. But, as is well known, Scotus’s mature position is the very opposite. In his Sentences commentary22 he argues that the concept of being must apply univocally (i.e. in the same sense) to all kinds of things: to God and creatures and to the ten Aristotelian categories. He was apparently the first to take this position, and it is considered one of his most important doctrines.23 According to Gilson, Scotus’s explicit rejection of univocity in his logical works “has been the despair of generations of Scotists,”24 as well as an embarrassment. Pini mentions several marginal notes to Scotus’s solution contained in the manuscripts, such as “this is not the view of the Doctor as is clear to anyone who diligently considers [the matter].” 25 Either Scotus is the author of QCat, in which case there is the difficulty of explaining this obvious inconsistency on a position for which he became famous, or he is not, in which case he probably wrote no logical works, which is an 21 See Vos 2006 (p. 34) and Normore 2000 for discussion of the relation between Scotus’ early and later views on modality. 22 Ord. I, d. 3, pars 1, q. 3, nn. 131-166, VC 3, 81-103. 23 Pini: “Scotus is the first and most important upholder of the doctrine of the univocity of the concept of being.” 24 Gilson 1927, p.105. 25 Haec non est opinio istius Doctoris sicut patet diligenter consideranti - QMet. IV.1, 315, ll. 12-17.

62

On the Authenticity of Scotus’s Logical Works

embarrassment, and a difficulty for those who want to understand the background to Scotus’s mature thought. However, thanks to some recent sleuthing by Georgio Pini,26 further developed here, the inconsistency may have a simple resolution. There is strong evidence that Scotus simply changed his mind between the early logical works, and the later QMet and the theological works. Pini notes the similarities between QCat 4 and Question IV.1 of Scotus’s QMet, particularly when later additions are removed. The extant version of QMet looks to be a conflation of two different drafts which the scribes assembled into one composite version. When the first draft is reconstructed it becomes clear that his position on univocity in QCat was the same as the position of the first draft QMet, which thus represents a halfway house between his early position, and his later and opposing position. Scotus’s revision of the question, if it had been completed, would have probably resulted in a full endorsement of univocity. This accounts for the curiously mixed character of QMet. IV.1. As it is now, QMet. IV.1 combines a first draft in favor of the equivocity of being and the first elements of a defense of univocity. The contrast simply reflects a wholesale redirection of thought that is not uncommon in the history of philosophy – Heidegger and Wittgenstein are only two examples. Pini’s discovery removes a significant and previously compelling objection to the authenticity of the logical works. Further comparison of QCat and QMet reveals similarities which are compelling positive evidence. First, a surprising number of the arguments that appear in QMet are present in QCat. Pini notes that ten of the main arguments for the univocity of being in QMet (nn. 1, 5-7, 9-12, 14, 16) are in QCat. There are also another two (3 and 15) if we include supporting arguments. Two of the four main arguments against univocity are in QCat (17, 20). Of Avicenna’s arguments for univocity, which Scotus deals with separately in QMet, one (nn. 31, 35) is in QCat 4 (14, 13). Avicenna’s arguments for univocity are followed in QMet. by seven objections against this position (nn. 50, 54, 57, 62-68), of which five (nn. 50, 54, 57, 62, 64) are in QCat 4 (nn. 18, 19 second half, 20-22, 24, 19 26

Pini 2005.

Edward Buckner

63

first half). (In QMet. IV.1, Scotus answers two of these arguments (nn. 63, 65), which in QCat. 4 remained unanswered.) The solution is similar in both questions (QMet n. 70, QCat. 4, nn. 37-38): being is equivocal for the logician and analogous for the metaphysician. Of the replies to the main arguments for equivocity (QMet nn. 71-85), 8 out of the 11 replies (nn. 76-81, 84, 85) are in QCat. 4 (nn. 40, 41, 43, 44, 47, 50, 48-49, 52). Of the three replies to Avicenna in QMet 86-90 two are in QCat (53, 54). The close parallel between the arguments of the different works is evidence that the author of QCat is the same as the author of QMet, who is unquestionably Scotus. To be sure, it cannot entirely be ruled out that this is an accident. Scholastic writers often copied the formulation of questions from each other, and the same question in the hands of different writers sometimes looks similar. But the close correspondence between multiple questions makes it difficult to explain as an accident. To take just one example, QCat. n. 3 has “Again, ‘a man is a substance, therefore a man is a being’ is valid, because the antecedent cannot be true without the consequent, but in equivocal [terms] there is no consequence;” QMet n.6 has “Again, ‘this is a substance, therefore [it is] a being’ is valid. The antecedent cannot be true without the consequent; in equivocal [terms] there is no consequence.” The parallel wording across many arguments and their replies is difficult to explain as a coincidence. Even the differences in form suggest a natural development of thought. In QCat n.12, the author argues that the ‘first principle’ (i.e. the principle of contradiction) is the most certain of all principles, and that if ‘being’ were not univocal, it would not be certain because of the ‘multiplicity of the terms’. In his reply (QCat. n.12 reply at n.52) he invokes Aristotle’s maxim that we know principles insofar as we know their terms,27 and so the truth of absolutely certain principles such as non-contradiction must be known so long as the terms are known, which is consistent with the terms being equivocal. In QMet., Aristotle’s maxim is not given in the reply but in the main argument itself,28 i.e. what in QCat is seen as part of an 27

Posterior Analytics Book I 72 b23, translated by James of Venice as “principium scientiae esse quoddam dicimus, in quantum terminos cognoscimus.” 28 QMet. n.16 “principium notissimum habet terminos notissimos, quia principia cognoscimus inquantum terminos cognoscimus.”

64

On the Authenticity of Scotus’s Logical Works

objection to univocity, now appears in support of it. Furthermore, in QMet he adds a confirmatio or corroborating argument: all principles which involve common conceptions, involve ‘transcendental’ terms, i.e. terms which lie outside the ten Aristotelian categories. But all common conceptions would be doubtful if ‘being’ were equivocal, which (we must infer) is absurd. This is remarkably similar to an argument he gives in the Ordinatio (I d.3 p.1 q.1-2 n.27), where he argues that an uncertain concept [intellectus] such as God’s being finite is composed of the certain concept of God’s being and the uncertain concept of God’s being finite. Since the former concept is certain, and since it is included in the uncertain concept of God’s being finite, it follows (according to him) that being is univocal. Underlying this is the implicit assumption that without univocity there can be no certainty, explicit in QMet n.16, implicit in QCat. Another development is suggested by the main argument against univocity in QCat n.16. “Porphyry, in the chapter on species, [says] that if a person should call all things beings, yet he will call them, so he says, equivocally, but not synonymously.”29 He misquotes Porphyry, who includes the word ‘says’ (inquit) because he is citing Aristotle: Porphyry is saying what Aristotle is saying. In QMet n.17 Scotus gives the same argument but this time includes ‘says’. Finally in Ordinatio I (D3 Q3 n.155) he gives the argument again, but this time in his reply (n.165) asks “Who is saying inquit?” Aristotle, of course, who Porphyry is speaking of. But this saying is not found in Aristotle’s logical works, says Scotus, but in the Metaphysics.30 Other arguments in QCat which are later developed in the theological works are: the claim that being is not a genus (QCat n.18, Ord. I D3 Q3 n.152); that metaphysics is a single science (QCat n.10, n.48, QMet n.14, I D3 Q3 n.153); that ‘a substance is a non-entity’ ‘cannot be true (QCat n.4, QMet n.1, I D3 Q3 n.156); that if being were univocal, it is either trivial to

29

Si quis omnia entia vocet, aequivoce nuncupabit. This may support Pini’s contention, following Donati, that there is a specifically English tradition, which combine the equivocity/analogy relationship with the difference between logic, on the one hand, and metaphysics and physics, on the other hand. Geoffrey of Aspall, writing about 1250-70, was one of these authors (Donati (2003) cites Galfridi de Aspall Quaestiones super Physicam, I, q. 26, ms. Oxford, Merton College, 272, f. 91ra), others were William of Chelvestun and William of Bonkes. 30

Edward Buckner

65

say of any existing thing that it exists, or contradictory to say it of a nonexisting thing (QCat n.20-22, I D3 Q3 n.157).

Scotus’s Questions on the Perihermenias Contemporary discussion of the relation between Scotus’s logical and theological works has focused on the question of univocity, and thus on the relation between the QCat and the later works. There has been relatively little work on the curious and obscure Questions on the Perihermenias (QPer).31 Scotus’s questions on the Perihermenias exists in two versions, Opus I (QPer I) and Opus II (QPer II). QPer I consists of only one book, corresponding to the first book of the Perihermenias. QPer II is shorter, but is on both books of the Perihermenias. It is difficult to say which came first. The traditional order, given in Wadding’s classic 1639 edition and in the later edition by Vives, compiles QPer I with the second book of QPer II, then follows with the first book of QPer II, the latter called the Quaestiones octo or ‘eight questions’. Both versions deal with topics we would now consider as belonging to the intersection between metaphysics and the philosophy of language: reference and signification, existence and essence, and truth and its connection with reality.32 The questions are loosely based on problems arising from Aristotle’s text, although many of these topics were also discussed in other genres, such as sophismata literature, during the second half of the thirteenth century. This is especially true of questions relating to the semantics of time, existence, and quantification. It is not known when they were written, but see the section below for an approximate dating. It is not clear why there are two versions, nor why QPer I consists of only one book. In the response to the second main argument of Q 9 (n32) of QPer I he says “perhaps, as we shall see afterwards, the content of the verb is not part of the predicate,” 33 apparently referring forward to the reply to 31

But see Marmo 2013, pp.250-1, for an overlapping discussion. Buckner and Zupko 2014, p.13. 33 Vel forte (ut videbitur post) res verbi non est pars praedicati. See also Q8 n42: “Quamvis haec responsio includat difficulatem de re verbi quando praedicatur 32

66

On the Authenticity of Scotus’s Logical Works

Q5 of QPer II book I (n29), which would be impossible if QPer II had been intended as a completely separate work, and suggests that QPer I originally included a second book, now lost. Alternatively, the first book of QPer II may have been intended as part of QPer I, although the manuscript evidence contradicts this: the two books of QPer II are always found together. There is no version containing QPer II on its own, whereas a number of versions contain only QPer I, suggesting that QPer I was the first to be attributed to Scotus, consistent with the fact that the versions containing QPer II, which are mostly 15th century, are later than the earliest version of QPer I (mid 14th century?). This suggests QPer II was discovered, or attributed to Scotus later.34 Nearly all the manuscripts containing the Perihermenias questions also contain the other two books on the Ars Vetus.35 Questions 12 and 13 are doublets for some of the questions in QPer I. Question 12 is a repeat of question 7 (whether ‘Caesar is a man’ is true or not). Question 13 is a repeat of question 9 (whether the predicate in a present tense proposition is restricted to denoting only objects existing in the present). However, the responses to the questions are different. The author of question 7 holds that ‘Caesar is a man’ (with Caesar not existing) is true, the author of question 12 that it is false. The author of question 9 holds that the predication is not restricted, the author of question 13 that it is.36 Note that while there are significant overlaps, there are arguments in 7-8 which are not addressed in question 12. If question 12 represents a later position of the author of 7-8, why did he not address some of the arguments that he was apparently persuaded by earlier? It is unclear why some questions are duplicated in QPer II, others not. The duplicated questions are question 2 of QPer I (Whether a name signifies a tertium, si sit pars praedicati (de quo forte dicetur post).” Vos 2006 p. 128 cites Bos 1987, p.123, as evidence that Opus II was later than Opus I. However, the reasoning is faulty: the reference in Opus I could easily be to a later version of Opus II, now lost, and so the surviving text of Opus II could be earlier. 34 Against, Antonius includes both in his digest. 35 E, L, N, R, A, J, K, Q, S, U, V, W. Those not: O, V. 36 See also Marmo 2013, who notes similar inconsistencies.

Edward Buckner

67

thing, or a species in the soul) and question 1 of QPer II.1 (Whether a name signifies a thing or an affection); question 3 of QPer I (Whether, given a transformation of the thing which is signified, there occurs a transformation in the signification of the utterance), and question 2 of QPer II.1 (Whether a name univocally signifies a thing, whether the thing exists or not). There is no other duplication. I.e. if QPer II were another version of QPer I, why does QPer II omit the interesting questions about empty terms, and no-longer-existing supposita? Questions 9-11, on whether a common term is restricted, are not found in QPer II (though they anticipate questions 7-9 of that work, on the problem of the sea battle). Conversely, why is there so little overlap between the other questions in Opus II and any questions in Opus I? A further curiosity is that some of the material in Antonius’ digest is missing from Scotus’s version. Questions 17 (“Whether the mode is predicated in modal propositions”) and 18 (“Whether modal and assertoric propositions are of distinct species”) in Antonius are missing completely, and, curiously, a fragment of Antonius’ Question 9 (“Whether a sign [of quantity] can be given on the side of the predicate”), appears verbatim in QPer I, Q.12, nn.17-18. This suggests that either Antonius took this fragment and turned it into a whole question, which seems unlikely, or that Antonius was working from a different source that included the whole question, or most of it. If so, it is an interesting question whether Questions 17 and 18 are authentic but undiscovered logical works of Scotus. The work is interesting because, more than with Scotus’ mature work, there is a clear and demonstrable connection to late thirteenth-century scholastic thinking about logic and semantics. The influence of modism is evident throughout. Modism was a school of thought that flourished in the second half of the thirteenth century, and which found its way to Oxford in the 1280s, when Scotus was probably a student there. At its heart is the principle that the way or ‘mode’ in which language signifies closely reflects the structure or ‘mode’ in which we think or understand, and that the structure of thought in turn reflects underlying reality. As Thomas of Erfurt expresses it in one of the works originally misattributed to Scotus, “Every mode of signifying is from some property of the thing.”37 It is

37

De modis significandi, 2.4; Bursill-Hall 1971.

68

On the Authenticity of Scotus’s Logical Works

uncertain, as Andrews has argued,38 whether the early Scotus agreed with the modists on every point,39 although Andrews argues that modism was transmitted to Oxford in the late thirteenth century by Andrew of Cornwall and propagated by Simon of Faversham to students such as Scotus.40 The focus of the work is almost exclusively on questions that fascinated the modists, such as whether identity statements like ‘Caesar is Caesar’ are true even when Caesar no longer exists, which has counterparts in Parisian and Oxonian discussions from the 1250s onwards, about whether a name signifies primarily something in reality, or a mental idea.41 A number of other logicians writing around the same time as Scotus was at Oxford42 were interested in the kinds of proposition (such as ‘a man is an animal’ or ‘a triangle has three angles’) which could be true even when no men or triangles exist. Scotus frequently appeals to the modist maxim “significare praesupponit intelligere” – “signifying presupposes understanding”. “Sicut igitur se habet ad esse, ita ad intelligere et significare” says Andrew of Cornwall43 – just as something is related to reality (esse), so it is related to understanding and signifying. Scotus holds, like Peter of Cornwall and other English writers influenced by Modism, that language signifies reality as it is understood.44 In QPer I, Q.2, he argues that a name signifies a thing, rather than an idea in the mind, but not as it exists in reality, but as it is 38

Andrews 2003. Pini, based on arguments in QCat, rejects the view that Scotus was influenced by modism, arguing that he rejects the possibility of a logico-semantic investigation of reality, and rejects any parallelism between modes of signifying, modes of understanding and modes of being. However, this is not supported by Scotus’s position in QPer I and II. 40 Andrews 1999. 41 See e.g. “Utrum rebus corruptis termini cadant a suis significationibus?” [Whether when the things (falling under terms) are destroyed, the terms fall away from their signification]” (NL 16135 54va-55ra, De Libera 1991 pp187-191). 42 Including Anonymous Wigorgensis. 43 Andrew of Cornwall, QPor q.12, f. 89va, quoted in Andrews 1999, p.112. 44 For a later example of this, see Lectura. I, d. 3, pars 1, q. 3, n. 191, VC 18 p. 302. “Quidquid igitur convenit rebus secundum esse intelligibile, convenit eis naturaliter respectu lucis increatae; sed res ut obiecta intellectui, inclinant intellectum naturaliter in esse intelligibile; et sic inclinare intellectum convenit essentiae lapidis vel aliorum in quantum sunt a Deo ut est lux quaedam, a qua prius sunt secundum esse intelligibile quam a voluntate”. 39

Edward Buckner

69

conceived by our understanding, which, as becomes clear, is the essence of the thing as signified by the definition of the name which signifies it.45 In reply to Q.3, he says that when the thing signified is destroyed, so is the sign (n.13). But even so, the thing is not destroyed as it is understood; thus, the significate of the utterance is not destroyed.46 Between existence in reality and existence in the understanding there must be some nature that remains the same, and which underlies the essence of reality, our understanding of that reality, and the signification of the language expressing that reality. This does not mean that Scotus is committed to the subjective idealist thesis that language signifies our understanding of reality, or our thought about reality, or concepts or ideas. Language, for the modist, signifies reality. But it signifies it as understood: reality as qualified by our understanding of reality, as Scotus makes clear in the extended argument of QPer questions 2, and 5-11, and much of the rest of that work. He says (n.44) that a nature conceived as ‘this’ has the reckoning (ratio) of the suppositum equally when it exists and when it does not. Later, in the Ordinatio he says that the relation between the common and the singular is not between being in the mind and being in reality (as Ockham later argued), because both the commonness and the singularity of a nature are outside the mind.47

45

This is in his second reply to the question (n.39) which, as we argue in the commentary below, is consistent with his later replies to QQ.3 and 5-8. 46 There are some similarities with the contemporary debate about whether language has a ‘semantic dependency’ or ‘object-dependency’ on external reality. Authors such as Nathan Salmon (Soames and Salmon 1988, p. 241 and passim) have argued that the semantic contribution of a proper name or demonstrative term is simply the referent of the term. Thus, a term’s signification is intrinsically linked to the existence of the objects it signifies or denotes, so that when the objects are destroyed, the signification is changed or destroyed. The position is known as ‘direct reference’ or ‘Millianism’, although it applies only to proper names, whereas in the thirteenth century, it applied to both common and proper names. This is the view that Scotus, along with most of his contemporaries, opposed. A name, including a proper name, retains its significance even when the object that falls under it is destroyed. Thus, it cannot have ‘semantic dependency’ in the modern sense. 47 Ordinatio II dist. 3 P1 Q1 n.42, VC7 p.410: Ad confirmationem opinionis patet quod non ita se habet communitas et singularitas ad naturam, sicut esse in intellectu et esse verum extra animam, quia communitas convenit naturae extra intellectum, et similiter singularitas.

70

On the Authenticity of Scotus’s Logical Works

This reflects a position, as well as the terminology, of other English modists of the period. Peter of Cornwall48 says that utterance primarily signifies a thing, although not absolutely, but rather ‘as it is apprehended’. He explains that a thing ‘as it is apprehended’ is not the same as a concept or intelligible species, because a thing is something in external reality. An anonymous writer, probably English and writing in the 1280s, mentions the argument that when a man has been corrupted, the species corresponding to a man remains in the understanding, so the thing in reality is not the cause of truth, but the thing as understood. He also says that ‘every man is an animal’ is necessarily true, because even when no man exists, the intellectual species corresponding to the term ‘man’ remains in the understanding.49 Given this connection to currents of thought in late 13th century Oxford, it is interesting how at least one of the questions in QPer seems to anticipate one of Scotus’s most well-known doctrines, namely the idea that a specific nature (e.g. man) has its own unity that is in some sense less than numerical unity, and that a specific nature is made individual (i.e. this man, Socrates) by the addition of some positive individuating factor, which Scotus calls the individuating difference (differentia individualis),

48

Probably writing before 1272, according to Lewry 1981 p. 246. Dico quod si vocamus intellectum speciem vel similitudem rei, quod vox non est primo modo signum illius [24] intellectus, et si hoc significet aliquo modo hoc est aequivoce; sed primo et principaliter significat rem, non tamen significat rem simpliciter sed secundum quod apprehenditur. “I say that if we call a concept (intellectum) a species or a likeness of a thing, an utterance is not primarily a sign of that concept, and if it does signify this in any way it is equivocal. Rather, primarily and principally it signifies a thing, yet it does not signify a thing absolutely, but as it is apprehended.” Peter of Cornwall, Omnis homo est; ms. Worcester 13, 49va, in Ebbesen 1987 p. 150, my translation. 49 Anonymous sophisma in Caius 611/341, f 59ra, quoted in Lewry 1981 p.268. Res non quocumque modo est causa veritatis, sed res [ut/unde] intellecta. Tunc ex istis suppositionibus argue sic: manente causa manet effectus; cum res secundum quod intellecta est semper manet, ergo cum propositio sit quidam effectus rationis, semper manebit eadem veritas in oratione. “The reality [res] is not the cause of truth in just any way, but rather the thing as [unde] it is understood. Then from those suppositions argue as follows: with the cause remaining, the effect remains; since the reality according as it is understood always remains, therefore since a proposition is a sort of effect of reason, the same truth will always remain in speech.”

Edward Buckner

71

or haecitas – ‘thisness’ or haecceity.50 In QPer I Question 6 Scotus asks ‘Whether there are any supposita belonging simpliciter to a common term signifying a true nature, apart from ones that exist?’ The scholastic terminology obscures a problem which is familiar to contemporary philosophers of language. In this context, a suppositum of a common term is broadly analogous to an ‘instance’ of a predicate. A scholastic logician would have spoken of Socrates being a suppositum of the common term ‘man’, just as we speak of him being an instance of the predicate ‘is a man’.51 Scotus is asking whether there can be non-existing supposita of a common term, i.e. asking whether instantiation is equivalent to existence. This is broadly analogous to the question, still hotly debated, of whether existence is a ‘thin’ conception, expressible in terms of quantification, or whether it is richer, as has been argued by members of the existentialphenomenological tradition.52 Scotus argues against those who would reduce existence to instantiation. After considering reasons for and against there being non-existent supposita, he maintains (n.45), consistent with his position elsewhere in QPer,53 that a thing is conceived in the same way when it does not exist, as when it does exist. He cites a passage from Aristotle’s Posterior Analytics, saying that there can be scientific demonstrations of things (e.g. eclipses) which have periodic or regular existence. By the same logic, a common 50

Also called an individual difference (differentia individualis), e.g. Ord. I d.7, n.48, VC 4 p. 128; Scotus’s followers adopted the term ‘haecceity’, which Scotus uses (as ‘haecitas’) only in QMet. 51 A significant difference is that in modern logic the predicate includes the copula ‘is’, and so the proposition has two logical parts, e.g. the predicate ‘—is a man’ and the subject ‘Socrates’. In scholastic logic, the predicate is the second term (‘man’), joined to the first term (‘Socrates’) by the verb ‘is’. This difference is not relevant to the present discussion. 52 See e.g. van Inwagen 2014, p.56. There is nothing new here. Ockham argues (Boehner 1974, Pars II c.2 p.249) that for the truth of ‘this is an angel’ it is both sufficient and necessary that ‘this’ and ‘angel’ supposit for the same thing, i.e. that there is at least one suppositum corresponding to ‘this’ and ‘angel’ and claims (Boehner 1974, Pars III-2, c.26, p.552) that “An eclipse exists” is equivalent to “Something is eclipsed,” and “Something having three angles equal to two right angles exists,” is equivalent to “Something is a thing having three angles equal to two right angles.” 53 See e.g. his reply to question 3 (n.10), whether there occurs a change in the signification of the utterance given that a change has occurred in the thing signified, where he quotes the same passage from the Posterior Analytics.

72

On the Authenticity of Scotus’s Logical Works

nature (such as human nature) conceived as ‘this’, i.e. as singular and ‘not predicable of many’ (indicibilis de pluribus), bears the same relation to the nature conceived under the aspect ‘predicable of many’ (dicibilis de pluribus), whether it is conceived of as existing or not. Therefore “it has the nature of a suppositum equally when it exists and when it does not.”54 So, he argues, (n.46) a nature understood as this is signified by terms like ‘Caesar’, whose referent no longer exists, or ‘the Antichrist’ whose referent does not exist now, but will exist in the future. Just as a general term like ‘man’ has a meaning whether or not any man exists, so a singular term like ‘this’ or ‘Caesar’ has a meaning whether or not the corresponding individual exists. Thus a suppositum is a nature conceived in the understanding as a ‘this’, and it is signified by utterances like ‘Caesar’. The suppositum is an instance of a common nature, but qua instance does not have to exist, since the suppositum corresponding to the term ‘this man’ does not have to exist; the term signifies the same whether or this man, or any man actually exists. The difference between being an individual or an instance, and being common, is something that can exist in thought as well as in reality. Being an instance of something is prior to being an existing instance. Existence is not the same as instantiation. The similarity to his later views on individuation is striking, particularly the strange grammatical distortion of the demonstrative pronoun into an adjective, as though signifying an attribute: “The common nature conceived as this.”55 Furthermore, the thesis that instantiation is prior to existence underpins the whole of his mature thought about individuation: Just as the highest point in a genus is found precisely by considering it under the aspect (ratio) of essence, so also are found the intermediate genera and species and differentiae; and the lowest point, namely the singular, is also found there, quite apart from actual existence – which is evidently clear, since this man no more includes actual existence formally than man does.56 54

n 43 Semper enim in natura communi concepta ut haec, intelligitur per se natura concepta absolute; ergo alia ab existentibus sunt simpliciter supposita termini communis. “For always in the common nature conceived as ‘this’, there is understood per se a nature conceived absolutely; therefore things other than existing things are the supposita absolutely of a common term [simpliciter].” 55 See, for example, the title question of QMet VII.13: “Whether the nature of a stone is of itself this’. 56 Ordinatio II dist. 3 P1Q3 n.63 VC7 p.419: … sicut invenitur supremum in genere praecise considerando illud sub ratione essentiae, ita inveniuntur genera

Edward Buckner

73

[…] The ultimate distinction in the categorial ordering is the individual distinction, and that exists through the ultimate actuality pertaining per se to the categorial ordering. But to that, actual existence does not pertain per se. Rather, actual existence is the ultimate actuality, but posterior to the entire categorial ordering.57

Existence is the ultimate actuality, but it is posterior to individuation, which is the ultimate distinction. Actual existence is an accident that lies outside the ten Aristotelian categories, and it is not the specific nature that is actualised or comes into existence, but rather an individual instance having the specific nature. There is a principle or individuating differentia that is prior to actual existence, and which has some kind of being qua specific nature plus individuating differentia. This is of a piece with the argument of QPer I. 6. The suppositum “is a nature conceived in the understanding under the aspect, ‘incapable of being predicated of many’,” and thus individuated. But there can be non-existing individualised natures, i.e. supposita. Existence is not the same as individuation.

The Formal Distinction The idea of a distinction which exists in reality but only as viewed through mental spectacles also reminds us of what is most interesting yet most problematic about Scotus’s later thought. The so-called formal distinction is articulated in Ordinatio II, in connection with his theory of individuating differentia. Two things a and b are formally distinct when a is really the same as b, but a has a different understanding or reckoning (ratio) from b. He says that the reality whereby we grasp the genus of something (e.g. animal) and the reality whereby we grasp its differentia (e.g. rational), from which we also grasp a specific reality (e.g. man), are realities of one

intermedia, et species et differentiae; invenitur etiam ibi infimum, scilicet singulare, omnino circumscripta exsistentia actuali, - quod patet evidenter, quia 'hic homo' non plus includit formaliter exsistentiam actualem quam 'homo'. 57 Ibid. n.65, p.420: Ita dico quod ultima distinctio in coordinatione praedicamentali est distinctio individualis, et illa est per ultimum actum, per se pertinentem ad coordinationem praedicamentalem, - sed ad hanc non per se pertinet exsistentia actualis; exsistentia autem actualis est ultimus actus, sed posterior tota coordinatione praedicamentali. See also Ord. I dist.3 Q3 n.133 VC3 p.82.

74

On the Authenticity of Scotus’s Logical Works

and the same thing, e.g. Socrates, although formally distinct.58 The formal distinction is based in reality and so external to the mind, and yet does not involve real separability. This is also anticipated by QPer I.6, in the form of the strange and puzzling identity that is not an identity: the nature conceived as singular which is identical with the nature conceived as common – and yet not identical. In one of the main arguments of Q6 (n.13) he considers the objection that something which is common cannot essentially be the same as what is singular, without existing, given that a non-being cannot be essentially the same as a true nature. Perhaps he has in mind those thirteenth-century arguments that the term ‘man’ cannot signify a dead man, for it is the nature of a man to be living and breathing and thus to exist. He replies (n.46) that the common nature, conceived as not predicable of many, i.e. conceived as this man, indeed is the same essentially as the nature conceived as predicable of many, i.e. conceived as man, but the way of conceiving (ratio concipiendi) this state of affairs is not such that it ‘falls into the union’ (cadat in unionem) created by the verb ‘is’. Earlier, in QPer I.2 (“Do Names Signify Things, or Species in the Mind?”) he considers the argument (n.15) that the proposition ‘a man is an animal’ would be false if ‘man’ and ‘animal’ signified species, because then it would be stating that one species is identical with another. In his second reply to the question (QPer I.2 n.41) he refers to the modist position that things are not signified as they exist, but as they are conceived or understood, and so the composition of man and animal in the understanding is not of intelligible species (or ‘ideas’) but of real things. However they are things as they are understood, and so there is no contradiction between the composite nature of the understanding (i.e. the fact that man plus animal is a composite of two non-identical things) and the unity that exists in reality. No matter how many things are signified by the same thing, of which one is signified insofar as it is the sign of the other, if it is compounded in an expression with the other, there is no composition of signs but of ultimate 58

Ordinatio II dist. 3, p.1, qq. 5–6, n. 188, VC7 p.484: Nec possunt istae duae realitates esse res et res, sicut possunt esse realitas unde accipitur genus et realitas unde accipitur differentia (ex quibus realitas specifica accipitur), - sed semper in eodem (sive in parte sive in toto) sunt realitates eiusdem rei, formaliter distinctae.

Edward Buckner

75

significates, which are not signs. And by the spoken assertion there is signified not the composition of species but of things, just as the composition of utterances is not signified by the written expression, but rather, the composition of things.59 See also his claim in Ordinatio I that a word is a sign of a thing and not an understanding, otherwise, given that our understanding of the predicate is different from our understanding of the subject, every affirmative statement would be false, even though the thing understood by the subject is identical with the thing understood by the predicate.60 This doesn’t altogether makes sense, for the reasons given later by Ockham in his commentary on the Sentences, namely that is impossible for things to differ formally unless they are distinguished in reality. If the common nature is in any way distinguished from the contracting individual differentia, they must be distinguished as one thing is distinguished from another thing.61 In Summa Logicae I.1662 he dismisses Scotus’s view as ‘utterly improbable’, arguing that the same thing cannot be both common and proper. If the common nature were the same in reality as the individual difference, then there would be in reality as many common natures as there are individual differences, and so none of them would be common, but each proper to the difference with which it is the same in reality. Whether it really does make sense is beyond the scope of this paper. What seems clear is that a view of this sort is anticipated in QPer. Modists say that language signifies reality as it is understood. Whatever composition exists in language must therefore exist in reality, at least qua understood. Therefore the composition of a common term with a 59

Op I .2 n33, transl. Buckner & Zupko 2014, p.33. Ord. I dist. 27 Q1 n.2, VC6 p.63: […] verbum autem exterius est signum rei et non intellectionis, - alioquin quaelibet affirmativa esset falsa in qua non praedicatur idem de se, quia intellectio subiecti non est intellectio praedicati, licet res sit res; ergo verbum est obiectum et non actualis intellectio. “but the external word is a sign of a real thing not of an intellection, – otherwise any affirmative proposition in which the same thing is not predicated of itself would be false, because the intellection of the subject is not the intellection of the predicate, although the real thing is the real thing; therefore the word is object and not an actual intellection.” 61 Scriptum in libros sententiarum Book I dist.2 Q6, Brown 1970, p. 173 and passim. 62 Boehner 1974, Pars I c.16, p.54. 60

76

On the Authenticity of Scotus’s Logical Works

quantifier (‘every’, ‘some’) or a determiner (‘this’, ‘that’) reflects a composition in reality, as far as it is understood. Perhaps there is a route from the modism of the Questions on the Perihermenias to a position for which Scotus became famous.

Differences There are also interesting differences between the doctrine of individuation outlined in QPer I.6 and the more detailed theory set out in the six questions of II.D3. For example, the idea of a ‘less than numerical unity’ (unitas minor unitate numerali) which he explicates and argues for in the first question of II.D3 is not explicit in QPer. However, there are two parts to this. The first is the negative thesis that the common nature does not of itself have numerical unity, which Scotus identifies with the kind of unity signified by the demonstrative ‘this’. For example, he says that common nature is not one of itself, for then it would be this by itself. But then anything that was a stone, would be this stone, which is clearly absurd.63 He identifies this absurdity with Plato, who supposed (he says) that an idea (e.g. man) could be a substance existing by itself, per se, in which there was the whole nature of the species man.64 The same thesis is also evident in QPer I.6. In his resolution at n.43 he says that the common nature is not attributed to a thing as it exists (i.e. to a self-subsisting entity), but rather as it is conceived in the understanding. Otherwise we are committed to Plato’s absurd doctrine of the ‘idea’, a self-subsisting common nature existing in reality.65 However, the second, positive, thesis, that the common nature has a kind of non-numerical unity, which he argues for in Ord. II/D3/P1Q1 nn.11-28, is absent. 63

II D3 P1Q1 #n1, VC7 p.392, see also QMet VII.13. Ordinatio II dist.3 P1Q6 n.160, VC7 p.470: In metaphysica sequuntur inconvenientia: Primum, quia poneretur idea quam posuit Plato. Plato enim posuit ideam esse substantiam exsistentem per se, naturam separatam, sine accidentibus (sicut sibi imponitur a Philosopho), in qua esset tota natura speciei “In metaphysics there follow absurdities. First, because we would have to suppose the Idea which Plato supposed. For Plato supposed that an Idea was a substance existing per se, a separate nature, without accidents (according to the view Aristotle attributes to him) in which there was the whole nature of the species.” 65 Ibid: … ponere aliquid existere secundum quod sibi attribuitur ratio communis est ponere ideas, sicut PLATO posuit. “.. to suppose that something exists according as the common nature is attributed to it is to suppose there are Ideas, just as Plato supposed.” 64

Edward Buckner

77

Also absent is any mention of Avicenna’s statement that ‘horseness is merely horseness, neither of itself one nor several, neither universal nor particular’,66 even though Avicenna is frequently cited elsewhere in the logical works. This is fundamental to Scotus’s distinction between universality and common-ness. For example, he says in the Ordinatio67 that commonness belongs to a nature of itself, whereas singularity belongs to it by the individuating difference which ‘contracts’ to it, but universality does not belong to a thing of itself. However, it is implicit in QPer. In the composite expression ‘this man’, the part corresponding to the common term ‘man’ cannot be singular or universal, because what is singular corresponds to the whole expression ‘this man’, and what is universal corresponds to the whole expression ‘every man’ or ‘a man’ in general. He says (QPer I.6 n.43a) that what is common, insofar as it has the reckoning (ratio) of what is common, is a nature conceived under the aspect ‘predicable of more than one’, whereas a suppositum or instance is the same nature conceived under the aspect not-predicable of more than one.68 That is, the common of itself is not singular – or universal – but only when it is conceived under a certain aspect. Similarly, in question 12 of QCat, on whether some man is a primary substance, he says that man of itself is indifferent to different senses, and can be taken for an utterance, a mental item (intentio), and for instances of actual men (supposita). But when the quantifiers ‘every’ and ‘some’ are added to it, they remove this indifference to utterances and mental items, and determine it to be taken only for instances of

66

Ordinatio II dist.3 P1Q1 n.31, VC7 p.402. Ordinatio II dist.3 P1Q1 n.42, VC7 p.410: communitas convenit ex se naturae, singularitas autem convenit naturae per aliquid in re contrahens ipsam; sed universalitas non convenit rei ex se. “Commonness belongs of itself to a nature, but singularity belongs to a nature through something in reality contracting it.” 68 Igitur terminus communis [commune] secundum quod habet rationem communis est natura prout concipitur sub ratione ‘dicibilis de pluribus’, et ita suppositum est natura concepta apud intellectum sub ratione ‘indicibilis de pluribus’. “Therefore a common term, according as it has the nature (ratio) of the common, is a nature according as it is conceived under the aspect predicable of several, and so a suppositum is a nature conceived in the understanding under the aspect ‘impredicable of several’.” 67

78

On the Authenticity of Scotus’s Logical Works

humanity.69 A common term is only truly universal or particular or singular when quantified, or when singularised by a demonstrative.

Dating The correspondence between QCat and the first draft of QMet gives a best estimate of 1295-96 for the authorship of the logical works, during Scotus’s stay in the Oxford Franciscan studium. This is because the first draft of QMet IV.1, as Pini has conclusively shown, rejects univocity, a position he embraces in the first book of the Lectura in Oxford, thought to have been written around 1297-98. This suggests the first draft of QMet IV.1 was slightly earlier, and therefore the logical works slightly earlier than that, at least assuming that the close correspondence between QCat 4 and QMet. IV.1 is explained by correspondence in time. Other estimates vary. Gal, quoted by Andrews,70 thinks they were written in Paris “where Antonius Andreas could have audited them, and thus around 1295-98.” Andrews, in the same paper, considers them to have been written in England around 1281-87, citing Brampton 1964 p.18 and Wolter 1993 p.9. Elsewhere, Wolter suggests that they were completed in the Parisian years, on the evidence of Antonius’ claim that his own logical 69 Andrews 1999 Q12 n13 p.358: Sed propter auctoritatem Aristotelis intelligendum quod ‘homo’ de se est indifferens ad multas acceptiones, scilicet pro voce, intentione, et pro suppositis. Signa ergo, ut ‘omnis’, ‘aliquis’ sibi addita, indifferentiam ad intentiones et ad vocem tollunt, et determinant ipsum ad acceptionem tantum pro suppositis. “But because of the authority of Aristotle, it must be understood that ‘man’ of itself is indifferent to many acceptances, namely for utterance, [mental] intention and for supposita. Therefore signs [i.e. of quantification] added to it, such as ‘every’, ‘some’, remove the indifference to intentions and to utterance, and determine it to the acceptance only of supposita.” See also Ord. I, dist. 2 Q3 n.183, VC2 p.238: Exemplum: cum dico ‘universale’, obiectum conceptum est pluralitas, sed modus concipiendi, id est modus sub quo concipitur, est singularitas; ita in intentionibus logicis cum dico ‘singulare’, quod concipitur est singularitas, sed modus sub quo concipitur est universalitas, quia quod concipitur, ut concipitur, habet indifferentiam ad plura. “For example, when I say ‘universal’, the object conceived is plurality, but the mode of conceiving, i.e. the mode under which it is conceived, is singularity, so in logical intentions when I say ‘singular’, what is conceived is singularity, but the mode under which it is conceived is universality, because what is conceived, as [my emphasis] it is conceived, has indifference to several.” 70 1999, “Reception of the Modistae” p.105.

Edward Buckner

79

writings are based on what he learned from Scotus at Paris.71 Vos, however, argues that this is unlikely.72 “The fact that Antonius witnessed him lecturing on such questions in Paris, however, does not prove that he wrote them during that period.” At the end of his commentary on Aristotle’s Categories, Antonius states that he is reporting the words of Scotus when he was “sitting on the Master’s chair” (sedentis super cathedram), presumably at Paris.73 However, as Pini notes, Antonius may just be saying that he prepared his work when Duns Scotus was a regent master. The material used for his redaction almost certainly originated from what Scotus taught previously, given the close correspondence between the texts attributed to Scotus, and Antonius’s work, and given Antonius’s excuses elsewhere for what may seem like plagiarism, this careful qualification may have been anticipating deflection of such a charge. Problems remain. There is the difficulty of why there are two separate versions of QPer, and also whether the second version is authentic. The arguments above establish conclusively that QPer I qq. 1-11 are authentic. There are doubts about the inconsistencies between the doublet questions 12 and 13 and the rest of the work.74 There is also the difficulty, noted above, of the ‘synchronic’ conception of modality which appears in Scotus’s mature work, and the position advanced in QPer II.1, QQ.7-9, which I do not discuss here. But these inconsistencies would be easily explained by Scotus’s inclination to change his mind.

Overall Summary In overall summary, it is beyond all reasonable doubt that the author of three of the logical works attributed to Scotus, namely the Questions on Porphyry, the Questions on the Categories, and Opus I of the Questions on the Perihermenias, is Scotus. The attribution of Opus II of the Questions on the Perihermenias is less certain, but there is no reason to strongly doubt it. The attribution is mostly on internal grounds. The external evidence – only one reliable reference to a work, twenty years after 71

Wolter 1995 p.3. Vos 2006 p.35. 73 P. Sagues Azcona 1968, cited above. 74 Also noted by Marmo (ibid), see also Buckner and Zupko (2014) p.18, pp.254255. 72

80

On the Authenticity of Scotus’s Logical Works

Scotus’s death, without any substantive indication of what work it is, and without any firm evidence that there was not already an existing misattribution – is scant, and less conclusive than previously supposed, given that Antonius on his own is not a reliable, witness. More recent work, particularly by Pini, removes significant negative doubts caused by contradictions within the internal evidence, and there is additional positive evidence. Finally, analysis of the Questions on the Perihermenias suggests a definite connection between Scotus’s mature thesis of an individuating differentia present in both thought and reality, and currents of thought in late thirteenth-century Paris and Oxford. The attribution of the Questions on the Sophistical Refutations still lacks conclusive evidence.

References Andreae, A. Quaestiones super duodecim libros Metaphysicae, Venice 1491. —. Scriptum Antonii Andreae in arte veteri et in divisionibus Boetii cum quaestionibus eiusdem i.e. no accreditation] Venice, March 1508. Andrews, Robert (1999). “Andrew of Cornwall and the Reception of Modism in England”. In Sten Ebbesen, Russell L. Friedman (ed.). Medieval Analyses in Language and Cognition: Acts of the Symposium, The Copenhagen School of Medieval Philosophy, January 10–13, 1996. The Royal Danish Academy of Sciences and Letters. pp. 105–115. —. (1999 ed.), B. Ioannis Duns Scotus. Quaestiones in Librum Porphyrii Isagoge; Quaestiones super Praedicamenta Aristotelis, edited by R. Andrews, G. Etzkorn, G. Gál, R. Green, T. Noone, and R. Wood, Opera Philosophica 1 (St. Bonaventure, N.Y.: The Franciscan Institute Press, 1999). —. (2003). “The Modistae and John Duns Scotus’s Questiones Super Perihermeneias.” In Aristotle’s Peri Hermeneias in the Latin Middle Ages: Essays on the Commentary Tradition, edited by Braakhuis, Henk A.G. and Kneepkens, Corneille Henri, 67-83. Groningen: Ingenium Publishers. —. (2004) ed., B. Ioannis Duns Scotus. Quaestiones in libros Perihermenias Aristotelis; Quaestiones Super Librum Elenchorum Aristotelis, edited by Robert R. Andrews, O. Bychkov, S. Ebbesen, G. Gál, R. Green, T. Noone, R. Plevano, A. Traver. (St. Bonaventure,

Edward Buckner

81

N.Y.: Franciscan Institute Press; Washington, D.C.: The Catholic University of America Press, 2004). Angel D’Ors, ‘Antoine Andre’, Archives d’histoire doctrinale et litteraire du moyen age 63 (Librarie Philosophique J. Vrin, Paris 1995), 28-30. —. “Utrum nomen significet rem vel passionem in anima”, Archives d’histoire doctrinale et litteraire du moyen age, 62 (1995), p. 9. Ashworth, E.J. (1987) “Jacobus Naveros on the Question: ‘Do Spoken Words Signify Concepts of Things?”, in Logos and Pragma, ed. de Rijk and Braakhuis, Nijmegen: Ingenium Publishers. Berube, C. “Antoine Andre, temoin et interprete de Scot”, Antonianum LIV (1979). Boehner 1974 (ed.) Guillelmi de Ockham Opera Philosophica I - Summa Logicae St. Bonaventure, N.Y.: Editiones Instituti Franciscani Universitatis S. Bonaventurae, 1974. 899 p., eds. Boehner, Philotheus, Gál, Gedeon, 1915- Brown, Stephen. Bos, E.P. “The Theory of Proposition according to John Duns Scotus’s Two Commentaries on Aristotle’s Perihermeneias,” in L.M. De Rijk and H.A.G. Braakhuis (eds.), Logos and Pragma. Essays in Honour of Professor Gabriel Nuchelmans, Nijmegen, Ingenium Publishers, 1987, pp. 121–139. Brampton, C.K. (1964) “Duns Scotus at Oxford, 1288-1301” Franciscan Studies 66: 5-20. Brown, S. & Gal, G., (ed.) Guillelmi de Ockham: Scriptum in Librum Primum Sententiarum Ordinatio, Opera Theologica II, 1970. Buckner, E. Zupko, J. (2014) Duns Scotus on Time & Existence: The Questions on Aristotle’s “De Interpretatione”, The Catholic University of America Press, Sep 2014. Bursill-Hall, G. L., 1971, Speculative Grammars of the Middle Ages: The Doctrine of the partes orationis of the Modistae, (Approaches to Semantics, 11), The Hague: Mouton. De Libera, A., 1991, César et le phénix: distinctiones et sophismata parisiens du XIIIe siècle / édités par Alain de Libera. Imprint Pisa: Scuola normale superiore, 1991. Donati, S. La discussione sull’unità del concetto di ente nella tradizione di commento della “Fisica”: commenti parigini degli anni 1270-1315 ca., in M. Pickavé (ed.), Die Logik des Transzendentale. Festschrift für Jan A. Aertsen zum 65. Geburstag, de Gruyter, Berlin - New York 2003 (Miscellanea Mediaevalia, 30), 60-139. Gensler, M. “Antonius Andreae - the Faithful Pupil?”, Mediaevalia Philosophica Polonorum XXXI (1992), p. 29-30.

82

On the Authenticity of Scotus’s Logical Works

Gensler, M. “The making of a doctor dulcifluus. Antonius Andreae and his position in formation of Scotism” KAL.: Tecnología filosófica 16, Universitat de Barcelona. E.I. KAL. Universitat de Barcelona. Facultat de Filosofia, 1998. Gilson, E., Avicenne et le point de départ de Duns Scot, Orient, 1927. Gölz, B. “Die echten und unechten Werke des Duns Scotus nach dem gegenwärtigen Stand der Forschung,” in Sechste und siebte Lektoren Konferenz der deutschen Franziskaner für Philosophie und Theologie (Werl i. W., 1934). Lewry (1981) Lewry, O., 1981b, “The Oxford Condemnations of 1277 in Grammar and Logic”, in Braakhuis , H.A.G., Kneepkens, C.H., de Rijk, L.M., (ed.) English Logic and Semantics, Nijmegen 1981, 235278. Marmo, C. “Scotus on Supposition” in Medieval Supposition Theory Revisited, BRILL, Oct 2013, pp. 233-259. McDermott, A.C.S. “Notes on the Assertoric and Modal Propositional Logic of the Pseudo-Scotus,” Journal of the History of Philosophy 10.3 (July 1972). Minges, P., Ioannis Duns Scoti doctrina philosophica et theologica, 2 vols. (Quaracchi: 1930). Normore, Calvin. “Duns Scotus’ Modal Theory.” In The Cambridge Companion to Duns Scotus, edited by Thomas Williams, 129–62. Cambridge: Cambridge University Press, 2002. Pini, G. “Scotistic Aristotelianism: Antonius Andreas’ “Expositio” and “Quaestiones” on the Metaphysics”, in: Atti del Congresso Scotistico Internazionale, Roma 1995. —. Categories and Logic in Duns Scotus – An Interpretation of Aristotle’s Categories in the Late Thirteenth Century —. (2008) “How Is Scotus’s Logic Related to His Metaphysics? A Reply to Todd Bates.” In Medieval Commentaries on Aristotle’s Categories, edited by L. Newton, 277–294 (Leiden and New York: Brill, 2008), 145–182. —. (1996) “Duns Scotus’ Literal Commentary on the Metaphysics and the Notabilia Scoti super metaphysicam” (Milan, Biblioteca Ambrosiana, C 62 sup., ff. 51r-98r) , Bulletin de Philosophie Médiévale, Brepols Publishers, Issue Volume 38, Volume 38 / 1996, pp. 141-142. —. (2005) “Univocity in Scotus’s Quaestiones super Metaphysicam: The Solution to a Riddle” (Medioevo 30 2005, 69–110). Richter, V. Studien zum literarischen Werk von Johannes Duns Scotus, Munich 1988.

Edward Buckner

83

Sagues-Azcona, P. “Apuntes para la historia del escotismo en Espana en el siglo XIV”, De doctrina Ioannis Scoti, Romae 1968 3-19. Nathan U. Salmon & Scott Soames (eds.), Propositions and Attitudes, Oxford University Press (1988). Smeets, Urïel, Lineamenta Bibliographiae Scotisticae (Rome: Commissio Scotistica, 1942), 8–9. Peter van Inwagen: Existence: Essays in Ontology, Cambridge University Press 2014. Vázquez Janeiro, Isaac, (1981) Rutas e hitos del escotismo primitivo en Espana, w: Acta V Congressus Scotistici Internationalis, Salamanticae 1981, Roma 1984, p. 432-436. Vos, A. (2006) The Philosophy of John Duns Scotus, Edinburgh University Press. Wodeham, Adam, Lectura secunda in primum librum Sententarium, Gedeon Gál and Rega Wood (ed), St. Bonaventure, NY: Franciscan Institute, 1990. Wolter, Allan B. (1993) “Reflections on the Life and Works of Scotus” American Catholic Philosophical Quarterly 57: 1-36. Wolter, Allan B. (1995) Duns Scotus, Metaphysician, Purdue University Research Foundation. Zupko, J., “How It Played in the rue de Fouarre: The Reception of Adam Wodeham's Theory of the Complexe Significabile in the Arts Faculty at Paris in the Mid-Fourteenth Century,” Franciscan Studies 54 (199497): 211-225.

SCOTUS ON THE SPECIES OF QUALITY LLOYD A. NEWTON

In his recent, magisterial treatment of the demise of scholasticism, Robert Pasnau claims that the “most important issue dividing the scholastics from their seventeenth-century critics is the status of qualities.”1 He adds that “it is the conviction that the scholastics were wrong about quality, more than anything else, that fueled the wholesale rejection of Aristotelianism.”2 Given the importance of qualities to the medieval, Aristotelian project, it is not unreasonable for us to focus on this topic. Although Aristotle discusses qualities throughout his writings, the place to begin any discussion of qualities, if not the locus classicus, is in chapter 8 of his Categories. There, Aristotle appears to divide the genus quality into four species: 1) habits and dispositions; 2) abilities and inabilities; 3) affections and affective qualities; and 4) form and figure. However, as every logician knows, the immediate division of a genus is typically into two species, not four. Consequently, Aristotle’s division into four species has generated a disproportionate amount of controversy amongst his followers. Surely Aristotle, the father of logic, did not make so rudimentary of a mistake as to think that a genus has four immediate species. Perhaps there is some twofold division prior to these four species. Or, perhaps the species are subalternated to one another, such that some are contained by others. Or, perhaps quality is not a single genus after all. Arguably, the way in which a philosopher solves this sort of dilemma reveals his skills as a commentator. Some, like the contemporary J.L. Ackrill, follow Porphyry’s lead and simply assert that there are four distinct species of qualities, without any attempt to deduce “them or

1

Robert Pasnau, Metaphysical Themes: 1274–1671, Oxford: Clarendon Press, 2011, p. 401. 2 Ibid.

86

Scotus on the Species of Quality

connect them on any principle.”3 Other commentators, like Aquinas, follow Simplicius’s lead and try to derive the four species in a systematic manner. Not surprisingly, in his commentary on the Categories, Duns Scotus also addresses these aporiai with the usual subtleties that earned him his sobriquet. In order to appreciate Scotus’s solution to these issues better, it will be helpful to contrast his views with two previous solutions, namely those of Simplicius and Aquinas. In what follows, we will briefly look at Simplicius’s attempt to deduce the four species of quality from a prior twofold division. Then, we will look at Aquinas’ attempt to derive the four species in a systematic manner. Once we have examined these two attempts, along with their shortcomings, we will then turn to Scotus’s treatment of the category of quality. In doing so, we will see that Scotus’s account avoids the problems facing his predecessors’ accounts and provides a more coherent account of quality than had previously been offered.

Simplicius’s Division of Quality In his commentary on the Categories, Simplicius posits a twofold division that theoretically is prior to the four species listed by Aristotle. He writes: But since Aristotle presents a fourfold division of qualities which manifests a sort of twofold division prior to the fourfold one, it is necessary to reveal the consequences of the division, starting from the twofold one . . . . For some qualities are natural, some acquired; the natural ones are those which are by nature innate and are always present, while acquired ones are those which are brought in from outside and can be lost. Of these latter, let states or conditions [i.e., habits and dispositions] be those which differ by being longer or shorter lasting and harder or easier to get rid of. Of natural qualities, let some be potential, others actual; the former are those because of which we are said to be capable of something, while of the latter some are deeply ingrained, as with affective qualities.4

3

Aristotle, Categories and De Interpretatione, trans. J.L. Ackrill, Oxford, Clarendon Press, 1963, p. 104. 4 Simplicius, On Aristotle’s Categories 7–8, trans by Barrie Fleet, New York, Cornell University Press, 2002, p. 228.15–25.

Lloyd A. Newton

87

As one can see, Simplicius’s first division is between natural qualities and acquired qualities, where acquired qualities are identified with Aristotle’s first species, namely habits and dispositions. Natural qualities are then subdivided between potential and actual qualities, where potential qualities are identified with Aristotle’s second species, namely capacities and incapacities. Finally, actual qualities are then subdivided between those that are ‘deeply ingrained’ and those that are superficial. Those that are deeply ingrained are identified with Aristotle’s third species, namely affections, while the superficial qualities, as the name suggests, are identified with the last species, namely shape and figure. Simplicius’s solution, though, is not without its difficulties, as he himself immediately admits. For as soon as he proffers this solution, he qualifies it with the following confession: But everyone ought to observe . . . that it is quite possible for figure to be acquired; for the same thing is at one moment triangular, at another rectangular when it is remodeled by art. Similarly colour is not [always] natural, for example the paleness of the coward. It is also possible for a condition to be natural and not acquired (good and bad health, for example, which could be states since they are in the domain of nature).5

As Simplicius realizes, this division has a number of problems. A thing’s shape or figure may be either natural or acquired, such as the original triangular shape of a branch that is trimmed by the carpenter into a rectangular shape, thus falling into both halves of the initial division. The same is true of color, which may be either natural or acquired, as occurs when someone turns pale because of fear. Such counter examples might even have occurred to Aristotle, leaving Simplicius to confess that: “That is, perhaps, why Aristotle rejected the division into two.”

Aquinas’ Division of Quality Given the problems with Simplicius’s division, it is not surprising that Aquinas puts forth his own solution. At the beginning of his discussion of habits in I–II of the Summa Theologica, Aquinas provides the following derivation of the four species of quality:

5

Ibid., p. 228.36–229.4.

88

Scotus on the Species of Quality For quality, properly speaking, implies a certain mode of substance. Now mode, as Augustine says (Gen. ad lit. iv, 3), “is that which a measure determines”: wherefore it implies a certain determination according to a certain measure. . . . Now the mode of determination of the subject to accidental being may be taken in regard to the very nature of the subject, or in regard to action, and passion resulting from its natural principles, which are matter and form; or again in regard to quantity. If we take the mode or determination of the subject in regard to quantity, we shall then have the fourth species of quality. . . . But the mode of determination of the subject, in regard to action or passion, is considered in the second and third species of quality. . . . On the other hand, the mode or determination of the subject, in regard to the nature of the thing, belongs to the first species of quality, which is habit and disposition.6

Unlike Simplicius’s attempt to ground the four species of quality in a prior twofold distinction, Aquinas’ derivation is quite a bit different. That derivation is grounded in the notion of a mode, which is defined by Augustine as “that which a measure determines” (quem mensura praefigit). Why Aquinas cites the authority of the Neoplatonist Augustine to solve a difficulty in Aristotle is unclear. However, Aquinas’ insight is clear enough, especially when one bears in mind the teleological nature of substances. Stated briefly, each substance has a final term towards which it moves, and by which it is measured. If it achieves its term, that is, if is determined in the right way, then it is measured favorably, whereas if it fails to achieve its end, it is measured and found wanting. However, before deriving the four species of quality, Aquinas makes a more fundamental distinction between qualities that are found in the genus substance and qualities that are accidents. According to Aquinas: Just as a quality is that in accordance with which the material potency [potentia materiae] is determined to its substantial being, which is a difference affecting the substance, so that is an accidental quality, in accordance with which the potentiality of the subject [potentia subiecti] is determined to its accidental being, which is also a kind of difference, as is clear from the Philosopher (Metaph. v, text. 19).7

Although Aquinas does not explicitly say so here, it is safe to say that the word ‘quality’ is analogical to those qualities in the category of substance 6 7

Summa Theologica I–II, Q. 49, art. 2. Ibid. [trans. mine]

Lloyd A. Newton

89

and to those qualities that are accidental. Thus, when matter is determined, or becomes fully actualized, it is proper to say that the form that actualizes or terminates the matter is a quality. Or, to view the same thing from the point of view of logic, when a genus is contracted to a species by means of a differentia, that differentia is properly called a quality. For example, when the genus animal is contracted to the species man, the differentia rational is said to be a quality. Analogously speaking, substances are also ordered toward various sorts of accidental existence, namely towards the four different types of qualities. In each instance, the goal or termination of the subject in a certain way is said to be an accidental quality, that is, how the subject is. With this understanding in place, let us turn to the four species of quality and see how Aquinas accounts for each one. Given that a material substance is a hylomorphic composite that in turn is the subject for quantity, Aquinas begins with the last species first, viz., shape or figure. According to him, shape or figure is simply the determination of the accident quantity. For example, the figure of a human being is simply the boundary or termination of the quantity of that human. Aquinas then adds: “And because quantity, considered in itself, is devoid of movement, and does not imply the notion of good or evil, so it does not concern the fourth species of quality whether a thing be well or ill disposed, nor quickly or slowly transitory.”8 These last two disjunctive qualifications, as we will see, are considered by many to be differentia within the category of quality, distinguishing one kind of quality from another. For now, it suffices to note that both disjunctions are absent from shape or figure, whereas they will be present in the other species. In contrast to figure and shape, which is rooted in the accident of quantity, the second and third species are derived from the two primary substantial principles of form and matter, respectively. More specifically, the form of a thing is the source of its active qualities, those innate abilities that are described by the second species of qualities, while the third species are rooted in the substantial matter. For example, the form of plants would account for their vegetative and nutritive capacities, whereas the sensitive soul in animals would account for their abilities to sense and move about locally, while the rational soul would be the source of the intellectual abilities unique to man. On the other hand, the material out of which a thing is composed would account for its affective qualities, namely its 8

Ibid.

90

Scotus on the Species of Quality

abilities to affect the senses. For example, if something is made primarily out of water, this would ostensibly account for the qualities of being wet or cold. Aquinas then adds: “And therefore in both, we take into account whether a thing be done with ease or difficulty; whether it be transitory or lasting. But in them, we do not consider anything pertaining to the notion of good or evil: because movements and passions have not the aspect of an end, whereas good and evil are said in respect of an end.” Like the fourth species of quality, the second and third species do not have anything pertaining to the notion of good or evil, which as we will see, pertain properly to the first species. However, the second species has the property of being done with ease or difficulty, while the third species has the property of being either transitory or lasting. With these three species accounted for, Aquinas then turns to the final one, habits and dispositions. According to Aquinas, it is not the quantity or the matter and form that is determined by this species, but the nature of the substance as a whole. Again, Aquinas adds: For the Philosopher says (Phys. vii, text. 17), when speaking of habits of the soul and of the body, that they are “dispositions of the perfect to the best; and by perfect I mean that which is disposed in accordance with its nature.” And since the form itself and the nature of a thing is the end and the cause why a thing is made (Phys. ii, text. 25), therefore in the first species we consider both evil and good, and also changeableness, whether easy or difficult; inasmuch as a certain nature is the end of generation and movement. And so the Philosopher (Metaph. v, text. 25) defines habit, a “disposition whereby someone is disposed, well or ill”; and in Ethic. ii, 4, he says that by “habits we are directed well or ill in reference to the passions.” For when the mode is suitable to the thing's nature, it has the aspect of good: and when it is unsuitable, it has the aspect of evil.

Unlike the last three species, the first one has two specific attributes: first, this species has the property of being related to good and evil, or being disposed well or ill towards a function. Second, like affections and affective qualities, this species also has the attribute of being easily changed or difficultly changed. In some ways, Aquinas’ derivation of the four species may seem to be an improvement over Simplicius’s account. It explains the various attributes of each species, such as whether a species is easily changed or not, or whether it is long lasting or not. Likewise, it accounts for the four species by reducing them to four prior, metaphysical principles, namely the substance, matter, form, and quantity. However, these advantages do not

Lloyd A. Newton

91

solve the fundamental problem with which we began: namely, how can a genus divide into four species and not into two? It is in answer to this problem, I contend, that Scotus’s account is superior, to which we must now turn.

Scotus’s Division of Quality In his youthful commentary on the categories, Scotus raises seven distinct questions pertaining to the category of quality. Four of those questions are about the four species of quality, while three of those questions are what I will call meta-questions, viz: 1) whether Aristotle assigns four species to the genus quality; 2) whether these four species are properly ordered; and 3) whether there are other species or not. As previously mentioned, the core problem seems to be how a single, univocal genus can be divided into more than two species. Unlike Simplicius, however, Scotus does not think that there is a prior fundamental division to these four species, and unlike Aquinas, he does not think that there is some essential derivation of these species from prior metaphysical principles. In fact, on the surface, Scotus seems quite skeptical that quality is a single univocal genus at all. Scotus’s response to the first question, of whether Aristotle assigns four species to quality, is twofold. First, he argues that “Aristotle does not assign [these four to be] the species of quality,”9 since they are not subalternated to one another and since some of them can be predicated of the same thing. Of course, since they are not species, then it would seem to follow that quality is not predicated univocally of them, which is exactly what Scotus further asserts. Scotus appeals to Aristotle, who begins his discussion of quality by admitting that quality is ‘said in many ways’. The Latin is: “‘Qualitas multipliciter dicitur’, id est, multis modis.” As one can see, Scotus interprets the fact that quality is ‘multipliciter dicitur’ to mean that it is said in many modes, or ‘multis modis’. This interpretation immediately raises two questions: 1) whether ‘said in many modes’ makes quality equivocal? and if not, 2) what does Scotus mean by a mode? Scotus is clear that when quality is said in many ways, this is not to be interpreted as though quality is said equivocally, since quality does have something like species, whereas an equivocal term does not have species in any way. Secondly, Scotus differentiates a species 9

Scotus’s Commentary on the Categories, qq. 30–36, n. 35.

92

Scotus on the Species of Quality

from a mode by claiming that a species “adds an essential difference, whereas a mode adds an accidental difference” (qq. 30–36, n. 36). Presumably, Scotus has in mind something along the lines of the fact that ‘rational’ adds an essential difference to the species ‘man’, whereas ‘risible’ adds only an accidental difference. Scotus is aware, however, that his position is controversial, and so he has a second response: If someone maintains that Aristotle enumerates species of qualities and not only modes, one can say that these differentiae ‘to be easily or difficultly moveable’, ‘to be the principle of operating or not’, ‘to imply an affection of sensation’, ‘to dispose a substance externally’ are accidental differentiae [with respect] to the genus of quality, which, nevertheless, Aristotle uses for essential differentiae, since the essential differentiae are unknown; and so those forms, which are primarily under those differentiae, are four species of quality. And this way of speaking does not differ much from the prior [way of speaking], except in this, that according to the prior way [of speaking], it is not necessary to concede that those four modes essentially [per se] follow distinct species of quality, but rather are able to be in the same quality according to species. But in this second way of speaking, it is necessary at least to concede that these differentiae – although [they are] accidental – are properties in four distinct species of quality; so that just as one of these species is not said of another nor are two [species said] of the same thing, so also none of these differentiae is said of another nor are two [differentiae said] of something the same, since ‘proper’ is said universally of that of which it only is (qq. 30–36, n. 65).

As one can see, Scotus is willing to concede that the four kinds articulated by Aristotle might be genuine species of a single genus. However, just as a species is always composed of a genus plus a specific differentia in other categories, so the same would have to be the case in the category of quality. That is, if quality is a genus, and if the four kinds listed by Aristotle are genuine species, then they would need to be differentiated from one another by specific differences. According to Scotus, though, such differentiae are not provided by Aristotle in the text. At most, Aristotle provides accidental differentiae which distinguish one kind of thing from another: namely ‘to be easily or difficultly moveable’, which differentiates habits and dispositions; ‘to be the principle of operating or not’ which differentiates abilities and inabilities; ‘to imply an affection of sensation’ which differentiates sensible affections; and ‘to dispose a substance externally’, which differentiates figure and form. Conceiving the four kinds in this way, though, does not substantially differ from

Lloyd A. Newton

93

conceiving of them as modes, except in the fact that by taking them in this latter way, the four kinds necessarily follow from and are proper to four distinct kinds of qualities. As one might expect, taking the differentiae to be accidental, rather than essential, raises other problems. Specifically, if the differences are accidental, and especially if the differences are not exclusively divisive, then one runs the risk of having the same quality in multiple classes. For example, as one objection points out, the heat of fire would be both an affective quality, since it is a proper sensible, and an innate principle of acting, and thus it would be in both the second and third species of quality. Likewise, since a figure is essentially sensible, it would be found not only in the fourth species but also in the third species. To this last objection, Scotus gives what I consider to be his most insightful observation. Scotus notes: To this [objection] one can say that the species that are understood through the accidental differentiae assigned here do differ. But neither are these differentiae proper to the species unless they are taken precisely. For instance: [if] that quality is divided in this way: one [part] is an interior disposition, the other an exterior [disposition]. The exterior: figure. The interior: either a primary sensible, and in this way it is the third species. If it is an interior disposition and not a primary sensible: either it is an innate principle of operating, and thus it is the second species; or it is an interior disposition and not an innate principle of operating, and thus it is the first species. Essential differentiae of the species must always be understood through these accidental differentiae. And so it is evident how one species includes a privative differentia with respect to another species; and so the one more perfect species ought always to be taken with precision with regard to its inferiors (qq. 30–36, n. 73).

As one can see, in this instance Scotus offers a prior division of quality, analogous to Simplicius’s fundamental division of quality into two species, one of which in turn is then subdivided in such a way to account for the other three basic kinds. The most basic division, in Scotus’s account, is between interior and exterior dispositions. The exterior disposition of an object consists of its form or figure, which is the fourth species of quality. The interior disposition, on the other hand, is either a proper sensible or not. If it is a proper sensible, it constitutes the third species of quality. If it is an interior disposition but not a proper sensible, then it is either an innate principle or not. The innate principle is identified

Scotus on the Species of Quality

94

with the second species of quality, whereas the other principle is identified with the first species, namely with habits and dispositions. To divide qualities in this way would produce the following outline: 1)

Exterior disposition (4th species): a. Figure – the exterior disposition of an inanimate thing b. Form – the exterior disposition of an animate thing

2)

Interior disposition a. Proper sensible, i.e., that which causes essential change to the senses (3rd species) b. Not a proper sensible i. Innate principle of acting or being acted upon (2nd species) ii. Acquired principle of acting, i.e., not an innate principle (1st species).

In contrast to Simplicius’s view, which posited a fundamental division between natural and acquired qualities, Scotus’s seems to posit a fundamental division between exterior and interior dispositions. Although he does not develop or defend this division at length, I think Scotus is onto something by dividing quality in this way.10 Earlier in the same question, Scotus cites the Metaphysics, where Aristotle says that a “disposition [dispositio] is the order [ordinatio] of the parts into [something] having parts.” 11 Looked at in this way, quality would divide immediately into an ordering of the parts that are either exterior or interior. As is well known, having parts is proper to the category of quantity, which is divided into those that have parts relative to one another and those that have parts that are not relative to one another.12 Aristotle identifies lines, planes and solids as instances of those quantities that have parts relative to one another. Of course, there are at least two questions that can be asked about the parts that are relative to one another, viz., 1) how many parts are there, and 2) what is the order of those parts? Arguably, an answer to this first question would fall into the category of quantity, whereas an answer to the 10 The same division of qualities is advanced in a contemporary study of the Categories by Paul Studtmann, in The Foundations of Aristotle’s Categorial Scheme, Wisconsin: Marquette University Press, 2008, see especially chapter 5, pgs. 101–124. 11 Metaphysics, V, ch. 19, (1022b 1–2). 12 Categories, ch. 6, 4b 20–24 and 5a 15–20.

Lloyd A. Newton

95

second question would fall into the category of quality. That is, the order of the exterior quantitative parts which are relative to one another is what we mean by the shape or figure of a thing, which is the fourth species of quantity. What does it mean, though, to speak of the disposition or ordering of the internal parts of a substance? Immediately after Aristotle identifies disposition with an ordering of the parts, he clarifies three ways in which parts can be ordered: either with respect to “place, or as to potentiality or as to species.”13 In his Commentary on the Metaphysics, Aquinas elaborates on this second kind as follows: “Disposition is used in a second sense inasmuch as the order of parts is considered in reference to potency or active power, and then disposition is placed in the first species of quality. For a thing is said to be disposed in this sense, for example, according to health or sickness, by reason of the fact that its parts have an order in its active or passive power.”14 Aquinas’ interpretation of this passage does not seem to square with his derivation of the species of quality that he provides in the Summa. As we saw earlier, in the Summa, Aquinas identifies the first species of quality with the nature as a whole, the second species with its active powers, and the third species with its passive powers. Here in the Metaphysics, however, he identifies the first species with both the active and passive powers. Setting aside whether Aquinas is consistent or not, he seems to be onto something when he identifies a disposition as an order or arrangement of the parts with respect to potency. Scotus too has a similar notion. Even though he does not comment specifically on this passage in his commentary on the Metaphysics, Scotus does identify the first three species of quality with the principles of acting.15 This still leaves us, though, with the question of how to differentiate the first three kinds of qualities insofar as they are principles of action. Given that Scotus identifies the first two kinds as belonging to a single genus that is coordinated with the third species, I suggest that the principle difference between them is whether or not the action resides in the subject of the quality or in another subject. Stated another way, if the first three species of qualities imply an internal ordering of the parts which potentially 13

Metaphysics, V, ch. 19, (1022b 1–2). Aquinas, Commentary on Aristotle’s Metaphysics, Lesson 20, 1058. 15 John Duns Scotus, Questions on the Metaphysics of Aristotle, bk. V, q. 14, n. 90. 14

96

Scotus on the Species of Quality

produces an action, then the difference between the first two species is that they primarily produce an action in the subject in which they inhere, whereas the third species primarily produces an action in a different subject, one in which the affections do not inhere. Such an interpretation, I contend, is borne out by some of the observations Scotus makes in passing in his commentary. For example, in his discussion of affections, Scotus says that an affection “expresses a comparison [comparationem] of a quality to a subject that is apt to be altered according to it; or [it expresses] a comparison of quality to sensation, to which it implies an affection.”16 A few paragraphs later, Scotus adds that “qualities of the third species are not said to be affective because they imply affections to their subjects – for no form is apt to have an affect [agere] on its own subject – but they are said to be affective because they imply an affection to the senses in some mode.”17 For example, colors, sounds, and odors are typical examples of affections. But these qualities do not, per se, affect the subject in which they inhere, but rather another subject, namely a sentient being that is able to perceive these affections. Furthermore, if the third species of qualities are the principles of acting with respect to another subject, then it would follow that the first two kinds of qualities are principles of acting with respect to the subject in which they inhere. Of course, this is exactly the way the first two qualities are described by Scotus. In response to the question about abilities and inabilities, Scotus describes an ability as “a principle of operation” and inability as “a principle of acting difficultly or of undergoing [something] easily (n. 49). Likewise, in another question, Scotus describes habits as “a natural principle for operating easily or not easily” (n. 35). Thus, both abilities and inabilities on the one hand, as well as habits and dispositions on the other, are principles of action with respect to the subject in which they inhere. Of course, to say that abilities and habits are principles of action is hardly novel. However, to identify these qualities in terms of dispositions, that is, of an ordering of the internal, material parts of a subject may be novel. To see how this is the case, consider the more recent work done by Erwin Straus, who claims that the upright posture, or the ordering of the interior parts of a human being, is crucial for the 16 17

Scotus’s Commentary on the Categories, qq. 30–36, n. 54. Ibid., n. 57.

Lloyd A. Newton

97

various functions unique to humans. One quotation from Straus will have to suffice: “Upright posture, while unique [to humans], is also essential. . . However, there is no doubt that the shape and function of the human body are determined in almost every detail by, and for, the upright posture.”18 Or, to take only the most banal of examples, the arrangement of the parts of the hand makes the activity of holding something possible. Let us return, though, to Scotus. If both habits and abilities are principles of actions, i.e., dispositions or arrangements of the material parts with respect to the subject in which they inhere, then we are still left with the question of how to differentiate one from one another? The most obvious answer is the one provided by Simplicius: abilities and inabilities are innate to a subject, whereas habits and dispositions are acquired. Although Scotus does not emphasize the point, perhaps because it is a wellestablished fact, he does mention that both abilities and inabilities are innate (n. 52). Presumably, then, habits and disposition would be acquired. I confess, though, that the connection between natural abilities, viewed as the arrangement of the internal parts, and the subsequent activities is much easier to grasp than the relationship between habits and activities. To return to the previous example, it is fairly obvious that the position or arrangement of the thumb in opposition to the fingers makes the activity of holding an object possible. However, it is not nearly as obvious how a habit can be identified as the arrangement of the parts. Stated bluntly, what does a habit have to do with the disposition or arrangement of the parts? Admittedly, this part of the paper is more speculative, but I want to suggest that habits can be identified with the arrangement of parts if one takes these parts as related to one another sequentially and not simply spatially. Consider, for example, Aquinas’ comments on the habit of a skilled grammarian and a skilled musician: Nor does the artisan deliberate insofar as he has the art, but insofar as he falls short of the certitude of the art. Hence the most certain arts do not deliberate, as the writer does not deliberate how he should form letters. Moreover, those artisans who do deliberate, after they have discovered the certain principles of the art, do not deliberate in the execution. Thus one who plays the harp would seem most inexperienced if he should deliberate in playing any chord. And from this it is clear that an agent does not 18

Erwin Straus, Phenomenological Psychology, trans. Erling Eng, New York, Basic Books, 1966, p. 138.

98

Scotus on the Species of Quality deliberate, not because he does not act for an end, but because he has the determinate means by which he acts. Hence, since nature has the determinate means by which it acts, it does not deliberate. For nature seems to differ from art only because nature is an intrinsic principle and art is an extrinsic principle. 19

In each of the examples provided by Aquinas, the sequence of the acts, or the sequence of the parts insofar as they move through time, is arguably what Aquinas, following Aristotle, means by a habit. Consider again the musician who plays the harp. His ability to play various notes in order is simply a sequential disposition or arrangement of the parts, so to speak. Furthermore, unlike natural abilities, the habit is an ability that is acquired rather than innate. Scotus makes the same point in his Commentary on the Metaphysics, where he says that Aristotle “divides potencies into those which are innate and those which are acquired, ‘and the latter come about by practice, like flute playing, or by learning, like the artistic powers’.”20

Conclusion As we saw at the beginning of this paper, qualities play a fundamental role in Aristotle’s physics, and were, according to Pasnau, the crucial factor that played a role in the wholesale rejection of Aristotelianism by the enlightenment philosophers. The fact that Aristotle lists what appear to be four species of qualities in his Categories certainly did not help matters. If qualities are as important as Pasnau claims, and I think he is largely right about this claim, then it is imperative for his interpreters to try to find some sort of rationale for why Aristotle posits four species and not two, as one would expect of a single genus. As previously mentioned, Simplicius tried to account for the division into four species by positing a prior division into natural and acquired qualities. Not satisfied with Simplicius’s account, Aquinas offered his own derivation, where each of the four species of quality is grounded in a prior metaphysical principle, viz., quantity, matter, form, or the composite nature. Although Scotus does not, as far as I can tell, directly acknowledge Aquinas’ derivation, one could argue that it is nonetheless conspicuously absent. For instead of trying to account for why there are four and only 19 Thomas Aquinas, Commentary on Aristotle’s Physics, trans. Richard Blackwell, et. al., Notre Dame: Dumb Ox, 1999, Lect. 14, n. 268. 20 Scotus, Metaphysics, bk. IX, q. 8, n. 5.

Lloyd A. Newton

99

four species, Scotus at first claims that the four species are not really species, in the strict sense, at all. Rather, quality is simply said of a substance ‘in many ways’ or ‘in many modes’. Indeed, he admits that there may be more kinds of qualities than the four mentioned by Aristotle. However, for the rigid Aristotelian, who insists that Aristotle meant to articulate four and only four species of quality, Scotus concedes that one can account for the four kinds by interpreting quality as a disposition, that is, by an arrangement of the parts. The arrangement of the parts is then divided into an exterior and an interior arrangement, where the latter is divided into those that affect a different subject and those that affect the subject in which they inhere. Finally, the last member is divided into those that are innate to a subject and those that are acquired. Admittedly, by interpreting him as I have, I run the risk of turning Scotus into a corpuscularian. For if all the species of quality are simply dispositions, or arrangements of the material parts, then one might wonder how Scotus’s position is fundamentally any different from a seventeenth– century corpuscularian? Of course, given that Scotus thinks that matter has its own existence and substantial parts, this may not be a problem.21 In fact, this may well be a virtue of his position. However, one fundamental difference reveals itself when one keeps in mind Scotus’s early qualifications: the qualities of bodies are modes or accidental differences of bodies, since the essential differences of substances are beyond our grasp. That is, the four species or kinds of qualities identified by Aristotle are not essential differences of substances but are rather accidental qualifications, which may or may not reveal the inner substance of a thing. Why is this the case? Arguably, as any good Aristotelian will realize, it is because all of our knowledge starts with the sensation of physical, or corporeal, substances, that is, with substances that have parts outside of parts. Furthermore, as we saw earlier, the two most basic questions that concern parts seem to be: 1) how many parts? and 2) how are those parts organized? It is my contention that the most basic answers to these two questions should be identified as quantities and qualities, respectively.

21

See Pasnau, Metaphysical Themes, p. 51.

LOGIC, ONTOLOGY, AND THE PSYCHOLOGY OF UNIVERSALS IN DUNS SCOTUS CRUZ GONZALEZ-AYESTA AND DAVID GONZALEZ-GINOCCHIO

One of the difficulties in framing Duns Scotus’s doctrine on universals is to choose the perspective with which to present it. Scotus faces the problem of universals in at least five well-known texts, which, cited in a chronological order, are: questions 4-9 of the Questions on Porphyry’s Isagoge, the first question of d. 3 in book II of the Commentaries to the Sentences of Peter Lombard (in both versions of the Lectura and Ordinatio); and questions 13 and 18 of book VII of the Questions on the Metaphysics of Aristotle. The context of some of these passages is strictly metaphysical, namely, the discussion of the principle of individuation, while in others the specific issue at hand is the first Porphyrian question, that is, whether universals exist in reality or just in the mind, mixing a metaphysical perspective with a logical-semantic-epistemological approach. Among the difficulties raised by the metaphysical perspective, one is undoubtedly the precise sense that should be given to what has technically been called the ‘common nature’; concretely, how should this ‘commonness’ be understood: should a positive state be attributed to it or should it be characterized as mere indetermination or indifference towards universality and individuality (see P. King 1992, T. Noone 2002, G. Pini 2007). From a logical-semantic-epistemological point of view, the crucial issue is what meaning corresponds to universal terms and whether abstractions refer to something real or are themselves fictions of thought. While some interpreters argue in favor of a connection between the natura communis and the universal notion as cause and effect through the psychological process of abstraction (C. Marmo 1989, T. Noone 2002), others defend a certain conceptualism (P. King 1992, M. Tweedale 1999, R. Cross 2005).

102

Logic, Ontology, and the Psychology of Universals in Duns Scotus

On the other hand, there is also some discussion on the proper relationship between the epistemological and the metaphysical perspectives. Some authors defend an intrinsic bond between ontology and semantics; others believe that semantic and epistemological considerations are not sufficient for a complete account of the constitution of the world (G. Pini 2007).

1. Status Quaestionis There are a great number of studies about universals in Scotus that range from classic manuals (E. Gilson 1965, F. Copleston 1980) to specialized investigations (J. Owens 1957, C. Marmo 1989, P. King 1992, G. Sondag 1996, M. Tweedale 1999, T. Noone 2002, G. Pini 2007). In spite of the risk of falling into a simplification, we believe these interpretations may be grouped in two general categories: (a) the one we will call the ‘classic view’, which highlights the Avicennian features of Scotus’s doctrine and pivots over the metaphysical perspective, and (b) the view that intends to approximate Scotus to Ockham along the lines of conceptualism. As far as we can see, the classic manuals of the History of Philosophy, such as those of F. Copleston and E. Gilson, as well as the studies of J. Owens, C. Marmo and T. Noone, should be ascribed to the classic view, even though each of these authors presents his own understanding of Scotus’s doctrine. Specifically, while Copleston, Gilson and Owens delineate Scotus’s thought against the background of the Thomistic synthesis, T. Noone describes Scotus’s doctrine in the context of his dialogue with his predecessors and followers. The common thread running through this kind of interpretation is the focus on the Avicennian roots of Scotist thought and the qualification of that thought as a moderate realism. Avicenna had said that essence in and of itself, while being the content of universal concepts, is not itself universal, and while being given in each individual, is not itself concrete. Scotus would accept this notion. Essence of itself is the content of universal concepts but is not universal, and can repeat itself in each individual, not being an individual itself. It is common. It is not universal in the proper sense or actual when first presented to the intellect. Universality actually belongs to it only once it has been abstracted. What the intellect acquires through abstraction, nevertheless, is something real, even though not separable from the individual. Nature is of itself formally distinct from the individual but constitutes with it a unity that is less than numerical and, therefore, has a certain degree of reality. The controversial issue in this interpretation is the ontological status of the natura communis and the explanation of its

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

103

necessary relation to universal concepts in a way that allows for them not to be mere mental fictions. The second view that we have described as conceptualist is harder to describe in a uniform way. It seems clear that some authors (P. King, G. Pini) are leaning toward a weak interpretation of the common nature. We understand as “weak” the kind of explanation according to which the ‘being common’ of common nature considered in an absolute sense is not a positive attribute (a third mode of being different from individual or universal being) but just a negative characterization: the indeterminacy or indifference regarding these two ways of being, universality and individuality. For these authors, the problem of the repetition of the ‘common nature’ in different individuals (as different cases or instances) presupposes a certain confusion as it implies the reification of common nature. On the other hand, not even common nature could withstand the universality ‘test’ in a strict sense, for it is not something that is predicable of many. Therefore it must be said that what exist are individuals, and the universal is only in the mind, while the common nature is indifferent and can contract to both ways of being without having of itself a positive form. These authors must recognize that Scotus argues for a unity that is less than numerical for the common nature. Even so, they are not moved by this fact to speak of the entity or reality of the common nature. A tendency can be seen here to compare Scotus’s position with Ockham’s, and even while admitting there is a strong variance between them (M. Tweedale and P. King, for example, draw attention to this), they perceive some points in which Scotus opens a breach for Ockhamist conceptualism. A particularly hard topic, which we will not consider here, is that of the univocal notions that correspond both to God and the creatures (M. Tweedale and R. Cross have touched on this subject).1 1

The problem appears when speaking of perfections that can have an infinite or finite mode according to their being given in God or the creatures. In that case, when the same concept and the same term are used to express them, a problem emerges: what is the proper meaning of this concept? Does it actually refer to two different natures? According to Duns, for science to be possible, concepts must remain univocal within an argument, so that they can be used as the middle term in a syllogism. Tweedale indicates that in this point a certain fracture can be seen in Scotus’s thought in the direction of Ockham (Tweedale 1999, 422). Cross, speaking about the case of ens, calls it a ‘vicious abstraction’, that is to say, a concept to which no real thing pertains, in the same way as it occurs in genera:

Logic, Ontology, and the Psychology of Universals in Duns Scotus

104

Giorgio Pini (2007, 397) intended to remedy what in his view are the major defects of the standard interpretations of Scotus: (i)

The intelligibility of the notion of commonality (or, as T. Noone calls it, “community”) as something positive that exists between universality and singularity (the difficulty of the interpretations from a metaphysical point of view);

(ii)

And the supposed crypto-Platonism of Scotus when his realism is defended on the basis of semantic considerations, namely, that for any general term there must be something in the world that conforms to it as its meaning (the difficulty of logical-semantic-epistemological interpretations).

While we will not consider Pini’s arguments or the pertinence of his accusations against other interpretations; we believe, nonetheless, that they bring to the fore two crucial points of the doctrine of universals that must be studied in the texts themselves. We will divide our exposition into three parts: in the first place, we will explain the various senses of the term ‘universal’ in Scotus’s texts, allowing for a clarification of the difference between the logical and the metaphysical universal (section 2); in the second place, we offer an explanation of the way of being of the natura communis (section 3); in the next section we will focus on what Scotus calls the ‘complete universal’ (section 4); finally, we will draw some conclusions. In this way, section 3 will allow us to face the first of the problems mentioned by Pini: the notion of commonness, a problem that affects the metaphysical view. The fourth section, on the other hand, will deal with the problems from the logical-semantic-epistemological approximation (the accusation of crypto-Platonism, in Pini’s terms), although in this context we wanted to give a special relevance to psychology: the explanation of the process of the formation of the universal will provide some keys to understand Scotus’s own perspective.

there is no corresponding res to ‘animal’ that is not a dog or a cat. He argues that there is semantic theory in Scotus that is not founded upon real shared properties (Cross 2005, 68-70). Other interpreters (Ginocchio 2013, 47-53. 124ff.) reject Cross’s reading.

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

105

2. Logical and Metaphysical Universal In the text of the proemium to question 4 of the Quaestiones in librum Porphirii Isagoge, Scotus establishes a primary classification of the senses of ‘universal’: There is a proper question about universals. And because that which is said by the name precedes every question, we should note that ‘universal’, like other concrete terms, can be said in a triple fashion. In one way it is said as the subject, that is to say, as a thing of first intention to which the universal intention is applied, and in this way every universal is the first object of the intellect. In another way it can be said as the form, that is to say, for a thing of second intention, caused by the intellect and applied to things of first intention, and this is how the logician speaks about universals in a proper sense. The third way is said as an aggregate of subject and form, and that is a being per accidens.2

Two clear senses of universal appear in this passage: that of the universal as the first object of the intellect, which is the one that the real sciences (those whose objects exist outside of the intellect) deal with, especially metaphysics, and that of the universal as a second intention, the object of logic. Real science is indeed about the universal, which is a certain reality, while logic is about the universal inasmuch as it is an intention.3 With this, Scotus acknowledges a first difference between what we may call logical and metaphysical universality. Logical universality designates, according to Scotus’s own explanation in his Quaestiones super libros Metaphysicorum Aristotelis, the formal or explicit consideration of the relationship between an abstract concept and the subjects that fall within its extension or meaning: 2

Quaestiones in librum Porphirii Isagoge, Proemium q. 4: “In speciali quaerendum est de universali. Et quia omnem quaestionem praecedit quid dicitur per nomen, ideo notandum est quod ‘universale’, sicut et cetera concreta, sumitur tripliciter. Quandoque pro subiecto, scilicet pro re primae intentionis cui applicatur intentio universalis, et hoc modo universale est primum obiectum intellectus. Quandoque sumitur pro forma, scilicet pro re secundae intentionis, causata ab intellectu et applicabili rebus primae intentionis, et sic loquitur logicus proprie universali. Tertio modo pro aggregato ex subiecto et forma, et illud est ens per accidens.” Our translation. 3 “Ad secundum dicendum quod scientia realis est de universali primo modo, quod est res; sed logica est de universali secundo modo, quod est intentio” (Quaestiones in librum Porphyrii Isagoge, q. 7-8, interpolatio a [n. 29]).

106

Logic, Ontology, and the Psychology of Universals in Duns Scotus Sometimes [the universal is taken] for a second intention, which is a certain relation of reason in the predicable directed to that of which it is predicable.4

When it designates this connection, “universal” means a second intention and does not stand for the content of the concept but rather for something formal in it, namely, one of the properties it has when it is in the mind, the property of being said of many. That is why, in the proemium, Scotus holds that “sometimes the universal is taken for the form, that is to say, for a thing of second intention”, while other times “it is said as the subject, that is to say, as a thing of first intention to which the universal intention is applied”.5 The passage is precisely contrasting the use described above – where the universal is taken for the proper relation of predicability of the concept (logical concept) – to another in which universal signifies a res (metaphysical concept). However, when Scotus says that the universal stands for a res, does he mean reality itself (the being or nature of things: being a man or a dog) or the content of the concepts as their objective or abstracted being given by the intellect (the content of a concept such as man or dog)? It seems that he means both aspects. In a parallel text corresponding to the just-quoted proemium of the Questions on the Isagoge, he writes: In another sense ‘universal’ is taken for that which is denoted by that intention, which is a thing of first intention, for second intentions are applied to first intentions. In this sense it can be taken in two ways: In one way for that which is denoted by this intention as a sort of remote subject. In another way, for the near subject. In the first way a nature taken absolutely is called universal, because it is not of itself this, and thus it is not contrary to it of itself to be said of many. In the second way only that is universal which is actually indeterminate in such a way that numerically

4

Quaestiones super libros Metaphysicorum Aristotelis, VII, q. 18, n. 38: Quandoque [universale sumitur] pro intentione secunda quae scilicet est quaedam relatio rationis in praedicabili ad illud de quo est praedicabile”. Translation taken – and partially modified– from Tweedale (1999), v. 1, p. 151, § 38. 5 Quaestiones in librum Porphirii Isagoge, Proemium q. 4

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

107

one intelligible is sayable of every suppositum. This is the complete universal.6

“Universal” designates what is meant by the intention that is predicable of many, but Scotus makes a distinction. What is meant in a proximate way is something universal in a complete sense; that is to say, something that is one in number and that may be predicated of each subject, or the essence itself, inasmuch as it is abstracted or has objective being. While in an ultimate or remote sense what is designated is nature in an absolute sense, that is to say, inasmuch as it is not incompatible of itself with existing in various individual subjects. This is a key for Scotus’s doctrine on universals and appears in many passages in his texts, though frequently under different names. For example, in the same book of the Quaestiones super libros Metaphysicorum Aristotelis, in the question dedicated to the principle of individuation (VII, q. 13), a text usually considered his definitive doctrine on the subject, he contrasts what he calls the complete universal (which in a proximate potency can exist in many and can be predicated of many) with the mere non-repugnancy to be in many that he qualifies as “capable of being common in the nature of things”: It has to be understood that the complete universal is what is in many and of many, not actually but by a close potential. Nothing is of this sort except on account of the consideration of the intellect. But that which does not reject being in many is said to be one in many potentially, where ‘potentially’ is taken in its logical sense, not in the sense it has in natural science. In this way something common can be in the nature of things […].7

6

Quaestiones super libros Metaphysicorum Aristotelis, VII, q. 18, nn. 39-41: “[39] Alio modo accipitur universale pro illo quod denominatur ab ista intentione, quod est aliqua res primae intentionis, nam secundae intentiones applicantur primis. [40] Et sic accipi potest dupliciter: uno modo pro illo quod quasi subiectum remotum denominatur ab ista intentione; alio modo pro subiecto propinquo. [41] Primo modo dicitur natura absolute sumpta universale, quia non est ex se haec, et ita non repugnat sibi ex se dici de multis. Secundo modo non est universale nisi actu indeterminatum, ita quod unum intelligibile numero sit dicibile de omni supposito, et illud est complete universale.” Trans. Tweedale (1999), p. 151, §§ 39-41. 7 Quaestiones super libros Metaphysicorum Aristotelis, VII, 13, n. 131: “Intelligendum quod universale complete est quod est in pluribus et de pluribus, non actu sed potentia propinqua. Tale nihil est nisi ex consideratione intellectus.

108

Logic, Ontology, and the Psychology of Universals in Duns Scotus

‘What is universal’ is the natura inasmuch as it is abstracted by the intellect, while ‘what is common’ is clearly the natura considered in an absolute sense. A similar distinction may be found in the corresponding passages of book II of the Commentaries on the Sentences (both in Lectura and in Ordinatio), even though he employs there a different terminology, speaking of universalitas in re and universalitas in actu vel formaliter. In any case, the difference described is the same, namely, that between what is not repugnant to being ‘universal’ (in the sense of being common to many, or being in many) but is not yet something numerically one and what, inasmuch as it is capable of being predicated of many, can effectively exist in many. This second attribute would be proper of the universal formaliter.8 In the Ordinatio he seems to be referring to this universal formaliter when he speaks of universale actu: he writes that the natura of itself is not a universal in act, unlike the universal that is an object of the intellect.9 In fact, it is not repugnant for a thing to possess a quiddity that something else also possesses, but the nature that has such an indeterminacy does not have a unity (praecisa) such that it can immediately be predicated of many subjects. This is only possible if the nature is not taken ‘absolute’ but ‘ut intellecta’. Indeed, as an object of the intellect, the nature is taken as something numerically one and predicable of all of the subjects for whom it is not repugnant.10 As an intellectual object, it is universal in act. Unum autem in multis potentia –ut accipitur ‘potentia’ logice, non naturaliterdicitur cui non repugnat esse in multis, et sic commune potest esse in rerum natura.” Trans. Tweedale (1999), p. 264, § 131. 8 “Dico quod talis est ‘universalitas in re’ cui non repugnat ‘esse universale’; sed istud non est universale formaliter, nam ‘universale est unum in multis et de multis” (Lectura II, d. 3, p. 1, q. 1, n. 34). 9 “Qualiter autem debeat intelligi, potest aliqualiter videri per dictum Avicennae V Metaphysicae, ubi vult quod ‘equinitas sit tantum equinitas’, – nec est de se una nec plures, nec universalis nec particularis. Intelligo: non est ‘ex se una’ unitate numerali, nec plures pluralitate opposita illi unitati; nec ‘universalis’ actu est (eo modo scilicet quo aliquid est universale ut est obiectum intellectus), nec est ‘particularis’ de se” (Ordinatio II, d. 3, p. 1, q. 1, n. 31). 10 “Ad primum dico quod universale in actu est illud quod habet aliquam unitatem indifferentem, secundum quam ipsum idem est in potentia proxima ut dicatur de quolibet supposito, (…). Nihil enim –secundum quacumque unitatem – in re est tale quod secundum illam unitatem praecisam sit in potentia proxima ad quodlibet suppositum praedicatione dicente ‘hoc est hoc’, quia licet alicui exsistenti in re non repugnet esse in alia singularitate ab illa in qua est, non tamen illud vere potest dici de quolibet inferiori, ‘quod quodlibet est ipsum’; hoc enim solum est possibile de

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

109

It can be seen that all these passages (Quaestiones super libros Metaphysicorum Aristotelis, Lectura, Ordinatio) contrast nature of itself or absolutely considered and nature as an abstraction and object of the intellect. In this case, contrasting does not mean separating. Nature absolutely considered is the remote reality to which the universal in act (complete or formaliter) corresponds. The complete universal is the nature with abstracted or objective being. According to Scotus, the difference between logical universality (i.e., that of a second intention) and the universality of a metaphysical concept (i.e., a concept of first intention) lies in the fact that the universality of the logical concept is part of the mode of being of what is conceived or thought (i.e., the mode of being of a concept of second intention), while in the case of the metaphysical concept the universality is just a way of thinking of a nature: “nature having its being in the intellect, does not have of itself, in a primary way, universality”.11 The objects of logic are universal because they are intentions of thought: their universality is a part of their being thought. In the case of scientific knowledge about real things, something that is not of itself universal can be understood in a universal way. That is precisely why complete or in actu universality requires the intervention of the intellect. What is the starting point of the psychological process of abstraction? That which can be known by the senses. Scotus consistently holds that that which is apt to be taken by the agent intellect and converted into a complete universal is not nature taken in an absolute sense, but rather the nature as singular in an individual. That is why he establishes a distinction, between the universal in re, the universality proper to nature in itself which he sometimes calls a universal negatively (nature considered in its indeterminacy in such a way that it is neither universal nor singular) and that universality which he calls a universal privatively (nature as is given in a singular).12 If nature of itself is obiecto eodem numero, actu considerato ab intellectu” (Ordinatio II, d. 3, p. 1, q. 1, n. 37). 11 Cf. Ordinatio II, d. 3, p. 1, q. 1, n. 33. 12 It may be noted here that Henry of Ghent speaks about a privative and a negative indetermination in the context of the kind of knowledge we can have about God (Summa a. 24, q. 7). There is a negative indetermination of attributes (e.g., justice) and a privative one. The negative indetermination corresponds to an attribute simpliciter: it cannot be determined and is applicable to God. The privative

110

Logic, Ontology, and the Psychology of Universals in Duns Scotus

considered universal because it is not repugnant to it to exist in many (commonness or communality, universal negative), singular nature is called ‘universal’ because it is the starting point of the process of abstraction, from which a complete universal is obtained, even if it is not strictly universal because it is singularized (universal privative). These distinctions are clearly stated in the following text, corresponding to an interpolation in book VII, q. 18 of the Quaestiones super libros Metaphysicorum Aristotelis: As can be gleaned from his statements in this question, in addition to the notion of universality, which is a second intention or a conceptual relation in the predicable to that of which it is predicable, there is a threefold universal. The first is a universal taken negatively and this is nature in itself taken absolutely, of which Avicenna speaks (Metaphysics V, ch. 1, Avil 231) concerning humanity which is neither universal nor particular. And it is said to be a universal negatively in the sense that taken this way it is not apt to be a complete universal. The second is the universal taken privatively and this is nature in a determinate individual and it can become a complete universal by [the action of] the agent intellect. The third is the universal taken contrarily and this is the complete universal as abstracted from its determination by the intellect.13

In this same text Scotus writes: “However, the universal taken in the first two ways is in reality, as is obvious, although differently. How the universal, taken in the third way, is in reality is also obvious, although its

indetermination considers the attribute as universal, one in many, and it must be determined in order to be predicated of creatures. Scotus deals with this terminology in the d. 3, p. 1 of his commentaries to Sentences I. While Scotus believes Henry is wrong, it is not necessary to debate the issue here. It may be, however, a possible source for the nomenclature discussed here. 13 Quaestiones super libros Metaphysicorum Aristotelis, VII, q. 18, interpolatio a [n. 45]: “Sicut potest elici ex dictis eius in ista quaestione: praeter intentionem universalitatis, quae est intentio secunda sive relatio rationis in praedicabili ad illud de quo est praedicabile, est triplex universale. Unum est universale dictum negative, et illud est natura in se absolute sumpta, quo modo loquitur Avicenna de humanitate quod nec est universalis nec particularis. Et dicitur universale negative quod non est natum fieri universale completum sub illa ratione. Secundum est universale dictum privative, et illud est natura ut in individuo determinata, quia potest fieri universale completum per intellectum agentem. Tertium est universale dictum contrarie, et illud est universale completum, abstractum per intellectum a determinatione.” Trans. Wolter (1998), p. 301, note 48.

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

111

being there is not the same as that which characterizes it as universal”.14 It seems clear, then, that only the logical universal is something that belongs exclusively to the intellect. In the rest of the cases we have different senses of the term “universal”, which conceive types of universals in re. In the case of the privative universal, it is improperly a universal and the title can only be justified inasmuch as the individual is the starting point of the process of abstraction. This ‘universal’ refers to singularized nature: inasmuch as it exists. To sum up, it may be said that Scotus distinguishes four different sense of “universal”, the first of which is a logical sense and the rest belong to the metaphysical domain. All of these four senses appear in the text which, in our opinion, is the most complete, namely the above quoted Quaestiones super libros Metaphysicorum Aristotelis: (i)

Universal as a second intention or relation of reason.

(ii) Universal in re as universal negative which is nature absolutely considered inasmuch as it is not ‘universal’ in a complete sense nor singular and yet it is not incompatible to exist in various individuals. Some texts call such an indetermination or non-repugnancy: ‘common being’. That is why the term “natura communis” was coined by Scotists to designate precisely the essence considered in this way. (iii) Universal in re as universal privative which is nature singularized in an individual from which the process of abstraction can be initiated in collaboration with the agent intellect. This is an improper sense of universality inasmuch as it is the ‘singularized’ nature: the individual. (iv) Complete universal (in act or formaliter): it is abstracted and contrary to singularity (that is why he also calls it universale contrarie). It has an objective being in the intellect.

14

“(…) Sed universale primis duobus modis est in re; patet de se, sed alio et alio modo. Hic universale tertium quomodo est in re, patet; licet non sub eodem esse sub quo dicitur universale” (Quaestiones super libros Metaphysicorum Aristotelis, VII, q. 18, interpolatio a [n. 45]). Trans. Wolter (1998),p. 301, note 48 [slightly modified].

112

Logic, Ontology, and the Psychology of Universals in Duns Scotus

However, a clarification is needed regarding the ontological status of the natura communis. An explanation is also needed about the universal contrarie: while Scotus holds that abstract nature is in re, it does not exist in the same way as it does in the intellect. These will be the topics of the next sections.

3. A Modal Reading of the Natura Communis When expounding the different interpretations of Scotus’s doctrine of universals in the first section, we referred to the different approaches of the classical interpretation and the one that we called ‘conceptualist’. The interpretative divergence can be very clearly seen in the question of the natura communis, the principal point of dissent being whether it should be characterized positively (considering the common nature as a certain entity with a peculiar unity and a particular ontological status) or in a merely negative way (which views the common nature as incompatible either with individuality or with universality). We will throw some light on this discussion by means of textual evidence.

Minor unitate numerali Scotus talks about nature in itself in the texts where he considers the question of individuation, so that his first task is to justify that things are not individual of themselves; rather, their singularity requires an explanative cause. Scotus argues that the world cannot be only composed of individuals, and that there is a form of unity that is less than numerical (individual). He sets forth different arguments to prove this thesis, which are more or less similar in their three different versions (Lectura II, d. 3, p. 1, q. 1, nn. 8-27; Ordinatio II, d. 3, p. 1, q. 1, nn. 7-28; and Quaestiones super libros Metaphysicorum Aristotelis VII, q. 13, nn. 65-71). The Ordinatio presents the best layout; as the chronologically posterior Quaestiones super libros Metaphysicorum Aristotelis offer no significant novelty, we will follow the Ordinatio. Scotus formulates a possible objection to the individuation principle: just as nature of itself has being outside the soul but its existence in the soul can only be explained by means of the intellect, so it is singular by its own causes but its universality requires a different explanation (nn. 5-6). He then offers various arguments to show that things are not singular by themselves or by their own causes. These arguments can be divided into three classes: logical, metaphysical and psychological. In general, Pini’s

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

113

observation seems agreeable: “These arguments are based on the way things work in the observable world, independently of the words we use to refer to them and the concepts we use to represent them” (Pini 2007, 402). There seem to be two psychological arguments. The first one is devised to show that things cannot be singular by themselves because, if they were, the object of knowledge – which is prior to the act – would be of itself also singular. The consequence would be that in knowing it in a universal way, the object would be known by a ratio that is contrary to its own (n. 7). The relevance of this reasoning will be shown in section 4 below. The other argument concerns only sensory knowledge and strives to prove that there is a real unity that is less than individual, i.e., numerical, unity. Scotus points out that the act of the senses is specified by an object that is one but not numerically: senses are indeed capable sometimes of discerning individual differences between objects that nevertheless they acquire under a certain unity (for example, the rays of the sun in relation to sight) (nn. 21-22). The logical-metaphysical arguments are also framed with the intent to demonstrate that there is a kind of unity that is less than numerical and is the one belonging to a nature (n. 8). Scotus gives five senses of unity: (i) the degrees in a genus that are measured by reference to a first unity in that genus, which is real and yet not individual (nn. 12-15); (ii) the difference between the unity of the genus and the unity of the species is real but the unity of the species refers to a unity that is less than numerical (nn. 16-17); (iii) the relation of similitudes is necessarily grounded on a real unity (as the relation of similitude is real) that is nevertheless not individual (n. 18); (iv) if there were not a kind of unity like this the only possible real opposition would be between individuals, but the relation of contrariety (for example black-white) is a real kind of opposition and is not between individuals (n. 19); (v) if there were only a numerical unity, there would only be individual or numerical diversity, but then Socrates would be different from Plato in the same way that he is different from a rock, which is false (n. 23). Scotus finally provides a physical argument to prove that the unity of the natura is not individual. He points to the phenomenon of univocal causality, which is independent of the intellect. In univocal generation, however, there is certain unity between that which generates and that which is generated; it is a kind of unity that is not created by the intellect but only recognized by it (n. 28).

114

Logic, Ontology, and the Psychology of Universals in Duns Scotus

Ens, communitas, modus When we consult the texts to determine more precisely what kind of reality is attributed to a nature, which has of itself a kind of unity that is neither individual nor universal, there is no shortage of passages (in all relevant texts) in which Scotus claims that the nature is “an entity with a proper unity” or “a real entity with a real unity”: To the argument for the other opinion I say that it holds what is false, because nature is only a being with its own unity, and this is less than the unity that is numerical. And the nature’s unity is something it has of itself, but it is not a numerical unity.15 Or the minor may be proved in this way. Every real being, as such has some real unity. […] Now the nature, according to its being as nature, is a real being. Therefore it is one by some unity, in reality, and that unity is not that of an individual …16

Nevertheless, these passages must be read very cautiously. Speaking of an entity does not necessarily mean attributing to it effective extra animam existence. As Marmo has correctly pointed out (based on an analysis of Quodlibet, q. 3) there is a certain equivalence in Scotus’s terminology between ens and res as well as a gradation in what can be considered a res. Basically, res-ens is opposed to nothingness. As “nothingness” can mean the unintelligible contradiction or, in a weaker sense, what cannot exist independently of the intellect, so res-ens can refer to that which is not contradictory (ens of reason) or to what exists outside the intellect (ens extra animam) (Marmo 1989, 145). Scotus calls res “that which does or can have a proper being outside the intellect” (Quodlibet, q. 3, nn. 2-3, cit. in Marmo 1989, 153). Some interpreters follow a certain ‘maximalist’ or ‘ontologizing’ interpretation of the natura communis. Owens, for example, claims that, 15

Lectura II, d. 3, p. 1, q. 1, n. 37: “Ad rationem alterius opinionem dico quod capit falsum, quia natura est tantum ens unitate propria, et haec est minor unitas quam unitas numeralis, - quam de se habet, et non unitatem numeralem.” Trans. Wolter (2005), p. 19. 16 Quaestiones super libros Metaphysicorum Aristotelis VII, q. 13, n. 64: “Vel probetur sic minor: omne ens reale, secundum quod tale, habet unitatem realem (…). Sed natura, secundum quod natura, est vere ens reale. Ergo est unum aliqua unitate reali, non unitate individuali….” Wolter (1998), p. 209.

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

115

because Scotus holds the principle of the unity of the transcendentals one and being, he must admit a corresponding entity to the common nature, such an entity that can only exist individually singularized in the individual or in the intellect with universal being, and as such it constitutes a quidditative entity so that it can be the object of a real metaphysical consideration (Owens 1957, 11). Owens further claims, and this seems confusing, that there are two entities: one corresponding to the common nature and another one corresponding to the singularized natura (Owens 1957, 12). Saying that there are two simultaneous entities in the individual (one proper to the common nature and the singular esse) is a strong ontologization indeed of the common nature, and to our mind a hermeneutical mistake. The error may not lie in attributing entity or reality to the natura communis, but rather in doing so in such a way that that reality is understood as a certain ‘position’ or esse that is different from that of the singular (extra animam) and that of the universal (in anima). It is at this point that Scotus’s doctrine of the natura communis would become unintelligible, as Pini has indicated. There is, however, a series of texts that seem to support such a reading. The first significant locus in this respect may be found in two passages of the Commentary on the Sentences (Lectura II, d. 3, p. 1, q. 1, nn. 30-32 and Ordinatio II, d. 3, p. 1, q. 1, nn. 31-34). Scotus is outlining his own opinion on the topic of individuation and he starts by clarifying the ‘status’ of the nature taken by itself, following the Avicennian dictum: equinitas est tantum equinitas (Lectura, n. 30 and Ordinatio n. 31). Here is the more complete passage of the Ordinatio: And just as the nature is not of itself universal in virtue of that being, but rather universality happens to that nature in virtue of the first character of it on account of which it is an object, so also in things outside where the nature exists with singularity the nature is not of itself determined to that singularity; rather it is naturally prior to the character that contracts it to that singularity, and insofar as it is naturally prior to that contracting factor it is not repellent to it to be without that contracting factor. And just as the object in the intellect in virtue of that entity of it and universality has true intelligible being, so also in reality the nature has in virtue of that entity true real being outside the soul, – also in virtue of that entity it has a unity proportional to itself which is indifferent to singularity in such a way that it does not of itself conflict with that unity which is given with any unity of singularity.17 17

Ordinatio II, d. 3, p. 1, q. 1, n. 34: “Et sicut secundum illud esse non est natura de se universalis, sed universalitas accidit illi naturae secundum primam rationem

116

Logic, Ontology, and the Psychology of Universals in Duns Scotus

In this text Scotus holds (i) that nature in re outside the soul is given singularized even though (ii) of itself it is prior to the cause by which it is individualized, in such a way that (iii) considered of itself nature is not incompatible to exist without that individuating principle. He also establishes a comparison between the kind of being the natura acquires in the intellect and the being it acquires, so to speak, outside the intellect. The natura possesses universality and intelligible being (objective being) inasmuch as it is an object of the intellect because universality and objective being are the way of being of the natura in the intellect, and yet the known object itself is not universal: universality is the mode of knowledge, not the mode of that which is known. What is known, in the end, is the natura whose unity, distinct from numerical unity, has been demonstrated (we must here remember the difference we have established between the logical universal, the complete or contrarie universal and the negative universal). These precisions are very important to establish the proper exegetical keys to the hardest passage in the text: “ita etiam in re natura secundum illam entitatem habet verum esse reale extra animam”. The two first words establish the comparative sense, which allows the text to be read in the following way: “in the same way in re, the (individualized) natura, according to that entity (which is proper to natura of itself) has a real true existence outside of the soul”. Just like the natura is only universal ut obiectum and not because it is the content of concepts, so the natura only has real existence ut singularis and not because it is by nature prior to such a singularity. The second locus that must be examined is another passage from Ordinatio which gives classical definitions for the three states of the natura: universal, common, and singular. The text closes with the question of individuation as a certain confirmation of what has been previously explained:

eius, secundum quam est obiectum, – ita etiam in re extra, ubi natura est cum singularitate, non est illa natura de se determinata ad singularitatem, sed est prior naturaliter ipsa ratione contrahente. Et sicut obiectum in intellectu secundum illam primitatem eius et universalitatem habuit vere esse intelligibile, ita etiam in re natura secundum illam entitatem habet verum esse reale extra animam, et secundum illam entitatem habet unitatem sibi proportionalem, quae indifferens est ad singularitatem, ita quod non repugnat illi unitati de se quod quacumque unitate singularitatis ponatur.” Trans. Tweedale (1999), p. 178, § 34.

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

117

In response to the defense of the opinion: it is clear that commonness and singularity do not relate to a nature in the way being in the intellect and true being outside the soul do, because commonness belongs to the nature as outside the intellect, as does singularity. Commonness belongs to the nature of itself, while singularity belongs to the nature through something in reality that contracts it. But universality does not belong to the thing of itself.18

The text speaks of the communitas as something that pertains to nature outside the soul, just like singularity, even though being common and being singular are clearly distinct modes of being of the essence. Singularity appears when something is added that contracts the natura while being common belongs to the natura of itself. Universality also begs for an explanatory cause and assumes being in the intellect, not outside the soul. The text may lead to the qualification of ‘commonality’ as a certain positive property corresponding to the nature of itself, just as individuality corresponds to singularized nature and universality corresponds to the nature when it is abstractly known. To a certain extent this is what remains unclear in a study by T. Noone. He suggests four formulas to describe the status of nature: “(1) the nature has a minor unity and entity of its own; (2) the nature is not of itself this, that is, its unity is a kind of community; (3) the nature of itself could be found together with a distinct individuating principle without contradiction (or repugnance); and (4) the nature is not actually universal when first present in the intellect” (Noone 2002, 110). There is a tendency to identify the ‘communitas’ with the unity that is proper to a nature and, thus, with a positive property that defines what the essence is when it is neither singular nor universal: what capacitates it to be repeatable. While it is possible that this is a linguistic problem (giving positive formulations instead of negative ones), it seems to us in any case that such a characterization leads us again to a certain misunderstanding of the sense that communitas has in Scotus’s writings.

18 Ordinatio II, d. 3 p. 1, q. 1, n. 42: “Ad confirmationem opinionis patet quod non ita se habet communitas et singularitas ad naturam, sicut esse in intellectum et esse verum extra animam, quia communitas convenit naturae extra intellectum, et similiter singularitas, - et communitas convenit ex se naturae, singularitas autem convenit naturae per aliquid in re contrahens ipsam; sed universalitas non convenit rei ex se.” Trans. Tweedale (1999), p. 183, § 42.

118

Logic, Ontology, and the Psychology of Universals in Duns Scotus

Actually, what does it mean that the nature is common? Scotus explains it in the same way in every text where he deals with the principle of individuation: a nature is common because it is non-repugnant to being in many in the nature of things; common is that which can (according to logical potency) exist in another one than the one in which it exists.19 Scotus’s doctrine seems clear: a nature of itself is not singular, it becomes singular or individual when the principle of individuation is added. That is why a unity that is less than numerical and constitutes its entity or reality is logically compatible with another individuating principle, even if it is actually linked to this or that. This logical non-repugnancy of the unity that constitutes the nature with any individuation principle different from the one that actually or effectively individualizes it is what Scotus calls “communitas”. It is therefore a negative characterization: a nonrepugnancy, and even though it can linguistically be formulated in a positive way as compatibility-with, it does not constitute a real property in the same way as the singularity that results from the individuating cause or a logical property in the same way as the universality of second intentions. This, however, does not appear to have clarified the question: How must the entity-reality of the natura communis be then understood? It seems Scotus’s texts offer themselves to a modal interpretation. Peter King has also upheld this view when he said that “the contracted nature is an intrinsical mode of the uncontracted nature” (King 1992, 61). While we follow his idea of a modal approximation to the problem, our interpretation is not exactly the same as his. King does not consider the texts referred to above where Scotus attributes to nature, as prior to singularity and universality, a certain entity, and therefore King concludes that “the un-contracted nature, as such, necessarily does not exist” (1992, 55). Undoubtedly, this statement is correct, but it does not give sufficient attention to the problem, namely, the difference between entitas-res and esse. King identifies the common nature with an object’s essence in the 19

“Intelligendum quod universale complete est quod est in pluribus et de pluribus, non actu sed potentia propinqua. Tale nihil est nisi ex consideratione intellectus. Unum autem in multis potentia –ut accipitur ‘potentia’ logice, non naturaliter – dicitur cui non repugnat esse in multis, et sic commune potest esse in rerum natura (Quaestiones super libros Metaphysicorum Aristotelis, VII, 13, n. 131). “(…). In creaturis tamen est aliquod commune unum unitate reali minor numerali, -et istud quidem ‘commune’ non est ita commune quod sit praedicabile de multis, licet sit ita commune quod non repugnet sibi esse in alio quam in eo in quo est” (Ordinatio II, d. 3, p. 1, q. 1, n. 39).

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

119

intellect (King 1992, 54), but Scotus also holds a certain reality in re for it, even if not real existence or facticity. We must therefore explain how it is that nature of itself has a certain entity but does not exist. If we might employ a Kantian anachronism, we might say it has reality but no position. Nature has reality or entity but no esse: it does not exist independently of one of its two ways: universal or singular. To sustain this modal reading we might bring into our account the following two passages. The first one contains a modal indication regarding the way in which the relation of natura communis-individualized nature is found: To the first of these: there are certain things that are not repugnant to a given nature in itself, and yet are repugnant to a nature already existing or having been made, such as ‘not-existing’ or ‘not-having been made’. Likewise, perhaps the nature in itself does not reject being separated from all individual grades, since in thinking the nature without those we do not think of contradictories. Nevertheless since in existence it does reject being separated from all, although it does not reject being separated from this (since it is possible that it be in that, and vice versa), it follows that it can occur only under some individual grade, and for that reason it and that cannot differ in reality (Quaestiones super Metaphysicorum).20

This text confronts “natura in se” and “natura posita in esse vel facere”. The articulation that follows between them is suggested as follows: when nature is given in re in a singularized way (natura posita in esse) a nature that in itself does not have to exist in such a way (natura in se) is separable from the actual individuation principle it is linked with, but it is not separable from any and every principle of individuation. This means that this nature is not repugnant to the former (not being united to this individuation principle) but it is to the latter (not being united with any 20

Quaestiones super libros Metaphysicorum Aristotelis VII, q. 13, n. 136: “Ad primum istorum: quaedam naturae in se non repugnant, et tamen repugnant naturae positae in esse vel factae, sicut non-esse, non-factum. Similiter, naturae in se non repugnat forte separari ab omnibus gradibus individualibus, quia intelligendo naturam sine istis, non intelliguntur contradictoria. Quia tamen in esse repugnat sibi quod separetur ab omnibus, non autem quod separetur ab hac – hoc enim est possibile, ut in illa, et e converso; non ergo potest fieri nisi sub aliquo gradu individuali, quare et ‘ille’, ‘iste’ non potest diferre re”. Trans. Tweedale (1999), p. 266, §136.[modified]

120

Logic, Ontology, and the Psychology of Universals in Duns Scotus

individuation principle). As separable from the principle to which it is linked (by a formal distinction) the natura in se is by nature prior to the singular and has a certain entity or reality, as held by Owens or Noone. Its entity consists in the ability to have individual existence. As not-separable from any and every principle of individuation, it must be held, as with King, that natura in se does not exist: it can only exist factually as contracted or singularized (natura posita in esse vel facta). Now, when it exists in an individual mode, there is no factual coexistence between two positive entities in things themselves, rather their existence is another mode of nature. That is why natura in se is opposed to natura posita in modal terms. Scotus seems to suggest a modal terminology to speak about the relation between natura communis and universal nature. This passage belongs to the same locus in the Ordinatio previously referred to, where Scotus gives his opinio propia on the individuating principle (Ordinatio II, d.3, p. 3, nn. 29-40). After quoting the Avicennian dictum (“equinitas sit tantum equinitas”, n. 31), he deploys his own comprehension: nature is not of itself individual, and what relation it has with the universal mode it has in thought. He writes: But not only is the nature itself of itself indifferent to being in the intellect and in the particular, and consequently to being universal and particular or singular, but also when it has being in the intellect it does not have universality primarily in virtue of itself. For although it is understood under the reason of universality as a mode of understanding it, still universality is not a part of its primary concept, because it is not a metaphysician’s concept but a logician’s, for according to him [Avicenna], the logician studies second intentions that are applied to first intentions. Therefore, the first intellection is of the ‘nature’ as not understood along with some mode, neither a mode which belongs to it in the intellect nor one which belongs to it outside the intellect, even though universality is the mode of understanding that understood item, but is not the mode of what is understood.21 21

Ordinatio II, d. 3, p. 1, q. 1, n. 33: “Non solum autem ipsa natura de se est indifferens ad esse in intellectu et in particulari, ac per hoc et ad esse universale et particulare (sive singulare), – sed etiam ipsa, habens esse in intellectu, non habet primo ex se universalitatem. Licet enim ipsa intelligatur sub universalitate ut sub modo intelligendi ipsam, tamen universalitas non est pars eius conceptus primi, quia non conceptus metaphysici, sed logici (logicus enim considerat secundas intentiones, applicatas primis secundum ipsum). Prima ergo intellectio est

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

121

It can be seen that Scotus indicates first that nature is indifferent to universal and singular being (esse universale et particulare) and clarifies its relation to universal being. When nature is known abstractly (though this is not the place to discuss it, Scotus distinguishes abstract and existential knowledge), it is not itself universal but rather conceived as such. To illuminate the difference between universality as a way of intellection and as a mode of what is known, he refers to the difference between the concepts of the logician and the metaphysician. The logician employs concepts that refer to universal objects (second intentions) and which are, so to speak, universal in a double sense: for universality is here both a way of knowing and a mode of what is known. The metaphysician, on the other hand, employs concepts of first intention22 that refer to objects that are not universal in themselves and yet, inasmuch as they are intellectual and not sensible objects are known in a universal way. Universality is not a property of what is known, even if it is a way of knowing (“licet illius intellecti modus intelligendi sit universalis, sed non modus intellectus”). Nature is precisely the content of the concepts of the metaphysician and that is why Scotus can claim that, when known abstractedly, it is known without its mode: we do not acquire the universality (the way in which it is known by the intellect) nor its factuality (the particular way of being outside the soul): “Prima ergo intellectio est ‘naturae’ ut non cointelligitur aliquis modus, neque qui est eius in intellectu, neque qui est eius extra intellectum”. What is most relevant about this fragment for our current purpose is the modal indication. Nature is known independently of its two possible modes because nature in itself can be given in these two modes, and is thus prior to both. This power to be given according to both modes constitutes its entity and reality, which nevertheless is only effectively realized according to one of these modalities. There is a distinction between a nature and its modes. The way a nature is united with an individual must be understood according to a doctrine of ‘containing unity’: the individual ‘contains’ a nature, but in an individuated ‘naturae’ ut non cointelligitur aliquis modus, neque qui est eius in intellectu, neque qui est eius extra intellectum; licet illius intellecti modus intelligendi sit universalitas, sed non modus intellectus”. Trans. Tweedale (1999), pp. 177-78, §33. 22 First intentions as they refer to sensible objects, but they are not a reflexive knowledge of sensible things, rather, they are the intellectual form of knowing sensible things.

122

Logic, Ontology, and the Psychology of Universals in Duns Scotus

(modal) way. This individual’s containment of a nature includes other ‘contracting’ qualities. But these issues – the modal distinctions of nature, the nature-individual distinction and the question of the haecceitas – transcend the current discussion.

4. Nature with Esse Obiectivum: Scotist Psychology The most useful text to study Scotus’s position on abstract nature as the proximate subject of universal intentions is Quaestiones super Metaphysicorum Aristotelis VII, q. 18. The quaestio asks whether the universal is something in things themselves. After the usual pro and con arguments (nn. 1-9 and n. 10-12), Scotus offers a status quaestionis according to the solutions of Plato (nn. 13-15) and Aristotle (nn. 16-25), he then explains and refutes another opinion about universals being only in the intellect (nn. 26-37), and next moves on to the solution. In order to do so, Scotus first establishes a distinction (nn. 38-41) – already referred to in section 2 above – between the three senses of universal. According to Scotus, by ‘universal’ we can refer to (1) a second intention (the universals of logic), (2) a remote subject to which first intentions correspond (i.e. nature taken in an absolute sense, neither individually nor universally) and (3) a proximate subject designated by an intention, that is to say, something intelligible with numerical unity that can be predicated of various subjects. Sense (3), a first intention that can be predicated of many subjects, is what Scotus calls the “complete universal” (n. 41). The next two sections of Scotus’s text try to determine whether this complete universal is in the intellect (nn. 44-57) and in re (nn. 58-69). This last section is not conclusive, but the distinction about the complete universal being in the intellect is detailed and allows us to predict Scotus’s solution to the question of their being in re. The relevant passage is therefore the discussion about the complete universal being in the intellect, nn. 44-57. This section alludes to Scotus’ psychological doctrine of abstraction, further explained in Ordinatio I, d. 3, p. 3 and Quodlibetum 15. Sondag also compares these three sources to explain Scotus’ doctrine regarding the universal, but we do not fully agree with his conclusions (G. Sondag 1996, 387).

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

123

Scotus explains that something can be in the intellect in a habitual (in actu primo) or actual way (in actu secundo), and applies this distinction to the case of the mental object (n. 44). When it is present as a motive or cause of apprehension, it is present habitualiter; on the other hand, when it is effectively what is known, its presence is actual. Scotus holds, and this is explained in more detail in other texts, that the species is present to the intellect prior to its intellection. He therefore accepts the presence of a species impressa. Many have discussed whether it is prior by nature logically or also chronologically; we do not have to answer the question here.23 This presence of the species before its actual apprehension allows Scotus to explain that the species is, along with the intellect, a co-cause of intellection. Scotus then holds that the universal (in the sense of a nature with an objective being) is in the intellect, not necessarily in an actual way, but at least habitualiter (n. 46). Scotus wonders then how the object can be a complete universal, that is to say, something intelligible and numerically one that can be predicated of many. He is asking not for the cause of intellection (which is the intellect co-causing with the species impressa), but for the cause of the intelligible species (n. 47). His answer alludes again to a co-causality: neither the ‘res’ nor the intellect (agent or possible) can by themselves be the only cause of the species: Therefore, the agent intellect in conjunction with a nature which is in some way indeterminate of itself is the whole effective cause of the object in the possible intellect in respect of its first being [or in actu primo]. And this is the case as regards the complete indetermination of the universal. And just as there is no reason why what is warm makes things warm other than the existence of an appropriate power, likewise there is no other cause why the agent intellect in conjunction with the nature makes the object exist in this way. Therefore, a nature has a remote potency toward the determination of singularity and toward the indetermination of a universal. And just as it is joined to singularity by its producer, so it is joined to universality by the thing as an agent in conjunction with an agent intellect.24 23

Spruit (1994, 257-266) considers a logical and chronological priority. Quaestiones super libros metaphysicorum Aristotelis, VII, q. 18, n. 48: “(…) Intellectus igitur agens, concurrens cum natura aliquo modo indeterminata ex se est causa integra factiva obiecti in intellectu possibili secundum esse primum, et hoc secundum completam indeterminationem universalis. Nec est alia causa quare intellectus agens cum natura facit obiectum sic esse nisi quia est talis potentia, sicut nec quare calidum calefacit. Est ergo natura in potentia remota ad

24

124

Logic, Ontology, and the Psychology of Universals in Duns Scotus

This passage holds that the agent intellect concurs with the indeterminate nature considered in itself to be the integral productive cause of the species impressa (the object in actu or in esse primo in the passive intellect). This is not the place to explain the doctrine of a partial efficient cause.25 There is however another aspect to the previous statement that deserves an examination in the context of the doctrine of the universal. How can the natura be made present to the intellect as a co-cause of intellection? Scotus had claimed in one of the interpolations to the question (the one after n. 45) that nature absolutely taken could be named “universal negative” precisely “because it is not apt to be made a complete universal under that ratio,” while the singularized nature in the individual can be named “universal privative” because “it can be made a complete universal by the agent intellect.” It seems therefore that the origin of abstraction is the singular nature, and not the natura communis. There is no contradiction in these statements. Abstract knowledge begins with the apprehension of a singular by the senses, and continues to the forming of an image or phantasma. The origin of the process, in this sense, lies with the individuals known sensibly. When the phantasma concurs with the agent intellect to form a species impressa, the image contributes the natura indeterminately considered (not in its individuality). In this way, the natura makes itself present in the phantasma. This may be seen more clearly if we analyze the causal process that constitutes the species. According to Scotus, there is in the first place a real action of the agent intellect that provides a term to the passive intellect, which really transforms it: the formaliter presence of the species, the intelligible object that is in the intellect in a way that it was previously not, as a proximal cause of intellection. Scotus claims that “the object” is in the intellect as capable of causing an intellection. The “object” means an intelligible content that intentionally represents the natura present in a singular mode in the phantasma. That is why Scotus speaks of the “abstraction of the object” and qualifies it as an action that is not real.

determinationem singularitatis et ad determinationem universalis; et sicut a producente coniungitur singularitati, ita a re agente et simul ab intellectu agente coniungitur universalitati.” Trans. Tweedale (1999), pp. 154-5, § 48. [Slightly modified] 25 Cfr. Ordinatio I, d. 3, p. 3, q. 2, n. 496.

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

125

When the object is in its primary being in the intellect, it relates to the intellect differently than before when it was not in its primary being. For now it brings about movement, but before it did not. Now the intellect understands on account of that movement, while before it did not. Therefore, some change was naturally brought about before the intellect understood. But obviously this was not in the phantasm nor in the agent intellect, nor in their conjunction, because they were always equally joined while the image was present. Therefore, we have to claim that there is in the possible intellect some change naturally prior to the understanding itself, on account of which the object comes to be in its first being, which it was not in before. This is allowed in such a way that the abstraction of the object is not some real action; rather the intelligible species is caused by the phantasm and the agent intellect together. And when it is caused in the possible intellect formally, at the same time the abstracted object is caused there non-formally, but rather objectively. In this way we can rightly save what Avicenna says in Metaphysics V, ch. 2, concerning the intelligible form. For although there are two modes of being in, still one necessarily goes with the other. For unless we posit this generative activity [gignitio] we cannot give any real action to the agent intellect. And consequently we cannot give it any action concerning an object, because this exists intentionally only on account of something real.26

We have said that, for Scotus, there is a real action (an action in the natural order) of the categorical genus ‘action’ (a gignitio) which presents the object formaliter in actu primo (habitualiter) in the intellect. This presence supposes a previous real transformation in the possible intellect. 26

Quaestiones super libros Metaphysicorum Aristotelis, VII, q. 18, n. 51: “(…) Obiectum, quando est in esse primo in intellectu, aliter se habet ad intellectum quam prius quando non fuit in esse primo, quia nunc movet, prius non; nunc intellectus intelligit ex ista motione, prius non. Ergo aliqua mutatio facta est prius naturaliter quam intellectus intelligat. Sed ista [mutatio] non est in phantasmate – patet – nec in intellectu agente, nec in eorum coniunctione, quia semper erant aequaliter coniuncta, ex quo illud phantasma infuit. Ergo oportet ponere in intellectu possibili aliquam mutationem priorem natura ipsa intellectione per quam obiectum sit in esse primo in quo non fuit [priore]. Hoc conceditur, ita quod abstractio obiecti non est aliqua actio realis, sed causatur species intelligibilis a phantasmate et intellectu agente simul; qua causata in intellectu possibili formaliter simul causatur obiectum abstractum ibi, non formaliter, sed obiective. (…). Licet enim sit duplex modus ‘essendi in’, tamen unus concomitatur alium necessario. Nisi ponatur ista gignitio, nulla actio realis potest dari intellectui agenti, et ita nec aliqua circa obiectum, quia illa numquam est intentionaliter nisi propter aliquam realem.” Trans. Tweedale (1999), pp. 155-6, § 51 [slightly modified and with added emphasis].

Logic, Ontology, and the Psychology of Universals in Duns Scotus

126

This action is caused by the agent intellect and the phantasma. At the same time, and dependent on this real action, the intellect and the phantasma ‘cause’ the intelligible species, that is to say, they cause the content of the image to acquire an intelligible being that represents the natura in its indetermination. The agent intellect provides, if it may be said so, the objective being, and the phantasma provides the content of the representation. The abstract object is thus the same common nature with an objective being by which the reality which it represents is intentionally presented. In Quodlibetum q. 15 Scotus qualifies this non-real causality as ‘metaphorical’; the above quoted passage indicates that it is not a real causality, but an intentional, objective one. He clearly means the transformation from the image to the intelligible object (abstract, universal). Scotus holds that, even though objective being is not real, such a transformation is not possible without a real action of the agent intellect and the image by which the species begins to be formally in the passive intellect. To sum up, (i)

The species is really generated by a real action of the agent intellect and the phantasma itself, from the singular thing (universal privative), from which the phantasma is obtained by sensible knowledge.

(ii)

By means of this real action, the species is formaliter present to the intellect as a proximal cause of intellection before the intellection (species impressa).

(iii) The species, inasmuch as it is intelligible, has an objective or abstract being (universal complete) which results from a non-real action of the intellect and the phantasma. (iv) The species, being given an objective being (as intelligible or abstract), represents or intentionally refers to the natura not as singular but as separable from its determinations (universal negative).

We can see the connection between the singularized nature (universal privative), the natura communis (universal negative) and the nature with abstract being (universal contrarie), and the way in which the common nature with objective being is an intelligible content predicable of many (universal complete or contrarie) and can exist in re. Scotus claims that, owing to the simultaneity of the real action and the metaphorical action, and the co-causality of the phantasma in both, the universal contrarie is

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

127

representative of the natura considered without the determinations that singularize it. We can now turn back to one of the arguments referred to in the previous section. Scotus held, against the opinion that considered natura as individual of itself: The object insofar as it is the object is naturally prior to the act itself, and, according to you, as prior the object is of itself singular, because this is always the case with a nature when it is not considered as qualified or in respect of the being which it has in the soul. Therefore an intellect that understands that object under the character of a universal understands it under a character opposed to its own character, because as it precedes that act it is determined of itself to the opposite of that character, i.e. of that character of a universal.27

The strength of the argument relies on the result that, if there were no such thing as what the most common terminology calls “common nature”, there would be a paradox in the psychological process of abstraction. The phantasma would not make present a natura obtained from a being that individually instantiates it but is not of itself intrinsically linked to such a determination, but instead a natura that is of itself linked to singularizing determinations. When the agent intellect concurred with the image in the real action, they would then cause a species impressa that reflected the singular character of the natura present in the image. This would make the abstraction of a nature from an object an unintelligible process, for the intelligible species is universal precisely because it is predicable of many. Indeed, while the natura communis is not of itself so indeterminate as to be a complete universal, it is at least compatible with objective or abstract being; a nature that is singular of itself would not be. A singular object could not possess sufficient indetermination. The consequence is obvious: the intellect would have to go against the proper ratio of its object (its singular character) when giving it objective being and effectively knowing it in a universal way. But then the intellect, when abstracting, would show 27

Ordinatio II, d. 3, p. 1, q. 1, n. 7: “Obiectum in quantum est obiectum, est prius naturaliter ipso actu, et in illo priore – per se – obiectum ex se singulare, quia hoc semper convenit naturae non acceptae secundum quid sive secundum esse quod habet in anima; igitur intellectus intelligens illud obiectum sub ratione universalis, intelligit ipsum sub ratione opposita suae rationi, quia ut praecedit actum determinatur ex se ad oppositum illius rationis, scilicet universalis.” Trans. Tweedale (1999), p. 169, § 7.

Logic, Ontology, and the Psychology of Universals in Duns Scotus

128

us things under a different ratio than their real ratio: it would ‘modify’ them, so to speak. Only if nature is not of itself singular can we avoid this hurdle. It is interesting to note, in our opinion, that Scotus finds in the psychological process of abstraction one more argument to defend the universal in re. We hope to have shown that the argument is grounded on the kinds of causes and processes implied in the process of abstraction and not only in the semantics of concepts or terms.

5. Conclusions (i)

Just as in In Metaph IV q 1, nn. 27-30 he had rejected the question of the univocity of being as a merely semantic matter, Scotus here rejects the problem of the common nature as a problem of semantics (both being similar issues, inasmuch as they deal with the real existence of a being with less than numerical unity). Not all universal terms respond in the same way to a reality with a unity that is less than numerical. Scotus is no Platonist. And yet, according to his doctrine, when we employ universals obtained through abstractions, these are not fictions. Scotus is neither a nominalist nor a conceptualist, even if these labels have raised many questions amongst scholars.

(ii)

When the intellect considers abstract terms (with objective being) and their property of being universal, that is to say their capacity to refer to many, another sense of universality arises: universality as a relation of reason between concepts (genera, species, etc.) and the subjects that fall under each. This is the kind of universality that is a second intention and the object of logic.

(iii) Scotus describes a sense that eventually came to be called “universal in re”. He could be therefore considered a moderate realist. Yet his characterization of this sense of the universal sets him apart from other moderate realists, specifically Aquinas. Although both of their positions are inspired by Avicenna , for Aquinas common nature is only intellectually separable from individual essence, whereas Duns Scotus draws a real-modal distinction (see n. VI). The “universal in re” is what he sometimes calls negative. From an ontological viewpoint, this universal is the “natura absolute considerata”: what has since been called natura communis. This is universal only in the (incomplete) sense that it is not of itself individual nor is it repugnant for it to be in many (that is to say, it has the logical potency of being in many). And yet, it only exists as singularized. Once individuated,

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

129

it is repugnant for it to be separated from all individuals (even if it is not repugnant for it to be separated from this or that individual). (iv) It cannot be said that Scotus gives natura communis any mode of existence, position or facticity independently of its individuation. He does concede to it unity and entity. This difference must be understood according to modal categories. The natura communis has entity because it can exist individualized outside the soul or have an esse in anima as universal, but of itself it possesses no position or existence. This distinction allows understanding of some interpretational divergences. (v)

Natura communis or absolute considerata is a co-cause in the psychological process of abstraction (through its presence in the phantasma). So it is that the species impressa is the natura communis itself with an objective being, that is to say, universally considered. In this way, even if the actual process of abstraction begins with the singular (universal privative), the object of concepts or first intentions is proximately the abstract nature (with objective being) that Scotus calls universal complete or contrarie. Remotely, its object is the natura communis itself (universal negative). Scotus does argue for the existence of common nature from the complete universal, based on the psychological process of abstraction, but not, as has been said, with exclusively semantic arguments (grounded only in signification).

(vi) Scotus’s description of the abstraction process provides new arguments to support his claim about the universalis in re. Indeed, only if the essence has its own unity and being, and if it is separable from its contracted and singular version in a real and modal way and not only intellectually, can the understanding abstract the essence (under a universal reason) from the singulars (which possess it in an individual way). This separability explains the particular status of the so-called common nature. Therefore Scotus does not ground his position exclusively on semantic arguments.

References Primary Sources Ioannis Duns Scoti: (1950), Opera Omnia, v. 1, C. Balic (ed.), Typis Poliglottis Vaticanis, Civitas Vaticana, pp. 140*-175* (Introductio generalis). (1973), Opera Omnia, v. 7: Ordinatio liber secundus d. 1-3, Typis Poliglottis Vaticanis, Civitas Vaticana.

130

Logic, Ontology, and the Psychology of Universals in Duns Scotus

(1982), Opera Omnia, v. 18: Lectura in librum secundum Sententiarum d. 1-6, Typis Poliglottis Vaticanis, Civitas Vaticana. (1999), Opera Philosophica, v. 1: Quaestiones in librum Porphirii Isagoge et Quaestiones super praedicamenta Aristotelis, R. Andrews et al. (eds.), The Franciscan Institute, St Bonaventure (NY). (1997), Opera Philosophica, v. 3: Quaestiones in librum super libros Metaphysicorum Aristotelis, R. Andrews et al. (eds.), The Franciscan Institute, St Bonaventure (NY).

Translations Decorte, J. and Teske, R. (2005), Henry of Ghent’s Summa. The Questions on God’s Existence and Essence (Articles 21-24), Peeters, Paris. Etzkorn, G. J. and Wolter, A. B. (1998), Questions on the Metaphysics of Aristotle by John Duns Scotus, v. 2, The Franciscan Institute, St. Bonaventure (NY). Tweedale, M. (1999), Scotus vs Ockham. A medieval dispute over universals, v. 1: Texts, The Edwin Mellen Press, Lewiston (NY)Qeenston (CA)-Lampeter (UK). Wolter, A. B. (2005), John Duns Scotus Early Oxford Lecture on individuation. Latin Text and English translation with an introduction and notes, The Franciscan Institute, St. Bonaventure (NY).

Secondary Sources Copleston, F. (2003), A History of Philosophy, v. 2: Medieval Philosophy, Continuum, London. Cross, R. (2005), Duns Scotus and Suárez at the origins of Modernity, in Wayne J. Hankey-Douglas Hedley (eds.), Deconstructing Radical Orthodoxy. Postmodern theology, rhetoric and truth, Ashgate, Aldershot (Hants)-Burlington (UT), pp. 65-80. Gilson, E. (1955), History of Christian Philosophy in the Middle Ages, Random House, New York. King, P. (1992), “Duns Scotus on the common nature and the individual differentia”, Philosophical Topics 20/2, 51-76. González Ginocchio, D. (2013), Ser e infinito en Duns Escoto, Eunsa, Pamplona. Marmo, C. (1989), Ontology and semantics in the logic of Duns Scotus, en U. Eco- C. Marmo (eds.). On the Medieval theory of signs, John

Cruz Gonzalez-Ayesta and David Gonzalez-Ginocchio

131

Benjamins Publishing Company, Amsterdam-Philadelphia, pp. 143193. Noone, T. (2003), Universals and Individuation en T. Williams, The Cambridge Companion to Duns Scotus, Cambridge University Press, New York, pp. 100-128. Owens, J. (1957), “Common nature: A point of comparison between Thomistic and Scotistic metaphysics”, Mediaeval Studies 19, 1-14. Pini, G. (2007), “Scotus on Universals. A reconsideration”, Documenti e Studi sulla tradizione filosofica medievale 18, 395-409. Sondag, G. (1996), Universel et Natura Communis dans l’ordinatio et dans les questions sur le Peri Hermeneias (Une brève comparaison), en L. Honnefelder-R. Wood-M. Dreyer (eds.), Johannes Duns Scotus. Metaphysics and Ethics, Brill, Leiden-New York-Köln, pp. 385-391. Tweedale, M. (1999), Scotus vs Ockham. A medieval dispute over universals, v. 2: Commentary, The Edwin Mellen Press, Lewiston (NY)-Qeenston (CA)-Lampeter (UK), pp. 409-434.

THE SEMANTIC UNITY OF THE ANALOGOUS CONCEPT ACCORDING TO JOHN CAPREOLUS DOMENIC D’ETTORE

In his Short History of Thomism, Romanus Cesario calls the French Dominican John Capreolus (1380-1444) “the champion of a small, antirevisionist movement that, in effect, became a nucleus of the Thomism that during the Italian renaissance flourished in its own circles and even influenced certain secular humanists.”1 Capreolus’s great work, Defensiones Theologiae Divis Thomae Aquinatis, “embodies the first comprehensive presentation of Thomist theology” and “merited Capreolus … the title Princeps Thomistarum, the Prince of Thomists.”2 This paper looks at Capreolus’s engagement with one much debated issue central to the foundations of Metaphysics and Natural Theology; namely, the unity of concept of being and other concepts signified across the categories and of God and creatures. Many later Thomists, including Cajetan, argue that these concepts can have an ontological or formal unity, but deny that their signification is semantically one or univocal. This article shows how John Capreolus, the first Thomist to fully address the issue, provides the Thomist tradition with an explanation of the unity of the analogous concept which overlaps with the position of John Duns Scotus yet does not reduce the semantics of analogy to univocity.3

1

Romanus Cesario, Short History of Thomism, 59 Ibid., 61. 3 Throughout this article, I am addressing the unity of what Capreolus calls the “formal concept” as distinct from the “objective concept.” The distinction between these two and the difference between their unities is important within John Capreolus’ overall account of analogy in Natural Theology and his defence of Thomism from Scotist criticism. In so far as I see Cross’s article as touching primarily upon the unity of the formal concept, I restrict the scope of this article to this aspect of the issue. For a treatment of the distinction between the formal and objective concepts and their unity in Capreolus’s doctrine of analogy, see Domenic 2

134

The Semantic Unity of the Analogous Concept according to Capreolus

Capreolus and the Thomists before and after him develop their accounts of the conceptual unity of analogous signification in response to an argument from Scotus recently taken up by Richard Cross in his 2012 article “Duns Scotus and Analogy: A Brief Note.”4 In Cross’s translation, Scotus’s argument runs as follows: When the intellect is certain, it is certain either of a concept that is one simply speaking, or not, but [certain] of [a concept that is one] by the unity of analogy. If the first – and [it is] not [certain] of either concept (since it is doubtful about each of these in particular) – then it is certain of some third concept that is simply one: which is what is proposed. If the second, it is true to the extent that [the concept] is one in that way [i.e., by the unity of analogy]. But I argue about the [concept] that is thus one, that [the intellect] cannot be certain of what is one by the unity of analogy unless it is certain of the two as they are two: therefore those two do not seem to the intellect to be one, because they are simultaneously conceived as distinct concepts. (O I, 8, 1, 3, 67 [4:183]).5 D’Ettore, “John Capreolus on Names said Analogously of God and Creatures,” The Thomist, vol. 77 (July 2013): 402-406. 4 Richard Cross, “Duns Scotus and Analogy: A Brief Note,” in The Modern Schoolman, vol. 89 (No. 3-4), July and October 2012: 147-154. 5 Cross, “A Brief Note,” 151. Cross discusses the “certain and doubtful argument” as it appears in Ordinatio 1, dist. 3. There he gives the following summary: “Suppose we accept that God is wise (Scotus’s example is (a) being, but I choose a less contentious case so that we are not misled into irrelevant complaints about God’s not being an object in the universe). We thus accept that God falls under the extension of the concept of wisdom… But this concept must be distinct from the further concepts finite wisdom and infinite wisdom, or from created wisdom and uncreated wisdom… For we can accept that God falls under the extension of the concept of wisdom, but yet not know whether God falls under the extension of finite wisdom or the extension of infinite wisdom… Furthermore, it seems that the concept of wisdom is included in both the concept of finite wisdom and the concept of infinite wisdom.” See Cross, Duns Scotus on God, Ashgate Studies in the History of Philosophical Theology (Aldershot, England/ Burlington, Vermot: Ashgate Publishing Limited, 2005), 252. See also John Duns Scotus, Ordinatio I, dist. 3, pars 1, q. 1-2, nn. 27-30 (VC, vol. 3, 18-20). Referring back to the above quote on next page, Cross writes “there is a (complex) concept proper to God, infinite wisdom, and a (complex) concept proper to creatures, finite wisdom; each concept includes the simple concept of wisdom, along with the further qualifier – infinite and finite, respectively” (Cross, Duns Scotus on God, 253). For a brief summary of all five arguments Scotus gives for univocity, see Luis Alberto De Boni's "Duns Scotus and the Univocity of the Concept of Being," in New Essays on Metaphysics as Scientia Transcendens, edited by Roberto

Domenic D’Ettore

135

The last line of the argument above pertains strictly to the theory of analogy proposed by Henry of Ghent, in which the intellect mistakenly fails to distinguish between two distinct but related concepts.6 Scotus proposes that such confusion cannot be the foundation for analogy. Instead, recognition of the distinct concepts themselves as distinct must precede recognising them as analogously similar. The point Scotus draws out of the argument is that the foundation for the intellect’s experience of doubt about whether one and the same thing is an instance of x or y is the intellect’s certitude that both x and y are instances of z. That is, the intellect can be uncertain of whether to attribute one of two concepts to something only because it has grasped some third distinct concept present within both of the first two. The ancient philosophers could be uncertain of whether to attribute the concept “created being” or the concept “uncreated being” to fire only because they were certain that they could attribute the concept “being” to fire. It is only because of this “third” concept contained in the other two that the intellect attributes the (analogous) likeness to the distinct complex concepts. As Cross puts it, To be “certain … of a concept that is one simply speaking” is to just grasp a univocal concept, and the first option that Scotus considers in the passage goes like this: if we fail fully to grasp the two analogous concepts, but do grasp that, whatever they are, they are related by analogy, we must “be certain of some third concept that is simply one”: i.e., the univocal concept that, according to the penultimate passage just quoted, is included in the two analogous concepts. It is in virtue of our grasp of this third concept that we can detect that analogical relation between the first two concepts. Hofmeister Pich, Textes et Etudes du Moyen Age, 43 (Louvain-La-Neuve: Federation Internationale des Instituts d’Etudes Medievales, 2007), 102-112. For a fuller description of all five arguments presented within the context of Scotus’ overall cognitional theory, see Mary Beth Ingham and Mechthild Dreyer, Philosophical Vision of John Duns Scotus: An Introduction (Catholic University of America, 2004), 38-47. 6 Henry of Ghent, Summa questionum ordinarium, a. 21, q. 2 ad 3 (Decorte and Teske ed., 60). On the direction of Scotus’ arguments at Henry of Ghent, see Alexander Hall, Thomas Aquinas and John Duns Scotus: Natural Theology in the High Middle Ages, Continuum Studies in Philosophy (London/New York: Continuum, 2007), 17, and Etienne Gilson, Jean Duns Scot. Introduction à ses positions fondamentales (Paris, 1952), 88.

136

The Semantic Unity of the Analogous Concept according to Capreolus The second option is a little more complex. Suppose we do indeed grasp the two analogous concepts – a situation here referred to as being certain of a concept that it is one “by the unity of analogy.” Doing this requires knowing how the concepts are related (i.e., by conceptual overlap), on pain of infinite regress. To know the two concepts as distinct but analogously related, on the composition criterion outlined above, requires knowing that they overlap in some univocal concept.7

In effect, apart from the univocal content, i.e. apart from the overlapping semantic content, there would be no reason for relating the two composite concepts and no reason to entertain doubt over the application of these particular concepts to the item under consideration. This is the case whether the composite concepts relate to each other univocally or analogously. You can be certain the object in the distance is an animal while uncertain if it is a horse or a donkey because of the univocal content contained in the concepts of “horse” and “donkey,” i.e., “animal” or “sensitive living body.” In just the same way, you can be uncertain if, for example, light is substance or an accident while certain that it exists because either way it is still a being. “Being” is the content univocal to the analogous complex concepts substance and accidents just as “animal” is the content univocal to specifically different complex concepts “horse” and “donkey.” The same semantic rules apply in Natural Theology. 8 In Cross’s words: Take the concept of wisdom. Add to it the notion of limitation or finitude. This gives a complex concept: finite wisdom. Take the same concept of wisdom, and add to it the notion of limitlessness or infinity. This gives a 7

Cross, “A Brief Note,” 152. Cross himself assumes “that Scotus would restrict his discussion of analogy to the context of the transcendentals” because “the context focuses entirely on concepts that are either simply transcendental or composites of two simply simple transcendental concepts…” Analogous concepts then are not limited to those we use when speaking about God, but Cross limits himself in his “Brief Note” to treating what is relevant to “our God-talk.” See Cross, “A Brief Note,” 150. On Scotus’ discussion of transcendentals, Cross refers his reader to Ordinatio I, 8, 1, 3, 97-115 [4:198-207]. Cross gives a brief explanation of Scotus’ position on the transcendentals in Duns Scotus on God, especially pp. 254-257. For a more detailed account of the transcendentals in the writings of John Duns Scotus, see Allan B. Wolter, The Transcendentals and their Function in the Metaphysics of Duns Scotus, Franciscan Institute Publications, Philosophy Series, 3 (St. Bonaventure, New York: The Franciscan Institute, 1946).

8

Domenic D’Ettore

137

further complex concept: infinite wisdom (O I, 3,1, 1-2, 39 [3:26-7]; for infinity, see O I, 3, 1, 1-2, 58 [3:40]). Each is a composition, and they overlap; and the overlap is, of course, univocal (see O I, 3, 1,1-2, 27 [3:18]). The question I address here is this: what does Scotus believe the semantic relation of these two complex concepts to be? And my answer, in short, is that the relation is analogy.9

In Scotist doctrine then, analogy is the semantic relation between “distinct [complex] concepts, overlapping at some univocal concept.”10 Because the composite concept “infinite wisdom” proper to God and the composite concept “finite wisdom” proper to creatures have overlapping semantic content, the two composite concepts are analogous to each other. And because any overlapping semantic content is univocal content, the foundation for analogy is univocity. Two concepts agree univocally when “they have some semantic content in common.”11 Two complex concepts agree analogously when they differ in degree or mode but agree in some “simply simple”12 univocal concept.13

9

Cross, “A Brief Note,” 150. Cross, “A Brief Note,” 152. See John Duns Scotus, Ordinatio 1, 8, 1, 3, 64 (VC, vol. 4, 182). See also Richard Cross, Duns Scotus on God, 257: “Thus far, I have presented Scotus’s claim that there are univocal concepts under whose extension both God and creatures fall. This claim does not mean that he has no theory of analogy; quite the reverse, in fact. In a late addition to the Ordinatio Scotus explicitly accepts such a theory in this context. He presents a version of the argument…that presents the unity between two complex concepts that share one simpler univocal concept as analogy. The idea is, presumably, that the complex concepts of (say) infinite wisdom and finite wisdom are analogous to each other. Now, the context is a refutation of the opinion of Henry of Ghent that there is no ‘real likeness’ between God and creatures, and hence that forms predicated of God and creatures are merely analogous to each other. So I take it that Scotus is in this context using the words of his opponent. But the position seems wholly compatible with the sort of theory that Scotus develops in his theological writings, and whatever label is given to the relation between the concepts of (say) infinite wisdom and finite wisdom, that relation is clearly closer than the relation that obtains between two concepts that have nothing in common.” 11 Cross, “A Brief note,” 152. See full quotation above in text. 12 Cross explains “Simply simple” as “Scotus’s jargon to describe a wholly noncompositional concept.” Cross, “A Brief Note,” pp. 149. 13 According to Cross, analogy differs from univocity in that the name has only a similar meaning and not precisely the same meaning. Wherever there is similarity there is something the same which is univocally common to the similar things. On 10

138

The Semantic Unity of the Analogous Concept according to Capreolus

Cross has argued that dissolving “analogy into univocity” allowed Scotus to succeed when others fail at providing a semantic account of analogy, “albeit that explanation is univocity.”14 I do not challenge Cross’s interpretation of Scotus, but present an old response to Scotus’s argument which challenges the reduction of analogy to univocity.

Enter the Thomists Generations of Thomists from the beginning of the 14th Century through the 15th Century treated Scotus’ “certain and doubtful” argument as one of the central critiques of the Thomist understanding of analogy. The 14th Century Thomists, Thomas Sutton and Hervaeus Natalis agree with Scotus (as interpreted by Cross) that if there is one and the same single concept (or ratio) contained within, for example, “wisdom” said of God and creatures, then that content would be univocal to creaturely and divine wisdom. To save Thomas Aquinas’ doctrine of analogy, both early Thomists steadfastly deny any single ratio for the names said of God and creatures.15 They explain the experience of certitude and doubt by saying Scotus’ doctrine of modes and degrees in the context of divine infinity, see Richard Cross, Dun Scotus, (New York/ Oxford: Oxford University Press, 1999), 39-42. See also Cross’s discussion of analogous unity and analogy in Duns Scotus, 33: “…we sometimes use a word in different but related or similar senses. In this case, we could say that we are using the word analogically…. I take it that we can talk of a word’s having two (or more) similar senses only if there is something in common between the two senses. But the senses can have something in common only if the attributes signified by the terms themselves have something in common. The attributes, presumably, include some more basic property that they have in common. If they did not, it would be difficult to see how we could claim that they were similar (rather than wholly different)… Now, we can presumably find, or invent, a term to signify any common basic attribute. And this term will be univocal: it will be used in the same sense in all statements.” See also Richard Cross, “Univocity and Mystery,” in New Essays on Metaphysics as Scientia Transcendens, ed. R. H. Pich Textes Et Etudes Du Moyen Age, 43 (Louvain-LaNeuve: Federation International des Institutes d’Etudes Medievales, 2007), 119. 14 Cross, “A Brief Note,” 153; see Hochschild, The Semantics of Analogy, 134. 15 See Thomas Sutton, Quaestiones Ordinariae 32, ad 12, in Zur Diskussion über das Problem der Univozität im Umkreis des Johannes Duns Skotus, edited by Michael Schmaus (München: Bayerische Akademie der Wissenschaften, 1957) [hereafter, Schmaus ed.], 40, and Hervaeus Natalis, Quolibet, 2, a. 7, in Quolibeta Heruei : subtillissima Heruei Natalis Britonis (Venice, 1513; repr. Ridgewood, N.J. : Gregg, 1966) [hereafter, Venice, 1513], 46v. For a discussion of 14th Century

Domenic D’Ettore

139

the intellect can be certain that something is a being while uncertain if it is a substance or an accident because the concept of “being” in such cases expresses the disjunction “substance or accident.” In other words, the certitude that something is a being is nothing other than the certitude that something is “either a substance or an accident.”16 By taking this position, the early Thomists could deny any univocal content to the concept of being and likewise to any name said of God and creatures. Effectively, they deny any “simply simple” concept of being distinct from the concepts of substance and accidents. Since there is no separate consideration of “being” from the consideration of substance or accidents, no underlying univocal concept of being need or can be posited to explain the experience of certitude and doubt. In the 15th Century, John Capreolus approached the matter differently. Breaking ranks with the early Thomists and establishing the future practice of the Thomist tradition, Capreolus admits Scotus’ argument shows that there is indeed one concept of being. Capreolus agrees with Cross’s Scotus that there is one concept (e.g. of wisdom) involved in the name said analogously, and that names said properly of God necessarily employ complex concepts.

Thomist doctrines of analogy, see especially E. J. Ashworth, “Analogical Concepts: The Fourteenth-Century Background to Cajetan,” Dialogue XXXI (1992), 399-413. Ashworth also briefly treats Capreolus, Dominic of Flanders, and Soncinas in this article. For treatment of the topic in the writings of 14th Century Thomists and other authors not treated in this article, see especially Silvia Donati, “La discussione sull’unità del concetto di ente nella tradizione di commento della Fisica: commenti parigini degli anni 1270–1315 ca.” in M. Pickavé (ed.), Die Logik des Transzendentalen. Festschrift für Jan A. Aertsen zum 65. Geburtstag. Miscellanea Mediaevalia 30 (Berlin, New York: Walter de Gruyter, 2003), 60-139. Bonino addresses the relation of Capreolus to the early Thomists in S. Bonino, “Le concept d'étant et la connaissance de Dieu d'après Jean Cabrol (Capreolus),” Revue Thomiste 95 (1995): 109-136. 16 See Thomas Sutton, Quaestiones Ordinariae, 32 ad 14 and ad 23 (Schmaus ed., 41, 45-46); Hervaeus Natalis, Quolibet, 2, a. 7 (Venice, 1513, 44v, 45v).

140

The Semantic Unity of the Analogous Concept according to Capreolus

Capreolus, Defensiones 1, dist. 22. John Capreolus engages the discussion of analogous names a number of times in Defensiones Theologiae divi Thomae Aquinatis, Book 1.17 Most scholarly treatment of Capreolus’ doctrine of analogy has focused on Defensiones 1, dist. 2, q. 1.18 While this text is indispensable for a full account of Capreolus’ position, here I point to something relevant from a question twenty distinctions later on whether God can be designated by any name properly, “utrum Deus possit aliquo nomine proprie designari.”19 Conclusions one through four of Distinction 22, q. 1 are: 1) No name is said about God properly as regards the mode of signifying.20 2) Some names are said about God properly as regards the thing signified, and some are not [said properly], but only metaphorically.21 3) No name imposed by us signifies the divine essence according to what it is in itself (i.e. essentially), but only in so far as it is represented in the perfections of creatures.22

17 Johannes Capreolus, Defensiones Theologiae Divi Thomae Aquinatis, In Primo Sententiarum, Prologus, Quaestio 1, edited by C. Paban and Thomas Pegues, 7 vols. (Turonibus: A. Cattier, 1900-1908; repr. Frankfurt: Minerva GmbH, 1967) [hereafter Frankfurt]. 18 The fullest English language treatment of this text is given in Lawrence Dewan, “Does Being Have a Nature? (Or: Metaphysics as a Science of the Real),” in Approaches to Metaphysics, edited by William Sweet (Dordrecht: Kluwer Academic Publishers, 2004), 23-59. See also S. Bonino, “Le concept d'étant et la connaissance de Dieu d'après Jean Cabrol (Capreolus), ” Revue Thomiste 95 (1995): 109-136, and Domenic D'Ettore, “John Capreolus on Names said Analogously of God and Creatures,” 395-418. 19 Capreolus, Defensiones 1, dist. 22, q. 1 (Frankfurt, vol. 2, 169b). 20 Capreolus, Defensiones 1, dist. 22, q. 1, a. 1 (Frankfurt, vol. 2, 170a): “Prima conclusio: quod nullum nomen proprie dicitur de Deo, quantum ad modum significandi.” 21 Capreolus, Defensiones 1, dist. 22, q. 1, a. 1 (Frankfurt, vol. 2, 170b): “Secunda conclusio est quod aliqua nomina proprie dicuntur de Deo, quantum ad rem significatam, et quadam non, sed solum metaphorice.” 22 Capreolus, Defensiones 1, dist. 22, q. 1, a. 1 (Frankfurt, vol. 2, 171a): “Tertia conclusio est quod nullum nomen a nobis impositum, significat divinam essentiam

Domenic D’Ettore

141

4) No name imposed by us signifies God according to a concept of Him simple and proper to Him alone.23

Capreolus argues for the first three conclusions almost exclusively through quotations from works of Thomas Aquinas, especially Prima Pars 13, Summa Contra Gentiles 1.30, and De Potentia 7.5. To paraphrase Capreolus’ adoption of these texts: Human speech acts reflect our modes of understanding. Since we cannot understand God as God is in Himself, we cannot signify God in speech acts in the way God is in Himself. Since creatures do imperfectly represent God in their creaturely perfection, we can imperfectly understand God in understanding creatures, and, consequently, a concept signifying a creaturely perfection will imperfectly signify God’s essence.

As Capreolus quotes De Potentia 7.5 in his argument for the third conclusion: “although names of this kind, which the intellect from such conceptions attributes to God, signify that which is the divine essence, they nevertheless do not signify it perfectly according to what it is (secundum quod est), but in so far as it is understood by us.”24 Arguing for the 4th Conclusion, Capreolus speaks in his own name, and explains his understanding of how a complex concept applied to God properly contains within it one and the same concept drawn from and applied to creatures: …For our intellect, in the state of the way, does not understand God in any concept unless it is a concept of a creature, such as the concepts of perfections exemplared (conceptus perfectionum exemplatarum) from divine perfection; or at least composed from such concepts as is the case with the concepts: first cause, infinite being, and the like. But we signify as we understand. Therefore, no name imposed by us is attributed to God by

secundum quod in se est, sed solum secundum quod repraesentatur in perfectionibus creaturarum.” 23 Capreolus, Defensiones 1, dist. 22, q. 1, a. 1 (Frankfurt, vol. 2, 171b): “Quarta conclusio est quod nullum nomen a nobis impositum, significat Deum secundum ejus simplicem et proprium ei soli conceptum.” 24 Capreolus, Defensiones 1, dist. 22, q. 1, a. 1 (Frankfurt, vol. 2, 171b): “...licet hujusmodi nomina, quae intellectus ex talibus conceptionibus Deo attribuit, significent id quod est divina essentia, non tamen perfecte ipsam significant secundum quod est, sed secundum quod a nobis intelligitur.”

142

The Semantic Unity of the Analogous Concept according to Capreolus us in such a way that the concept would signify God and no other, not even its [i.e., the concept’s] parts. These things are clear from what has been said. And it is confirmed. Because if there were any such name expressing any concept proper to God alone, this would seem most of all to be the name “God.” But it is not so about the term “God.” Hence, [it is the case] about none. The minor premise is clear. Because although this name “God” signifies the divine nature, nevertheless, it does not express a concept belonging to God alone; on the contrary, [it expresses] a complex concept from many concepts, which belong separately to creatures. For “God” signifies divine nature under the concept “universal provider” or “first cause,” or “first mover,” or something of this kind. But it is certain that the parts of these complex concepts belong to many creatures. And so the conclusion is clear. With which (conclusion) it stands that many names express simple concepts nevertheless common to God and creatures, such as wise, good, and the like.25

Effectively, any name said about God properly is a composite of names said properly about creatures. Even the name “God” itself does not signify God except through its signification of creatures. The term “God” signifies the divine essence under the aspect of some such complex concept as “first cause,” “universal provider,” or “first mover.” Each part of these complexes belongs properly to creatures. We encounter various “firsts” and “causes” among creatures, such as the “first person to enter a room” and “the arm causes the movement of the stick.” The intellect combines 25 Capreolus, Defensiones 1, dist. 22, q. 1, a. 1 (Frankfurt, vol. 2, 171b-172a): “Probatur conclusio ista ex praedictis. Nam intellectus noster, in statu viae, non intelligit Deum in aliquo conceptu, quin ille sit conceptus creaturae, sicut sunt conceptum perfectionum exemplatarum a divina perfectione; vel saltem compositus ex talibus conceptibus, sicut sunt isti conceptus: causa prima, ens infinitum, et hujusmodi. Sicut autem intelligimus, ita significamus. Ergo nullum nomen a nobis impositum, sic est Deo a nobis appropriatum, quod Deum significet et nihil aliud significet iste conceptus, nec ejus partes. Ista omnia patent ex praedictis. Et confirmatur. Quia si esset aliquod tale nomen exprimens aliquem conceptum soli Deo proprium, hoc videretur esse potissime hoc nomen, Deus. Sed non sic est de hoc termino, Deus. Ergo de nullo. Minor patet. Quia hoc nomen, Deus, licet naturam divinam significet, tamen non exprimit conceptum soli Deo convenientem; immo conceptum complexum ex multis conceptibus, qui divisim creaturis conveniunt. Significat enim divinam naturam sub hoc conceptu, universalis provisor, vel causa prima, vel primus motor, vel hujusmodi. Constat autem istorum conceptuum complexorum partes convenire multis creaturis. Et sic patet conclusio. Cum qua stat quod multa nomina exprimunt conceptus simplices, communes tamen Deo et creaturis, sicut sapiens, bonus, et hujusmodi.”

Domenic D’Ettore

143

these two concepts of “first” and “cause” together to produce the complex concept, “first cause.” The component parts of this complex concept signify creatures. Taken absolutely and as a unit “first cause” applies to nothing in our experience from which we can abstract a concept. Apart from its reference or signification of creaturely firsts and creaturely causes, the complex concept “first cause” has no meaning. So neither the name “God,” nor the names “first cause,” “universal provider,” etc. are altogether proper to God in their signification. They necessarily include within them creaturely reference to be meaningful.26 Capreolus’ position on forming a proper concept of God becomes clearer in his response to the third objection from Peter Auriol regarding reasoning with analogy. Auriol had argued that the intellect must be able to produce a concept proper to God alone, drawing on Aristotle’s example of reasoning by analogy from three knowns to a previously unknown fourth in the Physics demonstration of the existence of prime matter.27 Auriol’s argument tries to show that analogous reasoning provides an 26 Capreolus provides a clear and detailed example of what he means by forming a complex concept of God in Defensiones 1, dist. 2, q. 1, a. 1, Seventh Conclusion (Frankfurt, vol. 1, 123a-b): “Verb gratia: ex motu coeli, quem videmus sensu, concludimus coelom moveri ab aliquo alio; et ultra, per alias rationes, quod illud quod movet coelom est substantia. Item, ex eo quod movetur circulariter, arguimus quod substantia movens coelom est intelligens; et quia ille motus semper videtur durare, ulterius concluditur quod illud movens est incorporeum et immateriale. Et ex omnibus istis colligimus quod est aliquod ens quod est substantia immaterialis incorporea, qui conceptus est proprius substantiae separatae. Ulterius autem per alias rationes concludimus quod est aliquod ens quod est substantia immaterialis incorporea et prima omnium substantiarum separatarum, qui est conceptus proprius Dei, in quo conceptu ex post cognoscimus Deum.” “For example: from the motion of the heavens, which we see by sense, we conclude that the heavens is moved by some other. Through other reasons, we conclude further that the other which moves the heavens is a substance. Then, from the fact that it is moved in a circle, we argue that the substance moving the heavens is intelligent; and, because that motion seems to endure always, it is further concluded that the substance doing the moving is incorporeal and immaterial. From all those conclusions, we gather that there is some being which is an immaterial, incorporeal substance, which is the concept proper to separate substance. We conclude further through other reasons that there is some being which is an immaterial, incorporeal substance and the first of all separate substances, which is the concept proper to God, in which concept from that time we know God.” 27 Capreolus, Defensiones 1, dist. 22, q. 1, a. 2 (Frankfurt, vol. 2, 172b-173a). See Peter Auriol, Scriptum in I Sententiarum, dist. 22, a. 2 (Vaticana, 522b).

144

The Semantic Unity of the Analogous Concept according to Capreolus

alternative to composition for the intellect to become aware of and represent conceptually things with which it has no sense experience. Just as the intellect has no concept of prime matter through abstraction from a phantasm of prime matter but knows prime matter through analogous reasoning, the intellect can know and conceive God by engaging in analogous reasoning from three knowns. Responding to this objection, Capreolus writes: To the third I say that the wayfarer is able to have the fourth concept; but it will not be properly of God and simple, but the concept of being or some composite concept.28

He adds: … it is doubtful to me if we have a simple concept of prime matter. For it seems that always it is conceived by a certain complex concept, composed from an absolute concept and a relative concept with respect to form. And whatever may be [the case] about this, no effect of God can lead the wayfarer to knowing God in Himself, that is, by a simple concept proper to Him, by which no other thing is conceived. Nor does any concept held by creatures represent Him as He is in Himself, but always in another, whether [it represents] by negation, as the concept of infinite, or by causality, as first cause, first being, or by the way of eminence, as supergood, super-just, and the like; for our concepts either are likenesses of sensible creatures, or composed from them or with them, or drawn from them.29

28

Capreolus, Defensiones 1, dist. 22, q. 1, a. 2 (Frankfurt, vol. 2, 175a): “Ad tertium dico quod viator potest habere illum quartum conceptum; sed ille non erit proprius Dei et simplex, sed conceptus entis vel aliquis conceptus compositus.” 29 Capreolus, Defensiones 1, dist. 22, q. 1, a. 2 (Frankfurt, vol. 2, 175a): “Sed dubium tamen mihi est, si nos habeamus simplicem conceptum proprium materiae primae. Videtur enim quod semper concipitur quodam conceptu complexo, composito ex conceptu absoluto et conceptu relativo, per respectum ad formam. Et quidquid sit de hoc, nullus effectus Dei potest viatorem deducere ad cognoscendum Deum in se, id est, conceptu sibi proprio simplici, quo nulla alia res concipiatur. Nec aliquis conceptus habitus a creaturis, sive per negationem,ut conceptus infiniti, sive per causalitatem, ut causa prima, ens primum, sive per viam eminentiae, ut superbonus, superjustus, et hujusmodi, repraesentat eum ut in se est, sed semper in alio; conceptus enim nostri, vel sunt similitudines creaturarum sensibilium, aut compositi ex illis, vel cum illis, aut ab illis extracti.”

Domenic D’Ettore

145

Capreolus grants to Auriol the validity of analogous reasoning. He even grants that in some cases, such as Auriol’s example of bread, with sufficient prior knowledge of accidental features of a thing or its effects analogous reasoning can lead to “a simple, proper and distinct concept of the quiddity of a thing.”30 He denies that in the case of reasoning by analogy to God (or even to prime matter) the “fourth concept” derived from the previous three signifies God simply, properly, and distinctly; that is, signifies God without at once signifying creatures in its component parts. The response also adds to what has been said before in the question that the concepts with which the human intellect conceives God represent God by way of negation, causality, or eminence from creatures, “by negation, as the concept of infinite, or by causality, as first cause, first being, or by the way of eminence, as super-good, super-just, and the like.” All of the human intellect’s concepts of God are composed of the concept of a creaturely perfection and at least one negative or relative concept. Conceiving the Divine Attributes is similar to conceiving privations and negative concepts. The intellect understands darkness in the complex negative concept “absence of light.” The intellect understands “infinity” in a complex concept denying the limits of a perfection, and God’s infinity in the more complex concept denying the limits of any and all perfections. Presumably, the complex concept with which the human intellect represents God as wise begins with the concept of creaturely wisdom and adds to the concept both the relative concept of “more eminent than” and the negative concepts of “without limits, potency, or composition.”

Capreolus’s Answer to Scotus’s “Certain and Doubtful” Argument Returning to Scotus, Capreolus reminds his reader that he had taken up objections from Scotus in Distinction 2, question 1 to the position that no name is said univocally of God and creature and now he will take up those

30

Ibid., 175a: “”quando per accidentia rei vel effectus ejus sufficienter homo ducitur in notiam quod quid est, sicut exemplificat de materia et de pane; tunc enim, in illo casu, quis potest habere conceptum simplicem, proprium, et distinctum de quidditate rei, puta pantis, si habeantur tales tres conceptus, de quibis iste loquitur.”

146

The Semantic Unity of the Analogous Concept according to Capreolus

objections again under a different form.31 For this article’s purposes, we are interested in the first argument from Scotus against Capreolus’ fifth conclusion; namely, “no name imposed by us is said univocally of God and creatures” (Quinta conclusio est quod nullum nomen a nobis impositum, dicitur univoce de Deo et creatura).32 Capreolus offers a rather detailed version of the “certain and doubtful” argument, taken from Scotus’ Ordinatio 1, 3. I quote the skeleton of the argument here: Every intellect certain about one concept and doubtful about different things (de diversis) has one concept about which it is certain, [which is] other than the concepts about which it is doubtful. But the intellect of the wayfarer can be certain about God, that He is a being while doubting about [whether God is] a finite or infinite being, created or uncreated. Therefore the concept of being about God, is other than the concept about the latter and the former [i.e. finite being and infinite being; created being and uncreated being], and so [the concept of being is] neither [the latter nor former concept] in itself and is included in both of them….33

Rather than responding to this objection as his Thomists predecessors had done by denying any simple unity to the concept of being, Capreolus denies instead that the argument proves the one concept of being is a univocal concept. He writes as follows: To the first argument of Scotus against the fifth conclusion, with all its confirmations, it is said that it does not suffice for some name to be said univocally about two things that it signifies one concept, common to those two in any way; but it is required that that concept is participated by them equally, and not more perfectly by one than by the other; so that if that concept is more perfectly a likeness of one than of the other, on account of 31

Ibid., 173a: “Contra quintam conclusionem arguitur secundum Scotum. Licet ejus argumenta fuerint alias recitata (dist. 2, q. 1), non tamen sub eadem, qua nunc, forma.” The arguments are presented much more briefly and in greater numbers in distinction 2, q. 1, a. 2. 32 Capreolus, Defensiones 1, dist. 22, q. 1, a. 1 (Frankfurt, vol. 2, 172a). 33 Ibid., 173a: “Omnis intellectus certus de uno conceptu, et dubius de diversis, habet conceptum de quo est certus, alium a conceptibus de quibus est dubius. Sed intellectus viatoris potest esse certus, de Deo, quod sit ens, dubitando de ente finito vel infinito, creato vel increato. Ergo conceptus entis de Deo, est alius a conceptu illo et isto, et ita neuter ex se, et in utroque istorum includitur. Ergo est unicus.” Capreolus goes on to quote Scotus’s defence of the Major and Minor premise and the examples of the first philosophers certain that the first principle is a being but uncertain if it is fire or water, finite or infinite, created or uncreated.

Domenic D’Ettore

147

this, a name signifying that concept will not be univocal to those two about which it is said. Yet so it is for every concept which the wayfarer has about God. For every concept formed by it common to God and to creature represents God more imperfectly than the creature. For the concept of wisdom represents created wisdom more perfectly than uncreated; because it is taken immediately from created wisdom; and therefore it represents it more distinctly, as regards itself and its modes, than uncreated wisdom; therefore this name “wisdom” is not properly univocal. Unless you say that for univocation one concept of any kind suffices; but then it is a question about the name; because then that which I call analogous, you call univocal.34

Capreolus claims that a name only counts as univocal if the concept it signifies represents both the things named equally. The names said of God and creatures represent creatures more perfectly than God because the concept through which the name signifies is taken from created wisdom not from divine wisdom. In effect, the concept of wisdom is abstracted from created wisdom and never abstracted from divine wisdom. Divine wisdom is not known in itself, but only through creaturely wisdom. So even though the intellect uses the same concept of wisdom when it represents human and when it represents divine wisdom, the concept “wisdom” is not univocal to God and creatures. At beginning of the 15th Century changing the ground rules on what counts as univocity and what counts as analogy was already a familiar move in the on-going debate. Scotus’s own account of univocity does not exactly

34 Ibid., 175b: “Ad primum Scoti contra quintam conclusionem, cum omnibus confirmationibus suis, dicitur quod ad hoc quod aliquod nomen dicatur univoce de duobus, non sufficit quod significet conceptum unicum, communem illis duobus quocumque modo; sed requiritur quod ille conceptus aequaliter participetur ab eis, et non perfectius ab uno quam ab alio; ita quod si conceptus ille sit perfectius similitudo unius quam alterius, non, propert hoc, nomen significans illum conceptum erit univocum illis duobus de quibus dicitur. Sic autem est de omni conceptu quem habet viator de Deo. Nam omnis conceptus ab eo formatus, communis Deo et creaturae, imperfectius repraesentat Deum quam creaturam. Conceptus enim sapientiae repraesentat perfectius sapientiam creatam quam increatam; quia est immediate sumptus a sapientia creata; et ideo repraesentat eam distinctius, quaod se et modos ejus, quam sapientiam increatam; ideo hoc nomen sapientiae non est proprie univocum. Nisi dicas quod ad univocationem sufficit unicus qualiscumque conceptus; sed tunc est quaestio de nomine; quia tunc illud quod ego dico analogum, tu dicis univocum.”

148

The Semantic Unity of the Analogous Concept according to Capreolus

match up with Aristotle’s Categories 1.35 And Scotus’ arguments for the univocity of the concept of being presuppose a name can be said univocally of two things which do not agree in genus; a position explicitly denied by Thomas Aquinas in Prima Pars 13.5.36 Peter Auriol likewise had expressed his disagreement with Scotus on the univocity of being by explaining that univocity requires one distinct concept whereas Scotus had only shown there to be one concept indistinctly.37 John Capreolus’ own explicit explanation of unequal participation in a ratio appears in the context of interpreting Thomas Aquinas In 1 Sent., dist. 19, q. 5, a. 2, ad 1, where Thomas Aquinas presents his controversial “three ways something is said according to analogy” and concludes that 35

See John Duns Scotus, Ordinatio I, dist. 3, pars 1, q. 1-2, n. 26 (VC, vol. 3, 25), and John Duns Scotus, Quaestiones Super Libros Metaphysicorum Aristotelis Libri 4, q. 1, I. – Ad quaestionem, n. 27-28 in Opera Omnia, ed. R. Andrews (St. Bonaventure, NY: Franciscan Institute, 1997), vol. 3, 301. See also See Dumont, “Transcendental Being: Scotus and the Scotists,” Topoi 11, no. 2 (Sept 1992), 137: “Scotus defines a univocal concept as one whose unity is sufficient to cause a contradiction when asserted and denied of the same thing or, alternatively, one whose unity enables it to function as a middle term in a syllogism. These definitions of univocity, while not incompatible with, differ from the standard one taken from the opening lines of Aristotle's Categories, according to which something is univocal if both its name and essence or definition (ratio) are one. Scotus prefers instead to give a functional definition of the univocity he is about to demonstrate, namely, the concept of being will be univocal enough to guarantee the law of non-contradiction and to avoid the fallacy of equivocation in reasoning about God and substances.” 36 Thomas Aquinas, Prima Pars 13, 5, sed contra et ad tertiam. See also Prima Pars 3.4. 37 For these points, see Petrus Aureoli, Scriptum in I Sententiarum, dist. 2, a. 1-4 (Roma: Vaticana, 1596), 118b-121b, as well as Peter Auriol, Reportatio in I Sententiarum, dist. 2, p. 1, q. 1-2, edited by Stephen Brown, in Traditio 50 (1995), especially pp. 212-216 and 228. See also Petrus Aureoli, Scriptum in I Sententiarum, dist. 22, a. 2 (Roma: Vaticana, 1596), 517. For a brief exposition of Auriol’s position, see Stephen Brown, “Scotus’s Univocity in the Fourteenth Century,” in De doctrina Ioannis Duns Scoti: Acta Congressus Scotistici Internationalis (Rome: Scotist Commission, 1968): 35-41. For a more thorough account, see Brown’s work “The Unity of the Concept of Being in Peter Aureoli's Scriptum and Commentarium (with a critical edition of the Commentarium text)” (PhD diss., University of Louvain, 1964). See also the work of W. Goris, “Implicit Knowledge - Being as First Known in Peter of Oriel,” Recherches de Theologie et Philosophie Medievales 69, no. 1 (2002): 33-65.

Domenic D’Ettore

149

truth, goodness, wisdom, and the like have a “common nature” in God and creatures but differ “according to a ratio of greater and lesser perfection.”38 Capreolus explains Thomas’s text as follows: It is clear that the intentiones carried by these names commonly said about God and creatures are in God and in creatures in that way in which an intentio is said to be in a thing. That is, [the intentiones or rationes are in God and creatures] not subjectively, but as in the foundation of their truth. Now those intentiones have a more perfect foundation in God than in creatures even though it holds that they represent God more imperfectly than creatures because they are immediately taken from creatures and not God.39 38

See Thomas Aquinas, In 1 Sent., d. 19, q. 5, a. 2, ad 1 (Mandonnet, vol. 1: 492): “Ad primum igitur dicendum, quod aliquid dicitur secundum analogiam tripliciter : vel secundum intentionem tantum, et non secundum esse; et hoc est quando una intentio refertur ad plura per prius et posterius, quae tamen non habet esse nisi in uno; sicut intentio sanitatis refertur ad animal, urinam et dietam diversimode, secundum prius et posterius ; non tamen secundum diversum esse, quia esse sanitatis non est nisi in animali. Vel secundum esse et non secundum intentionem; et hoc contingit quando plura parificantur in intentione alicujus communis, sed illud commune non habet esse unius rationis in omnibus, sicut omnia corpora parificantur in intentione corporeitatis. Unde Logicus, qui considerat intentiones tantum, dicit, hoc nomen, corpus, de omnibus corporibus univoce praedicari : sed esse hujus naturae non est ejusdem rationis in corporibus corruptibilibus et incorruptibilibus. Unde quantum ad metaphysicum et naturalem, qui considerant res secundum suum esse, nec hoc nomen, corpus, nec aliquid aliud dicitur univoce de corruptibilibus et incorruptibilibus, ut patet, X Met., text. 5, ex Philosopho et Commentatore. Vel secundum intentionem et secundum esse; et hoc est quando neque parificatur in intentione communi, neque in esse; sicut ens dicitur de substantia et accidente; et de talibus oportet quod natura communis habeat aliquod esse in unoquoque eorum de quibus dicitur, sed differens secundum rationem majoris vel minoris perfectionis. Et similiter dico, quod veritas, et bonitas, et omnia hujusmodi dicuntur analogice de Deo et creaturis. Unde oportet quod secundum suum esse omnia haec in Deo sint, et in creaturis secundum rationem majoris perfectionis et minoris; ex quo sequitur, cum non possint esse secundum unum esse utrobique, quod sint diversae veritates.” 39 Capreolus, Defensiones 1, dist. 2, q. 1, a. 1 (Frankfurt, vol. 1, 125a): “Ex quibus patet quod intentiones importatae per haec nomina communiter dicta de Deo et creaturis sunt in Deo et in creaturis eo modo quo intentio dicitur esse in re, non scilicet subjective, sed sicut in fundamento veritatis suae; licet illae intentiones perfectius fundamentum habeant in Deo quam in creaturis; cum hoc tamen stat quod imperfectius repraesentant Deum quam creaturas : quia scilicet sunt immediate sumptae a creaturis et non a Deo.”

150

The Semantic Unity of the Analogous Concept according to Capreolus

So names said through one concept unequally participated are said analogously not univocally, and a concept is unequally participated when it represents one analogate more perfectly than another. In the case of names said of God and creatures, one concept represents a creature and it represents God less perfectly than the creature. Joining this text to what we found in Distinction 22, Capreolus tells us that the human intellect can produce a composite concept proper to God by adding to its concept of a creaturely perfection other concepts also derived from creatures involving negations, causality, and eminence.40 So there is an overlap of semantic content between the concept through which the intellect signifies wise creatures and the concept through which the intellect signifies divine wisdom. But Capreolus denies that this overlap is univocal because it is unequally participated.

Does Capreolus Found the Semantics of Analogy on a Univocal Common Core? Cross’s Scotus founds the semantics of analogy on overlapping common semantic core content, which, since (Scotus holds) any common semantic content is univocal, entails that the semantic foundation for analogy is univocity. Now we have seen that in Capreolus’s account of names said analogously of God and creatures, the concept abstracted from the creature must still carry some of the same semantic content within the composite concept applied properly to God; otherwise, the composite would signify the negation, causality, or eminence of nothing. The difference between Capreolus’s common concept and Scotus’s in analogy hinges on the influence the “composing” concept has on the common concept. Within Cross’s account of Scotus, adding a mode, such as finite or infinite, does not alter the semantic content of the concept with which it is composed. That is, the additional concept of “infinite” does not 40

Besides the texts already quoted to make this point, see also the response to the second objection from Scotus to the fifth conclusion (Frankfurt, vol. 1, 175b): “Dico tamen quod intellectus viatoris potest formare conceptum compositum de sapientia divina, puta concipiendo actum sapientificandi, independentem, non inhaerentem, etc. Et tamen ille conceptus adhuc esset, secundum unam partem sui, similitudo sapientiae creatae, superexcellens; et ideo nunquam esset univocus isti et illi.”

Domenic D’Ettore

151

in any way alter the concept “wisdom” when the two are composed in the complex concept “infinite wisdom.” “Wisdom” carries exactly the same semantic content in “infinite wisdom” and in “finite wisdom.” Capreolus treats the additional concept quite differently in Defensiones 1, distinction 22. When Capreolus forms the proper concept “divine wisdom,” the semantic content of the concept “wisdom” does not go unchanged. Instead of Scotus’ modes, the concepts added to “wisdom” include non-inherence, independence, and others expressing negation, causality, or eminence. Now “non-inherence,” for example, removes the genus of quality from the meaning of the concept “wisdom.” When the concept “wisdom” is taken from creatures, its qualitative character is included within the concept. The genus and difference are both essential constituents of the concept “wisdom.” The addition of “non-inherent” is necessary to apply the concept to God without falsely attributing composition to God. The “wisdom” attributed to God does not differ (only) by the modes “finite” and “infinite,” from the “wisdom” attributed to creatures. Something essential to “wisdom” as abstracted from creatures, namely its categorical genus, is removed or denied in conceiving “divine wisdom.” The “wisdom” then which Capreolus conceives of and attributes to God is quite specifically not identical to the “wisdom” he abstracts from creatures. So while Capreolus affirms that there is something in God corresponding to the concept of “wisdom” in creatures,41 what this wisdom in God is remains unknown to the created intellect. Scotus’s modes compose with the initial concept without changing its meaning or semantic content. Capreolus’ additions serve no other purpose than to show how one would be misled if one took the initial concept according to its proper meaning (i.e. the meaning it has when applied to creatures).

41

See especially Capreolus, Defensiones 1, dist. 8, q. 4, a. 1 (Frankfurt, vol. 1, 375a): “In ipsa re quae Deus est, est aliquid correspondens rationibus et conceptionibus attributorum. Ergo illae rationes sunt in Deo.” “In the very thing which God is, there is something corresponding to the rationes and conceptions of the attributes. Therefore, those rationes are in God.” Capreolus continues in the same passage to explain that the way a ratio is in a thing outside the soul is as the foundation of its truth, “…quia in re est aliquod respondet illi rationi, per similitudinem faciens ejus veritatem. Et ideo tales rationes dicuntur esse subjective in intellectu, sed in re sicut in fundamento suae veritatis.”

152

The Semantic Unity of the Analogous Concept according to Capreolus

The difference between Scotus’s and Capreolus’ semantics of analogy can also be seen by considering non-theological analogies. Scotus and Capreolus both treat analogy within the context of discussing names such as wisdom, goodness, being, etc. said of God and creatures. Yet analogy is useful in Theology and Natural Theology because it is something we already are familiar with outside Theology and Natural Theology. So the question arises, do the semantic principles which Cross’s Scotus and Capreolus invoke in Natural Theology cross over to apply to more mundane cases of analogous predication? Can their principles account for the old SAT and GRE analogies, such as “sharp” said of a knife and a comedian, or “falling” said of a rock pushed from a cliff and of a share price in a stock exchange, or of “star” said of Alpha Centuri and of Sidney Crosby, or of “seeing” said of the act of performed through the eyes and optic nerve and the act of recognising a pattern of historical events or the reasoning of an argument?42 The concept of “sharp” abstracted from the knife is clearly not identical to the concept of “sharp” abstracted from the comedian’s wit. One concept signifies the quality of a knife which allows it to cut well; another concept signifies a comedian’s ability to quickly and amusingly address an audience especially at someone’s expense. The similarity or unity between concepts signifying the knife’s capacity to affect a steak and a comedian’s capacity to affect the guy in the red shirt is of a different kind from the unity of the concept “animal” abstracted from sharks and beavers.43 They are also not entirely equivocal in the way the concepts of “bat” abstracted from Louisville Sluggers and of “bat” abstracted from the small flying mammals are purely equivocal. What can Capreolus and Scotus tell us about analogous “sharp,” “falling,” “star,” or “seeing”? Cross’s Scotus would have to say there is some one common univocal concept contained within the different complex concepts (or perhaps deny that there is analogy here at all). It is not apparent to me what the univocal core content would be. In the case of “sharp,” the univocal content could not be “cutting,” since that would just push the question further back. In 42 It may be objected that these are cases of metaphor rather than analogy. I maintain that these are analogies and not metaphors because there is no need to be familiar with the meaning of one concept or the use of one term in order to understand the other concept or the use of the other term. 43 To the doubtful, this can be proven by testing the effects of sharp knives on fellows in red shirts and of sharp comedians on steaks.

Domenic D’Ettore

153

what common sense of “cutting” do knives cut steaks and comedians cut audience members? The same would apply to “moving down” in the case of analogous “falling.” There is no apparent common meaning of “moving down” in “falling rocks” and “falling stocks.”44 The search for an overlapping univocal content, I suggest, would go on infinitely. What about Capreolus? Capreolus’ analogies involve the concept proper to one analogate serving as the imperfect representation of another analogate.45 In the cases Capreolus discusses (such as prime matter and the divine attributes), one of the analogates simply speaking cannot be properly represented by the human intellect, so the concept of one has to serve as the concept of the other if the other is to be known and signified at all. This necessity is clearly not present in the case of the mundane examples of analogy. Still, he can again say the concept proper to one is able to serve as an imperfect concept of the other, and vice versa. Indeed, the extent of the imperfection of the representation can reinforce his point that the human intellect does not achieve quidditative knowledge of God through analogy. That is, just as we can only know very imperfectly what it means for a comedian to be sharp if it is explained to us by reference to sharp knives, we can only know very imperfectly what it is for God to be wise by reference to Socrates’s wisdom. This puts Capreolus’s semantics of analogy in a different (and I think better) position than Cross’s Scotus to explain the familiar analogies.

Conclusion John Duns Scotus and his followers challenged the Thomists to explain how analogous signification could occur unless there is a univocal foundation to analogous concepts. John Capreolus answered this challenge by granting to the Scotists that the analogous concept does indeed have a formal unity while denying that this formal unity entails univocal content when the concept is signified analogously. He does this in Natural Theology by affirming that it is the same concept which is abstracted from creatures which is used in the signification of Divine perfection, and by denying that this one concept represents the analogates equally. Properly 44

I expect Cross and others could come up with better options than I have suggested here. I write this hoping for better light on the subject. 45 That is, the kind of analogy suitable for naming God and creatures has this kind of semantic unity.

154

The Semantic Unity of the Analogous Concept according to Capreolus

applied to God, the one concept is altered by the composing concepts, such that it no longer carries exactly the same semantic content as it does when it is properly applied to creatures. This alteration and inequality of representation renders the common content in the analogous concept nonunivocal. John Capreolus, thus, provides the Thomist tradition with an alternative to Scotus’s “reductionist” explanation of the semantics of analogy. Within the Thomist tradition, Capreolus’ account has the additional merit of preserving the non-quidditative character of the human intellect’s natural knowledge of God. It likewise provides, as I have suggested, a more adequate explanation of mundane analogies than Cross’s Scotus’s semantics of analogy. That said, if the Scotist does not accept Capreolus’ position that a concept is univocal only when it represents equally, the Scotist is likely to regard Capreolus as one more philosopher whose explanation of analogy is univocity.

CAN WE SPEAK ABOUT THAT WHICH IS NOT? ACTUALISM AND POSSIBILISM IN ANALYTIC PHILOSOPHY AND SCHOLASTICISM LUKÁŠ NOVÁK

1. Introduction The question raised by the title may seem bizarre: for it seems quite evident that the answer must be “yes”, for both a priori and a posteriori reasons. It is easy to verify empirically that one can speak about nonexistent things. For example, many people talk about hobbits nowadays, even call them by their names; and as for the a priori rationale – wouldn’t it be inconsistent to say about that which is not that it is impossible to say anything about it?1 In spite of these considerations, both of the two great philosophical traditions that combine metaphysical speculation with logico-semantic enquiry, viz. high scholasticism and analytic philosophy, notably feature thinkers or even entire schools of thought, sometimes dominant, that maintain the opposite. That which in no way is, they contend, cannot be truly spoken about. I will label this thesis for future reference as the Principle of Reference (PR): (PR) It is impossible to refer to that which is not.2

In this paper, after some preliminary clarifications, I will first document several different incarnations of this belief, first in the analytic tradition 1

This is just a version of the traditional proof of entia rationis (formulated e.g. by Suárez), dubbed by Daniel D. Novotný “the ontological argument for entia rationis”; see Daniel D. Novotný, Entia rationis from Suárez to Caramuel (New York: Fordham University Press, 2013), 51 (and 173). 2 The meaning of “is” is deliberately left vague at this point.

156

Can We Speak About That Which Is Not?

and then in scholasticism, and after that I will set out the theory of John Duns Scotus, amounting to what appears to be the first clear and wellargued rejection of the PR.3 In the final part of the paper I consider how these findings can help us to understand much more precisely the nature of the classical philosophical dispute between actualism and possibilism – for it allows us to perceive its ambiguity, rarely if ever noticed in the contemporary analytical philosophy.

2. Some clarifications What do I mean by the phrase “speak about/of something”? I hope that sufficiently precise meaning(s) of this phrase will become clear in the course of the analysis of the positions of the individual authors; but minimally, I will take the “possibility to speak about/of something” to require the capacity of that (non-existent) something to be singled out or grasped as an individual, by means of our thought or speech, in order to become the subject of a meaningful statement or judgement. The possibility to “speak about something” therefore involves primarily the possibility to refer to that thing; or even the possibility to refer to it rigidly (in Kripke’s sense), in a modal or generally (hyper-)intensional context. That means, the possibility to fix the respective individual across possible worlds, regardless of how it is characterized, and make de re statements about it. While it is beyond doubt that somehow, at least indirectly, non-existing things can be spoken about (the proof of which being this sentence), the answer to the question whether it is possible to single out a non-existing individual and meaningfully ascribe a property to it, is contentious.

3

I am not the first to write on the rejection of the PR in scholasticism. Gyula Klima made certain similar points – see esp. “Existence and Reference in Medieval Logic”, in New Essays in Free Logic, Applied Logic Series Volume 23 (Springer, 2001) (most of the content of this paper is reproduced in his John Buridan (Oxford University Press, 2008)); and “Being and cognition”, in Neo-Aristotelian Perspectives in Metaphysics, ed. D. D. Novotný and L. Novák (New York: Routledge, 2014). However, Klima maintains that the rejection of the PR was pretty general in scholasticism, and interprets the existence of logico-semantic devices like ampliation or natural supposition in scholasticism as evidence for that. I will discuss this point below (see ch. 6, and esp. notes 21, 23, 47, 49 and 53).

Lukáš Novák

157

The dispute between actualism and possibilism in modern analytical philosophy is a dispute over the ontological status of that which is merely possible, of the un-actualized possible worlds or ways how things might have been but are not. We shall return to this controversy in the final part of this paper; here it suffices to recall that actualism ascribes a certain ontological status only to that which is actual, attempting to “construct” the alternative possible worlds as other possible configurations of actual elements, to the effect that what is possible is derived from and conditioned by that which is actual. Possibilism, on the other hand, conceives of possibility as of an ontological status in its own right, presupposed by actuality, so that it is not the possible that depends on the actual, but vice versa. Thus, the possibilists maintain that although the unactualized possibilities do not actually exist, they nevertheless are there, in some irreducible sense; the actualists oppose this by claiming that whatever there is, is actually there and so actually exists. In order to avoid unnecessary complications, I will limit the scope of the following reflections to that which is not of the domain of the possible. I will leave aside things that, in addition to not existing, are incapable of existence, and refrain from exploring our prospective ability to refer to them by name. Moreover, I will leave aside, for brevity’s sake, the question of how possibility itself is to be understood – let everyone feel free to employ their own intuitions.

3. Gottlob Frege Analytical philosophy, one can say, imbibed the PR with its mother’s milk. Gottlob Frege, the father of analytical philosophy and of modern logic, concluded in his reflections on existence that there is no informative concept of existence that could be predicated of individuals: if we are to mean something non-trivial by existence, we must treat it not as a predicate of individuals, but as a statement of the non-emptiness of the extension of a concept (which can be expressed, in the present version of formal logic, by means of the existential quantifier): Here I would retort: If “Sachse exists” should mean “The term ‘Sachse’ is not an empty sound but it stands for something”, then it is correct to say that the condition that “Sachse exists” must be satisfied. This, however, is no new premise, but a self-evident presupposition of all our words. The rules of logic always presuppose that the words used are not empty, that the sentences express judgements, that we are not playing with mere words. Given that “Sachse is a man” is an actual judgement, the word

158

Can We Speak About That Which Is Not? “Sachse” has to stand for something; in which case I do not need any further premise to infer “There are men” from that. The premise “Sachse exists” is superfluous, as long as it means nothing over and above that selfevident presupposition of all our thought. Or can you produce an example where a sentence of the form “A is B” is meaningful and true, A being a name of an individual, and yet “There are B’s” is false?4

The existence of individuals is, according to Frege, a self-evident presupposition, implicit in the very rules of logic. If the names we use to refer to individuals did not stand for anything real, our utterances would have no meaning and would be reduced to a mere play with words. Let us take note: the presupposition that names stand for something, i.e., that they are not empty sounds, is automatically interpreted by Frege to mean that they stand for something that exists. What does not exist, therefore, cannot be named or referred to. Frege poses the rhetorical question as to whether it is possible to produce an example of a meaningful true statement of the form “A is B”, where A is a proper name, but there are no B’s. The challenge is not met by Frege’s opponent in the dialogue as preserved, but in case we attempted to suggest e.g. the sentence “Smaug is a dragon” as an obvious counterexample (assuming that, of course, there are no dragons), according to his principles Frege would be bound to saying that this is “not a real (i.e., meaningful) sentence”. This counterintuitive result was to be mitigated in a later phase of the development of Frege’s thought, as the distinction drawn between Sinn and Bedeutung enabled, under certain circumstances, expressions (including sentences) to have a Sinn 4

“Hierauf würde ich sagen: Wenn ‚Sachse existiert‘ heißen soll ‚Das Wort »Sachse« ist nicht ein leerer Schall, sondern bezeichnet etwas‘, so ist es richtig, daß die Bedingung ‚Sachse existiert‘ erfüllt sein muß. Dies ist aber keine neue Prämisse, sondern die selbstverständige Voraussetzung bei allen unseren Worten. Die Regeln der Logik setzen immer voraus, daß die gebrauchten Worte nicht leer sind, daß die Sätze Ausdrücke von Urteilen sind, daß man nicht mit bloßen Worten spiele. Sobald ‚Sachse ist ein Mensch‘ ein wirkliches Urteil ist, muß das Wort ‚Sachse‘ etwas bezeichnen und dann gebrauche ich eine weitere Prämisse nicht, um daraus zu schliesen, ‚Es gibt Menschen‘. Die Prämisse ‚Sachse existiert‘ ist überflüssig, wenn sie nichts etwas anderes bedeuten soll, als jene selbstverständige Voraussetzung bei allem unserem Denken. Können Sie ein Beispiel angeben, wo ein Satz von der Form ‚A ist ein B‘ einen Sinn hat und wahr ist, A der Name eines Einzelnen ist, während ‚es gibt B’s‘ falsch ist?” – Gottlob Frege, “Dialog mit Pünjer über Existenz” (between 1879 and 1884), in Schriften zur Logik und Sprachphilosophie: Aus dem Nachlass (Meiner Verlag 2001), 99, p. 11–12. Italics and emendation of “etwas” to “nichts” mine. All translations are mine as well.

Lukáš Novák

159

despite lacking a Bedeutung. Nevertheless, the fundamental principle of Frege’s theory, viz. that objects capable of being judged about are, trivially, exclusively existing objects, was never abandoned by Frege, nor did he ever consider it in the least controversial. Quite the opposite – Frege regarded it as empty and tautological, since the term “existing”, insofar as it is applied to individuals, is devoid of any content and as such it does not impose any extensional narrowing: If we attribute any content to the verb “to be”, to the effect that the sentence “A is” is neither superfluous nor self-evident, we shall have to concede that the negation of “A is” is, under certain circumstances, possible, viz. that there are subjects to which being must be denied. Then, however, the notion of “being” generally won’t be suited to be used as an interpretation of the meaning of “there is” any more, according to which “there are B’s” would be equivalent to “some being falls under the concept of B”. For if we applied this interpretation to the sentence “There are subjects to which being must be denied”, we would obtain the sentence “Some being falls under the concept of non-being”, or “Some being is not”. This is unavoidable, as soon as one ascribes any content whatsoever to the concept of being. If the interpretation that “there are B’s” means the same as “some being is B” is to be correct, then it is simply necessary that “being” be understood as conveying something completely self-evident.5

This fundamental intuition of Frege underlies a strong tradition of classical analytical thought, represented by figures like B. Russell, R. Carnap, W. V. O. Quine or P. F. Strawson (we will discuss most of them shortly). And although their approaches often differ or even contradict each other,

5

“Wenn man dem Worte ‚Sein‘ einen Inhalt geben will, so, daß der Satz ‚A ist‘ nicht überflüssig und selbstverständlich ist, wird man dazu genötigt, zuzugeben, daß die Verneinung des Satzes ‚A ist‘ unter Umständen möglich ist; d. h. daß es Subjekte gibt, denen das Sein abgesprochen werden muß. Dann aber ist der Begriff des ‚Seins‘ nicht mehr allgemein geeignet, zur Erklärung des ‚es gibt‘ zu dienen in der Weise, daß ‚Es gibt B’s‘ gleichbedeutend ist mit ‚Einiges Seiende fällt unter den Begriff B‘; denn wenden wir diese Erklärung an auf den Satz ‚Es gibt Subjekte, denen das Sein abgesprochen werden muß‘, so erhalten wir ‚Einiges Seiende fällt unter den Begriff des Nichtseienden‘ oder ‚Einiges Seiende ist nicht‘. Darüber ist in keiner Weise hinwegzukommen, sobald man dem Begriff des Seienden irgendwelchen Inhalt, sei es welchen es sei, geben will. Es ist eben nötig, wenn die Erklärung ‚es gibt B’s‘ ist gleichbedeutend mit ‚Einiges Seiende ist B‘, richtig sein soll, daß unter Sein etwas volkommen Selbstverständliches verstanden wird.” – ibid., 20-21; italics mine.

160

Can We Speak About That Which Is Not?

this fundamental Fregean intuition has never been called into doubt in their respective philosophies.

4. Bertrand Russell and Alexius Meinong Frege tackled the problem of being or existence, which had once fallen under the sovereignty of metaphysics or ontology, from the point of view of logical semantics – as he considered it to be in principle a problem of semantics. However, already Bertrand Russell, the apostle and critical continuator of Fregeanism, had explicitly rehabilitated the metaphysical aspect of the issue: viz. the question, what can actually be said, in a reasonable sense, to be. His critique of Alexius Meinong has become a classic in this respect: a paradigmatic incarnation of the actualismpossibilism dispute in analytical philosophy. Meinong’s position can be understood as a stubborn and thoroughgoing implementation of the PR: there can be no real judgement if there is nothing to be judged about, therefore, any judgement presupposes something, some pre-given object. On the other hand, in accordance with the common-sense practice, Meinong does not impose any limits on the sphere of what can be judged about: his “theory of objects” therefore assumes as “objects” not just actually existing beings (such as Pope Francis), but also merely possible beings (such as Frodo Baggins), inconsistent (a round square) or incomplete (a man with no specified weight) entities, and other items. All objects, according to Meinong, share a certain basal ontological status, independent of any cognitive operations – a kind of minimal mode of being called Außersein.6 So Meinong in fact 6 There exists a controversy concerning the true meaning of Meinong’s theory. Meinong was misinterpreted by the Russell-Quine tradition which conflated his notions of Außersein and subsistence, yet even keeping these notions distinct does not solve the question of whether Außersein prescinds from any being whatsoever or whether it is some minimal kind or degree of being. The former interpretation is emphatically defended by Bill Vallicella (cf. several contributions to his blog, Maverick Philosopher, listed under “Meinong Matters” at ‫ۃ‬http:// maverickphilosopher.typepad.com/maverick_philosopher/meinong-matters/‫ۄ‬. For the other view, see J. Marek, “Alexius Meinong”, in The Stanford Encyclopedia of Philosophy (Fall 2013 Edition), ed. Edward N. Zalta, URL = ‫ۃ‬http://plato.stanford.edu/archives/fall2013/entries/meinong/‫ۄ‬, ch. 4.3.1: “Meinong repeatedly ponders the question of whether outside-being is a further mode of being or just a lack of being […]. He finally interprets outside-being as a

Lukáš Novák

161

agrees with Frege that any meaningful expression stands for something that is – he just refuses to follow Frege in identifying this self-evident or matter-of-course being with actual existence and interprets it instead as the much broader notion of Außersein. Meinong is thus able to accommodate the common sense notion that we can refer to that which does not exist, merely subjoining that not to exist is not the same as not to be at all. In this way his system is bound to allow for various modes of being. Russell’s critique of Meinong has two facets: a logical one and a metaphysical one. From the logical point of view Russell objects that Meinong’s theory is inconsistent. Technical details of this way of criticism can be left aside here – we may just note that the inconsistency-objection seems to be valid against Meinong’s original conception; on the other hand, several consistent neo-Meinongian systems have been worked out that sidestep the technical problems of naïve Meinongianism (whether it can actually be attributed to Meinong or not).7 The metaphysical ingredient of Russell’s critique consists in a strong intuition of his, which he calls a “robust sense of reality”. “Logic,” he says, “must no more admit a unicorn than zoology;”8 when we say that borderline case of a kind of being. Every object is prior to its apprehension, i.e., objects are pre-given /vorgegeben/ to the mind, and this pre-givenness is due do the (ontological) status of outside-being.” For the purposes of this paper, I will ascribe this latter position to Meinong; for it in any case seems to be more in accord with the way his theory was perceived and interpreted by the Russellian tradition. 7 Cf. Marek, ibid., ch. 4.4.4. 8 “It is argued, e.g. by Meinong, that we can speak about ‘the golden mountain,’ ‘the round square,’ and so on; we can make true propositions of which these are the subjects; hence they must have some kind of logical being, since otherwise the propositions in which they occur would be meaningless. In such theories, it seems to me, there is a failure of that feeling for reality which ought to be preserved even in the most abstract studies. Logic, I should maintain, must no more admit a unicorn than zoology can; for logic is concerned with the real world just as truly as zoology, though with its more abstract and general features. To say that unicorns have an existence in heraldry, or in literature, or in imagination, is a most pitiful and paltry evasion. What exists in heraldry is not an animal, made of flesh and blood, moving and breathing of its own initiative. What exists is a picture, or a description in words. Similarly, to maintain that Hamlet, for example, exists in his own world, namely, in the world of Shakespeare’s imagination, just as truly as (say) Napoleon existed in the ordinary world, is to say something deliberately

162

Can We Speak About That Which Is Not?

Hamlet has some fictitious existence, we in fact don’t mean anything over and above the uncontentious platitude that there are our perfectly real thoughts or mental images of Hamlet. A fiction is a fiction precisely because there is absolutely nothing it could be said to stand for – i.e., not even some object in pure Outside-Being. There is just one reality and just one kind of being, which is actual existence. From what has been said one might surmise that Russell is committed to rejecting the PR, since on the one hand he concedes that it is possible to talk about Hamlet, but on the other hand rejects that one should need to assume any kind of being whatsoever for Hamlet in order for such talk to be possible. As a matter of fact, however, Russell regards the PR as no less obvious than his mentor Frege. For that reason, he faces a serious theoretical problem – in his (and his followers’) eyes actually one of the most fundamental problems of philosophy – namely, how to reconcile the (apparently) self-evident PR with his “robust sense of reality”. As a solution to this problem, Russell presented, in “On Denoting” of 1905, his famous “Theory of Descriptions”, accompanied with his descriptive theory of proper names. General empty terms (like “unicorn”) are not a problem for Russell, since these, on the Fregean interpretation, do not denote individuals but sets or properties, and sets can, assumedly unproblematically, be empty and properties unexemplified. A problem is posed solely by singular terms, which seem to refer to individuals – since the individuals apparently referred to often happen not to exist. There are two kinds of singular terms in the natural language, according to Russell: proper names, and definite descriptions. As for the latter, Russell’s proposed solution consists in the denial that they are, logically speaking, singular terms capable of referring to individuals. The actual logical form of statements that contain definite descriptions differs radically from the surface grammatical form in the natural language, to the effect that e.g. the statement confusing, or else confused to a degree which is scarcely credible. There is only one world, the ‘real’ world: Shakespeare’s imagination is part of it, and the thoughts that he had in writing Hamlet are real. So are the thoughts that we have in reading the play. But it is of the very essence of fiction that only the thoughts, feelings, etc., in Shakespeare and his readers are real, and that there is not, in addition to them, an objective Hamlet.” – B. Russell, Introduction to Mathematical Philosophy, 2nd ed. (Allen&Unwin, 1920), 196.

Lukáš Novák

163

The present King of France is bald

in fact does not mention the (non-existing) present King of France and ascribe the property of being bald to him, but has the following form instead: Some x is the present King of France, whoever is the present King of France is identical to x, and x is bald.

No non-existing individual is being referred to here, not even seemingly – for it is not a singular statement at all but an assertion that the extension of a certain intersection of properties (viz. “to be the present King of France”, “to be identical to whoever is the present King of France”, and “to be bald”) is not empty. By asserting that this is the “real” logical form of sentences employing definite descriptions Russell avoided the inconvenience of having to cope with referring to non-existent items; at the same time, however, he transformed this (according to him merely apparent) reference to a non-existent item into an implicit assertion of its existence, i.e., exemplification of the cluster of predicates into which the definite description has been transformed.9 Thus asserted existence, therefore, is not the existence of an individual (such remains trivial for Russell just like it was for Frege), but existence qua instantiation of a

9

B. Russell, “On Denoting”, Mind, New Series vol. 56, no. 14 (1905): 481–482: “It remains to interpret phrases containing the. These are by far the most interesting and difficult of denoting phrases. Take as an instance ‘the father of Charles II was executed’. This asserts that there was an x who was the father of Charles II and was executed. Now the, when it is strictly used, involves uniqueness; we do, it is true, speak of ‘the son of So-and-so’ even when So-and-so has several sons, but it would be more correct to say ‘a son of So-and-so’. Thus for our purposes we take the as involving uniqueness. Thus when we say ‘x was the father of Charles II’ we not only assert that x had a certain relation to Charles II, but also that nothing else had this relation. The relation in question, without the assumption of uniqueness, and without any denoting phrases, is expressed by ‘x begat Charles II’. To get an equivalent of ‘x was the father of Charles II’, we must add ‘If y is other than x, y did not beget Charles II’, or, what is equivalent, ‘If y begat Charles II, y is identical with x’. Hence ‘x is the father of Charles II’ becomes: ‘x begat Charles II; and “If y begat Charles II, y is identical with x” is always true of y’. Thus ‘the father of Charles II was executed’ becomes: ‘It is not always false of x that x begat Charles II and that x was executed and that “if y begat Charles II, y is identical with x” is always true of y’.”

164

Can We Speak About That Which Is Not?

property, the non-emptiness of a set, or the property of a propositional function to be true on some arguments.10 As regards proper names, they are, according to Russell, true referring expressions, and as such they are subject to the PR – that is, they require the existence of their referent, else they cannot exist in the language. There are, however, very few such genuine proper names: according to Russell, the existence of the denotate is guaranteed solely for the indexicals “this” and “that”, given that they are used to refer to something we are immediately acquainted with. Other expressions normally considered to be proper names are as a matter of fact shorthands for definite descriptions and so can be eliminated as described above.11 In Russell and Meinong we can see that although they maintain contrary metaphysical stances, the theoretical source of their respective views is the same: viz. the PR. Both take for granted that properly speaking it is impossible to speak about that which is not. However, whereas Meinong was led by this assumption to ascribe some kind of being even to things which do not exist (as we are able to speak of them), Russell the “robust realist” seized the other horn of the dilemma and concluded that

10

Cf. ibid., 481. B. Russell, The Philosophy of Logical Atomism (1924; 6th ed., Open Court, 1996), 62: “What pass for names in language, like ‘Socrates’, ‘Plato’, and so forth, were originally intended to fulfil this function of standing for particulars, and we do accept, in ordinary daily life, as particulars all sorts of things that really are not so. The names that we commonly use, like ‘Socrates’, are really abbreviations for descriptions; not only that, but what they describe are not particulars but complicated systems of classes or series. A name, in the narrow logical sense of a word whose meaning is a particular, can only be applied to a particular with which the speaker is acquainted, because you cannot name anything you are not acquainted with. You remember, when Adam named the beasts, they came before him one by one, and he became acquainted with them and named them. We are not acquainted with Socrates, and therefore cannot name him. When we use the word ‘Socrates’, we are really using a description. Our thought may be rendered by some such phrase as, ‘The Master of Plato’, or ‘The philosopher who drank the hemlock’, or ‘The person whom logicians assert to be mortal’, but we certainly do not use the name as a name in the proper sense of the word. That makes it very difficult to get any instance of a name at all in the proper strict logical sense of the word. The only words one does use as names in the logical sense are words like ‘this’ or ‘that’.”

11

Lukáš Novák

165

individuals which are not trivially known to exist cannot be actually spoken about, i.e., cannot be referred to as subjects of singular propositions.

5. W. V. O. Quine and Peter F. Strawson Russell’s solution of the conflict between the PR and his “robust realism”, which is in fact just actualism, enjoyed wide acceptance by the empiricist mainstream of analytical philosophy in the first half of the twentieth century and the logic of Russell’s approach was further developed by other thinkers, e.g. Rudolf Carnap and W.V. O. Quine. Serious criticism of Russell’s approach only came in 1950, in the paper “On Referring” by Peter F. Strawson, who challenged Russell’s theory of descriptions; and Russell’s descriptive theory of proper names has only began to lose its absolutely dominant status since the 1960s, as the “New Theory of Reference” of Saul Kripke and others came to gain support. Let us have a look, before we turn to the Schoolmen, at how the PR is upheld not just in Quine’s development of Russell’s conception, but also in the criticism thereof proposed by Strawson. In his paper “On What There Is” from 1948, Quine in the first place aptly sums up Russell’s stance and once again, with wit and scorn, rejects Meinongianism;12 he emphasizes the Russellian thesis that existence is just one, and expressly synthesizes Russell’s theory of descriptions with his descriptive theory of proper names (a point not entirely explicit in 12

W. V. O. Quine, “On What There Is” (1948, repr. in From a Logical Point of View, Harvard University Press, 1980), 4: “Wyman’s overpopulated universe is in many ways unlovely. It offends the aesthetic sense of us who have a taste for desert landscapes, but this is not the worst of it. Wyman’s slum of possibles is a breeding ground for disorderly elements. Take, for instance, the possible fat man in that doorway; and, again, the possible bald man in that doorway. Are they the same possible man, or two possible men? How do we decide? How many possible men are there in that doorway? Are there more possible thin ones than fat ones? How many of them are alike? Or would their being alike make them one? Are no two possible things alike? Is this the same as saying that it is impossible for two things to be alike? Or, finally, is the concept of identity simply inapplicable to unactualized possibles? But what sense can be found in talking of entities which cannot meaningfully be said to be identical with themselves and distinct from one another? These elements are well-nigh incorrigible. By a Fregean therapy of individual concepts, some effort might be made at rehabilitation; but I feel we’d do better simply to clear Wyman’s slum and be done with it.”

166

Can We Speak About That Which Is Not?

Russell). However, in his reception of Russell’s ideas Quine makes one inconspicuous but crucial modification: any mention of “genuine” proper names is left out, to the effect that all proper names are in fact disguised descriptions. In this way, our language is finally devoid of any means whatsoever to genuinely and uniquely refer to a fixed individual; in Quine’s conception, individuals can only be reached via (quantified) variables. Thus, according to Quine, we are not only denied the capacity to speak of that which is not, but we also cannot, directly and by name, speak of individuals that are: this is manifested e.g. in Quine’s rejecting not just “possible entities”, but de re modalities in general.13 Still, the PR is upheld by Quine, albeit in a modified, de-metaphysicized form. Russell, driven by his “robust realism” (which, after all, is a metaphysical stance), had interpreted this principle normatively, i.e., as a requirement on a language that its individual constants should have a priori guaranteed reference to what Russell regarded as real things, and also in a way that ensured that the domain of its variables be confined to such real things (excluding, for example, fairies or unicorns). In Quine’s hands, however, the principle takes on a purely descriptive meaning, stating merely what kind of entities a given theory presupposes to exist – namely those that can figure as values of bound variables in that theory. In 13

Quine, ibid.: “Possibility along with the other modalities of necessity and impossibility and contingency, raises problems upon which I do not mean to imply that we should turn our backs. But we can at least limit modalities to whole statements. We may impose the adverb ‘possibly’ upon a statement as a whole, and we may well worry about the semantical analysis of such usage; but little real advance in such analysis is to be hoped for in expanding our universe to include so-called possible entities. I suspect that the main motive for this expansion is simply the old notion that Pegasus, for example, must be because otherwise it would be nonsense to say even that he is not.” There is a conceptual connexion between rigid reference and de re modalities: whenever a term refers rigidly to an individual, any modality possibly ascribed applies de re, and vice versa: in order to apply a modality de re, one needs rigid reference across possible worlds. Truly enough, as one anonymous referee pointed out, total rejection of de re modalities does not strictly follow from the rejection of direct and rigid reference to individuals, since even barring individual constants we still have the apparatus in modal predicate logic for general de re statements (e.g. “horses are necessarily animals”, (x)(Hx ĺ ƑAx)). On the other hand, it would be really strange to allow for general de re statements while repudiating individual ones (e.g. to concede that horses are necessarily animals, but eliminate the possibility of saying that Bucephalus is necessarily an animal).

Lukáš Novák

167

other words, the Principle features in Quine as his famous criterion of “ontological commitment”, which, since language has been purged of all directly referring expressions, lacks any a priori connexion to reality and becomes rather a matter of pragmatic choice. In spite of this pragmatic turn, the core meaning of the PR remains unchanged: the assumption (however pragmatic) that an entity can figure as a value of a bound variable in his theory is, according to Quine, equivalent to the assumption that such an entity exists; it is impossible to quantify over entities of which existence is not, eo ipso, assumed. Put more precisely: according to Quine the notion of existence just means the capability of featuring as a value of a bound variable. To assume that something exists is to assume nothing less, and nothing more: To be assumed as an entity is, purely and simply, to be reckoned as the value of a variable.14

Strawson, in his essay “On Referring”, criticizes Russell’s semantic approach in many respects; an implicit tenor of his critique is, however, a thoroughgoing rejection of Russell’s revisionism as regards natural language. According to Strawson, the alleged “logical forms” of sentences proposed by Russell simply do not capture what we actually do and mean when we use the language. In particular, Strawson points out the empirical fact that we, at least sometimes, do use definite descriptions to refer, or at least to attempt to refer, to an individual. Furthermore, according to Strawson it is false to say that by asserting a sentence such as “The present King of France is bald” we (albeit implicitly) assert the existence of that hairless monarch.15 The existence of the referent of a description used in this way is not stated or asserted but presupposed or implied (in a peculiar sense not equivalent to entailed), while by such a “presupposition” or “implication” of a statement Strawson means any condition that has to be fulfilled in order that the statement using the given sentence have a truthvalue at all.16 In case the existential presupposition is not fulfilled, a 14

Quine, ibid., 13. Unlike Russell, the courteous Strawson does not speak of baldness but wisdom in his example. 16 One of Strawson’s main points is that presuppositions and therefore also truthvalues are pragmatic or pragmatically dependent (they apply to sentence uses, not sentences). Others, however, have defended a semantic notion of presupposition – cf. e.g. J. Raclavský, “Semantic concept of existential presupposition”, Human 15

168

Can We Speak About That Which Is Not?

“failure of reference” occurs, which results in the statement being neither true nor false, even though the sentence stated is perfectly meaningful. In Strawson’s censure of Russell one can, once again, notice the implicit presence of the PR. Despite the thoroughness of Strawson’s criticism, he shares an elementary principle with Russell – viz. that a reference to that which is not cannot (successfully) happen. The core difference between Russell and Strawson is in how they react to the implications of this assumption: whereas Russell undertakes a far-reaching attempt at reforming the language (presented as a search for the “true” logical form) in order to make such failures of reference a priori impossible, Strawson simply describes the circumstances under which the failures occur and their eventual implications, as he perceives them. Neither of the two thinkers, however, called into question the common assumption, which is just another incarnation of the PR.

6. Scholastic Possibilism We have reviewed several key figures of the analytical tradition of the end of the nineteenth and first half of the twentieth centuries who without exception shared the fundamental belief called here the Principle of Reference. We have also seen that, except for A. Meinong, the “analytics” are led by the said principle not to expand the domain of being, but to reduce the possibilities of reference (this can be explained by the lasting disgrace into which Meinongianism fell in consequence of Russell’s criticism, and also by the generally anti-metaphysical bias of the early analytic philosophy).17 In scholasticism the matters are rather more complicated. Generally speaking, the scholastics lacked the Russellian revisionist attitude towards natural language, and therefore they rarely explicitly challenged the obvious capacity of the natural language to refer to non-existents. Their approach was, generally, to explain and analyse, not to correct language –

Affairs 21 (2011): 249–261. For the present purpose the precise demarcation of the respective competences of semantics and pragmatics in this issue does not seem to be necessary. 17 Klima in “Existence and Reference”, 212 (and in John Buridan, 158), makes a similar observation concerning a common assumption of Meinongians and antiMeinongians which amounts to what I call the Principle of Reference.

Lukáš Novák

169

and so the standard scholastic theory of supposition (the mediæval counterpart of reference) naturally allows (via devices like ampliation etc.) for reference to non-existents.18 Nevertheless, it would be rash to conclude from that alone that the PR was generally rejected by the scholastics. It seems to have been operative already before Scotus, often exerting its effects not directly on semantic theories, but rather on their ontological background. Its chief source was the Aristotelian-inspired doctrine of the transcendentals, according to which being and ontological truth, i.e. knowability, are convertible19 – which means that only that which in some sense is may be the object of truth-relating discourse: a principle very close to the PR. Moreover, at least once the PR is explicitly stated and defended in semantic terms: About that which absolutely is not, it is not possible to state anything.20

A scholastic accepting this principle may indeed concede, as a matter of course, that prima facie reference to non-existents is possible, but then go on to construe the nature of non-existents so that ultimately they are somehow explained away or reduced to actual entities. This approach is exemplified in Aquinas21 and Boëthius of Dacia,22 and may have been 18

Cf. the works of G. Klima cited in note 3. Cf. Thomas Aquinas, STh I, q. 1°6, a. 3, co: “[S]icut bonum habet rationem appetibilis, ita verum habet ordinem ad cognitionem. Unumquodque autem inquantum habet de esse, intantum est cognoscibile.” 20 “De eo quod non est penitus non contingit aliquid enuntiare.” – Peter of Spain or Peter of Auvergne, as related by Sten Ebbesen, “The Chimera’s Diary”, in The Logic of Being, ed. S. Knuuttila and Jaakko Hintikka (Springer Science & Business Media, 1986), 134. 21 Quite often, when dealing with non-existents, the scholastics, including Aquinas, only consider entities existing in the future or past (even Buridan in his Summulae treatise on supposition limits ampliation thus and never mentions genuine possibilia – cf. Jean Buridan, Jean Buridan’s Logic. The Treatise on Supposition. The Treatise on Consequences, trans. and ed. P. King (Dordrecht: Reidel, 1985), 125ff – although he does mention genuine possibilia in his (early, according to Ria van der Lecq’s introduction to the edition) commentary on De Interpretatione: cf. Klima, “Existence and Reference”, 213, or Jean Buridan, 160–161). Note also that the prevailing conception of time before Scotus was eternalist: viz. the Boëthian theory of time as eternally present to God in its entirety; so that from God’s point of view the past, present, and future are equally actual. Cf. Aquinas, STh I, q. 14, a. 13, where contingents qua future are said to be unknowable. And when Aquinas does speak of pure entia in potentia (i.e. non-existent even in relation to God), he 19

170

Can We Speak About That Which Is Not?

typical for the more Aristotelian strands of thought in the pre-Scotian period.23 conceives them qua “in potentia Dei vel creaturae.” On this view, meaningful reference to true possibilia seems to be enabled only by positing some actually existing media (viz. their causes) “in which” the non-existents are known (and referred to): “[S]cientia non est nisi entis, eo quod ens et verum convertuntur. Dupliciter autem dicitur aliquid ens, uno modo, simpliciter, quod scilicet est ens actu; alio modo, secundum quid, quod scilicet est ens in potentia. Et quia […] unumquodque cognoscitur secundum quod est actu, non autem secundum quod est in potentia, scientia primo et principaliter respicit ens actu. Secundario autem respicit ens in potentia, quod quidem non secundum seipsum cognoscibile est, sed secundum quod cognoscitur illud in cuius potentia existit.” – STh III, q. 10, a. 3 (emphasis mine). Aquinas does not admit another way of reaching possibilia, no epistemic access to possibilia in themselves qua possibilia. If they cannot be known in their causes, they cannot be known at all (cf. De veritate, q. 2, a. 8, co.). 22 Cf. Ebbesen, “The Chimera’s Diary”, 133–134, where Boëthius is documented to employ the strategy of possibilia being known “in their causes,” in a manner quite similar to that of Aquinas. 23 Klima claims (“Existence and Reference”, 212ff, or John Buridan, 158ff) the scholastics rejected the PR almost universally and only cites Scotus (in note 33) as a possible exception. As regards Scotus, this is a mistake, since, as I am going to show, Scotus explicitly rejected the Principle, thus setting himself apart from an already established Franciscan-Augustinian tradition, represented by Henry of Ghent. Klima is misled by T. J. Cronin, Objective Being in Descartes and in Suárez (Gregorian Biblical BookShop, 1966), App. II, esp. p. 187, where Cronin ascribes to Scotus the view that the not-yet-existent objects of divine knowledge possess more reality than nothing and are true things. But Cronin is clearly in error here, since the text he quotes from Opus Oxoniense I, d. 36, n. 1 in support of his interpretation does not state Scotus’s own opinion but is one of the “principal arguments” commonly listed in the beginning of a question in order to motivate the problem and later to be responded to (cf. Opus Oxoniense, ibid., n. 11, which is Ordinatio I, d. 36, q. un., n. 48–52 (ed. Vat. VI, 290–292)). In his own determinatio quaestionis Scotus clearly rejects this (Henry of Ghent’s) position (see below part 7, esp. notes 37 and 38). Although Scotus need not have been the first to explicitly reject the PR, he seems to have been the first to have the theoretical means to make this rejection clearly and unambiguously: namely, Scotus is the first to jointly have: (1) a clear notion of objective potency in terms of synchronic modality, which enables a clear notion of a genuine possibile considered in itself – i.e., (i) not just a thing existing in the past or future, and (ii) not something considered (and referred to) merely in its causes; (2) a clear notion of the act–object distinction (or the distinction between “esse in anima subiective” and “esse in anima obiective”), which precludes confusing the notions of reference to (non-actual) possibilia and reference to (actual) mental acts (something that

Lukáš Novák

171

Another, perhaps more principled solution to this difficulty is scholastic possibilism:, i.e., the option to ascribe some ontological status to possibilia as such.24 Various positions of this kind were defended especially within the Augustinian-inspired “Franciscan school”, in which one must also count the secular Parisian master Henry of Ghent (ca. 1217–1293), who, although himself not a Franciscan, adopted a number of typically Franciscan views and thereby influenced this tradition (notably Duns Scotus). It was Henry who formulated the paradigmatic version of scholastic possibilism, built upon the notion of esse essentiae or essential being,25 i.e., a kind of being belonging to an essence (or identical to an essence)26 independently of whether it actually exists (i.e., whether it is endowed with esse existentiae). It was this version of possibilism that the Subtle Doctor criticised; oftentimes, however, his pupils reverted to some version of Henry’s position and the doctrine of esse essentiae seems to have been relatively common in realistic circles until Scotism established itself as a relatively disciplined “school” (i.e., by the end of the fifteenth century). For the purposes of the present exposition I chose just one of

arguably happens in the Aristotelian-Thomistic formal identity theory of intentionality) – cf. D. Perler, Theorien der Intentionalität im Mittelalter (2. Aufl., Frankfurt a. M.: Vittorio Klostermann, 2004), esp. 185ff. The absence of these two conceptions in pre-Scotian scholastic thought is obscured by the fact that as they became a universally accepted part of the scholastic conceptual apparatus, they came to be retrospectively interpreted into the older authors (e.g. into Aquinas by Hervaeus Natalis and Cajetan). See also below, note 49. 24 As already mentioned, I am leaving aside the particularly tangled problem of the possibility of reference to impossible beings in scholasticism. It cannot be waived merely by saying that such reference is simply impossible for the scholastics, since many scholastics purport to make many true claims about impossibilia or “entia prohibita”. Frankly, on my understanding, Scotus does in effect reject the possibility of genuine reference to impossibilia and generally has an (implicit) reductive theory of enita rationis, though my view is not generally accepted at present (I defend it in my Scire Deum esse (Praha: Kalich, 2011), English translation forthcoming). 25 The doctrine may have (some of) its roots in the “modi essendi” of the modistae: cf. Ebbesen, “The Chimera’s Diary”, 131ff. 26 According to Henry, there is an intentional distinction between an essence and its esse essentiae (see Henricus Gandavensis, Summa quaestionum ordinarium, a. 21, q. 4, co., and other places cited in Scotus, Ordinatio I, d. 36, q. un., n. 4 (ed. Vat. VI, 273)); Francis of Meyronnes, however, seems to identify essential being with the essence altogether, since he argues that to deny essential being just is to deny essences (see below note 31).

172

Can We Speak About That Which Is Not?

Scotus’ immediate pupils who rejected his criticism of Henry, viz. the teaching of the influential thinker Francis of Meyronnes (or Mayronis), set forth in his treatise De esse essentiae et existentiae.27 In Henry the notion of esse essentiae is connected with the Avicennian conception of “essence considered absolutely” (natura absolute sumpta), i.e., essence considered under abstraction from whether it exists as individualized in reality or universalized in the intellect.28 Francis of 27 The French Franciscan Francis of Meyronnes/Mayronis (ca. 1288–1328) was an immediate but not faithful pupil of John Duns Scotus. In 1320–1321 he was reading the Sentences of Peter Lombard in Paris; notable is his disputation over the Blessed Trinity with Pierre Roger (the future pope Clement VI). In 1323 Francis became Master in Theology and was afterwards elected the Minister General of his order. Later he was present at the papal court; he supported Clement’s predecessor John XXII in his struggle with Emperor Louis of Bavaria. In his times he was counted among the most outstanding theologians and philosophers, earning the titles “Doctor illuminatus” and “Doctor abstractionum”. According to a fifteenthcentury record Scotus’s pupils formed three factions after their master’s death: the Mayronists (adherents of Francis of Meyronnes), the Bonetists (adherents of Nicolaus Bonetus, another pupil of Scotus) and “pure Scotists”. The fame of Francis only came to be overshadowed by that of Scotus towards the end of the fifteenth century, as the Scotistic school in the strict sense was established, but he still remained a frequently read and appreciated author for some decades. Francis’ Platonizing realism, carefully formulated to remain within the confines of Catholic orthodoxy and set out with admirable clarity, seems to represent a welcome “politically correct” alternative to Wycliffism for the die-hard adherents of via antiqua in the rise of late mediæval nominalism. Francis’ vindication of possibilism is found in his treatise De esse essentiae et existentiae, also called Vinculum, which was later adapted as Question 8 of his Quodlibet. For a modern, semi-critical edition of the treatise see Franciscus de Mayronis, “Tractatus de esse essentiae et existentiae”, ed. M. Lánský, R. Mašek, L. Novák, and S. Sousedík, Studia Neoaristotelica 2, no. 2 (2005): 277–322 [henceforth Vinculum, page and lines of the mentioned edition are given in parentheses]; see also the accompanying essay in Czech, S. Sousedík, “Franciscus de Mayronis a jeho traktát De esse essentiae et existentiae”, ibid., 271–276. For an excellent treatment of Scotus’s and Francis of Meyronnes’ theory of possibilia see Tobias Hoffmann, Creatura intellecta: die Ideen und Possibilien bei Dun Scotus mit Ausblick auf Franz von Mayronis, Poncius und Mastrius (Münster: Aschendorff, 2002). 28 For a precise survey in Czech of Avicenna’s conception of the “essence absolutely considered” and the reception of this doctrine in Henry of Ghent (and Aquinas) see Václav NČmec, “Esence ‚uvažovaná absolutnČ‘ u Avicenny, Tomáše Akvinského a JindĜicha z Gentu”, in Univerzálie ve scholastice, ed. D. Heider a D. Svoboda, Studia Neoaristotelica, Supplementum I (2012): 72–96. Pasquale

Lukáš Novák

173

Meyronnes, however, takes esse essentiae unambiguously as the being of a real essence, i.e., the essence of a singular thing, which, as a matter of course, contains individuality.29 The essential being of an entity is therefore a certain degree of reality which belongs to that entity in virtue of its essential predicates. The essential being of Socrates therefore comprises such things as “being a man”, “being a substance” etc.30 The notion of essential being does not, of itself, entail possibilism. Even Duns Scotus makes use of this term and ascribes essential being (in addition to existential being) to actually existing beings. What distinguishes scholastic actualists from scholastic possibilists is the answer to the question whether beings have their essential being even if they do not exist (i.e., scholastically speaking, “before creation”) and whether in this ontological state they are sufficiently distinct both from God and from nothingness. Francis gives a positive answer to all these questions, an important motivation of his replies being precisely a certain version of the PR, or more precisely, a more general principle of which the PR is a special case.

Porro, “Henry of Ghent”, in The Stanford Encyclopedia of Philosophy (Fall 2008 Edition), ed. Edward N. Zalta, URL = ‫ۃ‬http://plato.stanford.edu/archives/fall2008 /entries/henry-ghent/‫ۄ‬, §6, downplays Henry’s possibilism claiming that “despite the ultra-essentialist interpretation of Henry’s metaphysics that began with Suárez, the syntagma esse essentiae does not designate a separate being, but only the fact that a res thus constituted has an objective content and so is objectively possible; that is, it can be placed in act by God.” But this would mean that Henry’s position on the ontological status of possibilia is almost indistinguishable from that of Scotus – who, however, clearly understood Henry as a true possibilist and criticised him as such. So regardless of Henry’s true intentions, the possibilist or “ultra-essentialist” interpretation of his theory was there right from the start and made a great impact on the via antiqua circles in the fourteenth and fifteenth centuries. 29 Francis makes this general claim in article 5 of the Vinculum (292:11–15), but further on he is driven to what looks like a partial retraction: see ibid, a. 14, diff. 2 (309:20–25); cf. also ibid., a. 15. 30 Vinculum, a. 1, co. (283:11–13): “[D]icitur, quod illud est esse essentiae, quod pertinet ad cuiuslibet entis quiditatem, quia in rebus habentibus essentiam quiditas et essentia idem sunt. Sicut cum dicimus hominem esse animal, illud pertinet ad esse essentiae, quia animal est de hominis quiditate. Et in hoc differt illud esse ab esse existentiae, quia illud non pertinet communiter ad quiditatem. Sicut cum dicitur ‚homo existit in rerum natura‘, quia illud est contingens.”

174

Can We Speak About That Which Is Not?

That the principle is operative can be detected already in the first article of his treatise, where he tries to vindicate the very notion of essential being against those who reject it altogether. According to Francis, to reject essential being is to reject all definitions and consequently all demonstrations, which would result in absolute scepticism. So even though Francis does not explicitly speak of being as a precondition of reference, he regards it as a sine qua non of defining and argumentation: But a difficulty occurs, for some claim that there is no such essential being as we posit. These however reject all definitions and consequently all demonstrations propter quid – for this essential being is nothing but quidditative being, and who rejects quiddities must reject everything.31

Francis of Meyronnes develops this approach further in the second article, where he demonstrates the independence of essential being from the contingent act of creation: the truths expressing essential being (for example that every man is an animal) are necessary truths, but no necessary truth can depend on something contingent, therefore no necessary truth can depend on the contingent act of creation. Necessary immutable truths, however, require a necessary and immutable object, in relation to which they are adequate; so essential being, the object of necessary truths, must also be necessary and immutable, indifferent to creation.32 “Production does not increase number but perfection.”33 31

“Sed oritur difficultas, quia dicunt aliqui, quod non est dare aliquod esse essentiae tale, sicut infertur. Isti autem negant omnem diffinitionem et per consequens omnem demonstrationem propter quid, cum istud esse essentiae non sit nisi esse quiditativum, et qui negant quiditates, habent omnia negare.” – Vinculum, a. 1, diff. (285:7–10). 32 Francis actually presents four arguments for the independence of essential being: (i) nothing necessary can depend on a contingent act; (ii) analytic (certae et per se notae) truths cannot depend on a contingent act; (iii) essences ‘before’ and ‘after’ creation have the same genus (because generic determination is necessary for a given essence) and being in a genus is (an aspect of) essential being; (iv) essences before and after creation have the same definition (since actual existence, which makes the difference, is not part of the definition of contingent beings), and therefore also the same quiddity and the same essential being: “Primo, quia nullum necessarium potest accipere esse per contingens, cum omnino dependens a contingente necessario sit contingens. Sed creatio est actio contingens, ut docet fides catholica, et esse essentiae est necessarium, cum omnis propositio de esse tali sit in naturali materia, quae distinguatur contra contingentem. Igitur ante creationem habent quiditates tale esse. Secundo, quia nullum certum et per se

Lukáš Novák

175

Above we saw how Russell denied our capacity to make statements about non-existing individuals, in order to avoid the necessity of denying that such statements have a truth-value, and how Strawson conceded the possibility of meaningful attempts at such statements, but on the same principle denied them having a truth-value. The logic of Francis of Meyronnes’ reasoning is similar: but since he regards as undisputable that non-existing beings can be the subjects not only of meaningful, but also of true, or even necessarily true statements, he has to admit their having some kind of being in consequence. For the third time and most obviously the principle is employed when Francis demonstrates, in the sixth article, that the essences in pure essential being are not pure nothing. He presents four structurally identical arguments, of which each derives the non-nothingness of the essences in pure essential being from the fact that they are objects of a certain relation: they are “something creatable” (and “something” is not nothing), intelligible notum in notitia abstractiva potest esse per actum contingentem. Sed veritates de illo esse sunt certae et per se notae, quia in primo vel in secundo modo dicendi per se. Ergo non sunt per creationem. Et sic antecedens maior probatur, quia species repraesentans quiditatem obiecti non repraesentat evidenter, nisi illa, quae insunt obiecto necessarie. Alioquin, si contingenter, deciperet, cum variatione facta in obiecto, quoad talia species non varietur, sed uniformiter repraesentet. Tertio, quia quiditates creabiles ante suam creationem fuerunt in potentia obiectiva. Sed ens in potentia obiectiva et in actu sunt in eodem genere. Et illud esse essentiae pertinet ad esse in genere. Ergo in illa potentia habent esse essentiae. Quod autem sunt in eodem genere, probatur: quia esse in genere non est contingens, ut patuit. Quarto, quia illae quiditates creabiles in potentia tali et in actu habent eandem penitus diffinitionem. Quia sicut angelus in actu est substantia spiritualis, ita etiam in potentia. Et tamen, quod in actu esse non sit de quiditate angeli, patet, quia non est necesse esse per fidem, et per consequens nec esse in potentia. Et sic remanet diffinitio penitus una. Ideo dico, quod quiditates creabiles ante suam creationem habent esse essentiae, quidquid sit illud esse. Quia ante suam creationem sunt entia in potentia, et actus et potentia non diversificant essentiam. Sic confitentur philosophi et constat, quod essentia et esse essentiae sunt idem.” – Vinculum, a. 2 (286–287). The principle defended here by Francis is practically the same as the “Principle of Independence” (of so-being on being), Prinzip der Unabhängigkeit des Soseins vom Sein, the explicit formulation of which is ascribed to Meinong’s pupil Ernst Mally (1879–1944), which says that objects can have properties regardless of whether they exist (see Marek, “Alexius Meinong”, ch. 4.1). Nihil novum sub sole – as usual. 33 “Productio non largitur multitudinem, sed perfectionem – scilicet modi” – Vinculum, a. 5, ad 4 (294:3–4).

176

Can We Speak About That Which Is Not?

and known (and who knows, knows something not nothing), they can be willed by God (likewise, who can will, can will something), and they are images of divine ideas-paradigms (and again, a paradigm is a paradigm of something).34 In a summary of these arguments, Francis reduces all of them to the following general principle: any object [of a relation] must have being in that way in which it is an object.35 This principle is nothing but the PR generalized to cover not just reference but any relation whatsoever. Thus, according to Francis, one not only cannot speak about that which in no way is, but one also cannot relate to it in any way whatsoever. This intuition, whether correct or not, seems to be the ultimate source of Francis’ possibilism.36

34

“Primo, quia Deus ab aeterno habuit potentiam creativam. Omnis autem potentia productiva est productiva alicuius producibilis. Et sic fuit ab aeterno aliquid producibile sibi correspondens, et quod est aliquid, non est omnino nihil. Secundo, quia omnis operatio intellectiva habet aliquod obiectum ad quod terminatur utpote obiectum, quod per ipsam intelligitur. Divina autem intellectio ab aeterno intelligebat omnia creabilia. Ergo ipsa erant aliquid, cum essent a Deo intellecta. Tertio, quia actus voluntatis semper habet obiectum. Divina autem volitio ferebatur ante creationem super ista creabilia. Et sic non erant omnino nihil, cum haberent rationem obiecti. Quarto, quia in Deo ante omnem creationem fuerunt ideae omnium creabilium secundum beatum Augustinum 83. q., et illa sunt exemplaria. Sed omne exemplar inquantum tale respicit exemplatum. Ergo quiditates istorum creabilium fuerunt tunc exemplatae, et non in esse existentiae. Et sic consequenter in esse essentiae.” – Vinculum, a. 6 (294–295). For Scotus’ summary of similar arguments by Henry and others see Ordinatio I, d. 36, q. un., n. 1–2 and 3–12 (ed. Vat. VI, 273–275). 35 “Ideo dico, quod licet quiditates illorum creabilium ante suam creationem fuerint pure nihil secundum carentiam esse existentiae, non tamen omnino nihil secundum carentiam ipsius esse essentiae. Quia essentiae creabilium ante omnem creationem sunt obiecta secundaria divini intellectus et omne obiectum praesupponit esse eo modo, quo habet esse obiectum: sicut pars in modo praesupponit suum totum.” – Vinculum, a. 6, co. (294:8–13); italics mine. 36 And, to a great extent, also of his attempt (quite unique among the scholastics) at repudiating the entia rationis altogether. Cf. questions 6 and 7 of his Quodlibet, reproduced in Francisci de Mayronis OFM Quodlibeti quaestiones VI. et VII. (De entibus rationis), ed. N. Cuhrová et L. Novák, Studia Neoaristotelica 3, n. 2 (2006): 198–239.

Lukáš Novák

177

7. John Duns Scotus Let us move now from Francis to the view of his teacher Scotus, so difficult to accept or even understand for his pupils. Scotus’ position can be simply characterized as a consistent rejection of the PR (which implies the rejection of the generalized principle employed by Francis of Meyronnes). According to Scotus, the objects of any intentional relations (whether it be the relation of a knower to the known, of a willing one to the willed, of a creator to the created, or a paradigm to its image) simply are not required to have any ontological status whatsoever, or, as Scotus puts it, any esse verum. The “being” expressed by the predicates exploited by Francis, like “to be known” (esse cognitum), “to be intelligible” (esse intelligibile), “to be an image of a paradigm” (esse exemplatum), “to be represented” (esse repraesentatum) and the like, is not real or true in any way, irrespective of whether the relation involved concerns God or man: For even if something is not, it can be known (both according to its essence and according to its existence) – and still we don’t say that it should have true essential or existential being. And in this regard there seems to be no difference between divine and human intellect, except that the divine intellect produces the intelligibles in intelligible being, whereas ours does not, primarily.37

It is not necessary to assume any esse essentiae in objects of knowledge: instead, Scotus speaks of “esse deminutum” here, but he points out emphatically that this “diminished being” is being only “secundum quid”, i.e., in an improper, qualified sense – this is the point of Scotus’ famous criticism of Henry of Ghent laid out in the unique question of dist. 36 of the first book of his Ordinatio. If you look for some real being in the object of intellection that it should have precisely in virtue of being such an object, there is none to be found. The only real being to be found here

37 “Quia si aliquid non sit, potest a nobis intelligi (et hoc sive essentia eius sive existentia eius), et tamen non propter intellectionem nostram ponitur quod illud habeat verum esse essentiae vel existentiae; nec est differentia aliqua – ut videtur – inter intellectum divinum et nostrum, quoad hoc, nisi quod intellectus divinus producit illa intelligibilia in esse intelligibili, noster non producit primo.” – Duns Scotus, Ord. I, d. 36, q. un., n. 28 (ed. Vat. VI, 281–282).

178

Can We Speak About That Which Is Not?

is the real being of the intellection, to which the esse deminutum of the intellected object is reduced: To the second objection I respond that this production [viz. of possibilia into intelligible being] is a production into being of a kind distinct from any sort of “being without qualification”. And it is not a production of a mere relation, but of its fundament as well – not, to be sure, according to essential or existential being (which is true being), but according to diminished being, which is “being in a qualified sense”. Still, it is the being of an absolute entity as well, which absolute entity is nevertheless accompanied by a relation of reason according to this diminished being […]38 And if you are looking for some “true being” of this object as such [viz. of the object qua conceived], there is none to be found over and above that “being in a qualified sense”, except that this “being in a qualified sense” can be reduced to some “being in an unqualified sense”, which is the being of the respective intellection. But this being in an unqualified sense does not belong to that which is said to “be in a qualified sense” formally, but only terminatively or principiatively – which means that to this “true being” that “being in a qualified sense” is reduced, so that without the true being of this [intellection] there would be no “being in a qualified sense” of that [object qua conceived].39 38 “Ad secundum dico quod productio ista est in esse alterius rationis ab omni esse simpliciter, – et non est relationis tantum, sed et fundamenti; non quidem secundum esse essentiae vel existentiae (quod est esse verum), sed secundum esse deminutum (quod est esse verum), quod esse est esse secundum quid etiam entis absoluti, quod tamen ‘ens absolutum’ secundum istud esse deminutum concomitatur relatio rationis.” – Ord. I, d. 36, q. un., n. 44 (ed. Vat. VI, 288). In the critical edition the part in parentheses “(quod est esse verum)” is placed after “sed secundum esse deminutum”; but Ȉ reads as given here. It may be that the misplacement of the phrase goes back to Scotus, but given the entire context it is out of the question that esse deminutum were consciously called “esse verum” by Scotus. The point of the entire passage is to distinguish esse deminutum which is esse secundum quid from esse verum (essentiae et existentiae) that can only be ascribed to that intellectio on which the esse secundum quid of the conceived object depends. The editors’ attempt to maintain the combination of esse verum with esse secundum quid motivates another evident misinterpretation of the text in n. 46 (see further the following quote and note 39). 39 Ord. I, d. 36, q. un., n. 46 (ed. Vat. VI, 289): “Et si velis quaerere aliquod esse verum huius obiecti ut sic, nullum est quaerere nisi ‘secundum quid’, nisi quod istud ‘esse secundum quid’ reducitur ad aliquod esse simpliciter, quod est esse ipsius intellectionis; sed istud ‘esse simpliciter’ non est formaliter esse eius quod

Lukáš Novák

179

In other words: if we were to make something like an inventory of reality, we should not list any objects having mere esse deminutum. By speaking about objects in intelligible being we do not take on any ontological commitment (to use the Quinean language) over and above the commitment to the existence of the intellections directed to these objects. As was already mentioned, according to Scotus this holds not just for human cognitive acts, which compare to their objects as “something measured to the measure”, and therefore are related to them by a real relation, but also for cognitive acts of God, where it is the other way around. Divine acts are absolute and identical to God’s essence, for which reason they are the measure of their objects and it is therefore the objects that are related to them.40 Nevertheless, even such absolute cognitive acts “produce” in their objects esse intellectum, since despite the absolute and object-independent nature of divine knowledge it still holds, objectively, that the object is known by God. All this having been said, the independence of divine intellection of its object (and the inverse dependence of the object on the intellection) still does imply a certain fundamental difference between divine and human cognition, which is most relevant for understanding the deeper background dicitur ‘esse secundum quid’, sed est eius terminative vel principiative, ita quod ad istud ‘verum esse secundum quid’ ‘verum esse’ [esse] ‘secundum quid’ reducitur sic quod sine isto vero esse istius non esset illud ‘esse secundum quid’ illius.” The editors enclose the phrase “verum esse secundum quid” into a single set of quotation marks, but this parsing is clearly untenable, as I show in “Divine Ideas, Instants of Nature, and the Spectre of ‘verum esse secundum quid’”, Studia Neoaristotelica 9, n. 2 (2012): 194–198. The term “esse” supplied in brackets may have been omitted by mistake of the scribe (given that at first glance it looks like erroneous repetition), but the sentence would make sense even without this editorial insertion (representing a typical example of Scotus’ terse Latin). 40 Cf. Ord. I, dist. 35, q. un., n. 27 (ed. Vat. VI, 256): “[R]elationes tertii modi differunt per se ab aliis aliorum duorum modorum, quia in tertio modo non est mutuitas sicut in aliis duobus modis, – et ex hoc sequitur […] quod terminus relationis est absolutum quid, ut absolutum. Sicut ergo obiectum intellectionis nostrae terminat relationem ipsius in quantum mere absolutum est, et sic ipsa est mensurata per ipsum, ita – videtur – cum intellectio divina simpliciter sit mensura omnium aliorum a se intellectorum, sequitur quod alia praecise referantur ad intellectionem divinam, et ipsa terminabit relationem istam sub ratione mere absoluti; […] Non opportet ergo propter intellectionem alicuius obiecti praecise, quaerere relationem.”

180

Can We Speak About That Which Is Not?

of Scotus’ repudiation of the PR. Whereas in the case of human cognition, a precondition of the actual “being known” of the object is its prior intelligibility, divine cognition cannot thus depend on prior intelligibility of its object. Divine intellection, therefore, confers on its object not only its knownness, but also its knowability: not just esse intellectum, but also esse intelligibile – as Scotus observed in the first quotation given above. And it is not so, as one might think,41 that God first (in the sense of atemporal priority of nature, or order of metaphysical dependence) made an object intelligible and then actually cognized it. Let me stress this point once more: divine cognition does not presuppose any prior intelligibility of its object, and for that reason it needs not first bestow it, in order to be able to actually grasp the object. Rather, the objects become intelligible only by the very fact of being actually cognized by God. In the famous passage where Scotus provides the analysis of the individual logical moments or “instants of nature” of God’s knowledge of the possibilia, God’s “producing the stone in intelligible being” and God’s “conceiving the stone” fall within one and the same instant of nature: In the first instant God conceives his own essence as something purely absolute. In the second instant he produces a stone in intelligible being and conceives it, so that there is a relation in the stone to the divine intellection, but still none in the divine intellection to the stone – but the divine intellection terminates the relation of the stone qua conceived to itself.42

That means, in Scotus’ conceptual framework, that there is not even the least distinction or ordering of any sort between these two quasi-processes – they are perfectly identical.43

41

Michael Renemann did, actually think something like this – see his Gedanken als Wirkursachen: Francisco Suárez zur geistigen Hervorbringung (Amsterdam– Philadelphia: B. R. Grüner, 2010); and “The Mind’s Focus as an Efficient Cause: Francisco Suárez’s Re-interpretation of the Traditional Understanding of the Idea”, American Catholic Philosophical Quarterly 84, No. 4 (2010): 704–707; and my criticism in Novák, “Divine Ideas”, 192ff. 42 “Deus in primo instanti intelligit essentiam suam sub ratione mere absoluta; in secundo instanti producit lapidem in esse intelligibili et intelligit lapidem, ita quod ibi est relatio in lapide intellecto ad intellectionem divinam, sed nulla adhuc in intellectione divina ad lapidem, sed intellectio divina terminat relationem ‘lapidis ut intellecti’ ad ipsam; […]” – Duns Scotus, Ord. I, dist. 35, q. un., n. 32 (ed. Vat. VI, 258). 43 For detail see Novák, “Divine ideas”, 192–193.

Lukáš Novák

181

And now the crucial point: it is precisely this intelligibility, imparted to the objects by the divine intellect, that makes human conceiving of the same objects possible, irrespective of whether they have any real being or not: This twofold causality of the divine intellect – viz. that [i] it is the true uncreated light which produces secondary objects [of the divine intellect] into intelligible being, and that [ii] it is that in virtue of which [thus] produced secondary objects actually move the intellect – can, as it were, be combined in a third sense in which we say that we “truly see [whatever we know] in eternal light”.44

In other words: the most fundamental reason why the PR is false is, according to Scotus, the fact that a sufficient condition of the human capacity to refer to something is the intelligibility of that something. This intelligibility, however, is bestowed on things in virtue of their being conceived, prior to creation, by the absolute divine intellect. This divine conceiving, however, neither produces nor presupposes any genuine being in the objects; for it is a universal truth that cognition is an immanent operation, one whose effect remains wholly in its subject (and so does not really affect its object) – in this elementary point divine cognition is not different from ours. Accordingly, objects need not have any being whatsoever in order to be capable of being referred to. This doctrine of Scotus not only explains how reference is independent of any genuine being of the object, but also opens the possibility of “eternal truths” without the necessity of positing any “eternal truth-makers” – be they divine ideas or essences in essential being. In his polemic against Henry’s elaboration of the Augustinian illumination theory, according to which eternal ideas are required to guarantee the necessity and immutability of eternal truths, Scotus re-interprets the notion of “idea”, identifying ideas with objects of the divine intellect qua in intelligible being – i.e., with the “not yet existing” creatable things. Thus he achieves several aims: First, in contrast to the Platonizing Augustinian-Franciscan line of thought that looked for the object of true knowledge outside the created individual things (viz. in God and his ideas), Scotus is able to 44

“Ista igitur duplex causalitas intellectus divini – quod est vera lux increata, videlicet quae producit obiecta secundaria in ‘esse intelligibili’, et quod est illud virtute cuius secundaria etiam obiecta producta movent actualiter intellectum – potest quasi integrare tertium membrum, propter quod dicamur vere videre in luce aeterna.” – Duns Scotus, Ord. I, d. 3, pars 1, q. 4, n. 267 (ed. Vat. III, 163).

182

Can We Speak About That Which Is Not?

maintain the robust Aristotelian realism, according to which it is possible to have universal, necessary, scientific knowledge of the world of material particulars which are subject to generation and corruption – since the realm of intelligible objects is the realm of ordinary particulars, of which some come to be created and so actually exist, while others remain forever merely possible. At the same time, however, Scotus need not sacrifice a single iota of the traditional Augustinian locutions, for insomuch as these particulars, which are objects of our human knowledge, are ab aeterno conceived by God, we are justified to call them “ideas”, “light” etc.;45 and the traditional term “uncreated light” can be applied to God’s intellect as the ultimate source of all intelligibility. For, as Scotus points out again, intelligible being has no reality in the intelligible objects, the entire “reality” of the intelligibility of things consists in the reality of God’s act of conceiving them: Insofar as these [things beside God] are the secondary object of the divine intellect, they do not have being but in a qualified sense. However, to that which is merely in a qualified sense a genuine real operation does not pertain in its own right, but it pertains to it, if at all, in virtue of something that has being in an unqualified sense. Therefore, these secondary objects only move [our] intellect, precisely speaking, in virtue of the being of the divine intellect, which is the being in an unqualified sense thanks to which these objects have their being in a qualified sense. Thus we “see in the light eternal (in a qualified sense)”46 [viz. in the essences in intelligible being] as in the proximate object, but “in eternal uncreated light” [viz. in the divine intellect] we see as in the proximate cause, in virtue of which the proximate object moves [our intellect].47 45

See Duns Scotus, Ord. I, d. 3, pars 1, q. 4, n. 261nn (ed. Vat. III, 160nn). The reason why the light is “eternal” merely in a qualified sense is that genuine eternity involves real being, whereas the objects which are this light just have, qua such, intelligible being. 47 “[I]lla ut sunt obiectum secundarium intellectus divini non habent ‘esse’ nisi secundum quid; operatio autem aliqua vera, realis, non competit alicui praecise ‘enti secundum quid’ virtute sui, sed si aliquo modo competit sibi, hoc oportet esse virtute alicuius cui competit ‘esse’ simpliciter; igitur istis obiectis secundariis non competit movere intellectum praecise nisi virtute ‘esse’ intellectus divini, quod est ‘esse’ simpliciter et per quod ista habent ‘esse’ secundum quid. Sic ergo in ‘luce aeterna secundum quid’ sicut in obiecto proximo videmus, sed in ‘luce aeterna increata’ videmus secundum tertium modum, sicut in causa proxima, cuius virtute obiectum proximum movet.” – Duns Scotus, Ord. I, d. 3, pars 1, q. 4, n. 265 (ed. Vat. III, 162). Compare this with Aquinas’s view sketched above: according to Scotus (cf. Ord. I, d. 43, q. un., n. 14 (ed. Vat. VI, 358)) possibilia are formally 46

Lukáš Novák

183

Third, Scotus is able in this way to blunt the edge of the Platonizing arguments that request eternal truth-makers for eternal truths, as really existing divine ideas or essences in essential being are replaced by objects existing merely intentionally – primarily and perfectly in God’s intellect, secondarily also in ours: How can a proposition [e.g. “A man is an animal”] be claimed to be necessary, given that identity of the subject and predicate can be destroyed [e.g. if men died out]? I respond: when a thing is not, its identity, for sure, is not real; but if in such a case it is in an intellect, the identity exists as an object conceived by the intellect, and it is necessary in a qualified sense. For although in such being the subject and predicate cannot be without such identity, the proposition itself can possibly not be, as either of its terms can possibly not be conceived by the intellect. In our intellect, therefore, the proposition is necessary in a qualified sense [only] – viz. insomuch as it is incapable of becoming false [although it is not incapable of ceasing to exist]. On the other hand, it is necessary without qualification solely in the divine intellect, as in no other being than in this “being conceived” [by God] do subject and predicate have an identity that is necessary in an unqualified sense.48

In order to save necessary truths it is therefore not necessary that the sciences be concerned with anything else than empirical particulars (e.g. with Platonic Forms or divine ideas), nor is it necessary to posit any intelligible in themselves (although they receive this intrinsic intelligibility “principiatively” from God’s intellect) – without having any real being whatsoever. Thus Scotus completely divorces intelligibility from actual entity; whereas Aquinas adheres to the principle that in order for something to be intelligible of itself, it must, to that extent, have being (cf. note 19). The intelligibility of non-entities, if admitted at all, is always “borrowed” in Aquinas, so to speak, and so although reference to them is possible to outward appearances, it is ultimately explained away in the theory. 48 “Contra: quomodo propositio ‘necessaria’ affirmatur, si identitas extremorum potest destrui? Respondeo: quando res non est, non est identitas eius realis, – sed tunc si est in intellectu, est identitas ut est obiectum intellectum, et necessaria secundum quid, quia ut in tali ‘esse’ extrema non possunt esse sine tali identitate; tamen illa potest non esse, sicut extremum potest esse non intellectum. Ergo ‘propositio necessaria’ in intellectu nostro secundum quid, quia immutabilis in falsam; sed ‘simpliciter necessaria’ non nisi in intellectu divino, sicut nec extrema habent identitatem simpliciter necessario in aliquo ‘esse’ nisi in illo ‘esse’ intellecto.” – Duns Scotus, Ord. I, d. 3, pars 1, q. 4, n. 248 (ed. Vat. III, 151–152). The inverted commas inserted by the Vatican editors are better ignored.

Can We Speak About That Which Is Not?

184

“eternal” aspect within these particulars (esse essentiae). In order that a thing be a possible object of science (and so function as the truth-maker of that science’s claims) it is not required to have real being. Intelligibility is sufficient, and that is provided by God, necessarily and since eternity.49

8. Two Actualisms and Two Possibilisms Quine’s criterion of ontological commitment (which, as we have seen, is nothing but the PR de-metaphysicized), has established itself in the analytical tradition as a standard conceptual instrument. Accepting this criterion, however, has some non-trivial implications, which affect (among other things) the common understanding of the nature of the actualismpossibilism dispute. If the ontological commitment of a theory is defined as the domain of bound variables occurring in that theory, then it follows that the metaphysical question (i)

do non-existent things have any ontological status?

coincides with the semantic question (ii)

49

can non-existent individuals become values of the bound variables of our theory?

From what has been said it follows that there are a number of actual properties that can be truly ascribed to non-existents – intelligibility in the first place. This further illuminates the originality of Scotus’s position. As Klima notes (“Existence and reference”, 210), the standard ampliation/natural supposition approach “has the philosophical advantage that it is not going to leave us wondering about the ‘weird features’ of ‘non-actualized entities’ causing so much headache to philosophers exploring, gardening or uprooting (depending on their temper) the Meinongian Jungle. For on this approach we can refer to nonexistents only in contexts in which no actual properties can be attributed to them. So when we are talking about them, we are not going to get the false impression of exploring a different realm of entities […]” According to Scotus, however, we need not be so small-minded. We are allowed to explore the “jungle” as we please without ontological scruples, because it simply need not be there at all in order to be explorable. In light of this comparison, it seems that even the ampliation/natural supposition approach is at least partially driven by the PR, the desire to avoid speaking of utter non-entities as far as possible. On the explanative power of the ampliation approach see further below note 53.

Lukáš Novák

185

Thus theories that allow their variables to take non-existent individuals as their values are automatically understood as possibilist, to the effect that those who share Quine’s dislike towards the overpopulated Meinongian slum feel under pressure to construe their theories so that they enable reference to actual entities only. That results in various technical problems (the Barcan Formula50 and the like) requiring sophisticated workarounds, which however tend to introduce various ersatz-entities into the actualist systems like individual essences (Plantinga) or bare individuals “in limbo” (Transparent Intensional Logic), in effect barely distinguishable from the abhorred possibilia. It seems to me that Scotus’ position allows us to perceive the gratuitous nature of this common assumption. For in order that we are able to describe Scotus’ position, we need to distinguish between two actualisms and two possibilisms, depending on which of the questions stated above they concern. On the one hand, there is metaphysical possibilism, a theory ascribing some kind of real being to merely possibles, and metaphysical actualism, a theory that denies any kind of real being to non-existents. On the other hand, there is semantic possibilism, a theory claiming that nonexistent things can be referred to and become values of bound variables, and semantic actualism, which rejects this. Scotus shows us that metaphysical actualism can be maintained together with semantic possibilism – an alternative unheard of among the analytics, as far as I know. It might be objected here that Scotus’ position is extremely implausible, since it makes general semantic principles (what can and what cannot be referred to) dependent on certain very special metaphysical claims concerning divine existence and intellectual operation.51 It may seem that God functions in Scotus’ theory as a kind of deus ex machina, as it is only the assumption of God’s existence and intellectual activity that allows Scotus to reject the PR and so reconcile semantic possibilism and metaphysical actualism.

50

If it is possible that there is an F, then (actually) there is something that is possibly an F: ¸x(Fx) ĺ x¸(Fx). 51 I owe this objection to Václav NČmec (personal communication).

186

Can We Speak About That Which Is Not?

However, in my opinion this objection fails to properly understand the logic of Scotus’ position. For one thing, Scotus’ God is not a deus ex machina, only thrust in to save a fancy philosophical position from a collapse. Scotus has an extremely elaborate and complex argument for God’s existence, one of the most robust in the entire history of philosophy. He argues that God’s existence follows from a mere analysis of the notion of “essential order”, which is one of the transcendental attributes of being – so that the necessity of God’s existence is woven into the very fabric of reality. So why should he feel obliged to bracket it out from his philosophizing? But even apart from this consideration it is not true that the rejection of the PR is somehow epistemically “saved” only by means of the assumption of God’s existence. Scotus’ rejection of that principle does not rely on God’s existence as on a premise that makes it epistemically justified. The epistemic justification of the rejection of the principle consists in the simple fact that we all know that we can refer to non-existing things (see: “Frodo Baggins does not exist”), and we all know that they are not there (Tolkien made him up).52 This is a pre-philosophical datum, which can be either respected and analysed, or attempted to be explained away; but since the latter option amounts to a gross revisionism of our pre52

It is sometimes objected that fictive entities like Frodo Baggins cannot be identified with possible individuals because fictions are necessarily incomplete or under-determined (for example, nothing in Tolkien’s fiction determines the number of hairs on Frodo’s left foot) – and so our apparent capability of referring to fictions cannot be interpreted as capability of referring to possibilia. However, I don’t think that fictions are incomplete or underdetermined: in Tolkien’s fiction, Frodo certainly was intended to be a possible, fully determinate individual. There are (rare) examples of truly incomplete fictional entities in literature (perhaps, e.g., the grin of the Cheshire Cat would qualify) – but Frodo Baggins is not one of them. Of course, Tolkien did not care about the number of his hairs, but he certainly conceived him as having some determinate – if unspecified – number of them. It is true that the cluster of descriptions used to conceive Frodo is satisfied by more than one possible individual – but it certainly is not satisfied by any incomplete entity. What we want to do when speaking about Frodo is to speak about a certain possible individual, of which the important thing is that he satisfies certain descriptions; and we just do not want nor need to specify precisely which of the matching possibilia it is (except that it always is the same arbitrary individual). It seems that there is hardly any other way how to speak about possible individuals than in this underdetermined way. But this underdetermination of our reference does not imply, or so I contend, the underdetermination of the intended referent.

Lukáš Novák

187

philosophical conceptual scheme, I daresay that the burden of proof rests with the revisionist party, not with Scotus. God’s existence, to be sure, provides an explanation of this prephilosophical datum, just as it provides an explanation of the prephilosophical datum of the existence of change.53 If the proof of God’s existence from change is valid, then it is true that change would not be possible if God did not exist. But from that it does not follow that we are not epistemically justified to believe in the existence of change, unless and until we are able to prove God’s existence. Analogically, if Scotus is right, God’s existence and intellectual activity is indeed a sine qua non of our capability to refer to non-existing individuals (and, by the way, also a sine qua non of our capability to refer to existing individuals). But this does not mean that we can come to know the possibility of referring to non-existing individuals solely as an implication of God’s existence.

53

There is no contention that the possibility of reference to non-existents needs explanation and analysis. Merely saying “we all know that we can refer to nonexistents” does not suffice. Moreover, though the doctrine of ampliation may analyse the phenomenon, it does not, as such, explain it. Klima (“Existence and reference”, 212, or John Buridan, 159) perceives the urgency of the question of how ampliation to non-entities is possible, how we can say we can refer to entities which are literally nothing, but the explanation he offers is circular: viz. that “to refer” (“supponere”) is itself an ampliation-inducing verb. Clearly, when the possibility of ampliation as such is the explanandnum, to use the doctrine of ampliation as the explanans just begs the question: either you buy the entire ampliation business as something more than a surface phenomenon of natural language – or you just don’t.

188

Can We Speak About That Which Is Not?

To conclude: The PR led Meinong, Henry of Ghent and Francis of Meyronnes to derive metaphysical possibilism from semantic possibilism; and it led the Frege-Russellian tradition to derive sematic actualism from metaphysical actualism. In contrast, John Duns Scotus suggests that it is possible and perfectly natural to maintain semantic possibilism and yet to remain a metaphysical actualist; that we can speak about that which is not, without turning our universe into that unwholesome slum ridiculed by Quine. The finding that from an accurate analysis of such a fact God’s existence would ultimately follow should not take us by surprise – after all, if Scotus is right, then God’s existence follows already from the mere possibility of being.54 Is not Scotus’ suggestion worth considering?55

54

I attempt a detailed defence of this thesis in my Scire Deum esse. This paper is based on a talk given on March 7th, 2013 at the Winter Conference UNCE organized by the University Centre for the Study of Ancient and Mediæval Tradition of Thought, Charles University, Prague. A slightly different earlier version of this paper in Czech is to appear in Studia Neoaristotelica 11 (2014), no. 3. I thank the conference participants (especially V. NČmec) for stimulating criticism, Bill Vallicella for reading and commenting on the draft of the paper, and the two anonymous reviewers of the Czech version and especially the editors of this volume, G. Klima and A. Hall, for a number of very important critical comments. Last but not least, I am indebted to SvČtla Hanke Jarošová for her help with correcting my English. Of course, I am fully responsible for all the remaining shortcomings. 55

APPEN NDIX

Volumes 122, 2014 The Proceeedings of thee Society for Medieval Loogic and Meetaphysics (P.S.M.L.M M.) is the pubblication of th he Society forr Medieval Logic L and Metaphysicss, collecting original o materrials presentedd at sessions sponsored s by the Socieety. Publicatioon in the Proceedings consstitutes prepu ublication, leaving the authors’ rightt to publish (aa possibly moodified version n of) their materials elssewhere unafffected. The Society for Medievall Logic and Metaphysics M (S S.M.L.M.) is a network of scholars founded withh the aim of fostering f collaaboration and d research based on thee recognition that t ™ recovvering the proofound metaph hysical insightts of medievaal thinkers for oour own philosophical thou ught is highly desirable, and, despite the vvast conceptuual changes in the intervvening period d, is still possiible; but ™ this rrecovery is onnly possible if we carefullyy reflect on th he logical frameework in whhich those in nsights were articulated, given g the paraddigmatic diffe ferences betw ween medievaal and moderrn logical theorries. The Societty’s web sitte (http://facu ulty.fordham.eedu/klima/SM MLM/) is designed to serve the puurpose of keeeping each otther up-to-datte on our current projeects, sharing recent r results,, discussing sccholarly questtions, and organizing m meetings. If you are innterested in joining, pleasee contact Gyuula Klima (Ph hilosophy, Fordham Unniversity) by e-mail e at: klim [email protected] © Society foor Medieval Logic L and Mettaphysics, 20114

CONTRIBUTORS

Edward Buckner, Independent Scholar Daniel Davies, Clare Hall, Cambridge Domenic D’Ettore, Marian University Cruz Gonzalez-Ayesta, University of Navarra David Gonzalez-Ginocchio, University of South Bohemia (Budweis)/ Universidad Internacional de La Rioja Lloyd Newton, Independent Scholar Lukáš Novák, Charles University, Prague Stephen R. Ogden, Yale University/Johns Hopkins University Kenneth Seeskin, Northwestern University Jamie Anne Spiering, Benedictine College