The Contents of Perceptual Experience: A Kantian Perspective 9783110372656, 9783110372649

Table of contents :
Acknowledgements
Introduction
I
II
III
IV
1 The Contents of Perceptual Experience: Opposing Views
1.1 The Content View
1.2 Sensory Content vs. Representational Content
1.3 Varieties of Content 1
1.4 Nonconceptual Content (NCC)
1.5 Varieties of Content 2
1.6 Phenomenology
1.6.1 Richness and Fineness of Grain
1.6.2 Perceptual Illusions
1.6.3 Memory
1.6.4 Demonstrative Identification and Reference
1.7 Perceptual Content and Empirical Beliefs
1.8 Content and Normativity
2 Are the Roots of the Debate Kantian?
2.1 McDowell’s Kant
2.2 Kant and the “Myth of the Given”
2.3 Kant’s Notion of Experience and the Conceptualism vs. Nonconceptualism Debate
2.4 Kant, Nonconceptual Content, and Body in Cognition
3 Kant on Nonconceptual Content: Sensations and Intuitions
3.1 Representational Content Revisited
3.2 Nonrepresentational Content: Sensations
3.3 Nonconceptual Content: Intuitions
3.4 The Forms of Intuition
3.4.1 Forms as a priori Intuitions
3.4.2 The Separation Argument
3.4.3 The Objects of Pure Intuition
3.5 The Ambiguity of “Perceiving”
4 Kant on Concepts in Experience
4.1 The Point of Departure
4.2 “The Same Function”
4.3 Three Syntheses: Does Apprehension Require Recognition?
4.3 Non-Conscious Spontaneity? Schematism and the Transcendental Imagination
4.4 The “I think” of the Transcendental Apperception
4.5 Syntheses and Intuitions: Are Space and Time “Given” or “Constructed”?
4.6 Non-Cognitive Perception
5 Nonconceptual Content and Transcendental Idealism
5.1 What Is Transcendental Idealism?
5.2 And How Does It Bear on Nonconceptualism?
5.3 The Empirical and the Transcendental
5.4 Sensory Content and Cognitive Constraints
5.4.1 “The problem of affection”
5.4.2 Space and Time
5.5 Empirical Realism, Transcendental Idealism, and Nonconceptual Content
5.6 Concepts, Intuitions, and Transcendental Idealism
6 Kant and Naturalism about the Mind
6.1 Spinoza and Hume
6.2 Nonconceptualism and Naturalism
6.3 McDowell’s Naturalism of the “Second Nature”
6.4 Kant’s “Transcendental Psychology” and Naturalism
Conclusion
Bibliography
Index

Citation preview

Anna Tomaszewska The Contents of Perceptual Experience: A Kantian Perspective

Anna Tomaszewska

The Contents of Perceptual Experience: A Kantian Perspective Managing Editor: Anna Michalska Language Editor: P. Christian Adamski

Published by De Gruyter Open Ltd, Warsaw/Berlin.

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 license, which means that the text may be used for non-commercial purposes, provided credit is given to the author. For details go to http://creativecommons.org/licenses/by-nc-nd/3.0/.

Copyright © 2014 Anna Tomaszewska ISBN 978-3-11-037264-9 e-ISBN 978-3-11-037265-6 Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.dnb.de. Managing Editor: Anna Michalska Language Editor: P. Christian Adamski www.degruyteropen.com Cover illustration: © Istock/lcsdesign

Contents Acknowledgements  Introduction  1 1.1 1.2 1.3 1.4 1.5 1.6 1.6.1 1.6.2 1.6.3 1.6.4 1.7 1.8 2 2.1 2.2 2.3 2.4



 1

The Contents of Perceptual Experience: Opposing Views The Content View  13 Sensory Content vs. Representational Content  16 Varieties of Content 1  17 Nonconceptual Content (NCC)  18 Varieties of Content 2  22 Phenomenology  25 Richness and Fineness of Grain  25 Perceptual Illusions  26 Memory  27 Demonstrative Identification and Reference  29 Perceptual Content and Empirical Beliefs  31 Content and Normativity  35

Are the Roots of the Debate Kantian?  39 McDowell’s Kant  39 Kant and the “Myth of the Given”  44 Kant’s Notion of Experience and the Conceptualism vs. Nonconceptualism Debate  47 Kant, Nonconceptual Content, and Body in Cognition  49

3 Kant on Nonconceptual Content: Sensations and Intuitions 3.1 Representational Content Revisited  54 3.2 Nonrepresentational Content: Sensations  60 3.3 Nonconceptual Content: Intuitions  67 3.4 The Forms of Intuition  73 3.4.1 Forms as a priori Intuitions  76 3.4.2 The Separation Argument  76 3.4.3 The Objects of Pure Intuition  77 3.5 The Ambiguity of “Perceiving”  80 4 4.1 4.2 4.3

 13

 54

Kant on Concepts in Experience  84 The Point of Departure  84 “The Same Function”  85 Three Syntheses: Does Apprehension Require Recognition?

 87

4.4

Non-Conscious Spontaneity? Schematism and the Transcendental Imagination  91 The “I think” of the Transcendental Apperception  93 Syntheses and Intuitions: Are Space and Time “Given” or “Constructed”?  95 Non-Cognitive Perception  100

4.5 4.6 4.7

5 Nonconceptual Content and Transcendental Idealism  104 5.1 What Is Transcendental Idealism?  104 5.2 And How Does It Bear on Nonconceptualism?  105 5.3 The Empirical and the Transcendental  108 5.4 Sensory Content and Cognitive Constraints  111 5.4.1 “The problem of affection”  112 5.4.2 Space and Time  113 5.5 Empirical Realism, Transcendental Idealism, and Nonconceptual Content  115 5.6 Concepts, Intuitions, and Transcendental Idealism  120 6 6.1 6.2 6.3 6.4



Kant and Naturalism about the Mind  126 Spinoza and Hume  126 Nonconceptualism and Naturalism  128 McDowell’s Naturalism of the “Second Nature”  132 Kant’s “Transcendental Psychology” and Naturalism  136

Conclusion Bibliography

 142  146

Acknowledgements This book is based on my doctoral dissertation defended in 2011 at the Jagiellonian University in Kraków. I would like to thank my doctoral supervisor, Professor Miłowit Kuniński, and reviewers, Professor Stanisław Judycki and Dr. Jens Timmermann, for their helpful comments, constructive criticism, and discussions. I am also grateful to the members of the Bednarowski Trust in Aberdeen for the research fellowship I was granted in 2012, which allowed me to work on some parts of the manuscript of this book. The remaining parts were drafted with the help of the funding within an EU-sponsored programme “Science – Environment – Technology,” which I received from the Jagiellonian University for the years 2013-2015. I am also grateful to the reviewers and editors of the manuscript and especially to Dr. Anna Michalska for her many helpful suggestions. Special thanks, however, I would like to address to Hasse, my husband, and to my family: Urszula, Krzysztof, and Natalia – for all their help and encouragement.

Introduction I Philosophy can be considered a collection of ideas and problems that remain the same across history and terminological changes. Different philosophical traditions, on such a construal, do not contribute to the development of philosophy but provide distinct conceptual resources by means of which to express perennial questions. Communication between representatives of different traditions in philosophy would thus resemble translation from one language into another, rather than a genuine dialogue. But philosophical problems can also emerge in response to certain developments witnessed in human history: for instance, to the development of scientific thought or to the alterations in social and political conditions. On this view, philosophy remains a collection of problems but philosophical problems undergo evolution as does everything else. It seems, however, that a combination of these two ways of understanding philosophy is also possible. In one of the introductory passages of John McDowell’s Mind and World, one can read: “It is true that modern philosophy is pervaded by apparent problems about knowledge in particular. But I think it is helpful to see those apparent problems as more or less inept expressions of a deeper anxiety – an inchoately felt threat that a way of thinking we find ourselves falling into leaves mind simply out of touch with the rest of reality, not just questionably capable of getting to know about it.”1

McDowell presents a modern philosophical problem, raised in a determinate historical context, as an instantiation of a more universal problem. In particular, the problem of justifying empirical beliefs can be viewed as a special case of the problem of the relation between thought and experience or, even more generally, between thought and reality, or mind and world. Thus, it can be concluded that, since particular problems, emerging in particular historical contexts, express more general or universal ones, parallels can be sought between problems across different philosophical traditions. As Wilfrid Sellars has put it: “The history of philosophy is the lingua franca which makes communication between philosophers, at least of different points of view, possible. Philosophy without the history of philosophy, if not empty or blind, is at least dumb. Thus, if I build my discussion of contemporary issues on a foundation of Kant exegesis and commentary, it is because, as I see it, there are enough close parallels between the problems confronting him and the steps he took to solve

1 J. McDowell, Mind and World, Harvard University Press, Cambridge MA 1996, p. xiii. © 2014 Anna Tomaszewska This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.

2 

 Introduction

them, on the one hand, and the current situation and its demands, on the other, for it to be helpful to use him as a means of communication, though not, of course, as a means only. In their most general aspect both his problems and our perplexities spring from the attempt to take both man and science seriously.”2

According to Sellars, historians of philosophy work on problems shared by philosophers both recently and in the past. It is because philosophy builds on rational argumentation that communication between philosophers, distant in time, is possible. What is more, Sellars seems to appreciate an approach, adopted by Hegel and later German philosophers, such as Dilthey, on which the work of a historian of philosophy is the work of a philosopher proper. The expression “man and science,” one may surmise, refers to two different domains: the “space of reasons” and the “realm of (causal) law.” The overarching problem is how to bring together these two domains, symbolizing human rationality, on the one hand, and nature, on the other. The problem is indeed Kantian in spirit and can be traced back to the Third Antinomy in the Critique of Pure Reason, where Kant investigates the possibility of reconciling the spontaneity of freedom with the determinism of nature.3 To take man and science seriously means to find room for both the freedom of will and the determinism of nature, thus to make morality compatible with science. But the problem of how to take “man and science seriously” also echoes the kind of concerns that were manifested at the beginning of the twentieth century by Edmund Husserl, the founder of phenomenology.4 These concerns were related to the critique of scientism, a philosophical outlook, represented by logical positivists, on which only empirically verifiable statements can pretend to truth-valuation, and all the “non-scientific” discourse of other disciplines (in particular metaphysics) is considered meaningless.

2 W. Sellars, Science and Metaphysics. Variations on Kantian Themes, Ridgeview, Atascadero CA 1992, p. 1. 3 An antinomy is a conflict of two theses which, taken together, generate a contradiction. Thus, in the third antinomy the first thesis reads: “Causality in accordance with laws of nature is not the only one from which all the appearances of the world can be derived. It is also necessary to assume another causality through freedom in order to explain them.” And the second thesis (“antithesis”) reads: “There is no freedom, but everything in the world happens solely in accordance with laws of nature.” I. Kant, Critique of Pure Reason, translated and edited by P. Guyer and A. W. Wood, Cambridge University Press, Cambridge 1998, pp. 484-5 (A 444/B 472-A 445/B 473). 4 According to Christopher Norris, contemporary analytic authors who recruit Kant in the debate on the relation between thought and experience have completely overlooked Husserl and the phenomenological tradition, to the detriment of the debate. See: “‘Second Nature,’ Knowledge, and Normativity: Revisiting McDowell’s Kant,” Diametros 27, 2011.



I

 3

In this book, I deal with a particular philosophical problem, raised within contemporary epistemology, philosophy of mind and perception, against the background of Kant’s theory of cognition. More specifically, my considerations focus on the question about the nature of representational content in experience. This is a very specific question that one can also address by studying the Kantian account of empirical cognition in general, and intuition in particular. Such an approach has been adopted by a number of authors: Wilfrid Sellars, John McDowell, Robert Brandom, and Gareth Evans, to mention but a few. Many authors work at the intersection of Kant commentary and contemporary theories of mind and cognition; as it seems, the majority of commentators and historians of philosophy recognize the great relevance of the Kantian doctrines to current issues.5 Thus, I do not aim to add up more than a voice in an ongoing debate that has both Kant and contemporary philosophers as its participants. Moreover, the problem to be dealt with here – namely, one of the nature of experience – can be connected with more general background provided by Kant’s transcendental philosophy. Again, these more general problems bear on the relation between thought and reality, as well as on the relation between man and nature. It is in Kant that one can find a thorough reshaping of the mind – world relation as a result of the so-called Copernican revolution. With his claim that objects must conform to the rules intrinsic to the human faculty of cognition, Kant goes beyond both classical metaphysics, which endorsed the idea of the universal intelligibility of the world, and some modern “veil of perception” doctrines, which equated objects with ideas in the mind. Kant’s “science of sensibility,” another name for the Transcendental Aesthetic, provides a framework within which to discuss the relation between man and nature, the “space of reasons” and the “realm of law,” the mental and the natural (in the sense of the Aristotelian “first nature”). As I read him, however, Kant does not offer a comprehensive picture of the relation, free from inconsistencies; rather, he leaves us with a number of puzzling questions.6 This signals a need to overcome the dichotomy

5 To name several examples: recent publications include a book on Kant and the content debate: D. Heidemann (ed.), Kant and Non-Conceptual Content, Routledge, London and New York 2013; and on Kant and the psychology of the unconscious: P. Giordanetti, R. Pozzo, M. Sgarbi (eds.), Kant’s Philosophy of the Unconscious, Walter de Gruyter, Berlin 2012. The list of names of scholars writing in a similar vein would be long and one could include therein: Henry Allison, Lucy Allais, Richard Aquila, Andrew Brook, Hannah Ginsborg, Paul Guyer, Dieter Henrich, Robert Hanna, David Hamlyn, Jaakko Hintikka, Patricia Kitcher, Beatrice Longuenesse, Christopher Norris, Leslie Stevenson, Wayne Waxman, Kenneth Westphal, Markus Willaschek, Crispin Wright, John Yolton, and many others. 6 To illustrate the point, one may consider the so-called problem of affection. According to Kant, the origin of the “matter” of cognition should be traced back to the affection relation between the mind and its object(s), which involves no more than the subject’s receptivity. How this relation should be construed has been subject to numerous debates. John Yolton, for example, distinguishes between

4 

 Introduction

in terms of which we tend to think about the issue: for we either consider mind as part of nature, or as radically autonomous from it. McDowell’s account of “second nature” marks an attempt at abandoning the dichotomy; in my opinion, though, it reaches piecemeal success only.

II Opening the first chapter of Mind and World, McDowell writes: “One of my main aims is to suggest that Kant should still have a central place in our discussion of the way thought bears on reality.”7

As much as I share this aim, I do not agree with McDowell that Kant can be read as an advocate of conceptualism, a view on which perceptual experience is conceptdependent, or structured by concepts. On the contrary, my aim is to show that Kant’s theory of empirical cognition has much more to offer to the proponents of the opposite view. Thus, it can be shown that Kant held the view that there is a conceptindependent and pre-conceptual way of representing objects. This way is provided by intuition or intuitive cognition (Anschauung). In a number of places, throughout his philosophical writings, both pre-critical and later, Kant furnishes arguments for this view. Let me give a few examples. In The Only Possible Argument in Support of a Demonstration of the Existence of God, Kant introduces a distinction between logical and real possibility. What can be thought without contradiction is logically possible. But what is really possible is thinkable on account of the fact that it has been or can be experienced. Thus, merely intellectual cognition and empirical cognition go separate ways in that the latter involves a nonconceptual ingredient: the experience of existence, or reality, of an object. Kant writes about the disparities between purely conceptual and empirical cognition, for instance, here: “The motive force of a body in one direction and an equally strong tendency in the opposite direction do not contradict each other. They are also really possible in one body at the same

a merely causal (hence physical) perceptual relation and a cognitive or epistemic relation that does not presuppose any temporal sequence of events in a causal connection. Kant’s affection relation would be of the second kind, according to Yolton. (Cf. Perception and Reality: A History from Descartes to Kant, Cornell University Press, Ithaca 1996, ch. 7.) A different reading can be found in: R. Aquila, Representational Mind: A Study of Kant’s Theory of Knowledge, Indiana University Press, Bloomington 1983. According to Aquila, the affection relation involves “stimulation of sense organs by an object” (p. 64) and results in producing sensory content. 7  J. McDowell, Mind and World, op. cit., p. 3.

II 

 5

time. However, one motive force annihilates the real consequences of the other motive force; and since the consequence of each motive force by itself would otherwise be a real movement, the consequence of both together in one subject is naught. That is to say, the consequence of these opposed motive forces is rest. But rest is, indubitably, possible. From this it is also apparent that real opposition is something quite different from logical opposition or contradiction, for the result of the latter is absolutely impossible.”8

In the same work, Kant makes a point that there are limits to conceptual analysis and that terms that defy any further analysis acquire their meaning by virtue of their relatedness to some sort of the “given.” A similar claim is advanced in a later essay, entitled Inquiry Concerning the Distinctness of the Principles of Natural Theology and Morality. The claim will later transform into a “principle of significance” according to which, roughly speaking, concepts derive their meaning from intuitions.9 Compare: “Suppose that you can now no longer break up the concept of extension into simpler data in order to show that there is nothing self-contradictory in it – and you must eventually arrive at something whose possibility cannot be analysed – then the question will be whether space and extension are empty words, or whether they signify something. The lack of contradiction does not decide the present issue; an empty word never signifies anything self-contradictory. If space did not exist, or if space was not at least given as a consequence through something existent, the word ‘space’ would signify nothing at all. As long as you prove possibilities by means of the law of contradiction, you are depending upon that which is thinkable in the thing and which is given to you in it, and you are only regarding the relation in accordance with this logical rule. But in the end, when you consider how this is then given to you, the only thing to which you can appeal is an existence.”10

And another passage: “Before I set about the task of defining what space is, I clearly see that, since the concept is given to me, I must first of all, by analysing it, seek out those characteristic marks which are initially and immediately thought in that concept. Adopting this approach, I notice that there is

8 In: I. Kant, Theoretical Philosophy, 1755 – 1770, translated and edited by D. Walford in collaboration with R. Meerbote, Cambridge University Press, New York 1992, p. 130 (2:86). 9 ����������������������������������������������������������������������������������������������������� More specifically, the “principle of significance” has been defined by Peter Strawson in the following manner: “This is the principle that there can be no legitimate, or even meaningful, employment of ideas or concepts which does not relate them to empirical or experiential conditions of their application.” The Bounds of Sense. An Essay in Kant’s ‘Critique of Pure Reason’, Methuen, London 1966, p. 16. My understanding of the principle above draws upon Kant’s famous claim, formulated in the introduction to the Transcendental Logic, that concepts unaccompanied by intuitions are “blind.” Cf. I. Kant, Critique of Pure Reason, op. cit., pp. 193-4 (A 51/B 75). 10  In: I. Kant, Theoretical Philosophy, 1755 – 1770, op. cit., pp. 125-6 (2:81).

6 

 Introduction

a manifold in space of which the parts are external to each other; I notice that this manifold is not constituted by substances, for the cognition I wish to acquire relates not to things in space but to space itself; and I notice that space can have only three dimensions etc. Propositions such as these can well be explained if they are examined in concreto so that they come to be cognised intuitively; but they can never be proved.”11

It would perhaps be slightly anachronistic to say that Kant approaches the view that logic is based on the mechanisms by means of which the mind works. However, he does emphasize an asymmetry between intuitive and conceptual cognitions. While there are intuitive cognitions that cannot be analyzed by means of concepts, or cognitions that originate directly from intuition, without conceptual mediation, grasping even basic logical principles, such as the law of contradiction, requires recourse to intuition. Thus, in Kant’s Inaugural Dissertation, marking transition into the critical period, and entitled On the Form and Principles of the Sensible and the Intelligible World, one can read: “Indeed, far from it being the case that anyone has ever yet deduced the concept of time from some other source, or explained it with the help of reason, the very principle of contradiction itself presupposes the concept of time and bases itself on it as its condition. For A and not-A are not inconsistent unless they are thought simultaneously (that is to say, at the same time), about the same thing, for they can belong to the same thing after one another (that is to say, at different times). Hence, it is only in time that the possibility of changes can be thought, whereas time cannot be thought by means of change, only vice versa.”12

By the end of the Introduction to the Critique of Pure Reason Kant raises a point about the priority of intuition to thought, by stating: “[T]he conditions under which alone the objects of human cognition are given precede those under which those objects are thought.”13

Since the conditions under which objects can be given in intuition are presupposed by the conditions of thought, they must be more basic and essential, and hence independent of the latter: at issue here is constitutive dependence rather than conceptual or genetic (temporal) priority of one kind of cognition to the other. It follows that objects can be given in intuition without at the same time having to be thought. Interestingly, whereas the pre-critical Kant acknowledged the autonomy of purely conceptual cognition, the critical Kant does not seem to appreciate this kind

11 Ibidem, pp. 253-4 (2:281). 12 Ibidem, p. 394 (2:401). 13 I. Kant, Critique of Pure Reason, op. cit., p. 152 (A 16/B 30).

III 

 7

of cognition. In addition to that, he thinks that it is intuition that makes conceptual cognition – thought and judgment – laden with meaning, or related to an object. This means that Kant endorses the reversal of the conceptualist claim that only concepts can endow experience with an intelligible structure. Therefore, as I argue in this book, Kant is much closer to nonconceptualism than to the opposite position.

III Yet one may assess the approach presented above as somewhat flawed. Indeed, its flaws may become evident at different stages and levels of the discussion. Some of them would result from what may look like a conflation of incompatible discourses and from bringing together philosophical traditions that do not have much in common, others would signal flaws inherent in the conceptualism vs. nonconceptualism debate itself. Let me briefly address these two possible objections. 1. The first objection would point to the fact that analytic philosophy came into being due to, roughly speaking, two factors: the development of mathematical logic, with the works of Gottlob Frege and Bertrand Russell, and the “linguistic turn,” with Wittgenstein as its main motive force. Neither of these two pillars of the analytic philosophy seems to be essential to Kant. With his invention of the transcendental logic – that is, logic discovering the necessary ways of thinking about the objects of cognition14 – Kant can be situated much closer to those cognitive scientists, like Eleanor Rosch or Peter Gärdenførs, who investigate the basic categories by means of which subjects organize the field of experience.15 What is more, Kant seems to overlook the importance of language as the main factor which affects the way we think about the world, and stay in line with the Cartesian “mentalist” paradigm. All this can be granted. However, one can observe that for the past (more or less) thirty years in contemporary philosophy the analysis of language has given way to the analysis of mind and cognition; analytic philosophy seems to have been outranked by cognitive science or at least by the kind of philosophizing that takes into account the results of scientific research. Since, on certain readings, Kant comes close to this paradigm, there is a good reason to look for similarities, interdependencies and inspirations between the two philosophical traditions. As I have pointed out, there are “big problems” behind the debate on the nature of perceptual experience, problems

14 See: ibidem, pp. 195-6 (A 56/B 80 – A 57/B 82). In Kant’s definition: “Such a science, which would determine the origin, the domain, and the objective validity of such cognitions, would have to be called transcendental logic, since it has to do merely with the laws of the understanding and reason, but solely insofar as they are related to objects a priori” (ibidem, pp. 196-7, A 57/B 82). 15 Cf. Robert Piłat, Doświadczenie i pojęcie [Experience and Concept], IFiS PAN, Warszawa 2006, ch. 1.

8 

 Introduction

whose origin one can trace back to Kant. Also, it is worth noting that Kant’s account of cognition significantly diverges from the Cartesian one, for example in the theory of judgment: whereas for Kant judgment is a complex representation, a combination of concepts, furnished with objective validity, that is (roughly speaking), reference to an object, for Descartes judgment consists in a mental act in which the will accepts or rejects a particular cognition (or idea, in the Cartesian parlance).16 Since judgments, according to Kant, mediate cognition, it would perhaps be more appropriate to think of them as intersubjectively shareable and therefore expressible in a language, rather than in terms of (private) mental acts. Last but not least, some scholars emphasize that it is in fact already in the British empiricist tradition that the importance of language for cognition comes into view,17 thus the roots of the “linguistic turn” could perhaps be shifted back as early as to Locke. Interestingly, also philosophers such as Thomas Reid, who founded the eighteenth-century Scottish school of common sense, claimed that language reflects the basic structure of human thought. 2. Another objection may state that participants in the conceptualism vs. nonconceptualism debate not only use certain concepts that do not figure in Kant’s vocabulary (such as intentionality or content), and not only do they use certain concepts in a different way than Kant does (as is the case with the concept of representation or experience), but there are also concepts which are employed throughout the debate without being properly defined. This pertains to, for example, the concept of concept: since it is difficult to see which theory of concepts underlies the debate, it may also be unclear what it means to claim that the content of perceptual experience is conceptual, or structured by concepts. And, since we do not know what concepts are, we cannot determine the conditions of concept-possession. Finally, even if particular authors explain how they understand concepts, there is no unitary account of concepts that would be accepted by all theorists of conceptual and nonconceptual content. This may lead to confusion and render the whole debate pointless. Again, I quite agree with this objection. Terminological confusion is a real problem. Whereas neo-Fregeans would define concepts as abstract entities encapsulating the senses of linguistic expressions, cognitive scientists would regard them as a kind of mental representations, an approach Kant would perhaps be more sympathetic to. Laurence and Margolis, in their anthology on concepts, provide a systematic overview of theories of concepts including: classical theory, prototype theory, theory-theory, neoclassical theory, and conceptual atomism – a wide palette of options from which

16  For more on the anti-Cartesian dimension of Kant’s theoretical philosophy, see: Arthur W. Collins, Possible Experience. Understanding Kant’s ‘Critique of Pure Reason’, University of California Press, Berkeley and Los Angeles 1999. 17 See, e.g.: A. Flew, “Was Berkeley a Precursor of Wittgenstein?” in: W. B. Todd (ed.), Hume and the Enlightenment: Essays Presented to Ernest Campbell Mossner, Edinburgh University Press, Edinburgh 1974.

III 

 9

to choose.18 McDowell speaks of “conceptual capacities” involved in experience but he is not at all explicit about what exactly possessing these capacities involves. Is it necessary for concept-possession to exercise certain abilities – such as discrimination, re-identification, recognition, belief-formation – taken jointly, or does any of these abilities, separately taken, suffice to have the “conceptual capacities”?19 If the conceptualist and nonconceptualist hold different requirements as to what qualifies as a concept, or a conceptual capacity, and what does not, a debate between them will unavoidably end up in mere disagreement about words. The debate I discuss in this book is flawed by the notorious lack of clarity about what concepts are. A way out of this predicament would be to admit that how concepts are understood throughout the debate is not really essential to it. The idea would be that the debate develops at a higher level of generality, which allows leaving certain issues, like that of the theory of concepts, unspecified. As much as Descartes, in the fifth part of the Discourse on Method, did not need to define language in order to formulate an argument for the distinctively human character of rationality, which precludes that animals have reason,20 philosophers who argue for or against the conceptual nature of experience would not need to provide a definition of concept because what is at issue is the rationality of experience and hence of the mind – world relation itself, a feature which implies conceptual representation, whatever the nature of concepts might be. In other words, one would not need to explain what concepts are in order to recognize the merely explanatory value of arguments appealing to conceptual and/ or nonconceptual content.

18 The most popular (perhaps until recently) classical theory of concepts is currently in retreat. One may criticize it for imposing too strict epistemic requirements on subjects; also, for its failure to provide a plausible account of the structure of most concepts (very few concepts have definitions, subjects can competently apply concepts without knowing their definitions, etc.). Cf. Stephen Laurence, Eric Margolis, “Concepts and Cognitive Science” in: E. Margolis, S. Laurence (eds.), Concepts. Core Readings, The MIT Press, Cambridge MA 1999, pp. 3-81. 19 Robert Brandom suggests a different approach: following one of Kant’s ways of understanding concepts, he defines concepts as rules in that he writes: “To call something ‘necessary’ is to say that it happens according to a rule, and everything that happens in nature, no less than everything done by humans, is subject to necessity in this sense. Concepts are rules, and concepts express natural necessity as well as moral necessity. So according to him [i.e., Kant] there is strictly no non-normative realm – no realm where concepts do not apply. Kant’s fundamental innovation is best understood to consist in his employment of a normative meta-language in specifying both what merely happens and what is done.” Making it Explicit. Reasoning, Representing, and Discursive Commitment, Harvard University Press, Cambridge MA 1998, p. 624. 20 R. Descartes, Discourse on Method, Optics, Geometry, and Meteorology, translated by P. J. Olscamp, Bobbs-Merrill, Indianapolis 1965.

10 

 Introduction

IV The aim of this book is twofold. First, it is to show that from Kant’s account of experience, and in particular sensible intuition, one can derive arguments relevant to the debate about the content of perceptual experience. Second, it is to provide a reading of Kant as an advocate of nonconceptualism, i.e. a theory of content which denies that concepts structure the (entire) content of experience and that concept-possession necessarily underlies conscious perceptual states. As such, this reading challenges McDowell’s influential interpretation of Kant’s doctrine as paradigmatically conceptualist. There are several reasons why predominantly McDowell’s conceptualism is addressed throughout the book, rather than nuanced versions of the view by other authors. First, unlike for example B. Brewer, McDowell consistently maintains his position; whereas Brewer, in Perception and Reason,21 develops an argument for conceptualism, in his later works on perception, where he defends disjuctivism, he withdraws from maintaining what he calls the “content view.”22 On this later account, concepts are introduced in perception on the level of interpretation or categorization of what perception otherwise directly manifests to the subject, rather than necessitating the very perception of objects. Second, McDowell, unlike for example W. Sellars, represents a radical version of conceptualism, on which perceptual content is structured by concepts through and through; even the most primitive impressions of the senses, like sensations of red or high-pitched tones, are, according to McDowell, conceptually informed. On the other hand, the difference between moderate conceptualism and nonconceptualism is merely nominal: both positions allow a degree of conceptual uninformedness of perceptual content. Thus, a real debate can be held between radical conceptualism and its adversaries. Third, McDowell’s conceptualism purports to form part of a comprehensive view of the relation between mind and world and human rationality. The employment of conceptual capacities sharply distinguishes rational beings, possessors of what McDowell – following Aristotle – calls “second nature,” from non-rational animals and other beings. It is, in my opinion, noteworthy that whereas McDowell construes the employment of concepts as indicative of rationality, Kant takes a step further in that he attempts to show that rationality has more basic, non-rational, underpinnings.

21 B. Brewer, Perception and Reason, Oxford University Press, Oxford 1999. 22 Cf. Bill Brewer, “Perception and Content,” European Journal of Philosophy 14 (2), 2006; Perception and its Objects, Oxford University Press, Oxford 2011. Also McDowell subscribes to the disjunctive theory of perception but without abandoning the “content view.” Cf. e.g. “Singular Thought and the Extent of Inner Space” in: P. Pettit and J. McDowell (eds.), Subject, Thought and Context, Clarendon Press, Oxford 1987; “The Disjunctive Conception of Experience as Material for a Transcendental Argument” in: F. Macpherson and A. Haddock (eds.), Disjunctivism: Perception, Action, Knowledge, Oxford University Press, Oxford 2008.

IV

 11

The book has the following structure. In chapter 1 I present the main positions in the debate on the content of perceptual experience. I distinguish between state and content conceptualism and, respectively, nonconceptualism. I also distinguish between the kinds of content involved in perception: from nonrepresentational sensations to representational, concept-dependent perceptual states – and outline the arguments, exploited most frequently by the parties in the debate. Chapter 2 starts by presenting Kant’s account of experience in the light of the moderately conceptualist interpretation of Sellars and radically conceptualist interpretation of McDowell. I signal where, on my reading, both interpretations go wrong: namely, in their inability to explain the status and the role of pure intuitions (the a priori forms of intuition) in cognition. I suggest then that the forms constitute the explanatory basis for the nonconceptual content of experience and formulate an interpretative hypothesis that they originate from the bodily nature of the subject. Textual evidence for this hypothesis may be searched in some pre-critical writings of Kant or, on a certain reading, in a section of the Critique of Pure Reason called the Refutation of Idealism. Emphasizing the role of body in Kant’s account of empirical cognition makes my proposal aligned with the nonconceptualism of Robert Hanna, Christopher Peacocke, or Gareth Evans. Chapters 3 and 4 have a mainly expository character. Chapter 3 opens with a discussion of Kant’s notion of representation and a number of ambiguities related to his construal of sensory content. Subsequently, I characterize intuitions, the deliverances of sensibility, which, according to Kant, can be described as heterogeneous and thus autonomous from concepts, the products of the understanding. Therefore, what I label Kant’s Cooperation Thesis does not, as McDowell takes it, invite conceptualism. Rather, Kant’s conception of intuition makes room for a distinction, advocated by Fred Dretske, between epistemic and non-epistemic senses of perception. Chapter 4 challenges the nonconceptualist reading of Kant’s account of experience in that it brings to the fore the aspects of this account that seem to testify to the plausibility of an opposite interpretation. These aspects involve the doctrine of syntheses and the “I think” of the transcendental apperception, present in the Transcendental Deduction of the Categories. I attempt to address these challenges by arguing that Kant was not clear enough whether all representational content derives from a synthesis in accordance with a determinate concept (if he held space and time for representations, he might have conceded the status of unsynthesized representations to them). And even if he claimed that all representations result from concept-involving syntheses, applying a concept in experience can be conceived in a way that does not preclude at least state-nonconceptualism: this is the case when a synthesis is rule-governed but proceeds non-consciously (or “blindly”). Finally, the “I think” passage, as I read it, expresses the requirement that all representations be poised for conceptualization, thus it can be modeled along the lines of Gareth Evans’ two-tier account of representational (informational) content.

12 

 Introduction

In chapter 5 I investigate the links between Kant’s transcendental idealism and Kant’s nonconceptualism. This is a relatively unexplored part of this area of Kant scholarship; Hanna, for example, thinks that the two doctrines are logically independent. Setting Kant’s against Hegel’s idealism, I explain why a nonconceptualist account of experience better motivates transcendental idealism, which I take to be (largely) a doctrine about epistemic constraints intrinsic to human mind, than does the opposite position. Since these constraints can ultimately be attributed to the bodily nature of the subject, in accordance with my hypothesis, chapter 6 raises a question whether reading Kant as a nonconceptualist opens up a way to naturalizing his theory of the mental. My suggestion is that such a move is far from necessary.

1 The Contents of Perceptual Experience: Opposing Views 1.1 The Content View When I observe a scene around me, I perceive a number of objects: I can see my desk with a pile of books and papers on it (in slight disarray), some writing implements and a notebook; I can hear a car passing the nearby road and friends chatting outside my window; I can sense the smell of freshly baked bread and feel the smoothness of my cat’s fur. What are the objects of my perception – in general? Surely, I can think about these objects, as well as imagine them. But there must be something in perception that makes it different from other mental states, acts or processes, such as thinking or imagining. For example, smelling a rose is different from merely thinking about a rose. When you smell a rose, you have certain sensations that you would not have if you were merely thinking about it. Moreover, you would not know what the scent of rose is like, you would not be able to recognize it as coming from your garden, and to tell it apart from the scent of lilacs that grow next to roses, if you did not have a perceptual experience of a rose. Accordingly, philosophers have claimed that what distinguishes perception from other kinds of mental states, acts, or processes, is its content. Since the objects that can be perceived can also be thought about or imagined, perceptual content is a special way or mode in which objects can be given, or presented, to us in experience or, more broadly, cognition. Let us call this view the content view (CV).23 On the CV, cognitive access to objects would be regarded as mediated, rather than direct. The role of the intermediary would be played by representational content. This view is deeply rooted in the history of philosophy, for example in the debates between direct and indirect realists, such as Thomas Reid, on the one hand, and John Locke, on the other.24 The latter, also known as representationists, would claim that “ideas,” which would be characterized as mind-dependent, or existing “in” the mind, mediate access to all kinds of objects. The standard objection raised against the CV states that it encourages skepticism about the properties of the objects outside the

23 For a discussion and critique of the view, see: B. Brewer, “Perception and Content,” European Journal of Philosophy 14 (2), 2006, pp. 165-181. 24 John W. Yolton suggests that this is actually a misreading of Locke’s view, promoted by Reid and endorsed by a number of commentators, such as Jonathan Bennett or John L. Mackie. See, in particular: J. Bennett, Locke, Berkeley, Hume: Central Themes, Oxford University Press, Oxford 1971; J. L. Mackie, Problems from Locke, Clarendon Press, Oxford 1976; J. W. Yolton, Perceptual Acquaintance from Descartes to Reid, Basil Blackwell and University of Minnesota Press, London 1984; J. W. Yolton, Perception and Reality: A History from Descartes to Kant, Cornell University Press, Ithaca 1996. © 2014 Anna Tomaszewska This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.

14 

 The Contents of Perceptual Experience: Opposing Views

mind by introducing a “veil of perception,” a term coined by Jonathan Bennett in his commentary on Locke.25 Besides, as Bill Brewer puts it, on the CV, “perceptual experience trades direct openness to the elements of physical reality themselves, for some intellectual act of classification or categorization.”26 But, it seems reasonable to surmise, direct openness to the world cannot always be the case, since subjects sometimes fall prey to perceptual illusions. Thus, it is claimed, the content of perceptual experience provides the “highest common factor”27 to the veridical and non-veridical perception, the element that both kinds of perception are supposed to share. The “highest common factor” can be equated with what philosophers also refer to as representational content, or simply – representation. Kazimierz Twardowski28 argued that each act of representation, each “representing,”29 has a “double object:”30 primary and secondary.31 The former is what is represented in the act of representation, what the act of representation refers to or is about; the latter is what the object represented in the act of representation is represented by. For example, when an artist paints a landscape on a canvas, the landscape, the object painted, is both the real object that the canvas is about and the object painted in the canvas, thus the latter’s content. But to what extent is Twardowski’s comparison accurate? One may argue that there can be hardly any phenomenological evidence for the double-object view: whatever object I perceive, I do not seem to be aware of the object of my perception and the thing by means of which the object is perceived (unless I happen to use special equipment, such as a magnifying glass or an AFM microscope). Rather, I simply

25 See the above footnote. 26 B. Brewer, op. cit., p. 18. I refer to the online version of the paper: http://www2.warwick.ac.uk/ fac/soc/philosophy/people/faculty/brewer/pc.ejp.pdf (accessed on 26th November 2012). According to Brewer, the consequences of the CV are that it “loses all right to the idea that it is the actual physical objects before her which are subjectively presented in a person’s perception, rather than any of the equally truth-conducive possible surrogates” (p. 18). Brewer contends that “content does enter a complete account of our perception of the world around us,” however “only as the result of an intellectual abstraction, or generalization, from the basic nature of such experience, given the mode of our attention to its constituent direct objects” (p. 27). Thus, he comes close to conceptualism concerning the contents of perception. 27 See, e.g.: D. Macarthur, “McDowell, Scepticism, and the ‘Veil of Perception,’” Australasian Journal of Philosophy 81 (2), 2003. 28 In: Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. Eine psychologische Untersuchung, Philosophia Verlag, München-Wien 1982. 29 This term has been used by Wilfrid Sellars in order to distinguish the acts from the objects of representation (“representeds”). See: W. Sellars, “Some Remarks on Kant’s Theory of Experience,” The Journal of Philosophy 64 (20), 1967. 30 K. Twardowski, Zur Lehre vom Inhalt und Gegenstand der Vorstellungen, op. cit., p. 14. 31 Ibidem, p. 18.



The Content View 

 15

perceive an object with its diverse properties. Also, in the cases of misperception, I do not seem to be aware of both the illusory content and the misperceived object, unless I realize that I have fallen prey to an illusion (thereby ceasing to be its prey, though). Furthermore, unlike the painter, I do not seem to be aware of the act of projecting the object I perceive on some “inner” canvas. Therefore, the content view might be considered as phenomenologically inaccurate. Some philosophers – e.g. John McDowell,32 M. G. F. Martin,33 P. Snowdon, Bill Brewer or J. M. Hinton – propose thus a disjunctive theory of sense perception in its place.34 In outline, according to the disjunctivists, objects enter into the description of the content of perception, or constitute it. The act of perception can be either veridical and relate to a fact, or a case of an illusion or hallucination and relate to an appearance of a fact. The disjunctivists claim that the “good” and the “bad” disjuncts do not share their contents, even though perceiving may be phenomenologically indistinguishable from hallucinating or undergoing an illusion. However, the disjunctive theory of perception would not silence the doubts of the skeptic: without accounting for the reliability of the process of perception (in causal or other terms), and without a theory of illusions and other states of misperception, the subject cannot justifiably claim that he is perceiving rather than misperceiving a given object, so a skeptic may argue. Addressing these concerns would require separate considerations which I have no space to engage in here; may it suffice to mention that the strategy pursued by the disjunctivists consists in removing the assumptions on which the skeptical challenge rests, such as the “veil of perception” and indirect realism, the transparency of the mind and the infallibility (or authority) of first-person experience. That much, it seems, has to be conceded if one wants to be able to think of experience as a kind of “openness to the layout of reality,” as McDowell puts it.35

32 E.g. in: “Singular Thought and the Extent of Inner Space” in: P. Pettit and J. McDowell (eds.), Subject, Thought and Context, Clarendon Press, Oxford 1987; Mind and World, Harvard University Press, Cambridge MA 1996; “The Disjunctive Conception of Experience as Material for a Transcendental Argument” in: F. Macpherson and A. Haddock (eds.), Disjunctivism: Perception, Action, Knowledge, Oxford University Press, Oxford 2008. 33 Cf. “Transparency of Experience,” Mind and Language 17, 2002. 34 Cf. A. D. Smith, “Disjunctivism and Illusion,” Philosophy and Phenomenological Research 80 (2), 2010. 35 J. McDowell, Mind and World, Harvard University Press, Cambridge MA 1996, p. 26. Elsewhere he writes: “the idea of environmental facts making themselves available to us in perception must be intelligible, because that is a necessary condition for it to be intelligible that experience has a characteristic that is […] not in doubt. The relevant characteristic is that experience purports to be of objective reality. […] [T]he transcendental argument shows that the disjunctive conception is required, on pain of our losing our grip on the very idea that in experience we have it appear to us that things are a certain way.” “Conception of Experience as Material for a Transcendental Argument,” op. cit., 380-2. The strategy seems to beg the question, though.

16 

 The Contents of Perceptual Experience: Opposing Views

1.2 Sensory Content vs. Representational Content In the previous section I made two commonsensical claims that one may take to be grounded in the phenomenology of experience. The first is that we quite naturally tend to distinguish perception from other kinds of cognitive states, acts, or processes, such as thought or imagination, by means of a specific kind of content. The second is that what we have called the content view (CV) does not adequately render the phenomenology of experience; simply, in perceptual experience, one cannot usually distinguish its object, the act of experiencing itself, and its representational content. Rather, the distinction must be based on abstraction or the analysis of the language used in the description of experience. But these claims seem to be incompatible: if the CV is false, then it is not true that we can distinguish perception from, for example, thought by means of a specific kind of content. However, the possible incompatibility builds on the ambiguity of the notion of content. For, apparently, it presupposes that all content is representational, that is, directed at a certain object. But the CV might be false and we may claim that what makes perception different from thought is a certain kind of content, provided that we endorse the idea that perceptual experience may also involve nonrepresentational types of content. Many philosophers have admitted the existence of nonrepresentational content and associated it with sensations.36 Hence, I will call content of this kind sensory content (SC). Accordingly, an opponent of the CV does not have to deny the existence of SC. SC involves both sensations, such as pain, and the so-called secondary qualities, such as colours, sounds, or tastes. Remarks on both kinds of SC can be found, for example, in Descartes’ Sixth Meditation. According to Descartes, the senses do not provide us with clear and distinct perceptions (ideas) of the nature of the objects perceived. Rather, properties of objects as they are “in themselves,” more specifically, geometrical properties, are perceived in abstraction by the intellect. Information delivered by the senses is aimed at preservation of the psychophysical unity, that is, the embodied conscious individual that happens to be identical with myself. Thus, when I feel pain in my leg, what I am supposed to do is not to form a belief about the putative cause or source of my pain but to identify and remove that cause or source. Hence, for Descartes, the purport of the information delivered by the senses is pragmatic, rather than epistemic.37

36 According to Rolf George these philosophers include: Étienne Bonnot de Condillac, Thomas Reid, Johann Nikolaus Tetens, Johann Gottlieb Fichte, Arthur Schopenhauer, Wilhelm von Humboldt, William Hamilton and Immanuel Kant. See: R. George, “Kant’s Sensationism,” Synthese 47 (2), 1981. 37 Cf. René Descartes, Meditations on First Philosophy. With Selections from the Objections and Replies, edited by J. Cottingham, Cambridge University Press, Cambridge 1996.

Varieties of Content 1 

 17

Kant also considers nonrepresentational sensory content and he identifies this content with sensations. Thus, in the so-called Stufenleiter passage in the Critique of Pure Reason, he characterizes sensations as belonging to the class of perceptions “that refer to the subject as […] modification[s] of its state.”38 The Transcendental Aesthetic lists such sensations as “colors, sounds, and warmth.” Kant points out that “the[se] are merely sensations and not intuitions,” and thus they “do not in themselves allow any object to be cognized.”39 In the Anthropology, he gives an example of a child and says that she has sensations and perceptions but no cognition of objects, hence no experience.40 Kant also claims that “the senses do not deceive” since they do not make judgments; therefore, epistemic errors can originate only from the understanding.41 On the basis of such statements, Rolf George holds an arguable position on which “Kant’s theoretical philosophy is a complex extension of sensationism,”42 a view that there are nonrepresentational, non-intentional mental states that lie at the basis of all empirical cognition. On such a view, the content provided by sensibility would be nonrepresentational, unlike on the CV, but it would nevertheless account for the distinction between perception and other mental states, acts, or processes.

1.3 Varieties of Content 1 The content of perceptual experience may thus be divided into representational and nonrepresentational. But within this division further ones are possible. The representational content of perception can be either conceptual or nonconceptual. Since the nonrepresentational content of perceptual experience is sensory, hence not conceptual, nonconceptual content can be either representational or nonrepresentational. Although it is the existence and cognitive significance of the representational nonconceptual content that has become of interest to contemporary philosophers of mind and perception – and we will deal with this for the remaining part of the present chapter – it is noteworthy that, on the basis of the distinction between conceptual content (CC), nonrepresentational (sensory) nonconceptual content (SC) and representational nonconceptual content (NCC), one can also distinguish a number of positions.

38 Immanuel Kant, Critique of Pure Reason, translated and edited by P. Guyer and A. W. Wood, Cambridge University Press, Cambridge 1998, p. 398 (A 320/B 378). 39 Ibidem, p. 178 (A 29/B 44). 40 I. Kant, Anthropology, History, and Education, edited by G. Zöller and R. B. Louden, translated by M. Gregor et al., Cambridge University Press, Cambridge 2007, p. 232 (7:128). 41 Ibidem, p. 258 (7:146). 42 R. George, “Kant’s Sensationism,” op. cit., p. 229.

18 

 The Contents of Perceptual Experience: Opposing Views

Thus, one can endorse both the existence and cognitive significance of CC, SC and NCC. Arguably, such a position can be attributed to Kant who distinguishes sensations, intuitions, and concepts as the components of the content of experience. But one can also endorse a view about the existence and cognitive significance of CC and SC alone. This view has also been attributed to Kant, on some interpretations, and it can be attributed to Christopher Peacocke (in his earlier writings) or Wilfrid Sellars, too. The view on which there is only CC involved in cognition seems fairly radical but, again on some readings, McDowell and Hegel can be said to have advocated it. One may also understand this view as implying the denial of any contribution to cognition from the senses. Such a position would be quite odd; it would describe an “angel-like” rather than human mode of cognition. A creature in whose experience NCC and SC, or SC alone, would figure, would most probably be a non-human animal or a human baby prior to her acquisition and/or use of conceptual capacities. Indeed, according to Kant, very young children are incapable of entertaining experiences; they, as it were, live among a “blooming buzzing confusion of sensations.”43 Finally, some philosophers would acknowledge the distinction between CC and NCC but without admitting the existence and cognitive significance of SC. On certain interpretations, this position can be attributed both to the early modern rationalists (Leibniz) and to the empiricists (Locke, Berkeley, Hume).

1.4 Nonconceptual Content (NCC) When one perceives an object, such as a rose, the object presents itself in a determinate way. One perceives it as, for example, purple and thorny, fully blossomed, and delicate, growing amidst other roses, on a flowerbed, in the rear of a house. All these ways in which one perceives an object constitute the contents of one’s particular perception.44 If perception had no content, one would not be able to perceive things as being certain

43 José L. Bermúdez, “Nonconceptual Content: From Perceptual Experience to Subpersonal Computational States,” Mind and Language 10 (4), 1995, p. 337. 44 Here is how Ch. Peacocke clarifies the notion of content in one of his later writings: “I use the phrase ‘the content of experience’ to cover not only which objects, properties, and relations are perceived, but also the ways in which they are perceived. The ways […] all contribute to the representational content of experience.” Ch. Peacocke, “Does Perception Have a Nonconceptual Content?” The Journal of Philosophy 98 (5), 2001, p. 241. However, the notion of the way in which one perceives something can be ambiguous: on the one hand, it points to the fact that when we perceive an object, we tend to perceive it as this or that or as such and such; on the other hand, it suggests that there is an irreducibly subjective component in the content of perceptual experience, a point of view which one adopts when entering into a perceptual relation with objects.



Nonconceptual Content (NCC) 

 19

ways. Of course, it does not have to mean that one would not be able to perceive any things then. What counts as perception – seeing, hearing, touching, smelling, tasting – depends on the understanding of perception that one may endorse.45 Thus, for someone who understands perception as a way of conveying information of a certain sort, what matters most is the function of perception; what matters less is whether perception is accompanied by consciousness or not. But, on such a construal, only a linguistic convention prevents one from attributing perceptual states to things like pressure gauges or thermometers.46 How can the way one perceives an object be further characterized? It seems quite evident that it is relational: an object appears to the subject in a certain way and the subject apprehends the object in a certain way. On what conditions is this possible? A provisional answer would amount to this: the object must be such that it is possible for the subject to apprehend it, and the subject must be equipped with the capacities that make the apprehension possible for him. If both the subjective and the objective conditions are satisfied, we can say that there is a certain match between the way in which the object appears to the subject and the way in which the subject apprehends the object. Whether there can be such a match is itself a transcendental question, that is, a question the answer to which would require the kind of considerations one can encounter in Kant’s Critique of Pure Reason, the writings of the German idealists (Fichte, Schelling, Hegel) or the phenomenological school (Husserl, Heidegger, or Sartre). But the question about the subjective conditions belongs within the area of epistemology. One way of addressing this question would be to say that the subject must be equipped with conceptual capacities or possess appropriate concepts. For example, what makes it possible for one to perceive a rose as purple and thorny are the concepts of these perceptible qualities of the rose which one possesses. But many philosophers have argued that concept-possession is neither a necessary, nor a sufficient condition of perceiving an object in a certain way, that is, of entertaining content-laden

45 Thus, whereas McDowell defends the view that all perception is factive, Fred Dretske distinguishes perception of facts (“perception-that”) from “mere” perception of objects (“simple seeing”), the former having an epistemic and the latter non-epistemic purport. McDowell contends that “in a particular experience in which one is not misled, what one takes in is that things are thus and so. That things are thus and so is the content of the experience...” Mind and World, op. cit., p. 26. For an opposite view see: F. Dretske, “Meaningful Perception” in: D. N. Osherson, S. M. Kosslyn et al. (eds.), Visual Cognition. An Invitation to Cognitive Science, vol. 2, The MIT Press, Cambridge MA 1995; “Simple Seeing” in: F. Dretske, Perception, Knowledge and Belief. Selected Essays, Cambridge University Press, Cambridge 2000; “What We See. The Texture of Conscious Experience” in: B. Nanay (ed.), Perceiving the World, Oxford University Press, Oxford – New York 2010. 46 An analogy with artificial technical devices can be found in: F. Dretske, Knowledge and the Flow of Information, The MIT Press, Bradford Books, Cambridge MA 1981; Naturalizing the Mind, The MIT Press, Cambridge MA 1995.

20 

 The Contents of Perceptual Experience: Opposing Views

perceptual states. This is because the content of perception can be nonconceptual. As Gareth Evans puts it: “In general, we may regard a perceptual experience as an informational state of the subject: it has a certain content – the world is represented a certain way – and hence it permits of a non-derivative classification as true or false. […] The informational states, which a subject acquires through perception, are non-conceptual, or non-conceptualized. Judgments based upon such states necessarily involve conceptualization: in moving from a perceptual experience to a judgment about the world (usually expressible in some verbal form), one will be exercising basic conceptual skills […]. The process of conceptualization or judgment takes the subject from his being in one kind of informational state (with a content of a certain kind, namely, nonconceptual content) to his being in another kind of cognitive state (with a content of a different kind, namely, conceptual content).”47

Evans bases his claim about the two kinds of content on a general observation that informational states are belief-independent:48 one can be in a non-doxastic informational state which, as one of its functions, can give rise to an informational state with a belief involved. Although Evans’ construal of the notion of information is far from clear (and he takes it as a primitive notion for philosophy),49 his idea seems to boil down to this: it is not necessary for an entity to be an information-carrier that it be equipped with any concepts which would be employed in rendering the information. York Gunther has listed a number of criteria defining conceptual content (CC).50 Should the content of perceptual experience satisfy these criteria, perception can be attributed CC; otherwise, the content of perception will be considered nonconceptual. The criteria include: 1. Compositionality: CC is functionally determined by its constituents; in other words, it is contingent upon the compositional variations among its constituents.

47 G. Evans, The Varieties of Reference, edited by J. McDowell, Oxford University Press, Oxford 1982, pp. 226-7. Evans’s claims regarding perception are subsidiary to his key considerations of the problem of demonstrative reference. 48 “[…] the subject’s being in an informational state is independent of whether or not he believes that the state is veridical.” Ibidem, p. 123. 49 Ibidem. “Information” derives from the verb “to inform” which literally means: “to put a form into something,” “to endow something with a form.” The notion of form is far from being “a primitive notion for philosophy.” Rather, it has motivated some major debates in the Western metaphysics, such as the debate about the existence of universals, or about the relation between the general and the particular. Also, it has figured in the chief dichotomies of metaphysics and the theory of cognition, such as one of form and matter in Aristotle or Kant. 50 Y. Gunther, “General Introduction” in: Y. Gunther (ed.), Essays on Nonconceptual Content, The MIT Press, Cambridge MA 2003, pp. 1-20.



Nonconceptual Content (NCC) 

 21

2. Cognitive Significance: (i) weak: a given state m has a conceptual content c if a subject S undergoing m state believes, or is able to believe that c (in normal epistemic conditions); (ii) strong: CC should agree with the principle of contradiction; a subject S cannot undergo a particular state m with CC of the form F and non-F. Cognitive Significance in its strong version might otherwise be called a criterion of rational constraint, for it requires that perceptual content be liable to the same kind of limitations that are imposed on all meaningful thought and speech. 3. Reference Determinacy: CC has a determinate semantic value understood along the lines of the Fregean doctrine that the sense of an expression (Sinn) determines its reference (Bedeutung). Metaphorically speaking, concepts provide “tools” by means of which rational subjects “capture” mind-independent realities. 4. Force Independence: CC can be individuated independently of the “force” (such as assertion, doubt or desire) of the state or act in which it figures. A number of examples illustrate the claim that perceptual content does not always satisfy the above criteria. Think about floor patterns in which the same tiles can be seen either as square or as diamond-shaped.51 The elements that build up what you can see, their properties (such as being interchangeably black and white) and relations between them (the length and proportion of sides) stay the same, yet what you can see depends on the way in which you see it. Further, think about your listening to a piece of music or looking at a meadow full of different kinds of flowers: could any of the beliefs you frame on the basis of your perception adequately render what you have heard or seen?52 The so-called waterfall illusion undermines the requirement that the content of perceptual experience conform to the principle of contradiction:53 look at a waterfall for about 30 seconds, then look at a stationary object like a stone – perceived relative to the waterfall the stone seems to stand still, whereas perceived relative to its stationary background the stone seems to move. Moreover, there are cases in which concepts do not suffice to determine or, conversely, in which concepts “over-determine” the reference of an empirical content:54 usually, demonstrative identification requires both a conceptual and a

51 The example can be found in: M. Tye, “Nonconceptual Content, Richness, and Fineness of Grain” in: T. Gendler, J. Hawthorne (eds.), Perceptual Experience, Oxford University Press, Oxford – New York 2005. 52 This, in brief, is the so-called richness argument which Evans encapsulated in the following question: “Do we really understand the proposal that we have as many colour concepts as there are shades of colour that we can sensibly discriminate?” G. Evans, The Varieties of Reference, op. cit., p. 229. 53 For the relevant argument, see: T. Crane, “The Waterfall Illusion,” Analysis 48 (3), 1988. 54 Ch. Peacocke, “Does Perception Have a Nonconceptual Content?” op. cit.; “Nonconceptual Content Defended,” Philosophy and Phenomenological Research 58 (2), 1998; Sean D. Kelly, “Demonstrative Concepts and Experience,” The Philosophical Review 110 (3), 2001.

22 

 The Contents of Perceptual Experience: Opposing Views

nonconceptual component in the sense that one can determine the reference of an expression such as “this chair” only in the presence of the object which becomes available as soon as one comes into a direct perceptual contact with it. Finally, on some accounts, certain kinds of content, such as emotional content, cannot be individuated independently of the force of the act in which they figure.55 Thus, one’s grief about a person’s failure differs from one’s contentment with the fact. Similarly, one’s seeing the sky’s blue differs from one’s seeing a carpet’s blue; for perception, it can be argued, is context- and object-dependent, and so its content varies according to the change in the context and the object.56 The above examples appeal to the phenomenology of experience. But philosophers have raised a number of points about the nonconceptual content of perceptual experience, which emphasize its explanatory role in the account of empirical cognition. By way of an example: Ch. Peacocke claims that NCC rationalizes our perception-based beliefs or judgments; that it makes CC available to the subject of experience; it accounts for the acquisition of empirical and, in particular, observational concepts; it explains why both concept-possessing creatures and those that do not employ concepts, such as human babies or non-rational animals, can entertain content-laden perceptual states; and it explains intentional action.57 Combined together, the phenomenological and the epistemological arguments have a potential to make a strong case for NCC.

1.5 Varieties of Content 2 Most of the examples provided above aim at illustrating the idea that it is not necessary that subjects possess appropriate concepts in order to be able to undergo content-laden perceptual states. Thus, the examples promote a view called state nonconceptualism – a view that refers to the concept-possession condition imposed on the subjects of perceptual experience.58

55 The point is advocated in: Y. Gunther, “The Phenomenology and Intentionality of Emotion,” Philosophical Studies 117, 2004. 56 The argument is discussed in: S. D. Kelly, “The Non-Conceptual Content of Perceptual Experience: Situation Dependence and Fineness of Grain,” Philosophy and Phenomenological Research 62 (3), 2001. The phenomenon of perceptual constancy illustrates what context-dependence comes down to: namely, it occurs when one perceives a colour, or a shade of colour as the same in different lighting conditions; or when one perceives an object as of the same size despite one’s varying distance from the object. Object-dependence, in turn, occurs when one perceives a certain property of an object x, such as x’s height or colour, in a different way than one in which one perceives the same property of an object y. 57 Ch. Peacocke, “Does Perception Have a Nonconceptual Content?” op. cit., p. 242. 58 In what follows, I distinguish state and content conceptualism and nonconceptualism, respectively, after: T. Crowther, “Two Conceptions of Conceptualism and Nonconceptualism,” Erkenntnis 65 (2), 2006.



Varieties of Content 2 

 23

According to state nonconceptualism, the condition of concept-possession does not have to be satisfied by perceivers, even if they otherwise happen to be capable of acquiring and exercising concepts. While it is evident that I am able to acquire and employ (for example) the concept of a birch tree, I do not have to possess this concept to be able to perceive birch trees. In other words, perception does not involve meeting the epistemic requirements that concept-possession does. If someone maintains the opposite of this view, he may be considered to represent state conceptualism. According to state conceptualism, perception necessarily involves the operations of conceptual capacities (to use McDowell’s term),59 hence it requires that its subject satisfy the concept-possession condition. On this account, my seeing a birch tree presupposes, or is contingent upon, my possession of the concept “birch tree.” Obviously, whether one’s views can be identified as state conceptualism or state nonconceptualism depends on the understanding of concepts in play. It can be argued that concept-possession consists in the ability to discriminate and re-identify particular items that fall under a certain concept but it can also be argued that the exercise of discriminatory capacities alone suffices to establish that a subject satisfies the condition of concept-possession.60 However, state conceptualism and nonconceptualism should be distinguished from content conceptualism and nonconceptualism, respectively. For whereas the former two pertain to the exercise of the subject’s capacities, the latter deal with the determinations of the ways in which experience presents its objects to the subject, that is, the content. It is one thing to think that seeing a birch tree requires that one possess the concept “birch,” but quite another to think that one’s perception of a birch tree implicates conceptual content; that what one sees when one sees a birch tree is conceptually structured. McDowell expresses the idea of content conceptualism in that he writes: “In a particular experience in which one is not misled, what one takes in is that things are thus and so. That things are thus and so is the content of the experience, and it can also be the content of a judgment: it becomes the content of a judgment if the subject decides to take the experience

59 In Mind and World McDowell does not attempt to clarify the notion of conceptual capacities. Elsewhere, though, he casts some light on its meaning in that he writes: “I do not see conceptual capacities as dispositions automatically triggered into actualization by appropriate stimuli. I see conceptual capacities as capacities of freedom, capacities whose paradigmatic actualizations are under the control by their possessors. That is what is conveyed by saying that those paradigmatic actualizations occur in acts of judgment.” J. McDowell, “Responses” in: M. Willaschek (ed.), John McDowell: Reason and Nature: Lecture and Colloquium in Münster 1999, LIT Verlag, Münster 2000, p. 96. 60 This remark is important to the extent that it helps one realize that as long as one does not have a determinate conception of what concepts are, most of the arguments for conceptualism or nonconceptualism may eventually turn out to be inconclusive. Cf. P. Chuard, “The Riches of Experience,” Journal of Consciousness Studies 14 (9-10), 2007.

24 

 The Contents of Perceptual Experience: Opposing Views

at face value. So it is conceptual content. But that things are thus and so is also, if one is not misled, an aspect of the layout of the world: it is how things are.”61

According to McDowell, perception is factive: what we “take in” when we undergo perceptual experiences are facts, and the structure of facts, in a way, mirrors the structure of propositions (and vice versa); in other words, propositions capture facts. Content nonconceptualists, in turn, hold that what we take in when we undergo perceptual experiences does not have a structure analogous to the structure of propositions. Content nonconceptualism about perceptual experience is a view that perception has its own specific kind of content, different from the content of beliefs, judgments, and thoughts. Thus, when I see a birch, there must be something in my perception of it that cannot be grasped by means of the concept “birch” which I possess and can correctly use. What is the relation between the two kinds of conceptualism and nonconceptualism? I claim that they are logically independent. Thus, one can coherently endorse both state conceptualism and content nonconceptualism, on the one hand, and state nonconceptualism and content conceptualism, on the other. Even if my conceptual capacities are involved in my perception of a birch, it does not follow that what I perceive is conceptually structured throughout. For in my perception two kinds of content may be involved: the content of my belief and the nonconceptual content on which the former can be taken as supervening.62 Content conceptualists reject this idea, though: on their tenets, featuring the content of perceptual beliefs as dependent on nonconceptual content results in providing an inadequate picture of the relation between perception and belief. Briefly, on such a picture, the relation between perception and belief could not be rational, and this is what some theorists, like McDowell, find objectionable. (I will return to McDowell’s argument for conceptualism from perceptual reasons later in this chapter.) Also, the converse of the above situation is possible. Kant’s doctrine of syntheses (to be discussed in chapter 4) can be taken to illustrate the point. Concepts may structure the content of what one perceives, hence perception can be rule-governed, without the subject being able to consciously employ the concepts in experience. Thus, I may perceive something as a birch tree without being able to say what birch trees are like.

61 J. McDowell, Mind and World, op. cit., p. 26 (italics mine). 62 The idea that CC supervenes on NCC can be found in: R. Hanna, “Beyond the Myth of the Myth: A Kantian Theory of Non-Conceptual Content,” International Journal of Philosophical Studies 19 (3), 2011.

Phenomenology 

 25

1.6 Phenomenology Whereas most arguments for the existence and the cognitive significance of NCC appeal to the phenomenology of perceptual experience, the most relevant arguments for conceptualism are framed against the epistemological background. Let us deal first with the phenomenological arguments. They include the following: (1) richness and fineness of grain; (2) perceptual illusions; (3) memory (both perceptual and iconic); (4) demonstrative identification and reference.

1.6.1 Richness and Fineness of Grain63 According to nonconceptualists, the ability to perceptually discriminate objects, properties, and relations outstrips the ability to do so by means of concepts alone. This is because there is more on the side of objects, properties, and relations than can be conceptually grasped. As Evans rhetorically asks, “Do we really understand the proposal that we have as many colour concepts thus there are shades of colour that we can sensibly discriminate?”64 Similarly, Richard Heck emphasizes the lack of conceptual resources available to capture the fineness of detail, or the determinacy, of perceptual experience: “Consider your current perceptual state – and now imagine what a complete description of the way the world appears to you at this moment might be like. Surely a thousand words would hardly begin to do the job […] my experience of these things represents them far more precisely than that, far more distinctively, it would seem, than any characterization I could hope to formulate, for myself or for others, in terms of the concepts I presently possess. The problem is not lack of time, but lack of descriptive resources, that is, lack of the appropriate concepts.”65

The richness argument builds on the assumption that if the content of perceptual experience is conceptual, the subject who undergoes content-laden perceptual states must be equipped with the concepts indispensable to articulate that content. But,

63 Philippe Chuard notices that richness and fineness of grain are two different properties of representational content. An experience is rich if it represents a variety of detail, but it is fine-grained if it conveys specific information about the objects, properties etc. represented. An experience can be scarce in detail, yet fine-grained, e.g. when you see a red dot against a uniform white background (you see the dot as shaded in a specific way); and an experience can be coarser in grain but convey rich information, as is the case e.g. when you walk a town street in thick fog. P. Chuard, “The Riches of Experience,” op. cit. 64 G. Evans, The Varieties of Reference, op. cit., p. 229. 65 R. G. Heck, “Nonconceptual Content and the ‘Space of Reasons,’” The Philosophical Review 109 (4), 2000, pp. 489-490.

26 

 The Contents of Perceptual Experience: Opposing Views

according to the authors quoted above, it is not the case that the subject undergoing content-laden perceptual states would always be equipped with the right kind of concepts. In such cases, they claim, the content of perception must be nonconceptual. Surely, there must be a way in which to show that the subject does undergo contentladen perceptual states despite his lack of proper conceptual resources, because the conceptualist could rejoin that, for the lack of proper conceptual resources, the subject’s experiential state lacks any content; in other words, it does not represent anything. Now, the nonconceptualist may reply that, in those frequent cases in which the variety of the detail perceived far outstrips our capacities of conceptualization, we can perceptually discriminate the detail. For instance, when presented with two samples of minimally different shades of blue, we can tell them apart. But this does not mean that we possess the concepts of these shades since we are unable, for example, to re-identify the instances of these shades later (let us suppose that the shades differ so slightly that we may re-identify them later as one and the same shade). Again, this might fall short of a satisfactory answer to the conceptualist. After all, perhaps in such cases we do make use of demonstrative concepts and “recognitional capacities” which are quite “short-lived,”66 though retainable in memory. But even if we do not make use of such demonstrative concepts, the very ability to discriminate objects, properties, and relations in a given perceptual experience might be taken as testifying to the fact that we possess appropriate concepts. For what conditions does the subject have to meet to possess a concept? And cannot a concept be possessed only partially or in degrees? Thus, the argument from the richness of perceptual detail might turn out, in the end, to be inconclusive: its success or failure would ultimately depend on the account of concept-possession one endorses.

1.6.2 Perceptual Illusions Recall the waterfall-illusion example. Think about the cases in which you perceive one and the same object, such as the Necker cube or Wittgenstein’s rabbit-duck, or a set of objects, such as black and white tiles on the floor, in two different ways. Think about “impossible objects,” such as Escher’s staircase or the Penrose triangle. Why would these examples play against conceptualism? The argument might go as follows: If the content is conceptual, then it conforms to certain criteria (one may call them rationality constraints), such as the principle of contradiction or the principle of identity. But there are cases in which perceptual content does not conform to the criteria at issue. Therefore, in such cases, the content of perceptual experience is nonconceptual.

66 Both expressions come from the passage in which McDowell presents the so-called “that-shade argument.” See: J. McDowell, Mind and World, op. cit., pp. 57-58.

Phenomenology 

 27

However, the conceptualist may reply that what these examples show is only that perception presents us with aspects of objects: when viewed from a certain perspective, the floor looks to be covered with square-shaped tiles; viewed from a different perspective, it looks to be covered with diamond-shaped ones. One perceives things under different aspects depending on the perspective one adopts. But this answer is somewhat unsatisfactory: the floor is made up of tiles (and grout fillings between them), the tiles are of a certain shape; no matter from which perspective we look at them, they have one and the same shape, one and the same set of geometrical properties. Besides, granting that we perceive things under different aspects, which depends on the point of view from which we view them, what are these aspects to be attributed to? An object cannot be at the same time identified as, say, a rabbit and a duck, or as moving and as standing still. Different, mutually exclusive determinations cannot be predicated of one and the same thing. But aspects are not independent entities, and we tend to think of them as aspects of something. Thus, recourse to aspects does not seem to satisfactorily explain why things can be seen in mutually incompatible, or even contradictory, ways. However, it may also be observed that from the fact that the content of perceptual experience does not conform to rationality constraints, it does not have to follow that it is nonconceptual. This is because, first, conformity to rationality constraints is a normative requirement which sometimes would not be met; and, second, analogously to perceptual illusions, or illusions of the senses, one can think of rational illusions, or illusions of the intellect (reason). The paralogisms of pure reason Kant discovers, in the arguments of rational psychology, provide a good example of the latter kind of illusions.67

1.6.3 Memory Arguments that appeal to memory, both perceptual and iconic, might be regarded as a version of the richness argument. For what they show is that the fineness of detail of our perception-based representations considerably exceeds our capacities of conceptualization. Memory stores these representations, which becomes evident when one reflects on some common occurrences or psychological experiments. Thus, Michael Martin describes a case in which a person recollects having left his cufflinks in

67 More specifically, Kant calls them “dialectical illusions of pure reason” and distinguishes them from empirical illusions “occurring along with determinate empirical cognitions.” I. Kant, Critique of Pure Reason, op. cit., p. 439 (A 396-7). He also discusses “transcendental illusions,” which are “natural and unavoidable” and which “rest on subjective principles and pass them off as objective.” Ibidem, p. 386 (A 297-8/B 354).

28 

 The Contents of Perceptual Experience: Opposing Views

a drawer, where he was previously looking into without noticing them at that time.68 Michael Tye, in turn, draws on Sperling’s experiment with iconic memory in order to reinforce the richness argument.69 In the experiment, an arrangement of letters is displayed to subjects for 50 milliseconds, after which they report a certain number of letters they have seen. But when cued with a tone shortly after the elapse of the 50millisecond time, subjects report all the letters from the row their attention has been drawn to. Therefore, it might be concluded that, as Fred Dretske puts it: “There is more information in the sensory store than can be extracted, a limit on how much of this information can be exploited by the cognitive mechanisms […]. Our own perceptual experience testifies to the fact that there is more information getting in than we can manage to get out.”70

But against the argument from memory, as much as against the richness argument, one can raise the charge that it builds on an “illusion of richness.” Thus, experiments on change blindness and inattentional blindness have shown that in perception we are aware of significantly fewer details than we tend to think we are. Consider a wellknown example: a group of people, while watching a game of basketball, are asked to count how many times players pass the ball to one another. Interestingly, hardly anyone performing the task notices the appearance of a man in a gorilla suit dancing a jig on the playground.71 A similar effect occurs when observers are not instructed about what they should keep track of when viewing a scene: then, if changes, especially background ones, come gradually, subjects usually fail to spot them. However, while it might be true that how much and what we perceive depends on how much and what we attend to,72 it is by no means evident that how much and what we attend to depends on what concepts we have. The links between perception and attention might be more primitive than the links between attention and conceptual capacities, just as the links between perception and action: if an animal does not attend to what is currently going on in its habitat, it may pay the highest price for its inattentiveness; but if it does not form beliefs about its surroundings, not much of the bad thing will happen. But, again, one can reply that attention presupposes discrimination (for I am attentive to x-s rather than y-s) and discrimination requires generalization, even if very basic. This means that while attending to something I should entertain a concept

68 M. G. F. Martin, “Perception, Concepts, and Memory,” The Philosophical Review 101 (4), 1992. 69 M. Tye, “Nonconceptual Content, Richness, and Fineness of Grain,” op. cit. 70 F. Dretske, “Sensation and Perception” in: Y. H. Gunther (ed.), Essays on Nonconceptual Content, The MIT Press, Cambridge MA, p. 35. 71 The example has been taken from: M. Tye, “Nonconceptual Content, Richness, and Fineness of Grain,” op. cit. 72 Cf. e.g. R. A. Rensink, J. K. O’Regan, J. J. Clark, “To See or Not To See: The Need for Attention to Perceive Changes in Scenes,” Psychological Science 8 (5), 1997.

Phenomenology 

 29

of it, even if that concept cannot be articulated in a language. But whether there can be such concepts that cannot be articulated in a language at all remains quite mysterious and takes us back to the question of how concepts are understood in the debate between conceptualists and nonconceptualists.

1.6.4 Demonstrative Identification and Reference When referring to objects directly perceived, we use phrases such as, e.g., “this chair” or “that car parked in front of Collegium Broscianum.” If the content of perceptual experience is conceptual, then our demonstrative concepts articulated by means of such phrases adequately pick it out. But do demonstrative concepts “cut” sufficiently finely to individuate this kind of content? Some philosophers have suggested that perceptual content is too rich and our concepts, including the demonstrative ones, are too coarse-grained for that; others have insisted that demonstrative concepts “slice too finely,”73 hence there is much less on the side of experience than our conceptual capacities can accommodate. McDowell has argued that demonstrative concepts render the details of perceptual experience in the right way; for, in the face of the richness of perceptual content, one can resort to a demonstrative concept that will capture the content adequately. Formulating the “that-shade” argument, McDowell contends: “When Evans suggests that our repertoire of colour concepts is coarser in grain than our abilities to discriminate shades, and therefore unable to capture the fine detail of colour experience, what he has in mind is the sort of conceptual capacities that are associated with colour expressions like ‘red,’ ‘green,’ or ‘burnt sienna’ […]. But why should we accept that a person’s ability to embrace colour within her conceptual thinking is restricted to concepts expressible by words like ‘red’ or ‘green’ and phrases like ‘burnt sienna’? It is possible to acquire the concept of a shade of colour, and most of us have done so. Why not say that one is thereby equipped to embrace shades of colour within one’s conceptual thinking with the very same determinateness with which they are presented in one’s visual experience, so that one’s concepts can capture colours no less sharply than one’s experience presents them? In the throes of an experience of the kind that putatively transcends one’s conceptual powers – an experience that ex hypothesi affords a suitable sample – one can give linguistic expression to a concept that is exactly as fine-grained as the experience, by uttering a phrase like ‘that shade,’ in which the demonstrative exploits the presence of the sample.”74

McDowell claims that phrases like “that shade” capture conceptual content in a given context of perceptual experience. But while it is obvious that “shade” expresses a concept, it is less so when it comes to “that” (or other demonstrative pronouns). In

73 Ch. Peacocke, “Nonconceptual Content Defended,” op. cit., p. 382. 74  J. McDowell, Mind and World, op. cit., pp. 56-57.

30 

 The Contents of Perceptual Experience: Opposing Views

the first chapter of part one of the Phenomenology of Spirit, Hegel famously argues that indexical expressions like “this,” “here,” and “now,” by means of which we refer to objects in perception, are general. On the Hegelian view, also indexicals mediate conceptual content.75 The claim may be considered contentious, though. “This,” “here,” and “now” can refer to many objects, indeed to an unrestricted scope of them. Nevertheless, they do not refer to objects by virtue of a mark, a feature, or an aspect which can be common to a number of things. Thus, what is general about indexical expressions is no more than the mode in which we use them, rather than their content. Suppose, however, that indexicals do refer to objects by virtue of a mark, a feature, or an aspect which can be common to a number of things and which a speaker intends while employing a demonstrative. But then, in the context of perceptual experience, we would have to be able to specify what exactly an indexical captures from the rich field of experience. Pointing at the colour of leaves of a tree growing by the river, one would have to be able to specify whether “this” identifies the particular colour, or the shade of the colour, or the shade of the colour as perceived in daylight, or as perceived on the background of the shiny surface of water, etc. – the list could, in principle, expand to infinity. Again, there seems to be no determinate aspect that a demonstrative pronoun would be meant to capture. I have pointed out that demonstrative concepts might be too “coarse” to capture the content of perceptual experience. But, according to Christopher Peacocke, these concepts are simply too “fine” to do the job.76 This is because two persons can apply different demonstrative concepts, for example “that shade,” “that red,” or “that scarlet,” while perceiving an object in the same way. Peacocke’s example resembles, to an extent, one of the examples of Kant from the Jäsche Logic, where he says that when a “savage” sees a house, he sees the same object as someone who has the concept “house,” although the former does not have this concept.77 But both Peacocke’s and Kant’s claims seem to rest on the assumption that things can be experienced as being a certain way despite the fact that subjects lack the right kind of concepts by means of which to grasp these ways. As it stands, the assumption may be ungrounded, though. For someone who employs the demonstrative concept “that red” while looking at the colour of a rose’s petals, may simply see the rose much differently from someone who employs the demonstrative concept “that scarlet” to articulate his experience of the same property. Likewise, a wine expert may discover a slightly different taste in the wine that both he and a layman are drinking; and a person who enjoys good eyesight will normally perceive green differently than one suffering from colour-blindness.

75 Cf. G. W. F. Hegel, Phenomenology of Spirit, translated by A. V. Miller, Oxford University Press, Oxford 1977. 76 Ch. Peacocke, “Nonconceptual Content Defended,” op. cit. 77 Cf. I. Kant, Lectures on Logic, translated and edited by J. M. Young, Cambridge University Press, New York 1992, pp. 544-5 (9:33).



Perceptual Content and Empirical Beliefs 

 31

1.7 Perceptual Content and Empirical Beliefs That perception gives rise to beliefs and provides a (or the) source of empirical cognition may pass for no more than a commonsense platitude. Yet, to establish what relation there can be between perception and empirical beliefs is a task for a philosopher. One can distinguish two kinds of relations at issue, one of which is epistemic and the other non-epistemic, i.e. justification and causation. The relation between perception and empirical beliefs is causal in that the former causes the latter, as in Locke’s theory of perceptual cognition in which objects perceived causally account for there being ideas in the mind. If the relation is one of justification, though, then perception logically accounts for a belief. However, both causal and justificatory relations between perceptual experiences and empirical beliefs seem quite problematic.78 In the first case, it is difficult to understand how a non-propositional occurrence or state can be related to the propositional content of a particular belief, as much as it is difficult to understand, how, in Descartes’ theory of the mind, the extended matter and the thinking mind can interact. In the second case, we have to assume that perceptual experiences have the same kind of content as empirical beliefs, i.e. propositional content. But how can perception entertain propositional content, yet at the same time differ from other mental states, acts, and processes, such as thought? Thus, a dilemma emerges: if perception provides causes for empirical beliefs, the latter cannot be (properly) justified, but if perception provides reasons for beliefs, this leads to an infinite regress. For reasons furnished by perceptual experiences must themselves be properly justified, unless we acknowledge that they rest on nonpropositional occurrences or states. In Mind and World, McDowell identifies the poles of the dilemma as coherentism and the “Myth of the Given,” and observes that both sever rational connections between perception and reality. Since, according to the coherentists, represented by Donald Davidson, beliefs can enter justificatory relations with other beliefs alone, perception cannot have any rational bearings on empirical beliefs. Since for the adherents of the “Myth” perception only causally accounts for beliefs, there cannot be rational relations between the former and the latter. In order to escape this “oscillation,” McDowell proposes a conceptualist account of

78 The problem can also be described as pertaining to the foundations of empirical knowledge. According to Laurence BonJour, there are two ways of addressing it, both of them unsatisfactory: either knowledge claims rest on something that defies further justification, although it belongs to the body of knowledge, or knowledge claims rest on other knowledge claims. In the first case, we are faced with incoherence, in the second – with infinite regress, and we must reject the very idea of the foundations of knowledge. Thus, either alternative leads to skepticism about the possibility of justifying empirical beliefs. See: L. BonJour, “Can Empirical Knowledge Have a Foundation?” American Philosophical Quarterly 15 (1), 1978.

32 

 The Contents of Perceptual Experience: Opposing Views

perceptual experience, a view on which perception consists in the endorsement of facts rather than mere “acquaintance” with objects.79 However, it may be argued that such a view on perception is epistemically too demanding: after all, there are creatures, like human babies and non-human animals, to whom we would deny the ability to “endorse facts” by means of structured propositions but not the ability to perceive.80 But this objection, a conceptualist may reply, can be easily fended off. For we can deny that the entities mentioned perceive facts; therefore, they cannot be in possession of reasons for perceptual beliefs. This does not imply that they do not perceive objects. But this is exactly what they have pre-cognitive intentionality. Surely, this pre-cognitive intentionality is too little to secure the epistemic significance of sensibility. Now, the conceptualist and the nonconceptualist alike can distinguish two senses of “perception:” epistemic and non-epistemic, and grant that both senses are valid. Further, the conceptualist may add that transition from the non-epistemic into the epistemic mode of perception is merely a matter of acquisition or development of certain capacities, which pre- or non-linguistic creatures do not or cannot have. Thus, an argument that appeals to such creatures must remain inconclusive as a case against conceptualism. I think, however, that a case against conceptualism can be made in that we reject the argument from justification which, in short, goes like this: (1) experience provides reasons for (empirical) beliefs; (2) reasons are thinkable items, and so they are conceptual; (3) hence, the content of experience is conceptual as well. Now, what can be done is this: it can be shown that (1), as it stands, is incoherent. First, it builds

79 According to McDowell, the “Myth of the Given” motivates the embrace of coherentism but the latter ends up again with the Myth. In both cases, the idea is that experience remains in a causal but not a rational relation with our perceptual beliefs. As a result, we are “exculpated” from but not justified in entertaining such beliefs. As he puts it: “But it is one thing to be exempt from blame, on the ground that the position we find ourselves in can be traced ultimately to brute force; it is quite another thing to have a justification. In effect, the idea of the Given offers exculpations where we wanted justifications.” J. McDowell, Mind and World, op. cit., p. 8. 80 It is worth to note that research in the area of cognitive and developmental psychology shows that pre-linguistic human babies, as well as some non-human animals, perceive objects rather than a “blooming buzzing confusion of sensations” in disarray, in particular due to their ability to exercise motoric skills, which are more primitive than the ability to employ concepts. For in order to perceptually track an object in motion, babies must be able not only to “parse” their perceptual field and “synthesize” their perceptions into representations of “quasi-objects,” but they must also be able to distinguish, albeit imperfectly, shapes, magnitudes and distances of objects. See: E. S. Spelke, G. Gutheil, G. Van de Walle, “The Development of Object Perception” in: S. M. Kosslyn, D. N. Osherson (eds.), Visual Cognition. An Invitation to Cognitive Science, vol. 2, The MIT Press, Cambridge MA 1995. The authors maintain that spatial organization of one’s perceptual field precedes quality recognition.



Perceptual Content and Empirical Beliefs 

 33

on an unclear notion of reason. Second, it fails to meet the challenge that might be posed by a skeptic. I will focus on the latter objection. For McDowell, reasons delivered by perception come down to what he calls “appearings.”81,82 Since the contents of appearings and judgments are of the same kind, McDowell contends that the former maintain a rational relation to the latter. But whereas judgments refer to the objects of cognition, appearings should rather be posited as merely subjective ways of perceiving objects. To illustrate the point: on the basis of it appearing to me, say, that the tower over there is round, I do not have a reason good enough to believe that the tower is round. On the contrary, if, as Descartes would say, my idea is not sufficiently clear and distinct (that is, if I have certain sufficiently strong reasons to doubt in the deliverances of my sensibility – and, after all, why should I not have them, considering the fact that my senses have deceived me so many times), then judging is the last thing that I, as a rational subject, should do. For Descartes, judging involves transition from subjective ideas (acts or contents of the mind) into the realm of the objective (the causes of the ideas); thus, it might be said that, metaphorically speaking, it involves bridging the gap between mind and the extra-mental reality. But, for Descartes, only judgments that are made on the basis of clear and distinct or intuitively certain ideas are sufficiently warranted. In all other cases by making a judgment one is permanently prone to err. However, Descartes makes a strong metaphysical claim in order to guarantee the reliability of intuition in that he introduces a non-deceiving God as the source of all clear and distinct ideas. A perfect match between cognition and reality is possible, but only upon the condition that God exists (and in the event that He does – that He is not a deceiver). Now, to return to McDowell: why should transition from perception to belief be possible? Why should subjective occurrences – “appearings” or phenomena – with which McDowell identifies perceptual reasons, suffice to justify beliefs about mindindependent reality? Whereas Descartes introduces God as part of his strategy of defusing skepticism concerning the source of knowledge (or ideas, to use a Cartesian term), McDowell seems to avoid the skeptical challenge by suggesting that it is spurious. “[M]y move is not well cast as an answer to skeptical challenges – he admits – it is more like a justification of a refusal to bother with them.”83

81 Until the end of this section I draw extensively on a part of my article “McDowell and Perceptual Reasons,” Forum Philosophicum 17 (1), 2012. 82 Those “appearings” – or, as we may call them, phenomena – he characterizes as “just more of the same kind of things [as] beliefs […]: possessors of empirical content, bearing on the empirical world.” J. McDowell, Mind and World, op. cit., p. 142. 83 J. McDowell, “Knowledge and the Internal,” Philosophy and Phenomenological Research 55 (4), 1995, p. 888n.

34 

 The Contents of Perceptual Experience: Opposing Views

And further: “The considerations I have offered suggest a way to respond to skepticism about, for instance, perceptual knowledge; the thing to do is not to answer the skeptic’s challenges, but to diagnose their seeming urgency as deriving from a misguided interiorization of reason.”84

The remedy for skepticism prescribed by McDowell turns out to consist in an externalist account of perceptual content and a disjunctive theory of perception. On this construal, what we take in when we perceive things are appearings of the world out there, independent of our subjective acts of content-acquisition. There are no intermediaries between the acts and the objects of perception (unlike on the “highest common factor” view mentioned at the beginning of this chapter). Accordingly, the fact that perception justifies empirical beliefs ceases to be mysterious: as in Moore’s (famous or notorious) example with two hands, the fact that one sees, say, a tree outside the window provides a reason for one’s belief that the tree is out there. In other words, perception is factive, and therefore it furnishes reasons for empirical beliefs.85 As McDowell puts it: “[K]nowledge is a status that one possesses by virtue of an appropriate standing in the space of reasons when [...] the world does one the favour of being so arranged that what one takes to be so is so.”86

Yet the skeptical adversary might reply that it is not obvious why the world should do one a favour. Perhaps no one, Descartes included, would sanely doubt that, for example, he has two hands. But, as pointed out by Wittgenstein in On Certainty, this kind of commonsensical belief presupposes a framework of beliefs, a certain background which a priori rules out skeptical scenarios. Our fact-endorsing perceptual beliefs would rest on “hinges” which we take in without questioning.87 On the Wittgensteinian account, such background beliefs would have a status similar to the status that would be held by religious or even ethical beliefs. Such beliefs are immune to evidence or justification, but they shape the way in which one perceives facts and matters “in the world.” By way of a conclusion: McDowell neither explains nor inquires into the very ground of the match between mind and world, which he seems to keenly endorse as a background assumption underlying the claim about the factivity of perceptual

84  Ibidem, p. 890. 85 For a discussion of McDowell’s Moorean strategy of “disarming” the skeptic, see: D. Pritchard, “McDowell on Reasons, Externalism and Scepticism,” European Journal of Philosophy 11 (3), 2003. 86 McDowell, “Knowledge and the Internal,” op. cit., p. 881. 87 L. Wittgenstein, On Certainty, edited by G. E. M. Anscombe and G. H. von Wright, Harper & Row, New York 1972, § 341.



Content and Normativity 

 35

reasons. Indeed, on McDowell’s tenets, the very fact that the contents of experience are conceptual ensures that “experiences have objective purport.”88 But why should there be any match between mind and world? McDowell can be said to ignore the question and, consequently, the skeptical problem that it generates, not because of the lack of philosophical refinement but since he thinks that an attempt at addressing it would be hopeless (for the adversary of the skeptic). Instead, he proposes a transcendental argument in which the “match” is posited as a requirement for maintaining that experience is “of objective reality.”89 Thus, the “match” is not something a philosopher is expected to prove; rather, it is a starting-point presupposition for a conception that purports to establish a proper way of thinking about experience as constitutive of our beliefs about the world.

1.8 Content and Normativity Let us assume that the claim that perception provides reasons for empirical beliefs holds true, regardless of the objections raised in the previous section. Even then it can still be argued that from the fact that perception provides reasons for empirical beliefs it does not follow that conceptualism is true. But before substantiating this claim, let me bring to the fore one more issue related to the debate on content and, as I hope will become clear by the end of this section, to the point under consideration. Thus, at issue would be a normative account of content, that is, an account which explains what it is for something, e.g. for a mental state, to have content, by appealing to normative facts. Normative facts, as defined by Mark Greenberg, “include facts about what is correct and incorrect, better and worse, proper and improper, ideal and defective; and about what standards or requirements people are subject to.”90 In the case of perception, what makes it plausible to talk about normativity is the fact that the content of perception can be thought to have correctness or accuracy conditions. Of course, normative facts about perception can be recognized if one endorses the content view (CV) which states that perception involves more than sensory content (SC), namely conceptual content (CC) or representational nonconceptual content (NCC). Also McDowell holds that normativity is involved in the perceptual relation between mind and world. This is because such a relation necessarily entails CC. Thus, in the Introduction to Mind and World he states:

88  J. McDowell, “Intentionality as a Relation,” The Journal of Philosophy 95 (9), 1998, p. 471. 89 J. McDowell, “The Disjunctive Conception of Experience as Material for a Transcendental Argument,” op. cit., p. 230. 90 M. Greenberg, “A New Map of Theories of Mental Content: Constitutive Accounts and Normative Theories,” Philosophical Issues 15, 2005, p. 311.

36 

 The Contents of Perceptual Experience: Opposing Views

“To make sense of the idea of a mental state’s or episode’s being directed towards the world, in the way in which, say, a belief or judgment is, we need to put the state or episode in a normative context. A belief or judgment to the effect that things are thus and so – a belief or judgment whose content (as we say) is that things are thus and so – must be a posture or stance that is correctly or incorrectly adopted according to whether or not things are indeed thus and so […]. This relation between mind and world is normative, then, in this sense: thinking that aims at judgment, or at the fixation of belief, is answerable to the world – to how things are – for whether or not it is correctly executed.”91

If a mental state or episode is “correctly executed” – one may add – it provides a good reason for the belief that one may entertain on the basis of the state or episode in question. But, as Uriah Kriegel points out, normativity does not pertain – “in the first instance” – to intentional properties of mental states or episodes. Rather, it emerges along with ascriptions of intentional properties to the states or episodes at issue.92 In other words, there are no normative facts associated with experiencing objects as such. Normative facts emerge when we reflect upon the experiences of objects we have undergone (or are undergoing). Only upon reflection on a particular experience can we evaluate it as correctly or incorrectly “executed.” McDowell would be ready to accept the point since he claims: “Active empirical thinking takes place under a standing obligation to reflect about the credentials of the putatively rational linkages that govern it. There must be a standing willingness to refashion concepts and conceptions if that is what reflection recommends.”93

Now, what fosters reflection may simply be information derived from other sources: for instance, one’s visual experience of a bent stick immersed in water may be taken as misleading upon one’s reflection on the additional information provided by one’s tactile perception of the stick (or upon comparing the two pieces of information). Or one’s visual experience of a mountain looming on the horizon may be falsified by one’s getting to know from a friend that this is not a real mountain but decorations for a film. But reflection upon one’s experience may well be replaced by another person’s evaluative judgments about one’s cognitive achievements. This is where there is a role to play for the so-called principles of charity.94 As Kriegel remarks, following Davidson,

��J. McDowell, Mind and World, op. cit., pp. xi-xii. 92 U. Kriegel, “Intentionality and Normativity,” Philosophical Issues 20, 2010, p. 187. 93  J. McDowell, Mind and World, op. cit., pp. 12-13. 94 “These include the principles that (by the interpreter’s own lights) persons’ beliefs are mostly true and coherent, that their desires are mostly good and mutually satisfiable (and/or suitably prioritized), and that their beliefs and desires mostly constitute good reasons for action. As Davidson (1970: 97) puts it, in constructing a theory of someone’s behavior, ‘we will try for a theory that finds him consistent, a believer of truths, and a lover of the good (all by our own lights, it goes without saying).’” U. Kriegel, “Intentionality and Normativity,” op. cit., pp. 188-189.



Content and Normativity 

 37

applying these principles to content-ascriptions makes the ascriptions imbued with normativity. To give an example: if subject A says: “There is a mountain over there,” pointing towards scene decorations, subject B, who has heard A’s statement, would ascribe to A a belief that there is a mountain over there and a (mistaken) perception of scene decorations as a mountain. B would charitably assume that A does not aim at deceiving her and that she reports on a mountain-like experience. In this way, B can appreciate the fact that A has a reason for believing what she does, even though B can see that A’s belief is false. And A’s belief is false because, for some reason or other, she has mistakenly categorized or conceptualized her experience. Her belief has a justification, hence it is based on a reason, but the justification is simply wrong. Therefore, employing an ancient distinction we can say that she does not have knowledge (episteme, science) but merely belief (doxa, opinion). Now let us get back to the point raised at the beginning of this section. There I noticed that from the fact that perception supplies reasons for empirical beliefs it does not follow that the content of perception is conceptual through and through or that what one comes to be perceptually related to depends on concepts. If my rendering of the above example is correct, this indeed must be the case. For, in the example, it is assumed that A and B see the same object but A, perhaps for the lack of some substantial information, misinterprets what she can see. As I take it, it does not make much sense to claim that A’s very experience is misleading: if you were A and undergoing A’s experience, you would certainly not question the credibility of the fact that you are having a particular experience. To use Descartes’ terms: one can err as to what the ideas one has refer to, but not as to the fact that one has these ideas. Kant, in the Anthropology, also seems to embrace a similar view in that he claims that it is not the senses that deceive us but the understanding that fails to make correct judgments on the basis of sensory impressions.95 Of course, one is in a way obliged to make correct judgments upon what one perceives, and here is where concerns about normativity arise. For, as Christopher Peacocke says, “the claim that [an] object really is the way it is experienced as being is one which has a correctness condition.”96 But in perception to get things right means to get the right interpretation of what one perceives, to categorize or conceptualize what one perceives in the best possible way (who sets the standards of correct perception and what these standards are is a separate question which we do not need to be troubled about here). Such an account of perception involving normative facts is compatible with both the CC- and the NCC-thesis, since it makes the obtaining of normative facts contingent upon conceptualization while showing that NCC must be shared by both correctly and incorrectly executed “interpretations” and provide a “supervenience base” for all acts of conceptualization.

95 Cf. I. Kant, Anthropology, History, and Education, op. cit., p. 258 (7:146). 96 Ch. Peacocke, “Does Perception Have a Nonconceptual Content?” op. cit., p. 241.

38 

 The Contents of Perceptual Experience: Opposing Views

The aim of this chapter has been to provide a brief introduction into the notions employed and issues raised most often in the debate between conceptualism and nonconceptualism. I have presented the standard arguments, employed on both sides of the debate, and contended that there is a risk that the arguments can be regarded as inconclusive. This is because the way in which concepts are understood throughout the debate remains unclear. This lack of clarity of the notion of concept will not be a problem for Kant who explicitly defines concept as a mediate, discursive and general representation which “contains that which is common to several objects, nota communis,”97 in opposition to intuition which is an immediate and singular representation of an object.98 In the following chapter, I will characterize Kant’s position against the background of the accounts of perceptual experience offered by such authors as Sellars and McDowell, and critically examine the Kantian inspirations in their works. At this point, only one thing remains to be made clear: Why is nonconceptual content an interesting issue at all? Why shall we talk about nonconceptual content rather than, for example, about sensations or sense data? It seems correct to reply by pointing out that the debate on nonconceptual content opens a wide area of investigations into the nature of intentionality in that it attempts to answer the question whether objects can be consciously represented without concepts and whether subjects can make use of such representations in both cognition and action.

97  I. Kant, Lectures on Logic, op. cit., p. 485 (24:752). 98  I. Kant, Critique of Pure Reason, op. cit., p. 399 (A 320/B 377).

2 Are the Roots of the Debate Kantian? 2.1 McDowell’s Kant99 In Mind and World, McDowell outlines a dialectical conflict between two positions, against the background of which he proposes his own “middle way” solution. Coherentism, represented by Donald Davidson, cuts empirical thought off from reality, because it holds that the content of perceptual experience does not suffice to provide the justification of empirical beliefs (beliefs can be justified only by other beliefs, on Davidson’s view; however, sensory content plays an important role in their acquisition). The Myth of the Given,100 in turn, with which McDowell charges philosophers such as Gareth Evans and Christopher Peacocke, is supposed to secure thought’s grip on reality; but it also thwarts rational connections between experience and thought. And this again makes it impossible to regard experiences as proper justifiers of empirical beliefs. As a way to escape the “oscillation” between the two positions, McDowell introduces the idea of conceptualized experience.101 He recognizes in Kant an advocate of the idea and one whose voice should be taken into account in the recent debate about the nature of empirical knowledge. Crucial in this return to Kant is McDowell’s appeal to what might be termed Kant’s Cooperation Thesis (henceforth, KCT), expressed as follows: “The original Kantian thought was that empirical knowledge results from a co-operation between receptivity and spontaneity. (Here ‘spontaneity’ can be simply a label for the involvement of conceptual capacities.) [...] Receptivity does not make an even notionally separable contribution to

99 A large part of this section is a modified version of one section of my paper “Experience and Conceptual Content in Kant and McDowell. Remarks on ‘Empty Thoughts’ and ‘Blind Intuitions,’” Diametros 28, 2011, esp. pp. 84-87. 100 McDowell borrows the term from Wilfrid Sellars, to be more precise: from his classic essay Empiricism and the Philosophy of Mind. There, in § 1, Sellars characterizes the “framework of givenness” and claims that “it has, indeed, been so pervasive that few, if any, philosophers have been altogether free of it; certainly not Kant, and, I would argue, not even Hegel, the great foe of ‘immediacy.’” Among the (immediately) “given” Sellars includes sensory contents, physical objects, universals, propositions, real connections, first principles. He admits that even the givenness itself can be regarded as “given.” W. Sellars, “Empiricism and the Philosophy of Mind” in: H. Feigl, M. Scriven (eds.), Minnesota Studies in the Philosophy of Science, vol. 1, University of Minnesota Press, Minneapolis 1956. I refer to the online edition of the text available at: http://www.ditext.com/sellars/epm.html, edited in Hypertext by A. Chrucky, 1995 (accessed on 21st December 2012). 101 For an argument for the claim that McDowell fails to overcome the “oscillation” and relapses into the Myth of the Given, see: C. Wright, “McDowell’s Oscillation,” Philosophy and Phenomenological Research 58 (2), 1998. © 2014 Anna Tomaszewska This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.

40 

 Are the Roots of the Debate Kantian?

the co-operation. The relevant conceptual capacities are drawn on in receptivity [...]. It is not that they are exercised on an extra-conceptual deliverance of receptivity. We should understand what Kant calls ‘intuition’ – experiential intake – not as a bare getting of an extra-conceptual Given, but as a kind of occurrence or state that already has conceptual content.”102

York Gunther restates McDowell’s claim: “In his slogan, ‘Thoughts without content are empty, intuitions without concepts are blind,’ Kant sums up the doctrine of conceptualism. While he specifically proposed the restriction for transcendental psychology, the latter half of the slogan neatly sums up the doctrine in any of its forms: religious, existential, aesthetic, epistemological, psychological, and so forth. According to the conceptualist, no intentional content […] is a content unless it is structured by concepts that the bearer possesses.”103

Also in his Woodbridge Lectures McDowell maintains the conceptualist reading of KCT in that he writes: “This picture of visual experiences as conceptual shapings of visual consciousness is already deeply Kantian, in the way it appeals to sensibility and understanding so as to make sense of how experiences have objective purport.”104

According to McDowell, KCT allows both (1) the avoidance of the notorious Myth by establishing that there are no cognitively significant concept-independent episodes that underlie empirical beliefs, and (2) the reinforcement of the connections between thought and reality by ensuring that perceptual experiences directly present us with objects. But does the meaning McDowell attaches to KCT in his account of perceptual experience correspond to the role KCT plays in Kant’s theory of empirical cognition? Arguably, what McDowell seems to take for granted, Kant endorsed as a result of a long development of his philosophical views. It seems, indeed, that McDowell builds on the assumption that KCT calls for no further justification. Moreover, he does not make an effort to question its coherence and, consequently, the coherence of his account of experience.105 Commentators have responded vividly to McDowell’s reading. Thus, Hannah Ginsborg, building on Kant’s account of imagination, argues that, according to

102  J. McDowell, Mind and World, Harvard University Press, Cambridge MA 1996, p. 9 (italics A.T.). 103 Y. Gunther (ed.), Essays on Nonconceptual Content, The MIT Press, Cambridge MA 2003, p. 1. 104  J. McDowell, “Lecture III: Intentionality as a Relation,” The Journal of Philosophy 95 (9), 1998, p. 471. 105 Importantly, McDowell does not see any problem in the application of concepts, which are general, to experience, which is subjective and context-specific, hence particular, by its nature. Kant takes up this issue in the Critique of Pure Reason, in the chapter on schematism.



McDowell’s Kant 

 41

Kant, perception is accompanied by “consciousness of normativity”106 which originates from the understanding but which “does not presuppose any antecedent grasp of concepts.”107 Therefore, perception is prior to concept formation and judgment but it is conceptually informed insofar as it is guided by the consciousness of normativity. By contrast, Robert Hanna appeals to Kant’s theory of intuition as a sui generis source of cognitions which it would not be possible to acquire otherwise than by direct acquaintance with objects. In particular, in his argument for the nonconceptualist reading of Kant, Hanna draws upon Kant’s doctrine of incongruent counterparts (enantiomorphic figures),108 that is, objects that Kant defines as “exactly equal and similar, and yet still […] so different in themselves that the limits of the one cannot also be the limits of the other.”109 Among incongruent counterparts Kant includes: two screws with their threads curved in the opposite directions; two spherical triangles of equal size; one’s right and left hands; or one’s hand and its mirror reflection. Although such objects do not differ with regard to their intrinsic properties, and so their descriptions would have to be identical, one object of this kind cannot occupy the space taken up by the other, which can be recognized in intuition. Also, Lucy Allais encourages a (moderately) nonconceptualist interpretation. On her reading, “one of Kant’s fundamental thoughts is that the kind of direct contact with objects that involves particular things being present to consciousness is essential for cognition.”110 Intuition “gives” us particulars, whereas concepts enable categorization and property-attribution; it is also due to concepts that relations between particulars become established (or revealed). A similar approach one can find in Richard Aquila’s Representational Mind, where he argues that intuition confers referential properties upon concepts, constitutes the basic form of object-directedness, or pre-conceptual intentionality, and sets up constraints to what can be represented and to the faculty of representation. As Aquila puts it, our cognition becomes constrained by the “nonconceptual features” of sense representations,111 that is, their spatiality and/or temporality.

106 H. Ginsborg, “Was Kant a Nonconceptualist?” Philosophical Studies 137 (1), 2008, p. 74. 107 Ibidem. 108 See, e.g., R. Hanna, Kant, Science and Human Nature, Oxford University Press, Oxford 2003; “Kant and Nonconceptual Content,” European Journal of Philosophy 13 (2), 2005; “The Myth of the Given and the Grip of the Given,” Diametros 27, 2011. 109 I. Kant, Theoretical Philosophy, 1755 – 1770, translated and edited by D. Walford in collaboration with R. Meerbote, Cambridge University Press, New York 1992, p. 369 (2:381). 110 L. Allais, “Kant, Non-Conceptual Content and the Representation of Space,” Journal of the History of Philosophy 47 (3), 2009, p. 392. 111 R. Aquila, Representational Mind: A Study of Kant’s Theory of Knowledge, Indiana University Press, Bloomington 1983, p. 99.

42 

 Are the Roots of the Debate Kantian?

Finally, a number of authors – e.g. Wilfrid Sellars,112 Lorne Falkenstein,113 Rolf George114 or Kenneth Westphal115 – opt for conceptualism but they recognize the important role of sensory (and non-intentional) content in providing the “matter” of cognition to the subject. But Kant’s account of experience, with what has been labelled Kant’s Cooperation Thesis (KCT) as one of its key constituents, can also be considered from an historical perspective. This may contribute to better understanding its significance for Kant’s transcendental theory of experience. Mario Caimi has offered a reading in this vein. He has argued that KCT (more specifically, its first part about thoughts without content being empty) is only seemingly tautological. For “Kant introduces, with this apparently tautological statement, a revolutionary novelty into logic as it was conceived in his times.”116 The novelty consists in the fact that “on Kant’s account, the richness or the emptiness of a concept does not depend merely on its logical possibility anymore”117 and in the fact that “sensibility becomes recognized as a legitimate source of cognition.”118 Thus, Caimi concludes, the understanding turns out to fall short of providing all (kinds of) cognitions. The title of this chapter raises a question whether the roots of the recent debate about the contents of perceptual experience are genuinely Kantian. For a number of reasons one can be skeptical about a positive answer to the question. First of all, one may claim that the overall aim of the doctrines offered in the first Critique has been to shed light upon the conditions of empirical or, more specifically, scientific cognition.119 It is overcoming the Humean challenge, arguably, that remains of primary importance, rather than building a theory of perceptual experience. Second, Kant is certainly neither the first nor the last thinker in the history of philosophy whose epistemological conceptions entail an account of perceptual

112 See, e.g., W. Sellars, “Some Remarks on Kant’s Theory of Experience,” The Journal of Philosophy 64 (20), 1967, pp. 633-647; Essays in Philosophy and its History, D. Reidel Publishing Co., Dordrecht 1974; Science and Metaphysics. Variations on Kantian Themes, Ridgeview, Atascadero 1992; P. Amaral (ed.), Kant and Pre-Kantian Themes: Lectures by Wilfrid Sellars, Ridgeview, Atascadero 2002. 113 L. Falkenstein, Kant’s Intuitionism. A Commentary on the Transcendental Aesthetic, University of Toronto Press, Toronto 2004. 114 R. George, “Kant’s Sensationism,” Synthese 47 (2), 1981. 115 K. Westphal, “Contemporary Epistemology: Kant, Hegel, McDowell,” European Journal of Philosophy 14 (2), 2006. 116 M. Caimi, “»Gedanken ohne Inhalt sind leer«,” Kant-Studien 96 (2), 2005, p. 141 (my translation). 117 Ibidem, p. 144 (translation A.T.). 118  Ibidem, p. 146 (translation A.T.). 119 An excellent interpretation along these lines can be found in: H.-J. Vleeschauwer, The Development of Kantian Thought. The History of a Doctrine, translated by A. R. C. Duncan, T. Nelson and Sons, Edinburgh 1962.



McDowell’s Kant 

 43

experience. Indeed, such an account typically belongs to any theory of cognition, and Kant of course provides such a theory. Third, the roots of the debate may be identified, perhaps more adequately, within an inquiry about the foundations or the justification of empirical knowledge, where the available options are skepticism, on the one hand, and foundationalism or coherentism, on the other.120 Finally, it can be observed, Kant’s theory of experience encourages incompatible interpretations. Indeed, it can be read as supporting the claim about the existence and cognitive significance of conceptual content (CC), as well as sensory content (SC) and nonconceptual content (NCC); or the existence and cognitive significance of CC and SC alone; or even, on McDowell’s reading, it has been taken as promoting radical conceptualism. Now, the above objections can be waived in the following manner. First, historically speaking, Kant’s aim (especially in the Critique of Pure Reason) did consist, as a matter of fact, in establishing the necessary conditions of mathematics and natural science but, since these results have been, for the most part, recently disputed, it would perhaps be more fruitful, from the epistemological perspective, to consider a more general framework which the transcendental philosophy offers. And this framework certainly makes room for reflection on the necessary (transcendental) conditions of cognition in general. Second, Kant embarks on a radical change of perspective on cognition and its conditions (this idea is captured as “Copernican revolution”); against this background, an account of perceptual experience forms part of more general considerations about the relation between cognition and reality. Third, while Kant focuses on mathematics and natural science, he thinks that the conditions of science are the same as the conditions of all empirical cognition, or human experience in general, a belief closer perhaps to Aristotle than the Cartesian tradition. Thus, it is entirely legitimate to focus on the account of experience and abstract from its relation to the account of science. Finally, I believe that there are reasons to read Kant as an advocate of one of the above views on perceptual content: (1) the view endorsing CC, NCC and SC; (2) the view endorsing only CC and SC; or (3) the view accepting CC alone, i.e. radical conceptualism. As I intend to show in the chapters to follow, a reading that opts for (1), that is, one that allows attributing (moderate) nonconceptualism to Kant, should be deemed most accurate.

120 The problem has been formulated in: L. BonJour, “Can Empirical Knowledge Have a Foundation?” American Philosophical Quarterly 15 (1), 1978.

44 

 Are the Roots of the Debate Kantian?

2.2 Kant and the “Myth of the Given” While we have a general grasp of the concept of myth, the philosophical concept of the “given” needs to be clarified. Its application falls within the area of the theory of cognition. The concept can be used in order to describe certain features of the content or the object of mental acts and processes. Drawing upon an information-theoretic metaphor, one may associate the given with the input rather than the output or the result of cognitive processes. As such, it may be characterized as simple, or at least simpler than the outcome of these processes. Since the given does not result from any cognitive processing, our cognitive access to it must be unmediated. What satisfies the description? According to Wilfrid Sellars, “many things have been said to be ‘given’: sense contents, material objects, universals, propositions, real connections, first principles, even givenness itself.”121 Thus, on the side of the given one may place items as different as innate ideas, on the one hand, and sense data, on the other. But the above characteristic of the given is quite rough; for, one may remark, cognitive immediacy does not have to entail the lack of mental processing. Thus, one can argue that we perceive colours immediately but underlying our perception of colours are certain subconscious or “subpersonal” representational states.122 The reason we perceive colours immediately would be that perception of colours does not require the involvement of other conscious cognitive acts or processes, such as inference. Therefore, the contents or the objects of mental acts or processes can be given in some sense but not given in another. Sellars criticizes the idea of the given, which he thinks to be a myth, in his seminal paper Empiricism and the Philosophy of Mind. The idea of the given is a myth because it generates an allegedly mistaken picture of empirical knowledge as built upon foundations which subjects have privileged access to. Such a structure cannot represent the edifice of knowledge in an adequate manner because, in Sellars’s view, none of the elements of the edifice can entertain such privileged status. Thus, for example, while the sense data language would be derivative of the language used in order to refer to physical objects in space and time,123 sense data (or sense impressions) would play a role in experience “guiding” our thought, or giving it a grip on the mind-

121  W. Sellars, “Empiricism and the Philosophy of Mind,” op. cit.; quoted from the website: http:// www.ditext.com/sellars/epm.html, § 1 (accessed on 31st Dec 2012). 122 An example of such states can be provided by Marr’s 2.5-dimensional “sketches” which can be separated from the chain of processes leading to vision. For a discussion of subpersonal states and their content see, e.g., J. L. Bermúdez, “Nonconceptual Content: From Perceptual Experience to Subpersonal Computational States,” Mind and Language 10 (4), 1995. 123 For an illustration of this claim, see: W. Sellars, Science and Metaphysics. Variations on Kantian Themes, Ridgeview, Atascadero 1992, ch. 1.



Kant and the “Myth of the Given” 

 45

independent reality.124 With this conception of sense data at hand Sellars reads Kant’s account of intuition. On Sellars’s construal, when Kant speaks about intuition, he refers to two different things.125 On the one hand, intuitions are what singular terms (“this chair,” “Professor Smith,” “the highest skyscraper in New York,” “this”) refer to. Rendered by such terms, intuitions are conceptualized. Singular terms form part of judgments in which the objects of intuition are predicated of. Judgments are truth-evaluable if their scope of reference does not go beyond spatiotemporal bounds. Therefore, the objects of conceptualized intuitions are to be identified with particulars in space and/ or time. On the other hand, Sellars also claims that the Kantian intuitions play an explanatory role in that they form part of a causal chain that starts from an object affecting a subject’s sense organ and ends with verbal behaviour as a response to the affection. Accordingly, experience imposes a kind of constraint upon thought in the sense that it prompts us to attribute the origin of our representations (perceptions, thoughts) to objects external to the mind. But we do not have to experience intuitions, understood along these lines, in an immediate manner, in order to grant an explanatory role to them. Rather, they may be posited as hypothetical or theoretical entities that account for some (in particular qualitative) aspects of our representations. This is where Sellars’s reading seems to derive its inspiration from Descartes’ account of sensory content and the explanation of its origin in the sixth meditation, and where it also itself seems to lapse into the Myth of the Given. McDowell critically assesses the reading in that he says: “Sellars’s idea is that for thought to be intelligibly of objective reality, the conceptual representations involved in perceptual experience must be guided from without. Indeed they are, I can say. But there is no need for manifolds of ‘sheer receptivity’ to play this guiding role. In a way we are now equipped to understand, given the conception of intuitions adumbrated in the passage from the ‘Clue,’ the guidance is supplied by objects themselves, the subject matter of those conceptual representations, becoming immediately present to the sensory consciousness of the subjects of these conceptual goings-on.”126

124 W. Sellars, Science and Metaphysics. Variations on Kantian Themes, op. cit. Cf. J. McDowell, Having the World in View: Essays on Kant, Hegel, and Sellars, Harvard University Press, Cambridge MA 2009. 125 See: W. Sellars, Science and Metaphysics. Variations on Kantian Themes, op. cit.; P. Amaral (ed.), Kant and Pre-Kantian Themes: Lectures by Wilfrid Sellars, Ridgeview, Atascadero 2000, ch. 7. 126  J. McDowell, “Lecture II: The Logical Form of an Intuition,” The Journal of Philosophy 95 (9), 1998, p. 467. The passage in the Critique of Pure Reason McDowell refers to reads: “The same function that gives unity to the different representations in a judgment also gives unity to the mere synthesis of different representations in an intuition, which, expressed generally, is called the pure concept of understanding.” I. Kant, Critique of Pure Reason, op. cit., p. 211 (B 105).

46 

 Are the Roots of the Debate Kantian?

Willem DeVries has argued that there is a certain role to play for mere impressions of the senses in both empirical and transcendental arguments.127 In the former, the appeal to the given helps explain phenomena like illusions, hallucinations or blindsight in terms of perceptual errors. In the latter, the given figures in the justifications of the claim that our perceptual experiences, and intentional states in general, are of mind-independent realities, the subject matter of our observational reports and beliefs. According to McDowell, this “guidance from outside” is unnecessary, since the object-relatedness of our perceptual states is ensured by the fact that conceptual capacities are involved in every act of perceptual experience. This is because, as noted in chapter 1, on McDowell’s view, perception is factive. Concepts make objects available in conscious experience because, on these assumptions, there is a correspondence between the structure of perceived facts, constituted by objects, and the structure of propositions which capture these facts, constituted by concepts. But, as stated in the previous chapter, there is not much more than a kind of rational faith in the intelligibility of the world that McDowell may offer when challenged by the skeptic. Sellars also does not provide a satisfactory means by which to fend off the challenge effectively; for appealing to a mind-independent reality in order to explain the qualitative aspect of experience assumes a form of a hypothesis which may be refuted once a more compelling one has been found to replace it. From the Kantian perspective, however, it seems that none of the interpretative proposals fits the bill in all relevant respects. Sellars’s reading does not adequately render Kant’s conception of intuition because it does not explain the fact that intuition accounts for intentional relatedness to particular objects in space and time. For sense impressions do not suffice to give us intuitions of spatiotemporal particulars because sense impressions do not convey information about spatial and temporal properties of objects. Also, it is difficult to see how concepts could do the job, since, according to Kant, cognition of particular objects does not derive from concepts alone; rather, the form of particularity originates from a faculty which significantly differs from the understanding. On the other hand, McDowell, with his “rationalist faith” in the correspondence between mind and world, apparently ignores transcendental idealism and its corollaries; indeed, he seems either to push Kant back on the track of the pre-critical, and perhaps even the pre-Cartesian philosophy, or to pull him in the direction of absolute idealism of the Hegelian kind, which Kant, from all we know, was all but willing to endorse. But before I argue that, for these reasons, Kant’s account of experience should rather be read as a version of nonconceptualism, I would like to address the concern whether it is even legitimate to formulate such a claim, considering the fact that the notion of experience can be understood and

127 W. deVries, “Sellars vs. McDowell on the Structure of Sensory Consciousness,” Diametros 27, 2011.



Kant’s Notion of Experience and the Conceptualism vs. Nonconceptualism Debate 

 47

applied in a different way by Kant and throughout the debate between conceptualists and nonconceptualists.

2.3 Kant’s Notion of Experience and the Conceptualism vs. Nonconceptualism Debate One may object that the attempt to take Kant as a side in the recent debate on perceptual experience results from misconstruing the purport of Kant’s theory of experience and misusing the concept of experience. For whereas contemporary theories aim at an account of perceptual experience (seeing, touching, hearing etc.), Kant would rather deal with empirical cognition, i.e., the kind of experience that leads to judgment. Thus, the question arises whether one may read the Critique of Pure Reason as an account of perceptual experience, with perceptual experience understood along the lines of contemporary theories. There is textual evidence to the effect that the subject matter of these theories was of relatively little interest to Kant and formed just part of his theory of empirical cognition. For example, Kant distinguishes between degrees of knowledge and identifies perception (Wahrnehmung) as the second lowest degree.128 Also, Kant describes perception, in the Lectures on Logic, in different ways, which sometimes seem incompatible. In the notes collected as Vienna Logic, he writes: “In perception we do not relate our cognitions to the object. Through sensation, good feeling, pain – one does not cognize an object. This is only a mode of representation.”129

And in Dohna-Wundlacken Logic one can read: “A representation combined with consciousness is called perception, perceptio. This perceptio becomes cognitio insofar as the representation is related with consciousness to an object.”130

128 The degrees encompass: (1) an idea (Vorstellung); (2) a conscious idea – perception; (3) knowledge of an object in relation to other objects, which requires the ability to identify and differentiate objects; (4) conscious knowledge of an object – cognition (according to Kant, animals can know objects but they cannot cognize them); (5) cognition by the understanding by means of concepts – conception; (6) rational cognition marked by perspicuity; (7) rational a priori comprehension marked by adequacy. See: Kant’s Introduction to Logic and his Essay on the Mistaken Subtlety of the Four Figures, translated by T. K. Abbott, Thoemmes Press, Bristol 1992, pp. 55-56. 129 I. Kant, Lectures on Logic, translated and edited by J. Michael Young, Cambridge University Press, New York 1992, p. 348. 130 Ibidem, p. 485.

48 

 Are the Roots of the Debate Kantian?

Perceptions (conscious representations) become then divided into intuitions and concepts. Kant seems thus to be ambivalent between construing perception in terms of sensing and construing perception in terms of representation of an object. In the first Critique, Kant does not even consider different sense modalities and speaks of sensibility in general; particular sense modalities are mentioned in the Anthropology from a Pragmatic Point of View, where Kant also distinguishes between the senses according to the degree of the objectivity of the information they convey (some senses, such as touch, sight and hearing, are more “objective,” whereas other senses are more “subjective,” e.g., taste and smell).131 Kant maintains there that children do not have experiences (Erfahrungen), but disconnected perceptions (Wahrnehmungen). Clearly, Kant distinguishes between sense perception and experience, and claims that only the latter brings about cognition in that it provides knowledge of objects and their relations.132 In the Lectures on Logic, he states this explicitly: “Experience is an operation of the understanding, hence perception does not at all suffice. […] Perception is only subjective – not yet experience. This is cognition with consciousness of the relation to the object. […] Experience means empirical cognition. With an empirical representation something must be perceived (sensed with consciousness). Hence experience presupposes empirical intuition and a concept. Sensation is indispensable for this.”133

Thus, according to Lorne Falkenstein, Kant’s theory of sensibility falls outside the scope of a theory of empirical cognition and belongs to the domain of physiology. On Falkenstein’s reading, “the distinction between the intuitive and the intellectual in Kant is properly to be seen as a distinction between the physiological and the psychic or, more exactly (if we do not want to presuppose dualism), between the physical and the cognitive.”134 However, despite these terminological differences, there are also points of convergence between Kant and the contemporary debate. For both, the problem of the relation between concepts and perceptions (or intuitions in Kant’s terminology), or perception and belief, remains relevant. We may thus try to see how Kant deals with the problem and whether the possible solutions he reaches can apply in the context of the debate between conceptualism and nonconceptualism. Thus, for example, Kant’s account of the transcendental syntheses may be useful in answering the question whether perceptual experience necessarily involves conceptual capacities, and if

131 I. Kant, Anthropology, History, and Education, edited by G. Zöller and R. B. Louden, translated by M. Gregor et al., Cambridge University Press, Cambridge 2007, pp. 265-6 (7:154). 132 Ibidem, p. 232 (7:128). 133 I. Kant, Lectures on Logic, op. cit., pp. 484-6 (24:750-3). 134 L. Falkenstein, Kant’s Intuitionism. A Commentary on the Transcendental Aesthetic, University of Toronto Press, Toronto 2004, p. 123.



Kant, Nonconceptual Content, and Body in Cognition 

 49

it does, what the involvement amounts to. Kant’s theory of the forms of intuition may, on the other hand, be useful in explaining key differences between perception, imagination, and thought. The doctrine of transcendental schematism provides the Kantian reply to the question about the possibility of connecting heterogeneous representations (concepts and intuitions) in experience. Kant’s transcendental idealism, which can be regarded as rooted in the distinction between (sensible) intuition and thought,135 may, in turn, be read as a doctrine concerning epistemic constraints intrinsic to (human) cognition. Locating these constraints in the constitution of the human sensibility can perhaps make it possible to account for the origin of nonconceptual content in a concise and unitary way.136 In the next section, I investigate into this possibility.

2.4 Kant, Nonconceptual Content, and Body in Cognition Advocates of nonconceptualism appeal predominantly, in their arguments, to the phenomenology of experience. For instance, they tend to emphasize the greater richness and fineness of grain of the content of perception in comparison with the content of beliefs or judgments. Nonconceptual content, it is also claimed, can be invoked to explain perceptual and memory illusions, or demonstrative identification and perceptual reference. Kantian advocates of nonconceptualism, such as L. Allais or R. Hanna, for their part, point to the fact that in perception the subject is presented with particulars and the experience of particulars cannot be rendered exclusively by concepts. But Kant offers more than a phenomenological description of experience. Already in the pre-critical writings he produces arguments for the distinction between sensible and intellectual cognition, in which the former one could take as equivalent to the nonconceptual constituent of experience. At the same time, Kant advances a claim – although calling it a hypothesis would perhaps bear more plausibility – about the source of the distinction being attributable to the sensible, or more properly: embodied, nature of the subject of experience. Thus, in his 1768 essay Concerning the Ultimate Foundation of the Distinction of Directions in Space, Kant identifies a set of cognitions that are due to the bodily constitution of the subject. The directionality of an object belongs to the set since it is in the relation to the subject’s body that the

135 This is largely a Hegelian view. On its justification, see: Paul Guyer, “Absolute Idealism and the rejection of Kantian dualism” in: Karl Ameriks (ed.), The Cambridge Companion to German Idealism, Cambridge University Press, New York 2000. 136 Robert Hanna points out that the lack of a unitary account of nonconceptual content is what troubles theorists, and suggests a Kantian version of nonconceptualism as a solution. See: “Kant and Nonconceptual Content,” op. cit., especially pp. 251-2, 278, 282.

50 

 Are the Roots of the Debate Kantian?

property can be determined. Here is an extended quotation from Kant’s text, which illustrates the point: “It is […] not surprising that the ultimate ground, on the basis of which we form our concept of directions in space, derives from the relation of [the] intersecting planes to our bodies. The plane upon which the length of our body stands vertically is called, with respect to ourselves, horizontal. This horizontal plane gives rise to the difference between the directions which we designate by the terms above and below. On this plane it is possible for two other planes to stand vertically and also to intersect each other at right angles, so that the length of the human body is thought of as lying along the axis of the intersection. One of these two vertical planes divides the body into two externally similar halves, and furnishes the ground of the difference between the right and the left side. The other vertical plane, which also stands perpendicularly on the horizontal plane, makes possible the concept of the side in front and the side behind.”137

Thus, says Kant, there are certain spatial determinations – “above,” “below,” “to the right,” “to the left,” “in front,” “behind” – the meaning of which is fixed in relation to the human body. We may also put it this way: the subject’s body structures the field of experience as long as we may find within this field, apart from objects, their spatial properties and relations, such as directionality.138 Since “above,” “to the left,” etc., acquire their meaning through a relation to a particular human body and in a particular context in which they are used, they are not general concepts. Since fixing their reference presupposes relation to a particular body, rather than for example mastery of the concept of space, or any concept, for that matter, proper understanding and application of these determinations involves nonconceptual content: the subject’s intuition of a particular body which in the end turns out to be identical with his own. Indeed, Kant contends, fixing the reference of these spatial terms would be grounded on a feeling which, perhaps with a grain of plausibility, may be identified as proprioceptive (more on that issue in chapter 5). Later in the essay one can read: “Since the distinct feeling of the right and the left side is of such great necessity for judging directions, nature has established an immediate connection between this feeling and the mechanical organization of the human body.”139

137  I. Kant, Theoretical Philosophy, 1755 – 1770, op. cit., pp. 366-7 (2:379). 138 The constitutive character of body in cognition is underlined in some recent Kant interpretations. See: R. Hanna, “The Inner and the Outer: Kant’s ‘Refutation’ Reconstructed,” Ratio 13 (2), 2000; M. S. Rukgaber, “‘The Key to Transcendental Philosophy:’ Space, Time and the Body in Kant,” Kant-Studien 100 (2), 2009. 139 I. Kant, Theoretical Philosophy, 1755 – 1770, op. cit., p. 369 (2:380).



Kant, Nonconceptual Content, and Body in Cognition 

 51

This “distinct feeling,” rather than any kind of mediate, conceptual cognition, accounts for the possibility of distinguishing between the sides of one’s body. To illustrate the point: it is not the case that in order to know, for example, that “this is my right hand,” I have to make an inference from the premise that “this is the hand I usually write with.” The cognition that “this is my right hand” comes immediately; in fact, this kind of knowledge is “already there,” before I even start to reflect upon it. Hence, it may be called “tacit knowledge.”140 Understanding concepts like “to the left,” “below,” “in front of,” etc., does not come through the mastery of concepts, or through other more basic cognitions (it is not because we have definitions of “the right” and “the left” that we do not confuse the right with the left sides of our bodies), and the distinctions between them can be clarified in relation to action. A contemporary philosopher, Gareth Evans, expresses a thought in a similar vein: “The subject hears the sound as coming from such-and-such a position, but how is that position to be specified? We envisage specifications like this: he hears the sound up, or down, to the right or to the left, in front or behind, or over there. It is clear that these terms are egocentric terms; they involve the specification of the position of the sound in relation to the observer’s own body. But these egocentric terms derive their meaning from their (complicated) connections with the actions of the subject. […] No one hears a sound as coming from the same side as the hand he writes with in the sense that, having heard it thus, he has to say to himself ‘Now I write with this hand’ (wiggling his right hand), ‘so that sound must be coming from over there’ (pointing with his right hand). Rather, he can immediately say to himself ‘It’s coming from over there’ (pointing with what is in fact his right hand), and may then reflect as an afterthought ‘and that’s the hand I write with.’”141

The quotation comes from Evans’s essay Molyneux’s Question, in which he develops an account of the representation of space. On Evans’s account, the representation of space is immediate, in the above described sense, hence nonconceptual, and amodal in that it does not require any specific sense modality in order to be entertained by a cognitive system. (This is why, perhaps, blind people can sometimes use sonar devices which help them in locating objects on the basis of sounds emitted by these objects.) In other words, the representation of space would be neither abstracted from information of a non-spatial kind, nor constructed on the basis of specific sensory content. The representation of space is thus a sui generis kind of representation and cannot be derived from what lacks any spatial organization.142 Evans’s account aligns,

140 Cf. Michael Polanyi, The Tacit Dimension, University of Chicago Press, Chicago 2009. 141 G. Evans, “Molyneux’s Question” in: A. Phillips (ed.), Gareth Evans: Collected Papers, Clarendon Press, Oxford 1985, p. 384. 142 The claim can be illustrated with a physiological example, given by Patricia Kitcher. The human eye receives sensory information as two-dimensional retinal arrays which require to be further pro-

52 

 Are the Roots of the Debate Kantian?

in this way, with Kant’s transcendental aesthetic. The latter introduces space as a form of intuition and characterizes it as not abstracted from “outer experiences;”143 an a priori “ground of all outer intuitions;”144 the source of geometrical truths;145 not discursive, hence not a concept but “a pure intuition;”146 and “a given infinite magnitude.”147 (Similar characteristics appear in the Inaugural Dissertation.) I started this chapter by discussing the Kantian inspirations behind McDowell’s and Sellars’s theoretical proposals. I pointed out that, as interpretations of Kant, these proposals fall short of being adequate since they misconstrue the cognitive status of the a priori forms of intuition, the pure intuitions of space and time. McDowell cannot account for the specific features of intuitions that make them generically distinct from concepts. Sellars, too, fails to provide such a theory, since he either equates intuitions with conceptualized representations, or with sensory content. Both fail to recognize the sui generis kind of the pure intuitions (and, accordingly, of all representations of sensibility). The characteristics of space in Kant’s Transcendental Aesthetic, which I have referred to, should be considered against the background of the eighteenth-century controversies over the nature of space and time, geometrical and mathematical cognition, and perception of distance (or the third dimension).148 In the controversies, Kant is situated, as a rule, between empiricists and rationalists, that is, between Locke (and Berkeley), on the one side, and Leibniz, on the other. According to Kant’s famous contention: “Leibniz intellectualized the appearances, just as Locke totally sensitivized the concepts of understanding […]. Instead of seeking two entirely different sources of representation in the understanding and the sensibility […] each of these great men holds on only to one of them […] while the other does nothing but confuse or order the representations of the first.”149

Certainly, there are overtones reminiscent of the debate between rationalists and empiricists in the contemporary debate on the content of perceptual experience. Since McDowell makes no room for nonconceptual content, he can be said to have

cessed in order to produce a three-dimensional representation of an object. The forms of intuition enable the interpretation of the two-dimensional arrays and the transition to full-fledged representational content. See: Kant’s Transcendental Psychology, Oxford University Press, Oxford – New York 1990, pp. 44-5. 143  I. Kant, Critique of Pure Reason, op. cit., p. 157 (A 23/B 38). 144  Ibidem, p. 158 (A 24/B 39). 145  Ibidem. 146  Ibidem, p. 158 (A 25). 147  Ibidem, p. 159 (A 25). 148  Cf. P. Kitcher, Kant’s Transcendental Psychology, op. cit., pp. 205-248. 149  I. Kant, Critique of Pure Reason, op. cit., p. 372 (A 271/B 327).



Kant, Nonconceptual Content, and Body in Cognition 

 53

intellectualized the Kantian pure intuitions.150 Sellars, who identifies nonconceptual content with sensations, can be said, on the other hand, to have sensitivized them. Evans, who offers an account of space as both a nonconceptual and a non-derivative representation, seems thus to represent Kant’s position most adequately. From this perspective, it can be admitted that, indeed, the roots of the contemporary debate are, at least in part, Kantian.

150  McDowell notices this mistake but criticizes Kant on account of having introduced sensibility as a “brute” fact and its representations – as the “given.” See: “Hegel and the Myth of the Given” in: W. Welsh, K. Vieweg (eds.), Das Interesse des Denkens: Hegel aus heutiger Sicht, Wilhelm Fink Verlag, München 2003.

3 Kant on Nonconceptual Content: Sensations and Intuitions 3.1 Representational Content Revisited In the context of the debate, characterized in chapter 1, philosophers often talk about “representational content” or “representational states.” This qualification captures a fundamental distinction between representational and nonrepresentational content of perceptual experience. There are, of course, philosophers who do not endorse the distinction. Thus, McDowell would identify content as “what is given by a ‘that’ clause in, for instance, an attribution of a belief.”151 On this account, should there be otherwise specifiable mental states, one would not attribute any content to them. In cognitive psychology and philosophy of mind, the concept of representation plays the key role in explaining cognition and action of conscious creatures. Two ideas are particularly closely related to the representational account of the mental: the idea of mediation and the idea of information. By means of representations subjects acquire cognitive access to their environment, both internal (states of the organism) and external (objects outside the mind). Since the mind can be considered active in that it builds its representations, mental states do not have to enable direct access to the environment, which would be the case if representations were somehow given to the mind (this being a necessary, rather than a sufficient, condition of immediacy). Such an account also precludes direct access to representations themselves, since they are regarded as, in a way, theoretical entities figuring at the level of the explanation of cognitive processes, rather than at the level of the latter’s content.152 Regarding the idea of information in the representational theory of the mind: on this theory, representations are construed as (physical) vehicles of information. They convey information about the environment, external or internal, to the representational system, in the way that speedometers, pressure gauges or thermometers do. Fred Dretske, a proponent of such an approach and the author of these “technical” metaphors, features representations as functions of the internal setup (which he calls design and calibration) of a representational system.153

151  J. H. McDowell, Mind and World, Harvard University Press, Cambridge MA 1996, p. 3. 152  These ideas are defended by Thomas Metzinger in: Being No One. The Self-Model Theory of Subjectivity, The MIT Press, Cambridge MA 2003. Interestingly, Metzinger makes use of one of the favourite ideas of phenomenologists, which is that, prereflectively, subjects tend to entertain naïve realist beliefs about the objects of cognition. This, according to the author of Being No One, has evolutionary underpinnings. 153  F. Dretske, Naturalizing the Mind, The MIT Press, Cambridge MA 1995. © 2014 Anna Tomaszewska This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.



Representational Content Revisited 

 55

On some readings, Kant also endorsed the representational theory. Andrew Brook, who offers a reading along these lines, characterizes the theory in the following way: “The basic idea behind the representational model is that the function of the mind is to shape and transform representations. Kant too had a representational model of the mind (a rather radical one […]), and his view of the mind as a system for applying concepts to percepts is entirely in line with functionalism. The three basic moves underlying Kant’s model of the mind are as follows. (i) Most or all representation is representation of objects; such objects are the result of acts of synthesis. (ii) For representations of objects to be anything to anyone, they must ‘belong with others to one consciousness’ […]; for this, the mind must synthesize its various objects of representation into […] the global object of a special kind of single complex representation called a global representation. (iii) Synthesis into either individual or global objects requires the application of concepts.”154

For our purposes, points (i) and (iii) are of special significance in that they bring out the following issues: whether representations, as construed by Kant, refer to objects, in a way, by their nature; whether representations are products of the “synthesizing” activity of the mind rather than the mere “given;” and whether the employment of concepts is necessary to account for the activity of the mind.155 Considering these issues will certainly shed some light on Kant’s ideas on the nature of the mind. In the Cartesian and British empiricist traditions, it was by means of the term “idea” that philosophers referred to the most basic mental items. Kant makes use of the term “representation” (Vorstellung) throughout the Critique of Pure Reason; but in the pre-critical writings he frequently employs such terms as “concept” (conceptus, Begriff) or “idea.” All these terms seem to be flawed with an ambiguity Descartes brought into light when characterizing ideas in a passage that precedes the main text of the Meditations on First Philosophy, where one can read: “[T]he word idea […] may be taken materially, as an operation of the intellect […]. Alternatively, it can be taken objectively, as the thing represented by that operation.”156

Sellars notes a similar ambiguity in Kant’s use of the term “representation,” and remarks:

154  A. Brook, Kant and the Mind, Cambridge University Press, Cambridge 1994, pp. 12-13. 155  Both Brook and McDowell claim that the employment of concepts or “conceptual capacities” is a prerequisite of forming a representation, according to Kant. But the point where similarity between the two readings disappears is this: according to Brook, concepts are directly applied to “percepts” delivered by sensibility; in McDowell’s view, concepts are “already there,” involved in the deliverances of the senses. 156 R. Descartes, Meditations on First Philosophy. With Selections from the Objections and Replies, ed. by J. Cottingham, Cambridge University Press, Cambridge 1996, p. 7.

56 

 Kant on Nonconceptual Content: Sensations and Intuitions

“Kant never tires of telling us that Nature and the Space and Time which are its forms exist as a system of ‘representations.’ Now a representation is either a representing or a something represented. Does Kant mean that nature is a system of representings? Or that it is a system of representeds? And, in any case, what would the claim amount to?”157

Thus, there are two alternative ways of understanding the term “representation,” as used by Kant, along the lines of the Cartesian distinction. Accordingly, one can think of representations either as acts of the mind, or as what the acts of the mind are about, what they refer to, i.e. their contents. Perceiving, imagining, remembering, thinking, etc., would belong to the first group. But what kind of items would belong to the second group? Answering the question about the kind of items which, for Kant, would fall under mental contents presupposes a determinate reading of Kant’s theory of experience, in terms of, for instance, McDowell’s conceptualist, Brook’s cognitive psychological, or Hanna’s nonconceptualist interpretation of the theory. In the Stufenleiter passage,158 Kant presents a kind of the Porphyrian tree for the notion of representation. There, he subsumes under that heading items such as sensations, intuitions, and concepts. But, one can note, even if the former two can be qualified as either acts or their contents, one may consider such a qualification incorrect with respect to concepts. For, arguably, concepts are distinct from particular acts of the mind,159 although conceiving belongs to the category of mental acts. Likewise, not much sense can be made of understanding concepts as the contents of mental acts, i.e. as what the acts are about, or what they refer to. This is not to imply that concepts do not figure in the specification of mental content. But in the sense that neither Sellars’, nor Descartes’ distinction seems to capture, the kind of representations that concepts are would play the role of intermediaries, things by means of which some other things can be carried out, hence a sort of “tools.” Let us now illustrate the kind of difficulties that arise once terminological considerations have been taken into account. To that end, I will consider the following passage from the Transcendental Aesthetic: “Time is the a priori formal condition of all appearances in general. Space, as the pure form of all outer intuitions, is limited as an a priori condition merely to outer intuitions. But since, on the contrary, all representations, whether or not they have outer things as their object, nevertheless as determinations of the mind themselves belong to the inner state, while this inner state belongs

157  W. Sellars, “Some Remarks on Kant’s Theory of Experience,” The Journal of Philosophy 64 (20), 1967, p. 633. 158  I. Kant, Critique of Pure Reason, translated and edited by P. Guyer and A. Wood, Cambridge University Press, Cambridge – New York 1998, pp. 398-9 (A 320/B 376-7). 159  Otherwise, no two individuals could possess the same concept; there would be no shared beliefs.



Representational Content Revisited 

 57

under the formal condition of inner intuition, and thus of time, so time is an a priori condition of all appearance in general, and indeed the immediate condition of the inner intuition (of our souls), and thereby also the mediate condition of outer appearances. If I can say a priori: all outer appearances are in space and determined a priori according to the relations of space, so from the principle of inner sense I can say entirely generally: all appearances in general, i.e., all objects of the senses, are in time, and necessarily stand in relations of time.”160

Kant argues that all appearances, inner as well as outer, are in time. The main premise of the argument is that all representations, as “determinations of the mind,” are in time. The term “appearance” Kant defines as referring to an “undetermined object of an empirical intuition.”161 Since empirical intuitions themselves constitute a subset of representations, one can contend that appearances constitute a subset of the objects of representations. Assuming that representations and their objects are identical with respect to their form, it can be stated that all objects of empirical intuitions, thus all appearances are temporal. But the argument is based on an unobvious assumption that there is a formal identity between representations and their objects. After all, it is not clear that representations and their objects can be temporal in the same sense, i.e. that temporality pertaining to the inner states of the mind is the same kind of temporality that pertains to outer objects represented in these states. (A supporter of Bergson’s theory of time, for example, would argue to the contrary.) It seems that the conclusion Kant wants to establish, namely, that both inner and outer phenomena conform to the same a priori form of sensibility, and hence belong to one and the same domain of empirical cognition, merely restates that disputable assumption. Furthermore, it is not obvious how to understand the claim that all representations are in time. The temporal flow of inner states, their ordering into past, present, and future, does not seem to be given in experience. This is the case because, strictly speaking, what one experiences is always, in a way, present.162 Even memories, whatever their content, are not experienced, as it were, in the past. It must be due to our observation of objects and states of affairs, different from our inner states, as changing that we become capable of understanding a variety of temporal determinations.

160  I. Kant, Critique of Pure Reason, op. cit., pp. 180-1 (A 34/B 50-1). 161  Ibidem, p. 172 (A 20/B 34). 162  “In the subjective time of human experience, only the present can be distinguished. The future is not yet there. The past has already gone. What exists is only the dynamic present, rich in the states of the subject, constantly emerging anew, whose contents become constituted by perceptions that constantly change.” M. Dzielski, “Newtona i Leibniza koncepcje czasu i przestrzeni” [“Newton and Leibniz on Space and Time”] in: G. Łuczkiewicz (ed.), Odrodzenie ducha – budowa wolności [A Spiritual Rebirth and Constructing Freedom], Krakowskie Towarzystwo Przemysłowe & Znak, Kraków 1995, p. 637 (translation A.T.).

58 

 Kant on Nonconceptual Content: Sensations and Intuitions

But let us apply Sellars’ distinction, which was to help explain the meaning of the term “representation” as employed by Kant, to the argument outlined above. The argument proceeds from a claim about “representings” to a claim about “representeds,” hence it makes use of the ambiguity characteristic of the notion of representation. Taken in the latter sense, it refers to the same kind of items as the term “appearance” does. As Sellars points out,163 Kant did not promote the absurd claim that empirical reality consists in a system of mental acts understood as occurrences in the consciousness of the (empirical) subject. But the Kantian notion of empirical reality cannot be construed in terms of contents that would correspond to mental states, either: empirical reality is not made of things as they are perceived, thought, etc., nor of things as they have been perceived, thought, etc. Rather, it consists of things which can, or could, be perceived, thought about, etc. Sellars refers to this kind of things as “representables,” a term that seems to express Kant’s intentions much more adequately than do the previous ones. The following are definitions and characteristics of representations encountered in Kant’s writings: 1. “[R]epraesentatio est determinatio mentis (interna), quatenus ad res quasdam ab ipsa (nempe repraesentatione) diversas refertur. It is that determination of the soul that is related to other things. But I call it related if its constitution is suitable to the constitution of the outer things, sive si rebus externis conformis est.”164 2. “We have representations in us, of which we can also become conscious. But let this consciousness reach as far and be as exact and precise as one wants, still there always remain only representations, i.e., inner determinations of our mind in this or that temporal relation.”165 3. “The genus is representation in general (repraesentatio). Under it stands the representation with consciousness (perceptio). A perception that refers to the subject as a modification of its state is a sensation (sensatio); an objective perception is a cognition (cognitio). The latter is either an intuition or a concept (intuitus vel conceptus). The former is immediately related to the object and is singular; the latter is mediate, by means of a mark, which can be common to several things. A concept is either an empirical or a pure concept, and the pure concept, insofar as it has its origin solely in the understanding (not in a pure image of sensibility), is called notio. A concept made up of notions, which goes beyond the possibility of experience, is an idea or a concept of reason.”166

163  W. Sellars, “Some Remarks on Kant’s Theory of Experience,” op. cit. 164  I. Kant, Notes and Fragments, edited by P. Guyer, translated by C. Bowman et al., Cambridge University Press, New York 2005, p. 34 (16:76-7). 165  I. Kant, Critique of Pure Reason, op. cit., p. 309 (A 197/B 242). 166  Ibidem, pp. 398-9 (A 319-20/B 376-7).



Representational Content Revisited 

 59

4. “[A]ll cognitions have, firstly, as cognitions, reference to an object, 2. to the cognition of the object. In the first case, it is a representation, in the second – consciousness is the general condition. Representation is intuition or concept; the cognition through concepts is thinking.”167 5. “Representation is either cognition, or sensation. The former – intuition or concept.”168 According to Kant, representations do not have to be conscious (cf. 2, 3), so he thinks that the notion of the mental and the notion of the conscious are not coextensive. In §5 of the Anthropology, one finds examples that, as Kant contends, show that the field of non-conscious representations extends much farther than the field of the conscious ones.169 On this account, consciousness is just a matter of attention and clarity: the more details in my perceptual field I follow and the more perspicuously they are presented to me, the more conscious I am. Here, only an empirical account of the non-conscious comes into play. But from the transcendental perspective all representations are of necessity at least potentially conscious.170 The reason why, in (2) and (3), thus in passages which come from the Critique of Pure Reason, it is suggested that we acknowledge the possibility of non-conscious representations, may come down to the fact that Kant occasionally switched the two perspectives, the transcendental and the empirical one. Kant seems to waver between two alternative conceptions of representation. Central to the first conception is reference to an object (cf. 1, 4). Importantly, representations must possess some characteristics in virtue of which they can correspond to the object they represent (at issue here is a structural rather than qualitative similarity).171 On the second conception, reference to an object is no longer

167  Kant, I., Gesammelte Schriften. Hrsg.: Bd. 1-22 Preussische Akademie der Wissenschaften, Bd. 23 Deutsche Akademie der Wissenschaften zu Berlin. Berlin 1900ff. AA 16:85.10-14 (translation A.T.). 168  Ibidem. AA 16:86.13-14 (translation A.T.). 169  Thus, there is an example with a man on the lea, seen at a distance, hence inconspicuously. We do not see his face clearly enough to discern all its elements, but since we know that we see a man, we infer that all the elements must be there. Kant says that, for that reason, we are indirectly aware of the man’s face. Other examples involve objects seen through a microscope or a telescope, and a musician who, while playing the organ, plunges into a conversation with a friend; surprisingly or not, this does not mar the beauty of his play. See: I. Kant, Anthropology, History, and Education, edited by G. Zöller and R. B. Louden, translated by M. Gregor et al., Cambridge University Press, Cambridge 2007, pp. 246-8 (7:135-137). 170  “The I think must be able to accompany all my representations; for otherwise something would be represented in me that could not be thought at all, which is as much as to say that the representation would either be impossible or else at least would be nothing for me.” I. Kant, Critique of Pure Reason, op. cit., p. 246 (B 132). 171  Cf. I. Kant, Notes and Fragments, op. cit., p. 35 (16:78).

60 

 Kant on Nonconceptual Content: Sensations and Intuitions

requisite for a mental state to be considered a representation (cf. 2, 3, 5). The above quotations give two hints at what might eventually turn out to provide a criterion for something to be a representation, no matter if the latter meets the requirement of objective reference or not. Accordingly, whatever qualifies as a “determination of the mind” would be regarded as a representation (cf. 3, 5). Alternatively, whatever possesses this characteristic along with temporal properties (more specifically, whatever stands in temporal relations to other determinations of the mind) – would also count as a representation (cf. 2). The first conception, in which representations are only “determinations of the mind,” encourages a Leibnizian interpretation of Kant’s views on the mind as a “representational engine.” For Leibniz, it is not the psychological aspect of representations that matters, but the informational one. This is perhaps why temporality does not have to be involved in the account of representation. The mind represents by its nature. Representing is dynamic, which means that it involves a continuous change of perceptions (Leibniz’s term) in a subject. The temporal aspect is not essential to the activity of representing. Rather, to put it somewhat metaphorically, time becomes generated along the way. However, there is much textual evidence in support of the claim that Kant was far from sympathetic to the ideas of Leibniz.172 Therefore, one should reject the first interpretative proposal. Instead, one should note that Kant endorsed the view on which representations are “determinations of the mind” that occur in temporal relations to one another.173 Underlying all representations, on that account, would be the form of intuition and a pure intuition – time. Time is not a derivative of representations; rather, it is constitutive of them. Since what needs foundations cannot supply the first building block for the construction of the edifice of knowledge, it follows that representations should not be construed as merely “given.”

3.2 Nonrepresentational Content: Sensations I have distinguished two conceptions of representation that can be found in Kant’s writings. One of them defines representations as determinations of the mind that occur in time and are temporally related to one another. The other, in turn, suggests that for a mental state to be recognized as a representation, it should refer to an object. Accordingly, the second conception excludes sensations from the category of mental representations, since sensations are defined as merely subjective modifications of the mind, as opposed to conscious objective representations, i.e. cognitions.

172  The main controversy concerned the status of what was later attributed the role of the forms of intuition, i.e. space and time. See, e.g., §14, 5 and §15 D in the Inaugural Dissertation. 173  Cf. the passage quoted before: I. Kant, Critique of Pure Reason, op. cit., pp. 180-1 (A 34/B 50-1).



Nonrepresentational Content: Sensations  

 61

Regarding the epistemological status of sensations: Kant’s account seems to offer two alternative options. The first one describes sensations as lacking objective reference in the proper sense of the term but it places them in the class of representations. The other one denies objective reference to sensations altogether, hence it also denies that they are representations. Before we take a look at these alternatives, let us briefly consider their historical background. No philosophical theory originates in complete isolation from other theories, let alone from their conceptual framework. This is also true of Kant’s philosophical theory, even if one should not overestimate the influence of others as long as they reach no further than into the matters of terminology. In her commentary on Kant’s transcendental psychology, in a chapter which addresses some issues related to spatial perception, Patricia Kitcher notices that “Eschenbach [a German translator of Berkeley – A.T.] introduced Johann Nicolaus Tetens to British philosophy, and in turn, Tetens’s work was closely studied by Kant.”174 Tetens was the author of a large volume entitled Philosophische Versuche über die menschliche Natur und ihre Entwicklung,175 a work that provides an overview of theories in what we can call philosophical psychology. Remarks on the nature of representations (über die Natur der Vorstellungen) open the first essay. Tetens acknowledges philosophers’ efforts to discover the common root or source of all the manifestations (Aeußerungen) of the activity (Kraft) of the mind. These manifestations divide into sensations (Empfindungen), representations (Vorstellungen), and thoughts (Gedanken). Accordingly, one can distinguish three kinds of activities of the mind: sensing, representing, and thinking. Tetens mentions two kinds of philosophical positions from which to answer the question concerning the “common root.” For philosophers who support the first position, i.e. sensationists, such as Helvetius, Condillac, Bonnet, and Search,176 all sorts of the mind’s activity are grounded in sensing (das Empfinden). In the opposite rationalist camp, one shall find philosophers, such as Leibniz and Wolff, who reduce all the activity of the mind to generating representations. According to the rationalists:

174  P. Kitcher, Kant’s Transcendental Psychology, Oxford University Press, Oxford – New York 1990, p. 34. Kitcher says that Kant’s readers were fairly well acquainted with the issue of spatial perception, in particular with the problem of perceiving objects at a distance (three-dimensionality). This is why, as she says, Kant is not very explicit, in the Transcendental Aesthetic, about one of his aims, that is, providing a solution to the problem of spatial perception (sometimes also referred to as Molyneux’s question). 175  That is, Philosophical Essays on Human Nature and its Development. All references in this chapter are to the original German edition of the book (Reuther & Reichard 1913). 176  J. N. Tetens, Philosophische Versuche über die menschliche Natur und ihre Entwickelung, Verlag von Reuther & Reichard, Berlin 1913, vol. 1, p. 4.

62 

 Kant on Nonconceptual Content: Sensations and Intuitions

“[T]he first basic activity would be that by means of which the soul makes its representations.”177

Tetens is moderately critical of both the sensationist and the rationalist positions. It is not clear, he contends, how to derive self-consciousness, or apperception, from mere sensing. On the other hand, explaining all kinds of mental activity in terms of representations runs the risk of blurring the essential distinctions between sensations and ideas. For Tetens, there are states of the mind that are not representations; they include sensations and mental occurrences such as joy, hope, and desire.178 (The latter are not representations in fact but, as the author claims, they can serve a means for representing other people’s feelings and mental states.) Representations, unlike sensations, correspond to the objects they were caused by and, in a way, present the features of their objects.179 Similar descriptions one can find in a number of Kant’s notes.180 Like Tetens, Kant criticized Leibniz’s theory of the mental. Rolf George suggested181 that one should consider Kant’s theory of experience a modified version of sensationism, i.e. as radically opposed to Leibniz’s representationism. On George’s reading, sensory states become downgraded to pre-cognitive mental occurrences. Consequently, Kant would sever the links between sensibility, the faculty that accounts for the fact that we have perceptions, and rational cognition. To a reader sympathetic to McDowell’s conceptualism, such a move would indicate relapse into the “Myth of the Given.” Although reconciling some versions of conceptualism with several key tenets of sensationism does not seem to fly in the face of reason,182 the final outcome must be the denial of the claim that sensations lay at the basis of cognition, and in particular perceptual experience. Rather, sensations, on this reading, leave the stage as (almost) cognitively irrelevant, as mere byproducts of the cognitive process, a residue of phenomenal consciousness. There are two more suggestions to consider: (i) we can admit that since objective reference does not provide a necessary requirement for a mental state to qualify as a representation, sensations are representations, albeit of a somewhat specific sort; or (ii) we can construe the nonrepresentational character of sensations in non-mentalist terms. In the latter case, sensations would not be representations, since they would be no psychological facts, no mental states at all.

177  Ibidem (translation A. T.). 178  Ibidem, p. 17. 179  Ibidem, p. 21. 180  See, e.g., I. Kant, Gesammelte Schriften, op. cit., AA 15:308; I. Kant, Notes and Fragments, op. cit., p. 34 (16:77); also Critique of Pure Reason, op. cit., p. 161 (A 28/B 44), where Kant gives an example of “the pleasant taste of a wine.” 181  R. George, “Kant’s Sensationism,” Synthese 47 (2), 1981, pp. 231-2. 182  Cf. K. Westphal, “Contemporary Epistemology: Kant, Hegel, McDowell,” European Journal of Philosophy 14 (2), 2006.



Nonrepresentational Content: Sensations  

 63

Why is it the case that sensations lack reference to objects? This is because, on Kant’s definition, sensations are no more than modifications of a subject’s mental states. Let us illustrate the point with an example of seeing a cube, on the one hand, and being in pain, on the other. To represent a cube means to perceive or imagine it as having a certain shape and colour, and perhaps texture; or it means to think that it has some shape, colour, etc. Objectivity enters my representation along with the fact that it is possible that I represent the cube the way it is. The content of my mental state is specified in that I am aware of the characteristics that a given object may actually possess. Contrastingly, the way one feels, e.g., pain would not tell one anything about the characteristics of any object other than one’s pain. It seems that, in this way, one could clarify Kant’s definition of sensations as subjective modifications of the mind. However, this leads to some oversimplifications. Surely, when one feels pain, one usually no longer has (cognitive) access to what has caused it; sometimes, getting to know the causes would require long examination that anyway might fail to inform one about the exact cause. But quite easily we distinguish between kinds of pains, let alone their qualities (a pain can be sharp, dull, throbbing, acute, chronic, and so on). Dressing wounds could not be urgently provided to bleeding patients if doctors, and other people, did not believe that this can alleviate the patients’ pains. In the argument I mentioned in the previous section, Kant assumes that all representations are in time. From this, of course, it does not follow that if a mental state is not a representation, it shall not be in time. But if it is in time, if it occupies a determinate “position” in a temporal sequence, it must conform to all the a priori determinations of the manifold of intuition. This is the case with all temporally structured occurrences, inner or outer, mental or bodily. Therefore, if sensations are in time, they must be related to other temporal occurrences determined, for example, by the category of cause. As one can read in §26, in the Transcendental Deduction of the Categories in the Critique of Pure Reason: “But in time, on which I ground the appearance as inner intuition, I represent necessary synthetic unity of the manifold, without which that relation could not be determinately given in an intuition (with regard to the temporal sequence). But now this synthetic unity, as the a priori condition under which I combine the manifold of an intuition in general, if I abstract from the constant form of my inner intuition, time, is the category of cause, through which, if I apply it to my sensibility, I determine everything that happens in time in general as far as its relation is concerned.”183

Thus, if one assumes that sensations occur in time, one shall also endorse the claim that they are causally determined. However, to form part of a causal chain means

183 I. Kant, Critique of Pure Reason, op. cit., pp. 262-3 (B 163).

64 

 Kant on Nonconceptual Content: Sensations and Intuitions

to be related to other members of the chain, hence to other elements within the boundaries of experience. Since subordination to the categories transforms a mental state into a proper object of judgment, to the extent that sensations could be regarded as parts of causal chains, one should also be able to formulate judgments about them. Linkages to objectivity would be guaranteed, for sensory occurrences, by the fact that judgments about them would belong to the experiential discourse. But are sensations temporal? In a short passage in the Transcendental Aesthetic, Kant gives examples of sensations. They include: tastes, such as “the pleasant taste of a wine,” colours, sounds, and warmth.184 The way he describes them suggests that Kant thought of sensations along the lines of the Lockean account of secondary qualities. About the taste of wine, for instance, he says that it belongs “to the particular constitution of sense in the subject that enjoys it,”185 rather than to the object itself. Crucial in Locke’s account is that, even if secondary qualities do not constitute real properties of the objects of experience, they maintain connections with the objects as the effects of the objects’ causal impacts on the sense organs of the subject of experience. Knowledge of the causal “powers” of objects and all essential facts about the constitution of sense organs should enable one to predict what kind of effects these impacts may bring about. (This does not imply that, on the basis of one’s knowledge of physical properties and laws, one should know everything about the nature of, say, experiences of colours.) Thus construed, sensations would belong to the realm of objects, with their subjective, phenomenal aspect remaining irreducible and ultimately unexplained. Noteworthy, though, several passages earlier Kant includes among sensations such qualities as impenetrability and hardness.186 This seems to invalidate the previous explanation of Kant’s account of sensations along the lines of Locke’s theory of secondary qualities, because the qualities mentioned by Kant would be considered, by Locke, as contingent upon primary qualities, the non-relational properties of the objects of experience. One may reply along the following lines: (i) either Kant shifts the boundary between primary and secondary qualities, so that the former may become associated with those features of the objects of experience that correspond to the a priori representations of the subject, and the latter may be associated with those features of the objects of experience that correspond to the a posteriori representations of the subject; or (ii) he does not conceive of sensations, as much as of the distinction between the subjective and the objective, in terms of the Lockean primary vs. secondary qualities distinction at all. Apparently, textual evidence provided above tells against the plausibility of the second supposition. Setting aside its plausibility, it seems to suggest, oddly enough,

184 Ibidem, pp. 160-1 (A 28/B 44). 185  Ibidem, p. 161 (A 28). 186  Ibidem, p. 156 (A 21/B 35).



Nonrepresentational Content: Sensations  

 65

that essential to Kant’s account of sensations would be neither the fact that they result from the impacts of objects on the subject’s sense organs, nor that they fail to represent objects or their properties. Let us now turn to the proposal on which sensations are neither temporal, nor representational in the sense that they do not refer to objects, because they are not mental states at all. It might be hard to accept this view and one cannot escape an impression that such an interpretation comes close to a misconstrual. But the objection is premature. Note that in the course of the current considerations we encountered a problem that undermines the coherence of Kant’s doctrine. The problem is that one cannot, within a unitary theoretical framework, think of sensations as both lacking objective reference and as manifesting a representational character. For, in the end, sensations turn out to have a kind of derivative objectivity; they refer to objects, as it were, through other representational states. Certainly, then, they are not entirely subjective. Furthermore, thinking of the Kantian sensations within the framework of “givenness,” or the notorious “Myth of the Given,” faces even more difficulties. This is because if we try to think of them as part of experience, the given turns out not to be what we have initially supposed it to be. As part of experience, it appears to be a theoretical “construction.” The difficulty may persist as long as we remain on one level of description, namely, on the phenomenological level. But one can dissolve it by recognizing the explanatory level of description and by making room in it for sensations. On this level, sensations would figure in the causal explanation Kant provides as part of his account of the genesis of experience187 and of there being qualitative, a posteriori factors involved in the experience of objects. As we read in §1 of the Transcendental Aesthetic: “The effect of an object on the capacity for representation, insofar as we are affected by it, is sensation […]. I call that in the appearance which corresponds to sensation its matter, but that which allows the manifold of appearance to be intuited as ordered in certain relations I call the form of appearance […] the matter of all appearance is only given to us a posteriori.”188

Appearances are undetermined objects of empirical intuition.189 These objects, Kant claims, result from the interplay of form and matter. Matter is that which corresponds to sensations, i.e. modifications of a subject’s mental state. The question is what this correspondence comes down to but, to my knowledge, Kant does not clarify the issue in a manner satisfactory enough. Assuming that Kant relates here the subjective with the objective, one may read his contention as establishing correspondence between

187  Cf. ibidem, p. 136 (B 1). 188  Ibidem, p. 155 (A 19-20/B 34). 189  Ibidem.

66 

 Kant on Nonconceptual Content: Sensations and Intuitions

these two domains. Matter, which would partly constitute the object of empirical intuition, would have its counterpart on the subjective side of the intuition, and this is what, in the above quoted passage, Kant calls sensation. We could adopt the Lockean reading again, and say that what accounts for sensations, on the part of the subject, are some properties of empirical objects that cause sensations. But then, why does Kant talk about a relation of correspondence rather than about a causal relation? The greenness of leaves I can see when looking out of my window is not, supposedly, what Kant means when he speaks about the matter of appearance. Being more on the objective side (it is the greenness of leaves out there, not of my inner states), it seems not to belong to sensation either. Perhaps a bit more elucidating on that matter would be the following remark in the foreword to Kant’s Anthropology: “He who ponders natural phenomena, […] can speculate back and forth (like Descartes) over the traces of impressions remaining in the brain, but in doing so he must admit that in this play of his representations he is a mere observer and must let nature run its course, for he does not know the cranial nerves and fibers, nor does he understand how to put them to use for his purposes. Therefore, all theoretical speculation about this is a pure waste of time.”190

The remark should be read against the background of Kant’s distinction between two kinds of anthropology: one dealing with the physiological aspect and the other one with the pragmatic aspect of the science of man. According to Lorne Falkenstein, this is one of the passages in which Kant suggests that a significant part of a theory of experience should be constituted by a physiological account of sense impressions which, however, as Kant makes it clear, goes beyond the scope of merely theoretical speculation. (I mentioned this issue briefly in chapter 2.) Thus, a philosopher must give way to a scientist. As Falkenstein says: “There is strong evidence […] that Kant did indeed take sensations, as ‘effects on the representative capacity’ arrayed in space as well as time, to be physiological states of the body of the perceiver. Following this line, it can be concluded that intuitions, which for Kant consist of sensations as matter, must, first of all, be effects on the subject and not objects the subject intends in so far as it is affected (since the sensations that make up the matter of intuition are effects on the subject and not intentional objects), and, second, be physiological effects (since space as well as time is a form of intuition). The distinction between the intuitive and the intellectual in Kant is properly to be seen as a distinction between the physiological and the psychic or, more exactly (if we do not want to presuppose dualism), between the physical and the cognitive.”191

190  I. Kant, Anthropology, History, and Education, op. cit., p. 231 (07:119). 191  L. Falkenstein, Kant’s Intuitionism. A Commentary on the Transcendental Aesthetic, University of Toronto Press, Toronto 2004, p. 123 (italics A.T.).



Nonconceptual Content: Intuitions 

 67

Again, let us note that an advocate of McDowell’s conceptualism could raise a number of objections to the idea that the above distinction runs along the boundary demarcated by Falkenstein. He could argue that the intuitive is conceptually shaped; thus, the boundary can indeed be fixed between the physiological and the psychic, or the physical and the cognitive, but neither the physiological, nor the physical equates, in Kant’s theory, as the conceptualist would take it, with the intuitive. The conceptualist might also be sympathetic to scientific explanations of cognitive phenomena, and recognize (though not without criticism) in Kant’s theory of sensations an attempt at incorporating scientific explanations into a theory of cognition.192 Figuring in a scientific theory, sensations would no longer be considered as psychological occurrences, but more as theoretical entities playing a role in the explanation of experience, from the point of view of physiology. But the advocate of conceptualism may insist on clearly separating a scientific from a rational inquiry, undertaken by epistemology, and he may criticize Kant for conflating the two domains.193

3.3 Nonconceptual Content: Intuitions Throughout his epistemological writings, Kant carefully distinguishes between two kinds of representations: concepts and intuitions. He adopts the distinction as early as in the pre-critical works. In the Inaugural Dissertation, for example, he calls the objects of the senses – sensible objects – phenomena, whereas those of the understanding he calls noumena. In the Introduction to the Transcendental Logic, in the Critique of Pure Reason, he formulates what we may call Kant’s Heterogeneity Thesis (KHT), in the following fashion: “If we will call the receptivity of our mind to receive representations insofar as it is affected in some way sensibility, then on the contrary the faculty for bringing forth representations itself, or the spontaneity of cognition, is the understanding. It comes along with our nature that intuition can never be other than sensible, i.e., that it contains only the way we are affected by objects.

192 The interest in scientific explanations of cognitive facts was common among modern philosophers, such as Descartes, Locke, or Berkeley. The analyses of how scientific and philosophical arguments intertwine in the account of cognition can be found in a number of monographs by John W. Yolton: Perceptual Acquaintance from Descartes to Reid, Basil Blackwell and University of Minnesota Press, London 1984; Perception and Reality: A History from Descartes to Kant, Cornell University Press, Ithaca 1996; and J. W. Yolton (ed.), Philosophy, Religion and Science in the 17th and 18th Centuries, University of Rochester Press, Rochester 1990. 193  In his lecture held at the University College Dublin, on 23rd April 2013, entitled “Can cognitive science determine epistemology?” McDowell insists that science itself relies on the conceptual framework that epistemology provides, rather than informing the way philosophical analysis should be carried out.

68 

 Kant on Nonconceptual Content: Sensations and Intuitions

The faculty for thinking of objects of sensible intuition, on the contrary, is the understanding. Neither of these properties is to be preferred to the other. Without sensibility no object would be given to us, and without understanding none would be thought. […] Further, these two faculties or capacities cannot exchange their functions. The understanding is not capable of intuiting anything, and the senses are not capable of thinking anything. Only from their unification can cognition arise. But on this account one must not mix up their roles, rather one has great cause to separate them carefully from each other and distinguish them. Hence we distinguish the science of the rules of sensibility in general, i.e., aesthetic, from the science of the rules of understanding in general, i.e. logic.”194

It is important to note that Kant attributes to sensibility and understanding two different kinds of representations. At least prima facie it is thus not the case that he understands representational content as “what is given by a ‘that’ clause in, for instance, an attribution of a belief,”195 as McDowell puts it. For Kant, representational content does not have to be rendered by a proposition, thus expressed in a language, in order to deserve its name. It does not have to be conceptually structured, since intuitions, the representations of sensibility, are not conceptual by definition. Therefore, at least prima facie, Kant is a nonconceptualist with regard to the representational content of experience. And one can find plenty of textual evidence for the claim that, throughout his writings, Kant considered concepts and intuitions heterogeneous representations. On the basis of this evidence, we can establish Kant’s Autonomy Thesis (KAT), which says that intuitions are concept-independent representations of objects. Thus, since concepts and intuitions are two different kinds of representational content, the subject can undergo content-laden representational states in which the only kind of representational content would be intuitions, regardless of the concepts that would correspond to these intuitions. The following text seems to confirm this contention: “For that objects of sensible intuition must accord with the formal conditions of sensibility that lie in the mind a priori is clear from the fact that otherwise they would not be objects for us; but that they must also accord with the conditions that the understanding requires for the synthetic unity of thinking is a conclusion that is not so easily seen. For appearances could after all be so constituted that the understanding would not find them in accord with the conditions of its unity […]. Appearances would nonetheless offer objects to our intuition, for intuition by no means requires the functions of thinking.”196

194  I. Kant, Critique of Pure Reason, op. cit., pp. 193-4 (A 51-2/B 75-6). 195  J. H. McDowell, Mind and World, op. cit., p. 3. 196  I. Kant, Critique of Pure Reason, op. cit., p. 223 (A 90-1/B 123; italics A.T.).



Nonconceptual Content: Intuitions 

 69

Accordingly, it does not seem to be necessary, for the subject to entertain representations of sensibility, that intuitions be accompanied by any concepts, be they empirical or pure. Objects can be intuited without the accompaniment of concepts simply because intuiting and thinking belong to two different faculties of the mind whose operations do not affect one another. As Kant remarks in the example with a “savage,” intuiting and thinking are two distinct “forms of knowledge;” thus the savage, who “sees a house in the distance the use of which he does not know,”197 does not need the concept of house to perceive a house. The concept of house is an empirical concept and one could argue that the Kantian savage, while lacking this concept, may still possess other concepts necessary to entertain content-laden perceptual states. But if intuiting and thinking are different functions of the mind to the extent that one does not require the other, then even if the savage had all the right concepts, the concept-independent content of intuition would have to be involved in his perception of a house. The point, thus, is not that sometimes, when we lack the right concepts, intuitions have to suffice in order to present us with objects, but that all cases of conscious perception of objects, no matter if our perceptions are accompanied by concepts or not, engage the concept-independent representations of sensibility. Against this background, what we have called Kant’s Cooperation Thesis (KCT) does not have to entail conceptualism, along the lines of McDowell. Importantly, it should be read as a thesis which says that the combination of concepts and intuitions is indispensable for cognition, rather than for entertaining representational states. KCT has been stated in the famous passage from the Introduction to the Transcendental Logic: “Without sensibility no object would be given to us, and without understanding none would be thought. Thoughts without content are empty, intuitions without concepts are blind. It is thus just as necessary to make the mind’s concepts sensible (i.e., to add an object to them in intuition) as it is to make its intuitions understandable (i.e., to bring them under concepts).”198

KCT does not entail conceptualism if it is combined with what I called Kant’s Autonomy Thesis (KAT), and KAT can be corroborated by textual evidence for the claim that intuitions are generically distinct from concepts. Thus, in Kant’s view, intuitions are immediate and singular representations of objects,199 whereas concepts are mediate

197  Kant’s Introduction to Logic and His Essay on the Mistaken Subtlety of the Four Figures, translated by T. K. Abbott, Thoemmes Press, Bristol 1992, p. 24. 198  I. Kant, Critique of Pure Reason, pp. 193-4 (A 51/B 75). 199  In Jaakko Hintikka’s interpretation, immediacy is no genuine characteristic of intuitions and can be reduced to singularity. Hintikka states: “According to [Kant’s] definition, presented in the first paragraph of his lectures on logic, every particular idea as distinguished from general concepts is an intuition. There is, we may say, nothing ‘intuitive’ about intuitions so defined. Intuitivity means

70 

 Kant on Nonconceptual Content: Sensations and Intuitions

and general. From the ontological perspective, intuitions have a mereological structure, whereas the structure of concepts is set-theoretic. This means that the basic relation by means of which to describe an intuition is a relation between parts and wholes. In the case of concepts, the basic relation entailed by their structure is that of the subsumption of a particular object under a determinate class of objects. Further, to distinguish intuitions from concepts, Kant also formulates, in the Jäsche Logic, what looks like a version of the richness argument: “Since only individual things, or individuals, are thoroughly determinate, there can be thoroughly determinate cognitions only as intuitions, but not as concepts; in regard to the latter, logical determination can never be regarded as completed.”200

Kant’s remark should be read against the background of Leibniz’s theory of substance. Leibniz held that each particular entity has a “complete concept” which encapsulates all of its characteristics. Such a concept can be analyzed into indefinitely (or infinitely?) many other concepts, since an individual substance could be attributed indefinitely (or infinitely?) many predicates. But according to Kant concepts do not determine particular entities throughout since, unlike intuitions, they do not refer immediately to objects; rather, they refer to classes of objects by virtue of marks common to the objects. Since only particulars are thoroughly determined and since only intuitions immediately refer to particulars, concepts cannot capture the complexity of particular objects. In contrast to Leibniz, Kant also emphasizes the sensible determinateness of empirical objects. Thus, he thinks that the sensible determinateness of objects remarkably outstrips their conceptual determinateness. On such a construal, Kant indeed phrases a version of the richness argument. However, Kant’s nonconceptualism may become problematic when we reflect on Kant’s theory of intentionality. As Derk Pereboom points out,201 on Kant’s theory

simply individuality.” See: J. Hintikka, “Kant on the Mathematical Method” in: Knowledge and the Known. Historical Perspectives in Epistemology, D. Reidel Publishing Co., Dordrecht 1974, p. 162. Kirk D. Wilson, by contrast, admits that singularity equals immediacy when it comes to reference, but he denies that they are also identical with regard to their senses. According to Wilson, singularity should be explained in terms of a specific mereological structure characteristic of intuitions. Since intuitions structurally differ from concepts, also their mode of reference to objects must be different. In other words, intuitions represent objects immediately because they are singular representations: their structure maps the structure of an individual object (spatiotemporality). Concepts do not represent individual objects immediately because their structure maps the structure of relations of a property that can be predicated of a number of objects. Concepts are abstract entities that, according to Kant, can be related to individual objects by means of other representations, more specifically, schemata. See: K. D. Wilson, “Kant on Intuition,” Philosophical Quarterly 25 (100), 1975. 200  I. Kant, Lectures on Logic, translated and edited by J. M. Young, Cambridge University Press, New York 1992, p. 597 (9:99). 201  D. Pereboom, “Kant on Intentionality,” Synthese 77 (3), 1988.



Nonconceptual Content: Intuitions 

 71

of representations, two characteristics can be attributed to intentional states: firstly, they are existence-independent – in that they are directed at objects regardless of whether these objects remain in causal relations to the representational states – and, secondly, they are concept-dependent, which means that their relatedness to an object is contingent upon the concepts possessed by the subject. Thus, one can be intentionally related to – let us say – the morning star, rather than to the evening star if one has the concept of the former but not the concept of the latter and does not know that the morning star and the evening star are one and the same object. Therefore, if perception is an intentional representational state, it must be concept-dependent. Concept-independent states would have to be nonrepresentational, then, that is, mere sensations (if anything). In this way, we are back at the position that can be called sensationist conceptualism: the position that endorses the nonrepresentational sensory content and claims that all representational content is concept-dependent. On such a view, the Kantian intuitions, as Sellars remarked, may be characterized in a twofold way: either as conceptualized singular representations of objects, or as the nonrepresentational content of experience. Another objection to reading Kant’s account of intuition in the nonconceptualist terms may build on the doctrine of threefold syntheses presented in the first-edition version of the Transcendental Deduction of the Categories in the Critique of Pure Reason. I will deal with this argument in more detail in chapter 4; now it suffices to note that the doctrine establishes that concepts are indispensable not only for making judgments, that is, for fully-fledged cognitions, but also for synthesizing the manifolds of intuition into representations of objects. In the latter sense, concepts play the role of rules, which Béatrice Longuenesse characterizes in this way: “The concept is a rule insofar as it is the consciousness of the unity of an act of sensible synthesis or the consciousness of the procedure for generating a sensible intuition. The first sense of rule anticipates what Kant, in the Schematism of the Pure Concepts of the Understanding, calls a schema. But the concept is a rule also in another, discursive sense. It is a rule in that thinking an object under a concept provides a reason to predicate of this object the marks that define the concept.”202

These two ways of understanding concepts as rules are complementary and can be traced back to the fact that, prior to their application to objects, concepts must be generated; both processes are governed by rules, that is concepts (though it seems that such an account runs the risk of falling either into circularity, or the infinite regress). Longuenesse says:

202  B. Longuenesse, Kant and the Capacity to Judge: Sensibility and Discursivity in the Transcendental Analytic of the ‘Critique of Pure Reason’, Princeton University Press, Princeton 1998, p. 50.

72 

 Kant on Nonconceptual Content: Sensations and Intuitions

“Because one has generated a schema, one can obtain a discursive rule by reflection and apply this rule to appearances.”203

To illustrate the point about concepts as rules for “generating” representations: when I draw a geometrical figure, or imagine a property of an empirical object, such as its shape, this procedure is guided by the right kind of concepts. Without the rule that I follow when drawing a figure or when imagining an object, these procedures could not be carried out – they would not result in singular representations of objects. This is because such representations are not given to us as the effects of causal impacts of objects on the senses, as it is the case in, for example, Locke’s theory, but have to be arrived at by way of synthesizing nonrepresentational manifolds, that is, through spontaneous operations of the mind. In order to address the above objections, we may point out that, first, Pereboom seems to conflate intentionality with intensionality; second, Kant indeed lacks an account of non-discursive spontaneity being available to finite subjects, although his theory of the forms of intuition seems to leave such a possibility open; and, third, it is by no means obvious that, according to Kant, all intuitions must be generated by rulegoverned syntheses, the pure intuitions of space and time being an exception. I deal with the last two replies in the rest of this chapter and in chapter 4; here, I briefly comment on the first one. Intensional contexts are contexts in which the reference of an expression cannot be regarded as the function of the reference of its constituents; in particular, the truth-value of a proposition cannot be regarded as a function of the truth-values of its constituents. In such contexts, verbs like “believe,” “think,” “regret,” “suppose,” etc., appear. Thus, for example, whereas it can be true that S believes that the morning star can be seen at dawn at such and such a position on the horizon, it does not have to be true that S believes that the evening star can be seen at the same position, both names referring to the planet Venus, because S does not know that the morning star and the evening star are the same object, or because S does not even know what the expression “the evening star” refers to. Now, we may agree that perception is intentional, that is an object-directed conscious state of the mind, that all perceptual states have an object, regardless of whether their object exists or not, but it does not seem to directly follow from this that the verb “perceive” functions analogously to the verbs “believe” or “think.” For we may say that S has seen the evening star, even though S does not know such a star himself; as much as we can say that a cyclist has noticed bumps on the road and adjusted his riding to the road conditions despite the fact that, when asked about what and how many bumps he encountered along the way, he would probably be unable to give a precise answer. Therefore, Pereboom’s argument seems to assume what is otherwise in need of justification, namely that the intentionality of perception implies that verbs

203  Ibidem.



The Forms of Intuition 

 73

describing perception introduce intensional contexts as a rule, which seems to be quite far from obvious.

3.4 The Forms of Intuition Cassirer, in his biography of Kant, features eighteenth-century Königsberg as a stronghold of the Aristotelian and Scholastic tradition,204 a fact which was reflected in the educational programmes of the Prussian gymnasia and universities of those times. The benefits Kant derived from his studies at Collegium Fridericianum would therefore amount, allegedly, to the mastery of Latin and classical authors. The need for knowledge of natural sciences he would have to satisfy by means of self-study and attending the lectures of Martin Knutzen, who would introduce his students to the scientific ideas of Newton and the Cartesians.205 The young Kant dedicated his efforts to studying physics and natural geography, a fact confirmed by his juvenile publications. Scientifically-oriented as he was, Kant must have taken perhaps less interest in metaphysics and the subtle teachings of the “Schools.” But his philosophical theories are by no means free from the influence of the traditional distinctions, as one can see from the Kantian terminology. One of these distinctions, the origin of which goes back to Aristotle, is Kant’s distinction between form and matter. It appears both in the Inaugural Dissertation and in the Critique of Pure Reason. Thus, in the Dissertation, one can read: “In a representation of sense there is, first of all, something which you might call the matter, namely, the sensation, and there is also something which may be called the form, the aspect namely of sensible objects which arises according as the various objects which affect the senses are co-ordinated by a certain natural law of the mind.”206

In all representations of the senses there are two components: matter and form. Kant identifies the former as that which corresponds to sensations and the latter as what arises due to the fact that things affecting the senses are coordinated “by a certain natural law of the mind.” It is not clear by means of what factor the unity of form and matter is brought about. It is pretty much clear, though, that form emerges on account of two factors: the

204 Ernst Cassirer, Kant’s Life and Thought, translated by J. Haden, Yale University Press, New Haven and London 1981. Hermann J. de Vleeschauwer (in The Development of Kantian Thought. The History of a Doctrine, translated by A. R. C. Duncan, Thomas Nelson and Sons, Edinburgh 1962) offers a slightly different outlook, on which Königsberg in Kant’s times was under the influence of the main intellectual currents of the eighteenth-century Europe. 205 E. Cassirer, Kant’s Life and Thought, op. cit., pp. 12-57. 206 I. Kant, Theoretical Philosophy, 1755-1770, translated and edited by D. Walford in collaboration with R. Meerbote, Cambridge University Press, New York 2007, p. 384 (2:392-3).

74 

 Kant on Nonconceptual Content: Sensations and Intuitions

affection of the senses by objects and an innate disposition of the mind to impose a determinate structure on the result of affection. Since form pertains to representations of the senses as a result of the coordinating activity of the mind, it cannot derive from the objects affecting the senses. The following passage confirms this conclusion: “[…] also the form of the same representation is undoubtedly evidence of a certain reference or relation in what is sensed, though properly speaking it is not an outline or any kind of schema of the object, but only a certain law, which is inherent in the mind and by means of which it coordinates for itself that which is sensed from the presence of the object. For objects do not strike the senses in virtue of their form or aspect. Accordingly, if the various factors in an object which affect the sense are to coalesce into some representational whole there is needed an internal principle in the mind […].”207

But the claim that objects do not affect the senses, so as to “transmit” their “form or aspect” to the mind, should not be accepted without justification. The view on which forms (species) inhere in the objects of cognition from which they become “transmitted” to the mind, can be traced back to Aristotle and the Scholastics. Early modern philosophers, such as Descartes, Malebranche, and Locke, would raise a number of objections against this view. Kant’s theory combines both ancient and modern ways of thinking about the objects of sensibility. First of all, his account seems to be based on the assumption that the senses do not present objects as they really are, which, clearly enough, is a skeptical assumption. Kant’s reasoning could be outlined as follows: if there is something the senses present to us, sensible representations, and if one cannot preclude the possibility, albeit theoretical, that these representations do not relate to objects that could be cognized otherwise than by means of sense experience, then one may claim that the senses present to us a specific kind of object. Second, Kant’s view is based on the distinction, accepted by the ancients, between the objects of sensibility and the objects of understanding. If sensibility presents things as they appear and the understanding presents things as they really are, and, moreover, if the latter does not get involved into the operations of the senses, then the objects of both faculties must differ in kind. Both the skeptical assumption and the dualism of cognitive faculties and their products can be considered as underlying Kant’s claim that the forms of sense representations originate from the subject’s way of perceiving, rather than from the constitution of objects themselves. However, if one maintains the dualism of sensibility and understanding and rejects the skeptical assumption, one will not arrive at the Kantian conclusion, but rather at the Aristotelian one.208

207  Ibidem, p. 385 (2:393). 208 In De Anima (On the Soul) III, 8, 432a, Aristotle distinguishes the objects of the senses and the objects of the intellect. Each kind of objects has its own specific kind of form. In cognition, the mind “becomes” its object as it “takes in” the form of the object.



The Forms of Intuition 

 75

Thus, the objects of cognition, according to Kant, can be considered the “results” of the combination of form and matter, form being provided by the subject and matter being derived from objects “outside” the mind. The form would inhere “in” the mind, whilst the object would deliver matter. But these metaphors may be fraught with difficulties: for, to take the first one, what does “in” mean in this case? Are the forms intentional objects of consciousness? Or perhaps they are only ways that make it possible for the subject to represent objects in intuition. By regarding them as intentional objects, we would postulate a kind of cognitive access to them. But if one thinks of the forms as ways in which objects are experienced, it may seem unclear why one should also think of them as constitutive of the objects of cognition, rather than of the representations of objects only. Here again the question about how Kant understands the notion of representation recurs. Furthermore, Kant’s account can be charged with some inconsistency. The inconsistency would be due to maintaining both that the objects of representational acts are unities of form and matter, and that also the components of representations, i.e. their form and matter, can be represented separately. In particular, the claim that the condition of representation can be represented, or made into an object of representation, may lead to difficulties. For, should the condition of representation itself be represented, there would have to be another condition for representing it. This leads to an infinite regress. Conversely, supposing that form and matter cannot be represented separately, how can one distinguish them in the object of representation? Although the former view seems to involve more difficulties, such as the abovementioned regress, in the Transcendental Aesthetic Kant approximates the forms-asrepresentations thesis.209 This he does in several steps, which can be outlined in the following way.

209  Interestingly, also from the psychological perspective this reading turns out to be more plausible. Quite commonly, psychologists agree that representing different properties of an individual’s surroundings presupposes a development of certain cognitive mechanisms; thus, e.g., perceiving objects (spatiotemporal physical particulars) should not be “taken for granted,” as a capacity given at the very start of one’s cognitive life. Apparently, though, both the evolutionary – “Piagetian” – approach and the nativist one are giving way to a more integrated and “moderate” view on human cognitive performance, in which innate capacities, dispositions or mechanisms, adaptation to the environment and exposure to socio-cultural influences play a due role. Such an “integrated” approach one can find in: Nora S. Newcomb, Janellen Huttenlocher, Making Space. The Development of Spatial Representation and Reasoning, The MIT Press, Cambridge MA 2000.

76 

 Kant on Nonconceptual Content: Sensations and Intuitions

3.4.1 Forms as a priori Intuitions The argument begins with a definition of pure representations as representations “in which nothing is to be encountered that belongs to sensation,”210 thus as that which has no empirical component. Accordingly, pure representations must also be “encountered in the mind a priori,”211 rather than deriving from objects. Since it is the forms of sensible intuition that are encountered a priori in the mind, pure representations and the forms of sensible intuition must be identical: “this pure form of sensibility itself is also called pure intuition.”212 However, the conclusion does not seem to be sufficiently warranted. From the premises that: (i) pure representations must be a priori and (ii) the forms of intuition lie a priori in the mind, one should not conclude that (iii) the forms of intuition are a priori representations, unless one assumes that (i’) the forms can be represented and that (ii’) there are no other things in the mind, apart from representations, which can be a priori. But perhaps there can be things in the mind which would be a priori but which we would hesitate to identify with representations, things such as, e.g., ways of cognizing or justifying certain truths. We do not need to appeal to experience to acquire justifications for geometrical truths, hence cognition of geometrical truths is a priori. But whether there are any a priori representations corresponding to the ability or disposition to cognize things independently of experience depends, in this case, on whether one can acquire a priori representations of what geometrical truths are about. From the fact that we have a cognitive capacity that can operate independently of experience (geometrical proofs can be valid regardless of whether any geometrical objects really exist), it does not follow that there are objects that we may cognize in this way.

3.4.2 The Separation Argument In the Transcendental Aesthetic Kant presents a thought experiment, to the following effect: “So, if I separate from the representation of a body that which the understanding thinks about it, such as substance, force, divisibility, etc., as well as that which belongs to sensation, such as impenetrability, hardness, color, etc., something from this empirical intuition is still left for me, namely extension and form. These belong to the pure intuition, which occurs a priori, even without an actual object of the senses or sensation, as a mere form of sensibility in the mind.”213

210  I. Kant, Critique of Pure Reason, op. cit., p. 156 (A 20/B 34). 211  Ibidem. 212  Ibidem. 213  Ibidem.



The Forms of Intuition 

 77

Let us comment briefly on the above quoted passage. 1. The “experiment” purports to show that after removing from a representation the sensational and the conceptual components one would arrive at a residue which Kant identifies as “extension and form” and puts on the side of “the pure intuition.” There is some phenomenological flavour to the argument: Kant begins with experience only then to suggest that we represent the experienced object in imagination. It is hard to establish, though, whether the “separation experiment” proves that subjects are capable of entertaining representations of the pure forms of intuition, or only that they cannot entertain a representation without also representing the condition that makes the representation possible. 2. Kant’s separation argument is supposed to show that forms cannot be reduced to any simpler or more basic kind of content; that they cannot be “constructed” out of any sort of original “input.” But the separation argument also illustrates Kant’s belief about the nonconceptual character of the forms of intuition; for it shows that a sense representation must contain an a priori component which cannot be contributed by the understanding (clearly, Kant says that when carrying out the experiment, one would have to take away, from a representation of an object, “that which the understanding thinks about it,” which could be understood as a demand, in the experiment, to imaginatively remove all contribution from the understanding, including the categories).

3.4.3 The Objects of Pure Intuition In the second part of the A 25 passage of the Transcendental Aesthetic, Kant claims that the a priori intuition of space underlies all spatial concepts. He also argues that from the a priori intuition of space cognitions of a special kind can be derived. These are the cognitions of geometry, and, according to Kant, the underlying pure intuition of space guarantees their apodictic certainty: “Thus also all geometrical principles, e.g., that in a triangle two sides together are always greater than the third, are never derived from general concepts of line and triangle, but rather are derived from intuition and indeed derived a priori with apodictic certainty.”214

Later, Kant restates the claim saying that space and time constitute two sources of synthetic a priori cognitions: “Time and space are accordingly two sources of cognition, from which different synthetic cognitions can be drawn a priori, of which especially pure mathematics in regard to the cognitions of

214  Ibidem, p. 159 (A 25).

78 

 Kant on Nonconceptual Content: Sensations and Intuitions

space and its relations provides a splendid example. […] [Time and space] thereby make possible synthetic a priori propositions.”215

Arguably, we can think of synthetic cognitions at least in two complementary ways: (1) as cognitions that result in synthetic judgments whose justification requires an appeal to intuition and (2) as cognitions that involve synthesis, a function of the mind performed in accordance with the rules of the understanding. Now, the question is, first, what the “appeal to intuition” comes down to in the case of synthetic and a priori cognitions; and, second, to what rules of the understanding cognitions originating from the a priori intuitions of space and time owe their synthetic character. Let us start with the second question. The rules required are furnished by the categories,216 more precisely: a special class of them. In the “Clue to the Discovery of All Pure Concepts of the Understanding,” Kant divides the categories into mathematical and dynamical. Whereas the mathematical categories refer to “objects of intuition (pure as well as empirical),”217 the dynamical categories are directed “at the existence of those objects (either in relation to each other or to the understanding).”218 This seems to suggest that there must be objects, given both in pure and empirical intuition, which the categories apply to. Does Kant thereby intend to say that the mathematical categories apply to mathematical objects? Some commentators deny that Kant was committed to the existence of the proper objects of mathematics. Michael Friedman, for example, says: “[P]ure mathematics is not a body of truths with its own peculiar subject matter for Kant. There are no ‘mathematical objects’ to constitute this subject matter, for the sensible and perceptible objects of the empirical world (that is, ‘appearances’) are the only ‘objects’ there are. For this reason, pure mathematics is not properly speaking a body of knowledge (cognition).”219

Indeed, the synthetic a priori judgments of mathematics deal with the formal conditions of knowledge, rather than with physical objects in space and time. These judgments

215  Ibidem, p. 166 (A 39/B 56). 216  The categories, and all derivative concepts, provide rules according to which the objects of mathematical judgments become represented. In order to represent a straight line, I must have the concept of unity, otherwise my representation would be good for nothing, since it would be produced randomly. Cf. J. Kim, “Concepts and Intuitions in Kant’s Philosophy of Geometry,” Kant Studien, 97 (2), 2006. On Kim’s interpretation, necessity originates from concepts (the understanding), and syntheticity – from the concepts’ ultimate deducibility from the categories. I think, however, that the author quite underestimates the contribution of intuition in grounding the truth of mathematical judgments; concepts do indeed determine intuitions but determination is not the same as production: what becomes determined is essentially heterogeneous from the determining factor, not a product thereof. 217  I. Kant, Critique of Pure Reason, op. cit., p. 215 (B 110). 218  Ibidem. 219 M. Friedman, “Kant’s Theory of Geometry,” The Philosophical Review, 94 (4), 1985, p. 504.



The Forms of Intuition 

 79

cannot, strictly speaking, be about objects, since otherwise they would not be a priori and would not qualify as cognitions that “precede” objects. Besides, space and time are no objects in the proper sense of the term. But, on the other hand, what do we refer to when we say, for example, that the sum of all the angles in a triangle equals two right angles? Furthermore, Kant holds that mathematics differs from philosophy in that the former constructs its concepts by providing them with corresponding intuitions. Philosophy undertakes no more than conceptual analysis.220 What is it then that a mathematician constructs in intuition if not an object of mathematics, or its instantiation? Now, let us turn to the first question raised above: what does recourse to intuition in the case of synthetic and a priori cognitions come down to? We can address this question by appealing to the doctrine of transcendental imagination and schematism. The transcendental imagination, on Kant’s account, is not a power to produce or reproduce images (a task of the empirical imagination), but a faculty that produces schemata, i.e. “representation[s] of a general procedure of the imagination for providing a concept with its image.”221 Importantly, “schematizing” mathematical (or, more specifically, geometrical) concepts does not amount to providing empirical images for them; one cannot, as it were, draw a schema on a piece of paper. As Kant says in the schematism chapter: “In fact it is not images of objects but schemata that ground our pure sensible concepts. No image of a triangle would ever be adequate to the concept of it. For it would not attain the generality of the concept, which makes this valid for all triangles, right or acute, etc., but would always be limited to one part of this sphere. The schema of the triangle can never exist anywhere except in thought, and signifies a rule of the synthesis of the imagination with regard to pure shapes in space.”222

Images are not candidates for representing a triangle as the subject matter of a geometrical proof. Pace Berkeley, no “picture in the head” can be suitable for representing a general property, at least as long as one regards images as particular, hence completely determined, representations. Denying the role, in mathematical reasoning, to images may imply that no specific kind of mental content is relevant to this kind of reasoning. Mathematical reasoning is not an operation on mental contents; one needs no images to do mathematics. On such an anti-mentalist reading of Kant’s theory of pure intuition, one of the defining features of intuitions, i.e. immediacy, could be explained in the following way.

220 “Philosophical cognition is rational cognition from concepts, mathematical cognition that from the construction of concepts. But to construct a concept means to exhibit a priori the intuition corresponding to it.” I. Kant, Critique of Pure Reason, op. cit., p. 630 (A 713/B 741). 221 Ibidem, p. 273 (B 180). 222 Ibidem, p. 273 (A 140-1).

80 

 Kant on Nonconceptual Content: Sensations and Intuitions

The idea that intuitions are immediate representations invokes the “given,” something present to the mind in its own right. Since mathematical objects, and other objects of intuition, do not seem to fall on the side of the given (can we understand what it would be to immediately represent a figure of one thousand angles?), what remains is our ability or disposition to employ the “rules of synthesis,” or the awareness of our capacity to engage in mathematical reasoning and arrive at necessarily true conclusions. Intuition could then be taken not as the source of specific objects of cognition but rather as a capacity by means of which to become aware of how to apply our mathematical concepts. Imagining a triangle would thus be reducible to actualizing a set of cognitive powers, such as the ability to prove certain theorems, rather than to perceiving an image “in one’s head.” The idea to think of the recourse to pure intuition in terms of the ability to correctly apply rules, in abstraction from the content of cognition, may seem appealing. It is reminiscent of theories in which subjects are attributed knowledge of structures or rules, which comes prior to experience. Chomsky’s theory of generative grammar provides a good example of such a theory. Also, the idea captures the common intuition that there is something ultimately unexplainable in the ability to successfully engage in mathematical reasonings and – I would add, following Kant – the ability to apply concepts to objects, general and discursive to singular representations (in the Critique of Pure Reason, Kant calls it “the power of judgment,” or “mother-wit” (Mutterwitz)223). But this idea does not allow for a satisfactory explanation of the fact that Kant equated pure intuitions with the forms of intuition and regarded them, already in his pre-critical writings, as a sort of the “given” unfit for further analysis. (I briefly touched upon this issue in the Introduction to this book.)

3.5 The Ambiguity of “Perceiving” When searching for arguments for the distinction between perception and thought, it seems fairly natural to consider the ways in which the distinction is reflected by language. For example, we say that we can see colours or coloured objects, hear sounds or melodies, feel warmth or the smoothness of a surface, but also that we are thinking about John or about yesterday’s meeting. When we say that we have heard about something, we do not mean that we have had an auditory experience of that thing but that we have got to know a fact about it. Indeed, in the way we use our language, we introduce the distinction between intellectual acts and those occurrences that engage our sensibility. The former are about objects; the latter seem to involve objects more directly. The object of thought is its subject matter, as is the object of a novel or a piece of music. In the act of seeing, hearing, touching, etc., unlike in the act of thinking, the

223 Ibidem, p. 268 (A 133/B 172).



The Ambiguity of “Perceiving” 

 81

object itself seems to be directly present. Whereas thought is a vehicle for making an object present to the mind, or recalling it, in perception an object is already there. This is not to deny that the immediacy with which objects appear in perception comes in degrees: when one touches the wooden back of a chair, the entire object is not present to one. Language reflects some intricacies of our thinking about perception but we do not have to rely on the arguments from the use of language alone and accept them without qualifications. Perhaps more in-depth reflection would reveal gross inconsistencies in the way we employ the terms that name perceptual acts and their objects. For how are we to explain the meaning of “perceive?” At first this seems to be a trivial question; perhaps answering it would not yield anything more than a list of philosophical platitudes. Surely, to perceive means to see, hear, touch, smell, and taste whatever object is proper to a particular sense modality. But a simple example suffices to undermine our confidence in our, so to say, “natural expertise” in the matters of perception. Let us imagine you ask me about the person I saw yesterday and I reply that it was Tim. Carrying on, you ask me about the colour of the tie Tim was wearing (suppose Tim wears a tie on every weekday). Now I may not know how to answer your question. Would I still, in that case, be right to claim that I saw Tim the day before? Most of us, I think, would be inclined to answer in the positive: indeed, I might have seen Tim, although I did not notice all the details of his garment. A philosopher, however, may ask new questions. The person I saw, he would say, was certainly wearing a tie coloured in a determinate way, so if I saw Tim, I must have seen his tie, even though I did not perhaps notice it. Now, the question is again: what do we understand by “seeing” (or, more broadly, “perceiving”)? Can one see a thing without noticing it? Two alternative meanings of “seeing” (or “perceiving”) could thus be established: (1) the doxastic or the epistemic sense and (2) the non-doxastic or the non-epistemic sense. Thus, I could say that I saw Tim’s tie in the second sense of “seeing” but I did not see Tim’s tie in the first of the senses of the term. Both replies rely on quite different theories of sense perception. When conceiving of perception as an epistemic process, we assimilate it to intellectual acts, such as thinking or judging. To perceive, in this sense, the subject must satisfy a number of criteria; analogously, to possess a concept, the subject must meet some epistemic requirements, such as knowledge of the concept’s definition, or at least the ability to subsume some particulars under the concept. By contrast, conceiving of perception in non-epistemic terms makes it possible to bring out its specific character and independence from higher cognitive processes. Thus, Fred Dretske distinguishes what he calls “simple seeing” from “seeing that thus-and-so is the case” about an object. According to Dretske, seeing an object does not have to entail a belief about that object. Forming a belief or judgment is not constitutive of a conscious experience: one can see an x without seeing that the x is F. (Of course, seeing that an x is F, in normal cases excluding blindsight, presupposes

82 

 Kant on Nonconceptual Content: Sensations and Intuitions

seeing the x.) As one can read in Simple Seeing: “Seeing x, just like stepping on x, is compatible with no beliefs about x.”224 This is because “people are not authorities about what they see.”225 For instance, I may see a robin without ever realizing I have seen one if I have no concept of robin, and hence cannot form any belief about robins. Elsewhere, Dretske distinguishes between thing-awareness and fact-awareness.226 By seeing things, their properties, and the relations they enter, we get to know facts about these things. But we do not have to know facts about things in order to be conscious of these things in experience. All in all, along the lines of Dretske’s proposal, my seeing Tim without noticing what colour his tie was provides an instance of simple seeing or thing-awareness without an accompanying belief. One could object, though, that the idea of conscious experience, developed by Dretske, falls short of being adequate. If I cannot even recollect that Tim was wearing a tie of some sort, coloured in a more or less familiar way, what justifies my claim that I saw Tim’s tie? Perhaps someone might insist on my having seen it, but after all it is up to me to confirm the fact. No one can persuade me into believing that I have experienced something I deny having experienced; the conscious subject I am is not a device registering alterations in the environment. Perhaps we should only admit that sometimes we fail to perceive things correctly, or that perceiving comes in degrees. To recall Leibniz: perceiving would differ from the acts of the understanding by the degree of clarity and distinctness it is accompanied by.227 Kant seems to capture Dretske’s distinction in his “savage” example (presented in chapter 1). The “savage” represents a house non-epistemically, since an epistemically valuable representation would have to involve both intuitions and concepts. Hence, we cannot ascribe empirical cognition to him despite his being perceptually related

224 F. Dretske, “Simple Seeing” in: Perception, Knowledge, and Belief. Selected Essays, Cambridge University Press, Cambridge 2000, p. 101. 225 Ibidem, p. 102. 226 Cf. F. Dretske, “Conscious Experience,” Mind 102 (406), 1993; F. Dretske, “What We See. The Texture of Conscious Experience” in: B. Nanay (ed.), Perceiving the World, Oxford University Press, Oxford-New York 2010. 227 One should not confound Dretske’s account of non-doxastic perception with accounts appealing to cases in which there is a conflict between the content of perception and the content of belief, irrespective of apparent similarities between the two accounts. In the Müller-Lyer illusion, the subject can see two equal lines as of different lengths and, if the subject knows the phenomenon, he would not tend to frame a belief that one of the lines is longer. Along the lines of Dretske’s proposal, what the subject of the illusion non-doxastically sees are two lines of equal length. But in order to adequately account for what happens in cases like the Müller-Lyer illusion, one needs to distinguish between the content of perception, which can be either conceptual or nonconceptual, and the belief or judgment which consists in either endorsing or rejecting the content of perception. Cf., e.g., G. N. A. Vesey, “Seeing and Seeing As” in: R. J. Swartz (ed.), Perceiving, Sensing and Knowing. A Book of Readings from Twentieth-Century Sources in the Philosophy of Perception, Anchor Books, New York 1965



The Ambiguity of “Perceiving” 

 83

to the object. What does his being perceptually related to the object amount to then? If Falkenstein’s observations about the distinction between the intuitive and the psychological (or cognitive) are correct, perhaps, unlike Dretske, Kant would not qualify non-cognitive mental occurrences as conscious experiences. Would he regard them as psychological rather than physiological phenomena? If not, the borderline between the intuitive and the conceptual, which he demarcates, would cut across the distinction between the physiological and the psychological (or the cognitive). If that were the case, however, Kant could not speak about intuitive, as opposed to conceptual, representations, thus about nonconceptual representational content, which he does, as a matter of fact. Dretske’s account faces a similar difficulty. For “conscious experience” can be construed either as an occurrence in the case of which an external observer might be the appropriate authority to assess whether it happens or not (likewise, a surgeon assesses whether an anaesthetized patient bleeds), in which case it ceases to be genuinely conscious, or as a psychological state like seeing, touching or smelling with its irreducibly subjective qualitative character. But making the two perspectives compatible presupposes a determinate ontology of the mental – an issue about which Kant, at least, was far from explicit. The main aim of this chapter has been to consider the notions of representation, sensation, and intuition, used by Kant. A number of problems emerged in the course. Two of them point in the same direction, that is, to the problem of the status of sensations and of the possibility of the non-epistemic nature of intuitions (or, in more contemporary parlance, perceptions, which could designate roughly the same thing). If sensations are treated as non-cognitive occurrences and non-epistemic intuitions as non-conscious ones, it may be plausible to suggest, following some commentators, that the Kantian theory of sensibility belongs to the field of science, more specifically, “physiology of the senses,” rather than to the theory of cognition. Such interpretations were indeed pursued in the past, for example in the nineteenth century by Helmholtz,228 and opened up a way to a naturalistic account of perception. But the discussion of the legitimacy of interpreting Kant in the naturalistic vein I have to postpone until chapter 6.

228 Cf. H. Helmholtz, “Facts in Perception,” translated by M. Lowe in: R. S. Cohen, Y. Elkana (eds.), Hermann von Helmholtz. Epistemological Writings, D. Reidel Publishing Co., Dordrecht-Holland, Boston-USA 1977.

4 Kant on Concepts in Experience 4.1 The Point of Departure According to Kant’s Cooperation Thesis (KCT), intuitions unaccompanied by concepts are “blind.” A “blind” intuition can be understood as an intuition that lacks reference to an object. Accordingly, intuitions unaccompanied by concepts must be non-intentional and also nonrepresentational. But nonconceptual content is both intentional and representational. Therefore, the nonconceptualist reading of Kant seems to lack sufficient warrant. In reply to this, the nonconceptualist may point out that KCT applies, quite limitedly, to a certain kind of representational states, namely those states that bear some cognitive value. Mere perception, or intuition in Kant’s language, does not have to bear any cognitive value, but this does not preclude its being object-directed. One can grant that not every case of perceiving an object is doxastic and hence not every instance of perception results in a judgment about an object. But the conceptualist may again offer a response to such an argument. He may appeal to the transcendental deduction of the categories and contend that this argument establishes that intuition are necessarily subordinated to the pure concepts of the understanding insofar as they are to represent objects. For, in order to bring about any kind of representation, including an intuition, there must be synthesis that proceeds in accordance with the categories – rules provided by the understanding. There can be no passively taken-in content of perception. All representations must therefore be products of the mind’s activity. To this, the nonconceptualist may reply by noticing that some kinds of the contents of experience, and hence some kinds of representations do not result from synthesis in accordance with the rules provided by the understanding. Space and time, the forms of intuition and pure intuitions, as such, are not results of the synthesizing activity of the mind. But space and time, the conceptualist may respond, are not objects of representation, let alone of sense perception. Besides, as some commentators have argued, space and time are the outcome of the synthesizing activity of the mind, or “flowing constructions,” as Arthur Melnick has called them.229 Therefore, they are contingent upon the conceptual furnishing of the mind, indeed, upon its categories. A move to be made now, by the nonconceptualist, comes down to the analysis of the arguments put forward by his adversary. If the arguments are based on an

229 “Indeed, I believe that what Kant means by space and time being given in pure intuition is that they are, or exist in, flowing constructions.” A. Melnick, “Categories, Logical Functions and Schemata in Kant,” The Review of Metaphysics 54 (3), 2001, pp. 617-618. © 2014 Anna Tomaszewska This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.



“The Same Function” 

 85

adequate rendering of the Kantian texts, the conceptualist wins; if there are gaps in the conceptualist reading, however, then the victory can be granted to his adversary, or to neither. In what follows, I will consider the arguments which make the conceptualist reading compelling and see to what extent this is indeed the case. Regardless of this, one could anyway argue for Kant’s nonconceptualism in that one would show that KCT applies primarily to the kind of experience that is valuable from the cognitive perspective. Thus, it does not encompass all possible kinds of experience, in particular the kind of experience that does not amount to cognition, such as aesthetic and ethical experience.

4.2 “The Same Function” Unlike general logic, which abstracts from the content of concepts and analyzes the forms of thinking in general, transcendental logic deals with concepts by means of which we cognitively relate to objects in an a priori manner.230 Transcendental logic studies the a priori conceptual content of cognition. That there is such content is not immediately evident because the formal conditions of representing objects could as well be limited to the forms of intuition, space and time. However, in both the Metaphysical and the Transcendental Deduction of the Categories, Kant purports to show that the forms of intuition are insufficient as conditions of representing objects. If these arguments are cogent, then there must be more than a grain of truth in the conceptualist reading of Kant’s theory of experience. The following statement from the third section of the “Clue to the Discovery of all Pure Concepts of the Understanding” seems to epitomize the idea of the conceptualist reading: “The same function that gives unity to the different representations in a judgment also gives unity to the different representations in an intuition, which, expressed generally, is called the pure concept of understanding.”231

In McDowell’s interpretation, what Kant suggests is that the same structure pertains to judgments and to intuitions.232 So, what one, for example, can see and what one

230 Cf. I. Kant, Critique of Pure Reason, translated and edited by P. Guyer and A. W. Wood, Cambridge University Press, Cambridge 1998, pp. 196-7 (A 57/B 82). 231 Ibidem, p. 211 (B 105). 232 More specifically, as Piotr Szałek notes, “in McDowell’s reading of this passage, Kant claims that intuitions (as ‘cases of sensory consciousness of objects’) have logical structures identical with logical structures of judgments.” P. Szałek, “Kant, Hegel and the Puzzles of McDowell’s Philosophy,” Diametros 29, 2011, p. 113. Cf. J. McDowell, “Hegel and the Myth of the Given” in: W. Welsh, K.

86 

 Kant on Concepts in Experience

judges to be the case is structurally identical, even though what one can see and what one judges to be the case does not have to be the same kind of things (for experience itself delivers mere “appearings,” rather than judgments proper233). But, in Kant’s idiom, structure could be identified with form. And to say that the categories provide forms for intuitions, as well as for thought, is to blatantly miss Kant’s point. Rather, the point is that, although the categories do not provide forms for intuitions, they refer to the objects that can be represented in intuition in an a priori manner, thus necessarily. The key word in the above passage from the Metaphysical Deduction is “unity.” For Kant does not claim that the objects represented in intuition are of necessity represented conceptually, as if intuiting, or perceptual experience was a kind of, or analogous to, thinking. The pure concepts of the understanding account for representing manifolds of intuition as unified and they account for unifying conceptual representations in a judgment. Thus, though it is not a kind of, or analogous to, thinking, intuiting must be rule-governed. Does it imply that without concepts the manifolds of intuition would be represented, as it were, in disarray or confusion? And does it imply that without concepts no directedness at objects, and no representation, would be possible? I think that this does not have to be the case. For why not concede that furnishing the manifold with a structure or order may come in degrees? The forms of intuition would provide the basic structure for representations, a kind of primitive intentionality,234 and the categories would add further structuring, which would not be indispensable to entertaining intuitive representations, nonetheless. As I suggested in chapter 2, where I discussed Kant’s claims from his essay on directions in space, the basic structuring of the field of experience – in terms of relations such as “to the left,” “in front of,” “behind,” etc., which exemplify spatial relations – would be due to the bodily constitution of the subject, rather than to the subject’s conceptual repertoire. To illustrate this point, let us think of a toddler’s way of getting acquainted with objects. For a baby that has not mastered any language yet, it may be difficult to parse her perceptual field in the same manner as adult humans do; what “competent” subjects regard as separate objects, for example food and a plate that the food is on, the baby may take as one object. But when she throws a ball on the floor, the child will easily perceive a difference between the ball and a cube, for instance. When thrown

Vieweg (eds.), Das Interesse des Denkens: Hegel aus heutiger Sicht, Wilhelm Fink Verlag, München 2003; “L’idealismo di Hegel come radicalizzazione di Kant,” translated by T. Fracassi, Iride 34, 2001. 233 McDowell defines “appearings,” the contents of perceptual experience, as being “just more of the same kind of things beliefs are: possessors of empirical content, bearing on the empirical world.” J. McDowell, Mind and World, Harvard University Press, Cambridge MA 1996, p. 142. 234 Richard Aquila furthers a reading along these lines in: Representational Mind: A Study of Kant’s Theory of Knowledge, Indiana University Press, Bloomington 1983.



Three Syntheses: Does Apprehension Require Recognition? 

 87

against the floor, the cube will not roll smoothly, unlike the ball. When catching the cube, the child will feel the protruding edges of the solid, unlike when catching the ball. Sensorimotor capacities enable thus the first stage of structuring one’s perceptual field, without the employment of concepts.235 But, according to Kant, at this stage experience manifests no unity; so, strictly speaking, it is no experience (no empirical cognition) at all. For, as Kant remarked in the Anthropology, a small child has no experience but mere perception.236 Granted, the notion of unity is somewhat ambiguous here; for it may refer to the unifying principle such as apperception (experience has one “owner” who can attribute it to himself) or to conceptual capacities, the understanding (concepts, like the category of cause, establish connections between perceptions, e.g. between a ball lying on a pillow and a dent in the pillow). But, endorsing either way of understanding the principle of the unity of experience one can see that, on the Kantian tenets, small children lack both apperception and conceptual capacities, which does not preclude yet that they have some degree of intentionality or conscious perception (unless intentionality and conscious perception presuppose the two principles of the unity of experience, which would be the case on a narrow conception of experience).

4.3 Three Syntheses: Does Apprehension Require Recognition? In the A-edition Transcendental Deduction Kant argues for the necessary involvement of the categories in experience. A programmatic statement, which one can find in A 97, reads: “Now these concepts, which contain a priori the pure thinking in every experience, we find in the categories, and it is already a sufficient deduction of them and justification of their objective validity if we can prove that by means of them alone an object can be thought.”237

Two claims seem to be worth underlining: (i) that every experience involves a pure form of thought, a category, and (ii) that only by means of the pure concepts can the objects of experience be thought. This is compatible with KCT and with understanding experience in the narrow sense – as empirical cognition rather than mere perceptual acquaintance with an object. That the narrow sense of experience is at issue here emerges from the following statement:

235 The sensorimotor theory of consciousness has been discussed in: J. K. O’Regan, E. Myin, A. Noë, “Sensory Consciousness Explained (Better) in Terms of ‘Corporality’ and ‘Alerting Capacity,’” Phenomenology and the Cognitive Sciences 4 (4), 2005. 236 Cf. I. Kant, Anthropology, History, and Education, edited by G. Zöller and R. B. Louden, translated by M. Gregor et al., Cambridge University Press, Cambridge 2007, p. 240 (7:128). 237 I. Kant, Critique of Pure Reason, op. cit., p. 227 (A 97).

88 

 Kant on Concepts in Experience

“If every representation were entirely foreign to the other, as it were isolated and separated from it, then there would never arise anything like cognition, which is a whole of compared and connected representations.”238

Thus, insofar as they are to refer to objects, representations must be connected, and this must be the case in virtue of a priori laws inherent in the mind. This connection of representations is contingent upon an activity of the mind, which Kant calls synthesis, and it pertains both to empirical and a priori, or pure, representations. In the A-edition Deduction, three kinds of synthesis are involved: a synthesis of apprehension in the intuition, a synthesis of reproduction in the imagination and a synthesis of recognition in the concept – according to the threefold division of cognitive faculties into sensibility, imagination, and understanding. The word “synthesis,” in Kant’s use, refers to an operation of the mind that consists in “putting together” simple mental items into more complex ones.239 For instance, a representation of seven green dots arranged in an array is a result of synthesis in that the mind, as it were, runs through each separate item and connects it with another one. The process occurs in time: for representing a row of seven green dots presupposes representing, first, one green dot and then another green dot, and then another one etc. Furthermore, to produce a complex representation of this kind, one needs a rule – in our case, a rule which says that we start from the first item and proceed “forwards” adding every next item to the sum of the preceding ones. In arithmetic, this operation would be defined as the function of successor. The three syntheses are interrelated and dependent upon one another; thus, the synthesis of apprehension, which consists in putting together a manifold of intuition into a unified representation, presupposes the synthesis of reproduction: for without retaining in mind the previous stages of the synthesis of a representation one would not get at the representation as a whole. For example, I could not represent seven dots in an array if I did not retain the moments of synthesizing the previous six ones. As Kant says: “[…] if I were always to lose the preceding representations […] from my thoughts and not reproduce them when I proceed to the following ones, then no whole representation and none of the previously mentioned thoughts, not even the purest and most fundamental representations of space and time, could ever arise.”240

The last synthesis – of recognition in the concept – rests on the consciousness of a rule, hence a concept, by means of which different representations have become synthesized. Without this consciousness it would be difficult to produce a

238 Ibidem, pp. 227-8 (A 97, emphasis mine). 239 Cf. ibidem, p. 210 (B 103). 240 Ibidem, p. 230 (A 102).



Three Syntheses: Does Apprehension Require Recognition? 

 89

representation of a determinate kind – thus, the representation of a row of seven green dots would involve the concept of number and the concept of succession. Therefore, to the extent that intuitive representations result from acts of synthesis, concepts must accompany them: for a concept is “this one consciousness that unifies the manifold that has been successively intuited, and then also reproduced, into one representation.”241 Now, it can be pointed out that the crucial argument for Kant’s conceptualism – i.e., the claim that concepts must accompany intuitions of objects – rests on a contentious assumption adopted by Kant. The assumption boils down to equating consciousness with self-consciousness, a move that some commentators, e.g. Paul Guyer, find illegitimate.242 This means that, for example, if one wants to ascribe to S an experience of seven green dots arranged in a row, one can do so only if such an ascription has been or can be made by S herself. Thus, it does not suffice that S synthesize her representations in accordance with determinate rules (the concept of number and the concept of succession), she must also possess the concepts that are employed in the synthesis. In other words – and to use a distinction introduced in chapter 1 – for Kant, content conceptualism would imply state conceptualism. How the assumption works in the section on syntheses in the A-edition of the Critique can be outlined in the following way. (1) Acquiring a representation of an object involves an act (or acts) of synthesis in accordance with a rule, or concept. (2) The act of synthesis presupposes transcendental apperception, or the a priori unity of consciousness, as its necessary condition,243 and it is only in the act of synthesis that we can become conscious of this unity of consciousness.244 (3) But by becoming conscious of the unity of consciousness, or self-conscious, we become at the same time conscious of an object that corresponds to it.245 (4) Therefore, since synthesis presupposes self-consciousness and since synthesis underlies a representation of an

241 Ibidem, p. 231 (A 103). 242 See: P. Guyer, “Kant on Apperception and ‘A Priori’ Synthesis,” American Philosophical Quarterly 17 (3), 1980. Cf. also: P. Guyer, Kant and the Claims of Knowledge, Cambridge University Press, Cambridge 1987. 243 “Now no cognitions can occur in us […] without that unity of consciousness that precedes all data of the intuitions, and in relation to which all representation of objects is alone possible. This pure, original, unchanging consciousness I will now name transcendental apperception.” I. Kant, Critique of Pure Reason, op. cit., p. 232 (A 107). 244 “Thus the original and necessary consciousness of the identity of oneself is at the same time a consciousness of an equally necessary unity of the synthesis of all appearances in accordance with concepts […]; for the mind could not possibly think of the identity of itself in the manifoldness of its representations, and indeed think this a priori, if it did not have before its eyes the identity of its action, which subjects all synthesis of apprehension (which is empirical) to a transcendental unity, and first makes possible their connection in accordance with a priori rules.” Ibidem, p. 233 (A 108). 245 Thereby, apparently, Kant establishes the principle of the intentionality of consciousness: consciousness is, in principle, “consciousness of an object.”

90 

 Kant on Concepts in Experience

object, it follows that a representation of an object presupposes self-consciousness. In particular, representing an object in intuition implies being conscious of one’s representing an object in intuition. In the light of these considerations, two issues emerge as worthy of closer examination: (i) Kant’s possible conflation of what – following Dretske – can be called thing-awareness and fact-awareness and (ii) the difference between employing concepts as rules for synthesizing representations and employing concepts in (selfreflective) judgments. One can be aware of an array of seven green dots – for instance, one can perceive or imagine an array of seven dots, or think of it – without at the same time being aware that there is an array of seven dots in the physical space. And it is only factawareness that implies a judgment; and only a judgment based on the awareness of a fact involves reference to the subject of experience, or self-ascription of an experience (such as, e.g., I can see that there are seven green dots arranged in a row). Also, only self-ascription of an experience requires that the subject possess the concepts that should be employed in characterizing the experience. But it is far from obvious why self-ascription of experiences, or at least the ability to ascribe experiences to oneself, would have to be presupposed by any conscious experience of an object; in this way, of course, many creatures lacking this ability but nevertheless apparently capable of conscious perception (human infants and non-rational animals such as dogs) would have to be excluded from the category of conscious perceivers. As regards the second issue: readers of the late Wittgenstein would find it obvious that one can follow a rule without being able to account for this process himself;246 the same would be claimed by the advocates of the conception of “tacit knowledge” or “knowledge-how,” such as Michael Polanyi or Gilbert Ryle.247 Also, commentators, such as H. Ginsborg248 or B. Longuenesse, distinguish the two senses of concepts applied by Kant: one in which concepts are considered as rules for synthesizing representations and the other in which concepts are considered as constituents of judgments; and it seems, at least prima facie, that concepts could be operative in experience in the first of these senses while not being employed in the second one. If applying concepts as rules presupposed self-ascription of experiences, there would not be much save a merely conceptual distinction between the two senses. At this point it becomes more evident why McDowell has taken Kant as an ally. If, for Kant, undergoing an experience were to entail the ability to ascribe the experience

246 Cf. C. Wright, Wittgenstein on the Foundations of Mathematics, Duckworth, London 1980; “RuleFollowing without Reasons: Wittgenstein’s Quietism and the Constitutive Question,” Ratio (new series) 20 (4), 2007. 247 M. Polanyi, The Tacit Dimension, University of Chicago Press, Chicago 2009; G. Ryle, The Concept of Mind, The University of Chicago Press, Chicago 1949. 248 H. Ginsborg, “Was Kant a Nonconceptualist?” Philosophical Studies 137, 2008.



Non-Conscious Spontaneity? Schematism and the Transcendental Imagination 

 91

to oneself, experiences would serve as reasons for beliefs in McDowell’s sense. For the subject of experience would possess such reasons, they would be his reasons. Since it is prerequisite for the self-ascription of an experience that one form a judgment on the basis of one’s experience, and judgments are structured by concepts, perceivers must possess appropriate concepts in order to entertain content-laden perceptual experiences. Therefore, conflating consciousness with self-consciousness would encourage state conceptualism (I described the position in chapter 1).

4.3 Non-Conscious Spontaneity? Schematism and the Transcendental Imagination It may seem quite puzzling why, following the Transcendental Deduction of the Categories, in both editions of the Critique of Pure Reason, Kant inserts a short chapter on the schematism of the pure concepts of the understanding. This is because the goal Kant undertakes in that chapter does not significantly differ from the goal he undertakes in the Deduction. The latter is to show that the involvement of the categories in experience is necessary since only due to them can objects be represented in intuition. For example, in § 20 of the Deduction, Kant sets off to prove that “all sensible intuitions stand under the categories, as conditions under which alone their manifold can come together in one consciousness.”249 And the conclusion from § 26 reads: “Consequently all synthesis, through which even perception itself becomes possible, stands under the categories, and since experience is cognition through connected perceptions, the categories are conditions of the possibility of experience, and are thus also valid a priori of all objects of experience.”250

Thus, clearly, Kant deals with the relation between concepts and intuitions in the Deduction. But in the chapter on schematism the problem recurs. This may be because, despite the argument in the Deduction, Kant has still not explained how two intrinsically heterogeneous kinds of representations can be combined together. The doctrine of schematism can thus be considered as an attempt at such an explanation that appeals to basic cognitive processes which constitute of the content of cognition. These basic cognitive processes are syntheses as a result of which transcendental schemata emerge. Kant calls the latter “mediating representations” and describes

249 I. Kant, Critique of Pure Reason, op. cit., p. 252 (B 143). 250 Ibidem, p. 262 (B 161).

92 

 Kant on Concepts in Experience

them as sensible and intellectual at the same time.251 Schemata are products of the imagination,252 which Kant calls “a blind though indispensable function of the soul, without which we would have no cognition at all, but of which we are seldom even conscious.”253 The inscrutable nature of the transcendental imagination is the focus of Kant and the Problem of Metaphysics, where Heidegger argues that the imagination constitutes the unknown “root” of sensibility and understanding.254 On W. Waxman’s interpretation, the transcendental imagination constructs the a priori representations of space and time,255 hence it conditions the possibility of both empirical and pure cognition. Hegel offers a similar reading in Faith and Knowledge and identifies the imagination with reason as operating within the sphere of the empirical consciousness.256 Transcendental imagination could be regarded as spontaneity without consciousness. It is spontaneous since it produces representations – schemata – by combining more primitive representations, that is, concepts and intuitions, or, more specifically, the categories and the pure intuitions.257 It operates without consciousness since its operations are more fundamental than entertaining conscious representational states, or experiences, and explanatory of the latter. If the imagination produces representational content, entertaining a contentladen representational state does not have to implicate a judgment in which a subject self-reflectively ascribes an experience to himself. Rather, the subject’s experience can be conceptually structured without the requirement of concept-possession being satisfied. Against this background, though the analogy is limited in scope, the operations of the transcendental imagination can perhaps be likened to cognitive states that arise on the so-called subpersonal level of cognition. Subpersonal states, as much as the workings of the Kantian imagination, would play an explanatory role in an account of cognition, inaccessible on the level of the phenomenological analysis of experience.258 As indicated, the analogy is limited in scope, which

251 Ibidem, p. 272 (A 138/B 177). 252 Ibidem, p. 273 (A 140/B 179). 253 Ibidem, p. 211 (A 78). 254 M. Heidegger, Kant and the Problem of Metaphysics, translated by R. Taft, Indiana University Press, Bloomington 1997. 255 W. Waxman, Kant’s Model of the Mind. A New Interpretation of Transcendental Idealism, Oxford University Press, Oxford – New York 1991. 256 G. W. F. Hegel, Faith and Knowledge, translated by W. Cerf and H. S. Harris, SUNY Press, New York 1977. 257 On the notion, role and importance of spontaneity in Kant’s system see a recent monograph: Marco Sgarbi, Kant on Spontaneity, Continuum, London 2012. 258 Cf. J. L. Bermúdez, “Nonconceptual Content: From Perceptual Experience to Subpersonal Computational States,” Mind and Language 10 (4), 1995; “What is at Stake in the Debate on Nonconceptual Content?” Philosophical Perspectives 21 (1), 2007. The term “subpersonal states” has been introduced to the philosophy of mind in: D. C. Dennett, Content and Consciousness, Humanities Press, New York 1969.



The “I think” of the Transcendental Apperception 

 93

is because it may blur the sense in which the operations of the imagination are spontaneous – since they defy an explanation in causal-theoretical terms. Indeed, the analogy fosters a naturalistic reading of the “non-conscious spontaneity,” an issue I return to in chapter 6.

4.4 The “I think” of the Transcendental Apperception In the B-edition Deduction, Kant makes a statement which apparently undermines the nonconceptualist reading: “The I think must be able to accompany all my representations; for otherwise something would be represented in me that could not be thought at all, which is as much as to say that the representation would either be impossible or else at least would be nothing for me.”259

Consider the arguments from the richness or fineness of detail of perceptual content. If the “I think” must be able to accompany all representational states, there cannot be a state, such as seeing a particular shade of green, which cannot be related to the capacity of thought. Hence perceptual experience, though at times it may outstrip subjects’ conceptual repertoire, must potentially be apt for conceptualization. One may currently lack the right concept for this particular shade of green and hesitate about whether John’s jacket and the carpet in one’s room have the same or only similar shade of green. But it is in principle possible to acquire and employ the concept of this particular shade of green – as a last resort one may avail oneself of a perceptual demonstrative like “that shade,” to recall McDowell’s suggestion. I think, however, that the claim made by Kant in the B 132 passage does not pose an insuperable challenge to the nonconceptualist interpretation. Before I say why, let me consider two possible ways of reading the “I think” passage – and their shortcomings. (1) The “I think” could be construed along the lines of the Cartesian Cogito from the fourth part of the Discourse on Method and the Second Meditation from the Meditations on First Philosophy. The Cartesian Cogito is a subject of mental acts such as “doubting, understanding, denying, being willing, being unwilling, and also imagining and having sensory perceptions.”260 In all such acts, by definition directed at their corresponding objects, equated by Descartes with ideas, the thinking subject is present; that is, it is conscious of the acts it performs. Clearly, this kind of conscious presence of the subject is not restricted to, or dependent upon, the employment of concepts. The “I think,” or self-consciousness, could be likened to a shaft of light cast

259 I. Kant, Critique of Pure Reason, op. cit., p. 246 (B 132). 260 R. Descartes, Meditations on First Philosophy. With Selections from the Objections and Replies, edited by J. Cottingham, Cambridge University Press, Cambridge 1996, p. 19.

94 

 Kant on Concepts in Experience

upon the objects within the field of its comprehension. In this sense, the “I think” may also accompany nonconceptual representations; indeed, for this reason, Kant identifies intuitions as conscious singular representations of objects.261 Read in this way, Kant’s claim about the “I think” which “must be able to accompany all my representations” indicates that there cannot be mental representational states that would be utterly non-conscious; states that would be inaccessible to the consciousness of the subject of experience or thought. To entertain such states would mean to entertain mental representational states that one could not regard as one’s own. Obviously, this entails a contradiction, and accordingly Kant’s claim must be analytic. But if this reading of the “I think” passage is correct, it becomes unclear whether the operations of the imagination may count as mental states at all, and whether the products of these operations can be construed as representations. In particular, the cognitive status of schemata would be in question, and, if we follow the constructivist reading, defended by Heidegger, Melnick, and Waxman, also the status of the pure forms of intuition, space and time, themselves pure intuitions and therefore representations, as Kant contends. (2) But the B 132 statement can also be read in a different way, which builds, to an extent, upon the previous interpretation saying that all one’s representations must be at least potentially self-conscious. For “I think” would, as a rule, be followed by a thatclause. The “I think” would thus be a tag for fact-awareness. We usually say: “I think that thus and so is the case.” Since facts are captured by propositions and propositions are conceptually structured, Kant’s statement could be taken as implicating that all representations must be able to adopt the structure of a proposition; or that all representations must be at least potentially structured by concepts. However, this reading generates a puzzling problem. As mentioned in (1), Kant defines intuitions as conscious objective representations, but intuitions, by definition, are not conceptually structured. Strictly speaking, some structuring is endowed upon intuitions by space and time. So, intuitions must be spatiotemporally structured. This is clear on the basis of Kant’s arguments presented in the Transcendental Aesthetic and the Inaugural Dissertation written a decade earlier. Also, Kant’s Heterogeneity Thesis (KHT, discussed in chapter 3) precludes that intuitions have the same kind of structure as conceptual representations. Thus, reading the “I think” passage in terms of fact-awareness makes Kant’s theory of empirical cognition inconsistent. We have seen that, on the basis of both (1) and (2), Kant could be charged with an inconsistency. But this is not the conclusion one would readily embrace unless one wants to bring one’s considerations to a halt. Let us then read a bit further:

261 See the so-called Stufenleiter passage: I. Kant, Critique of Pure Reason, op. cit., pp. 398-9 (A 320/B 377).



Syntheses and Intuitions: Are Space and Time “Given” or “Constructed”? 

 95

“That representation that can be given prior to all thinking is called intuition. Thus all manifold of intuition has a necessary relation to the I think in the same subject in which this manifold is to be encountered. But this representation is an act of spontaneity, i.e., it cannot be regarded as belonging to sensibility. I call it the pure apperception, in order to distinguish it from the empirical one, or also the original apperception, since it is that self-consciousness which, because it produces the representation I think, which must be able to accompany all others and which in all consciousness is one and the same, cannot be accompanied by any further representation.”262

There is a way to accommodate both Kant’s account of intuition as a nonconceptual kind of representation and the requirement that the “I think” accompany all of one’s representations. The interpretative maneuver is fairly simple and may be reminiscent of Gareth Evans’s “two-tier” theory of perception, referred to in chapter 1. On this theory, the nonconceptual content of intuition could be described as “the input to a thinking, concept-applying and reasoning system.”263 The “I think” requirement would then boil down to saying that all representations, including intuitions, must be poised for conceptualization and, accordingly, underlie perceptual judgments. Kant’s requirement that the nonconceptual content of intuition be eligible for conceptualization does not have to lead to an inconsistency. For such a requirement does not imply, without further premises, that intuitions themselves, singular representations of objects, must be conceptually structured.

4.5 Syntheses and Intuitions: Are Space and Time “Given” or “Constructed”? How to represent space was an issue for eighteenth-century philosophers and philosophical psychologists, as it is an issue for cognitive scientists and philosophers of mind today. In particular, early modern optics purported to address the question of the representation of distance and, accordingly, the three-dimensionality of objects, relying on physiological data and geometry. Geometrical optics would be challenged by the empiricists, in particular George Berkeley in his New Theory of Vision, primarily because it located the process of perceiving spatial properties outside the conscious

262  Ibidem, pp. 246-7 (B 132). 263 G. Evans, The Varieties of Reference, edited by J. McDowell, Oxford University Press, Oxford 1982, p. 158. More specifically, Evans discusses conditions that should be met by informational states if they are to be classified as conscious experiences. As he claims, this can be the case provided that an informational state “serves as the input to a thinking, concept-applying and reasoning system; so that the subject’s thoughts, plans and deliberations are also systematically dependent on the informational properties of the input. When there is such a link we can say that the person, rather than some part of his brain, receives and processes the information.” Ibidem. See also: F. Macpherson and J. L. Bermúdez, “Nonconceptual Content and the Nature of Perceptual Experience,” Electronic Journal of Analytic Philosophy 6, 1998 (http://ejap.louisiana.edu/EJAP/1998/bermmacp98.html, accessed on 5th July 2013).

96 

 Kant on Concepts in Experience

mind of the subject of experience. Berkeley explained perception of distance in terms of the empirical association of different ideas: visual, tactile, and kinesthetic. Thus, as Lorne Falkenstein suggested, we may call him a constructivist with regard to perception of spatial properties. Conversely, Kant would be an intuitionist because he thought that the representation of space (and time) was not a result of any more basic cognitive process or processes. Briefly, on Kant’s account, space (and time) would be “given” rather than “constructed.” Here is Falkenstein’s characteristic of the two positions: “If you believe that a certain output is already contained in the input to a processor, so that it does not require any process (other than transmission or attention) to become known, then you are what I call ‘an intuitionist’ with regard to that output. Aristotle, for example, gives an intuitionist account of colour perception in De Anima, II vii. According to this theory, colours exist in objects, and when a transparent medium between the object and the eye is activated by light, the colour imposes its form on the eye. This account is intuitionist, because the colour ultimately cognized by the soul is thought to be already contained in the sensory input, and nothing needs to be done to it by the senses or the soul other than to transmit and attend to it […]. If, in opposition to both sensationism and non-sensationist intuitionism, you believe that a given output is not already contained in the input to the cognitive system, but needs to be worked up out of that input by some process such as association, inference, comparison, abstraction, combination, or composition, then you are what I call ‘a constructivist’ with regard to that output. Berkeley’s NTV [New Theory of Vision] is the classic constructivist work on the cognition of visual spatial depth.”264

That space and time are “given” rather than “constructed,” according to Kant, clearly follows from the analysis of the concepts of space and time in their “metaphysical expositions” in the Transcendental Aesthetic, where Kant maintains his views on space and time formulated in the Inaugural Dissertation over a decade earlier. Thus, about space he claims that it “is not an empirical concept that has been drawn from outer experiences;”265 that it is an a priori, hence necessary, representation that underlies all intuitions;266 that it is “not a discursive or […] general concept of relations of things in general, but a pure intuition;”267 and that it is represented as infinite, that is unbounded.268 But we have seen that the Transcendental Deduction of the Categories allows us to challenge these views. Let us summarize the arguments that it can furnish against the intuitionist account of space and time, or against the claim that space and time are “given” rather than “constructed.”

264 L. Falkenstein, Kant’s Intuitionism. A Commentary on the Transcendental Aesthetic, University of Toronto Press, Toronto 2004, pp. 7-8. 265 I. Kant, Critique of Pure Reason, op. cit, p. 157 (A 23/B 38). 266 Ibidem, p. 158 (A 24/B 39). 267 Ibidem, p. 158 (A 25). 268 Ibidem, p. 159 (A 25).



Syntheses and Intuitions: Are Space and Time “Given” or “Constructed”? 

 97

1. Representations in general, and intuitions in particular, are complex mental “items”, results of syntheses of a manifold in accordance with rules. The rules are provided by concepts: empirical and pure (i.e. the categories) in the case of empirical representations and only pure in the case of the a priori (i.e. non-empirical) ones. Insofar as space and time are intuitions, hence representations, albeit singular, they must be the result of the act of synthesis. In other words, if all intuitions emerge as a result of synthesis – putting together simple mental items into a complex and unified one – also the pure intuitions must emerge in this way. 2. All representations are, of necessity, related to the unity of consciousness, or the transcendental apperception. That unity of consciousness, which in fact is selfconsciousness, consists in the consciousness of the activity of synthesizing the manifold of representations in accordance with rules. Hence, insofar as space and time are representations, they must be related to the unity of consciousness and we may cognize them only “in” the activity of putting together the manifold of intuitions in accordance with concepts. Since self-consciousness implies (the possibility of) judgment, our cognition of space and time would be dependent on (the capacity of) making judgments, which involves the categories. Importantly, this is the purport of the Analogies of Experience, the part of the Transcendental Analytic in which Kant shows in what way the categories operate in various principles constitutive of empirical cognition.269 3. The transcendental apperception furnishes representations, including singular representations (intuitions), with unity by means of concepts, rather than the forms of intuition. Since concepts are constitutive of singular representations, intuitions cannot be “given” but must be “constructed”, in the acts of conceptualization and/or judgment. Apparently, this idea has been encapsulated in the following, quite cryptic, footnote passage from the B-edition Deduction: “Space, represented as object (as is really required in geometry), contains more than the mere form of intuition, namely the comprehension of the manifold given in accordance with the form of sensibility in an intuitive representation, so that the form of intuition merely gives the manifold, but the formal intuition gives unity of the representation. In the Aesthetic I ascribed this unity merely to sensibility, only in order to note that it precedes all concepts, though to be sure it presupposes a synthesis, which does not belong to the senses but through which all concepts of space and time first become possible.”270

269 Cf. the Second Analogy. Ibidem, pp. 304-316 (A 189/B 232 – A 211/B 256). 270 Ibidem, p. 261 (B 160-1).

98 

 Kant on Concepts in Experience

This is one of the notoriously difficult passages in the First Critique.271 Kant distinguishes here the forms of intuition and formal intuitions. The forms of intuition enable the manifold of intuition to be given. But the formal intuitions account for the unity of the manifold. The unity of representation requires synthesis. Synthesis cannot be the function of sensibility, which is merely receptive. But the unity pertaining to formal intuitions does not originate from the understanding either; rather, it “belongs to space and time.” Perhaps this implies a division into a kind of lower-level and higherlevel unity, one that can be attributed to representations as such and the other that can be attributed to the representations of objects. Formal intuitions provide that unity which underlies “all concepts of space and time.” Since unity requires synthesis, all concepts of space and time must rest on a certain kind of synthesis. Synthesis is a function of the spontaneous faculty of the mind, the understanding and/or imagination. Formal intuitions must, therefore, be products of the higher faculty of cognition. This does not mean that Kant changes the status of space and time, which he characterized in the Transcendental Aesthetic, from intuitions, singular representations of sensibility, to general or discursive representations of the understanding. Space and time, in the footnote quoted, are not to be construed as concepts. This is because, by “concepts of space and time,” one can understand “mathematical concepts,” that is, concepts employed in the cognition of mathematical objects. Now, certainly mathematical cognitions, as expressible in judgments, must be products of the synthesizing activity of the mind, and so one may think of the formal intuitions as of the pure intuitions: space and time – considered as the objects of mathematical cognition. Some commentators, among them Wayne Waxman, object to reading the B 160 footnote as relevant to Kant’s account of mathematical cognition, directed at formal intuitions as its objects. Waxman suggests that the passage refers to the “metaphysical/ transcendental” space and time.272 Accordingly, space and time should be described as “entia imaginaria,” that is, objects of the transcendental imagination.273 Since, as

271 Waxman remarks that the passage has occasioned considerable controversy in the literature and that “some have even gone so far as to write it off as too obscure and convoluted to be made sense of.” W. Waxman, Kant’s Model of the Mind: A New Interpretation of Transcendental Idealism, op. cit., p. 80. 272 “As I read it, the footnote, by making clear that metaphysical/transcendental space and time are the result of a synthesis and a unity not belonging to the senses, buttresses his claim in the text that they are intuitive embodiments of the same synthetic unity of apperception of which the categories are the conceptual embodiments […]; accordingly, formal intuitions provide just the bridge between appearances and the categories that Kant was seeking in section 26.” Ibidem, p. 83. 273 “Of all the passages in the Critique of Pure Reason relating to the pure space and time of the Transcendental Aesthetic, none comes closer to an explicit, unequivocal statement that they are products of a synthesis of the imagination than the footnote at B 160, in section 26 of the second-edition version of the Transcendental Deduction of the Categories.” Ibidem, p. 79.



Syntheses and Intuitions: Are Space and Time “Given” or “Constructed”? 

 99

Kant emphasizes, the synthetic unity of apperception must be prior to the analytic one,274 there are operations of the mind which underlie the application of concepts in judgments. The transcendental imagination performs these operations. Thus, on Waxman’s reading, Kant’s account of space and time, developed in the Analytic does not contradict the account of space and time one can find in the Transcendental Aesthetic. Waxman purports to offer a solution to the problem of the “common root”275 of the representations of sensibility and the understanding. But whether such a solution can be given remains contentious. Many philosophers, Hegel and McDowell included, complained that Kant’s theory of space and time is evidence that he had not set himself entirely free from the empiricist way of thinking about the human mind, as at least partly receptive, or “constrained from outside.”276 But it is far from clear why the forms of sensible intuition, limiting cognition to appearances, the objects of sensibility, would have to be regarded as external rather than internal constraints – constraints originating from the constitution (or nature) of the subject. Only because these critics conceive of the embodied receptivity as external to the subject’s mind can they claim that constraints come “from outside.” However, such an assumption does not withstand all criticism. It is also worth to note that if the imagination were to account for all representations of the mind, as their “common root,” it would be difficult to see why Kant insists on the distinction between sensibility and understanding, and their corresponding kinds of representations: intuitions and concepts. Waxman’s interpretation downplays the importance of KHT and, consequently, it ignores the role the distinction between the faculties plays in the key Kantian doctrine: transcendental idealism, a topic to be discussed in chapter 5.

274 “Therefore it is only because I can combine a manifold of given representations in one consciousness that it is possible for me to represent the identity of the consciousness in these representations itself, i.e., the analytical unity of apperception is only possible under the presupposition of some synthetic one.” I. Kant, Critique of Pure Reason, op. cit., p. 247 (B 133-4).And further, in a footnote: “And thus the synthetic unity of apperception is the highest point to which one must affix all use of the understanding, even the whole of logic and, after it, transcendental philosophy; indeed this faculty is the understanding itself.” Ibidem. 275 In the Introduction to the Critique, Kant writes: “[T]here are two stems of human cognition, which may perhaps arise from a common but to us unknown root, namely sensibility and understanding, through the first of which objects are given to us, but through the second of which they are thought.” Ibidem, p. 152 (A 15/B 29-30). 276 See: J. McDowell, “Hegel and the Myth of the Given,” op. cit.. According to McDowell, Kant “sees our cognitive freedom as constrained, from outside, by the specific forms of our sensibility, which he leaves looking like a mere brute fact about the shape of our subjectivity.” Ibidem, p. 85.

100 

 Kant on Concepts in Experience

4.6 Non-Cognitive Perception What kind of objects can be perceived? Does perception require a special faculty of the mind? Are there special objects of perception? Is perception always focused on the cognition of objects? Such questions perplexed philosophers in the past and they continue to do so. The debate between conceptualists and nonconceptualists, whose main tenets and arguments I outlined in chapter 1, brings back old issues in a new verbal guise. But one can also rely on one’s linguistic intuitions, however philosophically unwarranted they may ultimately turn out to be, to try to answer these questions. Thus, for example, one can say that one perceives colours, sounds, and textures, but also cats, trees, and books in the cabinet. One can also say that one perceives the beauty of the nearby lakeside and joy in a friend’s tone of voice. And it seems right to say that one perceives the goodness of an action. All in all, it seems to make sense to claim that an object which can be perceived may belong to one of the following categories: (a) sensible qualities, (b) physical objects in space and time, (c) ethical and aesthetic properties or values, (d) feelings. I do not insist that the list is comprehensive and cannot be further expanded. Now, regarding the faculties requisite for perception of objects falling under one of the four categories in the above list: properly functioning senses should suffice to receive sensible qualities, but the grasp of physical objects in space and time would at least presuppose a synthesis of various sensible features of the objects or a judgment (belief) about their existence,277 and so it would involve faculties like the Kantian imagination and understanding. It becomes, however, markedly more difficult to specify the kind of faculty that would account for perception of ethical and aesthetic properties or values, and feelings. According to Kant, morality is the domain of practical reason, and aesthetics – the doctrine of the beautiful – is the domain of the faculty of judgment. Both ethical and aesthetic experiences contain non-cognitive concept-independent components, such as a (rational) feeling of reverence for the moral law and a feeling of pleasure in contemplating a beautiful object.278 If this is the case, then it is clear that cognition does not constitute the main end of perception. Besides, perception restricted to proper objects of the senses, sensible qualities, apparently fails to produce any cognition – for cognition requires the involvement of the capacity to judge. Cognition, therefore, is not essential to perception and a non-cognitive model of perception could perhaps be built on the basis of Kant’s ethical and aesthetic considerations. But it may also be the case that perception,

277 Such is the view of, e.g., Thomas Reid, an eighteenth-century Scottish philosopher, representing the so-called school of common sense. 278 These “rational” feelings become introduced in the Critique of Practical Reason and Critique of the Power of Judgment, respectively.



Non-Cognitive Perception 

 101

which involves the faculty of sensibility, is of hardly any importance to the latter: the feeling of reverence for the moral law and the capacity to distinguish between morally right and morally wrong actions, manifested both by children and adults, originates from reason in its practical application.279 Likewise, the feeling of pleasure in contemplating a beautiful object originates from the aesthetic judgment.280 Within the current framework, at issue is the kind of perception that involves sensibility and produces intuitions, in accordance with their forms. Thus, consistency would require that we investigate to what extent sense perception is constitutive of the non-cognitive kinds of mental content. Certainly, the senses provide the most basic kind of information about the objects one may perceive as exemplifying some general features, or about the objects one may further identify as moral agents,281 the so called “rogue objects,” to use Robert Hanna’s term,282 or bearers of aesthetic properties. More specifically, perceptual states convey information about the spatial and/or temporal location of objects. This is precisely the kind of information that makes it possible to refer to objects, and keep track of them within the field of one’s conscious experience and in the course of one’s engaging with objects as cognizer, moral agent, and contemplator of beauty. And this is also the kind of information that theorists would be prone to characterize as

279 In the Critique of Practical Reason, Kant gives an example of a 10-year old boy who, though without a highly sophisticated ethical idiom, is able to see that lying against an innocent person, even in the case in which one’s life is under threat, would be morally corrupt. To moral intuitions Kant refers also when discussing the example with a poor depositary, obliged to pay his debts to his rich benefactor even after the latter’s death. See: I. Kant, Critique of Practical Reason, translated by W. S. Pluhar, Hackett Publishing Co., Cambridge/Indianapolis 2002, p. 195 (5:156). 280 “Now this merely subjective (aesthetic) judging of the object, or of the representation through which the object is given, precedes the pleasure in it, and is the ground of this pleasure in the harmony of the faculties of cognition.” I. Kant, Critique of the Power of Judgment, translated by P. Guyer and E. Matthews, Cambridge University Press, Cambridge – New York, Cambridge 2002, p. 103 (5:218). For pleasure arises from the “free play” of faculties: imagination and understanding. A “judgment of taste,” based on this feeling, does not “presuppose a determinate concept,” and so no cognition of the contemplated object. Ibidem, p. 103-2 (5:217-18). 281 For example, Patrick Kain remarks that the Kantian moral agents must be both persons, that is rational entities equipped with a will, and perceptible objects, that is objects given in experience. Towards an imperceptible person it would be impossible for us to act and therefore we could not have any duties towards such a “person.” Equally, an imperceptible person could not be morally obligated towards us. “Kant’s Defense of Human Moral Status,” Journal of the History of Philosophy 47 (1), 2009. 282 “Rogue objects,” or “elusive objects,” are objects which do not fit into the framework outlined in the Analogies of Experience, which hence are “nomologically ill-behaved.” In this category Hanna includes: “objects of intuition that engage in systematically counter-nomological behavior (magic), purely random or indeterministic behavior (pure chance), or spontaneous goal-directed behavior (life, consciousness, freedom).” “Kant’s Non-Conceptualism, Rogue Objects, and the Gap in the B Deduction,” International Journal of Philosophical Studies 19 (3), 2011, p. 409.

102 

 Kant on Concepts in Experience

nonconceptual content, since its acquisition and possession do not necessitate the employment of conceptual capacities. Thus, nonconceptual content can be claimed to underlie perceptual states, both aimed at cognition and non-cognitive. That the acquisition of spatial (and temporal) information is concept-independent, on Kant’s account, comes to be evident, for example, once we read the passages in which Kant claims that orientation in the spatial (or spatiotemporal) framework is rooted in a certain kind of feeling. One of these passages comes from an essay on the “orientation in thinking,” and reads: “Now if I see the sun in the sky and know it is now midday, then I know how to find south, west, north, and east. For this, however, I also need the feeling of a difference in my own subject, namely, the difference between my right and left hands. […] If I did not have this faculty of distinguishing, without the need of any difference in the objects, between moving from left to right and right to left and moving in the opposite direction and thereby determining a priori a difference in the position of the objects, then in describing a circle I would not know whether west was right or left of the southernmost point of the horizon […]. Thus even with all the objective data of the sky, I orient myself […] through a subjective ground of differentiation […]. But in fact the faculty of making distinctions through the feeling of right and left comes naturally […] – it is a faculty implanted by nature but made habitual through frequent practice.”283

Since, as Kant argued in his 1768 essay on directions in space, cited in chapter 2, it is determining the basic spatial relations, such as “to the left/right,” “above/below,” “in front of/behind,” that allows one to locate objects in space and relate them to one another, and since the feeling which makes this structuring possible originates from the body of the subject, it follows that underlying the very basic ability to refer to objects in perceptual experience are not conceptual capacities but the skills of the human body, including sensorimotor skills. It is because Kant assumes that the space of our experience is egocentrically oriented, rather than a result of abstraction from the data of experience or construction from them, as Leibniz and Berkeley would contend, that we can attribute to him the claim that both cognitive and non-cognitive relatedness to objects is grounded in spatiotemporally structured nonconceptual content. The basic form of intentionality – in cognition, moral agency and aesthetic experience – consists in pre-conceptual relatedness to the realm of objects of an embodied, sensing subject, equipped also with faculties such as imagination, understanding, reason, and the capacity to judge. I have argued, throughout this chapter, that even those passages in Kant’s writings, which prima facie corroborate the conceptualist interpretation, can be read as compatible with the nonconceptualist reading. Thus, the B 105 passage from the

283 I. Kant, “What Does It Mean to Orient Oneself in Thinking?” in: I. Kant, Religion and Rational Theology, translated and edited by A. W. Wood and G. Di Giovanni, Cambridge University Press, Cambridge – New York 1996, pp. 8-9.



Non-Cognitive Perception 

 103

“Clue to the Discovery of All Pure Concepts of the Understanding,” about concepts that provide the same kind of structure to intuitions and judgments, is compatible with the claim that intuitions manifest also a more basic kind of structuring. The doctrine of the three syntheses, developed in the A-edition Deduction of the Categories, leaves it open whether all representations, including the pure intuitions: space and time, result from the rule-governed operations of the higher mental faculty, called synthesis. (Admittedly, it is also not entirely clear what it means to represent space and time as pure intuitions.) But even if all representations are products of synthesis, carried out by an intermediary faculty – the transcendental imagination, which operates “blindly,” i.e. without the subject’s consciousness – there is still space for nonconceptual content as “material” used by the imagination and nonconceptual states as states which are governed by concepts, though without the subject’s ability to self-attribute the contents of these states in a judgment. Furthermore, the B 132 “I think” passage is not any more helpful in establishing the conceptualist reading since it only states that all representations must be potentially suitable for conceptualization for the sake of forming a judgment on their basis. This conforms with a “two-tier” account of perceptual content, along the lines of Gareth Evans. Finally, the B 160 footnote does not subvert the Kantian doctrine of pure intuitions, formulated in the Inaugural Dissertation and the Transcendental Aesthetic. This is because “formal intuitions” can be construed as mathematical objects, more plausibly than as space and time produced by the transcendental imagination. If all representations originated from one “common root,” with concepts and intuitions being only a kind of aspects of representational content, there would be no need to distinguish between sensibility and understanding, which Kant starts to do in his precritical writings and does consistently in the Critique of Pure Reason, building up the doctrine called transcendental idealism.

5 Nonconceptual Content and Transcendental Idealism 5.1 What Is Transcendental Idealism? Transcendental idealism can be conceived as a conjunction of two claims, one of which is positive and the other negative. The positive claim deals with the objects of cognition and the negative one with cognitive constraints. Thus, the former says what we can know and what the status of things we can know is; the latter what we cannot know and what the status of things we cannot know is. Accordingly, in its positive guise, transcendental idealism can be summarized by the claim that we can have knowledge of the objects of experience, which are spatiotemporal (or temporal) particulars. The objects of experience, however, have the status of appearances, rather than things in themselves, the subject matter of the negative claim. In its negative formulation, transcendental idealism can be presented as a claim about constraints intrinsic to our cognition, which says that we cannot cognize things in themselves. Kant also calls the objects of cognition phenomena, whereas the objects that are cognitively inaccessible (to finite subjects) he identifies with noumena. The term “transcendental idealism” is closely related to such terms as “transcendental cognition” and “transcendental philosophy,” defined by Kant in the following way: “I call all cognition transcendental that is occupied not so much with objects but rather with our a priori concepts of objects in general. A system of such concepts would be called transcendental philosophy.”284

Transcendental cognition – pertaining to the necessary conditions of our knowledge – motivates transcendental idealism in both its positive and negative formulations. For Kant establishes identity between the conditions of knowledge and the conditions of the objects of knowledge.285 Thus, any object that might exist independently of the conditions at issue cannot be cognized by subjects like us unless it is cognized under the conditions of our cognition.

284 I. Kant, Critique of Pure Reason, translated and edited by P. Guyer and A. W. Wood, Cambridge University Press, Cambridge 1998, p. 133 (A 12). 285 See, e.g., the following statement that neatly sums up the results of the Transcendental Deduction of the Categories in the B-edition of the Critique: “since experience is cognition through connected perceptions, the categories are conditions of the possibility of experience, and are thus also valid a priori of all objects of experience.” Ibidem, p. 262 (B 161, italics A.T.). © 2014 Anna Tomaszewska This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.



And How Does It Bear on Nonconceptualism? 

 105

Many authors have considered transcendental idealism a metaphysical oddity. It has been charged with incoherence286 and also read in different ways. Thus, it has been read as encouraging the dualism of appearances and things in themselves, the merely thinkable objects radically transcendent to (human) cognition.287 But it has also been read as encouraging two perspectives on one and the same reality: one from which objects present themselves as conditioned by the structures of the subject’s cognitive faculties, and the other from which objects are considered in abstraction from their relation to the faculties of cognition.288 However, it seems that those who tend to charge Kant’s transcendental idealism with incoherence tend to forget that with empirical realism it makes two different sides of one and the same “coin.” Thus, it cannot only be a claim about mysterious supersensible objects beyond the realm of empirical cognition. It is also a claim about constraints intrinsic to cognition, conditioned by certain a priori factors.

5.2 And How Does It Bear on Nonconceptualism? Supposing then that transcendental idealism is a doctrine about constraints intrinsic to empirical cognition, how does it relate to the idea that nonconceptual content constitutes the content of perceptual experience, or that conceptual capacities are not indispensable for some representational states to occur? I mentioned earlier that transcendental idealism provides an additional argument for Kant’s nonconceptualism

286 For example, Sebastian Gardner, in his commentary on Kant’s Critique of Pure Reason, divides interpretations of Kant’s theoretical philosophy into analytic and transcendental idealist and notices that the former usually discard the doctrine of things in themselves as incoherent. On the analytic interpretations, stress is put on the structure of the so-called transcendental arguments and Kant’s empirical realism, the analysis of the conditions of knowledge, which might otherwise be considered as the conditions of meaning of a certain class of statements. These conditions are not to be found in the structure of the reality or objects, but in the structure of our concepts, or language. An analytic reader of Kant will be interested in finding out conceptual presuppositions of certain claims that are considered true. But even if he arrives at the rock-bottom structure of our beliefs, he will not claim to have dug up to the ultimate structure of reality. Cf. S. Gardner, Kant and the ‘Critique of Pure Reason’, Routledge, London 1999. 287 Such an interpretation has been favoured by a number of authors from diverse philosophical camps, e.g.: J. Bennett, Kant’s Analytic, Cambridge University Press, Cambridge 1966; G. W. F. Hegel, Faith and Knowledge, translated by W. Cerf and H. S. Harris, SUNY Press, Albany 1977; J. H. McDowell, Mind and World, Harvard University Press, Cambridge MA 1996; P. F. Strawson, The Bounds of Sense. An Essay in Kant’s ‘Critique of Pure Reason’, Methuen, London 1966. 288 A reading along these lines can be found, e.g., in: H. E. Allison, Kant’s Transcendental Idealism. An Interpretation and Defense, Yale University Press, New Haven and London 2004; G. Prauss, Erscheinung bei Kant, de Gruyter, Berlin 1971; G. Prauss, Kant und das Problem der Dinge an sich, Bouvier, Bonn 1974; M. Siemek, Idea transcendentalizmu u Fichtego i Kanta. ������������������������� Studium z dziejów filozoficznej problematyki wiedzy, PWN, Warszawa 1977.

106 

 Nonconceptual Content and Transcendental Idealism

– but how is this supposed to be the case? And, more interestingly, can we read Kant’s doctrine as also providing an argument for other theories of nonconceptual content? Assuming that Kant’s transcendental idealism can be read in such a way, one may develop an argument for nonconceptualism in the following way. If transcendental idealism is true, our cognition faces intrinsic constraints. These constraints are of such a sort, however, that they do not prevent certain kinds of cognition, whereas they thwart the possibility of some other kinds of cognition. More specifically, the constraints do not preclude mathematical cognition but they preclude metaphysical cognition. Now, once we have recognized the constraints, an attempt at a diagnosis about what underlies them can be made. Since the objects of our cognition qualify as appearances, mind-dependent objects, the constraints cannot be attributed to the constitution of objects. Rather, they should be ascribed to the subjective constitution of the mind. Thus, there must be a factor, attributable to the finite mind, which accounts for the constraints within our cognitive capacities. This factor, one may note, could be identified with sensibility, the receptive faculty of the mind. Sensibility accounts for intuitive cognition, both pure and empirical. Since intuitive cognition originates from sensibility, it does not suffice to warrant metaphysical cognition, which, according to Kant, requires a non-sensible kind of intuition. Thus, it is due to our sensible mode of cognizing that our cognition faces limitations. Since sensibility delivers intuitions and, on Kant’s Autonomy Thesis (KAT) expounded in chapter 3, intuitions are nonconceptual representations of objects, transcendental idealism motivates nonconceptualism. Obviously, a number of objections can be raised against the above argument. First of all, one may notice that the epistemic reading of transcendental idealism does not do justice to the metaphysical purport of the doctrine. While it focuses on cognitive constraints, it fails to account for the fact that the doctrine deals with the objects of our cognition as well. Furthermore, it explains cognitive constraints by appealing to the constitution of the human mind, that is, to certain psychological or anthropological factors. Indeed, the outlined reading seems to fail to distinguish between what has been called the empirical and the transcendental level of investigations. Certainly, contemporary theorists of nonconceptual content do not distinguish between the two levels either, even though, as noted in chapter 1, descriptive accounts of nonconceptual content can be separated from the normative ones. Second, one can also argue that KAT does not motivate nonconceptualism, especially when combined with what we earlier called Kant’s Cooperation Thesis (KCT). For Kant may distinguish cognitive faculties and their products but at the same time claim that perceptual experience of particular objects in space and time presupposes some kind of “consciousness of normativity”289 or reference to the

289  Cf. H. Ginsborg, “Was Kant a Nonconceptualist?” Philosophical Studies 137, 2008, pp. 71-76.



And How Does It Bear on Nonconceptualism? 

 107

framework of the categories.290 This would imply that the involvement of conceptual capacities (or the Kantian understanding) is necessary for perceptual experiences of particular objects. Furthermore, even if one concedes that KAT as such is incompatible with conceptualism, since it at least leaves room for the involvement of nonconceptual content in perceptual states, it does not have to follow that nonconceptualism is more in line with transcendental idealism. For it can be pointed out that what may account for the cognitive constraints may be sensory content (SC), rather than nonconceptual content (NCC), which is representational. In this way, cognition of objects as they are in themselves would be no option because the only kind of receptivity available to finite subjects would be provided by the senses, but the senses would not be able to deliver any representations of objects. All representations of objects, therefore, would depend on the activity of the mind to produce representations. As a result, the objects of representations could not be given to us as they are in themselves but only in minddependent appearances. Thus, transcendental idealism would motivate the claim that the involvement of the spontaneous faculty of the mind, the understanding, as A. M. Griffith has argued, is essential to generate representations of objects, and so it would lend support to conceptualism.291 Finally, one may argue that there is little in common between transcendental idealism and the idea that perceptual experience involves nonconceptual content, too little to suggest a kind of logical entailment between claims asserting the former and the latter. For example, Robert Hanna thinks that an argument for the Kantian nonconceptualism can rest on Kant’s account of incongruent counterparts (enantiomorphic figures), which was formulated in the pre-critical period, so the argument is “in fact neutral with respect to realism and idealism about space.”292 Besides, there are other, and better, ways of arguing for nonconceptualism than by appealing to a doctrine that many have regarded as a historical oddity. Whether the objects of cognition are appearances, hence mind-dependent, or things in themselves, existing outside the mind; whether we have cognitive access to reality as it is or to its appearances – seems to be quite irrelevant to the question about the constituents of perception. In other words, it seems perfectly plausible to adopt both nonconceptualism and what Kant labeled transcendental realism. It is

290 For an argument defending the claim that for Kant perceiving particulars requires the application of the categories as the rules for synthesizing manifolds of both pure and empirical intuition, see: Aaron M. Griffith, “Perception and the Categories: A Conceptualist Reading of Kant’s ‘Critique of Pure Reason,’” European Journal of Philosophy 20 (2), 2010. 291 A. M. Griffith concludes his paper with a similar observation: “Indeed, I take it to be a mark of Kant’s idealism that the content of perception – particularly the formal features of appearances – depends in part on the conceptual activity of the understanding. Thus, for Kant, perception is not merely a passive state in which we are receptive to the way the world is, but it involves our spontaneous capacity for conceptual activity.” Ibidem, pp. 216-17. 292 “Kantian Non-Conceptualism,” Philosophical Studies 137 (1), 2008, p. 55.

108 

 Nonconceptual Content and Transcendental Idealism

far from evident why the endorsement of the claim that sensibility furnishes us with a sui generis kind of cognitive content, or the endorsement of the claim that perceptual experience involves nonconceptual content or nonconceptual representational states, would be entailed by the view that this kind of cognition represents objects as they appear and not as they really are. In the following sections, I discuss the above objections. To that effect, first, I inquire whether my account of Kant’s nonconceptualism and its relation to the doctrine of transcendental idealism results in incoherence because it conflates the empirical and the transcendental level of considerations; and whether the confusion marks specifically the reading I suggest. Second, I inquire whether Kant’s transcendental idealism would also be compatible with the claim that the constraints originating from sensibility are provided by SC rather than NCC. Third, I look into possible connections between the Kantian account of experience and the debate between idealism and realism, primarily as presented by Kant in the Refutation of Idealism in the First Critique.

5.3 The Empirical and the Transcendental Supposing one does not reject transcendental idealism as an incoherent doctrine, one can endorse one of its prevalent interpretations: (1) the so-called two-worlds reading or (2) the so-called two-perspectives (or double-aspect) reading. Kant’s texts seem to provide evidence for both interpretations. Consider the following passage from the chapter on the paralogisms of pure reason in the A-edition of the Critique: “I understand by the transcendental idealism of all appearances the doctrine that they are all together to be regarded as mere representations and not as things in themselves, and accordingly that space and time are only sensible forms of our intuition, but not determinations given for themselves or conditions of objects as things in themselves.”293

The passage can be read as suggesting that there might be two modes in which objects can exist: either as appearances, or as things in themselves. Now this statement can be taken in two senses: either that for any object there are two modes in which this object can exist, or that for any object it is the case that it exists either as an appearance, or as a thing in itself. The second sense in which we can read the statement at issue encourages the two-worlds interpretation. On this interpretation, there are two kinds of entities constituting the Kantian universe: appearances and things in themselves. The former stand for the objects of cognition, the latter fall outside the scope of cognitively accessible objects. In principle, the relation between

293 I. Kant, Critique of Pure Reason, op. cit., p. 426 (A 369).



The Empirical and the Transcendental 

 109

appearances and things in themselves is inconceivable, on such a reading, because all the categories available to the subject can be meaningfully applied only to the domain of appearances. However, other passages make it clear that Kant would probably be more in favour of the alternative interpretation. Consider the following passage from one footnote in the B-edition Preface to the Critique of Pure Reason: “[…] the same objects can be considered from two different sides, on the one side as objects of the senses and the understanding for experience, and on the other side as objects that are merely thought at most for isolated reason striving beyond the bounds of experience. If we now find that there is agreement with the principle of pure reason when things are considered from this twofold standpoint, but that an unavoidable conflict of reason with itself arises with a single standpoint, then the experiment decides for the correctness of that distinction.”294

This passage clearly confirms the reading on which one and the same object can be regarded either as an appearance, or as a thing in itself, thus it can be viewed from two different perspectives. Now, the distinction of perspectives from which to consider objects brings no novelty from the purely historical point of view. After all, it is present in Spinoza’s Ethics, where one can read about two ways of looking at objects: sub specie aeternitatis and sub specie durationis, or from the point of view of natura naturans and natura naturata. Also Leibniz makes use of the distinction between the realm of appearances and the realm of things in themselves, which he calls monads. The significance of the distinction within Kant’s system derives from the recognition that things can be considered from different perspectives and hence that our, human, perspective of considering things is not in any sense unique or privileged as a cognitive perspective.295 Kant expresses this view many times when he speaks of possible entities equipped with a sensibility that differs from ours, or with a non-sensible kind of intuition.296 The kind of cognition that he provides an account

294 Ibidem, p. 111 (B xviii-xix). Interestingly, Kant maintains the two-perspectives view also in his post-critical period. See: Erich Adickes (ed.), Kants Opus postumum, Topos Verlag, Vaduz 1978, pp. 696-7 (C 563f.) 295 In other words, the finite (human) point of view on things does not guarantee the possibility of absolute knowledge – this, rather than the “anthropocentric turn,” is the true meaning of the socalled “Copernican revolution” of Kant. After all, with Copernicus, the picture of the earth as in the center of the universe has been replaced with the picture of the earth as one of many planets in the solar system. Cf. Murray Miles, “Kant’s ‘Copernican Revolution’: Toward Rehabilitation of a Concept and Provision of a Framework for the Interpretation of the Critique of Pure Reason,” Kant-Studien 97 (1), 2006. 296 See, e.g., the following passage from the Transcendental Aesthetic: “For we cannot judge at all whether the intuitions of other thinking beings are bound to the same conditions that limit our intuition and that are universally valid for us.” I. Kant, Critique of Pure Reason, op. cit., p. 177 (A 27/B 43). And: “It is also not necessary for us to limit the kind of intuition in space and time to the sensibility of human beings; it may well be that all finite thinking beings must necessarily agree with human beings in this regard (though we cannot decide this), yet even given such universal validity this

110 

 Nonconceptual Content and Transcendental Idealism

of, in the Critique of Pure Reason, should be viewed as cognition of a certain type of subjects. Hegel realized this fact and made a charge against the coherence of Kant’s transcendental philosophy out of it. To that effect, in Faith and Knowledge, he says: “Kant has simply no ground except experience and empirical psychology for holding that the human cognitive faculty essentially consists in the way it appears.”297

If Hegel’s observation is correct, one may claim that Kant’s theory of cognition is simply inconsistent: for, on the one hand, it purports to proceed in an a priori manner, when establishing the necessary conditions of cognition, but on the other hand it builds on an empirical observation that these conditions are valid only for subjects of a certain kind. In this way, transcendental considerations become confused with empirical psychology or anthropology. However, it is easy to fend off the Hegelian charge by pointing out that it is the core of Kant’s philosophical method to start from an empirical fact and proceed to revealing the conditions that make this fact possible. In the light of this, Hegel’s charge may appear as quite trivial. But it will seem less so if we consider it as expressing criticism towards Kant’s conflation of the empirical and the transcendental level of considerations. Henry Allison characterizes these levels in the following way: “[…] the distinction between the transcendental and the empirical, as it operates in the Critique, is perspectival rather than substantive. This means that it does not serve to distinguish two kinds of objects, modes of being or activities, but two perspectives or standpoints from which the above can be considered.”298

Thus, Hegel’s charge would boil down to a remark that Kant conflates the two perspectives from which to talk about objects: when he distinguishes between sensibility and understanding, he speaks from the empirical point of view, but when he refers to the a priori representations of these faculties, he adopts the transcendental perspective. Accordingly, claims about the origin of certain representations in the constitution of the subject seem to be made from the empirical point of view, whereas claims that certain representations make determinate kinds of cognition possible can be recognized as belonging to the transcendental level of considerations. In this way, Kant becomes subject to the charge of psychologism by grounding the transcendental

kind of intuition would not cease to be sensibility, for the very reason that it is derived (intuitus derivativus), not original (intuitus originarius), thus not intellectual intuition, which for the ground already adduced seems to pertain only to the original being, never to one that is dependent as regards both its existence and its intuition (which determines its existence in relation to given objects).” Ibidem, pp. 191-2 (B 72). 297 Georg W. F. Hegel, Faith and Knowledge, op. cit., p. 89. 298 H. E. Allison, “The ‘Critique of Pure Reason’ as Transcendental Phenomenology” in: D. Ihde, R. N. Zaner (eds.), Dialogues in Phenomenology, Martinus Nijhoff, The Hague 1975, p. 145.



Sensory Content and Cognitive Constraints 

 111

structure of cognition in the constitution of an empirical subject – ironically enough, since, as a matter of fact, Kant held empirical psychology in quite low esteem.299 Particia Kitcher also identifies an inconsistency in Kant’s transcendental theory of cognition but in the context of the problem of the ideality of time, a claim characteristic of transcendental idealism. Also, this issue casts light on a fundamental difficulty within Kant’s system, which Kitcher describes in the following way: “The inconsistency is between the theory of apperception and transcendental psychology generally and the doctrine of the ideality of time. If time is not real, then the accounts of the identity of a mind through time, and of the cognitive processes that enable us to have knowledge, are incoherent. Conversely, if the theories of synthetic processing and apperception are true, then time is real. Under these circumstances I see no choice but to reject the metaphysical claim, which is, in any case, independently problematic. This is a drastic move within Kant’s system. Still, it does not mean a total rejection of transcendental idealism. For in keeping transcendental psychology, we also keep the source of all the arguments that our knowledge is influenced by the structure of our minds. And that doctrine is at least as central to transcendental idealism as the theories of the ideality of space and time.”300

In light of these considerations, one can only reiterate that transcendental idealism, as a doctrine about the epistemic constraints of finite subjects, rests on a psychological or anthropological assumption. And it is on the basis of this assumption that Kant builds his model of cognition. Whether this assumption – concerning the division of human cognitive faculties into sensibility and understanding – is warranted or not is a separate issue.301 But, on closer examination, one can see that Kant’s dualism of sensibility and understanding followed decade-long debates with the Leibnizian philosophy, and its introduction in the Critique of Pure Reason is not without any grounding. (Of course, this does not prove yet that Kant’s arguments are plausible.) To these, mainly historical, issues I return by the end of this chapter.

5.4 Sensory Content and Cognitive Constraints I have claimed that, in Kant’s theory of cognition, nonconceptual content (NCC) accounts for constraints formulated in terms of transcendental idealism. But it also seems to be possible to conceive of cognitive constraints as coming from sensory

299 Cf. I. Kant, Critique of Pure Reason, op. cit., p. 700 (A 849/B 877). 300 P. Kitcher, Kant’s Transcendental Psychology, op. cit., p. 141. 301 For example, Lorne Falkenstein remarks that “the sense/intellect distinction is not one for which Kant ever obviously argues, or even explains in anything more than the most perfunctory way,” and adds that Kant must have borrowed the distinction from Aristotle and the Schools. L. Falkenstein, Kant’s Intuitionism. A Commentary on the Transcendental Aesthetic, University of Toronto Press, Toronto 2004, p. 29.

112 

 Nonconceptual Content and Transcendental Idealism

content (SC). In this case, transcendental idealism would in fact be motivated by conceptualism. This is because transcendental idealism would be entailed by the fact that all representational content would be originating from the spontaneous activity of the mind, and so would not be “given” by the objects of experience. Indeed, in the Transcendental Deduction of the Categories, Kant argues that perception of objects presupposes an activity called synthesis, and this cannot set off without the involvement of the understanding, albeit indirectly, that is, through imagination. A number of Kant scholars have advocated a reading in which Kant combines conceptualism with sensationism. Among these scholars one can find W. Sellars, L. Falkenstein, R. George and K. Westphal. The reading is compatible with conceptualism since it supports the claim that all representational content is conceptual through and through. Indeed, Kant seems to side with conceptualism when he states that “the manifold in a given intuition […] necessarily stands under categories.”302 Now, the reading under consideration is in conformity with sensationism as a view about the existence of non-intentional, nonrepresentational states, which do not refer to any objects but which, in a way, underlie cognition, providing the “given” which does not presuppose any cognitive activity. However, arguments can be raised to show that reading Kant’s theory of experience in terms of sensationist conceptualism may reveal a few inconsistencies within Kant’s system. First of all, it leads to the so-called “problem of affection” or “noumenal causality.”303 Second, this reading leaves the non-conceptual nature of the pure intuitions of space and time unexplained: for if they are representations, they must be conceptual, but if they are conceptual, they cannot be intuitions. Let me take a brief look at these two charges against the interpretation that promotes sensationist conceptualism.

5.4.1 “The problem of affection” Suppose that our mind becomes equipped with SC by means of objects affecting it. But this affection by objects does not suffice to equip the mind with representations of objects, since representing objects presupposes the synthesizing activity of the mind. What do we then refer to when we speak about the object or objects that account for the affection relation? The answer is that we refer to an unknown or hypothetical

302 I. Kant, Critique of Pure Reason, op. cit., p. 252 (B 143). 303 The problem is discussed, e.g., in: H. E. Allison, Kant’s Transcendental Idealism. An Interpretation and Defense, op. cit. (Chapter 3: “The Thing in Itself and the Problem of Affection”); Nicholas Rescher, “Noumenal Causality” in: Lewis White Beck (ed.), Proceedings of the Third International Kant Congress, D. Reidel Publishing Co., Dordrecht 1972; N. Rescher, “On the Status of ‘Things in Themselves’ in Kant,” Synthese 47 (2), 1981.



Sensory Content and Cognitive Constraints 

 113

object, if to anything. But how can we refer to unknown (and unknowable) objects by means of the concepts (or categories) we use when we refer to the objects we can acquire cognition of? Legitimately, of course, we cannot. Therefore, by supposing that constraints on our cognition derive “from without” the mind we get involved in incoherence. Read in this way, transcendental idealism remains a paradoxical position, as noted by Friedrich H. Jacobi.304

5.4.2 Space and Time In the Inaugural Dissertation and the Transcendental Aesthetic in the Critique of Pure Reason, Kant argues in favour of the nonconceptual character of the pure intuitions of space and time. If all, pure and empirical, intuitions were conceptual or conceptdependent, space and time would not be representations, thus they would not be intuitions in the end. But this is clearly false on Kant’s account. On the other hand, pace Kant’s separation experiment,305 it is difficult to imagine the objects of pure intuitions. For what objects do space and time represent? If I have an intuition of a dog, then I represent a certain particular object with a number of features and falling under the concept “dog.” But the same is simply not true in the case of the intuitions of space and time. Space and time cannot be objects of my representations in the same way as dogs can. One cannot refer to space and/or time by means of a demonstrative expression “this x,” which one can use in order to refer to particular objects in space and/or time. But this is no insurmountable challenge, considering the fact that also pure concepts do not refer to objects in the same way as empirical concepts do. For, according to Kant, a priori representations refer to objects of possible experience. One could say, thus, that space and time represent possible objects of experience – as much as an empty yard in front of a house represents (a place for) a possible garden. Since, as Kant remarks in the Introduction to the First Critique, the conditions of sensibility are prior to the conditions of the understanding,306 the possibility of thought and judgment about objects must be determined by the possibility of representing these objects in space and/or time. If all intuitive representations were conceptual or concept-dependent, it is difficult to see how space and time could provide limits to predication of objects in thought and judgment.

304 F. H. Jacobi, David Hume über den Glauben; oder Idealismus und Realismus: Ein Gespräch, Loewe, Breslau 1787. Quoted after: W. Martin, “From Kant to Fichte” in: G. Zöller (ed.), The Cambridge Companion to Fichte, Cambridge University Press, Cambridge (forthcoming). 305 Cf. I. Kant, Critique of Pure Reason, op. cit., p. 156 (A 21/B 35). 306 Ibidem, p. 152 (A 16/B 30).

114 

 Nonconceptual Content and Transcendental Idealism

Transcendental idealism restricts our cognition to appearances, objects dependent in their constitution on the mind’s cognitive capacities. But constraints can be taken as originating either from outside the domain of empirical cognition, or from within that domain. Consider the first option. If there were no constraints coming from outside, from the unknown object or objects affecting the mind, the “space of concepts” would be “unbounded,” as McDowell puts it;307 there would be no reason why thought and judgment would have to face any limitations. This is the case in Hegel’s absolute idealism, from which the unknown “thing in itself” has been rejected. With the thing in itself in his system, Kant remains subject to the “Myth of the Given” objection; by rejecting it, he would make a room for an unrestricted activity of the mind, inviting back again the illegitimate metaphysics. If all representational content of experience were conceptual, or concept-dependent, space and time would have to be considered either as products of the synthesizing activity of the mind, in which case they would cease to be constitutive of experience, or as the nonrepresentational kind of content, in which case they would be equated with sensations, an equally implausible suggestion. Consider, however, the second option on which cognition is restricted to appearances because of some intrinsic features of human cognitive capacities. On such an account, time and space would provide the most basic forms of intentionality; they would determine the possible scope of objects that our minds can be related to. Accordingly, one may observe, as much as for Leibniz “to be” means “to be possible,” where “to be possible” means “to be thinkable,” or “to be non-self-contradictory,” for Kant “to be” also means “to be possible,” where “to be possible” means “to be really, not only logically possible,” thus “to be representable in space and/or time,” rather than merely “representable in thought.” Consequently, transcendental idealism does not have to be equated with the endorsement of the two-worlds view with its mysterious relation between the objects of experience and the unknown things in themselves. On Kant’s account, one can realize that cognition is intrinsically limited in the course of reflecting upon its necessary conditions, a task to be carried out from the transcendental, rather than empirical, perspective, pace Hegel. (In fact, Kant applies deduction where Locke would be content with reflection. Thus, rather than about “reflecting” upon the necessary conditions of cognition, one should talk about the conditions being deduced on the basis of a given possibility, a method diametrically opposed to the straightforward introspective reflection.)

307 See the second chapter of Mind and World, “The Unboundedness of the Conceptual.”



Empirical Realism, Transcendental Idealism, and Nonconceptual Content 

 115

5.5 Empirical Realism, Transcendental Idealism, and Nonconceptual Content When asked to report on what one can see when looking around one’s environment, one would usually indicate physical objects in space as the constituents of one’s perceptual field. Often, one would take these objects to be located within a considerable distance from oneself as a perceiver. Certainly, one would describe them as existing outside one’s mind. For, we can assume, in ordinary life and practice direct realism about the objects of perception seems to be the most common attitude. This view, represented by Aristotle, Thomas Reid, John McDowell, and others, equates the objects of perception with mind-independent and (usually) physical objects in space, rather than with ideas, sense data, or other kinds of mental content. Thus, direct realism comes close to disjunctivism in that it also rejects the “highest common factor” or the “veil of perception” doctrine. It is when reflecting upon particular cases of perceptual illusions or hallucinations, or even fairly common phenomena related to perception of size and distance, that one may start to see problems in direct realism. The moon you can see at night is, let us say, four inches in diameter, but you certainly know that the real moon’s diameter is much more than four inches in length. Looking straight into the sun on a bright summer day for a couple of minutes makes you see patches of yellow, somewhere in your visual field, as soon as you shift your gaze onto a less bright and shiny object. Houses and trees kilometers away from the place you are in appear to be smaller than matches in a box you take out from your pocket; but of course in reality they are not. So, it seems that no highly sophisticated experiments are needed in order to challenge the direct (or naïve) realist view which equates the objects of perception with physical objects in space and/or time. The question about what we perceive relates to both the object and the content of perception. But the distinction between content and object may be challenged: first, for being at odds with the phenomenology of experience and, second, for encouraging skepticism about the possibility of knowing objects through perception. One who adopts the distinction may think that objects are experienced as being, looking, or appearing in certain ways. These ways make the content of experience. So, if the moon you see at night is four inches in diameter, it is not because it is of that size in reality, but because you perceive it as being four inches, or because it looks to you as such. Thus, distinguishing between content and object motivates the distinction between things as they appear and things as they are in reality. By denying direct epistemic access to the latter, while taking the former for the primary objects of perception, one makes a step towards epistemological idealism. By denying any access to things as they really are, not even through their representations or ideas, one commits oneself to idealism entirely.

116 

 Nonconceptual Content and Transcendental Idealism

Kant, it seems, endorses the distinction between content and object on the transcendental level of considerations, but he rejects the distinction on the empirical one. From the transcendental perspective, the content of cognition, representations, becomes sharply distinguished from the objects existing outside the epistemic relation, things in themselves. Since there is no cognitive access to the latter, Kant’s position is clearly idealistic. From the empirical point of view, however, representations, or appearances, are the objects of cognition: there are no intermediaries between perception and the objects perceived. Consequently, if one considers Kant’s theory of experience against the background of the conceptualism vs. nonconceptualism debate, which assumes the distinction between content and object, one cannot ignore the transcendental aspect of this theory. This is why I do not agree with Robert Hanna when he says that an argument for nonconceptualism, which “has a distinctively Kantian provenance” is “defensible on grounds that are altogether logically independent of the question whether transcendental idealism is true or false.”308 Kant dedicates some space to discussing realism and idealism. He does this, for example, in the Refutation of Idealism in the B-edition of the Critique of Pure Reason,309 in the section on the fourth paralogism of the A-edition,310 or in an appendix to the Prolegomena, where he addresses charges and misunderstandings, which emerged among his readers after Christian Garve’s review of the First Critique.311 Thus, Kant distinguishes material idealism from transcendental idealism, which he also calls formal or critical. Material idealism, according to Kant, can be either dogmatic, as in the case of Berkeley, or problematic, as in the case of Descartes. Dogmatic idealism, Kant explains, takes space to be a mere illusion, whereby it denies existence to objects outside the mind. Problematic idealism, in turn, allows of the existence of mind-independent objects in space but as a mere hypothesis, which cannot ultimately be proved. Yet, transcendental idealism claims that the existence of objects in space (as appearances, not as things in themselves) cannot be questioned. For this reason, it can be equated with empirical realism, which stands in opposition to transcendental realism and empirical idealism. Kant contends: “The transcendental idealist […] can be an empirical realist […], i.e., he can concede the existence of matter without going beyond mere self-consciousness and assuming something more than the certainty of representations in me […]. For because he allows this matter and even its inner possi-

308 R. Hanna, “Kantian Non-Conceptualism,” op. cit., p. 55. 309 I. Kant, Critique of Pure Reason, op. cit., p. 326 (B 274-5). 310 Ibidem, pp. 426-31 (A 369-380). 311 I. Kant, Prolegomena to Any Future Metaphysics. With Selections from the Critique of Pure Reason, translated and edited by G. Hatfield, Cambridge University Press, New York 2004, pp. 123-134 (4:372-83).



Empirical Realism, Transcendental Idealism, and Nonconceptual Content 

 117

bility to be valid only for appearance – which, separated from our sensibility, is nothing – matter for him is only a species of representations (intuition), which are called external, not as if they related to objects that are external in themselves but because they relate perceptions to space, where all things are external to one another, but that space itself is in us. […] Thus the transcendental idealist is an empirical realist, and grants to matter, as appearance, a reality which need not be inferred, but is immediately perceived.”312

In the Refutation of Idealism, Kant argues that we can be certain of the existence of physical objects in space and that we perceive these objects immediately.313 The spatiality of objects guarantees their existence outside the mind, understood as a collection of “inner” states, temporally structured representations of the inner sense. But because space, from the transcendental perspective, is a form of sensible intuition, objects perceived in space depend on the mind, understood as a certain model: a set of necessary conditions of empirical cognition. (In the transcendental sense, all representations, and so all possible objects of cognition, have a status analogous to the status of modes in Spinoza’s ontology.) Kant’s argument can be outlined in the following way: 1. We have empirical self-consciousness, that is, consciousness of our mental states, given in inner sense, and consciousness of our own cross-temporally persisting self. 2. Perception of a cross-temporally persisting object is perception of something permanent, which remains the same throughout changes in time. 3. Perception of a permanent object is not given in inner sense, nor could perception of time itself furnish us with perception of a permanent object (in fact, we do not perceive time). 4. Since inner sense does not furnish us with perception of a permanent object, and since, nevertheless, we do have consciousness of a cross-temporally persisting object, which is our own self, the source of the consciousness of our own crosstemporally persisting self must be sought elsewhere than the inner sense. 5. Objects can be perceived, or represented, either in inner sense, the form of which is time, or in outer sense, the form of which is space. 6. Therefore, perception of a permanent object must be possible through outer sense, that is, in space. 7. And the empirical self-consciousness entails and/or presupposes, hence must also be immediate with, the consciousness of an object in space. In other words, to the extent that self-consciousness is an immediate experience, so is the consciousness of an external object, an object “outside” the mind.

312 I. Kant, Critique of Pure Reason, op. cit., p. 426-7 (A 370-1). 313 Ibidem, pp. 326-9 (B 274-9).

118 

 Nonconceptual Content and Transcendental Idealism

We can observe that, on the phenomenological level, consciousness of one’s crosstemporally identical self, of one’s mental states and external objects in space occurs, as it were, simultaneously. This is reminiscent of what cognitive scientists call proprioception: an immediate and “internal” experience of one’s own body, literally, perception of one’s body as one’s own. More specifically, proprioception can be defined as “intracorporal tracking of somatic location and limb position” and it “accounts for one’s ability to detect limb position and bodily posture from the inside.”314 Kant refers several times to experiences which could be classified as proprioceptive, or experiences of one’s own body: in the 1768 essay on directions in space, where he describes a “feeling” by means of which we distinguish between the right and the left side of our body,315 in the 1786 essay What Does It Mean to Orient Oneself in Thinking?316 or in the Prolegomena, where he discusses enantiomorphic figures (incongruent counterparts), such as one’s left and right hand, or one’s hand and its reflection in a mirror.317 On the basis of the argument outlined above and the textual evidence one may surmise that Kant conceived of the subject of cognition as embodied. An alternative argument supporting this supposition could be developed thus. Consciousness of one’s self and of one’s mental states is contingent upon the possibility of representing the objects of outer sense, the form of which is space. This means that a subject that could not represent any objects (material bodies) in space, could not entertain an empirical consciousness of his own mental states. According to Kant, “the consciousness of my own existence is at the same time an immediate consciousness of the existence of other things outside me.”318 Now, if the consciousness of other objects, existing outside me, in space, entails the capacity of locating these objects relative to a reference point, connected with or within the scope of the subject’s perceptual field, then the subject of experience could be conceived as an embodied entity, since the subject’s body would provide the best candidate for such a reference point.

314 Ellen Fridland, “The Case for Proprioception,” Phenomenology and the Cognitive Sciences 10 (4), 2011, p. 523. Some authors, like Shaun Gallahger and J. L. Bermúdez, argue that proprioception fails to meet the criteria of object perception, such as the identification constraint. That is, in the case of proprioception the body is not directly present to us unless we direct our attention to it or its parts. But one could reply that without the awareness of one’s body, albeit non-conscious, it would not be possible to carry out even simple tasks that require integration of the input of visual, haptic, etc. perception. 315 I. Kant, “Concerning the Ultimate Ground of the Differentiation of Directions in Space” in: Theoretical Philosophy, 1755 – 1770, translated and edited by D. Walford in collaboration with R. Meerbote, Cambridge University Press, New York 1992, p. 369 (2:380-1). 316 I. Kant, “What Does It Mean to Orient Oneself in Thinking?” in: Religion and Rational Theology, translated and edited by A. W. Wood, G. Di Giovanni, Cambridge Univeristy Press, Cambridge – New York 1996, pp. 8-9 (8:134-5). 317 I. Kant, Prolegomena to Any Future Metaphysics, op. cit., pp. 37-8 (4:286). Kant’s examples of enantiomorphic figures are mentioned in chapter 2. 318 I. Kant, Critique of Pure Reason, op. cit., p. 326 (B 276).



Empirical Realism, Transcendental Idealism, and Nonconceptual Content 

 119

This consciousness of external objects, relativized to the subject’s body, although a complex state, does not seem to depend on the conceptually-structured knowledge of relations between objects. Rather, it underlies reference to objects in thought, belief, or judgment. Thus, its content is nonconceptual and may be likened to Peacocke’s “scenarios” – body-relative arrangements of perceptual content.319 What are the implications of these considerations for the debate concerning the content of perceptual experience? I have suggested that different views on the distinction between content and object motivate different idealist and realist positions. But also the distinction between conceptual and nonconceptual content proves relevant, to an extent, to the idealism vs. realism debate. For example, if we endorse the conceptualist claim that the content of experience is conceptual throughout, we may arrive at a view on which there are no constraints to conceptual capacities, along the lines of Hegel’s absolute idealism. There is then no way to “save” the reality, since reality is identified with what falls within the sphere embraceable in concepts. For McDowell, the problem is spurious since, on his account, the objects of cognition yield the required constraints. But, as I argued in chapter 1, this position is fairly contentious, as it requires that one accept the “factive” account of perceptual experience, based on the assumption, which is far from obvious, about correspondence between thought and reality. We have seen that Kant accepts this assumption as an empirical realist but rejects it as a transcendental idealist. It may seem that sensationist conceptualism accommodates realist intuitions. After all, it secures reference to mind-independent reality by stipulating that the reality furnishes material, i.e. sensory content, for cognition. Constraints would then come from the impingements of objects upon cognitive faculties. But we have seen that, by Kant’s lights, such a supposition remains problematic. If constraints upon cognition came from outside the mind and all representational mental content were conceptual, it would be difficult to explain the subject-dependent, nonconceptual and cognitively basic status of space and time, the a priori forms of intuition. Indeed, denying the cognitively basic status of space and time would take Kant’s doctrine close to Berkeley’s “dogmatic idealism” (in Kant’s terminology). Berkeley’s empiricism about space perception leads to reductionism since space perception, in this view, becomes construed in terms of more basic cognitive occurrences, such as tactile, visual, and kinesthetic sensations. Such an account encourages constructivism about the representation of space; only sensations can be considered as “given.” Strictly speaking, for Berkeley, there are no objects in space, a claim incompatible with Kant’s empirical realism.

319 Peacocke characterizes scenarios as “spatial types,” more specifically, as “a way of locating surfaces, features and the rest in relation to […] a labeled origin and family of axes.” “Scenarios, concepts and perception” in: T. Crane, The Contents of Experience, Cambridge 1992, p. 107.

120 

 Nonconceptual Content and Transcendental Idealism

However, the nonconceptualist reading accommodates, first, the account of space and time as cognitively basic (non-derivative), non-conceptual (intuitive) representations; second, the need for constraints on cognition, which become established on the transcendental level of considerations; third, the intuitions of empirical realism on which we directly cognize external objects in space, rather than having access to them through the representations of “inner sense.” All these three elements are crucial components of the Kantian account of experience.

5.6 Concepts, Intuitions, and Transcendental Idealism320 The idea of transcendental idealism boils down to the claim that the only domain which subjects (“like us”) have cognitive access to is one of appearances, not things in themselves. But Kant’s faculty dualism, on which some philosophers have claimed that transcendental idealism rests, may be taken as based on psychological or anthropological observations. This can indeed make it liable to a number of objections, such as one formulated by Hegel, who wrote (as we remember from the beginning of this chapter): “Kant has simply no ground except experience and empirical psychology for holding that the human cognitive faculty essentially consists in the way it appears, namely in this process from the universal to the particular or back again from the particular to the universal.”321

Hegel rightly attributes to Kant the view that cognition consists in the “transition” from the particular, i.e. the object given in intuition, to the general, i.e. judgment. However, in contrast to Kant, he postulates construing the whole process as not involving two distinct faculties producing representations the combination of which requires a special act of the mind. Combining heterogeneous representations gives way to an activity of the mind that consists of two “moments” which involve the particular and the general. The activity is performed by imagination, thus by one faculty. In Faith and Knowledge, Hegel says: “This power of imagination is the original two-sided identity. The identity becomes subject in general on one side, and object on the other; but originally it is both. And the imagination is nothing but Reason itself […] But it is only Reason as it appears in the sphere of empirical consciousness.”322

320 This section is a modified version of a part of my article “Transcendental Idealism, Intuitions and the Contents of Perceptual Experience” in: S. Bacin, A. Ferrarin, C. LaRocca, M. Ruffing (eds.), Kant und Philosophie in weltbürgerlicher Absicht. Akten des XI. Kant-Kongresses 2010, De Gruyter, Berlin/ Boston 2013. 321 G. W. F. Hegel, Faith and Knowledge, op. cit., p. 89. 322  Ibidem, p. 73.



Concepts, Intuitions, and Transcendental Idealism 

 121

This is an abstruse way of rendering Kant’s characterization of the transcendental imagination as embodying both spontaneity and receptivity. According to Hegel, imagination instantiates “fusion” of the understanding and sensibility. Hegel purports to overcome Kant’s faculty dualism by pointing to a third mediating faculty as a kind of “coincidence of opposites.” By identifying imagination with reason or an intuitive understanding, he intends to show that there are no intrinsic bounds to cognition, and therefore that transcendental idealism is false. Since imagination combines the features of the faculty of concepts with the features of the faculty of intuitions, Hegel equates it with an intuitive intellect. Thus, he suggests that already within Kant’s system there are ways by which to overcome the limitations imposed on human cognition by the doctrine of transcendental idealism. As defined by Kant, the doctrine states that no objects are cognizable save ones we can be acquainted with through sensibility. The possible objects of sensibility, the “undetermined object[s] of an empirical intuition,”323 Kant calls appearances. Objects considered outside the epistemic relation, thus in abstraction from the sensible mode of cognition, are things in themselves. Insofar as restoring the role of the “intuitive intellect” as a faculty of cognition is to invalidate transcendental idealism, it should be tantamount to extending cognition over things in themselves. However, saying that subjects are capable of cognizing things in themselves is not exactly the same as saying that subjects are able to cognize by means of intellectual intuition. We can conceive of cognition for which no engagement of intuition would be necessary (of course, this is not Kant’s model). Cognition of this kind would be intended at things in themselves, but it would not have to make use of the intuitive understanding. But, to an extent and in an important respect, Hegel’s interpretation is right; at least it allows us to identify a problem which Kant’s theory of experience might have difficulties solving. The problem emerges with the conception of judgment as a “process from the universal to the particular or back again from the particular to the universal,” which we agreed to attribute to Kant. The quoted passage may be paraphrased as stating that judgment paradigmatically involves concepts and intuitions. Accordingly, intuitions would have to be produced by the higher faculty of the mind, since they form part of judgments, which are generated by the understanding. For Kant, this is clearly false. But then it is hard to see how to provide a satisfactory account of judgment. For when one denies that intuition co-constitutes the content of judgment, one cannot maintain that judging consists in relating general representations (concepts) to singular representations (intuitions). If intuition does not co-constitute the content of judgment, it remains fairly unclear how judgments can relate to objects representable in intuition. At any rate, Hegel’s reading of Kant’s theory of empirical cognition, though heterodox, prompts reflection upon its fundamental presuppositions.

323  I. Kant, Critique of Pure Reason, op. cit., p. 155 (A 20/B 34).

122 

 Nonconceptual Content and Transcendental Idealism

If Kant’s faculty dualism can be superseded along the lines of Hegel and McDowell, and if this dualism underlies Kant’s overarching project, the doctrine of transcendental idealism, then undermining the former leads outright to undermining the latter. Key in the Hegelian interpretation is equating the cooperation of sensibility and understanding with the synthesizing activity of the imagination. Contributions of the two faculties are then in effect regarded as two aspects of one and the same thing. This ultimately invalidates Kant’s Autonomy Thesis (KAT). However, an easy rejection of both transcendental idealism and the faculty dualism cannot be afforded. Kant’s arguments supporting the claim that intuition provides its own concept-independent contribution to cognition are diverse and intricate and were developed for more than a decade preceding Kant’s critical period.324 Concepts without intuitions are empty, that is, objectless and meaningless. But, according to the Stufenleiter passage, concepts refer to objects by definition – for Kant calls them “objective perceptions,” that is “cognitions.”325 Since intuitions are required to furnish concepts with content, the mere objective character of a representation is not tantamount to its being content-laden. Accordingly, objectivity might be taken in two senses: either as a “formal” feature of a representation, in which case concepts are object-directed apart from and regardless of their being associated with any corresponding intuitions; or as a “material” feature of a representation which it acquires by virtue of becoming related to an object in intuition. A minimal requirement for a concept to be objective in the first of these senses is that it should not be self-contradictory. In the chapter on phenomena and noumena, in the Critique of Pure Reason, Kant writes: “For every concept there is requisite, first, the logical form of a concept (of thinking) in general, and then, second, the possibility of giving it an object to which it is to be related. Without this latter it has no sense, and is entirely empty of content, even though it may still contain the logical function for making a concept out of whatever sort of data there are. Now the object cannot be given to a concept otherwise than in intuition […].”326

And in an essay entitled What Real Progress Has Metaphysics Made in Germany since the Time of Leibniz and Wolff? one can read: “The possibility of a thought or concept rests on the principle of contradiction, e.g., that of a thinking immaterial being (a spirit). The thing of which even the mere thought is impossible (i.e. the concept is self-contradictory) is itself impossible. But the thing of which the concept is possible is not on that account a possible thing. The first possibility may be called logical, the second,

324 Cf. Kant’s essays, such as: Inquiry Concerning the Distinctness of the Principles of Natural Theology and Morality (1764), Concerning the Ultimate Ground of the Differentiation of Directions in Space (1768), On the Form and Principles of the Sensible and the Intelligible World (1770). 325 I. Kant, Critique of Pure Reason, op. cit., pp. 398-9 (A 320/B 376f). 326 Ibidem, pp. 340-1 (A 238f./B 298).



Concepts, Intuitions, and Transcendental Idealism 

 123

real possibility; the proof of the latter is the proof of the objective reality of a concept, which we are entitled to demand at any time. But it can never be furnished otherwise than by presentation of the object corresponding to the concept; for otherwise it always remains a mere thought, of which, until it is displayed in an example, it always remains uncertain whether any object corresponds to it, or whether it be empty, i.e., whether it may serve in any way for knowledge.”327

Thus, Kant distinguishes between the real and the logical possibility of objects. He also employs such terms as “objective validity” and “objective reality” which can be regarded as synonyms. An object is logically possible if there is a concept of this object which is not self-contradictory or which does not entail a contradiction. But an object is really possible if there is an intuition that corresponds, actually or in a possible course of experience, to the concept of this object. The distinction between the real and the logical possibility of objects could not be made without the assumption that intuition provides its own kind of contribution to cognition, which could not be afforded by means of concepts alone. In other words, the distinction at issue presupposes the dualism of concepts and intuitions. Now, according to Kant’s Cooperation Thesis (KCT), cognition demands a joint contribution from sensibility and understanding, hence of necessity it involves both intuitions and concepts. Therefore, since only concepts marked by objective reality are accompanied by intuitions, it follows that only really possible objects can be cognized, as opposed to merely logically possible ones. In this way, Kant removes the objects of the rationalist metaphysics from the realm of the objects of cognition. Intuition makes its separate contribution to cognition. If this contribution provides a necessary condition of cognition, then cognition faces limitations. Indeed, Kant curbs pretensions of the rationalist metaphysics, but at the same time he maintains a rationalist demand for absolute certainty in science. He achieves this by offering a theory of synthetic a priori judgments. Such judgments are both necessary and objectively valid. They are necessary due to their apriority that, for Kant, amounts to independence from experience. They are synthetic by virtue of being related to really possible objects. This means that there must be intuitions which correspond to, or which are synthesized with, concepts employed in this kind of judgments. The paradigmatic instances of synthetic a priori judgments are provided by the judgments of pure mathematics, in particular geometry – the subject matter of the Transcendental Aesthetic. This claim builds on the assumption that we have pure intuitions of space and time, the forms of all empirical intuitions. Let us grant that we can have the intuitions of space and time only if we assume that they are the forms of all empirical intuitions. If space and time are the forms of intuition, they are neither things in themselves, nor relations between things in themselves.328 This means that

327 I. Kant, Theoretical Philosophy after 1781, edited by H. Allison and P. Heath, translated by G. Hatfield et al., Cambridge University Press, New York 2002, p. 406 (20:325-6). 328 Cf. § 8 of the Transcendental Aesthetic in the Critique of Pure Reason.

124 

 Nonconceptual Content and Transcendental Idealism

they are transcendentally ideal. Thus, Kant’s faculty dualism underlies his theory of pure mathematics as synthetic a priori cognition and motivates transcendental idealism. Concepts that are not self-contradictory express a logical possibility of the objects they are supposed to represent. But only those concepts that are accompanied by intuitions are objectively real, thus they represent really possible objects. According to KCT, only such concepts are suitable for the cognition of objects. Since this principle obtains universally, in order to argue that cognition can go beyond the bounds of sensibility, one would have to demonstrate that, in some cases, an alternative, nonsensible mode of intuiting becomes involved. For the above stated principle pertains not only to the cognition of appearances but also to noumenal cognition, and to the cognition of the putative intelligible objects. However, Kant insists that this alternative kind of intuition must be denied to our cognition; this claim reinforces transcendental idealism. Kant writes: “To be sure, above we were able to prove not that sensible intuition is the only possible intuition, but rather that it is the only one possible for us; but we also could not prove that yet another kind of intuition is possible.”329

Why does Kant preclude the possibility of “another kind of intuition”? Why does he deny the ability to intuit to the understanding? Does he bring in an unwarranted anthropological assumption based on empirical observations, as Hegel observed? There are philosophical theories that combine a minimalist approach to metaphysics with an epistemological position in which an appeal to non-sensible intuition plays a significant role. Descartes’ theory of intellectual intuition, presented in the Rules for the Direction of the Mind as one of the ways in which to acquire unshakeable truths, provides a good example, as does Husserl’s theory of eidetic intuition by means of which to arrive at knowledge of the essences of particular objects. Neither for Descartes, nor for Husserl does non-sensible intuition open a way to metaphysical cognition; it does not guarantee direct access to the objects of metaphysics. (According to Descartes, for example, we have resources to prove the existence of God, but we have no insight into God’s nature, which we cannot comprehend, or to the meaning of the dogmas of theology, which surpass our cognitive powers, although we can be certain of their truth.) I would venture to reply to the above questions in the following way: (1) Kant rejects intellectual intuition as a candidate for a source of cognitions since he intends to avoid blurring the distinction between the real and the logical possibility, which would be the case if the understanding were taken to account for both the form and the content of concepts. Besides (2), according to Kant, intuitions rest

329 I. Kant, Critique of Pure Reason, op. cit., p. 349 (A 252).



Concepts, Intuitions, and Transcendental Idealism 

 125

on affections, whereas concepts rest on functions. The understanding, if it were to deliver intuitions, would have to be affected by objects (or the object?) – the “ground” of representations. But then it would be, at least partly, receptive, a conclusion Kant would not at all readily acknowledge. By contrast (3), if intuitions were produced by the understanding, the spontaneous faculty of the mind, their objects would no longer play any role in the explanation of the nature of representations, and in effect the nature of appearances, the objects of cognition. This would entail reducing all the contents of cognition to representations generated solely by the mind, a corollary most unwelcome to Kant. In other words, slighting the independence of intuitions from the faculty of concepts could lead to the kind of idealism that Kant, unlike Hegel, was all but willing to endorse. Since Kant’s dualism of concepts and intuitions underlies transcendental idealism, and since the dualism is to be read as entailing KAT, which is definitive of the Kantian nonconceptualism, it follows that Kant is a nonconceptualist about (at least part of) the representational content of experience. In this chapter, I have argued that considering Kant’s theory of experience as a version of nonconceptualism is more compatible with the kind of idealism Kant professes, namely transcendental idealism, than doing so in terms of the competitive positions. I have interpreted transcendental idealism as a theory about epistemic constraints, intrinsic to the faculty and process of cognition. The source of these constraints, I have suggested, can be found in those aspects of cognition which involve the subject’s body, a hypothesis I earlier raised in chapter 2. Against this background, I have offered a reading of Kant’s argument from the Critique of Pure Reason, called the refutation of idealism, which I take to establish the hypothesis. In the next chapter, which closes the book, I address the question whether my interpretation necessitates attributing to Kant a commitment to naturalism.

6 Kant and Naturalism about the Mind 6.1 Spinoza and Hume In the previous chapters, I suggested to read Kant as a supporter of the idea of the embodiment of the subject of cognition. As I pointed out, one can argue that the origin of the forms of intuition can be sought in the bodily nature of the subject and that the awareness of oneself and of one’s mental states is contingent upon the awareness of a distinguished object in space, that is, one’s body. The bodily nature of the subject can also account for the cognitive constraints that transcendental idealism is about. These constraints can be claimed to yield a pre-cognitive form of intentionality, which is also nonconceptual.330 The claim about the embodied nature of the subject and the claim about the pre-cognitive relatedness to objects can be jointly held as definitive of Kant’s nonconceptualism, although they seem to be logically independent. But the claim about the involvement of body in cognition seems, prima facie, to motivate naturalism about cognition and the mind. Naturalism is a view which says that everything, including human beings and minds, belongs to one and the same order of nature and is subordinated to a unitary set of laws. In the early modern philosophy, naturalism had its representatives, as it has its representatives today. For example, in the third part of the Ethics, Spinoza said that such typically mental occurrences as passions or feelings should be explained in terms of the laws that are also appealed to in the explanation of physical phenomena.331 In particular, Spinoza contended that the mind is not ontologically different from the body; the mind is the idea or the representation of the body, which, in Spinoza’s terms, means that the mind is identical with all the knowledge that can

330 An account of the forms of intuition as determining basic pre-conceptual relatedness to objects one can find in: R. Aquila, Representational Mind: A Study of Kant’s Theory of Knowledge, Indiana University Press, Bloomington 1983. See also Benedykt Bornstein, Zasadniczy problemat teoryi poznania Kanta [The Main Problem of Kant’s Theory of Cognition], Warszawa 1910. 331 According to Spinoza, “nothing happens in Nature which can be attributed to any defect in it, for Nature is always the same, and its virtue and power of acting are everywhere one and the same, that is, the laws and rules of Nature, according to which all things happen, and change from one form to another, are always and everywhere the same. So the way of understanding the nature of anything, of whatever kind, must also be the same, namely, through the universal laws and rules of Nature. The affects, therefore, of hate, anger, envy, and the like, considered in themselves, follow with the same necessity and force of Nature as the other singular things. And therefore they acknowledge certain causes, through which they are understood, and have certain properties, as worthy of our knowledge as the properties of any other thing, by the mere contemplation of which we are pleased.” B. de Spinoza, The Ethics and Other Works, edited and translated by E. Curley, Princeton University Press, Princeton 1994, p. 153. © 2014 Anna Tomaszewska This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.



Spinoza and Hume 

 127

be had of the body.332 Spinoza’s theory of the mental has been read as relevant to contemporary philosophy, e.g. to Donald Davidson’s anomalous monism333 or the “dual aspect theory” advocated by Thomas Nagel.334 Hume also advocates naturalism. Thus, in chapter 9 of the Enquiry Concerning Human Understanding, he emphasizes similarities between the ways in which human and animal minds operate. Causal connections are learned habitually in the case of adult humans, human infants, and non-human animals alike.335 A belief that, for example, heat hurts one’s limbs if one keeps them too close to the fire, arises by way of habituation, rather than conscious reasoning, and is equally accessible to rational and non-rational animals. Accordingly, Hume shares the general idea of Spinoza that man is subject to the same laws as other entities belonging to the order of nature. Besides, on Hume’s approach, the mental and the physical can be studied, in principle, in the same way. Hume’s “metaphysics of the mind,” or “mental geography” is not methodologically divergent from physics, the science of bodies. Since all we can build our cognition on are constant conjunctions of perceptions, based on the latter’s succession in experience, we are as much entitled to claim that the mind is causally determined as we are entitled to claim that physical objects are. Does Kant subscribe to naturalism in any of the above outlined senses? Considering the doctrine of the spontaneity of the understanding, the constitutive role of the transcendental apperception, transcendental idealism and the necessary (for morality) requirement of freedom, one should answer in the negative. But, supposing that there is some plausibility in my reading, spatiotemporal representations, the Kantian forms of intuition, are rooted in the bodily nature of the subject. Since body belongs to the order of nature, so must the subject. Yet, even if this is granted, can mental, and in particular cognitive, processes be explained in the same way as physical processes are, on the Kantian doctrine? Can Kant’s transcendental theory of cognition be read along the lines of Hume’s “mental

332 These claims are made in part two of The Ethics: “P11: The first thing which constitutes the actual being of a human Mind is nothing but the idea of a singular thing which actually exists. […] P12: Whatever happens in the object of the idea constituting the human mind must be perceived by the human mind, or there will necessarily be an idea of that thing in the mind; that is, if the object of the idea constituting a human mind is a body, nothing can happen in that body which is not perceived by the mind. […] P13: The object of the idea constituting the human mind is the body, or a certain mode of extension which actually exists, and nothing else.” Ibidem, pp. 122-3. 333 Cf. M. Della Rocca, Spinoza, Routledge, London – New York 2008, pp. 89-136. 334 Cf. The View from Nowhere, Oxford University Press, Oxford – New York 1986. 335 According to Hume: “Animals, therefore, are not guided in these inferences by reasoning: Neither are children: Neither are the generality of mankind, in their ordinary actions and conclusions: Neither are philosophers themselves, who, in all the active parts of life, are, in the main, the same with the vulgar, and are governed by the same maxims. Nature must have provided some other principle, of more ready, and more general use and application.” An Inquiry concerning Human Understanding, edited by T. L. Beauchamp, Oxford University Press, New York 1999, p. 166.

128 

 Kant and Naturalism about the Mind

geography?” For the reasons valid for Kant himself, the answer is: no. For, first of all, Kant’s transcendental considerations do not constitute an instance of empirical, descriptive psychology.336 Second, for the reason that motivates Kant’s solution to the Third Antinomy in the Critique of Pure Reason,337 the subject cannot be considered as causally determined through and through. Rather, as Beth Lord has argued, the Kantian subject transcends nature – as an intermediary between the realm of nature and the intelligible or spiritual realm of freedom.338

6.2 Nonconceptualism and Naturalism According to Spinoza, mind is a complex idea of a body. Since to all bodies, or modifications of the substance considered from the point of view of the attribute of extension,339 correspond their ideas,340 it follows, somewhat oddly, that all extended entities, including cats, stones, and trees, have a mind. Conversely, since for all ideas, or modifications of the substance considered from the point of view of the attribute of thought,341 there are objects they correspond to, there can be no mind without an extended object which it is a mind of. For Spinoza, mentality is not a distinctive feature of a particular kind of entities; nor does consciousness constitute the essence of the mental. If we take Spinoza’s ideas as representations, in the more contemporary sense of the term, that is, as basic information-carrying units,342 we arrive at the idea that,

336 Kant held empirical psychology in quite low esteem and considered it part of anthropology, an empirical doctrine of nature, rather than a proper part of metaphysics, already extant or to be established in the future. In the chapter dedicated to the “architectonic of pure reason,” one can read: “Empirical psychology must thus be entirely banned from metaphysics, and is already excluded by the idea of it. Nevertheless, in accord with the customary scholastic usage one must still concede it a little place (although only as an episode) in metaphysics, and indeed from economic motives, since it is not yet rich enough to comprise a subject on its own and yet it is too important for one to expel it entirely or attach it somewhere else where it may well have even less affinity than in metaphysics. It is thus merely a long-accepted foreigner, to whom one grants refuge for a while until it can establish its own domicile in a complete anthropology (the pendant to the empirical doctrine of nature).” I. Kant, Critique of Pure Reason, translated and edited by P. Guyer and A. W. Wood, Cambridge University Press, New York 1998, p. 700 (A 849/B 877). 337 The reason is, of course, that freedom, i.e., independence from the causal laws of nature, must be retained on pain of the idea of morality losing its significance. 338 See: B. Lord, Kant and Spinozism. Transcendental Idealism and Immanence from Jacobi to Deleuze, Palgrave Macmillan 2011, pp. 175-183. 339 Cf. B. de Spinoza, The Ethics and Other Works, op. cit., p. 115 (D1). 340 “P7: The order and connection of ideas is the same as the order and connection of things.” Ibidem, p. 119. 341 Ibidem, p. 116 (D 3). 342 Such a definition of representation can be found in: F. Dretske, Naturalizing the Mind, The MIT Press, Cambridge MA 1995.



Nonconceptualism and Naturalism

 129

in the Spinozean world, there cannot be objects that would embody no information at all. Indeed, on this view, objects may be regarded as “chunks” of information. Since for Spinoza consciousness plays no role in the account of the mental, there is also no need for concepts in the account of representation based on Spinoza’s theory. Similarly, there is no need for concepts in Dretske’s theory of content, although Dretske is not as radical as Spinoza in granting representational properties to all objects. A mouse can have an olfactory representation of a burning toast,343 as much as a speedometer can represent the velocity of a vehicle running on the road.344 Also, “simple seeing” provides an example of perceptual representing without the employment of concepts.345 There is nothing distinctive about representation, on Thus, Dretske’s account, nothing that makes representing a distinctively mental activity; rather, representation is a function of certain systems, a function that can be either intrinsic (natural), or acquired. Dretske’s account of representation promotes both naturalism and nonconceptualism. But it is far from obvious whether representation does not require consciousness, as the accounts of Spinoza and Dretske imply. And even if consciousness does not constitute the essence of the mental, and is no prerequisite of representation, as Spinoza and also Leibniz contend,346 it does not necessarily entail that representational states are causally determined. One can believe that objects are “chunks” of information and, at the same time, think that Spinoza was mistaken in regarding informational, or rational, relations as (ontologically) identical with the causal ones. Certainly, we should not take at face value a doctrine that encourages conflating reasons with causes, and the more so if the causes are understood in the early modern sense, that is, only as efficient causes in the sense of Aristotle. There is, of course, some appeal in what may seem to be the central tenet of Spinoza’s system, i.e. in the endeavour to construe the intelligibility of the world in terms of the causal mechanistic explanation alone, but we do not have to endorse the idea without qualifications. As the history

343 The example can be found in: F. Dretske, “Conscious Experience,” Mind 102 (406), 1993. 344 The speedometer example has been borrowed from: F. Dretske, Naturalizing the Mind, op. cit. 345 See: F. Dretske, “Meaningful Perception” in: D. N. Osherson, S. M. Kosslyn et al. (eds.), Visual Cognition. An Invitation to Cognitive Science, vol. 2, The MIT Press, Cambridge MA 1995; F. Dretske, “What We See. The Texture of Conscious Experience” in: B. Nanay (ed.), Perceiving the World, Oxford University Press, Oxford – New York 2010. 346 In the Monadology, Leibniz attributes perceptions to all substances, which he calls monads. Perception is to be distinguished from both apperception and consciousness. Whereas perception is a representational state of the monad, apperception or consciousness indicate the accompanying awareness of such a state, which is not essential to its occurrence and does not pertain to all entities (those lower in the hierarchy of monads may not have it). Leibniz’s theory of “petites perceptiones” has given rise to later theories of non-conscious cognition and, according to some commentators, influenced Kant’s theory of cognition, too. See: P. Kitcher, “Kant’s Unconscious ‘Given’” in: P. Giordanetti, R. Pozzo, M. Sgarbi (eds.), Kant’s Philosophy of the Unconscious, Walter de Gruyter, Berlin 2012.

130 

 Kant and Naturalism about the Mind

of philosophy shows, such ideas have proved insufficient – for example, to explain the nature of living things. Kant, for instance, in the Critique of Judgment, in the second part entitled “Critique of Teleological Judgment,” postulated that organisms be explained in terms of final-causal principles, apart from the mechanical ones.347 In chapter 1 of this book two kinds of conceptualism and nonconceptualism were distinguished. Whereas state conceptualism requires that subjects possess concepts employed in the ascriptions of representational states to them, state nonconceptualism rejects this requirement. Content conceptualism makes a substantial claim about perceptual content in that it stipulates that it be conceptually structured. In McDowell’s version, content conceptualism is a view that representational states consist in the endorsement of facts, and there is a structural similarity between facts and representational states that can be articulated by propositions or judgments. Content nonconceptualism, by contrast, claims that some representational states defy conceptual structuring in that they have a sui generis kind of content. An illustration of state nonconceptualism can be provided by the speedometer example. Although a speedometer represents a vehicle’s velocity, it would be absurd to attribute the concept of velocity to the device. Likewise, the Kantian “savage,” to take a different example, can be said to represent a house despite his lacking the proper concept. The “savage” can see a house without, in a way, being aware that he is seeing one. But one could point out that, in fact, the analogy with the technical device misses the point. First, it is due to a particular designer’s craftsmanship that an electronic device represents a certain property of the environment. The states of the device and the properties of the environment do not merely correlate with one another; for the device is also equipped with a kind of sensitivity to selected features of the environment, according to its original setup. Thus, there is consciousness, or intentionality, involved in assigning the function of representation to a physical system. In the same way, the “savage” is attributed the representation of a house from the point of view of an external observer, who has the concept of house. Second, both for the speedometer and for the uneducated man in front of a building there is something like “getting things right” in the course of performing their representational functions. If the pointer of the speedometer indicates the same velocity regardless of the vehicle changing its speed, or if the “savage” is under the influence of a hallucinogenic substance or blind, they cannot be said to represent what they are supposed to. Let us now turn to content nonconceptualism. It has been stated that perceptual content is richer and more fine-grained than the content of beliefs or judgments. We can also see Kant endorsing what might seem to be a version of the richness

347 Cf. B. Lord, Kant and Spinozism, op. cit., ch. 4: “Critiques of Teleological Judgement,” pp. 80104.



Nonconceptualism and Naturalism

 131

argument.348 Perhaps the richness of perceptual content explains the fact that it is not entirely suitable for conceptualization. Richness would make it impossible to articulate the content of perception by means of concepts. But from this it does not follow that the experience of those perceptual properties of objects that cannot be articulated by concepts is not conscious. Again, it is far from obvious that consciousness involves the ability to ascribe the content of one’s experiences (and beliefs) to oneself. Rather, this ability should be associated with self-consciousness, a capacity that small children and non-human animals may lack, despite their ability to undergo perceptual experiences. Besides, as it was pointed out in the previous chapters, although perceptual experience may not require the conscious command of concepts by the subject, there is a sense in which concepts can be involved in experience. Kant’s doctrine of syntheses performed by the transcendental imagination illustrates the point. Accordingly, what we perceptually experience turns out to be rule-governed and rules, another name Kant uses for concepts, are imposed on the content of experience prior to the subject’s becoming aware that experience is, so to say, imbued with them. This non-conscious spontaneity cannot be further explained because it is to explain the facts (i) that perception is rule-governed and (ii) that some subjects, like small children, may not possess concepts used in the ascriptions of perceptual experiences to them. Since this interpretation defends state rather than content nonconceptualism, it can be shown that state nonconceptualism does not invite the naturalism of the mental. State nonconceptualism makes room for the spontaneity of the understanding as the ultimate source of representational content but it does not necessitate the conscious command of concepts as constitutive of this spontaneity. To ask about the source of spontaneity would be somewhat preposterous. The spontaneity of the understanding has to be, in a way, taken for granted, as the starting point of Kant’s theory of cognition. Clearly, the Kantian spontaneity parallels McDowell’s “conceptual capacities,” operating already at the level of the intake of empirical content, characterized in the following way: “[T]he original Kantian thought was that empirical knowledge results from a co-operation between receptivity and spontaneity. (Here, ‘spontaneity’ can be simply a label for the involvement of conceptual capacities.) […] receptivity does not make an even notionally separable contribution to the co-operation. The relevant conceptual capacities are drawn on in receptivity […]. It is not that they are exercised on an extra-conceptual deliverance of receptivity. We should understand what Kant calls ‘intuition’ – experiential intake – not as a bare getting of an extraconceptual Given, but as a kind of occurrence or state that already has conceptual content.”349

348 Cf. I. Kant, Lectures on Logic, translated and edited by J. M. Young, Cambridge University Press, New York 1992, p. 597 (9:99). 349 J. McDowell, Mind and World, Harvard University Press, Cambridge MA 1996, p. 9.

132 

 Kant and Naturalism about the Mind

But an account inviting non-conscious spontaneity may be flawed, both with regard to “spontaneity” and with regard to the “non-conscious.” First, a rationalist may object that to introduce a faculty, whose operations defy any explanation, thwarts the principle of sufficient reason. Second, the acknowledgment of the Kantian spontaneity and McDowell’s conceptual capacities does not stop short of the notorious “Myth of the Given.” The conceptual “Given” becomes substituted for the sensory “Given.” The “brute facts” of perceptual experience are exchanged for the “brute facts” of thought. Finally, it may be difficult to make sense of the claim that conceptual capacities can operate non-consciously. It is as if, in the case of the spontaneous operations of the understanding, at issue would be a kind of “blind” rule-following. The understanding would operate analogously to muscles and tendons in the legs of a bicycle rider, taking curves and adjusting his velocity to road conditions without deliberation or inference from the facts about road conditions to determinate actions. But while the operations of the cyclist’s body can be accounted for in terms of the physiology of the motor system, the operations of the understanding resist a satisfactory explanation.

6.3 McDowell’s Naturalism of the “Second Nature” I have talked about naturalism against the background of the philosophy of Spinoza and Hume, and observed that it encourages the following claims: (1) human beings, and in particular human minds, belong to the same domain as other entities, such as tables, chairs, trees and cats, and hence they should be explained or understood in terms of the same laws in which those other entities are explained or understood; (2) there is no distinctive feature, such as consciousness, which would characterize a distinguished set of entities; (3) human beings, and in particular human minds, are determined by the same kind of laws as other entities, such as tables etc., that is, they are determined by causal laws. I think we may call these claims: (1) a unitary explanatory framework thesis, (2) a non-essentialism about consciousness thesis and (3) a causal determinism thesis, respectively. It seems that claims (2) and (3) are definitive of naturalism due to a determinate conception of nature, inherited from the early modern philosophy. According to McDowell, this conception of nature as equivalent to the realm of law, contrasted with the space of reasons, is too narrow and therefore distorted.350 It dismisses the theoretical possibility of a conception of “second nature” in which nature is no longer alienated from the “space of reasons.” It is in terms of the second nature that McDowell

350 These terms are used throughout Mind and World and many other writings of McDowell. See also, e.g.: J. McDowell, “Précis of ‘Mind and World,’” Philosophical Issues 7, 1996; R. F. Gibson, “McDowell’s Direct Realism and Platonic Naturalism,” Philosophical Issues 7, 1996.



McDowell’s Naturalism of the “Second Nature” 

 133

purports to explain human nature, and in particular the nature of the mind. Thus, in Mind and World he says: “Our nature is largely second nature, and our second nature is the way it is not just because of the potentialities we were born with, but also because of our upbringing, our Bildung. Given the notion of second nature, we can say that the way our lives are shaped by reason is natural, even while we deny that the structure of the space of reasons can be integrated into the layout of the realm of law. This is the partial re-enchantment of nature that I spoke of.”351

McDowell’s account of nature is situated against the background of “bald naturalism” and “rampant platonism.” Whereas the former attempts to reduce the space of reasons, structured by logical relations, to the realm of law, structured by the causal ones, or “reconstruct [the space of reasons] out of conceptual materials whose primary home is the logical space in which empirical description […] functions,”352 the latter construes the realm of law and the space of reasons as entirely heterogeneous and, consequently, severs all logical connections between them. As R. F. Gibson aptly remarks, “[o]n this view, humans are partly in nature and partly outside nature,” which “makes the human capacity to respond to reasons look like an occult power.”353 Thus, while bald naturalism fails to recognize that there is something special about human rationality, which makes minds different from planks of wood or mere clouds of molecules, rampant platonism cuts off the links between human rationality and the empirical world, regarding rationality as something supernatural. McDowell considers both accounts unsatisfactory and offers a remedy that consists in reconceiving the natural, to the following effect: “The natural, in the sense of second nature, embraces concepts that function in the logical space of reasons, sui generis though that logical space is.”354

The claim that human nature is “second nature” can be understood in the light of both the Aristotelian and, I think, the Wittgensteinian tradition. For Aristotle, the second nature is rationality and its upshot: language and social relations that it enables to cultivate;355 for (the late) Wittgenstein, humans have a “natural history”

351 J. McDowell, Mind and World, op. cit., pp. 87-88. 352 J. McDowell, “Précis of ‘Mind and World,’” op. cit., p. 235. 353 R. F. Gibson, “McDowell’s Direct Realism and Platonic Naturalism,” op. cit., p. 279-280. 354 J. McDowell, “Précis of ‘Mind and World,’” op. cit., p. 236. 355 “In Aristotle’s conception of human beings, rationality is integrally part of their animal nature, and the conception is neither naturalistic in the modern sense (there is no hint of reductiveness or foundationalism) nor fraught with philosophical anxiety. What makes this possible is that Aristotle is innocent of the very idea that nature is the realm of law and therefore not the home of meaning. That conception of nature was laboriously brought into being at the time of the modern scientific revolution. I am not urging that we should try to regain Aristotle’s innocence. […] But instead of trying to integrate the intelligibility of meaning into the realm of law, we can aim at a postlapsarian or knowing

134 

 Kant and Naturalism about the Mind

which involves typically human activities related to the use of language.356 Nature, on McDowell’s account, encompasses both the realm of law and the space of reasons. Therefore, rational relations, such as justification, rather than merely causal relations, must be natural, too. In this way, one can think of experience not as a product of causal impacts of physical objects on sense organs but as a result of the operations of conceptual capacities, natural to human beings. To remind one of the main arguments for conceptualism, discussed in chapter 2: since experience provides reasons for empirical beliefs and relations between reasons and beliefs are rational, rather than causal, hence conceptually structured, the content of experience must be conceptual. Conceptual capacities, the analogue of the Kantian spontaneity, turn out to be natural. Thus, in McDowell’s view, it is naturalist – in the broader sense of “naturalism” – to describe perceptual experience as providing reasons for beliefs and so implicating concepts. Why is conceptualism supposed to be an antidote to the narrowly construed (“bald”) naturalism? I would venture a claim that this theoretical move follows from McDowell’s recognition of the dualism of reasons and causes. McDowell seems to acknowledge that there can be two kinds of relations between objects in the world: causal relations and rational relations. Conceptualism secures rationality in the relations between experiences and empirical beliefs because it establishes rational relations between them. Nonconceptualism, in turn, thwarts the possibility of rational relations between experiences and beliefs because, on McDowell’s tenets, it can only offer causal relations between the two kinds of occurrences. Since rationality is constitutive of the second nature and thus belongs to nature in the extended sense, and conceptualism secures the empirical rationality, one of the possible types of rationality, conceptualism should be endorsed to avoid the consequences of “bald” naturalism. But why should the dualism of reasons and causes represent a plausible view? Why to take it at face value? Why, when providing an account of the relations between the content of representational states and the objects of these states, should one be constrained to two options only? It may as well be the case that objects in the world in general, and representational states in particular, can be related not only either causally or rationally, but also in a third different way. Indeed, Husserl introduced such a third class of the relations between mental states and their objects, which he

counterpart of Aristotle’s innocence. We can acknowledge the great step forward that human understanding took when our ancestors formed the idea of a domain of intelligibility, the realm of natural law, that is empty of meaning, but we can refuse to equate that domain of intelligibility with nature, let alone with what is real.” J. McDowell, Mind and World, op. cit., p. 109. 356 “Giving orders, asking questions, telling stories, having a chat, are as much a part of our natural history as walking, eating, drinking, playing.” L. Wittgenstein, Philosophical Investigations, translated by G. E. M. Anscombe, P. M. S. Hacker and J. Schulte. Revised fourth edition by P. M. S. Hacker and J. Schulte, Basil Blackwell, Oxford 2009, p. 16e (italics A.T.).



McDowell’s Naturalism of the “Second Nature” 

 135

called relations of motivation.357 Mental states, on this account, do not cause one another, nor are they merely caused by things in the world, otherwise psychology could formulate laws as exact as the laws of physics; nor are they logically linked like premises and conclusions in syllogistic inferences. McDowell seems to have a narrow conception of rational relations; he would reduce them to logical relations, in which case of course conceptualism would be the only option to secure the rationality of second nature. But, by expanding the conception of rational relations, room for nonconceptual content in an account of human rationality can be found, an account that would accommodate the idea of second nature, too. Let us now go back to Kant. Considered against the background of Kant’s theory of experience, McDowell’s proposal seems to conflate the sensory and the intuitional content of experience. Sensory content is nonrepresentational and results from causal interactions between objects and the faculty of cognition. Intuitions, however, are intentional and nonconceptual, especially the pure intuitions that, according to some authors, like Lorne Falkenstein, Richard Aquila, or Robert Hanna, do not result from the synthesizing activity of the mind. As ways of informing sensory content, spatiality and temporality are “given” in experience as irreducible aspects of its objects, intrinsic to its representational content, rather than caused in the course of an interaction between objects and the faculty of cognition. For McDowell, conceptualism and second nature naturalism seem to be related in such a way that the former is explained by the latter. It is because human nature is second nature that our experience stands in a rational relation to the world. For McDowell, the rational relation between experience and the world consists in the fact that experience provides reasons for beliefs about the world. Reasons are conceptually structured; therefore, also experience is conceptually structured. But McDowell’s conception of reasons delivered by experience is quite unclear. One may argue that it takes him close to the “oscillation” which he attempts to find a remedy for, and, in the end, makes his proposal – i.e. that conceptualism about the contents of perceptual experience helps secure the naturalism of second nature – inconsistent. For what are McDowell’s perceptual reasons?358 If we consider them as endorsed facts about the world, then in what sense do they differ from judgments? If they are judgments, it is legitimate to ask about their justification. No one can be said to have good reasons to endorse a belief if one’s belief lacks proper justification. But

357 E. Husserl, Ideas Pertaining to a Pure Phenomenology and to a Phenomenological Philosophy. Second Book: Studies in the Phenomenology of Constitution, translated by R. Rojcewicz and A. Schuwer, Kluwer Academic Publishers, Dordrecht/Boston/London 1989, § 56. 358 Considerations on this issue can be found, e.g., in: Ch. Larmore, “Attending to Reasons” in: N. H. Smith (ed.), Reading McDowell: On Mind and World, Routledge, London and New York 2002.

136 

 Kant and Naturalism about the Mind

providing a justification and providing a reason is one and the same thing – and in the end, we arrive at an account on which the series of reasons justifying empirical beliefs would have to proceed in infinity. In other words, we fall back onto one side of the “seesaw:” the coherentist side. However, we may also come to the point at which we recognize that justification, the series of reasons must terminate at some point. That is, we may recognize that experience delivers reasons that cannot be further justified, but such reasons should not adopt the form of beliefs or judgments – in which case we land back in the Myth of the Given. Alternatively, perceptual reasons may be both incapable of further justification and conceptually structured, pieces of information, fact-based propositions, “messages” that the world conveys to us – an alternative which sidetracks the skeptical challenge by pronouncing what looks like a kind of rationalist faith in the intelligibility of the world.359

6.4 Kant’s “Transcendental Psychology” and Naturalism360 Considered from the empirical, rather than the transcendental, perspective, the Kantian subject belongs to the order of nature and is subordinated to the same kind of laws as other entities that constitute it. Many authors object, though, to attempts at naturalizing Kant’s transcendental philosophy. Allison emphasizes the spontaneity of the Kantian mind, which does not depend on empirical determinations.361 McDowell connects the spontaneity of the higher cognitive faculties with the subject’s freedom.362 Allison remarks that ultimately it is in the

359 This rationalist “faith” can be mitigated if we turn it into a hypothesis which best explains the possibility of empirical knowledge, as McDowell indeed does in: “The Disjunctive Conception of Experience as Material for a Transcendental Argument” in: F. Macpherson and A. Haddock (eds.), Disjunctivism: Perception, Action, Knowledge, Oxford University Press, Oxford 2008. But then the argument for conceptualism becomes considerably weakened. 360 This section has been written on the basis of parts of my article: “Spinoza i Kant o naturze ludzkiego umysłu” [“Spinoza and Kant on the Nature of Human Mind”], Filo-Sofija 17 (2), 2012. 361 See: H. Allison, “On Naturalizing Kant’s Transcendental Psychology” and “Kant’s Refutation of Materialism” in: Idealism and Freedom: Essays on Kant’s Theoretical and Practical Philosophy, Cambridge University Press, Cambridge 1996. An argument against naturalism in the philosophy of mind, based on Kant’s theory of cognition, in which the spontaneity of the mind synthesizes the manifold “given” in accordance with the rules originating from the understanding, can be found in: S. Judycki, Umysł i synteza. Argument przeciwko naturalistycznym teoriom umysłu [Mind and Synthesis. The Case against Naturalistic Theories of Mind], RW KUL, Lublin 1995. 362 “When Kant describes the understanding as a faculty of spontaneity, that reflects his view of the relation between reason and freedom: rational necessitation is not just compatible with freedom but constitutive of it. In a slogan, the space of reasons is the realm of freedom.” J. McDowell, Mind and World, op. cit., p. 5.



Kant’s “Transcendental Psychology” and Naturalism 

 137

understanding that the origin of the basic categories, the so-called pure concepts, which enable syntheses of the manifolds of intuition, should be sought. Since the act of spontaneity – the “I think” from paragraph B 132 of the Critique of Pure Reason – must be able to accompany all representations entertained by the mind, there cannot be mental processes that the subject would not be aware of and that would be causally related to the representational states or contents of the mind. For Allison, the Kantian “I” of the transcendental apperception is “systematically elusive” and ineliminable from experience.363 Since the “I” cannot be represented, it does not meet the conditions of representability, such as being spatiotemporal. Besides, from the Transcendental Deduction of the Categories it follows that each particular experience implies the possibility of forming a judgment on the basis of it; and forming a judgment presupposes the ability to recognize an object as falling under a determinate concept.364 Such an act cannot be explained solely in terms of objects causally affecting the mind. However, there is an argument that can challenge the above views, defended by Allison and McDowell. This argument, if true, would make it possible for us to attribute to Kant the three theses, introduced in the previous section, which define the naturalist position: the unitary explanatory framework thesis, the causal determinism thesis, and the thesis expressing non-essentialism about consciousness. The argument could be formulated with the textual support of the introductory passages from the A- and B-edition of the Critique of Pure Reason, which read: 1. “Experience is without doubt the first product that our understanding brings forth as it works on the raw material of sensible sensations.”365 2. “In whatever way and through whatever means a cognition may relate to objects, that through which it relates immediately to them, and at which all thought is directed as an end, is intuition. This, however, takes place only insofar as the object is given to us; but this in turn is possible only if it affects the mind in a certain way. […] Objects are therefore given to us by means of sensibility, and it alone affords us intuitions […]. The effect of an object on the capacity for representation, insofar as we are affected by it, is sensation.”366

363 “The moral of the story is that I is not only ineliminable, but also, in the language of Gilbert Ryle, ‘systematically elusive.’ […] the attempt to grasp the I in thought can be compared to the attempt to catch one’s own shadow. […] this elusiveness can be construed as a consequence of the ‘logic’ of its use. The use is transcendental, not empirical.” H. Allison, “Kant’s Refutation of Materialism,” op. cit., p. 96. 364 I. Kant, Critique of Pure Reason, op. cit., pp. 230-234 (A 103-110). 365 Ibidem, p. 127 (A 1, italics A.T.). 366 Ibidem, p. 155 (A 19-20/B 33-4, italics A.T.).

138 

 Kant and Naturalism about the Mind

These passages seem to suggest that, for Kant, cognition starts with objects affecting the mind in such a way that they produce sensations. Since by means of sensations the mind does not represent anything, and since representational states involve consciousness and the “I think,” sensations cannot be the content of conscious mental states. Therefore, since they cannot be conscious, they must result from a causal process, like a dent in a wall that results from a metal ball hitting the wall with considerable force. Because underlying conscious representational states would be non-conscious causal processes, Kant’s theory of cognition seems to have naturalistic underpinnings. The unitary explanatory framework can be provided by cognitive science, which, as Patricia Kitcher has argued, does share a number of common assumptions with what she calls Kant’s transcendental psychology.367 Does this argument suffice to promote a naturalistic interpretation of Kant’s theory of cognition? I think that it would – if only Kant was clear about how to understand the “affection” relation. But since his doctrine is notoriously ambiguous at this point, we would rather suspend definitive conclusions about Kant’s alleged commitment to the naturalistic account of cognition. For, first of all, the affection of the mind by objects does not have to be taken as a causal relation; nor does it have to be construed as analogous to it. Certainly, if Kant had wanted to characterize the precognitive relation between mind and objectivity as causal, he would have applied the concept of cause. But the affection relation can be understood in terms of the mind’s receptivity, or sensitivity, to the presence of objects; in other words, it is from the relation between mind and object(s) that cognition starts – and no more than that. This assumption, indeed quite minimalist, amounts to the claim that underlying all representational states is an intentional relation, the object-directedness of the mind.368

367 According to Kitcher, both Kant and cognitive scientists provide models of the mind and its cognitive functions, and, in that they do so, they endorse similar methodology. They also take interest in explaining certain issues, like the role of concepts in cognition, or spatial perception. See: P. Kitcher, Kant’s Transcendental Psychology, Oxford University Press, Oxford – New York 1990. 368 Some authors argue that Kant’s doctrine of affection does not lead to naturalism. See: H. Allison, Kant’s Transcendental Idealism: An Interpretation and Defense, Yale University Press, New Haven and London 2004; N. Rescher, “Noumenal Causality” in: L. W. Beck, Proceedings of the Third International Kant Congress, D. Reidel Publishing Co., Dordrecht 1972. Rescher distinguishes between an “authentic causality” and “a not properly causal generic grounding which is merely intelligible.” Whereas the principle of causality helps establish a definite type of relations within the realm of experienced objects, it is, according to Rescher, the epistemologically reinterpreted principle of sufficient reason that establishes relations between experience and objectivity (construed as the unconditioned). According to Rescher, reason postulates the existence of mind-independent reality to endow cognitions with objective validity and to block the infinite regress in the series of conditions necessary for a natural event to occur. For more information on this issue, see my paper: “Transcendental Object and the ‘Problem of Affection.’ Remarks on Some Difficulties of Kant’s Theory of Empirical Cognition,” Diametros 21, 2007, p. 75.



Kant’s “Transcendental Psychology” and Naturalism 

 139

Second, Kant does not specify what kind of object(s) is involved in the affection relation with the mind; if the affection is not causal, hence not empirical, then also the object should be considered from a different perspective than the empirical one. Indeed, such an interpretation seems to accommodate several passages in the first Critique: both the introductory ones and, importantly, the passages referring to the transcendental object (an unknown “X”), which Kant discusses in the chapter following the A-edition account of the threefold syntheses. This unknown object is featured in the following way: “And here then it is necessary to make understood what is meant by the expression ‘an object of representations.’ […] It is easy to see that this object must be thought of only as something in general = X, since outside of our cognition we have nothing that we could set over against this cognition as corresponding to it. […] All representations, as representations, have their object, and can themselves be objects of other representations in turn. Appearances are the only objects that can be given to us immediately, and that in them which is immediately related to the object is called intuition. However, these appearances are not things in themselves, but themselves only representations, which in turn have their object, which therefore cannot be further intuited by us, and that may therefore be called the non-empirical, i.e., transcendental object = X. The pure concept of this transcendental object (which in all of our cognitions is really always one and the same = X) is that which in all of our empirical concepts in general can provide relation to an object, i.e., objective reality.”369

But it seems that objections can also be made against identifying the object Kant mentions in the introductory passages of the Critique of Pure Reason with the “transcendental object = X.” The latter corresponds to the activity of the mind in synthesizing the manifold of representations; it is, as it were, the object-side of the unity that the mind itself endows upon the representations it binds together. As such, the transcendental object seems to be somehow present, even if in an inarticulate manner, to consciousness.370 But the object that constitutes one side of the affection relation falls entirely outside the scope of consciousness. Furthermore, it is the nonrepresentational sensory content – the matter of cognition – that results from the affection of the mind by object(s), which makes affection similar to a physiological process that some philosophers, such as Locke, considered as giving rise to perceptual cognition. Again, if we think of the affection relation in terms of a physiological process, we end up with a naturalistic claim that underlying representational states is a causal relation. According to Kitcher, this is indeed the case:

369 I. Kant, Critique of Pure Reason, op. cit., pp. 231-3 (A 104-9). 370 What this kind of consciousness would consist in is a topic for a further debate; it may be argued that this consciousness would be pre-cognitive and therefore nonconceptual, as much as the consciousness of one’s existence that the “I think” from paragraph B 132 gives expression to. Cf. K. Crone, “Pre-conceptual Aspects of Self-consciousness in Kant’s Critique of Pure Reason” in: P. Giordanetti, R. Pozzo, M. Sgarbi (eds.), Kant’s Philosophy of the Unconscious, op. cit.

140 

 Kant and Naturalism about the Mind

“Still, despite the human incapacity to know how cognition works in any detail, some general features are known (and admitted by all in debates on cognition): knowledge of particular states of affairs and of particular concepts can be acquired only by causal interaction with objects. Hence the confident assertion at B 1 that there can be no doubt that cognition begins with objects stirring the senses. […] the production of the intuition is not a rational but a causal process, since both the materials and the processing of them are unconscious.”371

We have come to the point at which the interpretation may veer in one of the two directions: in the naturalistic direction, advocated by P. Kitcher and A. Brook,372 or in (what I would call) the intentionalist direction, advanced by H. Allison and N. Rescher. Indeed, already the first commentators of Kant’s critical philosophy noticed that the doctrine of affection marks an inconsistency in the Kantian programme. Kant himself struggled with the problem of the origin of what he called the matter of cognition long after the publication of the three Critiques and well into the period in which he sketched an outline of a theory that aimed to explain – more comprehensively than the Transcendental Deduction of the Categories was able to do – how the pure concepts of the understanding relate to the physical reality, an outline known as Kant’s Opus postumum.373 Perhaps it would not be too far-fetched to draw a comparison between Kant and Spinoza in this respect: whereas interpretations of the former oscillate between naturalism and anti-naturalism, interpretations of the latter support either materialism, or pantheism (also panpsychism), its opposite. But as much as understanding the Spinozean substance merely in terms of nature, i.e., a system of causally conditioned physical entities, betrays one-sidedness, so does reading the Kantian object(s) affecting the mind merely in terms of a physical object interacting with our sense organs as part of a physiological process. To the effect that Spinoza’s substance grounds all its infinitely numerous attributes, or “essences,” among which two – extension and thought – are known to us, it is neither exclusively mental, nor exclusively physical but bears both kinds of properties which in the Cartesian tradition would be taken to exclude one another. Likewise, the idea of the transcendental object and the idea of the unknown “ground” of affection could be regarded as two different ways of accommodating two different requirements of reason (or as concepts denoting one and the same reality from two different perspectives). Accordingly, whereas the transcendental object, the correlative of representations, could accommodate the need for the intelligibility of the object of cognition, its accessibility to the mind,

371 P. Kitcher, “Kant’s Unconscious ‘Given,’” op. cit., pp. 30-2. 372 A. Brook, Kant and the Mind, Cambridge University Press, Cambridge 1994. 373 An instructive account of how Kant attempted to solve the problem of the origin of the matter of cognition, long after publishing the first Critique, can be found in: B. Lord, “Spinozism in the Ether: Kant’s Opus Postumum” in: Kant and Spinozism. Transcendental Idealism and Immanence from Jacobi to Deleuze, Palgrave Macmillan 2011, pp. 155-174.



Kant’s “Transcendental Psychology” and Naturalism 

 141

the object(s) affecting the senses could accommodate the idea of its (radical) mindindependence. But whether such a reading can be properly warranted is an open question – indeed, substantiating this interpretation would require a separate and detailed discussion, which, for the shortage of space, I cannot offer here. In the Conclusion below, I say a few words about where the considerations presented in this chapter may take us, and briefly recapitulate the argument presented in this book.

Conclusion In this book, I have argued in favour of a nonconceptualist interpretation of Kant’s theory of empirical cognition. Thus, I have tried to show that Kant can be attributed a view on which intentionality can also be understood as pre-conceptual relatedness to objects and rationality can be taken as emerging from, or constituted upon, more primitive, proto-rational structures. To that end, I discussed Kant’s account of intuition (chapter 3) and especially pure intuitions (chapter 2) as a kind of concept-independent representations, and I formed an hypothesis, based on relevant passages in Kant’s pre-critical writings (concerning, for example, the “differentiation” of directions in space, or “incongruent counterparts”) and the Critique of Pure Reason (such as the Refutation of Idealism), that the source of nonconceptual content could be traced back to the bodily constitution of the subject of empirical cognition (chapter 2 and 5). I have also pointed out that, pace Robert Hanna, Kant’s nonconceptualism in fact underlies the doctrine that our cognition is constrained to spatiotemporal appearances and does not reach into the realm of things in themselves – the doctrine Kant refers to as transcendental idealism (chapter 5). Throughout the book I have abstracted from the concern that most of the arguments used in the debate between conceptualism and nonconceptualism may be regarded as inconclusive since they build on an ambiguous and insufficiently fleshed out notion of concept (chapter 1). However, I did consider possible challenges to my nonconceptualist reading, based on certain doctrines of the First Critique, such as the B 105 claim about the structural identity between intuitions and judgments, the A-edition account of three syntheses in the Transcendental Deduction of the Categories, or the claim about the “I think,” expressed in the B 132 passage and relating to the transcendental apperception which must be able to accompany all representations entertained by the subject (chapter 4). As I argued, all these doctrines can be made compatible with reading Kant as an advocate of nonconceptualism. I ended this book raising a question whether the reading presented in it may support a naturalistic account of Kant’s theory of the mind (chapter 6). I did not provide a conclusive reply to the question but only suggested possible answers, which turned out to be mutually exclusive, even though apparently equally legitimate. But one objection to the interpretation suggested in this book seems to be still pending. If the Kantian finite subject of cognition is construed as embodied, then it not only belongs to the realm of objects determined by causal laws, but also, and quite obviously, to the domain of objects which belong to the spatiotemporal framework, that is, appearances. Accordingly, conditions of empirical cognition would be grounded in the existence of an object for which we need the very same conditions to cognize it. The procedure of deriving these conditions, apart from its apparent circularity, would thus have to be a posteriori, dependent on experience. But for Kant the subject, with the a priori structures determining the mode in which it cognizes, plays a constitutive role with regard to experience, hence it must transcend the realm conditioned by © 2014 Anna Tomaszewska This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.

Conclusion 

 143

space and time, causality, and other categories. This is why it remains “notoriously elusive,” as Henry Allison put it – one cannot get to know what ontological reality underlies the conditions at issue, since any knowledge presupposes these conditions. The problem, therefore, is not solely that by advancing the interpretation presented in this book one incurs ontological commitments but, perhaps even worse, that one thereby conflates the empirical and the transcendental level of considerations, which leads to undermining the core tenets of Kant’s idealism. However, it should be remembered that the distinction between the empirical and the transcendental, or the phenomenal and the noumenal, is not one between the physical (material) and the mental (spiritual). Rather, it is a distinction between two modes of existing or two ways of considering an object. Both chairs and minds, for example, can be considered as phenomena, that is, (existing) in relation to empirical cognition, and as noumena, that is, (existing) in abstraction from empirical cognition. If there is then any problem in attributing the spatiotemporal constraints on cognition to the bodily constitution of the subject, it lies not in the fact that body is a physical, hence causally determined object, but rather in the fact that we make claims about the inaccessible noumenal realm of objects. This expressly goes against the grain of the following formulation one can find in the A-edition paralogisms chapter in the Critique of Pure Reason: “The transcendental object that grounds both outer appearances and inner intuition is neither matter nor a thinking being in itself, but rather an unknown ground of those appearances that supply us with our empirical concepts of the former as well as the latter.”374

Kant’s reluctance to inquire into the ontological reality that underlies the acquisition and employment of different kinds of representational content seems, in a way, to directly correspond to our propensity to ask this ontological question. This propensity we may owe, at least partly, to Kant himself, who assigns different kinds of representational content to different faculties of the mind. Concepts originate from the understanding and intuitions – from sensibility; and it seems fairly natural to think that the sensible mode of cognition, with the affection of sense organs by objects, implicates a certain kind of physical relations between mind and world. Also, tracking objects in space (and time) seems to require a capacity of locating them in relation to a reference point the best candidate for which would be the subject’s body. But all these apparently natural ways of thinking are based on overlooking the fact that Kant considers cognitive faculties, as it were, in abstraction from the ontological reality underlying them, rather by focusing on their functions. (A robot equipped with sensibility and understanding would thus be as good as flesh and blood for that purpose.)

374 I. Kant, Critique of Pure Reason, translated and edited by P. Guyer and A. W. Wood, Cambridge University Press, Cambridge 1998, p. 431 (A 379-80).

144 

 Conclusion

Kant’s approach centered on faculties and their functions is perhaps reminiscent of the approach adopted by psychology of cognition and cognitive science which, without inquiring into matters of ontology, purport to explain what makes perceiving different from other cognitive acts, such as engaging in reasoning and drawing inferences, recalling past occurrences and anticipating future ones, or imagining; and what accounts for certain phenomena, such as depth-perception, hallucinations, iconic memory, inattentional blindness, illusions and other. As much as positing nonconceptual content may help explain these phenomena, so may positing the subject’s embodiment help explain the properties of nonconceptual content, such as fineness of grain or situation-dependence and particularity. The former would be derivative of the sensational and analogue (to employ Dretske’s term) character of perceptual experience; the latter would relate to the subject’s locatedness in space, in relation to other physical objects. Situating the reading presented in this book within a merely explanatory framework opens a possibility to postulate a research programme without promising to establish irrevocable claims about the ultimate ontological grounding of the investigated phenomena. Last but not least, as I tried to argue in chapter 6, naturalism does not have to be narrowly construed, that is, in terms of universal causal determinism and materialism. Apart from the naturalism of second nature, introduced by McDowell and building on the Aristotelian conception of human nature, there is also naturalism of what one may regard a Spinozistic sort. What distinguishes the Spinozistic kind of naturalism from the other ones are several factors, possibly quite unfamiliar to a contemporary naturalist. One of such factors is that it consistently applies the principle of sufficient reason according to which no occurrence in nature and no entity can lack a cause or reason, which is the same thing, determining its nature and the consequences following from it. Another factor is that it defies reductionism of any kind, which means that one sort of occurrences (e.g. mental ones) cannot be explained away in terms of a different sort of occurrences (e.g. physical ones). This would entail the infringement of an “explanatory barrier” which Spinoza adopts throughout his metaphysical system.375 Finally, one of the important traits of the Spinozistic naturalism is that it locates the ultimate source of all phenomena in the whole that transcends the sum of its parts.376 Spinoza’s substance (the “God or Nature”) is as much inscrutable as Kant’s “transcendental ground” of all appearances, inner and outer. Yet, for one who pursues knowledge, it may serve as the idea encapsulating the final goal of cognitive pursuits. In other words, Spinoza is a naturalist to the extent that he postulates a unitary framework within which to explain all phenomena, rather than in the

375 Cf. M. Della Rocca, Spinoza, Routledge, London and New York 2008. 376 Cf. B. Lord, Kant and Spinozism: transcendental idealism from Jacobi to Deleuze, Palgrave Macmillan 2011, pp. 1-19.

Conclusion 

 145

sense of reducing one sort of phenomena to another. If Kant were a naturalist, he would perhaps be more sympathetic to the Spinozistic kind of naturalism – but a unitary and comprehensive framework against the background of which to explain all phenomena must, for him, remain no more than a regulative idea. Constructing explanatory models of cognition one may perhaps approximate this regulative idea but no more than that. Thus, if naturalism were to imply a determinate ontological commitment, the very logic of Kant’s philosophical system would prevent him from taking it on.

Bibliography Adickes, E. (ed.), Kants Opus postumum, Topos Verlag, Vaduz 1978. Allais, L., “Kant, Non-Conceptual Content and the Representation of Space,” Journal of the History of Philosophy 47 (3), 2009, pp. 383-413. Allison, H. E., “The ‘Critique of Pure Reason’ as Transcendental Phenomenology” in: D. Ihde, R. N. Zaner (eds.), Dialogues in Phenomenology, Martinus Nijhoff, The Hague 1975, pp. 136-155. -----------------, Idealism and Freedom: Essays on Kant’s Theoretical and Practical Philosophy, Cambridge University Press, Cambridge 1996. -----------------, Kant’s Transcendental Idealism. An Interpretation and Defense, Yale University Press, New Haven and London 2004. Amaral, P. (ed.), Kant and Pre-Kantian Themes: Lectures by Wilfrid Sellars, Ridgeview, Atascadero 2002. Aquila, R., Representational Mind: A Study of Kant’s Theory of Knowledge, Indiana University Press, Bloomington 1983. Aristotle, De Anima, translated by Ch. Shields, Oxford University Press, Oxford 2008. Bennett, J., Kant’s Analytic, Cambridge University Press, Cambridge 1966. -------------, Locke, Berkeley, Hume: Central Themes, Oxford University Press, Oxford 1971. Bermúdez, J. L., “Nonconceptual Content: From Perceptual Experience to Subpersonal Computational States,” Mind and Language 10 (4), 1995, pp. 333-369. -------------------, “What is at Stake in the Debate on Nonconceptual Content?” Philosophical Perspectives 21 (1), 2007, pp. 55-72. BonJour, L., “Can Empirical Knowledge Have a Foundation?” American Philosophical Quarterly 15 (1), 1978, pp. 1-14. Bornstein, B., Zasadniczy problemat teoryi poznania Kanta, Księgarnia G. Centnerszwera i Ski w Warszawie oraz H. Altenberga we Lwowie, Warszawa 1910. Brandom, R., Making it Explicit. Reasoning, Representing, and Discursive Commitment, Harvard University Press, Cambridge MA 1998. Brewer, B., Perception and Reason, Oxford University Press, Oxford 1999. -------------, “Perception and Content,” European Journal of Philosophy 14 (2), 2006, pp. 165-181. -------------, Perception and its Objects, Oxford University Press, Oxford 2011. Brook, A., Kant and the Mind, Cambridge University Press, Cambridge 1994. Caimi, M., “»Gedanken ohne Inhalt sind leer«,” Kant-Studien 96 (2), 2005, pp. 135-146. Cassirer, E., Kant’s Life and Thought, translated by J. Haden, Yale University Press, New Haven and London 1981. Chuard, Ph., “The Riches of Experience,” Journal of Consciousness Studies 14 (9-10), 2007, pp. 20-42. Collins, A. W., Possible Experience. Understanding Kant’s ‘Critique of Pure Reason’, University of California Press, Berkeley and Los Angeles 1999. Crane, T., “The Waterfall Illusion,” Analysis 48 (3), 1988, pp. 142-147. Crone, K., “Pre-conceptual Aspects of Self-consciousness in Kant’s ‘Critique of Pure Reason’” in: P. Giordanetti, R. Pozzo, M. Sgarbi (eds.), Kant’s Philosophy of the Unconscious, Walter de Gruyter, Berlin 2012, pp. 131-146. Crowther, T. M., “Two Conceptions of Conceptualism and Nonconceptualism,” Erkenntnis 65 (2), 2006, pp. 245-276. Della Rocca, M., Spinoza, Routledge, London and New York 2008. Dennett, D. C., Content and Consciousness, Humanities Press, New York 1969. Descartes, R., Discourse on Method, Optics, Geometry, and Meteorology, translated by P. J. Olscamp, Bobbs-Merrill, Indianapolis 1965. © 2014 Anna Tomaszewska This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.

Bibliography 

 147

----------------, Meditations on First Philosophy. With Selections from the Objections and Replies, edited by J. Cottingham, Cambridge University Press, Cambridge 1996. Dretske, F., Knowledge and the Flow of Information, The MIT Press/Bradford Books, Cambridge MA 1981. --------------, “Conscious Experience,” Mind 102 (406), 1993, pp. 263-283. -------------, “Meaningful Perception” in: D. N. Osherson, S. M. Kosslyn et al. (eds.), Visual Cognition. An Invitation to Cognitive Science, vol. 2, The MIT Press, Cambridge MA 1995, pp. 331-352. -------------, Naturalizing the Mind, The MIT Press, Cambridge MA 1995. -------------, Perception, Knowledge and Belief. Selected Essays, Cambridge University Press, Cambridge 2000. -------------, “Sensation and Perception” in: Y. H. Gunther (ed.), Essays on Nonconceptual Content, The MIT Press, Cambridge MA 2003, pp. 25-41. -------------, “What We See. The Texture of Conscious Experience” in: B. Nanay (ed.), Perceiving the World, Oxford University Press, Oxford – New York 2010, pp. 54-67. Dzielski, M., “Newtona i Leibniza koncepcje czasu i przestrzeni” in: G. Łuczkiewicz (ed.), Odrodzenie ducha – budowa wolności, Krakowskie Towarzystwo Przemysłowe & Znak, Kraków 1995, pp. 619-697. Evans, G., The Varieties of Reference, edited by J. McDowell, Oxford University Press, Oxford 1982. ------------, “Molyneux’s Question” in: A. Phillips (ed.), Gareth Evans: Collected Papers, Clarendon Press, Oxford 1985, pp. 364-399. Falkenstein, L., Kant’s Intuitionism. A Commentary on the Transcendental Aesthetic, University of Toronto Press, Toronto 2004. Flew, A., “Was Berkeley a Precursor of Wittgenstein?” in: W. B. Todd (ed.), Hume and the Enlightenment: Essays Presented to Ernest Campbell Mossner, Edinburgh University Press, Edinburgh 1974, pp. 153-163. Fridland, E., “The Case for Proprioception,” Phenomenology and the Cognitive Sciences 10 (4), 2011, pp. 521-540. Friedman, M., “Kant’s Theory of Geometry,” The Philosophical Review 94 (4), 1985, pp. 455-506. Gardner, S., Kant and the ‘Critique of Pure Reason’, Routledge, London 1999. George, R., “Kant’s Sensationism,” Synthese 47 (2), 1981, pp. 229-255. Gibson, R. F., “McDowell’s Direct Realism and Platonic Naturalism,” Philosophical Issues 7, 1996, pp. 275-281. Ginsborg, H., “Was Kant a Nonconceptualist?” Philosophical Studies 137 (1), 2008, pp. 65-77. Giordanetti, P., R. Pozzo, M. Sgarbi (eds.), Kant’s Philosophy of the Unconscious, Walter de Gruyter, Berlin 2012. Greenberg, M., “A New Map of Theories of Mental Content: Constitutive Accounts and Normative Theories,” Philosophical Issues 15, 2005, pp. 299-320. Griffith, M., “Perception and the Categories: A Conceptualist Reading of Kant’s ‘Critique of Pure Reason,’” European Journal of Philosophy 20 (2), 2010, pp. 193-222. Gunther, Y. H., (ed.), Essays on Nonconceptual Content, The MIT Press, Cambridge MA 2003. --------------, “The Phenomenology and Intentionality of Emotion,” Philosophical Studies 117, 2004, pp. 43-55. Guyer, P., “Kant on Apperception and ‘A Priori’ Synthesis,” American Philosophical Quarterly 17 (3), 1980, pp. 205-212. -----------, Kant and the Claims of Knowledge, Cambridge University Press, Cambridge 1987. -----------, “Absolute Idealism and the rejection of Kantian dualism” in: K. Ameriks (ed.), The Cambridge Companion to German Idealism, Cambridge University Press, New York 2000, pp. 37-56. Hanna, R., “The Inner and the Outer: Kant’s ‘Refutation’ Reconstructed,” Ratio 13 (2), 2000, pp. 146-174.

148 

 Bibliography

------------, Kant, Science and Human Nature, Oxford University Press, Oxford 2003. ------------, “Kant and Nonconceptual Content,” European Journal of Philosophy 13 (2), 2005, pp. 247-290. ------------, “Kantian Non-Conceptualism,” Philosophical Studies 137 (1), 2008, pp. 41-64. ------------, “Beyond the Myth of the Myth: A Kantian Theory of Non-Conceptual Content,” International Journal of Philosophical Studies 19 (3), 2011, pp. 323-398. ------------, “Kant’s Non-Conceptualism, Rogue Objects, and the Gap in the B Deduction,” International Journal of Philosophical Studies 19 (3), 2011, pp. 399-415. ------------, “The Myth of the Given and the Grip of the Given,” Diametros 27, 2011, pp. 25-46. Heck, R. G., “Nonconceptual Content and the ‘Space of Reasons,’” The Philosophical Review 109 (4), 2000, pp. 483-523. Hegel, G. W. F., Faith and Knowledge, translated by W. Cerf and H. S. Harris, SUNY Press, New York 1977. -------------------, Phenomenology of Spirit, translated by A. V. Miller, Oxford University Press, Oxford 1977. Heidegger, M., Kant and the Problem of Metaphysics, translated by R. Taft, Indiana University Press, Bloomington 1997. Heidemann, D. (ed.), Kant and Non-Conceptual Content, Routledge, London and New York 2013. Helmholtz, H. von, “Facts in Perception,” translated by M. Lowe in: R. S. Cohen, Y. Elkana (eds.), Hermann von Helmholtz. Epistemological Writings, D. Reidel Publishing Co., DordrechtHolland, Boston-USA 1977, pp. 115-185. Hintikka, J., Knowledge and the Known. Historical Perspectives in Epistemology, D. Reidel Publishing Co., Dordrecht 1974. Hume, D., An Inquiry concerning Human Understanding, edited by T. L. Beauchamp, Oxford University Press, New York 1999. Husserl, E., Ideas Pertaining to a Pure Phenomenology and to a Phenomenological Philosophy. Second Book: Studies in the Phenomenology of Constitution, translated by R. Rojcewicz and A. Schuwer, Kluwer Academic Publishers, Dordrecht/Boston/London 1989. Jacobi, F. H., David Hume über den Glauben; oder Idealismus und Realismus: Ein Gespräch, Loewe, Breslau 1787. Johnson, M., The Meaning of the Body. Aesthetics of Human Understanding, The University of Chicago Press, Chicago 2007. Judycki, S., Umysł i synteza. Argument przeciwko naturalistycznym teoriom umysłu, RW KUL, Lublin 1995. Kain, P., “Kant’s Defense of Human Moral Status,” Journal of the History of Philosophy 47 (1), 2009, pp. 59-101. Kant, I., Gesammelte Schriften. Hrsg.: Bd. 1-22 Preussische Akademie der Wissenschaften, Bd. 23 Deutsche Akademie der Wissenschaften zu Berlin. Berlin 1900ff. ---------, Kant’s Introduction to Logic and his Essay on the Mistaken Subtlety of the Four Figures, translated by T. K. Abbott, Thoemmes Press, Bristol 1992. ---------, Lectures on Logic, translated and edited by J. M. Young, Cambridge University Press, New York 1992. ---------, Theoretical Philosophy, 1755 – 1770, translated and edited by D. Walford in collaboration with R. Meerbote, Cambridge University Press, New York 1992. ---------, Religion and Rational Theology, translated and edited by A. W. Wood and G. Di Giovanni, Cambridge University Press, Cambridge – New York 1996. ---------, Critique of Pure Reason, translated and edited by P. Guyer and A. W. Wood, Cambridge University Press, Cambridge 1998. ---------, Critique of the Power of Judgment, translated by P. Guyer and E. Matthews, Cambridge University Press, Cambridge – New York 2002.

Bibliography 

 149

---------, Critique of Practical Reason, translated by W. S. Pluhar, Hackett Publishing Co., Cambridge/ Indianapolis 2002. ----------, Theoretical Philosophy After 1781, edited by H. Allison and P. Heath, translated by G. Hatfield et al., Cambridge University Press, New York 2002. ---------, Prolegomena to Any Future Metaphysics. With Selections from the Critique of Pure Reason, translated and edited by G. Hatfield, Cambridge University Press, New York 2004. ---------, Notes and Fragments, edited by P. Guyer, translated by C. Bowman et al., Cambridge University Press, New York 2005. ---------, Anthropology, History, and Education, edited by G. Zöller and R. B. Louden, translated by M. Gregor et al., Cambridge University Press, Cambridge 2007. Kelly, S. D., “Demonstrative Concepts and Experience,” The Philosophical Review 110 (3), 2001, pp. 397-420. --------------, “The Non-Conceptual Content of Perceptual Experience: Situation Dependence and Fineness of Grain,” Philosophy and Phenomenological Research 62 (3), 2001, pp. 601-608. Kim, J., “Concepts and Intuitions in Kant’s Philosophy of Geometry,” Kant-Studien 97 (2), 2006, pp. 138-162. Kitcher, P., “Discovering the Forms of Intuition,” The Philosophical Review 96 (2), 1987, pp. 205-248. -------------, Kant’s Transcendental Psychology, Oxford University Press, Oxford – New York 1990. -------------, “Kant’s Unconscious ‘Given’” in: P. Giordanetti, R. Pozzo, M. Sgarbi (eds.), Kant’s Philosophy of the Unconscious, Walter de Gruyter, Berlin 2012, pp. 5-36. Kriegel, U., “Intentionality and Normativity,” Philosophical Issues 20, 2010, pp. 185-208. Larmore, Ch., “Attending to Reasons” in: N. H. Smith (ed.), Reading McDowell: On Mind and World, Routledge, London and New York 2002, pp. 193-208. Longuenesse, B., Kant and the Capacity to Judge: Sensibility and Discursivity in the Transcendental Analytic of the ‘Critique of Pure Reason’, Princeton University Press, Princeton 1998. Lord, B., Kant and Spinozism. Transcendental Idealism and Immanence from Jacobi to Deleuze, Palgrave Macmillan 2011. Macarthur, D., “McDowell, Scepticism, and the ‘Veil of Perception,’” Australasian Journal of Philosophy 81 (2), 2003, pp. 175-190. Macpherson, F., J. L. Bermúdez, “Nonconceptual Content and the Nature of Perceptual Experience,” Electronic Journal of Analytic Philosophy 6, 1998 (http://eprints.gla.ac.uk/584/1/macpherson_ ejap.pdf). Margolis, E., S. Laurence (eds.), Concepts. Core Readings, The MIT Press, Cambridge MA 1999, pp. 3-81. Martin, M. G. F., “Perception, Concepts, and Memory,” The Philosophical Review 101 (4), 1992, pp. 745-763. ---------------------, “The Transparency of Experience,” Mind and Language 17, 2002, pp. 376-425. Martin, W., “From Kant to Fichte” in: G. Zöller (ed.), The Cambridge Companion to Fichte, Cambridge University Press, Cambridge (forthcoming). McDowell, J., “Singular Thought and the Extent of Inner Space” in: P. Pettit and J. McDowell (eds.), Subject, Thought and Context, Clarendon Press, Oxford 1987, pp. 137-168. ----------------, “Knowledge and the Internal,” Philosophy and Phenomenological Research 55 (4), 1995, pp. 877-893. ----------------, Mind and World, Harvard University Press, Cambridge MA 1996. ----------------, “Précis of ‘Mind and World,’” Philosophical Issues 7, 1996, pp. 231-239. ----------------, “Intentionality as a Relation,” The Journal of Philosophy 95 (9), 1998, pp. 471-490. ----------------, “The Logical Form of an Intuition,” The Journal of Philosophy 95 (9), 1998, pp. 451-470. ----------------, “L’idealismo di Hegel come radicalizzazione di Kant,” translated by T. Fracassi, Iride 34, 2001, pp. 527-548.

150 

 Bibliography

----------------, “Hegel and the Myth of the Given” in: W. Welsh, K. Vieweg (eds.), Das Interesse des Denkens: Hegel aus heutiger Sicht, Wilhelm Fink Verlag, München 2003, pp. 74-88. ----------------, “The Disjunctive Conception of Experience as Material for a Transcendental Argument” in: F. Macpherson and A. Haddock (eds.), Disjunctivism: Perception, Action, Knowledge, Oxford University Press, Oxford 2008, pp. 376-389. ----------------, Having the World in View: Essays on Kant, Hegel, and Sellars, Harvard University Press, Cambridge MA 2009. Melnick, A., “Categories, Logical Functions and Schemata in Kant,” The Review of Metaphysics 54 (3), 2001, pp. 615-639. Metzinger, T., Being No One. The Self-Model Theory of Subjectivity, The MIT Press, Cambridge MA 1995. Miles, M., “Kant’s ‘Copernican Revolution:’ Toward Rehabilitation of a Concept and Provision of a Framework for the Interpretation of the Critique of Pure Reason,” Kant-Studien 97 (1), 2006, pp. 1-32. Nagel, T., The View from Nowhere, Oxford University Press, Oxford – New York 1986. Newcomb, N. S., J. Huttenlocher, Making Space. The Development of Spatial Representation and Reasoning, The MIT Press, Cambridge MA 2000. Norris, Ch., “‘Second Nature,’ Knowledge, and Normativity: Revisiting McDowell’s Kant,” Diametros 27, 2011, pp. 64-107. O’Regan, J. K., E. Myin, A. Noë, “Sensory Consciousness Explained (Better) in Terms of ‘Corporality’ and ‘Alerting Capacity,’” Phenomenology and the Cognitive Sciences 4 (4), 2005, pp. 369-387. Peacocke, Ch., “Scenarios, concepts and perception” in: T. Crane, The Contents of Experience, Cambridge University Press, Cambridge 1992, pp. 105-135. ------------------, “Nonconceptual Content Defended,” Philosophy and Phenomenological Research 58 (2), 1998, pp. 381-388. -----------------, “Does Perception Have a Nonconceptual Content?” The Journal of Philosophy 98 (5), 2001, pp. 239-264. Pereboom, D., “Kant on Intentionality,” Synthese 77 (3), 1988, pp. 321-352. Piłat, R., Doświadczenie i pojęcie, IFiS PAN, Warszawa 2006. Polanyi, M., The Tacit Dimension, University of Chicago Press, Chicago [1966] 2009. Prauss, G., Erscheinung bei Kant, Walter de Gruyter, Berlin 1971. ------------, Kant und das Problem der Dinge an sich, Bouvier, Bonn 1974. Pritchard, D., “McDowell on Reasons, Externalism and Scepticism,” European Journal of Philosophy 11 (3), 2003, pp. 273-294. Rensink, R. A., J. K. O’Regan, J. J. Clark, “To See or Not to See: the Need for Attention to Perceive Changes in Scenes,” Psychological Science 8 (5), 1997, pp. 368-373. Rescher, N., “Noumenal Causality” in: L. W. Beck (ed.), Proceedings of the Third International Kant Congress, D. Reidel Publishing Co., Dordrecht 1972, pp. 462-470. --------------, “On the Status of ‘Things in Themselves’ in Kant,” Synthese 47 (2), 1981, pp. 289-299. Rukgaber, M. S., “‘The Key to Transcendental Philosophy:’ Space, Time and the Body in Kant,” KantStudien 100 (2), 2009, pp. 166-186. Ryle, G., The Concept of Mind, The University of Chicago Press, Chicago 1949. Sellars, W., “Empiricism and the Philosophy of Mind” in: H. Feigl, M. Scriven (eds.), Minnesota Studies in the Philosophy of Science, vol. 1, University of Minnesota Press, Minneapolis 1956, pp. 253-329. -------------, “Some Remarks on Kant’s Theory of Experience,” The Journal of Philosophy 64 (20), 1967, pp. 633-647. -------------, Essays in Philosophy and its History, D. Reidel Publishing Co., Dordrecht 1974. -------------, Science, Perception and Reality, Ridgeview, Atascadero 1991. -------------, Science and Metaphysics. Variations on Kantian Themes, Ridgeview, Atascadero 1992.

Bibliography 

 151

Sgarbi, M., Kant on Spontaneity, Continuum, London 2012. Siemek, M., Idea transcendentalizmu u Fichtego i Kanta. Studium z dziejów filozoficznej problematyki wiedzy, PWN, Warszawa 1977. Smith, A. D., “Disjunctivism and Illusion,” Philosophy and Phenomenological Research 80 (2), 2010, pp. 384-410. Spelke, E. S., G. Gutheil, G. Van de Walle, “The Development of Object Perception” in: S. M. Kosslyn, D. N. Osherson (eds.), Visual Cognition. An Invitation to Cognitive Science, vol. 2, The MIT Press, Cambridge MA 1995, pp. 297-330. Spinoza, B. de, The Ethics and Other Works, edited and translated by E. Curley, Princeton University Press, Princeton 1994. Strawson, P. F., The Bounds of Sense. An Essay in Kant’s ‘Critique of Pure Reason’, Methuen, London 1966. Szałek, P., “Kant, Hegel and the Puzzles of McDowell’s Philosophy,” Diametros 29, 2011, pp. 110-123. Tetens, J. N., Philosophische Versuche über die menschliche Natur und ihre Entwickelung, Verlag von Reuther & Reichard, Berlin 1913. Tomaszewska, A., “Transcendental Object and the ‘Problem of Affection.’ Remarks on Some Difficulties of Kant’s Theory of Empirical Cognition,” Diametros 21, 2007, pp. 61-82. ---------------------, “Experience and Conceptual Content in Kant and McDowell. Remarks on ‘Empty Thoughts’ and ‘Blind Intuitions,’” Diametros 28, 2011, pp. 82-100. ---------------------, “McDowell and Perceptual Reasons,” Forum Philosophicum 17 (1), 2012, pp. 73-88. ---------------------, “Spinoza i Kant o naturze ludzkiego umysłu,” Filo-Sofija 17 (2), 2012, pp. 101-110. ---------------------, “Transcendental Idealism, Intuitions and the Contents of Perceptual Experience” in: S. Bacin, A. Ferrarin, C. LaRocca, M. Ruffing (eds.), Kant und Philosophie in weltbürgerlicher Absicht. Akten des XI. Kant-Kongresses 2010, De Gruyter, Berlin/Boston 2013, pp. 443-456. Twardowski, K., Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. Eine Psychologische Untersuchung, Philosophia Verlag, München-Wien 1982. Tye, M., “Nonconceptual Content, Richness, and Fineness of Grain” in: T. Gendler, J. Hawthorne (eds.), Perceptual Experience, Oxford University Press, Oxford – New York 2005, pp. 504-530. Vesey, G. N. A., “Seeing and Seeing As” in: R. J. Swartz (ed.), Perceiving, Sensing and Knowing. A Book of Readings from Twentieth-Century Sources in the Philosophy of Perception, Anchor Books, New York 1965, pp. 68-83. Vleeschauwer, H. J. de, The Development of Kantian Thought. The History of a Doctrine, translated by A. R. C. Duncan, T. Nelson and Sons, Edinburgh 1962. Vries, W. de, “Sellars vs. McDowell on the Structure of Sensory Consciousness,” Diametros 27, 2011, pp. 47-63. Waxman, W., Kant’s Model of the Mind. A New Interpretation of Transcendental Idealism, Oxford University Press, Oxford – New York 1991. Westphal, K., “Contemporary Epistemology: Kant, Hegel, McDowell,” European Journal of Philosophy 14 (2), 2006, pp. 274-301. Willaschek, M. (ed.), John McDowell: Reason and Nature: Lecture and Colloquium in Münster 1999, LIT Verlag, Münster 2000. Wilson, K. D., “Kant on Intuition,” Philosophical Quarterly 25 (100), 1975, pp. 247-265. Wittgenstein, L., On Certainty, edited by G. E. M. Anscombe and G. H. von Wright, Harper and Row, New York 1972. -------------------, Philosophical Investigations, translated by G. E. M. Anscombe, P. M. S. Hacker and J. Schulte. Revised fourth edition by P. M. S. Hacker and J. Schulte, Basil Blackwell, Oxford 2009. Wright, C., Wittgenstein on the Foundations of Mathematics, Duckworth, London 1980. -------------, “McDowell’s Oscillation,” Philosophy and Phenomenological Research 58 (2), 1998, pp. 395-402.

152 

 Bibliography

-------------, “Rule-Following without Reasons: Wittgenstein’s Quietism and the Constitutive Question,” Ratio (new series) 20 (4), 2007, pp. 481-502. Yolton, J. W., Perceptual Acquaintance from Descartes to Reid, Basil Blackwell and University of Minnesota Press, London 1984. ---------------- (ed.), Philosophy, Religion and Science in the 17th and 18th Centuries, University of Rochester Press, Rochester 1990. ----------------, Perception and Reality: A History from Descartes to Kant, Cornell University Press, Ithaca 1996.

Index “I think” 11, 93-96, 105, 139, 141 Absolute idealism 47, 50, 116, 121, 149 Abstraction 14, 16, 81, 97, 104, 107, 123, 145 Accuracy conditions 35 Accuracy 35 Action 10, 15, 22, 28, 37-38, 52, 55, 90, 102, 138 Affection 3-4, 46, 75, 114-115, 140-142, 145 Analogies of Experience 98, 103 Animals 9-10, 22, 32, 48, 91, 129, 133 Anomalous monism 129 Anthropology 17, 37, 49, 60, 67, 88, 112, 130, 151 Antinomy 2, 130 Appearance 15, 28, 58-59, 64, 66-67, 110-111, 119 Appearings 33-34, 87 Apperception 11, 63, 88, 90, 93-94, 96, 98-100, 113, 129, 131, 139, 144, 149 Aristotle 10, 20, 44, 74-75, 97, 113, 117, 131, 135-136, 148 Aspect 2, 24, 30, 47, 61, 65, 67, 74-75, 118, 129 Attention 14, 28, 60, 97, 120 Awareness 81, 91, 120, 128, 131 Bald naturalism 135 Belief 16, 19-20, 24, 31, 33-34, 36-37, 44, 49, 55, 69, 78, 82-83, 102, 121, 129, 137, 149 Blind 1, 5, 40-41, 52, 70, 85, 93, 132, 134 Blindsight 47, 83 Bodily nature 11-12, 128-129 Body 4, 11, 31, 49-53, 67, 77, 79, 104, 120-121, 127-130, 134, 145, 150 Categories 7, 11, 64-65, 72, 78-79, 85-88, 92-93, 97-100, 102, 105-106, 109, 111, 114-115, 139, 142, 144-145, 149 Categorization 10, 14, 42 Causal relations 72, 136 Causal 2, 4, 15, 31-32, 46, 64-67, 72-73, 129-131, 134-137, 139-142, 144, 146 Causality 2, 114, 140, 145 Cause 16, 64, 67, 69, 88, 137, 140, 146 Change blindness 28 Change blindness 28 Children 18, 49, 88, 103, 129, 133 Cognition 3-4, 6-8, 11, 13, 17-20, 22, 31-33, 38, 41-45, 47-53, 55, 58-60, 63, 68-70, 75-77, 79-81, 84, 86, 88-89, 92-93, 95, 97-100, 102-104, 106-116, 118-129, 131, 133, 137142, 144-147, 149 Coherentism 31-32, 40, 44

Complete concept 71 Compositionality 20 Concept 5-9, 11, 23-24, 26, 28-30, 38, 42-43, 45-46, 48-49, 51, 53, 55-57, 59-60, 70-72, 79-80, 82-83, 86, 89-91, 94, 97, 103, 111, 115, 124-125, 132, 139-141, 144 Conceptual capacities 9-10, 18-19, 23-24, 28-29, 40-41, 47, 50, 56, 88, 104, 107, 109, 121, 133-134, 136 Conceptual cognition 6-7, 52 Conceptual content 17, 20-21, 23-24, 29-30, 36, 40-41, 44, 86, 133 Conceptual content 17, 20-21, 23-24, 29-30, 36, 40-41, 44, 86, 133 Conceptual 1, 4-10, 17-21, 23-26, 28-30, 33, 3537, 40-41, 44, 46-47, 50, 52, 56, 62, 68-69, 71, 78, 80, 83-88, 91, 94-95, 99, 104, 107, 109, 114-116, 121, 132-136 Conceptualism 4, 7-8, 10-11, 14, 23-26, 32, 35, 38, 41, 43-44, 47-49, 63, 68, 70, 72, 90, 92, 109, 114, 118, 121, 132, 136-138, 144, 148 Conceptualization 11, 20, 26-27, 38, 94, 96, 98, 105, 133 Conditions of knowledge 80, 106-107 Conscious Experience 19, 47, 83-84, 91, 103, 131, 149 Conscious 10, 16, 19, 45, 47-49, 55, 59-61, 70, 73, 83-84, 88, 90-91, 93-96, 103, 129, 131, 133, 140, 149 Constructivism 121 Content conceptualism 11, 23-24, 90, 132 Content conceptualism 11, 23-24, 90, 132 Content nonconceptualism 24, 132-133 Content view 10, 13, 15-16, 36 Content 3-4, 8-11, 13-31, 33-38, 40-46, 49-72, 74, 76, 78, 80-87, 90, 92-94, 96, 103-124, 126-127, 131-133, 136-137, 140-141, 144146, 148-151 Copernican revolution 3, 44, 111 Correctness 35, 37, 111 Correctness 35, 37, 111 Demonstrative concepts 21, 26, 29-30, 151 Demonstrative identification 21, 25, 29, 50 Demonstrative 20-21, 25-26, 29-30, 50, 94, 115, 151 Determinism 2, 134, 139, 146

154 

 Index

Direct Realism 117, 134-135, 149 Direct Realism 117, 134-135, 149 Directions 42, 50-51, 87, 104, 120, 124, 142, 144 Discrimination 9, 28 Disjunctive theory of perception 10, 15, 34 Disjunctive 10, 15, 34-35, 138 Disjunctivism 10, 15, 117, 138 Dogmatic idealism 118, 121 Double-object 14 Dual aspect theory 129 Dualism of concepts 125, 127 Dualism of sensibility 75, 113 Dualism 49-50, 67, 75, 107, 113, 122-127, 136, 149 Dynamical categories 79 Efficient causes 131 Eidetic intuition 126 Embodied 16, 50, 100, 104, 120, 128, 144 Embodiment 128, 146 Embodiment 128, 146 Empirical cognition 3-4, 11, 17, 22, 31, 41, 44, 48-49, 58, 84, 88, 95, 98, 107, 116, 119, 123, 140, 144-145 Empirical concepts 115, 141, 145 Empirical psychology 112-113, 122, 130 Empirical psychology 112-113, 122, 130 Empirical realism 107, 115, 117-119, 121-122 Empirical realism 107, 115, 117-119, 121-122 Empirical representation 49 Empirical 1, 3-5, 11, 17, 21-22, 27, 31, 33-37, 4041, 43-45, 47-49, 58-60, 66-67, 70-71, 73, 77, 79-80, 84, 87-90, 93, 95-98, 107-110, 112-113, 115-123, 125-126, 130, 133, 135136, 138-141, 144-145, 148 Empiricism 40, 45, 121 Enantiomorphic figures 42, 109, 120 Epistemic constraints 12, 50, 113, 127 Epistemology 3, 19, 43, 63, 68, 71, 150 Experience 1-4, 7-30, 32-38, 40-41, 43-51, 53, 55, 57-59, 63, 65-69, 72, 75, 77-78, 81, 83-88, 90-98, 100, 102-104, 106-112, 114-123, 125, 127, 129, 131, 133-134, 136-140, 144, 146, 148-149, 151 Extension 5, 17, 77-78, 129-130, 142 Fact 4, 7-8, 15, 18, 22, 26-28, 30, 33-35, 37, 4344, 47-48, 50, 52, 54, 60, 63-66, 69, 72-74, 77, 80-81, 83-84, 91, 98, 100, 104, 108109, 112-116, 119, 132-133, 137, 144-145 Fact-awareness 83, 91, 95 Factive 19, 24, 34, 47, 121

Feeling 48, 51-52, 102-104, 120 Fineness of Grain 21-22, 25, 28, 50, 146, 151 Force Independence 21 Form of intentionality 104, 128 Form of intuition 53, 61, 67, 98 Form of intuition 53, 61, 67, 98 Form 6, 10, 16, 20-21, 28, 42, 46-47, 51, 53, 57-58, 61, 64, 66-67, 74-78, 83, 87-88, 92, 97-98, 104, 119-120, 123-124, 126, 128, 138, 151 Formal intuitions 99, 105 Forms of intentionality 116 Foundationalism 44, 135 Freedom 2, 23, 58, 100, 103, 129-130, 138, 148 General logic 86 General representation 38 Geometrical optics 96 Geometry 9, 78-79, 96, 98, 125, 148-149, 151 God 4, 33, 126, 146 Hallucination 15 Highest common factor 14, 34, 117 Idealism 11-12, 47, 50, 93, 99-100, 105-110, 112130, 138, 140, 142, 144-146, 148-149, 151 Ideality of time 113 Identity 26, 58, 90, 100, 106, 113, 122, 144 Illusion 15, 21, 28, 83, 118, 148 Image 59, 80-81 Imagination 16, 42, 50, 78, 80, 89, 91-95, 99100, 102-105, 114, 122-124, 133 Immediacy 40, 45, 55, 70-71, 81-82 Impression 66 Inattentional blindness 28, 146 Inattentional blindness 28, 146 Incongruent counterparts 42, 109, 120, 144 Indirect realism 15 Information 16, 19-20, 25, 28, 36-37, 47, 49, 5253, 55, 96, 103-104, 131, 138, 140, 149 Informational state 20, 96 Inner sense 58, 119, 122 Intellect 16, 27, 56, 75, 113, 123 Intellectual cognition 4, 50 Intellectual intuition 112, 123, 126 Intelligibility 3, 47, 131, 135-136, 138, 142 Intensional contexts 73-74 Intensional 73-74 Intentional properties 36 Intentional 22, 36, 41, 47, 67, 71-73, 76, 85, 137, 140 Intentionality 8, 22, 32, 35-38, 41-42, 71, 73, 8788, 90, 104, 116, 128, 132, 144, 149, 151

Index  Intuition 3-4, 6-7, 10-11, 33, 38, 41-42, 46-47, 49-51, 53, 58-61, 64, 66-81, 84-87, 89, 91-92, 95-100, 103, 108-112, 114-115, 119, 121-126, 128-129, 133, 139, 141-142, 144145, 151 Intuitive cognition 4, 108 Intuitive cognition 4, 108 Intuitive intellect 123 Intuitive understanding 123 Intuitive understanding 123 Intuitive 4, 6, 49, 67-68, 70, 84, 87, 90, 98-99, 108, 115, 122-123 Justification 31-35, 37, 40-41, 44, 50, 73, 75, 79, 88, 136-138 Knowledge 1-2, 4, 10, 15, 19, 31, 33-34, 37, 40, 43-45, 48-49, 52, 61, 65-66, 70, 74, 79-83, 87, 90-91, 93, 106-107, 111-113, 121-122, 125-126, 128, 133, 138, 142, 145-146, 148151 Law of contradiction 5-6 Law 2-3, 5-6, 74-75, 102-103, 134-136 Linguistic turn 7-8 Logic 5-7, 30, 38, 43, 48-49, 68-71, 86, 100, 133, 139, 147, 150 Logical possibility 43, 125-126 Manifold of intuition 64, 89, 96, 99 Material idealism 118 Materialism 138-139, 142, 146 Mathematical categories 79 Mathematical cognition 53, 80, 99, 108 Mathematics 44, 79-80, 91, 125-126 Matter of cognition 141-142 Matter 3, 20, 27, 31-32, 43-44, 46-48, 51, 60-61, 66-67, 70, 74, 76, 79-80, 82, 84, 106, 113, 118-119, 125, 141-142, 145 Memory 25-28, 50, 146, 151 Mental content 35, 57, 80, 103, 117, 121, 149 Mental geography 129 Mental states 13, 17, 31, 36, 55, 59, 63-64, 66, 95, 119-120, 128, 136-137, 140 Mental 3, 8, 12-13, 17, 31, 35-36, 45, 55-57, 5961, 63-66, 80, 84, 89, 94-95, 98, 103, 105, 117, 119-121, 128-131, 133, 136-137, 139140, 142, 145-146, 149 Metaphysical cognition 108, 126 Metaphysics 2-3, 20, 43, 45-46, 74, 85, 93, 116, 118, 120, 124-126, 129-130, 150-151 Monad 131 Natural history 135-136 Naturalism 127-142, 146-147, 149

 155

Nature 2-4, 7, 9-12, 14, 16, 23, 38, 40-42, 50-51, 53, 56-57, 61-62, 65, 67-68, 84, 93, 96, 100, 104, 114, 126-130, 132-138, 142, 146, 150-151 Nonconceptual Content 8-9, 11, 17-22, 24-25, 28-30, 36-38, 41-42, 44-45, 49-51, 53-56, 58, 60, 62, 64, 66-72, 74, 76, 78, 80, 82, 85, 93, 96, 104-110, 112-122, 124, 137, 144, 146, 148-151 Nonconceptualism 7-8, 10-12, 23-24, 38, 44, 47-50, 71, 86, 105, 107-110, 118, 127-133, 136, 144, 148 Non-conscious spontaneity 91-92, 94, 133-134 Nonrepresentational Content 16-17, 55, 59, 61, 63, 65, 72 Nonrepresentational 11, 16-17, 55, 59, 61, 63, 65, 72-73, 85, 114, 116, 137, 141 Non-sensible intuition 126 Normative account of content 35 Normative facts 35-36, 38 Normative 9, 27, 35-36, 38, 108, 149 Normativity 2, 35-37, 42, 109, 151 Noumena 68, 106, 124, 145 Noumenal Causality 114, 140 Noumenal Causality 114, 140 Object 3-4, 7-8, 14-19, 21-22, 26-27, 30, 32, 3738, 42, 45-46, 48-49, 51, 53, 56-61, 64-67, 69-73, 75-78, 80, 82-85, 87-88, 90-91, 9799, 102-103, 106, 110-111, 115-125, 127-130, 134, 138-145 Objective reality 15, 35, 46, 125, 141 Objective validity 7-8, 88, 125, 140 Objective 7-8, 15, 19, 27, 33, 35, 41, 46, 49, 59, 61-63, 65-67, 88, 95, 104, 124-125, 140-141 Outer sense 119-120 Pain 15-16, 48, 64, 130 Paralogism 118 Perception of distance 53, 97 Perception 3-4, 10-11, 13-24, 26-28, 30-38, 42, 45, 47-50, 53, 59, 62, 68, 70, 72-74, 81-85, 88, 91-92, 96-97, 101-103, 109-110, 114, 117-121, 131, 133, 138, 140, 148-151 Perceptual content 10, 13, 21, 26, 29, 31, 33-34, 44, 94, 105, 121, 132-133 Perceptual content 10, 13, 21, 26, 29, 31, 33-34, 44, 94, 105, 121, 132-133 Perceptual reasons 24, 33, 35, 137-138 Perceptual reasons 24, 33, 35, 137-138 Perceptual 4, 7-8, 10-11, 13-14, 16-36, 38, 40-41, 43-48, 50, 53, 55, 60, 63, 68, 70, 73, 82, 87-

156 

 Index

88, 92-94, 96, 103-105, 107-110, 117, 120122, 131-134, 136-138, 141, 146, 148, 151 Phenomena 33, 47, 58, 67-68, 84, 106, 117, 124, 128, 145-147 Phenomenology 2, 16, 22, 25, 27, 29-30, 50, 88, 112, 117, 120, 137, 148-150 Physical objects 14, 40, 45, 80, 102, 117, 119, 129, 136, 146 Physiology 49, 68, 84, 134 Platonism 135 Possibility 2, 4-6, 31, 43, 50, 52, 59-60, 73, 75, 84, 92-93, 98, 106, 108, 111, 115-118, 120, 124-126, 134, 136, 138-139, 146 Pre-cognitive 32, 63, 128, 140-141 Primary qualities 65 Principle of identity 26 Principle of significance 5 Principle of sufficient reason 134, 140, 146 Principle 5-6, 21, 26, 30, 58, 75, 88, 90, 94, 111, 124, 126, 129, 134, 140, 146 Problem of Affection 3, 114, 140 Problematic idealism 118 Proposition 69, 73, 95 Propositional content 31 Proprioception 120, 149 Proprioceptive 51, 120 Psychology 3, 27, 32, 41, 53, 55, 62, 112-113, 122, 130, 137-141, 146, 151 Pure concept 46, 59, 86, 141 Pure intuition 53, 61, 77-78, 81, 85, 97 Qualities 16, 19, 64-65, 102 Rational illusions 27 Rational psychology 27 Rational relations 32, 136-137 Rational 2, 10, 21, 24, 27, 31-33, 36, 40, 47-48, 63, 68, 80, 102-104, 120, 129, 131, 136138, 142, 150 Rationalist metaphysics 125 Rationality 2, 9-10, 26-27, 135-137, 144 Real possibility 4, 125 Realism 15, 107, 109-110, 115, 117-119, 121-122, 134-135, 149 Realm of law 3, 134-136 Reason 2, 5-11, 17, 19, 23, 27, 33-34, 36-38, 41, 44-46, 48, 53, 56-61, 63-64, 68-70, 72, 74, 77, 79-81, 86, 88, 90, 92-95, 97, 99-100, 102-107, 109-116, 118-120, 122-127, 130, 134-135, 138-142, 144-146, 148-151 Receptivity 3, 40-41, 46, 68, 100, 109, 123, 133, 140

Recognitional capacities 26 Reference Determinacy 21 Reference 8, 20-22, 25, 29, 46, 50-51, 60-64, 66, 71, 73, 75, 85, 91, 96, 109, 120-121, 145, 149 Representation 8-9, 11, 14, 38, 42, 48-49, 52-57, 59-61, 63-64, 66, 74-80, 84-85, 87, 89-91, 94, 96-99, 103, 121, 124, 128, 130-132, 139, 148 Representational content 3, 11, 13-14, 16-18, 25, 53, 55, 57, 69, 72, 84, 93, 105, 114, 116, 127, 133, 137, 145 Representational system 55 Representational 3-4, 11, 13-14, 16-18, 25, 36, 42-43, 45, 53, 55-57, 61, 66, 69-70, 72, 75-76, 84-85, 87, 93-95, 105, 108-110, 114, 116, 121, 127-128, 131-133, 136-137, 139141, 145, 148 Richness 21, 25-29, 43, 50, 71, 94, 132-133 Rule 5, 9, 53, 72-74, 80, 89-91, 95 Rule-governed 11, 24, 73, 87, 105, 133 Rules of synthesis 81 Savage 30, 70, 83, 132 Scenarios 34, 121 Schemata 71, 80, 85, 92-93, 95 Schematism 41, 50, 72, 80, 91-92 Science 2-3, 7, 9, 19, 28, 32, 37, 40, 42-46, 6769, 84, 125, 129, 131, 140, 146, 149-150 Second nature 2, 4, 10, 133-137, 146 Secondary qualities 16, 65 Sensation 28, 48-49, 59-60, 66-67, 74, 77, 84, 139, 149 Sensationism 16-17, 43, 63, 97, 114, 149 Sense data 38, 45-46, 117 Sense impressions 45, 47, 67 Sense 3-5, 8-9, 13, 15, 21-22, 36-38, 41, 43, 4547, 49, 52, 57-59, 62, 65-67, 72, 74-75, 78, 80, 82, 85, 88, 92, 94-95, 99, 102-103, 107, 110-111, 113, 117, 119-120, 122, 124, 130131, 133-137, 142, 145, 147 Sensibility 3, 11, 17, 32, 41, 43, 49-50, 53-54, 56, 58-59, 63-64, 68-70, 72, 75, 77, 81, 84, 89, 93, 96, 98-100, 103, 105, 108, 110-113, 115, 119, 123-126, 139, 145, 151 Sensible intuition 10, 69, 72, 77, 100, 119, 126 Sensorimotor 88, 104 Sensory content 4, 11, 16-17, 36, 40, 44, 46, 5253, 72, 109, 111, 113-114, 121, 137, 141 Sensory content 4, 11, 16-17, 36, 40, 44, 46, 5253, 72, 109, 111, 113-114, 121, 137, 141 Simple seeing 19, 82-83, 131

Index  Singular representation 38 Skepticism 13, 31, 33-34, 44, 117 Space of reasons 2-3, 25, 34, 134-136, 138, 150 Space perception 121 Space 2-3, 5-6, 10-11, 15, 25, 34, 42, 45-47, 50-54, 57-58, 61, 67, 73, 76, 78-80, 85-87, 89, 91, 93, 95-100, 102, 104-105, 108-111, 113-122, 124-125, 128, 134-136, 138, 143146, 148, 150-151 Spatial relations 87, 104 Spatial 32, 47, 51-52, 62, 76, 78, 87, 96-97, 103104, 121, 140 Sperling’s experiment 28 Spontaneity 2, 40, 68, 73, 91-94, 96, 123, 129, 133-134, 136, 138-139 State conceptualism 23-24, 90, 92, 132 State nonconceptualism 23-24, 132-133 Subpersonal states 45, 93 Synthesis of apprehension 89-90 Synthesis of recognition 89 Synthesis of reproduction 89 Synthesis 11, 46, 56, 72, 79-81, 85, 89-90, 92, 98-99, 102, 105, 114, 138, 149 Synthetic a priori 78-80, 125-126 Tacit knowledge 52, 91 Temporal properties 47, 61 Temporal relations 61 Temporal 4, 6, 47, 58-59, 61, 64-66, 103-104, 106 That-shade argument 26 Theory of concepts 8-9 Thing in itself 110-111, 114, 116 Thing 5-6, 14, 23, 27-28, 32-34, 38, 56, 81-82, 84, 110-111, 114, 116, 124, 128-129, 138, 146 Thing-awareness 83, 91 Time 2, 5-6, 11, 25, 27-28, 31, 45-47, 50-51, 53, 57-58, 61, 64, 67, 73, 78-80, 85-86, 89-91, 93, 95-100, 102, 105, 108, 110-111, 113-117, 119-122, 124-125, 131, 135, 145 Transcendental Aesthetic 3, 17, 43, 49, 53, 57, 62, 65, 67, 76-78, 95, 97, 99-100, 105, 111, 113, 115, 125, 149

 157

Transcendental apperception 11, 90, 93-94, 98, 129, 139, 144 Transcendental Argument 10, 15, 35, 138 Transcendental cognition 106 Transcendental cognition 106 Transcendental Deduction 11, 64, 72, 85-86, 88, 92, 97, 100, 106, 114, 139, 142, 144 Transcendental idealism 12, 47, 50, 93, 99-100, 105-110, 112-130, 140, 142, 144, 146, 148, 151 Transcendental idealism 12, 47, 50, 93, 99-100, 105-110, 112-130, 140, 142, 144, 146, 148, 151 Transcendental imagination 80, 91-93, 99-100, 105, 123, 133 Transcendental imagination 80, 91-93, 99-100, 105, 123, 133 Transcendental Logic 5, 7, 68, 70, 86 Transcendental Logic 5, 7, 68, 70, 86 Transcendental object 140-142, 145 Transcendental object 140-142, 145 Transcendental philosophy 3, 44, 51, 100, 106, 112, 138 Transcendental Psychology 41, 53, 62, 113, 137141, 151 Transcendental Psychology 41, 53, 62, 113, 137141, 151 Transcendental schematism 50 Transcendental 3, 5, 7, 10-12, 15, 17, 19, 27, 35, 41, 43-44, 47, 49-51, 53, 57, 60, 62, 6468, 70, 72, 76-78, 80, 85-86, 88, 90-95, 97-100, 105-130, 133, 137-142, 144-146, 148-149, 151 Understanding 1, 5, 7-9, 11, 17, 19, 23, 37, 41-43, 46-49, 51-53, 57-59, 68-70, 72, 75, 77-79, 83, 85-89, 92-94, 99-100, 102-105, 109, 111-115, 123-129, 133-134, 136, 138-139, 142, 145, 148, 150 Unity of consciousness 90, 98 Unity 16, 46, 64, 69, 72, 74, 79, 86-88, 90, 98100, 141 Veil of perception 3, 14-15, 117, 151 Waterfall Illusion 21, 148