Kant's Aesthetic Epistemology: Form and World

‘A brilliantly executed presentation of a case for a new and fruitful way of perceiving Kant’s Critical project.’ John L...

Author: Fiona Hughes

42 downloads 574 Views 10MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

‘A brilliantly executed presentation of a case for a new and fruitful way of perceiving Kant’s Critical project.’ John Llewelyn, Reader in Philosophy, University of Edinburgh

Cover design: Cathy Sprent

KANT’S AESTHETIC EPISTEMOLOGY FORM AND WORLD

KANT’S AESTHETIC

EPISTEMOLOGY FORM AND WORLD

Fiona Hughes

Drawing on resources from both

the Analytical and Continental traditions, Fiona Hughes argues that a comprehension of Kant’s aesthetics is necessary for grasping the scope and force of his epistemology. She draws on phenomenological and aesthetic resources to bring out the continuing relevance of Kant’s project. One of the difficulties faced in reading The Critique of Pure Reason is finding a way of reading the text as one continuous discussion. This book offers a reading at each stage of Kant’s epistemological argument, showing how various elements of Kant’s argument, often thought of as extraneous or indefensible, can be integrated.

FIONA HUGHES

Cover image: Ian Hamilton Finlay, Little Sparta Photograph by Andrew Lawson

KANT’S AESTHETIC EPISTEMOLOGY

Fiona Hughes is Lecturer in Philosophy at the University of Essex. She is the author of published articles on Kant and aesthetics and on Nietzsche, aesthetics and hermeneutics. She is currently writing a book on aesthetics and the arts.

This incisive study, arguing for the centrality of aesthetics in philosophy, and within experience in general, challenges a blind spot in the AngloAmerican tradition of philosophy and will contribute to a growing interest in the general signiﬁcance of aesthetic culture.

Edinburgh University Press 22 George Square Edinburgh EH8 9LF www.eup.ed.ac.uk ISBN 978 0 7486 2122 4

Edinburgh

Fiona Hughes

M710 HUGHES PRE M/UP.qxd

29/3/07

8:11 AM

Page i Gary Gary's G4:Users:Gary:Public:Gary's Job

Kant’s Aesthetic Epistemology


29/3/07

8:11 AM

Page ii Gary Gary's G4:Users:Gary:Public:Gary's Jo

‘Music . . . is not an expression of what life is, but an expression of what life could be, or what it could become.’ Daniel Barenboim. Reith Lectures 2006

In memory of George Elder Davie (1912–2007), student of Kemp Smith and scholar of the Scottish (and Irish) Enlightenment.


29/3/07

8:11 AM

Page iii Gary Gary's G4:Users:Gary:Public:Gary's J

KANT’S AESTHETIC EPISTEMOLOGY

FORM AND WORLD 2 Fiona Hughes

Edinburgh University Press


29/3/07

8:11 AM

Page iv Gary Gary's G4:Users:Gary:Public:Gary's Jo

© Fiona Hughes, 2007 Edinburgh University Press Ltd 22 George Square, Edinburgh

Typeset in Sabon by Servis Filmsetting Ltd, Manchester, and printed and bound in Great Britain by Biddles Ltd, King’s Lynn, Norfolk A CIP record for this book is available from the British Library ISBN 978 0 7486 2122 4 (hardback)

The right of Fiona Hughes to be identified as author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.


29/3/07

8:11 AM

Page v Gary Gary's G4:Users:Gary:Public:Gary's Job

Contents

Acknowledgements References to Kant’s Works

vi viii

Introduction

1

1

The Centrality of the Problem of Formalism

8

2

Formalism and the Circle of Representation

49

3

Formal Idealism and the Aesthetic Condition of Experience

86

4

The Deep Structure of Synthesis

112

5

The Completion of the Subjective Deduction in the Deductions of the Critique of Judgement

169

A Priori Knowledge as the Anticipation of a Material Given and the Need for a Spatial Schematism

207

Empirical Systematicity and its Relation to Aesthetic Judgement

248

Aesthetic Judgement’s Exemplary Exhibition of Cognition

277

6

7

8

Afterword

311

Bibliography

316

Author/subject index

321


29/3/07

8:11 AM

Page vi Gary Gary's G4:Users:Gary:Public:Gary's Jo

Acknowledgements

This project has been a long time in gestation and I owe many thanks to particular people and the creative milieu in a number of different institutions. My acknowledgements have to reach back at least as far as my days as an undergraduate at the University of Edinburgh and to a circle of friends. Later, at the University of Oxford I gained another informal research group, some philosophers like Sally (Sarah) Gibbons, some not. My supervisor, Michael Rosen, was unfailingly helpful. At Oxford I found further resources to help me situate my philosophical interests within a more general framework. The motivation to do so – traces of which I hope are still visible at the margins of this book – arose in great measure from the intellectual culture in Edinburgh and the influence of George Davie’s The Democratic Intellect. From my appointment at Essex as a new lecturer, I had the great fortune to expand my horizons through exposure to the style of thinking of the philosophical art historian, Michael Podro, who became a true colleague and friend. For many years, we ran joint seminars together and if the students enjoyed them and learnt as much as I did, I am satisfied. Over the years, my students at Essex – mainly in the philosophy department, but also those working in other disciplines – have been an ongoing source of inspiration. I am particularly grateful to a number of people who gave their valuable time to read and comment on parts or the whole of my manuscript. Needless to say, any remaining problems are my own, not theirs. These include Neil Cox, Douglas Burnham, Sebastian Gardner, Dana MacFarlane, Wayne Martin and Michael Podro. I want to especially thank John Llewelyn. He has been an inspiration and a great support since my Edinburgh days. I cannot adequately capture his scrupulous attention to detail, combined with unstinting generosity – not only in reading my manuscript, but also in philosophy and in life in general. Other friends helped on specific points and kept my spirits up when the going was tough. In 2003 the Arts and Humanities Research Board (now AHRC) awarded me a Research Leave Scheme Grant allowing one term’s vi


29/3/07

8:11 AM

Page vii Gary Gary's G4:Users:Gary:Public:Gary's J

Acknowledgements leave from teaching. In conjunction with sabbatical leave from the University of Essex, this allowed me to complete a first draft of my manuscript. Many individual members of staff at Edinburgh University Press have been very helpful. I must mention, in particular, Jackie Jones and Carol Macdonald, my editors, who, without fail, have been both constructive and insightful. I am grateful to Nicola Gray, whose perceptive interventions were extremely helpful in the preparation of my manuscript for publication, and to James Corby for preparing the index. I would also like to especially acknowledge here the great support given me by my mother and also by my father, when he was alive. From early on, they both were prepared to make many sacrifices in support of my education. Finally, thanks to John Walshe, who gives so much encouragement and, refreshingly, not necessarily in ways I might expect!

vii


29/3/07

8:11 AM

Page viii Gary Gary's G4:Users:Gary:Public:Gary's

References to Kant’s Works

All references to Kant’s works are to Kant’s Gesammelte Schriften herausgeben von der Königlich Preußischen Akademie der Wissenschaften [Berlin 1902– ], commonly referred to as the ‘Akademieausgabe’ [ AA from now on]. When necessary I offer my own translations. Otherwise, I make use of the following standard versions in English: [1781 and 1787] (1933, 2nd edn) Critique of Pure Reason, trans. N. Kemp Smith, London and Basingstoke: Macmillan [ CPR]. Page references to the first edition of 1781 are preceded by A, while those to the second edition of 1787 are preceded by B. Material is reprinted by permission of Macmillan Publishers Ltd. The A edition appears in Volume IV of the Akademie edition, while Volume III comprises the B edition. [1790] (1987) Critique of Judgment, trans., W. Pluhar, Indianapolis, IN: Hackett. All references to this text are preceded by AA. Material is reprinted by permission of Hackett Publishing Company, Inc. All rights reserved. The main text appears in Volume V of the Akademie edition. An alternative, unpublished version of the Introduction, commonly referred to as the ‘First Introduction’ appears in Volume XX. References to the earlier and longer version will be distinguished by a prime mark, for example, AA 191′. Other translations referred to: (2004) Metaphysical Foundations of Natural Science, ed. M. Friedman, Cambridge: Cambridge University Press. (1963) On History, ed. L. W. Beck, Indianapolis, IN: BobbsMerrill for ‘Perpetual Peace’ and ‘The End of All Things’. (1991) Political Writings, ed. H. S. Reiss, Cambridge: Cambridge University Press.

viii

M710 HUGHES TEXT M/UP.qxd

29/3/07

8:12 AM

Page 1 Gary Gary's G4:Users:Gary:Public:Gary's Jo

Introduction

In ‘Little Sparta’, a garden in the Pentland hills outside of Edinburgh, two planks stretch across a stream. On each is inscribed: ‘That which joins and that which divides is one and the same.’ The inscription on one faces that on the other. At first sight we see a bridge, but then we see two planks divided from one another; looking again we see the two planks as one structure, realising that each bears the same inscription. The inscription stretching the length of each plank takes time to read and there is no one perspective from which both can be apprehended. Our realisation of the similarity of the two planks is only achieved through a process of recognition, for after an initial impression of unity, they appear as disrupting one another. Our aesthetic response is of unity achieved only through a process linking two distinct components. Much more could be said about this contemporary artwork by the Scottish artist Ian Hamilton Finlay, his use of a fragment from Heraclitus and the relation in which the work stands to the philosophical ideas that follow. In general my point is that through an aesthetic presentation – in this case, visual – a train of thought is instigated. What we see makes us think, and what we think makes us look further, in turn making us think more. While a whole philosophical treatise may, and indeed will in what follows, attempt to express a thought, this artwork quickly and effortlessly brings us into the centre of a complex set of associations. We are led in by what we see, just as we might be led across the stream by the bridge. We find that we are already in the midst of a way of seeing the world and of orienting ourselves within its perspective. Both artworks and aesthetic experiences of nature have the capacity to bring about such an insight, which counts as a form of reflection on the world and our place in it. This, I believe, is what Kant discovered in his Critique of Aesthetic Judgement. More specifically, I will argue in what follows that he came to the conclusion that a harmonious combination of mental capacities, characteristic of aesthetic judgement, allows for a reflection on a more general cooperation that is necessary for any experience whatsoever. 1


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology Finlay’s work is particularly pertinent to establishing this role for aesthetics, for the cooperation of orientations in question is not one of bland uniformity. We are complex beings bound to the world by a sensory or aesthetic capacity to take things in, namely, receptivity. But we are also able to stand back from our environment in a reflective stance. That which joins aesthetic receptivity with reflection is a third capacity, Kant tells us, imagination. Imagination makes possible a unification of what is ineliminably different – that is, our capacity to be affected and our capacity to reflect on the world. The mediating faculty does not achieve this by eliminating the distinctiveness of the two faculties, but by finding a process that allows for a relation between one side of our disposition and the other, just as the bridge at ‘Little Sparta’ makes possible a crossing from one side of the stream to the other. In the philosophical case under consideration in this book, imagination effects a crossing by making possible synthesis, a process of connecting something given to our senses with a concept that aims to make sense of it. I will argue that aesthetic phenomena allow us to reflect on the process of synthesis that is the condition for subjects such as ourselves having access to an external world, which is ours, as our environment, but is not in our possession. This complex relation is also expressed in Finlay’s heraclitean reflection. The relation in which aesthetic receptivity stands to reflection is thus one that occurs across a gap. Synthesis in general and aesthetic harmony, in particular, are not in plainsong; their harmony is not uniform. The unity that is a result of synthesis and the harmony characteristic of aesthetic judgement both have difference as a constitutive moment, just like the bridge at ‘Little Sparta’. Unity and harmony are constructed out of distinctive ways in which we relate to the world and yet these orientations could not prevail in isolation from one another. We can only be receptive and we can only be reflective insofar as we are both at one and the same time. The background problem, out of which the particular arguments in this book emerge, is one of the place of aesthetics within philosophical practice and its relevance for experience in general. The way in which this problem is viewed can be captured as giving rise to two responses. In – for want of a better title – the Continental idiom of practising philosophy, it is widely accepted that aesthetics shares philosophy’s project and has no small part to play within it. Meanwhile – and, again, for want of a better expression, for there are many different types of philosophical analysis – in Analytical philosophy we find more of a hesitation in granting a central role to aesthetics. 2


29/3/07

8:12 AM


Introduction Aesthetics or the philosophy of the arts is practised seriously and to great local effect, but there is little confidence that aesthetics is anything more than a regional activity, secondary to the more essential sub-disciplines of epistemology, philosophy of mind and moral philosophy. The disciplinary constitution of the aesthetic is sometimes reinforced by a more general suspicion that aesthetic appreciation is, after all, an optional activity and may even be a distraction from the central questions of philosophy and of life. While I do not pretend that I will be able to answer this general problem of the status of aesthetics in one book on Kant’s epistemology and aesthetics, I do hope to make some progress by establishing the importance he gave to the relation between aesthetic judgement and epistemology. Many other writers on Kant – a number of whom I will discuss – have made significant progress in uncovering the role played by aesthetic judgement within his philosophical system as a whole; however, I believe that, as yet, there is no convincing account of how the systematic link between cognitive and aesthetic judgement might be defended in a thoroughgoing fashion. My hope is that this work will help secure the commitment to aesthetics held by many Continental philosophers. Meanwhile, the way in which I pursue my account of the link between epistemology and aesthetics may persuade some of an Analytical persuasion to venture onto the bridge. Ian Hamilton Finlay’s visual metaphor is useful here, for there is no middle ground and no well-established territories on either side. ‘Analytical philosophy’ is a plurality of practices and so too is ‘Continental philosophy’. We fool ourselves to think that one well-defined practice opposes another. But what we can hope for is that there may be crossings or mediations of the stream. In a heraclitean frame of mind, I am inclined to think that the different ways of engaging in philosophy occur on the bridge, while philosophy is the stream. Why is it important to establish a role for aesthetics? I believe it is because aesthetics allows for an insight into the plural constitution of our experience in general and that once we lose sight of this, dogmatism threatens. If aesthetics does this – something I will argue through my interpretation of Kant in this book – it does so indirectly in the course of a preoccupation with an aesthetic phenomenon or event. I will argue that the indirectness of the reflection is a necessary one. There are two ways in which the term ‘aesthetic’ will be used in this book. The first sense is the common one of a pleasurable experience arising from the senses – be the event artistic or natural – and giving rise to a judgement of taste. The second is a more primary one 3


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology referring to any sensory experience whatsoever. In this book I will seek to show how, for Kant, the two senses of aesthetics are necessarily intertwined. The second sense of ‘aesthetic’, which figures directly in the title of this book, marks the status of knowledge from Kant’s point of view.1 Knowledge emerges in space and time, these being the context within which any sensory affect arises. Knowledge requires not only concepts, but also sensory or aesthetic givens, which he calls intuitions. I will argue that the wider ramifications of this stance, which is well known in the philosophical world at least, is that knowledge arises as a project undertaken over time and across space. Knowledge is not simply a fait accompli, but a process that marks any achievement that is reached.2 I will further argue that Kant comes to the view that aesthetic judgement – that is, ‘aesthetic’ in the first or common sense of the word – has the power to reveal this deeper truth about knowledge. The picture that emerges is one of a subject bearing certain broadly mental capacities with which he or she is able to take up and make sense of an environment or world of objects. I thus argue that the attention Kant pays to what he calls the ‘faculties’ or subjective capacities is not a hindrance for his intent of establishing the possibility of objective knowledge, but rather the means by which he achieves it. The engagement with Kant that emerges in the pages that follow is at times closely textual, but it is not a work in pure descriptive interpretation. I read Kant because he can help me make sense of problems that exist outside of the technical content that he himself developed. I believe that Kant also viewed philosophy in this way and that is why he has a continuing fascination for us. Although it is impossible to read Kant without entering into a series of technical debates about competing interpretations, the problem is institutionally elaborated, not institutionally constituted. In what follows, I will give an overview of what may be expected. I will not enter into the details of my arguments; there will be time enough for that later. Although my goal is to establish the importance of Kant’s aesthetics for his epistemology, my initial investigations are focused strictly on his theory of knowledge. More specifically, I begin by situating the problems I will address in later chapters within the context of the very rich interpretative reception of Kant’s epistemology that has developed over the last forty or so years. In later chapters I will argue that Kant’s aesthetic formalism – despite it often being the subject of criticism – provides resources for establishing the character of his less familiar formalist account of knowledge. 4


29/3/07

8:12 AM


Introduction Chapters 1 and 2 establish the centrality of cognitive formalism for both critics and defenders of Kant’s epistemology alike. A critical assessment of some of the major contributions to current debates provides a backdrop against which I will situate my own account of Kantian formalism. In Chapter 1 I am concerned with the way in which Kant has been accused of ‘impositionalism’, that is, the position that the mind imposes form or order on the world. The problem that seems to ensue is that, according to this view, knowledge would be not so much of things in the world, but rather a merely subjective projection of mind’s own structures. I examine how this criticism emerges from the readings of Robert Pippin, Paul Guyer, Dieter Henrich and Peter Strawson. In Chapter 2 I turn to some of Kant’s major defenders – namely, Gerd Buchdahl, Henry Allison and Béatrice Longuenesse – and show how they have argued that formalism does not lead to a subjective status for knowledge. While suggesting the beginnings of a way of defending Kant against the charge of impositionalism, I argue, nonetheless, that his defenders have not supplied an account of formalism that convincingly rebuts this charge. In Chapter 3 I begin my interpretation of formalism, starting from Kant’s own distinction between formal and material idealism. I suggest that the commitment to a material correlate of formal idealism is to be found in the initial paragraphs of the ‘Transcendental Aesthetic’. This is the material dimension of the aesthetic or sensory side of the dualist project. In Chapter 4 I turn to the reflective or conceptual side of dualism, as the condition of the possibility of synthesis. I argue that synthesis arises as a cooperation of a plurality of faculties and offer the ‘Transcendental Deduction’, in both its editions, as case studies. I conclude that only aesthetic judgement reveals the deep structure of the cooperative structure of synthesis within experience. In Chapter 5 I reassess the importance of the subjective deduction, first mentioned in the Preface to the A edition of the Critique of Pure Reason. I argue that the subjective deduction is the necessary corollary of the objective deduction of the categories and that the full account of the subjective side of cognition can only be found in certain passages in the Critique of Judgement. Establishing the relation between knowledge and aesthetic judgement cannot be achieved in a linear fashion. Just because the relation is reciprocal, there is no singular point of origin out of which the other term can be deduced. In Chapter 6 I turn back to the cognitive side of the relation, looking in more detail at the objective deduction, which 5


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology was addressed in nuce in Chapter 4. Having established the role of synthetic process in the preceding chapters, I argue that knowledge arises out of a process of subjective anticipation of a material given in experience. In Chapter 7 I take up the question of the purposiveness or systematicity of nature, introduced by Kant in the Critique of Judgement. The latter addition is important because, as I will show, it is necessary for empirical cognition. I argue that we can make sense of Kant’s claims that there is a link between systematicity and aesthetic judgements, once we distinguish between two levels at which purposiveness operates, while, at the same time, recognising that it combines a subjective and an objective orientation. I conclude that aesthetic judgements are exemplary for a general purposiveness of nature. In Chapter 8 I fine-tune this account, characterising the exemplarity of aesthetic judgements as contrastive or, as I call it, ‘contrapuntal’. Aesthetic judgement is not in any sense cognitive; nevertheless, in our aesthetic appreciation of artworks and of nature we are able to engage in an implicit reflection on the relation in which we stand to the world. This relation of subject and object is the fundamental engagement out of which knowledge emerges. In my Afterword I make some suggestions as to how I would further explore the relation in which the harmony characteristic of aesthetic judgements stands to the Kantian sublime. I return to Ian Hamilton Finlay’s heraclitean work at ‘Little Sparta’ to explain why, although the sublime has not been directly addressed in the body of this book, it still has a necessary role to play. The harmony of the faculties arises across difference; thus, the possibility of disruption is always already implicit. The beautiful marks a moment when the senses make sense of something in the world, while the sublime arises when the senses fail us. The beautiful is singular and, thus, indirectly reveals the fragility of our sensory grasp of the world or the possibility that sense might break down. The beautiful and the sublime cannot be articulated or experienced without the trace of the other.

Notes 1.

Throughout this book I will refer to ‘knowledge’ rather than to ‘cognition’, which is the more accurate translation of Erkenntnis. Kant’s project in the Critique of Pure Reason is to establish the possibility of objectively valid cognition, which would qualify as knowledge. He is not concerned with cognitions such as opinions or beliefs. So although 6


29/3/07

8:12 AM


Introduction

2.

Erkenntnis has a much broader range than ‘knowledge’, I think it is justifiable to use a term that is more intuitive and requires less qualification than the more technical ‘cognition’. My primary concern will be to establish the anticipatory status of formal, a priori cognition.

7


29/3/07

8:12 AM


1


What is the nature of Kant’s epistemological formalism, as his critics see it? I will argue that there are several versions, but that in all cases the criticism of formalism amounts to the charge that the outcomes of Kant’s method are ultimately subjective. Critics of Kant claim that his insistence that the form of experience arises from our minds, finally makes the empirical world of objects dependent on a form of subjectivity. If our minds introduce the form of experience, so the argument goes, then the objects we experience are only objects ‘for us’ and there is no access to a genuinely extra-mental world. In replying to this in later chapters I will argue that Kant’s formalism is an attempt to systematically link the subjective and objective conditions of experience. So while Kant’s critics are right to link formalism with a turn to the subject, they are telling only one side of the story. The project of the Copernican revolution is exactly that of showing how the turn to the subject will secure the possibility of knowledge of an objective world. But the full version of this project does not involve reducing objectivity to the subjective conditions of experience, which are necessary and not sufficient. Later I take up the problem of the empirical applicability of the subjective forms of thought by emphasising the aesthetic dimension of Kant’s critical turn. In this chapter I will show how formalism has been seen as a central issue for Kant’s epistemology. All of the authors I examine in this chapter conclude that Kant’s position is a formalist one. Paul Guyer and Peter Strawson argue that formalism results in the position that mind imposes order on objects.1 This is the thesis that is referred to as ‘impositionalism’. Robert Pippin and Dieter Henrich resist this conclusion, although I will show that ultimately both fail to provide a sufficiently robust defence of Kant’s position. Paul Guyer has been the most robust defender of the view that Kant’s epistemology portrays mental formative activity as an imposition on matter. As I will discuss in detail later, Guyer distinguishes between two models of the relation between mind and the material given. If the conformity between mind and reality is always already 8


29/3/07

8:12 AM


The Centrality of the Problem of Formalism guaranteed – a thesis to which he believes Kant is committed – then this can only be explained by: the mind’s ability to impose its rules on a pliable or formless material, or at the very least on a material any intrinsic form of which is inaccessible to the mind and can be replaced by a form of the mind’s own making.2

Guyer contrasts this to the view that: it is possible to think that we can know in advance of successful experience what objects must be like if we are to experience them but that we have no special power to make objects be like that, thus to make experience possible.3

Thus the impositionalist interpretation of Kant attributes to him the view that mind imposes form on a material given and that this results in our being able to produce the objects of experience.4 Now why should attributing impositionalism to Kant lead to difficulties in his account of knowledge? Although none of the interpreters discussed in this chapter would deny that the subject plays a necessary constitutive role in knowledge acquisition, the worry is that, from the perspective of the formalist method, the subject is faced only with our own mental structure, in the absence of any real encounter with an object. Robert Pippin’s Kant’s Formalism, which I discuss in the first section of this chapter (pp. 10–17), still counts as seminal for this debate. His critical, but sympathetic and subtle, reconstruction of Kant’s epistemology sets the standard for subsequent discussions. His eventual rejection of Kant’s project arises from the latter’s aim of establishing an a priori framework for knowledge that relies on a questionable account of pure intuition. An important aspect of his account is the claim that reflective judgement amounts to a mental projection that reinforces the subjectivist status of Kant’s epistemological project. Paul Guyer, whom I discuss in the second section (pp. 17–25), is, in general, critical of the formalist dimension of Kant’s method. However in his assessment of the latter’s epistemology, his criticism is, more precisely, of Kant’s ‘impositionalism’. Guyer’s aim is to rehabilitate a non-formalist core ultimately to be found only in certain manuscripts not prepared for publication by Kant. His view is that Kant often resorts to arguing that the mind imposes form on reality. In contrast to this, Guyer aims to unearth a ‘realist’ Kant who eschews the assumptions of transcendental idealism and is committed to the 9


29/3/07

8:12 AM

Page 10 Gary Gary's G4:Users:Gary:Public:Gary's J

Kant’s Aesthetic Epistemology independent existence of an extra-mental world accessed by a causal perceptual process. Like Pippin’s, this criticism amounts to the charge that Kant’s formalism leads him into an unwarranted subjectivism vis à vis knowledge. Dieter Henrich, whom I discuss in the third section (pp. 25–35), has been hugely influential in the reception of Kant both in the German and English speaking worlds. His reconstructive argument has greatly encouraged Guyer and others for whom the absolute Cartesian certainty of Henrich’s starting place entails impositionalism. It is for this reason that I include a discussion of Henrich at this stage of my account, even though he defends Kant and denies that his interpretation entails impositionalism. An earlier critique of Kant’s formalism is to be found in the work of Strawson, whom I discuss in the final section. He sees the dominant problem with Kant’s approach as his dependence on ‘faculty talk’. Instead of focusing exclusively on the objective empirical contents of experience, Kant repeatedly expresses his position in terms of the relations between mental capacities. The danger is that the critique of knowledge falls into psychologism. Importantly, Strawson also accuses Kant of impositionalism. The supposed problem, once again, is that Kant’s formalism – now specifically the claim that experience is dependent on certain formal mental capacities – results in knowledge arising from the mind enforcing its patterns onto reality. In the course of this chapter we will see that formalism takes on a specific shape for each of these authors. Interestingly, all of them, except Strawson, entertain the possibility that there may be an alternative version of formalism that does not fall into the impositionalist trap.

I Pippin’s Critique of Kant’s Formalism Robert Pippin’s Kant’s Theory of Form offers a thoroughgoing and systematic account of Kant’s epistemological project. While Pippin is critical of Kant’s formalism for resulting in a merely subjective legitimation of knowledge, he nevertheless resists interpretations that accuse Kant of impositionalism, the thesis that order arises in experience due to the imposition of mental forms. Kant, as interpreted by Pippin, fails to fulfil his deepest held undertaking to achieve a formal epistemology applicable to empirical experience, despite his commitment to ‘empirical guidedness’, that is, 10


29/3/07

8:12 AM


The Centrality of the Problem of Formalism the thesis that the forms of experience must be applied to empirical matter.5 Following the transcendental turn, empirical matter guides experience strictly in accordance with rules we introduce ourselves.6 Consequently, Kant’s position, which combines transcendental idealism with empirical realism, can only be ‘a comprehensive attempt to determine what can count as experience of objective reality “for us” ’.7 While Kant’s idealism is, Pippin believes, defensible, his formalism is ultimately subjectivist. From the outset Pippin disagrees with those critics for whom Kant’s epistemology amounts to the view that the subject projects or imposes form or structure onto a material world that, in itself, has no potential meaning. After the transcendental turn, matter can only have meaning in relation to conceptual form, but at the same time concepts are only meaningful insofar as they apply to something given in experience. This is the force of Pippin’s concern with empirical guidedness, which explains why Pippin finds himself in opposition to Norman Kemp Smith, Hans Vaihinger and others who he considers have too mentalist an interpretation of Kant.8 For Kemp Smith, consciousness confers meaning and indeed the objects of consciousness are, strictly speaking, other mental states.9 In particular, necessity and universality are ‘imposed by the mind’.10 Vaihinger insists that the mind projects order on the world.11 In contrast, Pippin is committed to the view that, for Kant, extra-mental reality is not simply the passive recipient of projections on the part of the subject. The world is given to us, admittedly in an indeterminate and indirect way. It is indeterminate because it is given only in sensation and indirect because sensations without concepts are blind.12 For Pippin, this conclusion is necessarily drawn from Kant’s dualism. Pippin is equally critical of Guyer’s impositionalist interpretation, which he sees as rooted in what he calls a literal interpretation of pure a priori synthesis. Guyer concludes that the latter is responsible for the subjectivism of Kant’s formalist position. Pippin’s reply is restricted to consideration of an article published in 1980, which sets the scene for Guyer’s systematic interpretation of Kant published as Kant and the Claims of Knowledge seven years later.13 Pippin argues that although Kant seems to commit himself to an independent synthesis prior to empirical synthesis, this is misleading.14 In contrast, Pippin reads Kant’s use of the a priori as an adverb.15 The point Pippin makes is that to talk of an a priori synthesis is not to talk of a synthesis that literally comes before experience, but rather of a synthesis that is independent of experience and yet serves as a ground for 11


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology our knowledge of it. Elsewhere Pippin further counters the challenge to Kant’s formalism saying that to assert that the unity of apperception is ‘formal’ means to assert merely that judgement, cognitive activity or experience are only possible if the contents of consciousness are brought to a unity.16 He also stresses that formal talk refers to the activity of mind and not to some special non-material object.17 Pippin agrees with Guyer that the way in which we ultimately understand what Kant means by a priori synthesis depends on how we understand the idea of ‘mind-imposed unity’.18 If we read the latter phrase literally then we are inevitably led to interpret a priori synthesis as an independent act occurring prior to empirical experience. But if we adopt Pippin’s adverbial reading, we need only interpret the a priori as the formal condition of any empirical experience and not as a distinctive act.19 We will see, however, that Pippin retains worries about the status of a priori formal intuition. Pippin argues that the impositionalist interpretation cannot be right because Kant does not simply claim that the categories of the understanding – the most pure concepts with which we operate in unifying any object of experience – establish how an object appears for us, but says that any object that appears for us must appear under certain formal conceptual conditions.20 This is the notion of ‘comprehensiveness’ and is the key to Pippin’s rejection of the impositionalist reading. Kant starts by trying to establish our understanding of the general structure of reality, but he also shows that this is the only way in which experience could be understood by human subjects. He thus establishes the necessity of the formal structure we in fact adopt. ‘This formal unity is not imposed; it just is the only type of unity we could understand in our experience.’21 Pippin thus distinguishes between a merely subjective imposition of the structure or form of reality and a comprehensive account that would establish the necessity of this formal structure for our experience. In the end Pippin’s judgement is that the comprehensiveness claim is not made good, thus undermining Kant’s commitment to empirical guidedness. But the fact that Kant aims at the latter goal counts as the major reason for Pippin’s opposition to readings like Guyer’s. Importantly, in his defence of Kant, Pippin contrasts ideas and concepts insisting that the latter are not subjectively imposed.22 However, if concepts are to qualify as empirically guided, then this must be discoverable in their empirical application. Even if a priori concepts are not subjective, this will count for little if empirical concepts are. The aim of empirical guidedness will fail, if empirical concepts turn out to 12


29/3/07

8:12 AM


The Centrality of the Problem of Formalism be instances of subjective projection and this is just what Pippin concludes, as I will discuss below. Pippin also disagrees with those interpreters for whom Kant’s major failing is his insistence on the role of subjective faculties. His argument here is principally addressed to Strawson’s critique of Kant’s idiom of the faculties. For Strawson, Kant’s supposed psychologism amounts to the claim that the mind imposes form on matter.23 While Pippin concurs that Kant’s position is ultimately subjectivist, he insists that it is a concern with the ‘activity of the subject’ and does not amount to psychologism.24 In making this argument he insists that faculty talk is best understood as a consideration of ‘forms of our conceiving’.25 However a suspicion of Kant’s faculty talk reemerges in his discussion of a priori intuition, especially with reference to ‘The Schematism of the Pure Concepts of Understanding’.26 Now, despite the distance at which Pippin stands to Guyer and to Strawson he nevertheless concludes that Kant falls into a subjectivism that should be ruled out by his commitment to empirical guidedness. How does this come about? The crux of Pippin’s critique of Kant is the latter’s formalism, which he believes counts as an ‘independent analysis of the “rules” for knowledge, prior to and divorced from experience of all kinds’.27 The aim of achieving comprehensiveness and therefore of avoiding impositionalism fails because Kant attempts to secure the legitimacy of claims to knowledge and their applicability to experience within the sphere of subjectivity alone. His objective of comprehensiveness, that is, of showing that knowledge is not simply for us but reveals the necessary conditions of all possible objects of experience, depends on achieving formal comprehensiveness as we have seen.28 Pippin does not immediately conclude that the formality of this move entails impositionalism. Indeed, on the contrary, he says that the formal status of the comprehensiveness claim means ‘not just that we impose’ order, but that this is the only possible ordering.29 However the distinction between subjective and objective ordering that this claim to comprehensiveness implies is not delivered: ‘Just how are we to understand this notion of a subjective unity of consciousness “depending” on, being “derivable” from, or being “grounded” in an objective unity?’30 Pippin’s adverbial defence of Kant’s talk of a priori synthesis is, as we have already seen, that it amounts to the claim that there are certain epistemic conditions for experience. This, he holds, should not be read as evidence of either impositionalism or psychologism, but rather reveals the formal, methodological or meta-level status of 13


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology Kant’s project.31 But the solution Pippin offers us now turns out to be problematic: how do we get from a comprehensive system of the subjective conditions of experience to securing the application of those epistemic conditions in empirical experience? As he puts it in an earlier pithy statement ‘If that formal idealism is to be successful, we must be able to understand its connection with this “material” or empirical realism . . .’.32 Elsewhere he says that if Kant is to avoid constructivism, by which he means subjectivism or impositionalism, he needs a ‘less metaphorical interpretation’ of the relation between the forms of experience and ‘empirically apprehended “matter” ’.33 The epistemic solution has reintroduced subjectivism by the back door. Pippin believes that Kant’s theory of judgement could have provided a solution to the problem. In the ‘Schematism’ Kant tries to escape from a strictly formal level of analysis, but fails due to his continuing to operate at an a priori level.34 It is his reliance on ‘his even more obscure theory of pure intuition (which itself stresses all over again a more constructivist theory of phenomenal unity)’ that undermines the potential for eventually achieving an adequate account of empirical guidedness through the strategy of comprehensiveness.35 Thus an attempt at showing how the epistemic subjective forms of experience genuinely grasp material reality, a problem that was promised specific attention in the ‘Schematism’, falls back into subjectivism by relying on an unexplained claim that our subjective power of intuition is capable of constructing reality. Pippin believes that the ‘Schematism’ chapter is further flawed insofar as it reveals a ‘serious instability in the understanding/sensibility distinction so very essential to the Critique’.36 While, as I have already mentioned, formalism does not necessarily entail impositionalism in Pippin’s eyes and indeed signals Kant’s intent of providing an alternative to the latter, a priori formalism is finally the problem. Thus Pippin says that for Kant ‘the homogeneity of the manifold is just due somehow to the demand of thought’.37 In other words it would appear in the end that the charge of impositionalism is correct. Combination of sensory input comes from thought alone and there is no effective empirical guidedness from the side of sense. He then immediately goes on to say: ‘And, if transcendental philosophy is to remain formal and a priori, it is hard to see how the situation could be otherwise.’38 A formalism that was not a priori might avoid the fate of impositionalism, but Kant’s version cannot do so on Pippin’s reckoning: 14


29/3/07

8:12 AM


The Centrality of the Problem of Formalism Once Kant argued for the ‘spontaneous’ nature of the understanding, and the indeterminate nature of the material of sensibility, we naturally wanted to know ‘by what right’ the understanding can so spontaneously, a priori, and with certainty and authority of necessity, legislate, and we are here to receive Kant’s unique answer: that the ‘ground’ or basis for this legislation rests in the requirements for a unified, self-identical subject.39

Pippin adds that the unity of the subject must be understood formally.40 Formalism ultimately situates the claimed legitimacy of the categories in subjectivity and, more precisely, in the unified selfidentical subject.41 It would appear that the distinction between concepts, which are not subjective and ideas, which are, has collapsed.42 Empirical guidedness turns out to be a straw man, insofar as it only arises in accordance with formal structures that we introduce a priori. One of the most important questions that Pippin raises is that of the legitimation or ground of empirical concepts. This point is central to his assessment of Kant’s success in establishing empirical guidedness.43 If empirical concepts cannot be shown to arise at least in part from conditions in extra-mental reality, then their fate is no better than that of pure a priori concepts. We have already seen that Pippin distinguishes concepts from ideas on the grounds that the former are not subjectively imposed while ideas are. However we have also seen that, on Pippin’s reading, while Kant intended to avoid the conclusion that concepts impose form on matter, he finally did not succeed in this aim. Thus, concepts can, in the end, do nothing other than subjectively impose a priori form on the external world. The remaining question is whether empirical concepts have a ground independent of pure concepts. Pippin considers various options; most importantly, for our discussion, that reflective judgement provides the ground for empirical concepts. In the Critique of Judgement Kant introduces a distinction between determining judgement that subsumes a particular instance or intuition under a universal rule, principle or law and reflective judgement that seeks a rule or concept for a given particular.44 Kant argues in the Introductions to the third Critique that reflective judgement provides the principle necessary for the systematic employment of empirical judgement and, arguably, also for empirical concepts.45 But the problem, as Pippin sees it, is that reflective judgement is a development of the regulative use of ideas in the Critique of Pure Reason. As such, reflective judgement, like regulative ideas, is projected by the mind onto nature. Pippin thus rejects the possibility that reflective judgement might be the sought-for autonomous ground of 15


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology empirical concepts, as it is simply too subjective to render it capable of securing the necessary reference to objectivity. ‘While the most general explanation for our association of markers into concepts, species, and so forth, it [reflective judgement] is also the furthest removed from any connection with any “objective” or empirically based origin.’46 Thus, Kant’s account of reflective judgement could, or even should, have given us the means for understanding how empirical concepts (and, we might add, perhaps even indirectly pure a priori ones) are capable of being empirically guided. This, however, is impossible because reflective judgement is a subjective projection. Both in his book and in a roughly contemporary article, Pippin does not argue for the premise that secures this conclusion, citing only Buchdahl in defence of it.47 In order to show how far Pippin’s final verdict lies from his initial defence of Kant, I must cite him again at some length: In sum, however much Kant intends to avoid the ‘imposition’ interpretation, it is still the case that, to use his own favourite metaphor for his formalism, reason legislates to nature, or even commands to nature, so formally, so independently of any ‘material’ reflection about nature, that the very possibility of ‘obedience’ to such laws, or the role of reason’s ‘subjects’ in formulating them seem considerations much too abstractly excluded. It is not incidental, I would suggest, that something of the same problem can be said to emerge in Kant’s moral, aesthetic, and teleological formalism.48

So while Pippin does not state that Kant’s position is impositionalist, it is clear that, in Pippin’s view, Kant ultimately falls into this trap. Pippin goes on to say that Kant would have needed ‘a more complicated account of the relation between form and a perhaps metaphysical “content” than his transcendental methodology allows . . .’.49 Pippin’s own alternative, he admits, is only glimpsed at the margins of his critique of Kant. The principal elements of his own position are a comprehensiveness that does not fall back into subjectivism;50 an account of subjectivity that is not indeterminate;51 and a theory that looks for its confirmation within actuality.52 These features suggest a broadly Hegelian programme, especially when combined with Pippin’s contrast between idealism and formalism. Pippin admits the limitations of his own alternative to Kant: Admittedly, of course, the idea of ‘breaking down’ Kant’s sensibility/ understanding dualism, or ‘opening the door’ to a fuller explanation of the 16


29/3/07

8:12 AM


The Centrality of the Problem of Formalism substantive or metaphysical foundation for pure intuition, concept, schema, and idea, or even just recognizing the necessity for a substantive account of the subject which thinks rather than just the rules of its thinking, are all fairly abstract conclusions to draw from the problems presented here.53

I will answer Pippin’s critique by supplying an alternative account of formalism. We will then be in a position to recognise that Kant is closer to Pippin than Pippin’s interpretation of Kant would suggest. I will argue that Kant’s conception of a formal system does not necessarily fall back into subjectivism and that his account of the ‘activity’ of the subject is less rigid than Pippin assumes. I will show how form arises in anticipation of matter and this allows us to construe empirical guidedness as more than an empty promise. I will not, however, argue that Kant offered a more substantive metaphysical foundation for the thinking subject. That would be to go beyond formal idealism towards the absolute idealism of Hegel. The account of the subject offered here will, rather, situate the formal framework for thought within the empirical world.

II Guyer’s critique of impositionalism Guyer believes that he can uncover within Kant’s writings an alternative to his official position. Guyer has great sympathy for this unofficial position, which he believes counts as a transcendental theory of experience. It is to be found in unpublished manuscripts and in a nascent form in the ‘Refutation of Idealism’ in the Critique of Pure Reason.54 While the unofficial position avoids metaphysical dogmatism, the official one does not and counts as impositionalist. Guyer’s extensive analysis of Kant’s epistemological project appeared in his book Kant and the Claims of Knowledge.55 Its core distinction was established in his earlier article to which, as we saw in the last section, Pippin responded.56 Central to Guyer’s reading of Kant’s epistemology is a distinction between two orders of necessity. He expresses the first as follows: a conditional necessity would arise were Kant claiming that if I experience an object, this requires that I am in fact aware of regularity among my representations of it.57 All this necessitates is, if I am to experience an object, it must in fact be of such a form as to be accessible to my means of grasping it cognitively.58 In contrast, absolute necessity would be required if experience of an object entails awareness of a necessary regularity among my representations of it.59 The absolute necessity of this condition can only be ensured if the mind 17


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology imposes this form.60 This is contrasted to what Guyer considers a distinctive metaphysical view of the relation between mind and reality where the conditions of the possibility of experience restrict rather than produce what can be experienced.61 Guyer reassesses the so-called overlooked alternative discussed in the nineteenth century by Adolf Trendelenburg: Kant assumes that space and time must be features either of our representations or of objects but not both, rather than that space and time may be either properties of objects or both necessary constraints on our perception of objects and genuine features of the objects we do succeed in perceiving.62

Guyer argues that, from Kant’s perspective, space and time could not be features of both subjectivity and of objects, because in the latter case they would not deliver the absolute necessity that is required. Only if space and time are subjective can they give rise to certain knowledge of appearances in space and time. Thus, the alternative was not so much overlooked as excluded by Kant.63 I will now sketch Guyer’s argument as presented in his earlier article before examining in more detail some of its principal claims.64 He contends that Kant is committed to an a priori synthesis of objects in all cases of knowledge, not just in mathematics.65 This a priori synthesis amounts to constitution, insofar as it imposes order on the manifold of empirical intuition.66 Anticipating attempts to interpret the priority of synthesis as the claim that all empirical manifolds are merely constrained by certain a priori conditions, Guyer identifies further passages to support his reading that a pure a priori synthesis precedes empirical syntheses. If empirical syntheses were simply constrained by a priori rules, then we would have the conditional necessity of Guyer’s preferred transcendental theory of experience. This would mean that our minds were restricted to a certain range of accessible objects. In contrast to this, he finds that Kant is committed to there being a prior pure synthesis that produces its own objects by imposing form upon matter. This counts as impositionalism and is a consequence of attributing certainty to the transcendental unity of apperception. He goes on to argue that Kant’s thesis that we are certain about the unity of apperception can only be construed as a synthetic and empirical claim, not analytical and a priori as Kant suggests, and, finally, that it rests on a conflation of consciousness and self-consciousness.67 Guyer’s alternative strategy is to argue that Kant could have proposed a transcendental, but not a priori formal theory. 18


29/3/07

8:12 AM


The Centrality of the Problem of Formalism Despite my disagreement with Guyer’s diagnosis of the identity of the ‘official’ Kant, my interpretation shares some features with Guyer’s unofficial transcendental theory of experience, being one in which mind is responsive to an external world not of its own making. In this regard, I would agree with Guyer that there is no necessary conflict between formal idealism and a version of realism. The difference between our positions is that I do not believe we have to divest Kant of transcendental idealism and most of his published works in order to be able to establish this. In later chapters I will show that there is a way of reading the a priori and formalism that illuminates the relationship between mind and object, showing their necessary relation without attributing impositionalism to Kant. Guyer admits that there are passages that could be interpreted according to a more modest interpretation of the activity of the mind. These would be in line with what in his book he calls conditional necessity. In his article he describes this tendency thus: ‘the purely hypothetical and analytic assertion that if the ascription of a representation to an identical self is possible, then it must also be possible to connect that representation to others ascribed to that self by a certain kind of synthesis’.68 But there are other passages that he believes cannot be so interpreted. These commit Kant to the absolute necessity of transcendental apperception and amount to the ‘thoroughly existential and synthetic claim that all of one’s representations can be ascribed to oneself’.69 Only this position requires a theory of imposition. Is there really a clear distinction between these two sorts of passages? Guyer supplies three examples of passages that, he concedes, commit Kant only to the claim that a priori syntheses bear conditional necessity. The first example Guyer says is concerned with a priori synthesis of the pure intuitions of mathematics, which he concedes implies no imposition on experiential objects as such.70 Meanwhile, the second and third examples are concerned with experience yet do not commit Kant to an imposition of pure synthesis. The second passage Guyer cites claims that appearances ‘must stand under a priori rules of their synthetic unity’,71 while in the third Kant says that appearances ‘are subjected to a priori conditions, with which their synthesis . . . must thoroughly accord’.72 Guyer is ready to concede that such passages do not suggest that there is a pure synthesis prior to empirical syntheses, but rather that certain a priori constraints govern the latter. However he insists that there is plenty of evidence for the stronger claim for absolute rather than conditional necessity. He begins with 19


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology the synthesis of reproduction that Kant says counts as one of the ‘transcendental actions of the mind’ and is ‘grounded prior to all experience’.73 Guyer then refers to another passage from the first edition of the ‘Transcendental Deduction’, or A ‘Deduction’ as opposed to the second edition or B ‘Deduction’. Here Kant says that apperception is the source of all synthetic unity and that it presupposes a productive synthesis of imagination, which is ‘prior to apperception’ and is ‘the ground of all cognition’.74 Guyer sees this as compelling evidence for Kant’s commitment to a pure synthesis prior to any empirical synthesis. Finally Guyer turns to a passage from the B ‘Deduction’ where we are told that the synthetic unity of the manifold ‘precedes a priori all my determinate thought’.75 He does not comment on the complex relation between the faculties that emerges in the second of these passages. He is intent simply on insisting that Kant is committed to certain pure operations of the mind that are prior to all experience.76 What exactly does Guyer mean by prior in these cases? He clearly does not mean temporally prior. Nor can he mean apperception is simply logically prior, for if he did there would be no reason to make the distinction between passages that express a restriction on our mental activity and those that supposedly express an imposition of form on matter. Logical priority would only entail that apperception was a necessary condition for the possibility of experience of objects, as we have already seen in our discussion of Pippin’s adverbial reading. This would be quite consistent with Guyer’s preferred conditional necessity and would not entail that apperception was the sufficient condition of that experience, insofar as it imposed form on appearances. For Guyer the priority at issue is synonymous with absolute necessity. This is the real crux of the matter and lies beneath what Pippin calls Guyer’s literal reading of the priority claim. A pure synthesis is prior in the sense that apperception produces its own objects by imposing forms on nature. Thus mind produces forms that are not only necessary, but also sufficient conditions for the order of nature. Is Guyer’s literal reading of the priority of pure synthesis proven by the textual references he supplies, or is it rather that his independent conviction that Kant is committed to impositionalism leads him to read these passages as literally as possible? The latter view is encouraged by the next move in Guyer’s argument. He says that even were the passages initially discussed in his article inconclusive, it cannot be doubted that Kant is committed to the view that the mind 20


29/3/07

8:12 AM


The Centrality of the Problem of Formalism constitutes ‘the objective affinity of nature as underlying all particular empirical investigation’.77 After all, Kant has said that it is we who introduce order and regularity into appearances or nature.78 How, asks Kant, ‘should we be able to get such a unity started a priori, if subjective grounds of such unity were not contained in the original sources of cognition in our minds?’79 Admittedly, Kant’s account at this stage of the first edition ‘Deduction’ raises some problems. How, crucially, is he sure that the unity achieved will be an ‘a priori certain unity of the connection of appearances’?80 Moreover, the statement that this certainty is somehow contained in the original sources of cognition seems to support Guyer’s belief that Kant is committed to an absolute necessity imposed transcendentally by the faculties on appearances and ultimately on matter. Guyer thinks that the evidence is incontrovertible and that we must read Kant as an impositionalist. This leads him to miss a possibility latent in the passages just mentioned. Might there not be a tacit distinction between the subjective and objective grounds of the same unity? If this suggestion is taken up, then Kant may only be committed to the view that the subjective conditions of cognition supply necessary, though insufficient, conditions of the formal structure of nature. The subjective grounds would then only ‘get things started’ [auf die Bahn bringen]: they initiate unity, but they do not achieve it.81 The grounds of necessary unity are contained in our minds only in an initial fashion. The original German offers some potential support for my suggestion. The crucial two sentences read: For this unity of nature should [soll] be a necessary, that is an a priori certain unity of the connection of appearances. How should we be able to get an a priori synthetic unity started [auf die Bahn bringen literally means ‘to bring onto the track’82], if there were not subjective grounds of such unity a priori in our original cognitive powers and if these subjective powers were not objectively valid insofar as they are the grounds of the possibility of knowing any object whatsoever in experience.83

The modal verb sollen suggests a task, even an epistemic duty. Kemp Smith translates ‘soll’ as ‘has to be’. This expression also bears a future connotation, although it can be mistaken for a simple statement of necessity. The modal verb soll could also be translated as ‘is to be’, resulting in the claim that the unity of appearances ‘is to be’ constructed as the result of a project. Thus it would not be contained already, other than in principle, in the initiating conditions, that is, in 21


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology the power of synthesis. And although translating ‘auf die Bahn bringen’ by ‘to establish’, as Kemp Smith does, is accurate, this English expression eliminates the future nuance of the German phrase, which could suggest a task that is only just beginning. Focusing on the future orientation in Kant’s expression allows us to reassess the status of the immediately preceding claim that ‘the order and regularity in the appearances, which we entitle nature,we ourselves introduce [bringen wir selbst hinein]’.84 While we introduce order, we may not construct or impose it. So when Kant says that ‘[t]he synthesis of the manifold through pure imagination, the unity of all representations in relation to original apperception, precede[s] all empirical knowledge [gehen aller empirischen Erkenntnis vor]’, the priority in question need only be the initiating condition of necessity.85 While the hints afforded by the text support, but are not conclusive as to the correctness of my interpretative suggestion, they should at least make us pause for thought before concluding that the priority that Kant seeks to establish entails that the mind imposes order on empirical matter. And if this is so, these passages need not commit Kant to impositionalism, although they certainly aim to secure more than a merely conditional or factual coincidence between mind and objects. Guyer’s suspicion is that the only way such an a priori certain unity could be achieved would be if the mind were capable of imposing its own necessary structures on objects. At this stage I will simply sketch a possible alternative solution, which I will develop in greater detail in later chapters. The key here is to distinguish between two possible levels at which nature may be unified. When Kant says that there should be an a priori certain unity of appearances, he may mean that there must be an a priori framework that guarantees the possibility of connection among empirical appearances. While the framework is certain at the a priori level, this does not entail that there is such certainty with regard to empirical experience. We will see later that in both editions of the ‘Deduction’ Kant insists that the categories do not fully determine empirical laws and thus reveals that there is a gap between the a priori framework and empirical nature.86 If I am right, Kant is only committed to certainty at the transcendental level as a framework for the possibility of experience, while at the empirical level, certainty is an ongoing project rather than a fait accompli. Are there significant developments in the arguments Guyer presents in his later book? At the outset of his later account Guyer fails to mention that Kant’s supposed problem arises from a confusion 22


29/3/07

8:12 AM


The Centrality of the Problem of Formalism between consciousness and self-consciousness, but otherwise his argument is the same.87 [Kant] simply assumes that we know certain propositions as universal and necessary truths; he then argues that such claims to knowledge of necessary truth can be explained only by our antecedent possession of certain conceptions and capacities which we must, in turn, be able to impose upon a reality which does not itself, even contingently, conform to these conditions . . . 88

Toward the end of the book he reconstructs the ‘error’ in three stages. Once again Kant is found to have presupposed the validity of a claim to knowledge of a universal and necessary truth. Second, we are told that ordinary experience can never justify such a claim, which must therefore have an a priori basis. Third, Kant concludes that only certain forms of intuition, concepts or principles of judgement could satisfy this requirement.89 This is a more nuanced account than was provided in the earlier article, but the nub is still the same: namely, Kant assumes that we are capable of a priori certainty about the unity of appearances and the only way this can be explained is to posit an a priori capacity or synthesis that produces such certainty by projecting its unity onto the manifold in apprehension. This claim is nothing other than the impositionalist thesis. In Kant and the Claims of Knowledge, Guyer declines to answer critics of his earlier article and presents instead a more detailed account of Kant’s argument.90 Following an analysis of Kant’s arguments in which he attempts to deduce the validity of the categories starting from concepts of the object or of judgement, he now returns to the argumentative territory of his earlier account in analysing those arguments that start from the self: [H]e discovered what he thought was a simple but powerful argument from what he took to be our a priori knowledge of the necessary unity or identity of the self throughout all of our experiences to the need for an a priori synthesis of the manifold of intuition, conducted in accord with a priori concepts.91

Thus the certainty a subject has in its identity throughout experience entails that it must be able to achieve an equally certain synthesis of the manifold of apprehension.92 It does so, as we have already seen in his earlier article, by imposing the forms of intuition onto that manifold in accordance with the categories. Guyer goes on to say that this ‘supplement[s] his [i.e. Kant’s] in any case considerable confidence in the deduction of the categories from 23


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology the conceptions of object and judgement’.93 So it sounds as if this ‘powerful’ argument is not an alternative to those other strategies, but rather complementary to them. Yet in what follows it becomes clear that, just as Guyer held in his earlier article, what he considers to be a new argument is, in his opinion, Kant’s primary one. The linchpin of Guyer’s argument is once again Kant’s claim that ‘the understanding is itself the source of the laws of nature’.94 This is what convinces Guyer that Kant’s position must be that ‘the mind can impose an “affinity” on all appearances’ for only this secures Kant’s conclusion that they must be ‘associable in themselves and subjected to universal rules of thoroughgoing connection’.95 This, he thinks, is ‘the only possible explanation of the premise of this new argument, the assumption that unity of apperception is a priori certain or guaranteed to obtain under all possible circumstances’.96 Thus only if there is an a priori synthesis that unifies all experiences, can the unity of apperception in which we are already confident be validated. The unity apperception introduces into experience is, moreover, an absolute necessity. This is so because our minds are capable of synthesising or determining all appearances that arise for us. I have argued that Guyer’s evidence for claiming that Kant is committed to the absolute necessity of apperception is not conclusive. He certainly has evidence that Kant holds that the transcendental unity of apperception is a necessary condition of any empirical knowledge, but does he have the further proof that the latter arises from a pure a priori synthesis that imposes form on matter? In 1987 Guyer’s evidence is similar to that presented in his earlier article. He still insists that it is Kant’s ‘unequivocal view that we can impose our categories on any data of sensibility whatever’.97 In other words, we simply must read Kant’s statements about a priori synthesis literally, because the passages and the argument would not bear any other interpretation. For Guyer, this is the crux of transcendental idealism, for it is only the transcendental ideality of space – that is, its status as a mental form – that ensures the transcendental affinity of order among objects qua appearances necessary for the certainty to which he believes Kant is committed.98 Interestingly, in 1987 Guyer does not oppose what he calls formal idealism, which he identifies with Kant’s mature position.99 The latter is ‘compatible with a realistic interpretation of the intended conclusion of the refutation of idealism’.100 What he objects to is the interpretation of form as imposed on objects of experience. This interpretation entails that transcendental idealism counts not as a doctrine 24


29/3/07

8:12 AM


The Centrality of the Problem of Formalism of epistemological modesty pace Henry Allison, but rather as a metaphysical dogma.101 Hence Guyer is opposed to the formalism correlative to transcendental idealism, which he takes to be the thesis that form is imposed on objects by the mind. He remarks that Kant’s mature position commits him only to the ideal status of the forms and not of the existence of the objects of intuition.102 I hope to have shown in a provisional way that even in the admittedly difficult territory of the ‘Transcendental Deduction’ there is no conclusive reason why Kant should be consigned to impositionalism. Objects have an ideal dimension insofar as they are formed, but this does not necessarily entail that form is imposed on matter. I have suggested that Kant’s formalism can be seen as the initiation of order through the synthetic activity of our subjective cognitive faculties. In later chapters I build on this suggestion.

III Henrich’s Cartesian Certainty – A Basis for Impositionalism? While Dieter Henrich, one of Kant’s most influential contemporary German interpreters, ultimately aims to defend Kant against the charge of impositionalism, Kant’s critics have found encouragement in the interpretation he offers in Identity and Objectivity.103 Guyer, in particular, relies on Henrich’s reading, something I will establish by an examination of a review written by Guyer just before his publication of the article considered in the previous section of this chapter.104 Thus, while it may seem odd to include Henrich among Kant’s critics, it is helpful to include a discussion of his highly reconstructive reading at this point. Guyer’s view that for Kant apperception is apodeictically certain reveals his debt to Henrich, for whom apperception counts as a sort of Cartesian awareness of what he calls the numerical identity of the self.105 Against this Allison argues that knowledge requires only the awareness of ‘the “fact” that this identity must be presupposed as a necessary condition of knowledge’, in other words, it requires only ‘the necessity of a possibility’, that is, the possibility of being aware of numerical identity ‘but not its actuality or necessity’.106 As Allison also puts it: ‘The consciousness of this act . . . is . . . the consciousness of the form of thinking’.107 Importantly he says that Kant’s account is ‘a formal model or schema for the analysis of the understanding and its “logical” activities’ and concludes: ‘As such, it is neither a bit of introspective psychology nor an idealistic ontological 25


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology thesis concerning the manner in which the mind “creates” the phenomenal world by imposing its forms upon the given sensible data.’108 Despite the apparent difference between Allison’s formal reading and Henrich’s Cartesian reading of apperception, we will see later in this section that the contrast is quite hard to draw.109 Henrich argues that, for Kant, the certainty that we have in the numerical identity of the self depends on our ability to employ certain forms of judgement. Guyer agrees with this entailment and the consequent denial of an immediate certainty of self in detachment from our judgemental capacity. However, while Henrich argues that Kant is not, or at least should not be, committed to ‘constitution-theoretic talk’ – that is, to impositionalism – Guyer argues that the latter is a necessary corollary of Henrich’s correct diagnosis of the basic premise of Kant’s deduction, namely, certainty about numerical identity: In fact, we could be certain a priori that we will be conscious of our continuing identity so long as we are conscious of any representations at all only if we can always process or force our representations to conform to the conditions of our consciousness of identity, which is, I take it, the gist of any theory of constitution.110

So whereas Henrich believes that Kant only slips into impositionalism because of an unnecessary confusion, Guyer insists that it is the unavoidable corollary of the certainty of apperception.111 It is because the numerical identity of the self can be established only if it is discovered in all appearances, that the former as the transcendental identity of apperception must impose unity on the latter. Therefore impositionalism leads on directly from Henrich’s insistence on the certainty of apperception. Significantly, Guyer’s core distinction is first aired in response to Henrich’s interpretation. The theory of constitution arises from what Guyer calls Henrich’s key argument, that is, ‘that we know a priori that we will be conscious of our numerical identity in whatever representations we will have’ in contrast to the weaker claim that ‘we can be conscious of our identity in any given sequence of representations only if they conform to the necessary conditions for our consciousness of identity’.112 This is the by now familiar distinction between absolute and conditional necessity.113 As we saw, Guyer concludes that Kant’s impositionalism arises from a confusion between consciousness and self-consciousness.114 Henrich’s insistence that the legitimacy of the categories requires a priori knowledge of the self – and not just, as Allison argues, 26


29/3/07

8:12 AM


The Centrality of the Problem of Formalism consciousness of the form of thinking – once again is a point d’appui for Guyer’s critique.115 It appears that Henrich has set the stakes too high. He concludes that the presupposition of Kant’s thought is ‘wholly original’, ‘highly productive’ and ‘exceptionally convincing’ and sums it up thus: ‘one sees in the notion of the ego [im Ich-Gedanken] a consciousness which demonstrates the highest degree of certainty and which is, so to speak, a priori superior to everything.’116 It is not possible to enter into a detailed reconstruction and assessment of Henrich’s argument here. Nevertheless we must consider the way in which Henrich rejects the impositionalist conclusion Guyer draws from his Cartesian premise. Henrich’s rejection of impositionalism emerges principally from the important distinction he draws between the identity of selfconsciousness and the identity of the act.117 In Henrich’s view, Kant’s tendency to focus on the act serves as an ‘unfavourable influence’ on the development of his proof.118 Henrich interprets this development in Kant’s account as entailing that the act of the mind ‘subjects appearances to rules’ and concludes that this commits Kant to a thesis of ‘real constitution’ rather than simply logical derivation.119 Henrich is referring to the passage from the first edition of the ‘Deduction’ where Kant famously states: For the mind could never think its identity in the manifoldness of its representations, and indeed think this identity a priori, if it did not have before its eyes the identity of its act which subjects all synthesis of apprehension (which is empirical) to a transcendental unity and thereby renders possible its connection according to a priori rules.120

Henrich comments that this implies a distinction between particular acts of synthesis and the act of synthesis itself that generates the a priori rules necessary for any synthesis whatsoever.121 His concern is that Kant suggests that we can move directly from the identity of the subject to the act of synthesis and this in turn suggests that the self ‘subjects appearances to rules’.122 I am not convinced that it is necessary to make the distinction Henrich suggests, as I read this passage as a not yet well-formulated expression of Kant’s view that the transcendental synthesis of apperception entails some awareness of its own activity. If so, Kant may not be guilty of falling into impositionalism as charged. In my view the real problem lies not in Kant’s focus on the act of apperception, but rather in the speed with which he thinks he can 27


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology move from a priori synthesis to empirical determination. Kant is not clear about the level at which his argument operates. This is evident not only in the sentence Henrich holds up for criticism, where he already talks about empirical apprehension, but in the preceding one where Kant claims that the ‘Deduction’ achieves the determination of objects, without making clear whether he is talking about empirical objects or an object ‘in general’, that is, the conditions of possibility for objectivity.123 There is, however, a way of decelerating his claim. We can take it to imply that the synthesis of apprehension ultimately falls within the general framework of transcendental unity, but not that the latter imposes form on empirical objects, there being further conditions of empirical unity yet to be established. Nevertheless, it seems probable that at the time of writing the first edition ‘Deduction’ Kant thought he could conclude his argument much more quickly than turned out to be the case.124 Henrich responds to the tendency in Kant’s position associated with the talk of an ‘act’ and which we can identify as impositionalist, as follows: Now the subject is certainly the agent [Akteur] of a number of activities. It can invariably reflect upon itself and initiate syntheses. . . . it still does not always follow that it is in the subject and its acts that the ground must be sought of all the conditions without which the concept of the subject cannot be thought and without which the subject could have no knowledge of itself.125

It is crucial here that not only self-knowledge, which would imply empirical intuition, but also the very thought of the concept of the subject, the ‘I’, may require conditions beyond the subject. In an earlier passage Henrich insists that it is not the case that the subject autonomously guarantees its own unity through its activity of synthesis.126 This is because ‘the continuance of its activity’ is not dependent on itself.127 What other grounds are there for the conditions of self-consciousness and self-knowledge? One promising suggestion may be found in a yet earlier passage where he explains why Kant’s commitment to the subject’s dependence on what is given to it in intuition, results in the fact that he could not have adopted Leibniz’s strict conception of identity.128 Drawing out Henrich’s general position in this way allows us to clarify his distinction between real constitution and logical derivation. The first of these would amount to the position that appearances are directly extracted from subjectivity. The second would coincide with the more complex thought that while the 28


29/3/07

8:12 AM


The Centrality of the Problem of Formalism transcendental unity of apperception is the ground of the connection between subject and world, self-consciousness necessarily stands in relation to a given and ultimately to objects in the world. What Henrich means by logical derivation is clarified by the distinction he draws between numerical identity and the simplicity of the subject. The deduction of the categories – that is, the proof of their legitimacy for all appearances – only succeeds through a ‘complex proof-procedure . . . which must make reference first to the identity of the subject, but then, necessarily to its simplicity as well’.129 Henrich’s view is that Kant tends to concentrate on simplicity at the expense of identity. This is what leads Kant to move so quickly from ‘the subject as a principle of identity to the assumption of a priori rules which apply to all apprehensions’.130 A more mediated relation between subject and rules is required if Kant is not to fall into a theory of ‘real constitution’. This supports my suggestion that Henrich is committed to the subject’s numerical identity arising from a mediate relation to objectivity. For, surely, the rules are ones that put the subject in a necessary relation to the possibility of experiencing objects. However, Henrich’s analysis remains oriented towards subjectivity. Moreover, as we will see, he undermines his commitment to objectivity, which I do not doubt he holds, by insisting that the rules that secure numerical identity are pre-temporal. Henrich’s commitment to the link between apperception and objectivity thus remains problematic. These passages reveal that, for Henrich, impositionalism is avoided principally insofar as Kant establishes complexity within subjectivity. Self-consciousness is certain strictly in the sense that it is underived from any other consciousness.131 This entails only that selfconsciousness is the ineliminable starting point or origin of all knowledge, but not that it contains the sufficient conditions of the latter. The crux of Guyer’s dissatisfaction with Henrich’s account and, at the same time, its productivity for his own attribution of impositionalism to Kant, lies at the other pole of the subject–object relation. This is the implication that we can know a priori that all our representations will be synthesisable under rules that are the conditions of transition between different states of the subject.132 The problem, Guyer’s argument goes, is that this commits Kant to absolute necessity, that is, to the position that anything we experience is necessarily conditioned by rules that arise from self-consciousness. But the question here is surely, how maximal or minimal is the condition? Is the claim that the categories provide the broad framework within which 29


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology experience is possible, or that they determine the shape or form of any empirical experience whatsoever? Given that Guyer argues that Henrich’s position entails impositionalism, he clearly believes that Henrich is committed to the latter of these two options. In attributing a commitment to absolute necessity to Kant as interpreted by Henrich, Guyer represents Henrich’s position as entailing that ‘all of our representations can be known a priori to be subject to synthesis into a “nature” or world of objects themselves completely interrelated by these rules’.133 In the first passage referred to by Guyer, Henrich states that all objects are subject to universal laws insofar as they belong to nature, while in the second he claims ‘It is in relation to this manifoldness of fundamental forms of judgement that the collective unity of objects in One Nature must be thought’.134 In the latter passage Henrich is explicitly concerned with the question of objectivity, however there is no reason to think that he is talking about empirical objectivity. The nature he has in view in both passages is surely that of the most general order of things in relation to the transcendental unity of apperception. Nature at this level is not yet equivalent to the system of empirical nature. This leaves open the possibility that while nature in general is subject to a priori synthesis, empirical nature in its particular detail is not. Were Henrich speaking of nature at the empirical level, this would follow, but we need not conclude that he is doing so when he has made no mention of the empirical. My suggestion gains additional plausibility from Henrich’s insistence that the subject is the underivable starting place for any knowledge, not its sole condition and from his claim that it is not necessary to read Kant as committed to impositionalism. Nevertheless, a statement of the limitations of the argument of the ‘Transcendental Deduction’ is sorely missing in Henrich’s account. My suggested defence of his account in the face of Guyer’s critique entails attributing to him the position that the ‘Transcendental Deduction’ is only the beginning of the project of legitimating the categories and will be supplemented by further arguments showing how they apply to empirical objects. His insistence on the ‘Deduction’ as the sole locus of legitimation for the categories tells against such an extension of his account. As Karl Ameriks reports an observation by Ralph Walker, Henrich’s reading of the ‘Deduction’ may establish nothing more than the necessary application of categories to ‘a blue qualia (sic) that is square’.135 And without an account of the broader epistemological project within which the ‘Deduction’ arises, Guyer’s interpretation is far from unreasonable. 30


29/3/07

8:12 AM


The Centrality of the Problem of Formalism But, what of Henrich’s relationship to formalism? The point at which Henrich most clearly adopts a formalist position comes in the course of a discussion of the way in which the mind ‘can ascertain its identity solely in view of a regulated synthesis’.136 Henrich favours moderate numerical identity in contrast to Leibniz’s strict identity. Moderate numerical identity allows for a change in the states of a thing.137 Henrich argues that it must be possible to know something about transitions taking place under the subject’s identity conditions ‘without being able, in this knowledge, to refer to any definite transitions in which the subject has found itself’.138 What Henrich means is that there is a formal knowledge of transitions or changes prior to any identification of change in the subject at the empirical level. The functions governed by the categories count as modes of transition which: ‘must be constant, because it is only in this way that they are suitable for rendering possible, independently of all experience and in unconditional universality, the knowledge of the subject’s identity, which is at all times possible.’139 Henrich concludes that knowledge of such modes of transition or functions is a necessary condition of the subject’s a priori awareness of itself as identical. Thus self-certainty will only be confirmed as apodeictic insofar as it can be guaranteed that it perseveres throughout such modes of transition. This process is not to be understood as empirical, but rather as prior to experience.140 Henrich next considers the objection that too much may have been derived from a ‘completely formal principle’ and that the mere thought of a sequence of thoughts in time would suffice.141 The objection under consideration is that the temporality of consciousness would be sufficient for establishing the numerical identity of the self. Henrich replies that the transitions he is analysing are prior to temporality and count as analogous to rules of logical derivation.142 It might appear that Henrich’s method is formalist in the particular sense that it is modelled on logic. Yet we have already seen that logical derivation implies a complex derivation of numerical identity as opposed to the claim that knowledge is simply entailed in the subject per se.143 Moreover, Henrich distances himself from ‘a theory of purely formal principles of knowledge (which can also be called a logical theory of experience)’.144 He insists on the need to combine the logical and the real trajectories of Kant’s argument, instead of choosing between them as, he claims, the latter’s commentators have done.145 It is only by combining both dimensions in Kant that the 31


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology categories can be grounded and, significantly, Henrich suggests that this will entail a formal element: Starting from the fundamental notion of the spontaneous and reflectively acting subject, one can ground the validity of necessary rules for all appearances only by means of the other basic notion of the formal and a priori principle of subject identity.146

There are therefore two basic notions: the real one of the consciousness of the ‘I think’, and the formal or logical one of the numerical identity of the subject over time. This reveals that Henrich’s position is a hybrid one, combining form with the particularity or reality of the Cartesian subject. But it is only numerical identity that secures the validity of the categories. Interestingly, this shows his commitment, once again, to the view that the capacity for self-reflection is not merely formal, as Allison would have it, but also real. This is but one of the many complexities of the position he adopts. The interest of this for my own account is the way in which his, admittedly underdeveloped, account of formalism challenges the prevalent equation between the latter and logicism or mere subjectivism. In a later article, Henrich distances himself from logical formalism such as the neo-Kantian project to establish ‘a merely formal property of thoughts which can itself never become an instance of consciousness’ in contrast to an actual thought of self-identity.147 The position he favours insists once again that, although formal, self-consciousness is a ‘real act’ of consciousness.148 But this is not to say that it is actual at all times: ‘it is always possible for self-consciousness to become actual, it does not have to be actual’.149 The form of consciousness is constant and I can be certain of it a priori. The fact that I achieve this level of self-consciousness is, however, not known a priori.150 It is now clear that Henrich’s Cartesian certainty is the formal, but not merely formal, possibility that I can be conscious of my identity in relation to any representation. How great a distance, then, is there between Allison’s account of formal possibility and Henrich’s Cartesian position? We have seen that in Henrich’s account of the apodeictic certainty of apperception, he insists that a Cartesian reflection must be possible on at least some occasions. Allison, we have seen, states that what is at issue is in no sense introspective, being only a form of thought. But it is arguable that self-reflection on the activity of thinking is not introspection. And now it seems that Henrich would be happy with Allison’s claim for the necessity of a possibility of such self-consciousness, for, as we have 32


29/3/07

8:12 AM


The Centrality of the Problem of Formalism just seen, the former holds that we are a priori certain only of the form of consciousness. Henrich insists that actual self-consciousness cannot remain a mere possibility and must be capable of actualisation. But does not Allison’s claim for the necessity of a possibility commit him to the same position? To say this is surely to say more than that only a formal awareness is possible and to commit to the actualisation of that form as a possibility. A fine distinction remains and it seems to be one of emphasis: while Henrich insists the actualisation of the possibility must be possible, Allison says only that the possibility and ‘not its actuality or necessity’ is necessary.151 The final distinction is not as clear-cut as Allison seems to think and is very difficult to pin down, but we can say that, for Henrich, consciousness is referenced to self-consciousness more explicitly than it is for Allison. An assessment of the structure of Henrich’s Cartesianism goes beyond the scope of this discussion. Nevertheless, we have seen that the formal moment in Henrich’s approach aims to establish a set of rules prior to experience and to time. In my opinion this is where the real problem lies. The formal Cartesian status of apperception may not in itself lead to impositionalism, but a-temporal a priori rules appear to do so. While the subject’s identity can only be established in relation to a range of possible judgements, this occurs prior to any actual application within experience. Now in a sense, from a Kantian position, this is uncontroversial. The categories are the formal framework for experience and as such are prior to any actual experience. But the question is: is this a set of, in principle, formal moves that will only be fully worked out in relation to experience, or are they fully formed prior to any application? Their being a priori leaves open both options, but in the second case the order in intuition is imposed. There are two ways in which we can develop the idea that the rules operate in principle. In the first case, the categories, although a-temporal, require temporal schematisation if their full objective validity is to be established.152 But this would mean that the forms as presented in the ‘Transcendental Deduction’ are only provisionally stated. Henrich’s insistence on the primacy of the ‘Deduction’ does not suggest he would wish to adopt this strategy. A stronger position would involve arguing that even apperception stands in some relation to temporality, which is fine-tuned in the ‘Schematism’ and ‘Principles’ chapters of Kant’s Critique of Pure Reason. Thus the categories, while a priori, are temporal at a formal level and are only the first stage of the determination of the rules for the unity of apprehension. It is clear that Henrich would not want to accept that apperception is in any sense 33


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology temporal, so this option is definitely not open to him. Once again, it would appear that the additional moves that Henrich could make in order to avoid impositionalism are not ones he could or would take. But is Henrich’s insistence on the pre-temporal status of the rules that express the possible modifications of numerical identity, in effect the categories, not wholly faithful to Kant? After all, is it not the case that, for Kant, apperception is distinct from temporality? It would not be possible to supply a thorough answer to this complicated question here, but I will provide a sketch developing the second interpretative option just mentioned. The transcendental unity of apperception is the form of thought and thus is peculiarly linked to the understanding. However, apperception is a synthetic unity and what it synthesises is the manifold in intuition. All intuition, without exception, is temporal. Thus apperception is the exercise of the understanding in relation to intuition. The initial condition of apperception lies on the intellectual side of dualism, which taken in abstraction from the activity of synthesis counts as a-temporal. But this initial condition initiates a synthesis that is necessarily temporal. The transcendental unity of apperception is the synthetic unity of the temporal achieved by the understanding. This synthesis operates at a formal level and thus entails no empirical time consciousness, but it counts as the form of temporality that necessarily precedes any particular temporal determination.153 We can conceive of a version of this account more closely modelled on Henrich’s interpretation. Self-certainty is only achieved in relation to a number of rules that govern appearances, including, ultimately, empirical ones. These rules are nothing other than the rules for synthesis of the manifold of intuition. Certainty attaching to apperception thus arises only in relation to a synthesis that necessarily has intuition as one of its terms. In effect, as in my original account of this version, apperception entails the exercise of understanding on intuition. The certainty of apperception cannot be established in detachment from this synthesis. It is neither Henrich’s intention, nor is it a necessary corollary of his Cartesian interpretation of Kant, that his account results in impositionalism. We have seen, however, that insufficient attention to the levels at which Kant is concerned with nature encourage interpretations such as Guyer’s. I have argued, nevertheless, that the real danger of impositionalism derives from Henrich’s insistence that experience is based on an a-temporal a priori set of rules. Apperception is thus set outside or at the limits of the spatio-temporal world. In a more recent 34


29/3/07

8:12 AM


The Centrality of the Problem of Formalism publication he goes so far as to suggest a reciprocal relation between subject and world: ‘But the source from which a world originates is equally dependent on that world.’154 Both mind and world are cooriginary, yet Henrich does not sufficiently develop the object pole of the subject–object relation. The result has been that his Cartesian position has been able to serve as a basis for impositionalist readings of Kant despite Henrich’s strong intentions to the contrary.

IV Strawson’s critique of transcendental psychology I now turn to consider the interpretation of Peter Strawson, whose Bounds of Sense (1966) has been so influential within the Englishspeaking world. The distinctive contribution of his interpretation is his diagnosis of faculty talk as characteristic of Kant’s formalism and as the source of the latter’s ‘impositionalism’. For Strawson, impositionalism and formalism are inextricably entwined. Kant’s epistemology rests on the distinction between a world of appearances and a world of things in themselves, while his account of our knowledge of the world of appearances depends on a metaphysics of mind, that is, a faculty theory. The latter is what Strawson calls ‘the idiom of the faculties’.155 According to Strawson, it is the second thesis that does all the argumentative work. It is because Kant insists that the world is only accessible to us through the apparatus of the faculties that he relies on a ‘two-worlds’ metaphysics. Strawson adopts an approach that has become common within Anglo-American approaches to Kant and other systematic thinkers. Some supposed kernel of critical philosophy is defended, while an array of other, apparently central, theses are rejected. A general suspicion of ‘systematicity’ supports this approach. Strawson intends to retain Kant’s commitment to ‘the principle of significance’. This is the maxim 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’.156 According to this principle, transcendent metaphysics is rejected because of its lack of empirical significance. It is therefore replaced by a transcendental ‘investigation of that limiting framework of ideas and principles the use and application of which are essential to empirical knowledge, and which are implicit in any coherent conception of experience which we can form’.157 Strawson claims that empiricism is in broad agreement with the principle of significance and with its rejection of transcendent metaphysics. 35


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology However Strawson believes that the search for an a priori framework leads Kant to straddle the boundary that he has shown to be impassable. Operating at an a priori level would entail exercising reason beyond the bounds of sense, thus entering into the transcendent metaphysical area Kant himself proscribed. While Strawson maintains that Kant does not slip back into a traditional form of transcendent metaphysics, he claims that a new and equally illicit metaphysical position emerges as ‘transcendental psychology’. If there is no transcendent beyond, there is instead a transcendent mind that imposes structures on nature. This runs very much against the grain of Strawson’s own empiricism: Is it not, after all, easy to read the very formulation of the programme – ‘the determination of the fundamental general structure of any conception of experience such as we can make intelligible to ourselves’ – in such a way as to suggest the Kantian-seeming thought that any necessary limits we find in such a conception are limits imposed by our capacities?158

Strawson’s impositionalist conclusion arises from failing to distinguish two ways in which subjectivity could enter into Kant’s argument. Subjective forms could count as necessary conditions of determination – this is in fact what Kant holds and it is a highly defensible position – or they could be the necessary and sufficient conditions of determination.159 I have suggested in my response to Guyer that Kant’s claim is that mind initiates order, but not that the latter is imposed. According to Strawson, Transcendental Idealism necessarily entails the stronger claim. He concludes that this is not just a possible reading of Kant, but the one he favours.160 Kant’s mistake, in Strawson’s eyes, is to make experience wholly dependent on the exercise of mental faculties that impose limits or forms. Thus this interpretation, which started by distinguishing the transcendent from the transcendental, ends up accusing Kant of muddying the divide. The transcendental faculties that impose form or limitation on matter have become the new transcendent and reopen a ‘two-worlds’ ontology despite the Copernican turn, which sought to eliminate all objects not capable of being experienced by us. As Strawson crucially remarks: ‘The doctrine is not merely that we can have no knowledge of a supersensible reality. The doctrine is that reality is supersensible and that we can have no knowledge of it.’161 Strawson sums up the supposed difficulty thus: The doctrine of transcendental idealism, and the associated picture of the receiving and ordering apparatus of the mind producing nature as we 36


29/3/07

8:12 AM


The Centrality of the Problem of Formalism know it out of the unknowable reality of things as they are in themselves, are undoubtedly the chief obstacles to a sympathetic understanding of the Critique.162

He proceeds to consider, briefly, that there may be other ‘weakened interpretations of these doctrines’. Maybe this talk of faculties is only an expository device or ‘a device for presenting an analytical or conceptual inquiry in a form readily grasped by the picture-loving imagination’.163 He dismisses both options as imputing a ‘half-conscious irony’ foreign to Kant.164 One cannot but agree that Kant must have intended more than expository clarity by such an intricate device. And although unconscious irony cannot be ruled out, half-conscious irony indeed seems alien to the generally sober tone of the Critique. Another possibility is that the faculty idiom does some real transcendental work in Kant’s account. Strawson’s lack of sympathy with Kant’s systematic project is transparent in the ‘General Review’ of The Bounds of Sense, where ‘the story of synthesis’ is rejected on the basis that ‘we can claim no empirical knowledge of its truth’.165 He goes on to raise the very pertinent issue of what resources this leaves for addressing the question: ‘when we speak of the necessary unity of experiences, what are the items so unified, and in what does their necessity consist?’166 His answer is as follows: First, the unified items are just the experiences reported in our ordinary reports of what we see, feel, hear, etc. . . . Second, the unity of these experiences under the rules embodied in the concepts of objects is just what is exemplified in the general coherence and consistency of our ordinary descriptions of what we see, hear, feel, etc. The employability of such concepts as these, hence the objectivity of experience in general, is necessarily bound up with the fulfilment of this requirement of consistency or unity.167

According to this view, the necessary unity within experience does not imply a priori forms generated by a spontaneous mind. Rather, our ordinary language descriptions pick out the structure of experience, which is now to be understood as amounting to its ‘coherence and consistency’. As Strawson later remarks, ‘no high doctrine’ – that is, theory of a priori form – is necessary.168 Kant’s insistence that only a transcendental form of experience will vindicate claims to validity within empirical experience, has been replaced with the claim that empirical experience itself – or at least our ordinary linguistic descriptions of it – exhibits its validity insofar as it is consistent and coherent. We no 37


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology longer find a transcendental theory of knowledge but, rather, empirical coherentism. I agree with Strawson that dualism and faculty talk are two sides of one coin, but I understand both of these and the relation in which they stand to one another in a quite different fashion. I will argue in what follows that Kant’s commitment to an investigation of the internal complexity of mind does not entail his giving up on a commitment to the empirical significance of those structures. Kant transformed the ‘faculty talk’ already current in the philosophical vocabulary.169 His faculty talk allows him to express a new model of reason as a dynamic interaction between a plurality of mental capacities or orientations. It also allows him to present experience as an ongoing task, rather than as a fait accompli. Subjective conditions initiate the task of knowledge, but they do not complete it. This positive account of ‘faculty talk’ distinguishes my position not only from those such as Strawson and Guyer who criticise it, but also from others like Allison, Pippin and Buchdahl who seek to replace it with a wholly logical concern with ‘epistemic conditions’. In the same vein, I take issue with Strawson’s understanding of Kant’s concern for limits.170 While Strawson follows Locke’s stricture that philosophy should practise a hygiene of self-regulation, the Copernican revolution introduces a radical reinterpretation of the role of limit in philosophy.171 Locke’s version expresses a pragmatic interest and intellectual modesty, whereas Kant moves the discussion to a metaphysical level insofar as he is concerned with limits as conditions of possibility for physis or nature.172 Where Strawson thinks that limits should only be abstracted from experience, Kant holds that limits are necessary conditions for the possibility of experience and thus count as a priori. Transcendental Idealism insists that empirical realism will only be secured if we are able to think the possibility of experience, not just its actuality. This entails thinking of limits in a more radical way than is allowed by a Lockean economy and leads to a fundamental disagreement about the role of form in experience. The a priori framework is the limit beyond which knowledge is impossible, but it is also the condition of the possibility of knowledge and experience. Our subjective cognitive faculties are the condition for any experience whatsoever and as such count as the limiting framework of that experience. However, as I have already suggested and will argue in what follows, these conditions initiate and do not impose order on nature. 38


29/3/07

8:12 AM


The Centrality of the Problem of Formalism Strawson’s critique of ‘faculty talk’ still has a significant influence on how Kant is read, despite the ways in which contemporary commentators distance themselves from his interpretation. It is likely that it is because of the way in which the ‘idiom of the faculties’ has been tainted that Kant’s supporters try to avoid the issue if at all possible. Meanwhile we have found that his critics return to Kant’s talk of the faculties. We have seen that Pippin rejects Strawson’s critique, preferring a metaphysically neutral reading over a psychological one. Pippin’s analysis of Kant’s supposed failure focuses on the latter’s aim of establishing an ‘independent analysis of knowing’ and not on faculty talk per se. However, it is faculty-talk that prevents Kant from escaping impositionalism via the theory of judgement. The reliance on an ‘obscure theory of pure intuition’ undermines the project of achieving a theory of knowledge that is comprehensive and at the same time ‘empirically guided’.173 And while Guyer seeks to distance his account from Strawson’s, I have suggested that his distinction between absolute and conditional necessity is more reliant on a concern with the ‘idiom of the faculties’ than he admits. Passages Guyer relies on for establishing Kant’s commitment to the absolute necessity refer to the faculties, while others that Guyer concedes display only a conditional necessity use the more neutral terminology of ‘conditions’. While Guyer’s criticism of Kant here is not in direct lineage with Strawson’s, it is nevertheless Kant’s concern with the internal structure of the mind and its status as a ground for experience that leads, in both their views, to impositionalism.

Conclusion In this chapter we have seen that all four interpreters considered present Kant’s position as formalist. The precise nature of the formalism attributed to him is varied, but in every case the latter is seen as resulting in the subjectivism of his epistemological project. Moreover for all of these authors, Kant’s project offers not just an investigation of the subjective necessary conditions of experience, but also the claim that its foundational ground lies in the subject alone. This leads Guyer and Strawson to conclude that Kant is impositionalist. While Pippin and Henrich resist this view, I have argued that their interpretations ultimately invite the conclusion that Kant’s formalism results in an overly subjective account of knowledge. 39


29/3/07

8:12 AM



Notes 1. For now I will leave open the ontological status of objects, which could be understood in a realist or phenomenalist manner. In what follows, I hope to establish that these are not the only options. 2. Paul Guyer, Kant and the Claims of Knowledge, p. 55. 3. Ibid., For an extended account of Guyer’s contrast between these two positions, see pp. 53–61 and my discussion below, pp. 17–18. 4. I will investigate how we might understand this claim in what follows. 5. Robert Pippin, Kant’s Theory of Form ( KTF), p. 223. See also pp. 29–30 on the notion of ‘guidedness’ and pp. 46–53. 6. Ibid., p. 51. 7. Ibid., p. 218. See also pp. 46 ff. on the problem of Angewiesenheit or guidedness and on ‘affection’. 8. Ibid., p. 219, n. 5. See Norman Kemp Smith, Commentary to Kant’s Critique of Pure Reason, and Hans Vaihinger, Kommentar zu Kants Kritik der reinen Vernunft. 9. This is the core of Kemp Smith’s phenomenalist reading of Kant, which is, he believes, equivalent to an objective idealism. 10. Kemp Smith, Commentary, p. xxxiii. 11. See, for instance, Vaihinger, The Philosophy of ‘As if’, pp. 30–1, where the categories are characterised as ‘analogical fictions’ that do not give rise to knowledge of reality. Interestingly, Pippin uncritically adopts Gerd Buchdahl’s characterisation of reflective judgement as a mental projection. Buchdahl’s account can be seen as a much more sophisticated reworking of Vaihinger’s interpretation. See discussion of Buchdahl in Chapter 2 (pp. 59–60). 12. See Pippin, KTF, pp. 34–5 on indeterminacy and pp. 29–30 on indirectness. 13. See Paul Guyer, ‘Kant on Apperception and A Priori Synthesis’; and Kant and the Claims of Knowledge. See Pippin, KTF, p. 102, n. 16. 14. Pippin, KTF, p. 102. 15. Ibid., p. 103. 16. Ibid., p. 219. 17. Ibid., p. 219. 18. Ibid., p. 102, n. 16. 19. See, for instance, Pippin, KTF, p. 100, n. 13, for the agreement in which Pippin stands to Melnick for whom ‘categories are the features objects must have if they are to be subject to our forms of thought’. Melnick uses the expression ‘epistemic concepts’ to capture the formal characteristic of the categories. Pippin refers to Melnick’s Kant’s Analogies of Experience, pp. 40 ff. especially p. 45. This account seems very close to Allison’s characterisation of the categories as epistemic conditions. See my discussion of Allison in the next chapter. The 40


29/3/07

8:12 AM



20. 21. 22. 23. 24. 25. 26. 27.

28. 29. 30. 31. 32. 33. 34.

35. 36. 37.

38. 39.

categories provide the form of experience, rather than imposing form on the latter. See Pippin, KTF, p. 102: ‘The question the deduction will pose is thus not: Can we discover in the understanding (as birthplace) concepts which lie there (like seeds) prior to any actual experience? but: Can we identify and justify a use of concepts which establishes a relation to all possible objects of experience which does not justify that use by appeal to what we have experienced?’ See Pippin, KTF, p. 185. Ibid., p. 221. Ibid., pp. 209–10, n. 38. See also p. 221. See the final section of this chapter (pp. 35–9). See, for instance, Pippin’s statement that Kant has a ‘metaphysically neutral notion of intellectual activity’, KTF, p. 221. See also p. 76. Ibid., p. 92 (Pippin’s emphasis). Critique of Pure Reason, A 137–47, B 176–87. Pippin, KTF, p. 87. Pippin’s restriction of Kant’s project to a system of rules shows, I believe, his agreement with Allison’s reading of transcendental idealism as a theory of epistemic conditions. See Chapter 2 of this volume (pp. 62–3). However, as we will see later in this chapter, Pippin’s account is also reminiscent of Henrich’s reading of Kant, where pre-temporal rules provide a formal framework. Whereas both Allison and Henrich hold that a formalist interpretation secures the success of Kant’s epistemic project, Pippin believes that Kant’s formalism is the source of its failure. See also Pippin, KTF, pp. 154 ff. Ibid., p. 185. See ibid., p. 171. The methodological is equated with the formal ibid., p. 196. The latter is equated with a meta-level analysis on p. 217. Ibid., p. 25. Ibid., p. 223. Ibid., p. 142; see also pp. 221–2. Pippin also suggests that Kant’s theory of judgement could have provided a solution to the problem of applicability. I hope to show in what follows that judgement is indeed the clue to the possibility of applicability. Ibid., p. 222. Ibid., p. 222. In Chapter 4, pp. 113–16, I argue that dualism is less rigid than Pippin suggests and entails a plurality of the faculties. Ibid., p. 142 (Pippin’s emphasis). The tension in Pippin’s position is evident in the move between a rejection of ‘any simplistic resort to the imposition metaphor’ on p. 226 and his final diagnosis of Kant’s formalism on pp. 227–8. Ibid., p. 142. Ibid., p. 152. 41


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 40. Ibid., p. 152. 41. This sounds very similar to Henrich’s interpretation as we will see below. 42. Pippin, KTF, pp. 209–10. See discussion earlier in this section, p. 12. 43. See especially ibid., pp. 143–50. 44. Critique of Judgment, AA 179. 45. See discussion of the systematic function of reflective judgement in Chapter 7, pp. 249–55. 46. Pippin, KTF, pp. 118–19 (my addition). See also p. 221. 47. Ibid., p. 118, n. 41. See also Pippin, ‘Kant on Empirical Concepts’, p. 15: ‘It is the subjective nature of this demand or ‘need’ for unity which, while it reveals how deeply connected Kant’s version of empirical knowledge is with his theory of reflective judgement, is of little help with our problem here.’ In both cases Pippin gives exactly the same reference to Buchdahl’s Metaphysics and the Philosophy of Science, pp. 504–6. Interestingly, in later work, Pippin introduces a quite different interpretation of reflective judgement as involving an orientation to nature. See his ‘The Significance of Taste’, especially pp. 567–9. See also ‘Avoiding German Idealism’, especially pp. 987–8 and pp. 990–1. 48. Pippin, KTF, pp. 227–8. 49. Ibid., p. 228. 50. Ibid., p. 230: ‘The broad problem at stake in all such critical enterprises is how to determine in general proper “account-giving” in philosophy, and indeed in all knowledge or discourse. To make such a determination implies some sort of comprehensive perspective on what can or cannot be said, or known . . . I have suggested that, however important and interesting many of Kant’s reflections on the problem of form are, this comprehensiveness is not attained.’ 51. See ibid., pp. 5–6, n. 6, on Hegel. 52. Ibid., p. 212: ‘If we are in fact “called on” to make assumptions about experience, which assumptions can direct but whose terms transcend such experience, then which assumptions would seem to be much more a result of reflection on the actual, scientific questions asked in various contexts, than on the formal nature of human reason. If in fact this were true, then it would also follow that the more effective such “holistic” assumptions were, especially in increasing actual empirical knowledge, the more inclined we would be to treat such assumptions less as regulative maxims, and more as correct explanations of nature.’ 53. Ibid., p. 228. 54. See, for instance, Paul Guyer, ‘Kant on Apperception and A Priori Synthesis’, p. 211. 55. Paul Guyer, Kant and the Claims of Knowledge ( KCK). 56. Guyer, ‘Kant on Apperception’; Pippin, KTF. 42


29/3/07

8:12 AM


The Centrality of the Problem of Formalism 57. Guyer, KCK, p. 123. 58. Ibid., p. 55: ‘he does not suggest that the mind is such that it can always ensure that experience is possible, but rather that it is so constituted that experience will be possible only if the objects of experience, as a matter of fact, conform to the requisite conditions’. 59. Ibid., p. 123. 60. Ibid., p. 55 (quoted on p. 9, this chapter). 61. Ibid., p. 55. 62. Ibid., p. 363. Guyer refers to the discussion of the debate between Trendelenburg and Kuno Fischer in Vaihinger, Kommentar (vol. 2, pp. 290–326), and in Kemp Smith, Commentary, pp. 113–14. 63. I agree with Guyer’s diagnosis that, for Kant, the fit between the spatiotemporal features of mind and of objects cannot be merely coincidental. However, this does not mean that form is imposed on nature, but rather that form supplies a structure necessary for any experience. This is only the beginning and not the conclusion of how order arises within experience. 64. See Karl Ameriks’ excellent account in ‘Kant and Guyer on Apperception’, in Archiv für Geschichte der Philosophie, vol. 65. 65. Guyer, ‘Kant on Apperception’, p. 206. 66. Ibid., p. 206. 67. Ibid., p. 206. 68. Ibid., p. 208. 69. Ibid., p. 208. 70. Ibid., p. 206, A 99–100: ‘synthesis of apprehension must also be exercised a priori, that is, in respect of representations which are not empirical. For without it we should never have a priori the representations either of space or of time.’ Guyer notes in this article and at the outset of his book that the translations he uses are normally his own, although guided by Kemp Smith. I leave his translations unaltered in this section. 71. Ibid., p. 206. References are to A 99–100 and to A 110. 72. Ibid., p. 206. The reference is to A 113. 73. Ibid., p. 206. The reference is to A 101–2. 74. Ibid., p. 206. The reference is to A 117–18. 75. Ibid., p. 206. The reference is to Critique of Pure Reason ( CPR), B 134. The emphasis is Guyer’s. 76. This suggests that Guyer’s critique of Kant may be more indebted to Peter Strawson’s The Bounds of Sense than he suggests. See the discussion of Strawson’s critique of Kant’s use of an idiom of the faculties in the final section of this chapter. Guyer refers to Strawson at the outset of his article, seeking to distinguish his own position from the latter’s. See Guyer, ‘Kant on Apperception’, p. 205. In the final section, Guyer claims that ‘“transcendental psychology” may be removed from “what is living” in Kant’s philosophy’ (p. 211). 43


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 77. Guyer, ‘Kant on Apperception’, p. 207. 78. Ibid., p. 207. Reference also to CPR, A 130. 79. Ibid., p. 207. The reference is to CPR, A 125. Guyer’s translation is slightly different from Kemp Smith’s. See also the discussion of A ‘Deduction’ in Chapter 4, pp. 121–32. 80. Guyer, ‘Kant on Apperception’, p. 207. 81. CPR, A 125. 82. Thus I agree with Guyer’s translation on this point. 83. CPR, A 125–6 (my translation). 84. Guyer, ‘Kant on Apperception’, p. 207. The reference is to CPR, A 125 (Guyer’s emphasis). 85. Ibid., p. 207. The reference is to A 130. 86. See discussion of the hierarchical nature of the full version of Kant’s deduction of the categories in Chapter 6 below. 87. Guyer, ‘Kant on Apperception’, p. 206. 88. Guyer, KCK, p. 6. 89. See ibid., p. 418, for Guyer’s analysis of the structure of Kant’s ‘transcendental deductions’ in contrast to ‘transcendental proofs’ understood more widely. 90. Ibid., p. 438, n. 2. 91. Ibid., p. 131. 92. As we will see in the next section (pp. 25–35), Guyer’s account of the identity of the self throughout experiences owes a debt to Henrich. 93. Guyer, KCK, p. 131 (my addition). Once again this is very close to Henrich as we will see in the following section (pp. 25–35). See Guyer, ‘Review of Identität und Objectivität’, p. 153, where he refers to pp. 16–53 of Henrich’s German edition. 94. CPR, A 127; cited in Guyer, KCK, p. 132. 95. CPR, A 122; see Guyer, KCK, p. 132. 96. Guyer, KCK, p. 132. 97. Ibid., p. 26. 98. In contrast Guyer asks why we shouldn’t rehabilitate ‘transcendental realism’ regarding space and time. See ibid., p. 349. 99. In Guyer’s view this is to be found principally in some unpublished remarks and in a letter written in 1792 to Beck and not in the CPR. It thus stands in contrast to what Guyer interprets as Kant’s official position. 100. Guyer, KCK, p. 415. 101. Ibid., p. 414. See discussion of Allison’s Kant’s Transcendental Idealism in Chapter 2, pp. 61–9. 102. Ibid., p. 414. 103. In this section I will principally discuss Henrich’s Identität und Objectivität. An English version, ‘Identity and Objectivity’, is to be found in Henrich, The Unity of Reason. 44


29/3/07

8:12 AM


The Centrality of the Problem of Formalism 104. See Guyer’s ‘Review of Identität und Objectivität’. 105. We will come back to what is meant by this particular kind of identity. See Henrich, ‘Identity and Objectivity’. See especially pp. 186–7 on numerical identity and Cartesian certainty (Identität und Objectivität: pp. 86–7). See also Henrich, ‘The Identity of the Subject in the Transcendental Deduction’, in Schaper and Vossenkuhl, Reading Kant, especially pp. 254 and 261–6 on ‘numerical identity’. 106. Henry Allison, Kant’s Transcendental Idealism ( KTI) (1983), p. 140. (This discussion is omitted in the 2004 edition.) In a note, Allison says that Guyer criticises Henrich for his Cartesian reading of Kant (see p. 353, n. 23). But it is rather that Guyer criticises the Cartesian tendency in Kant and agrees with Henrich’s reading. See Guyer, ‘Review’, p. 162. Speaking of Henrich’s claim that, for Kant, we have a priori knowledge of numerical identity, Guyer states: ‘I have no doubt that Kant does subscribe to this argument’. Admittedly Guyer questions Henrich’s use of the term ‘Cartesian’. He suggests that to call the certainty of apperception ‘Cartesian’ is misleading as Descartes was referring to a singular starting point, whereas Henrich seeks to establish the foundation of a plurality of certain forms of judgement. Moreover, for Descartes the certainty of the Cogito does not imply certainty about the continuing existence of a thinking substance or of the conservation of finite substances, for the latter requires God. However, it is likely that Henrich uses the term ‘Cartesian’ in a way similar to Husserl, that is to express an apodeictic starting place, rather than to subscribe to Descartes’ wider metaphysics. See Husserl, Cartesian Meditations. See Chapter 2, pp. 61–9 below for an extended discussion of Allison. 107. Allison, KTI (1983), p. 144 (my emphasis). 108. Ibid., p. 144. 109. This is in part because (as we will see later), for Henrich, apperception is not only real but also formal. 110. Guyer, ‘Review’, p. 166. See also p. 167, where he says that absolute certainty entails ‘we can always force our future representations to conform to our categorial rules’. 111. See discussion of logical derivation and real constitution later in this section. 112. Guyer, ‘Review’, pp. 166–7. 113. Yet again Guyer refers to Kant’s account of the ‘affinity’ of nature as his trump card. See ‘Review’, p. 166, n. 12. The reference is to CPR, A 121–125. See my discussion of this in the previous section. 114. See discussion of Guyer’s article, ‘Kant on Apperception’, on p. 18. 115. See in particular Guyer’s discussion of Henrich’s insistence on the necessity of self-consciousness on pp. 163 and 164 of his ‘Review’. 45


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 116. Henrich, ‘Identity and Objectivity’, p. 198; references to the German edition follow in brackets (p. 100). 117. Guyer does not discuss this specific argument in his ‘Review’. 118. Henrich, ‘Identity and Objectivity’, p. 201 (p. 103). 119. Ibid., p. 201 (p. 104). 120. CPR, A108. Cited by Henrich, ‘Identity and Objectivity’, pp. 199–200 (p. 102). 121. Henrich, ‘Identity and Objectivity’, p. 200 (p. 102). 122. Ibid., p. 201 (p. 104). 123. CPR, A 108: ‘The original and necessary consciousness of the identity of the self is thus at the same time a consciousness of an equally necessary unity of the synthesis of all appearances according to concepts, that is, according to rules, which not only make them necessarily reproducible but also in so doing determine an object for their intuition, that is, the concept of something wherein they are necessarily interconnected’ (my emphasis). 124. See Chapter 6 on Kant’s discussion of special laws in the ‘Deduction’. 125. Henrich, ‘Identity and Objectivity’, p. 201 (p. 104). Note Henrich’s mention of the capacity of the subject to initiate synthesis. 126. Ibid., p. 186 (pp. 85–6). 127. Ibid., p. 186 (p. 85). 128. Ibid., p. 181: ‘for the states of the representing subject are dependent upon what is given to it in intuition’ (p. 79). 129. Ibid., p. 204 (p. 107). 130. Ibid., p. 203 (p. 106). 131. Ibid., p. 164 (pp. 58–9). 132. Guyer, ‘Review’, pp. 152–3. Guyer argues that it is Henrich’s commitment to the certainty of self-consciousness that leads to absolute necessity, but it is the conclusion of this entailment that really worries him. 133. Ibid., p. 153. Guyer refers to pp. 52 and 109 of the German edition. 134. The first passage is from Henrich, ‘Identity and Objectivity’, pp. 159–60 (p. 52). The second is from Ibid., p. 206 (p. 109). See Guyer’s ‘Review’, p. 153. 135. Karl Ameriks, ‘Recent Work on Kant’s Theoretical Philosophy’, p. 16. He refers to Ralph Walker, A Selective Bibliography on Kant, p. 195. 136. Henrich, ‘Identity and Objectivity’, p. 187 (p. 87). 137. See ibid., p. 180 (p. 78): ‘Particular things can remain the same throughout change in their states as long as several of their constitutive properties remain unchanged or as long as the thing in question endures through the continuous, but never total change of its states.’ In contrast, Leibniz’s strict numerical identity is committed to the view that ‘all states of substances are present in them from the beginning onward’ (p. 179 (p. 77)). This results in the theory of substances as monads. 46


29/3/07

8:12 AM


The Centrality of the Problem of Formalism 138. 139. 140. 141. 142. 143. 144. 145. 146. 147.

148. 149. 150.

151. 152. 153.

154.

155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166.

Ibid., p. 187 (p. 86). Ibid., p. 188 (p. 88). Ibid., p. 188 (p. 88). Ibid., p. 189 (p. 89). Ibid., p. 189 (p. 90). Ibid., p. 201 (pp. 103–4). See discussion on pp. 28–9. Ibid., p. 207 (p. 111). Ibid., pp. 207–8 (pp. 111–12). Ibid., p. 208 (p. 112). Henrich, ‘The Identity of the Subject in the Transcendental Deduction’, Section 6: ‘Identity as a Formal Property of Self-Consciousness’, in Schaper and Vossenkuhl, Reading Kant, pp. 266–70. See in particular p. 268. Ibid., p. 268. Ibid., p. 268. Ibid., p. 269. It should be noted that Henrich’s ‘fact’ is not the same as Allison’s. See Allison, KTI, p. 140, cited on p. 25 above. Whereas Henrich’s ‘fact’ would be a successful actualisation of form and would not count as a priori, Allison’s is the transcendental presupposition of a formal structure. Allison, KTI (1983), p. 140. See discussion on pp. 25–6 above. Chapter 6 is a working-through of this perspective. I present a version of this argument in my ‘Kant’s Phenomenological Reduction?’, where I argue that aesthetic judgement reveals the primary temporalisation and spatialisation necessary for any judgement (Hughes 2007). ‘The Moral Image of the World’, in Henrich, Aesthetic Judgement and the Moral Image of the World, p. 3. At the same point he also says ‘it can be shown that the unity of self-consciousness could not even be conceived unless that very unity functions as the point of departure for constituting a world of objects’. Strawson, The Bounds of Sense, p. 30. Ibid., p. 16. Ibid., p. 18. Ibid., p. 44 (Strawson’s emphasis). See Chapter 5 on the subjective side of the deduction, and Chapter 4, pp. 116–19, on the status of faculty talk. Strawson, The Bounds of Sense, p. 44. Ibid., p. 38. Ibid., p. 22 (my emphasis). Ibid., p. 22. Ibid., p. 22. Ibid., p. 32. Ibid., p. 32. 47


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 167. Ibid., p. 32. 168. Ibid., p. 44. 169. For a related claim, see Henrich, ‘Kant’s Explanation of Aesthetic Judgment’, in Aesthetic Judgement and The Moral Image of the World. See especially pp. 32 and 54. 170. Strawson, The Bounds of Sense, p. 44. 171. See John Locke, An Essay Concerning Human Understanding ‘Introduction’ pp. 46–7: ‘Our Business here is not to know all things, but those which concern our Conduct’. See also p. 44. 172. Kant was undoubtedly metaphysical in this sense, as is any philosophical position that goes beyond bare naturalism. If any version of transcendence is in question, metaphysics is at issue. But this is not to say that the projection of a transcendent or dogmatic position is necessary for metaphysics. If transcendence emerges within, or at least in relation to, experience, metaphysics can be critical or dualist and avoid dogmatism. 173. See discussion on Pippin, pp. 10–17 ff. above.

48


29/3/07

8:12 AM


2


The purpose of this chapter is to show how formalism is defended by some of Kant’s most important recent supporters. Gerd Buchdahl’s Metaphysics and the Philosophy of Science, first published in 1969, set the scene for the riposte to the dominant Strawsonian critique of Kant, published three years previously in 1966.1 Henry Allison openly acknowledges his debt to Buchdahl, as does Pippin in his less sympathetic interpretation of Kant. Béatrice Longuenesse’s direct reference to Buchdahl is restricted to one critical note, but his influence can be indirectly traced through Allison.2 While in the accounts offered by Buchdahl, Allison and Longuenesse we will find more positive accounts of formalism than offered by Kant’s critics, I will nevertheless argue that we discover an emphasis on subjective structure at the expense of a convincing account of the relation between subjective form and the material given in experience. Both Buchdahl and Longuenesse, for all the many strengths of their interpretations, risk falling into what I call a circle of representation. Allison has a more convincing account of affection, but still fails to provide a sufficiently robust account of the material side of experience.

I Buchdahl’s Reductive Formalism Buchdahl’s intricate and forceful interpretation of Kant has not often been given due attention. While his extensive discussion of Kant in Metaphysics and the Philosophy of Science is regularly cited, it is not so often analysed or assessed. Nevertheless, Buchdahl’s interpretation of Kant has been of great importance for Pippin, as we saw in the previous chapter, and also for Allison, whom I will discuss in the following section. Buchdahl’s reading diagnoses a hierarchy of formal frameworks for experience in Kant’s epistemology. He argues that the categories of the understanding provide a provisional template for the derivation of methodological principles that establish how we should judge in empirical cases.3 The categories do not, however, automatically supply the content or even the form of empirical judgements.4 49


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology His later work, Kant and the Dynamics of Reason, develops this perspective and, in particular, the view that an approach inspired by Husserl’s phenomenological reduction allows for a new assessment of the status of Kant’s ontological commitments. I will concentrate exclusively on establishing the formalist status of Buchdahl’s interpretation and on showing how his version of formalism leads to a problem in the account he gives of extra-mental reality. Buchdahl insists that Kant’s critical project is not intended to secure the certainty of particular descriptions of the world, but only the formal conditions which first make possible the certainty of those propositions: We are no longer concerned with the question of the certainty of our knowledge of such and such a set of statements, but only of the formal conditions that have to be satisfied before questions of certainty can even be asked.5

This is critical philosophy’s transcendental turn to the conditions of the possibility of experience. Only once transcendental conditions have been established will it be possible to assess the validity of concrete propositions about experience. Despite widespread recognition that conditions of possibility are the trademark of the critical turn, there remains, in Buchdahl’s opinion, much confusion between and conflation of transcendental and empirical levels of analysis. One leading example of this is the assumption that in the analysis of causality in the Second Analogy, Kant is already referring to actual causal connection, rather than investigating the ground of the possibility of any causal connection.6 Kant’s focus, Buchdahl insists, is on the transcendental form of experience, although he prefers to refer to this as structure, and, in particular, as transcendental structure.7 In addition to the distinction between transcendental and empirical levels of experience, Buchdahl insists on further distinctions at the transcendental or formal level. In Metaphysics and the Philosophy of Science, Buchdahl distinguishes between the categorical and regulative orders of experience.8 While the first coincides with the most general order of experience or phenomena, the second aims at establishing an order among the range of particular natural laws. Thus there are two orders of nature. The first arises from the understanding and its generation of categories, while the second arises only insofar as reason projects an orientating idea that makes possible order in empirical nature. Buchdahl argues that the second order is only loosely or analogously modelled on the first, which thus does not 50


29/3/07

8:12 AM


Formalism and the Circle of Representation provide the validating grounds of the former.9 While categorical principles are objective in the sense that, without them, the experience of objects would be impossible, regulative principles are only subjective in that they arise from an intervention on the part of the subject.10 The order among empirical laws of nature is a construct that facilitates the investigation of that world, particularly by science. In Kant and the Dynamics of Reason, Buchdahl distinguishes three levels of formal structure. These coincide with what he calls general, special and systems ontology. All three operate at the level of possibility, that is, they establish the possibility of a certain level of experiencing nature. General ontology investigates the possibility of nature in general, while special ontology is concerned with material (or physical) nature and systems ontology with ‘nature as an “ordered” system of objects and of the empirical laws that govern their behaviour’.11 The first coincides with the constitutive principles and the third with the regulative principles. The second denotes the territory of Kant’s Metaphysical Foundations of Natural Science and ‘investigates the possibility, and thus intelligibility, of the basic concepts and laws of Newtonian science’.12 The identification of this distinctive ontology reinforces Buchdahl’s point that the Critique of Pure Reason does not directly offer a foundation for Newtonian physics.13 In laying out the three levels at which nature can be ordered, I follow Buchdahl’s order of presentation. However it may be useful to remember that the hierarchical sequence from generality towards empirical application would, rather, be from general via systems to special ontology. Systems ontology provides the mediating principle that allows for the transition from general ontology to the construction of a formal structure for Newtonian physics. Primarily Buchdahl is concerned with the first and third levels of ontology, that is, with ‘the real possibility of objective cognition or experience in general’ and with ‘the problem of the validation of the methodological maxims and ideas of natural science’.14 As in his earlier work, the analysis of causality is a touchstone for the distinction between levels. General ontology establishes only that a concept of causality must be presupposed. Any actual causal connection can only count as contingent.15 Buchdahl goes so far as to say that the transcendental category of causation ‘might be quite compatible with the absence of a network of laws, or of any laws whatsoever’.16 Systems ontology alone is capable of establishing that experience as a whole is causally connected, but it only operates regulatively or subjectively, that is, from the point of view of the judging subject. 51


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology The methodological approach, which treats nature as a system of empirical laws, allows for the emergence of a further level of analysis of nature as material or physical. This is the level of special ontology, which, while it does not directly coincide with Newtonian physics, establishes the philosophical perspective from which the latter operates. Each level of possibility, or ontology, ultimately counts as formal. In general ontology the forms of space and time and the categories supply the framework of nature in general. In systems ontology this same framework indirectly supplies a form for empirical laws. The Metaphysical Foundations of Natural Science offers a further determination of what is ultimately the same form by establishing the basic concepts and laws of Newtonian science, focusing principally on the concept of force. Importantly, even at this level of analysis Kant is concerned not with empirical experience and only with its form. In analysing the formalist status of Buchdahl’s interpretation of Kant, it is illuminating to examine the relation the former draws between the transcendental and formal status of the frameworks under discussion. The following passage from his earlier work addresses what he later refers to as general ontology: [T]he ‘critical’ approach will (as already mentioned) formulate the ground of the possibility of an individual material thing as such in terms of transcendental structure, i.e. of construction in space and time, under the guidance of the categories: an essential requirement for the definition of the ‘reality of the data’ being that the formal framework should contain an a posteriori element, i.e., be anchored in an instance of actual experience of the completed object (qua ‘phenomenon’).17

A problem arises that will turn out to be crucial for Buchdahl’s formalist interpretation of Kant. Buchdahl claims that the formal framework must ‘contain’ or ‘be anchored in’ an a posteriori element. How could this be, if the qualifying characteristic of the transcendental level of analysis is that it operates strictly at an a priori level? Does this suggest that the formal framework and the transcendental structure are not equivalent? Moreover, is there not a contradiction between saying on the one hand that the formal framework contains the a posteriori while on the other hand it is anchored in the latter? The metaphors seem to push in two directions, suggesting contrary theses. We will see later that this tension arises from Buchdahl’s commitment to there being a balance between the framework and the contents of experience, that is, between the a priori and a posteriori. 52


29/3/07

8:12 AM


Formalism and the Circle of Representation I need to pre-empt the conclusion to my later discussion of this problem and Buchdahl’s solution, because his fashion of resolving the issue allows us to clarify the relation between transcendental structure and formal framework. Buchdahl rearticulates what was to have been a balance in a onesided manner, for he claims that the a posteriori is finally grounded in the a priori. Thus transcendental structure and formal framework are identical here – despite the fact that the latter contains the a posteriori – because, in the final analysis, the given arises from the spontaneous contribution of the subject.18 Clearly, though, this solution severely compromises any suggestion that the formal framework is anchored in or, in any meaningful sense, balanced by experience. Later I will look at Buchdahl’s account of the status of the given in more detail. Buchdahl is hesitant as to whether the second level of framework, i.e. that of the fundamental laws of the natural sciences, qualifies as transcendental. On occasion he signals ambivalence as to this and even as to its formal status: Particularly in the case of EC [Special Ontology] it is not so much any formal ‘validity’ that is in question, as whether the explication involved is fruitful, anchored in the general structure of physical experience. In the case of EC, the argument is only barely transcendental. . .19

Later he takes a negative position on the question saying that ‘in Kant’s technical terminology of this it is not something “transcendental” but an analysis in “special metaphysics” ’.20 We can conclude that special metaphysics provides a framework, but not necessarily a transcendental one and thus it may not be formal in the strong sense where the two terms are equivalent. The discussion preceding Buchdahl’s negative judgement on the transcendental status of special metaphysics reveals that the other two levels of framework do so qualify.21 This is also clear from the following more moderate contrast: [l]eaving it open whether Special Ontology is transcendental in quite the same way in which General Ontology is construed; unlike the case of Systems Ontology, where we found that there is a relatively clear-cut distinction between the employment of the methodological framework, as yielding the phenomenology of a scientific theory, and its employment as generating a Systems Ontology.22

This passage identifies the stakes in qualifying not only as formal but also transcendental. What is required is that there be a clear distinction 53


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology between the framework or form and the particular descriptions of the world founded on that framework. General and systems ontologies achieve this and thus count as transcendental. In contrast, Special Ontology lacks a sufficient gap between ontology or structure and phenomenology, that is, the range of experiences falling under that structure.23 Buchdahl’s criterion for transcendental status would bear much more examination, but we can sum it up quickly for our purposes by saying that only if a clear distinction is achieved between the a priori and the a posteriori can the former operate as the condition of the possibility of the latter. In Buchdahl’s account, form is equivalent to determination and matter is the determinable.24 Form and matter are two sides of the experiential world of appearance qua forms of intuition and empirical content. As such, they also function as the poles of what he calls the reduction of that appearance to its transcendental conditions. This aspect of Buchdahl’s approach is most developed in Kant and the Dynamics of Reason but is already latently present in Metaphysics and the Philosophy of Science where he says that the notion of the object is ‘“bracketed” by virtue of the transcendental condition’.25 The Husserlian idea is that the transcendental conditions of the possibility of experience, which Buchdahl, unlike Husserl, calls its ontology, can be accessed by abstracting specific qualities from objects and returning to their transcendental or formal conditions. This strategy of reduction turns out, on Buchdahl’s reading, to be the crux of the critical turn in philosophy. The reduced aspect of form is captured in Husserlian terminology thus: ‘the noetic aspect of form is defined as “determination” [Bestimmung], “relation”, or again, as “copula”.’26 Thus form is an element – and, we will find, threatens to become the unique element – of the transcendental structure that emerges under reduction. Buchdahl goes so far as to say: ‘the ground of the world has become “transcendental form” ’.27 But Buchdahl is insistent that this strategy does not signal a subjective turn. The logical or grammatical reduction of the phenomenal world does not undermine the latter’s empirical reality. It is only its ontological status that is put in question, in that the transcendental method, as interpreted by Buchdahl, eschews reliance on an external ontological source for the being of objects. What is left is the mere ‘fact’ that objects are given to us in experience. Buchdahl refers to this by the German term Sachheit.28 Givenness, considered at the transcendental level, i.e. at the level of the possibility of experience, is nothing but the fact that there is something rather than nothing. 54


29/3/07

8:12 AM


Formalism and the Circle of Representation While there are causes within experience, there is no cause, no double affect, of experience. Talk of the given at the transcendental, as opposed to the uncritical, level is simply ‘Pickwickian’, that is, merely a form of speaking that arises from the internal logic of a technical discourse.29 In other words, we speak of and, perhaps, have to think of reality standing at the limits of experience, yet this image is misleading when we assume that there is something outside experience.30 The form of experience is reached by a reduction, but lived experience is re-attained through a construction that realises the form within a concrete experience. Realisation constructs a particular object through giving spatial and temporal form to matter. Thus the two sides of the reduction are repeated or realised at the empirical level. Buchdahl calls his method in general ‘Reduction-Realization Process’, and usually refers to it as RRP.31 Only what he calls ‘Transcendental Reflection’ can reduce experience to its transcendental ground and then progressively construct, that is, realise that ground at different levels of actuality.32 Matter is undifferentiated prior to its formal construction. Buchdahl concludes that there is only one object that is analysed or interpreted in many ways or at many different levels. Only realisation differentiates the singular transcendental object into a plurality of empirical objects. As I mentioned earlier Buchdahl is committed to the view that there is a balance between form and materiality.33 While this way of describing experience is intended to show that Kant’s transcendental project does not threaten the objective existence of objects, the account given by Buchdahl in both Metaphysics and the Philosophy of Science and in Kant and the Dynamics of Reason is unlikely to convert the unconverted. Not only is the objective situated entirely within the immanent, it is ultimately given a subjective ground. For while Buchdahl distinguishes the subjective and objective trajectories within constitutive judgement, it turns out that the given or objective side ultimately comes from the subject: It follows further that the character of the ‘given’ as ‘objective’, i.e., as other-than-self, will likewise have to be construed as a contribution made a priori. The character of the object as being ‘other-than-self’ will have to be located, not in what is received, but in what is spontaneously contributed by the cognitive subject in the context of judging. We might describe this by saying that the ‘self’ is here regarded as the seat of the a priori, and of what is ‘other-than-self’: objectivity being located within the framework of possible experience, instead of residing in a realm that transcends experience.34 55


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology The given as objective is grounded in the spontaneity of the subject. If this is what balancing amounts to, it is very one-sided. While Buchdahl could have argued that what is other-than-self always stands in relation to the self, he goes a step further by saying the former is located in the contribution made by the subject. Moreover, he says this contribution is spontaneous, not receptive, and that there is no role for the given. The suspicion is strengthened that his initial insistence that there is no perspective beyond the world of the subject results in the subjectivisation of the object. However, might not Buchdahl simply be committed to the position that the form of the given comes from the subject? This would be quite uncontroversial for it is unquestionably Kant’s position that the form of sensibility is subjective. If this is right, Buchdahl’s argument is only that our ability to take up the given – and thus to refer to an object in the first place – requires the a priori capacity of intuition insofar as the latter introduces spatial and temporal form into the manifold of apprehension. And this would mean that he is simply not concerned with the extra-mental element of the given. But the problem is exactly one of omission. Buchdahl fails to adequately address the question of affection, or at least the side of it that, in passing, he refers to as ‘otherthan-self’.35 He concedes that what is passively received is one of the component elements of the given, but quickly resolves the element that is ‘other-than-self’ into a discussion of the a priori structure of the given, which he says ultimately derives from the self.36 Buchdahl is quite right to say that what is given is not simply passively received because of the formal element Kant discovers in intuition. However, his accounts of both receptivity and what is received by our power of intuition are inadequate.37 Givenness is reduced to the form of matter that can be realised or constructed as empirical objects. In Kant and the Dynamics of Reason there is independent and suggestive evidence that Buchdahl does not intend to adopt a subjectivist position. He argues that the transcendental object [Buchdahl’s abbreviation is To], which is arrived at in the reduction, ‘ “determines” what particular concrete empirical intuitions will come up a posteriori’.38 To, in one of the many varieties Buchdahl distinguishes, qualifies as a ‘boundary notion [Grenzbegriff] (Ton) which “limits sensibility”, and thereby allows the “problematic” possibility of something “positive beyond the circumference [Umfang] of sensibility” ’.39 Buchdahl is most concerned to counter the view that affection amounts to the causal affect of objects on us and his alternative is that To is the ground of appearances. For this reason he rejects Graham Bird’s interpretation 56


29/3/07

8:12 AM


Formalism and the Circle of Representation of the transcendental object as a logical myth.40 Affection [Affizierung] is best understood not as a causal relation, but as ‘ “being brought to” – or “being in” a state of conscious awareness’.41 These passages reveal that while the transcendental object does not causally influence us, no more is it a mere thought experiment. Admittedly Buchdahl says that Sachheit is ‘the merely conceptual aspect of the holding of a certain state of affairs; in particular, to the fact that such and such is the case’.42 But this does not mean that he considers experience as a logical construction. Instead he wants to emphasise the contingency of experience and the fact that it ‘involves a ‘finding’ as well as an ‘originating’; that there is a particular state of affairs rather than another is simply the case.43 This is not something that bears further reflection, but is the beginning point for all reflection. The ‘fact’ of experience does not render the latter subjective, but merely establishes that its contingency is not open to further analysis. Buchdahl’s intent of avoiding attributing to knowledge a merely subjective status is further evident in a comment he makes on Fichte: It may be of interest to note that this reading echoes the approach taken by Fichte already during Kant’s lifetime, except that we have sought to replace the Fichtean somewhat misleading emphasis on the ‘self’, the Ich, by the more neutral language of the process of reduction and realization.44

While he concedes his project is comparable with Fichtean idealism, Buchdahl seeks to establish the distinctiveness of their approaches. So what is it that is to be denied in Fichte’s approach? Buchdahl does not accept that experience in general arises from self-affection. This is particularly clear earlier in a harsher judgement on Fichte: If this is not an ‘external cause’, neither should it on any account be described as standing for the ‘affection of the self by the self – the disastrous Fichtean direction of post-Kantian idealism. The self does not ‘posit’ itself as an entity affecting itself; rather, it is our analysis of realization which posits the self (sensibility) as affected or determined, a concept intended to render nothing more than the aspect of ‘the given’ in the object of that analysis.45

Buchdahl wants to replace Fichte’s subjective idealism with language that establishes the logical or grammatical ground of experience in a neutral place between subject and object. It remains to be seen whether the turn to the logical subject and the ‘Reduction-Realization Process’ accessed only by transcendental reflection is as neutral as Buchdahl’s emphasis on balance would require. 57


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology In Metaphysics and the Philosophy of Science Buchdahl discusses a ‘shift in the grammar of the Kantian “subjectivity” ’.46 There is ‘no reason to locate concepts, just because they are logical entities, in a realm of subjectivity’.47 Further the subjective is not to be equated with the psychological or empirical.48 This may seem rather confusing. Are concepts subjective or not? But as Buchdahl is about to make clear, what is at issue here is what he calls a sharpening of Kant’s use of subjectivity. What he means is that Kant’s range of reference is always subjective, where this term corresponds with experience in general in contrast to the thing-in-itself.49 We can therefore only make distinctions between different degrees of subjectivity within that broader horizon. In other words, Kant says that experience and the categories that provide order are subjective so as to emphasise that we are not concerned with a world beyond the bounds of possible access. Our world is always just that, the world that is accessible to us. Within our world, the subject has a greater or lesser role to play depending on the level of analysis. For instance, the projective role of reason involves a much more active and explicit role for the judging subject than does the relatively uncreative role of the understanding.50 On the face of it, this is a convincing representation of Kant’s transcendental or Copernican turn. Buchdahl’s aim, then, is to reappraise the relation in which subjectivity and objectivity stand to one another within the Kantian project. This will be of the greatest importance for, as he says: ‘[t]his “bringing together” of the “subjective” and the “objective”, of the “act” and what results from the act, is the very epitome of the Kantian method.’51 However the suspicion remains that what results is a rather subjective synthesis of subject and object, as the following passage suggests: This ‘sharpening’ of the Kantian meanings is graphically illustrated in putting Kant’s main point by saying that the ‘objectivity’ of the phenomenal object is not ‘objectively given’ but wrought ‘subjectively’. This at once makes it clear that the second ‘objectivity’ in this account no longer has a normal sense.52

Admittedly, Buchdahl simply says that the objectivity of the phenomenal object and not the phenomenal object per se is ‘wrought subjectively’. On the face of things this, once again, is a straightforward interpretation of Kant: for an object, strictly speaking, arises from the subsumption of an intuition under a concept of the understanding and thus entails the operation of our subjective faculties. But the 58


29/3/07

8:12 AM


Formalism and the Circle of Representation problem with Buchdahl’s position is, as previously remarked, one of omission. Buchdahl omits to mention that our intuition operates on something given in experience. This suggests that the given too – not just the forms of intuition – comes from the subject. The result is that subjectivity contains objectivity or the a posteriori within it, rather than what was at first presented as its equivalent, that the a priori is anchored in the latter. When we recall that one page earlier, Buchdahl has declared Pickwickian the status of the given, he is left with a very unbalanced account of the relation between the form and the matter of experience. What has happened to dualism? Certainly for Kant the forms of intuition are subjective, but they establish the possibility of our receptivity to something given in experience and, in the case of spatial form, something external. Buchdahl is, of course, right to insist that there is no absolute externality qua thing-in-itself. But it seems that the sharpening of subjectivity situates the limits and not just the form of experience within the subject. Such an interpretation, while offered in defence of Kant, opens the way for critiques such as Pippin’s. We have seen that the account given by Buchdahl in his earlier work of the balance within general ontology is, in his own terms, rather Pickwickian. But we have also seen that, especially in Kant and the Dynamics of Reason, Buchdahl seeks to establish distance between his position and subjectivism. We might hope that we will discover a more convincing account of objectivity in the contrast he draws in the later work between general and systems ontologies. There is no balance between the activity of reason and the empirical world into which it introduces order as a system of empirical laws. For this reason Buchdahl considers systems ontology to be subjective. In contrast, there is a balance between the categories of the understanding and the matter that is taken up and formed as an object: There is however a difference between the two cases. The categories of the understanding (e.g. causality) are valid only if they apply in the context of a sensory manifold; Kant calls this their ‘constitutive employment’. Only in this context is the notion of an ‘object’ admissible; only thus can the understanding ‘create’, so to speak, its object. For reason the case is different . . . nothing as such ‘corresponds’ to the theoretical frameworks of science; the ‘unity’ which such theories ‘mirror’, is not ‘given’ but – as Kant ceaselessly reiterates – only ‘projected’.53

The objectivity of general ontology arises from the fact that it is applied to the sensory given. However, as the use of the word creates 59


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology suggests, the understanding is spontaneous and does not merely mirror an already existing order. The latter claim is, of course, correct, but does not entail that the understanding creates order. In the end, Buchdahl’s distinction is difficult to pin down. While reason ‘projects’, understanding ‘injects’ order into nature at the most general level of its constitution. ‘The understanding is said to proceed “spontaneously”, so that the feature of “objectivity” which Kant seeks to define as “possible” is secured by showing it to be “selfwrought”.’54 Buchdahl insists that a balance between transcendental structure and a posteriori given qualifies general ontology as objective. Yet on examination there seems to be at best a difference of degree between an understanding that injects order and a reason that projects it.55 The way in which this plays into the hands of interpretations such as Pippin’s and indeed Guyer’s and Strawson’s is by now clear and becomes even more explicit when Buchdahl declares: As with the understanding, reason must not ‘borrow’ unity from what is presented to it, but instead ‘impose it’. The explanation of this procedure in both cases is the same: one can extract only as much with ‘apodeictic certainty’ from ‘the world’ as is injected into it by the spontaneous procedures of understanding and reason.56

In this passage ‘imposition’ can be substituted for both ‘injection’ and ‘projection’ and it would appear that the crucial distinction between understanding and reason collapses, leading us to conclude that not only reason, but also understanding, necessarily imposes form on matter. Despite Buchdahl’s talk of balance between the a priori and the a posteriori, the distinction between general ontology and systems ontology is in danger of disappearing. While it is clear that Buchdahl intends to secure a meaningful sense in which there is an objective element within constitutive judgement, it is not clear that he can achieve this within the perspective he develops. Buchdahl recognises the problem only too well. In Metaphysics and the Philosophy of Science, Buchdahl remarks that it is very difficult to discover what Kant believes is ‘brought into being’ and what is ‘found’ or ‘given’.57 His response to this dilemma in Kant and the Dynamics of Reason is to conclude that: ‘the kantian world is entirely “enclosed” within its transcendental framework’.58 It would appear that, in the end, not even a Fichtean Anstoß sets experience in motion. Reinforcing Buchdahl’s failure to adequately address the given as 60


29/3/07

8:12 AM


Formalism and the Circle of Representation given is his resistance to the philosophical relevance of applicability. He denies that Kant is in any way motivated by the ‘empirical question whether what we observe to be the case . . . is in fact the case’.59 He says this can only count as trivial in contrast with his project of establishing the possibility of cognition of objects in general.60 Buchdahl is, of course, correct that Kant is not directly concerned with empirical validity, but what he underestimates is that the validity of the formal structure of objects in general will only be established once Kant has shown that the latter applies to empirical objects. Admittedly, he only ever addresses this question at a general level, but the formal framework must be shown to be the form of matter and this cannot be achieved if the latter is merely viewed in a Pickwickian fashion. Kant is committed to an extra-mental given, as I will argue in the next chapter, and unless he can show how the latter is taken up through the forms of experience, these will only count as logical forms and not as forms of experience.61 The fact that matter is empirical and that we are not affected by things-in-themselves, does not rule out that what is given in experience is genuinely external to the subject. Kant holds that we are affected by something given in experience and this is not merely a technical issue. Admittedly, the otherthan-self can only affect us insofar as we have a capacity to be affected, but Buchdahl focuses on that capacity at the expense of what affects us and, moreover, characterises the object as rooted in spontaneity rather than, equally, in receptivity. He thus denies that the question of affection is a genuine philosophical question. What we have found in our investigation of Buchdahl’s account is that he establishes a powerful formalist interpretation of Kant’s epistemology. This version of formalism establishes the need for a reciprocity or balance between form and matter. However we have also discovered that, in the last instance, the objective side of the balance achieved by his version of formalism falls back into a monism of form or subjectivity. This has encouraged Pippin and others in their critiques of Kant’s supposed subjectivism. It remains to be seen whether there is an alternative elaboration of formalism that does not end up in the same cul-de-sac.

II Allison’s Methodological Formalism Allison’s Kant’s Transcendental Idealism has transformed Kant studies in the English speaking world and beyond by rediscovering a Kant not covertly dependent on a hidden noumenal world of 61


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology things-in-themselves.62 Allison establishes the human focus of critical philosophy and shows how the latter is still of great interest for current debates on the ‘transcendental’ in general. Allison shares much with Buchdahl’s formalist reading, insisting that the a priori must be understood as the form of experience and not as a self-standing apparatus. In Kant’s Transcendental Idealism, Allison enters into neither analysis nor criticism of Buchdahl, simply appealing to his authority on a number of occasions.63 A notable similarity between their readings is the way in which they interpret the relation between the ‘Transcendental Deduction’ and the sections that follow on from it. For both, the full story of the legitimation of the categories for experience is only to be found in the ‘Analytic’ as a whole. Allison’s particular take on Kant’s epistemology is to insist that the transcendental turn uncovers the ‘epistemic conditions’ of experience. These provide the fundamental framework without which experience would not be possible. I will argue that although it may seem that Allison engages in what Pippin calls an ‘independent analysis of rules for knowing’, his commitments are more complex.64 Importantly, Allison does not adopt Buchdahl’s talk of imposition, an interpretation he ascribes to the ‘standard picture’ for which Kant counts as a phenomenalist.65 Allison favours an analysis of epistemic conditions rather than one of the conditions of the possibility of experience because of the way in which the former focuses attention on the conditions for knowledge in contrast to logical, psychological and ontological conditions.66 It is straightforward to establish that Allison’s epistemic conditions should be construed as formal in status. In an early section of Kant’s Transcendental Idealism entitled ‘Transcendental Idealism as Formal Idealism’, Allison remarks that Kant notes in at least two places that formal idealism is an appropriate alternative term for his approach.67 Allison goes on to say that Kant would have done better to have followed his own recommendation as this may have discouraged the phenomenalist standard picture that has been so prevalent in interpretations of Kant. He continues: ‘Kant’s formal idealism is “formal” in the sense that it is a theory about the nature and the scope of the conditions under which objects can be experienced or known by the human mind.’68 This short passage holds within it the most important elements of the formalist status of Allison’s interpretation. The Copernican turn is a formalist one to the epistemic conditions of experience and, as 62


29/3/07

8:12 AM


Formalism and the Circle of Representation such, necessarily turns our attention to the role of mind in experience. However, Allison insists that the mind is to be understood logically and not psychologically. At this stage of his argument the importance of the relation between form and mind is clear, while its specific nature is less so: . . . behind Kant’s formal idealism, lies a principle that is implicit in the Critique as a whole, but is nowhere made fully explicit: that whatever is necessary for the representation or experience of something as an object, that is, whatever is required for the recognition or picking out of what is ‘objective’ in our experience, must reflect the cognitive structure of the mind (its manner of representing) rather than the nature of the object as it is in itself.69

The question arises, what is meant by the claim that the objective reflects the structure of the mind? Is the object other than the mind or merely a reflection of it? Admittedly, as Allison says in this passage, the object is not in-itself, but this does not eliminate the possibility that the object is empirical and yet still external to the mind. The question is: in what way can objectivity be dependent on subjectivity without Kant’s account falling back into the trap of impositionalism? Later, Allison emphasises the centrality of mental form for his epistemic or methodological reading thus: Once again, then, the point is that the appeal to the formal, a priori conditions of human experience and their characterization as ‘epistemic’ are the defining features of Kant’s idealism. The position is idealistic because, as we have seen, it grants to these conditions the function of defining the meaning of ‘object’ or equivalently, of determining what can count as ‘objective’ for the human mind.70

Epistemic conditions are formal in status and establish the conditions of objectivity. It is thus that Allison insists on the indexing of objectivity to the mind, without falling into psychological introspection. Objectivity does not so much mirror subjectivity as depend on the latter for its conditions of possibility. Nevertheless, it is still not clear in what sense objectivity requires that there is something outside the subject. If it does not, then the spectre of impositionalism – with its attendant inattention to the affective or aesthetic side of experience – returns. Reinforcing the equivalence between forms and epistemic conditions, Allison tells us: ‘[c]orrelatively, “form” must be taken to mean condition, while “matter” means that which is conditioned or determined by the form’.71 Were there any doubt, he sums up his account 63


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology of the sensible conditions of experience thus: ‘[i]n the last analysis then, everything turns on Kant’s claim that the representation of space functions as a form or condition of human experience.’72 The epistemic conditions comprise the forms of intuition and the categories of the understanding. Distancing himself from orthodox interpretations – and, importantly, anticipating Longuenesse’s extended defence of the relevance of the ‘Metaphysical Deduction of the Categories’ for the Critique as a whole – Allison argues that this section counts as a genuine first step in Kant’s argument. The ‘Table of Judgements’ presents the ‘conditions of all human thought, for all such thought is judgemental, and they are the conditions of judgement’.73 Judgemental functions require certain concepts, but the objective or empirical reality of these concepts requires the ‘Transcendental Deduction’.74 Having thus rehabilitated the beginning of the ‘Analytic’, Allison continues to show how its full scope is necessary if Kant is to establish that the categories are the conditions not just of thought but also of experience. Two stages of the ‘Deduction’, the ‘Schematism’ and the ‘Principles’ chapters, are all necessary for this. While the ‘Schematism’ chapter establishes that a schema ‘furnishes a meaning condition for a corresponding pure concept’, the ‘Principles’ chapter establishes that the same schema provides ‘the conditions of the determination of appearances in time and thus of the possibility of experience’.75 The second Analogy and its transition from ‘the subjection of the perceptions to the rule to the subjection of the perceived event’ is the ‘culmination’ of the ‘Transcendental Analytic’.76 Allison follows Buchdahl in favouring the weak interpretation of this portion of the text, arguing that Kant’s intent is only to show that there is some antecedent condition in any succession that counts as an event and not the stronger claim that other such events will follow the same pattern.77 Thus concludes Allison’s consistently formalist and hierarchical reading of Kant’s epistemology as presented in the ‘Analytic’ of the first Critique. We have already seen the problems Buchdahl falls into through his insistence that the given has a merely Pickwickian status. Despite the temptation to conclude that Allison’s focus on epistemic conditions amounts to what Pippin calls ‘an exclusive, foundational concern with our “mode of knowledge” ’, there are reasons for suspecting that this is not Allison’s intention. 78 He contrasts his position to that of the phenomenalist interpretation, which he calls the standard picture, for which an appearance is equivalent to a mere representation.79 64


29/3/07

8:12 AM


Formalism and the Circle of Representation Empirical realism commits Kant, he argues, to the position that ‘our experience is not limited to the private domain of our own representations, but includes an encounter with “empirically real” spatiotemporal objects’.80 Explaining the significance of the Copernican Revolution’s insistence that objects must conform to our knowledge, he comments: ‘But the mind-dependent objects of philosophers such as Berkeley and Hume (ideas and impressions) can no more be said to “conform to our knowledge” than can a humanly inaccessible “object” such as Lockean real essence’.81 Allison’s point is that a merely ideal object would not conform to our knowledge, simply because it would already be internal to the latter. Epistemic conditions are the mental conditions that make the representation of an object possible, and thus render the latter knowable. Mind supplies the structure or form of objects, not their material or substance. Epistemic conditions are, thus, not intended as an independent alternative for, but rather make possible an intentional reference to objects. Nevertheless Allison’s relative inattention to the affective side of Kant’s project – a topic that Kant also insufficiently elaborated – has the result that the epistemic conditions seem to promise a neutral and self-standing apparatus. Objectivity is explicated wholly with reference to epistemic conditions. What is ‘“objective” in our experience, must reflect the cognitive structure of the mind (its manner of representing) rather than the nature of the object as it is in itself’.82 This is unquestionably true, for Kant holds that an object that can be known must conform to the structure arising from our minds. But although it sometimes sounds as if this is not only the necessary but also the sufficient condition of objectivity, I will show in the next chapter that Kant’s position is rather more complex. In the early stages of the original edition of Kant’s Transcendental Idealism, there is no treatment of the material condition of objectivity. Allison’s view, quite simply, is that an object is whatever conforms to the mind’s conditions (both sensible and intellectual) for the representation of it as an object’.83 If matter has a necessary role to play in objectivity, he gives us little in the way of a clue as to what this would involve. However, in the final section of the first edition Allison discusses the question of affection in relation to the thing-in-itself and the transcendental object, while in the second edition he treats this topic in the revised first part of the book. 84 In both editions his account is based on a distinction between two levels at which affection can be approached within the transcendental perspective. 65


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology Empirically formed matter is the result of affection: it therefore cannot be also the latter’s cause or ground. To consider the ground of matter is to view it as not yet empirically formed, even though we cannot know or experience it as such.85 When we abstract from experience in this way, we are thinking of an indeterminate affect that gives rise to sensation insofar as it is taken up through sensibility. Allison concludes that it is justifiable to refer to these material conditions of human cognition methodologically as the ground or cause for the matter in experience.86 This establishes a distinction between empirical matter, which is always already formed, and its ground that counts as transcendental matter.87 We only think of the latter insofar as we abstract from the empirical character of an object and consider the empirical object as if it is unconditioned by the forms of space and time or the categories of the understanding. Allison argues that many of Kant’s discussions of the transcendental object can be understood as a treatment of the matter that is presupposed in any empirical object. But instead of treating matter in a transcendent fashion as a thing-in-itself, a transcendental approach allows Kant to consider what is only ever experienced in relation to epistemic conditions as if it were not constrained by the epistemic conditions that are characteristic of anything represented by us. This thought experiment, whereby matter is abstracted from empirical experience, makes use of the notion of the transcendental object.88 This development reinforces other evidence already mentioned above that Allison does not intend to merely provide an independent analysis of rules for knowing. A philosophical reconstruction of our affection by the object is expressed by reference either to the object considered as in itself or as the transcendental object.89 This means that affection can be recognised as the ‘necessary (material) condition of the possibility of experience, and in this sense as part of a “transcendental story” ’.90 Experience is only possible insofar as we are affected by something. It is therefore a necessary condition of a transcendental reconstruction of experience that the something in question is viewed as a ‘ “something in general = x” ’ that is, as the transcendental object.91 This account of affection makes a crucial adjustment to the significance of Allison’s account of epistemic conditions. In addition to the latter, which are mental in status, there is a necessary material condition. Matter must be given to us if experience is to be possible. While matter can be thought in abstraction from experience and is not reducible to epistemic conditions, it is nevertheless only known in rela66


29/3/07

8:12 AM


Formalism and the Circle of Representation tion to the latter. The material condition of experience always precedes any experience or reconstruction we can make of it. Nevertheless this does not render it extraneous to the transcendental system, but as Allison insists, makes it ‘a necessary condition of a transcendental account’.92 As we saw in the previous section, Buchdahl insists that Sachheit, or the fact that something is given, is a condition of experience. However he denies that this is of philosophical significance, saying that matter is only ever empirical. Allison would, of course, agree if this were taken to mean that we are only ever affected by appearing objects and not, indirectly or otherwise, by things in themselves. Nevertheless the advance in Allison’s account is that he is willing to concede the need for a philosophical consideration of the limits of experience. We must be able to transcendentally reflect on matter as given on the basis of a ground not conditioned by our sensibility. Recognition of this is not only philosophically possible, but necessary to a transcendental account. Nevertheless, Allison’s relative inattention to the role of affection, especially in the early chapters of the first edition of Kant’s Transcendental Idealism, has encouraged the view that the a priori form of experience can be discussed without any reference to the affective side of experience. Even when he eventually addresses the material element in experience, it is treated as simply the other side of the formal story. This is particularly marked in the new edition where he describes the claim that matter has a transcendental ground as an ‘analytic claim’.93 Is there any sense in which form must anticipate matter or stand in a dynamic relation to the latter? Allison does not raise any of these questions. That Allison is committed to the material side of the transcendental story has been established, but as the material given is merely an analytic entailment of epistemic form, it is questionable as to whether he has, in fact, escaped the charge that he offers nothing but a reflection on ‘our mode of knowledge’. If he is to show how the latter intentionally grasps an object, he would have to provide more than a merely analytic account of the relation between form and matter. This would entail focusing on the way in which epistemic conditions take up the affective given. My suggestion will be that a deeper examination of the subjective faculties reveals how the subjective conditions of experience provide access to extra-mental objects.94 Allison, like Buchdahl before him, seeks to establish that the formal framework of experience is not psychological. However his determination to avoid any trace of 67


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology subjectivism, leads to a failure to adequately explore the subjective conditions that have now been established qua epistemic conditions as the conditions of objectivity. Allison rejects Strawson’s view that formalism is inextricable from Kant’s use of the idiom of the faculties. A methodological consideration of epistemic conditions is meant to avoid any suggestion of experience being ultimately grounded in a mental or psychological source. Allison clearly shares Buchahl’s view that it would be preferable to adopt a more neutral or logical vocabulary than Kant does. Nevertheless, Allison does not reject the value of faculty talk outright – his answer to the question of whether there is any justification for undertaking an examination of the subjective side of cognition being difficult to pin down. In Idealism and Freedom, which gathers together articles originally published between 1986 and 1995, Allison defends Kant’s transcendental psychology or the ‘subjective deduction’ as ‘part of the project of justifying or establishing the objective validity of the categories’.95 Arguing against Patricia Kitcher’s empirically oriented defence of ‘faculty talk’, Allison says that the subject of transcendental psychology is neither phenomenal nor noumenal, being rather, using an expression introduced by Gilbert Ryle, ‘systematically elusive’.96 This suggests that it may be defensible to appeal to faculty talk, albeit indeterminately. In another article where he discusses his teacher Aron Gurwitsch, Allison highlights what he sees as the negative results of focusing on the subjective apparatus of cognition. He argues just as strongly against the phenomenological orientation of Gurwitsch, as against the empirical materialism of Kitcher. His major disagreement with the former is that despite Gurwitsch’s recognition of the need to read the three stages of the A ‘Deduction’ methodologically – that is, as outlining three conditions of experience – he still tends on occasion to treat apprehension as if it were ‘a unique form of pre-conceptual awareness that does not involve the categories’.97 This would suggest a pre-conceptual order ‘ratified’ by the categories.98 Allison strongly contests this, saying that through the introduction of the rule of the category we are able to represent ‘a distinctive objective order through these perceptions, not an order of them’.99 Gurwitsch’s focus on the aesthetic component within Kant’s epistemology has, he thinks, severed what must stay in necessary conjunction, namely the intuitive and conceptual elements of cognition. A focus on the more subjectively oriented A ‘Deduction’ results in 68


29/3/07

8:12 AM


Formalism and the Circle of Representation the undermining of the methodological approach Gurwitsch established. While in these two quite differently oriented critiques Allison argues against taking faculty talk as substantive – that is, as referring to empirical mental entities or pure acts – he does not rule out that there is a valid way of talking about the faculties as formal. Indeed, his response to Kemp Smith’s charge that Kant confuses two conflicting notions of form suggests that he may be in favour of this.100 Allison draws our attention to an ambiguity in Kant’s talk of the form of intuition as implying both the ‘form of intuiting’ and the ‘form of the intuited’ and defends Kant’s suggestion that space as the form of intuiting outer objects is the subjective formal source of the manifold.101 Nevertheless Allison insists that the ‘Transcendental Aesthetic’ is principally concerned with the form of the intuited, or the objective side of the formal analysis of intuition. It has become clear that Allison need not merely offer ‘an exclusive, foundational concern with our “mode of knowledge” ’.102 However, just as his account of affection is underdeveloped as to the relation in which the formal conditions of experience stand to the material given, correspondingly he is relatively silent on how a subjective representation takes up an affect. While Allison’s account does not enclose us in a circle of representation, it also fails to explain how representations could stand for something other than themselves. Interestingly, it may be the relative inattention he pays to the subjective side of experience that results in the enduring suspicion that formalism somehow misses the extra-mental status of the object.

III Longuenesse’s functionalist formalism: the appropriation of sense and imagination Béatrice Longuenesse’s Kant and the Capacity to Judge directly addresses the question of the relation between representations and the objects of representations. 103 Her claim that the ‘Table of Judgements’ in the Critique of Pure Reason offers a key for answering this question is an important and challenging addition to the recent literature. I will show how her account, focusing on the formal status of transcendental idealism, highlights what she calls the collaboration of the faculties necessary for knowledge. However, I will also argue that, despite her insistence on the dualist character of Kant’s project, the working through of her complex position finally puts at risk the affective side of dualism insofar as she insists that the 69


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology same function of understanding is the source both of the forms of understanding and of intuition. In ‘The Divisions of the Transcendental Logic and the Leading Thread’, Longuenesse rehearses the Leitmotiven of her earlier book.104 Her article gives a textual reading of the introductory passages of the ‘Transcendental Analytic’ where Kant presents the ‘Table of Judgements’ and claims that it is the ‘leading thread’ for the discovery of a priori categories governing our experience.105 Longuenesse argues that this often criticised claim, and the accompanying apparatus of logical judgement, is defensible if we read these passages as an analysis of the form of thought in general, a form that anticipates the categorial table. Longuenesse identifies the central role of form in transcendental idealism. From the outset she talks not only of ‘forms of sensibility’ but also of ‘forms of the understanding’.106 Whereas talk of intellectual capacities was already widespread, what is striking in Kant’s project is his transcending of the ‘quite standard empirical-psychological meaning’ of capacity.107 Indeed she insists that ‘the originality of Kant’s position lies in his attributing to each capacity a specific form, or mode of combination and ordering of representational data’.108 Instead of simply describing conventionally distinguished mental activities, Kant moves to the transcendental level insofar as he attributes to each capacity a specific way or mode of combining and ordering the elements of experience. Whereas intuition allows us to combine sensible data, understanding is the source of order among concepts. As Longuenesse expresses the situation, it is the collaboration between two kinds of forms that properly qualifies Kant’s logic as transcendental.109 Kant’s formalism is thus linked to faculty talk, which has been reidentified as transcendental rather than psychological. Longuenesse points out that Kant is operating with a quite different understanding of the relation between formal logic and psychology than would be the case in the twentieth and twenty-first centuries, especially in the analytical tradition: Note that for Kant – contrary to the view of logic prevalent after Frege –the opposition between logic and psychology does not rest on the fact that the former has nothing to do with our mental activities. For Kant the opposition resides, rather, in the normative character of logic, as opposed to the descriptive character of empirical psychology. But even the normative character of logic is, in the end, the expression of the normative capacity of our minds . . .110 70


29/3/07

8:12 AM


Formalism and the Circle of Representation Faculty talk is normative in that it analyses how thought should operate, rather than describing how we in fact think. Thus there is no need to equate Kant’s position with faculty psychology. This is an important corrective not only to Strawson, but also to Guyer and Pippin, whom I have argued suggest indirectly that Kant’s faculty talk contributes to his supposed impositionalism. Longuenesse here takes up a theme that Allison and Buchdahl try to avoid. Longuenesse remarks on the particular alliance between form and general logic insofar as the latter ‘deals with the mere form of thought’.111 The form of thought is set in opposition to the matter or object of thought, that is, its content. Formal logic is concerned only with the combination of concepts regardless of their content – in contrast to transcendental logic, which is concerned with the content of thought, at least at its a priori level.112 Longuenesse’s account of the status of content is persuasive. She insists that transcendental logic is concerned with the way in which our mode of sensibility forms empirically given data and not with the mechanics of physical affect.113 Longuenesse also says that both transcendental and general logic are pure insofar as they ‘borrow nothing from psychology’.114 We can conclude that both levels of logic are concerned only with the normative dimension of thought insofar as they deal with the a priori form of thought. Form implies ‘universality insofar as by virtue of this form alone concepts can be combined’.115 Prima facie it seems difficult to link this statement with aesthetic form. Admittedly Longuenesse’s account echoes Kant’s view that aesthetic form is the condition of an intuition’s unification under a concept. But the directness with which she links form and conceptualisation risks losing sight of what Longuenesse herself has recognised as a collaboration between two forms. The relative bias toward conceptual form is further evidenced by the way in which she contrasts conceptual function to sensible affection, finally claiming that ‘form’ and ‘function’ are used interchangeably, distinguished only by the latter term capturing the process of which the former is the result.116 She is clearly right in attributing to Kant the view that function and conceptual form are equivalent terms. The question is: is sensible form distinct from function? If form and function are equivalent and function is distinct from affection, then where is the place in her account for the latter? The suspicion arises that not just affection as a mechanical or physical process, but affectivity per se has been moved outside the bounds of her 71


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology philosophical horizon. Agreed, the form of sensibility is a priori and thus is not given, being rather the form of the given. Its a priori status reveals that it arises from the subject, but Kant makes a distinction between the unifying forms of the understanding and the receptive forms of sensibility. The forms of intuition make possible our capacity for sensibility, although, admittedly, this is a peculiar ability to take in or receive. The distinction between the two forms of experience is constitutive of the possibility of experience. The link Longuenesse makes between function and form is explained by her chief thesis of synthesis speciosa. Longuenesse argues that the same function of understanding is the source for the forms of sensibility and for the categorical forms. The initial analysis of that general ability of understanding – Vermögen zu urteilen or the capacity to judge – is laid out in the ‘Table of Judgements’. Longuenesse could have claimed simply that the ‘Table of Judgements’ anticipates the ‘Table of Categories’ by laying out a sketch of the general form of thought necessary for any judgement.117 It would then have been possible to combine this with the claim that judgement requires a source independent from thought, that is, our capacity for taking up the sensible given. But Longuenesse wants to say that even the latter is governed by the Vermögen zu urteilen. ‘His goal is to argue that these same functions by means of which judgments are formed (by analysis of the sensible given), first guide synthesis of sensible manifolds, with a view to analysis.’118 Synthesis relies on imagination or sensibility, but ‘in order to subsume a sensible manifold under concepts, one first needs to combine it in such a way that it is recognizable under concepts’.119 This bringing to concepts presupposes a unity of synthesis. In Longuenesse’s Kant and the Capacity to Judge this leading idea is identified as the thesis of synthesis speciosa. The categories play a double role: first preparing the sensible given for determination by introducing unity, and then determining what they have already unified: the categories have a role to play ‘at each end’ of the activity of judging: as mere ‘logical functions of judgement’, they play a role in guiding reflection of the sensible given with a view to forming empirical concepts, and for this they must first also guide synthesis in view of reflection. But as concepts, they are applied only in judgements of experience.120

Elsewhere it emerges that the goal of applying the categories within judgements of experience is what first gives the analysis of the logical 72


29/3/07

8:12 AM


Formalism and the Circle of Representation functions of judgement its role as ‘guiding thread’. This is a functionalist argument. The application of the forms of thought as categories within experience legitimates the presupposition of them at the outset of Kant’s argument. Were this the whole story it would be persuasive and count as a helpful expansion of Allison’s assessment of the role played by the ‘Metaphysical Deduction’. However, as the passage just cited reveals, Longuenesse’s position is more radical than this, for she is committed to the view that the categories in their initial guise as logical functions of judgement guide not only thought but also sensibility. Double accounting for the role of the categories is the key to Longuenesse’s account. It is also the key to grasping how she risks reducing dualism to a monism of a broadened understanding where the original distinction between two forms is traced back to one form or function, that of thought or judgement in general.121 The form of intuition has been transformed from the basis of our affective access to objects in the world into a subsidiary part of the internal apparatus of representation. The standard reading is that conceptual synthesis alone achieves the unity characteristic of determining judgement or knowledge. Nevertheless, this unity is only achieved insofar as sensibility combines the given within the forms of space and time. ‘Synthesis of apprehension’ is a distinct, though not a discrete, necessary condition of the unification of a sensible given under a concept that qualifies as knowledge. This account does not entail a commitment to intuition achieving a unity prior to that of the understanding. The holding together achieved in intuition is a component in the process of unification only finally completed in a concept.122 The standard account also holds that the forms of intuition, in Allison’s sense of forms of the intuited, are indeterminate until they are explicitly unified or determined under a concept.123 Longuenesse disagrees with this position. Crucially, she holds that the intuitions of the ‘Transcendental Aesthetic’ are already unified. 124 Kant, she claims, rarely speaks of form as the mere capacity to receive representations, and, retrospectively, in Section 26, we discover that in the ‘Transcendental Aesthetic’ intuitions already qualify as formal intuitions, that is, they have content. What in the ‘Transcendental Aesthetic’ are called forms of intuition are in fact formal intuitions and arise from the pre-formation of sensibility by the understanding.125 The claim that forms of intuition arise from the Vermögen zu urteilen derives from Longuenesse’s reading of Section 26 in the 73


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology second part of the B ‘Deduction’. The second step there, as she reconstructs it, is that the forms of space and time ‘being themselves unified intuitions, are under the transcendental unity of apperception, which is the source of the categories’.126 Longuenesse goes on to explain an ambiguity in Kant’s position in that he claims both that space and time are determined by the understanding and that they are prior to concepts.127 She resolves this apparent paradox in the following way. The forms of space and time are prior to any determinate concept and to the categories in their second role as reflective universal representations, but they are not prior to the first exercise of understanding as synthesis speciosa, that is of a self-affection of the understanding on our sensibility. This is an ingenious way of explaining what might otherwise appear to be a non sequitur at B160. Despite its initial plausibility, I believe it purchases the consistency of Kant’s position at too great a price.128 Longuenesse cites another passage in support of her reading: The same function, which gives unity to the various representations in a judgment also gives unity to the mere synthesis of representations in an intuition . . . The same understanding . . . brings by means of the synthetic unity of the manifold in intuition a transcendental content to its representations.129

Longuenesse reads this passage as establishing that the same logical function gives unity to concepts in a judgement and to the synthesis of intuition.130 In a sense this is incontestable, while in another sense it is highly contestable. Most readers of the Critique would agree that insofar as a sensible given is determined as to its unity, synthesis by the understanding is necessary. Many interpreters, although not all, hold that this determined intuition is what Kant means when he talks of a formal intuition as opposed to a form of intuition, where the latter is a pure term and the former a hybrid one.131 But most would not agree that the fact that the unity or unification of intuition can be traced to the understanding means that the latter is the source of the form of intuition. Instead, the usual interpretation would be that the application of the unifying power of understanding to the independent capacity for sensibility gives rise to knowledge. And there could be no independent capacity for sensibility, should the form of the latter ultimately resolve into an activity of understanding. The problem as to the identity of the faculties reveals a substantive problem about Longuenesse’s characterisation of Kant’s account of experience. If the understanding – even in its broader sense as intellect 74


29/3/07

8:12 AM


Formalism and the Circle of Representation in general – is ultimately the source of the form of sensibility, then it is difficult to maintain that the latter is affected by something other than the mind. We would be very close to saying that the data of sensibility has its source in the mind itself. Of course, Longuenesse strictly says only that space and time as unified forms fall under the understanding. However, she gives no serious role to the forms of intuition prior to unification with the result that the logic of her argument is that intuition does not so much cooperate with understanding, as count as an outpost of the latter. While Longuenesse is right to stress that for Kant there is no absolute given, there is just as much danger in suggesting that there is nothing other than constructions of the mind.132 The issue here is not the mechanics of physical affect, which Longuenesse is quite right to exclude from Kant’s transcendental account, but rather the very possibility of an affective element at the experiential or phenomenological level. For this we need a faculty, namely intuition, as the ground of the possibility of affection. And we still need to know how representations relate to objects at this level. Longuenesse does not altogether omit affection from her discussion, but she does underplay its importance. She rightly insists that Kant rearticulates ‘the manner in which things are given to us’.133 There is no brute given in Kant’s account of sensibility, but only a given that is formed in space and time. Longuenesse characterises the situation as one of a ‘transformation’ from outer affect to intuition.134 It would appear that outer affect is the physical cause of an experience, but in order for it to count as anything for us, it must be taken up as a representation through the synthetic power of imagination. And is this not Kant’s position? Longuenesse has argued that the form of sensibility is an exercise of the understanding, not that the matter of experience is. She is committed not to internal realism but to indirect realism.135 The primacy afforded to the understanding is at the level of representation and not of the represented. But Longuenesse’s thesis of synthesis speciosa turns sensibility towards the internal system of representation, giving no account of our capacity for receptivity of the represented. This is particularly evident in the contrast she draws between the position she attributes to Allison and her own. While, for the former, what she calls the ‘“internalisation of the object to representation” would simply mean internalisation to the point of view of (human) representing subjects’, her own preferred position is that ‘objects “as” objects-of-representation are appearances “in us” ’. She concludes that ‘things do exist in themselves outside us, 75


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology but we perceive them only by means of states of consciousness’.136 Allison’s epistemically formalist reading has been replaced by a newly tuned version of what he calls ‘the standard picture’. Objects, as things-in-themselves, stand hidden behind representations. Under this description, representations cannot count as epistemic conditions, that is, as our modes of access to objects. A further problem arises for Longuenesse’s interpretation of Kant’s epistemology in relation to his account of aesthetic judgement. Although addressing this issue may seem to take us onto quite different terrain, Kant claims that aesthetic judgement reveals the conditions of ‘cognition in general’, a claim which entails that a successful interpretation of his epistemology must also be able to account for aesthetic judgement, as discussed in the Critique of Judgement.137 Consideration of this issue allows us to reconsider the first stage of Longuenesse’s argument, namely, the view that apprehension already entails the determining power of understanding. Longuenesse’s thesis of synthesis speciosa entails that understanding ‘appropriates’ imagination in contrast to the view that the latter is the mediator between sensation and intellect.138 She speaks of the imagination in this context and not of sensibility, for it is the former in its guise as productive imagination that generates the forms of intuition.139 She concedes that there is also an activity of imagination that is not appropriated by the understanding, but goes on to say that Kant is only marginally interested in this ‘merely sensible/ associative/ fictional activity of imagination’.140 What Kant is really interested in, she believes, is the rule-governed activity of imagination at the behest of the understanding exercised in the interests of a cognitive goal. But in the Critique of Judgement, Kant focuses on a free exercise of the imagination that is autonomous and most definitely not appropriated by the understanding. Aesthetic judgement arises as a ‘harmony of the cognitive powers’ in which the faculties of imagination and understanding combine with one another in the absence of a rule.141 Moreover he argues that the ‘free lawfulness’ of the imagination, that is, the latter faculty exercised beyond the legislative authority of the understanding, reveals the subjective conditions of all judgement or ‘cognition in general’.142 Thus the unappropriated imagination turns out to be far from marginal to Kant’s project and to his cognitive project in particular. Later I will take up the relation between appropriated and unappropriated imagination. For the moment, I simply want to show that if Kant’s characterisation of 76


29/3/07

8:12 AM


Formalism and the Circle of Representation imagination as ‘free’ in aesthetic judgement is later shown to be defensible, then Longuenesse’s account of synthesis speciosa is put in further question. The contribution of aesthetic judgement is marked by the fact that it displays a free harmony between imagination and understanding and that it does not give rise to cognition. This would not be possible if imagination were already appropriated by understanding. If imagination is always already directed by the Vermögen zu urteilen, then it is difficult to see how we could employ our imagination without being oriented towards a cognitive goal. 143 Judgement (urteilen) is, for Longuenesse, cognitive judgement. In later chapters of this book I will challenge this equation.144 From her perspective, aesthetic apprehension can only be a failed or, at least, a not yet achieved cognition. Allison also suggests that Longuenesse is wrong to characterise reflective judgement as ‘failing to reach determination under a concept’.145 Longuenesse’s account differs in many important respects from the impositionalist one: namely, in her sensitivity to the relations between the faculties, in her sympathy for formalism and her insistence on the reciprocity between matter and form, and in her subtle reading of understanding as involving different levels of operation – to name but a few. Nevertheless, the internalisation of the relation between mind and world effected by the thesis of synthesis speciosa indirectly brings about the same consequence as impositionalism. The understanding comes to be the source of order among all sensory data. In a sense she goes even further than impositionalism insofar as she makes sensibility not just the recipient of cognitive activity, but a colonised outpost of the latter. Longuenesse insists she is using the term ‘understanding’ in a broader sense than usual. The general capacity for judgement analysed in the ‘Table of Judgements’ is not simply that of the determining understanding. It is, rather, that of intellect in general, covering determinant judgement (‘understanding’ in a narrow sense), reason and judgement. This distinction, she thinks, will prevent her from falling into the trap allegedly fallen into by Kant himself, that is, of making the whole project of the first Critique subservient to determinant judgement or understanding in the restrictive sense.146 Longuenesse hopes to open up the breadth of intellect and the interconnections between different branches of it. One example of her success in this project is the way in which she has revealed the necessity for reflection within the scope of determining judgement.147 Nevertheless the synthesis speciosa argument does not so much 77


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology achieve her intended goal as put it at risk. While not all judgements are cognitive, they are all governed by the Vermögen zu urteilen, which sets a broadly cognitive end for experience.148 Longuenesse’s goal of identifying the full range of judgement could be attempted in another way, by contrasting between a narrower and wider sense of reason, where the latter includes the whole range of the three Critiques.149 Cognition would then be situated within the systematic framework of reason without necessitating that all judgement is cognitive. This would also allow for what I call a pluralist account of the faculties in which they express distinct orientations open to us in experience while only ever being experienced as cooperating within experience.

Conclusion I hope to have established in this chapter that Buchdahl, Allison and Longuenesse each offer a highly sophisticated defence of Kant’s formalism. In their subtly different accounts, they rebut the charge that Kant offers a merely subjective epistemology, resting on a questionable faculty psychology. However, I have also argued that, due to their inattention to the role played by affect within experience, they risk falling back into impositionalism. Buchdahl and Longuenesse, in particular, are in danger of suggesting that Kant is trapped in a circle of representation, while Allison gives an insufficiently dynamic account of the relation in which matter stands to form. In the next chapter I commence my own account in which the affective or aesthetic side of experience is shown to be necessary for formalist idealism. In later chapters and in contrast to Buchdahl and Allison in particular, I will argue that it will only be possible to establish that formalism escapes the charge of subjectivism once the subjective element in knowledge is given its proper due and revealed as a necessary part of the project of establishing objectivity.

Notes 1. Gerd Buchdahl, Metaphysics and the Philosophy of Science: The Classical Origins Descartes to Kant (= MPS). 2. See Béatrice Longuenesse, Kant and the Capacity to Judge (= KCJ), p. 223, n. 21. It is also worth noting that Graham Bird published a quite alternative interpretation to that of Strawson in 1962. See his Kant’s Theory of Knowledge. Bird rejects the two worlds interpretation of Kant and should have been recognised as a viable alternative to Strawson. While Allison cites Bird favourably, he does not enter into a 78


29/3/07

8:12 AM



3. 4.

5. 6.

7. 8. 9. 10.

11. 12. 13. 14. 15. 16. 17.

18. 19.

20. 21. 22.

detailed exchange with an account he considers to be close to his own, but from which he differs ‘on many issues’. See Allison, KTI, p. 333 n. 9; see also p. 334, n. 12. Buchdahl does not reference Bird in MPS and neither does Longuenesse. For illumination on how Buchdahl uses the term ‘methodological’, see MPS, pp. 511–12. My reconstruction of Kant’s broader epistemological argument in Chapters 6 and 7 is deeply influenced by what I call Buchdahl’s ‘hierarchical’ reading of Kant. Buchdahl, MPS, p. 483, n. 1. Buchdahl is comparing Kant’s position to that of his predecessors discussed in previous sections of MPS. For Buchdahl’s account of the Second Analogy, see MPS, pp. 648–51. For his account of the relation between transcendental and empirical analyses of causality and his critique of competing positions, including Strawson’s, see pp. 651–65. Buchdahl, MPS; see pp. 489, 490 and 494. For a discussion of the relation between these, see Buchdahl, MPS, pp. 501–2. Buchdahl, MPS, p. 501. For an extended discussion of the relation, see pp. 651–65. Buchdahl, MPS, pp. 592–3. See also his Kant and the Dynamics of Reason (= KDR), pp. 183–4. It is Buchdahl’s view that regulative principles are not needed for the empirical experience of an object, but only for the lawful relations between empirical objects. Contrast my account in Chapter 7, pp. 254–5. Buchdahl, KDR, p. 56. Ibid., p. 56. Ibid., p. 170. Ibid., p. 56. Ibid., p. 205. Ibid., p. 204. Buchdahl, MPS, p. 494. Note again that, for Buchdahl, general ontology establishes the conditions of possibility for an individual material thing. I will argue in Chapter 7 (pp. 249–55) that full determination of the possibility of a material empirical thing is not supplied by constitutive principles alone. Buchdahl, MPS, p. 629 (discussed below). Buchdahl, KDR, p. 34 (my addition). EC is Buchdahl’s abbreviation for Special Ontology. This stands for ‘Explicative Component’. See KDR, p. 25, Fig. 1.5. Ibid., p. 90. Ibid., p. 89. Ibid., p. 34. I explain the idea of a framework yielding a phenomenology below. Buchdahl calls this the process of realisation. 79


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 23. 24. 25. 26. 27. 28.

29.

30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40.

Ibid., p. 34. Ibid., p. 62. Buchdahl, MPS, p. 638. Buchdahl, KDR, p. 62. Buchdahl refers to CPR, A 266, B 322. Buchdahl, MPS, p. 632, n. 4. Buchdahl, KDR, p. 118. Kant uses the term at CPR, A 143, B 182, in the ‘Schematism’ in a discussion of the schema of reality and, in particular, the material component in things, which he equates with things in themselves, Sachheit and reality. Kemp Smith, who translates Sachheit as ‘thinghood’, inserts a ‘not’, negating the sense of the original as stated in the Akademische Ausgabe version of both editions. Buchdahl remarks that this mention of Dinge an sich refers to the sense of the latter phrase as Transcendental Object, or To. Compare Allison’s interpretation of transcendental object as the material dimension in the object discussed in the next section pp. 65–7. Buchdahl, MPS, p. 639. In the first chapter of Charles Dickens’ The Pickwick Papers a row develops between Mr Pickwick and another member of the Pickwick Club, a speculative debating society. Other members are horrified that one member should accuse the other of being a humbug, while the accuser refuses to withdraw this insult. Nevertheless, calm is restored when Mr Pickwick’s opponent clarifies that he used the term in its Pickwickian, not its common, sense. Clearly, to label something Pickwickian is to say that its usage is limited to a restricted technical discussion and is about as relevant to our everyday concerns as Mr Pickwick’s seminal dissertation entitled ‘Speculations on the Source of the Hampstead Ponds, with some Observations on the Theory of Tittlebats’ – or, as some might say, it is ‘merely academic’! I give a quite different account of the philosophical importance of affection in Chapter 3. Buchdahl, KDR, pp. 53–103, especially p. 66. Ibid., pp. 79 ff. Buchdahl frequently mentions a question of balance. See, for instance, MPS, pp. 608 and 618. Buchdahl, MPS, p. 629 (Buchdahl’s emphasis). Ibid., p. 629. Ibid., p. 629. See my insistence on the two-sided character of sensibility in Chapter 3, pp. 105–8. Buchdahl, KDR, p. 51. Ibid., p. 44. He refers to CPR, A 255, B 311. Ibid., p. 73. Buchdahl refers to Bird, Kant’s Theory of Knowledge, pp. 68–9 and pp. 76–81. Compare Allison’s account of the transcendental object discussed on pp. 65–7. 80


29/3/07

8:12 AM


Formalism and the Circle of Representation 41. Buchdahl, KDR, p. 126. 42. Ibid., p. 118 (Buchdahl’s emphasis). 43. Ibid., p. 119. See also p. 132 on the ‘misleading empirical model of an external world acting on the subject’. There is nevertheless a ‘nonsubjective element’ in experience, namely that we find ourselves with such-and-such sensations or representations. 44. Ibid., p. 149. 45. Ibid., p. 75. Just to be clear, I include this characterisation of Fichte strictly to illuminate Buchdahl’s position. In no sense should this be taken as an attempt to characterise the relation between subjectivity and objectivity for Fichte. 46. Buchdahl, MPS, p. 639. 47. Ibid., p. 639. Buchdahl thus distinguishes his position from Strawson’s critique of faculty talk. 48. Ibid., p. 640. This distinction is extremely important for both Allison and Longuenesse. 49. Ibid., p. 640. 50. Below I will suggest that Buchdahl does not render this distinction as clearly as he intends. 51. Buchdahl, MPS, p. 593. In my own account, my aim is to demonstrate the intertwinement of the subjective and objective trajectories of critical idealism. 52. Ibid., p. 640. 53. Buchdahl, KDR, p. 170. 54. Ibid., p. 170. 55. Buchdahl, MPS, p. 640. See above. Moreover, I will argue in Chapter 7 that the system of nature has an objective trajectory insofar as it is necessary for empirical knowledge of objects. 56. Buchdahl, KDR, p. 170. 57. Buchdahl, MPS, p. 619. 58. Buchdahl, KDR, p. 146. 59. Buchdahl, MPS, p. 624. 60. Ibid., p. 624. 61. I develop this argument in Chapter 6. 62. Henry Allison Kant’s Transcendental Idealism was first published in 1983. A revised and enlarged edition appeared in 2004. It has not been possible to incorporate an assessment of the new edition in this work, but I believe that the issues addressed here are not substantially changed in the new edition. In other respects, however, the second edition offers so many developments in Allison’s thinking that Pippin’s remark cited on the dust jacket of the 2004 edition is just right, namely, that ‘the result is almost a new book altogether’. It is tempting to think that this is Allison’s B edition! 63. See Allison, KTI (1983), p. 97, on the indeterminacy of space (2004), 81


29/3/07

8:12 AM



64. 65. 66. 67. 68. 69. 70. 71. 72.

73. 74. 75. 76.

77. 78. 79. 80.

81. 82. 83. 84. 85. 86. 87.

p. 116; (1983), p. 231 on the second Analogy; and p. 234 on the importance of regulative principles. See Chapter 1 for a discussion of Pippin. Allison, KTI (1983), p. 4; (2004), pp. 38–42. Ibid., p. 11; (2004) p. 11. Ibid. (1983), p. 26. Allison refers to the appendix to the Prolegomena Ak. IV, 375, and to CPR, B 519. See (2004), p. 35. Allison, KTI, p. 26 (Allison’s emphasis); (2004), p. 35. Ibid. (1983), p. 27; (2004), p. 37. Ibid. (1983), p. 61. The Antinomies are given an extended treatment in the new edition, see pp. 357–95. Ibid. (1983), p. 106. Allison also recasts this argument radically in the 2004 edition, see pp. 122–8. Ibid. (1983), p. 111. Another passage important for establishing the formalist status of Allison’s interpretation is found on p. 106: ‘a form of appearance is a feature of the appearance in virtue of which its elements are viewed as ordered or related to one another in experience’. As noted above, this section is altered in the new edition. See (2004) pp. 122–8. Ibid. (1983), p. 122. For a revised account of ‘discursivity’ and the conceptual side of dualism, see (2004) Chapter 4, pp. 77–96. Ibid. (1983), p. 129; (2004), p. 156. Ibid. (1983), p. 195; (2004), p. 225. Ibid. (1983), pp. 226 and 216. Allison’s analysis of the Analogies is greatly extended in the new edition, where he endeavours to show the interconnectedness of all three principles of time determination. See (2004) pp. 229 ff. Ibid. (1983), pp. 230–1; (2004), pp. 255–6. Pippin, KTF, p. 23. See Chapter 1, p. 41. Allison, KTI (1983), pp. 5–6; (2004), pp. 4–6. Ibid. (1983), p. 7. See (2004), p. 14 ‘our intuition and, more generally, that of any finite cognizer, must be sensible, that is, receptive, resulting from an affection of the mind by objects’. Ibid. (1983), p. 30. For a revised version of the relation in which Kant stands to Berkeley, see (2004), pp. 38–42. Ibid. (1983), p. 27; (2004), p. 37. Ibid. (1983), p. 30. See (1983) Ch. 11, especially pp. 247–54. For the revised version in the new edition, see (2004), Ch. 3 pp. 50–73, especially pp. 64–73. Ibid. (1983), p. 250; see (2004), p. 67 ‘the object remains completely indeterminate’. Ibid. (1983), pp. 250–2; (2004), pp. 71–3. Ibid. (1983), p. 253. Allison prefers to refer to empirical matter and its transcendental ground, rather than transcendental matter. However, in 82


29/3/07

8:12 AM



88.

89. 90. 91.

92. 93.

94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107.

both editions he notes that Kant employs the latter expression on occasion. See, for instance (2004), p. 72. Allison’s account of the transcendental significance of the transcendental object can be compared to Gerold Prauss’s interpretation of philosophy’s role in relation to the natural sciences. For both Allison and Prauss, transcendental reflection allows us to grasp what cannot be understood at the empirical level. Prauss describes transcendental philosophy as the ‘non-empirical science of the empirical’. See Prauss, Kant und das Problem der Dinge an sich, pp. 205–27 and in particular pp. 212–13. Allison, however, distinguishes his position from Prauss’s, which he sees as rendering unintelligible the very concept of a non-empirical affection. See Allison, KTI (1983), pp. 250 and 364, n. 26; (2004), pp. 68 and 461. Allison, KTI (1983), p. 249. For a revised discussion of these, see (2004) pp. 51–64. Ibid. (1983), p. 249; (2004), p. 67. Ibid. (1983), p. 250. In the new edition, the ground of matter is aligned with things in themselves, see (2004), p. 72. This, however, is quite consistent with Allison’s account of the relation in which the transcendental object stands to things in themselves. Ibid. (1983), p. 250 (Allison’s emphasis); (2004), p. 67. Ibid. (2004), p. 72. See also his claim on p. 73 that all talk about things as they are in themselves, noumena and the transcendental object, that is, the ground of matter, counts as technical. This sounds very like Buchdahl’s claim that the given is strictly ‘Pickwickian’ in status. See Chapter 5 and subsequent chapters. ‘On naturalizing Kant’s transcendental psychology’, in Allison, Idealism and Freedom, p. 58. Ibid., pp. 65–6. ‘Gurwitsch’s interpretation of Kant’, ibid., p. 74. This article was first published in Kant-Studien in 1992. Ibid., see pp. 72 and 77–8. Ibid., p. 78. Allison, KTI, p. 97. Ibid., p. 97. Pippin, KTF, p. 23. See Chapter 1, p. 13. Longuenesse, Kant and the Capacity to Judge (= KCJ), p. 18. Longuenesse, ‘The Divisions of the Transcendental Logic and the Leading Thread (= ‘Leading Thread’). CPR, A 66, B 91, and A 67, B 92. Kemp Smith translates Leitfaden simply as ‘clue’. Longuenesse, ‘Leading Thread’, p. 131. Ibid., p. 131. 83


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 108. Ibid., p. 133. In this and subsequent passages, the emphasis is Longuenesse’s. 109. Ibid., p 133. 110. Ibid., p. 135. 111. Ibid., p. 134. Her reference is to A 54, B 78. 112. Ibid., p. 136. 113. Ibid., p. 136. 114. Ibid., p. 135. 115. Ibid., p. 136. 116. Ibid., pp. 139 and 143. 117. This is Allison’s position, as we saw in the previous section. 118. Longuenesse, ‘Leading Thread’, p. 151. 119. Ibid., p. 152. 120. Longuenesse, KCJ, pp. 199–200, n. 3. See also p. 196. 121. Understanding is ‘broadened’ because it now includes imagination and is even the source of the forms of sensibility. See Longuenesse, KCJ, p. 63, where understanding is identified as ‘the unity of apperception in relation to the synthesis of imagination’. 122. See my discussion of the conditions of unification of objects as presented in the A edition of the ‘Deduction’ in Chapter 4. 123. See, for instance, Allison, KTI, p. 97. 124. Longuenesse, KCJ, p. 221. Longuenesse compares her position to Heidegger’s. Indeed Heidegger already identifies the problem of the source of synthesis speciosa (see Heidegger’s Kant and the Problem of Metaphysics). But his solution is to deny that Kant subsumes imagination under understanding, preferring to construe the two faculties as standing in ‘the rich totality of a complex activity’. See especially pp. 67–8 (German edition p. 61). 125. Longuenesse, KCJ, p. 223. See also her interesting claim that the forms of intuition are really potential forms (p. 221). A very similar idea is to be found in Heidegger’s Kant and the Problem of Metaphysics, p. 149 (German edition p. 137) where he cites Kant’s statement that ‘space and time are the pre-formative forms [Formen der Vorbildung] in pure intuition’. He references Erdmann, Reflexionen II, 408, Kant’s Posthumous Works in Manuscript Form, Vol. V, No. 5934. 126. Longuenesse, KCJ, p. 214 (Longuenesse’s emphasis). 127. She is referring to Kant’s famous footnote at B 160–161. 128. I give an alternative reading of the crucial footnote at B 160–161 in Chapter 4. 129. Longuenesse, ‘Leading Thread’, p. 153. CPR, A79, B 105 (Longuenesse’s emphasis). 130. Longuenesse, ‘Leading Thread’, p. 153. 131. See Allison, KTI, p. 97. 132. Longuenesse, ‘Leading Thread’, p. 154: ‘so no object for a concept is 84


29/3/07

8:12 AM



133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143.

144. 145. 146.

147. 148. 149.

simply “given” ’. See, in this regard, the attack on the ‘myth of the given’ in John McDowell’s Mind and World. Longuenesse, KCJ, p. 213 (Longuenesse’s emphasis). Ibid., see pp. 213 and 220. Ibid., p. 20, n. 9. Ibid., p. 20, n. 9. For instance, Critique of Judgment, AA 218. See discussion in Chapter 5. Longuenesse, KCJ, p. 207. Ibid., p. 206. Ibid., p. 208. CJ, AA 218. CJ, AA 240, is a development of the account at AA 218. My account is in direct opposition to the interpretation offered by Avner Baz in ‘Kant’s Principle of Purposiveness and the Missing point of (Aesthetic) Judgements’, who relies on Longuenesse, among others, for his reading of Kant. See, in particular, Chapters 5 and 7. Allison, Kant’s Theory of Taste, pp. 353–4, n. 2. The reference is to Longuenesse, KCJ, p. 164, n. 47. See Longuenesse, KCJ, pp. 203–4, for a very interesting, acute critique of Heidegger’s reading of Kant. See also my discussion of Heidegger in Chapter 4. See, for instance, Longuenesse, KCJ, p. 112, n. 17. This is what encourages Baz’s reception of her interpretation. See Baz, ‘Kant’s Principle of Purposiveness’, p. 14. See discussion of the tripartite supersensible in the penultimate section of Chapter 8, pp. 299–302.

85


29/3/07

8:12 AM


3

Formal Idealism and the Aesthetic Condition of Experience

In the previous chapter I have identified a problem in the accounts given by some of Kant’s principal defenders of the relation between representations and objects. While Buchdahl, Allison and Longuenesse provide a robust defence of Kant’s formalism, each of their accounts fail to give an adequate account of the relation in which form stands to matter. I will now offer an alternative account of the relation between representation and material given. In the first section of this chapter, I discuss how Kant’s characterisation of his epistemological project as amounting to formal, as opposed to material, idealism reveals his commitment to an extramental world of objects. Only the form of experience is initiated by our minds. In the second section, I turn to Kant’s ‘Copernican revolution’, which has often been taken to be a corner stone of his commitment to the view that mind imposes order on the world. I suggest that it need not be read in this way, because of a limitation on the range of its scope and also because it comprises two stages, the first of which concerns our receptivity to an affect given in experience.1 Next I examine some of the passages in the ‘Transcendental Aesthetic’ that most invite an impositionalist reading and suggest an alternative perspective according to which representation is our mode of access to an extra-mental given. Having established this in principle in the third section, in the fourth I argue that the account of affectivity such a perspective would require is in fact provided in the opening paragraphs of the ‘Transcendental Aesthetic’. In the fifth section I give a resumé of the complex identity of receptivity that has emerged. Sensibility is dual-faced: its mental component being the formal condition of our capacity to be affected by something given in experience. This is the first and crucial step in taking seriously the subjective dimension of Kant’s epistemology, while not falling into the impositionalist readings examined in Chapter 1 and insufficiently 86


29/3/07

8:12 AM


Formal Idealism and the Aesthetic Condition of Experience parried by those interpretations discussed in Chapter 2. Nevertheless, Kant did not provide an adequately forceful account of the affective side of sensibility, often leaving the impression that the a priori forms of intuition were not so much modes of receptivity, as that they actively shape matter. This would, however, be a curious result as sensibility is also frequently characterised by Kant as ‘passive’. In fact, the truth lies somewhere between these extremes in the receptive capacity to be open to something other than ourselves.2

I Transcendental Idealism as Formal Idealism: The Refutation of Material Idealism In a footnote Kant makes the following remark about transcendental idealism: ‘I have also, elsewhere, sometimes entitled it formal idealism, to distinguish it from material idealism, that is, from the usual type of idealism which doubts or denies the existence of outer things themselves.’3 The final phrase in this passage may appear to suggest that Kant here commits himself to the existence of things-in-themselves (Dinge an sich), that is, things regarded without any reference to human experience of them. However there is no mention of Dinge an sich, but only of außere Dinge selbst. The ontological status of these outer things is left open by the latter phrase and does not necessarily commit Kant to a two-worlds ontology of appearances and things-inthemselves.4 Kemp Smith omits the subsequent sentence: ‘In many cases it seems advisable to use this [i.e., ‘formal idealism’] rather than the latter expression [i.e., ‘transcendental idealism’], in order to avoid all misinterpretation.’5 At times even Kant seems to lose sight of this important advice. My intention is to rearticulate the formal status of transcendental idealism and show how we can defend the latter against the charges of impositionalism and subjectivism. Kant’s point in this crucial note is that, in contrast to his own position, material idealism ‘doubts or denies’ that there are independently existing objects. Who are Kant’s opponents here? Certainly he has within his sights Berkeley, who denied the external existence of objects and insisted that the latter are only ideas in the mind.6 Formal idealism, in contrast, does not deny the extra-mental existence of objects. But no more is it committed to the extra-experiential existence of things-in-themselves. It must be that ‘outer things themselves’ in the passage under consideration refers to objects independent of the mind, but nevertheless experienced by us. These ‘außere Dinge selbst’ are accessible to the mind, but are not projected by it. Formal 87


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology idealism preserves a crucial role for the mind as objects are only identified as such insofar as they have a certain structure or form, which is mind-dependent. Formal idealism is, however, contrasted from the material variety insofar as the former does not reduce all experience to the activity of mind. In the ‘Refutation of Idealism’, ‘material idealism’ refers not only to Berkeley but also to Descartes.7 Kant contrasts the dogmatic idealism of Berkeley to the problematic idealism of Descartes. The latter species of idealism is sceptical, or at least uncommitted as to the independent existence of extra-mental objects. It ‘doubts’ though does not ‘deny’ the existence of outer things. All that is certain is the Cogito. It is only with the additional premise of a beneficent deity that our certainty in the existence of external objects can be reasserted. In contrast, for Kant, any certainty that we have in our activity of thinking must be based not only on a concept but also on an intuition. In order to experience myself as thinking, I must do so over time. And, to experience myself as temporal, is to mark out my experience of self against a spatial experience of external objects. Thus, Kant concludes, we simply cannot maintain a certainty with regard to the self and be sceptical about the existence of external objects. We cannot have one without the other and it is clear that, for Kant, the only reasonable conclusion is that our common sense faith both in the existence of the thinking subject and in external objects is reliable.8 Formal idealism is thus distinguished both from the Berkeleian and Cartesian versions of material idealism insofar as it starts from a commitment to the existence of extra-mental objects. Something must be given to the mind if experience is to be possible.9 However we have seen in Chapter 1 that had Kant consistently talked of formal instead of transcendental idealism, his critics would have been no more convinced. It is the very formal status of his method that – for Pippin, Guyer and Strawson – results in the conclusion that he is committed to subjectivism or the imposition of mental structures on the external world. In contrast to what they see as a reduction of empirical experience to an order projected on that world by the form-giving mind, we could say in our terms, not theirs, that they insist on the aesthetic or the given.10 For each of them, Kant fails to account for what is outside the mind and slips back into a tooeasy model of ‘impositionalism’. Meanwhile the authors we examined in Chapter 2 admit the importance of mind in Kant’s method, while insisting that what is in question here is the logical and not the psychological activity of mind. 88


29/3/07

8:12 AM


Formal Idealism and the Aesthetic Condition of Experience Mind gives structure to objects, but it does not generate them. Thus far, the approaches taken by Buchdahl, Allison and Longuenesse tally with Kant’s distinction between formal and material idealism, as I have read it. As I have argued, however, none of these three commentators focuses sufficiently on the relation in which form stands to the affective side of experience. This is certainly not to say that they qualify as material idealists in either of Kant’s senses. They neither deny nor doubt the external existence of things. Rather, they start from the presupposition that the aesthetic dimension of experience is unproblematic, philosophically speaking.

II Why an impositionalist interpretation of the Copernican Revolution is not necessary It cannot be denied, however, that Kant’s presentation of his method at times invites the impositionalist reading developed by his critics. Famously, in the account of the Copernican revolution in philosophy, he says that there must be a reversal in the relation between knowledge and objects.11 In this passage Kant consistently uses Gegenstand rather than Object, suggesting strongly that he is concerned with the possible experience of extra-mental objects and not merely with constructions of thought. Indeed, no revolution would be necessary if he merely wanted to establish the possibility of access to the latter. If we are to attain anything worthy of the title of knowledge, then the latter cannot simply arise from the causal effect of objects on our minds. The status of the empiricist causal account of knowledge became explicit in Hume’s claim that causality is nothing other than a customary association of certain events we call causes with certain others we call effects.12 If this model were applied to the case of knowledge in general, there would be no necessary connection between object and representation and, as a result, our claims to knowledge would not qualify for the title. The link between representation and object that is sought cannot be derived from experience, which as Hume rightly concluded, gives us nothing but associative patterns. Kant’s alternative is to suggest that the only way of securing that our claims to knowledge do, in fact, grasp a necessary connection between object and representation is if the mind – or, more strictly, our cognitive capacity – introduces an a priori element. The reversal Kant achieves is not, however, that now the mind – or the a priori form arising from it – causes objects. Causal relations are to be understood strictly within the range of experience. Asking 89


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology the question as to the cause of experience is to pose an illicit, transcendent and ultimately vacuous question. A priori elements, we will see, provide the structure or form of experience and not the cause of the latter. Nevertheless, does this not mean that the mind imposes its own order on objects? We will see in our discussion of the structure of the ‘Analytic’ in Chapter 6 that this is a question that returns to haunt Kant at many levels.13 For now I intend to concentrate on evidence internal to Kant’s initial account of the Copernican revolution in order to show that at this early stage of his account, we need not leap to conclude that his position entails impositionalism. Having begun with the claim that objects must conform to knowledge, Kant proceeds to develop his insight at two levels. In effect, he separates out two conditions of knowledge, saying first that objects must conform to the faculty of intuitions: A similar experiment can be tried in metaphysics [i.e. similar to that of Copernicus’ reversal of the attribution of movement], as regards the intuition of objects [Gegenstand]. If intuition must conform to the constitution of the objects, I do not see how we could know anything of the latter a priori; but if the object (as object of the senses) must conform to the constitution of our faculty of intuition, I have no difficulty in conceiving such a possibility.14

This is the aesthetic condition of knowledge. Kant asserts that if we are to understand how we achieve knowledge of objects of the senses, then we should not construe those objects as the causes of our knowledge. It must rather be the case that the objects conform to our power of intuition. In the ‘Transcendental Aesthetic’ we will discover what Kant has in mind. If we are to come to attain knowledge of sensory objects, there must be something in us that allows the given in sensation to be taken up, that is, we must be capable of being receptive to that something. The second condition is that objects must conform to concepts: Since I cannot rest in these intuitions if they are to become known, but must relate them as representations to something as their object, and determine this latter through them, either I must assume that the concepts, by means of which I obtain this determination, conform to the object, or else assume that the objects, or what is the same thing, that the experience in which alone, as given objects, they can be known, conform to the concepts. In the former case, I am again in the same perplexity as to how I can know anything a priori in regard to the objects. In the latter case the outlook is more hopeful.15 90


29/3/07

8:12 AM


Formal Idealism and the Aesthetic Condition of Experience Knowledge, although not perhaps some lower level of consciousness, requires that we have not only an intuition, but also a concept. In the ‘Transcendental Deduction’ we learn that the forms of intuition are only the initial conditions of the combination of the sensible given which, in turn, facilitates its being unified by the concepts of the understanding.16 In the light of this, it is surprising that the second stage does not say that intuitions must conform to concepts. Intuitions give sensible content for concepts, while objects are the composite of intuitions and concepts. Yet in this passage Kant uses not only ‘object’ but also ‘experience’ as referring to the sensory given that has not yet been determined or unified by the understanding. Later I will suggest that Kant uses these terms in both a technical and a non-technical sense. Kant’s general point is that the extra-mental given in experience can only be taken up into consciousness and thus qualify as known by us insofar as we supply an a priori element that comes from mind alone. There are two elements of form, both aesthetic and conceptual, and they conjointly allow the unification of the sensory object under the rules of understanding so as to give rise to knowledge. Insofar as an object can be experienced – which for Kant is equivalent to its being knowable – then it must ultimately be subsumable under certain fundamental concepts that function as rules: For experience is itself a species of knowledge which involves understanding; and understanding has rules which I must presuppose as being in me prior to objects being given to me, and therefore as being a priori. They find expression in a priori concepts to which all objects of experience necessarily conform, and with which they must agree.17

This is the locus classicus of the impositionalist interpretation of Kant. Knowledge arises neither from our intuitions conforming to the constitution of objects, nor from our concepts conforming to objects. It can only arise insofar as objects conform to our concepts. The case seems to be clear. The mind introduces order into objects and it does so without any contribution from those objects. It seems difficult to avoid the conclusion that the mind imposes order on objects. Before considering that the impositionalist reading may simply be too quick, we should consider what Kant means by the term ‘object’ in the passage just cited. I have already noted that he uses the term indeterminately in anticipation of a later, more rigorous usage. But what in general does Kant intend by the term in this context? It would seem reasonable to assume that he is referring to experiential or empirical objects. We can establish this by a via negativa, for the force 91


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology of these passages is that objects fail to ground knowledge because they are empirical. A further illumination of the empirical status of objects comes later in the Preface in the course of a discussion of the limitation of knowledge in the interests of our practical or moral orientation. Kant makes the point that if our knowledge is of appearances and not of things-in-themselves, then a space is left open for the thought of the latter within the practical domain.18 We can never have knowledge of these speculative or moral entities, but we can have practical faith in them.19 By contrast, the objects of knowledge are appearances and thus empirical. In insisting that the objects that we know are ‘appearances’, Kant is saying that they must stand in relation to mind. They are objects that appear to us. Now insofar as they are marked by the representational relation in which they stand to mind, they share the formal structures characteristic of mind. Only insofar as they are formed by the structures introduced by our mental powers are they capable of counting as appearances. But to say as much is not to say that they are mere projections on the part of mind. They are syntheses of the material given, which arise from the unifying capacity of mind. Synthesis must still be effected on a given that precedes mental activity and as such synthesis is not a mere projection but rather arises from the intervention of mind within the material world. The form of objects, admittedly, arises from the mind alone, but only as conditions for synthesising the given. As I will argue in the next chapter, this synthesis arises as a cooperation of a plurality of different mental orientations in response to an extra-mental given. We now must consider a limitation on the Copernican Revolution, before moving on to insist on its full scope. The limitation is that Kant is here concerned only with the a priori element in knowledge. The ‘objects’ – be they things in themselves or appearances – would be incapable of supplying this dimension of knowledge. Things-in-themselves are alien to the cognitive process, while appearances count as the ‘undetermined object [Gegenstand] of an empirical intuition’.20 Neither is capable of supplying the element that makes possible the determining function characteristic of knowledge. Only an a priori element that arises from the mind makes possible determination or unification. But this a priori element is merely the form or structure of knowledge, that is, it provides only the initial conditions of knowledge and is not the latter’s sole source. Admittedly, Kant’s presentation of the Copernican revolution seems to suggest a less limited reading where knowledge arises from the mind alone. However, seen in conjunction 92


29/3/07

8:12 AM


Formal Idealism and the Aesthetic Condition of Experience with his characterisation of transcendental idealism as ‘formalist’, a differently nuanced interpretation becomes possible. The a priori forms supplied by the mind are the initiating conditions with respect to which any claim to knowledge must arise, but the a priori is operative – and indeed valid – only insofar as it is set in relation to what is given in experience. The empirical or a posteriori object arises from the combination of a priori and the given. The task of achieving empirical knowledge will require much more than the merely formal conditions, both aesthetic and conceptual.21 Recognising this limitation goes hand in hand with establishing the full scope of the Copernican revolution. If the structures introduced by mind are to bring about order in the extra-mental world, mind must be capable not only of unifying the given through concepts, but also of being receptive to what is given in experience. This is the first and aesthetic condition of knowledge that I identified above. Receptivity also has a formal structure, namely, the forms of intuition. But the forms of intuition allow us to take up something extramental. This hidden side of sensibility has often been underestimated, resulting in the view that the forms of intuition impose order rather than make possible our reception of or response to objects. Indeed, Kant’s statement that objects must conform to intuition seems to suggest just this. However, Kant’s account of sensibility is more complex, as I will argue later.22 For the moment I will simply claim that the first stage of the Copernican revolution would not be distinct from the second, if the forms of intuition were not receptive rather than spontaneous. And we will see that the defining characteristic of intuition is receptivity, that is, our capacity for taking in something given to us via our senses. If receptivity is a necessary component of knowledge, then it cannot simply be the case that mind imposes order on the world. Admittedly the concept brings unity into intuition, but both steps are necessary for knowledge to arise. This brings to mind another classical reference point in the Critique of Pure Reason: ‘Thoughts without content are empty, intuitions without concepts are blind.’23 Once we recognise the limitation of the Copernican Revolution to the a priori and its full range of reference, expressed in two stages, both aesthetic and conceptual, it is no longer necessary to read the passage as advocating impositionalism. Only the form of experiential objects comes from the mind and, moreover, aesthetic form is a capacity for receptivity to something given, not a projection on the world. Admittedly, it is possible to read the aesthetic side of 93


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology transcendental idealism as merely passive and not receptive in any meaningful sense, but I believe this should be resisted.24 The remainder of this work is devoted to bringing out how the formal structure of transcendental idealism makes possible access to the material but empirical given. There are at least three elements necessary for the cognition of an empirical object.25 First, something must be given to us in intuition; second, we must be receptive to what is given through the forms of intuition, space and time; and, third, an a priori concept is required for determination of the given. In his account of the Copernican revolution in philosophy, Kant has often been taken to be concerned only with the third element. And even though I have emphasised the necessity of recognising the second element in the first stage of the Copernican revolution, it must be admitted that at this stage of his discussion the first element is not positively addressed. This has the result that the complex structure of sensibility is not made fully evident, and, until it is, the distinctiveness of the aesthetic and conceptual dimensions of experience cannot be adequately grasped. Later in this chapter we will see that Kant pays more attention to the first element of knowledge at the outset of the ‘Trancendental Aesthetic’. It will then become more apparent that the three elements of knowledge are part of one complex story. The limitation of the Copernican Revolution to the a priori and its composition in two stages, both aesthetic and conceptual, has important implications for what Kant intends by ‘knowledge’. Insofar as knowledge has an aesthetic dimension, it is temporal and arises with regard to spatial objects. The conceptual dimension emerges as a task of determination of the determinable, that is, of appearances, which are identified as the undetermined.26 This means that while the Critique remains an examination of the a priori, it must also give some account of the way in which the latter provides the form of the a posteriori. Only the form of experience, or a priori knowledge, arises from the mind alone and even it must anticipate application to empirical instances. Empirical knowledge has an even more complex genesis.27 Were the impositionalist interpretation correct, the remainder of the Critique would merely be an exposition of the apparatus of mental projection. Contrary to this, I will argue that the Critique furnishes a lengthy argument for the complex operation necessary for the possibility of any concrete knowledge in the world. This is the reason why Kant has to develop his argument at a series of different levels.28 94


29/3/07

8:12 AM


Formal Idealism and the Aesthetic Condition of Experience Through the progressive development of his argument, Kant gradually approaches a point at which the a priori structures – which are, indeed, initiated by the mind – come to grasp the genuinely extramental. This process entails setting the initiating conditions, both conceptual and aesthetic, in relation to the affective condition of experience. The aesthetic forms are particularly crucial because they provide the point of access to an affective given. Without them the conceptual forms would have nothing to operate on. One common way of presenting the situation is to say that Kant aims to get beyond the polarities of the rationalist and empiricist worldviews and their concentration on thought or sensation, respectively. Kant’s alternative path is to establish the relation between thinking and sensing. In the next two chapters I will show how this relation arises as the activity of synthesis. Synthesis is the capacity to take up the sensory given through the introduction of formal structures. The Copernican revolution first uncovers this initiating condition of experience, while also revealing that mental activity has a dual structure, being both receptive and unifying. This puts in question the thesis that the mind projects an order onto an indifferent world. The alternative is that knowledge arises as a process of negotiation between mind and world. My account will not suggest that the world independently displays structural features that just happen to fit with the mind’s own order. This would be to revert to Trendelenburg’s overlooked alternative rehabilitated by Guyer.29 This would, in turn, destroy any claim that the mind provided an a priori structure for experience, because the conformity between mind and world would simply be contingent and not necessary. To say that there is a necessary connection between subjective representations and objects is to say that mind actually gets hold of what is given to it in experience. We do not reach the right conclusion by coincidence. The necessity characteristic of knowledge arises neither from a contingent nor a pre-established harmony between mind and world, but only from our ability to determine things through the use of our cognitive powers, that is, our faculties of understanding and intuition. The alternative I offer – both to the view that the mind imposes order on the world and to the position that there is a mere coincidence between mental order and order in the world – is that there is an ongoing dynamic between mind and world initiated at a formal level by the mind. But this will only be established by the full extent of Kant’s argument, of which the Copernican Revolution is merely the beginning. 95


29/3/07

8:12 AM



III An alternative to the circle of representation: representation as the mode of access to objects I have suggested that the mind initiates the possibility of knowledge insofar as it provides forms for experience. I have highlighted that the conceptual determination of objects is only possible insofar as we are capable of receptivity to something given in experience. This is the aesthetic dimension of Kant’s formal idealism. These two insights into Kant’s position have been extracted from his statement of the Copernican Revolution in philosophy. In this section I will argue that we can recognise the centrality, and indeed the necessity, of representation without falling into a circle of representation. My aim is not to try to return to the affect (or even the ‘double affect’) of a thing in itself. It is rather to reinstate the force of the affect of the appearing thing in Kant’s account. I have argued that while the ‘outer things themselves’ that Kant mentions in his contrast between formal and material idealism are not things in themselves, they are extra-mental entities.30 Their defining characteristic is that they are accessible to us, and in this sense are ‘for us’ or for the mind. Their status is not privatively subjective, but relational. The question is: how are we to account for the affect of a thing that is related to mind and yet extra-mental? Throughout the ‘Transcendental Aesthetic’, Kant repeatedly insists that we encounter objects as representations for our minds. Crucial to this perspective is the claim that both time and space are modes of representation – that is, they are nothing if we abstract from ‘that mode of representation [Vorstellungsart] which is peculiar to us’.31 But I will argue that these claims are consistent with the position that the forms of intuition are modes of access to extra-mental things. Kant appears to go one step further when he says ‘what we call outer objects [Gegenstände] are nothing but mere representations [Vorstellungen] of our sensibility, the form of which is space . . .’32 This could sound as if what we call outer objects are merely subjective mental representations. Even this claim, however, is compatible with the view that we are affected by something extra-mental. Representations belong to sensibility insofar as the latter makes them possible, but representations arise from the affect on us of something other than our representations. This something other than ourselves, the source of the affect, can only affect us insofar as our consciousness represents things given in experience. Were this not the case, the object would count as a thing-in-itself. 96


29/3/07

8:12 AM


Formal Idealism and the Aesthetic Condition of Experience It might seem that Kant goes yet further when he says that form represents only what is ‘posited’ by the mind and thus is ‘nothing but the mode [Art] in which the mind is affected through its own activity (namely, through this positing of its representation), and so is affected by itself’.33 But in this case Kant is referring strictly to the formal conditions of experience, which do indeed arise from the mind. He need not be taken as suggesting that what is taken up within form – that is, representations in general or objects – are posited by the mind. Only form, which is our mode or style [Art] of being affected counts as self-affection. Admittedly, Kant’s account is misleading. It would have been better had he said that form is the dimension of our receptivity that comes from us. Nevertheless, this passage does not lead to material idealism. In yet another passage, Kant says that ‘all our intuition is nothing but the representation of appearance’.34 This, too, can be read in accordance with his formalist perspective. Representation can now be recognised as a necessary condition of our access to appearing objects, yet we really do access something extra-mental when we have a representation. In other words something does appear to us as a representation. So far I have parried what appears to be an invitation to an impositionalist reading in a number of passages in the ‘Transcendental Aesthetic’. Now I turn to another passage where Kant appears to take the opposite tack, going so far as to suggest he may have reverted to transcendental realism. Late in the ‘Transcendental Aesthetic’, Kant writes ‘Our mode of intuition is dependent upon the existence of the object [das Dasein des Objekts], and is therefore possible only if the subject’s faculty of representation is affected by that object’.35 Affect, it would appear, arises from the object, which sets the faculty of representation into action. This could count as a restoration of the causal account of affection against which the Copernican revolution was directed. Would this mean that things in themselves are the basis for our representations of things as appearances? It is thus not surprising that the interpreters discussed in the previous chapter emphasise the priority of representation over any talk of access to objects and restrict philosophical investigation of objects within the scope of what I have called the circle of representation. But Kant avoids transcendental realism just as much as he avoids empirical idealism. The last cited passage merely states that the appearing thing affects us insofar as it is accessible to us in experience and that the form of intuition is only exercised in conjunction with 97


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology such an affect. Now does this not lead to the danger raised by Allison and mentioned in the previous chapter, namely, that the empirical object, which is the result of affection, is conceived as the latter’s cause? We have seen that the problem can be avoided, as Allison suggests, once we recognise that the matter in the empirical object can be considered in philosophical reflection in abstraction from the epistemic conditions that are necessary for experience.36 We can thus maintain that it is the empirical object that affects us and that strictly speaking there is nothing other than empirical matter.37 An empirical object comprises a material and a formal component. The formal component establishes the possibility of introducing order into experience and arises from ourselves, thus counting as a priori. The material component is the given element in the empirical object. We only have access to this given insofar as we have capacities – both sensible and conceptual – for introducing order into matter. Matter requires form. However form is the form of matter. In explaining the material given we do not need to proliferate levels of matter, although we do need to distinguish different ways in which we consider it. When we consider an object as empirically formed, we view it as the result of a process whereby matter has been given a formal structure. The empirical object viewed in this way does not reveal the initial affect that first gave rise to that formative activity, because we are concerned with the object as formed matter. However, we can also view the same empirical object in respect of its constitutive elements, that is, as to its matter and its form, even though within experience these are never encountered discretely. From the vantage point of philosophical reflection, I consider the object as material, that is, I focus on the matter in the object. It is true that this matter has only affected me insofar as I am the bearer of formal capacities that give shape or figure to the material given, however there must also be matter that is formed and this matter is nowhere other than in the empirical object. The thought experiment I am engaged on is one in which I imagine what always occurs together as prised apart. The gain from doing so is that I am able to focus on the matter in the appearing object and do not need to conjecture a further transcendental matter beyond experience. The transcendental move involves reflecting in an alternative fashion on the experience with which we are already engaged. Kant’s transcendental turn is necessarily committed to a material given, and, at the same time, to the position that the latter can be viewed not only 98


29/3/07

8:12 AM


Formal Idealism and the Aesthetic Condition of Experience empirically but also from the standpoint of transcendental reflection that reveals the unformed element within experience. But so far, my account treats matter simply as accessible from the perspective of transcendental reflection. I have not addressed how empirical matter affects the subject. Empirical matter in the appearance has an affect on us, even though this only happens insofar as matter arises within a form, that is, as a representation. While Allison’s account suggests that matter is simply the obverse of form, a distinction between matter understood as result and as the event of affection will highlight the process through which matter is taken up by form and thus gives rise to an affect. While only formed entities can be known, transcendental philosophy must face the question of the limits of experience and it does so insofar as it addresses the point at which matter becomes accessible to us. The transcendental turn entails that this event of affection does not happen outside of experience, but rather at its limits. The event of the emergence of a thing for consciousness is usually simply assumed by us, even insofar as we are philosophers.38 Kant’s occupation of what could seem like opposing positions, at times insisting on the priority of mind, at others on the priority of the object, can now be explained. His general project is one of expressing the relation in which the formal structure initiated by mind stands to given objects in the world. His commitment to the relation between form and world is what leads him sometimes to emphasise one term and sometimes the other. Admittedly, the affect things have on us occurs at the limit of our powers of representation and hence of reflection. Nevertheless, this does not render the question of affect philosophically redundant. If Kant is to avoid material idealism he must show how it is that we can be in contact with real things and not just mental projections. How are we capable of being affected by something other than our own subjective impressions and yet only in the medium of our representations? He must show how our faculty of representation takes up something given in experience and does not simply generate the latter. What affects our power of representation will necessarily stand at the limit of our experience, but this is not to say that the object stands outside it in the way that a thing in itself would. Clearly the authors I examined in Chapter 2 do not think that the faculty of representation generates affect. However, by inadequately addressing the philosophical relevance of the question of affect, they invite the rejoinder of interpreters such as Pippin, Guyer and 99


29/3/07

8:12 AM

Page 100 Gary Gary's G4:Users:Gary:Public:Gary's

Kant’s Aesthetic Epistemology Strawson for whom Kant’s account gives no convincing story as to how the mind encounters something extra-mental and for whom instead the mind imposes order on a neutral given. The stakes are high, because I do not want to revert to the position that experience arises out of an unknown something, ‘I know not what’. How are we to understand the claim that appearances are both merely for our power of representation and yet, at the same time, arise out of an affect on that power of representation? The model I will introduce in the next section is one in which there is a necessary relation between mind’s formative or representational power and the matter given in experience, that is the empirical matter, that could not even be recognised as such were there not a representational form in which to take it up. Correspondingly, mental formative activity would be incapable of initiating experience were it not prompted or affected by the material in appearance.

IV A reading of the initial paragraphs of the ‘Transcendental Aesthetic’ In the previous section I established a possible reading of passages in the ‘Transcendental Aesthetic’ where Kant appears most guilty of adopting an impositionalist position. But I now need to establish that Kant actually presents affection as a condition of representation. In this section I will show that Kant provides a model of affectivity in the initial paragraphs of the ‘Transcendental Aesthetic’ before he moves to his principal topic, the status and structure of the forms of intuition, which, as we have seen are modes of representation of appearances. Intuition is the means through which knowledge stands in immediate relation to objects.39 While an object, strictly understood, is the result of the determination of an intuition under a concept, Kant initially uses ‘object’ in a less technical sense to refer to something extramental given to our senses. The object in this context is the underdetermined given in intuition. I have already used this distinction to clarify the Copernican revolution. I will now show that my suggestion is supported by the initial account Kant gives in the ‘Transcendental Aesthetic’. The relation to things through intuition is only possible insofar as the object is given to us.40 This can only occur insofar as ‘the mind is affected’.41 The mind can only be affected insofar as we have a capacity of receptivity. Sensibility is our capacity ‘for receiving [bekommen] representations through the manner in 100


29/3/07

8:12 AM


Formal Idealism and the Aesthetic Condition of Experience which we are affected by objects’.42 We can now draw together these claims. Sensibility is our capacity for being affected by objects and, for us, being affected entails that we have representations of objects. Sensibility is the capacity to take up affects as representations. This is confirmed when Kant goes on to say that sensibility ‘yields us intuitions’.43 But affection is the necessary condition of there being representations in the first place. Intuitions are representations and they arise from sensibility, our capacity to be affected. We can so far conclude from the first paragraph of the ‘Transcendental Aesthetic’ that our ability to be affected by objects is, at the same time, the initial condition of our representational capacity. Insofar as we are, through sensibility, capable of being affected by objects, we are capable of having intuitions of them, that is, of relating to them immediately. Intuitions are Vorstellungen, that is, they are modes of knowledge that allow an object, literally, to be placed before us (vorstellen). In having an intuition we stand in relation to things that are accessible to us. These things are appearing objects or appearances. In the second paragraph, Kant characterises an appearance as ‘the undetermined object of an empirical intuition’.44 We can now understand the implication of this characterisation. An appearance is an object that affects us through our capacity for sensibility. Our only access to the object is in a representation. At this stage of Kant’s argument, the appearance is undetermined simply because it affects us only as an intuition and has not yet been taken up by understanding, the capacity for determination of the given so as to achieve knowledge. The initial, non-technical sense of object refers us to our sensory awareness of what would, if it were determined by a concept, count as an object in the strict sense. In the first sentence of the ‘Aesthetic’ Kant says that all thought is finally directed to intuition, our ability to stand in immediate relation to objects.45 This reveals that it is not only the case, as McDowell says, that concepts ‘go all the way down’, but also that intuition goes all the way up.46 And it would appear that not all empirical intuitions are taken up in thought. In the ‘Transcendental Deduction’, Kant says that objects may appear to us without ‘their necessarily relating to functions of the understanding’.47 Now although this passage could be taken to suggest that we can have intuitions without their standing in any relation to concepts, it is not necessary to come to this conclusion. All Kant need say is, first, that intuition is a necessary and irreducible condition of an object in 101


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology the full sense of the term. Second, sensibility provides us with objects in a non-technical sense, that is, with appearances in intuition. Objects in this weaker sense count as representations, yet ones that are not determined by the faculty of understanding. However, Kant’s claim that intuitions are not necessarily connected to the functions of the understanding is misleading. He should have said that, insofar as our intuition is not intellectual and is necessarily combined with a capacity for conceptualisation, intuitions must necessarily relate to the possibility of determination under a concept. What is not necessary is that intuitions are in fact determined by a concept. This maintains his commitment to the dual necessity of intuitions and concepts for experience, without concluding that all intuitions give rise to cognition. Leaving open the possibility of indeterminate intuitions is important and not just insofar as it leaves a gap for underdetermined awareness in everyday life, without which Kant’s account of experience would be implausible. It also leaves a place for aesthetic judgements in which intuitions relate to concepts, but only indeterminately.48 It seems likely that the ambivalent nature of Kant’s account of the relation in which intuitions stand to concepts in the first Critique arises from his not yet having established the distinction between determinate and reflective judgement.49 At the most primitive level, an appearance is the representation of an object’s affect on us. It is empirical because the object is only ever that thing that affects us within the range of experience. This thing can only be accessed through our modes of knowledge, in this case the forms of intuition that give rise to representations. But representations would not arise if the object did not affect our sensibility. The relation between object (in the non-technical sense) and representation is a two-sided one. Without representation there would be no affect by the thing, without the affect by the thing there would be no representation. This account of the role of representation highlights the necessity of affect and thus does not fall back into the circle of representation. The conclusion is that an appearance is an object (in the looser sense) given in intuition and only accessible to us through our powers of representation. If that appearance were determined by a concept it would count as knowledge of an object (in the technical sense). In the second paragraph of the ‘Aesthetic’ Kant says that sensation is the effect (Wirkung) of an object upon the faculty of representation insofar as we are affected (affiziert) by it.50 The question is not here of the effect of some unknown cause, but rather of the affect that 102


29/3/07

8:12 AM


Formal Idealism and the Aesthetic Condition of Experience makes possible an empirical intuition.51 In the third paragraph, Kant develops his point by making a distinction within appearance between matter and form. Matter in an appearance is that which corresponds to sensation.52 Sensation is the empirical result of our capacity to be affected by an object. The form of appearance, in contrast, is that which allows the manifold, that is, the given, to be ordered in certain relations.53 The ordering of sensation cannot arise from the given itself, just because the latter is empirical. Order must arise from the mind, as only as a priori can its necessity be secured. Any appearing thing arises out of the conjunction of the material given, which is empirical and thus a posteriori, with the formally a priori. Matter is thus established as necessarily empirical. It is easy to conclude that this entails that the a posteriori is simply the after-effect of a priori form. But, as we saw in the previous section, there are two levels at which the matter necessary for intuition can be analysed. Matter experienced as one of the component parts of an appearance is viewed as the result of a process of formation. We only experience matter insofar as it is taken up by representation, that is, in relation to form. However this result arises out of two irreducible elements. Although we cannot experience matter in abstraction from form, we can think it. I suggested in the previous section that we can express the affect of matter as an event, in contrast to its result in sensation. Thinking matter as event allows us to consider the a posteriori in a different way. The matter of an appearance counts as a posteriori now in the sense that it is what is necessarily given as an affect. The a posteriori cannot come from the mind and must be given. This reveals an alternative nuance of the a posteriori, where the latter is not simply subsidiary to the a priori, but radically distinct from it. The radical sense of the a posteriori is that of the necessity of the givenness of matter within the empirical. Matter viewed in this way is the necessary complement of the a priori.54 Thus while sensation viewed as the material aspect of appearance counts as an effect or result, we can also think about affect as a necessary condition of experience or event. While we only ever experience the result or trace of this event – that is, some combination of form with matter – there would be no material given, were there not an initial event of affection.55 Matter is that in the appearing thing that sets our representational capacities to work. Neither the affective condition of representation nor its a priori form is temporally prior, so it is not that the object first exists and then has an affect on us. Rather, to view something as an object is to consider it as standing in an affective relation to us, that 103


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology is, as accessible to our representational powers. The affect and the representation arise in one and the same event as a relation between mind and world. This is the characteristic status of an appearance in contrast to a thing-in-itself that would not stand in any affective relation to the representational mind. To say that the object only exists as such insofar as it affects us and is capable of being taken up by our powers of representation may appear to fall back into material idealism, albeit of a new form. However, Kant is not committed to the position that we produce objects, but simply that we only attend to those that our subjective faculties are capable of taking up. This does not entail the existence of another category of objects we cannot access.56 Rather, we relate to objects as our environment; they are part of the world we also inhabit and it is only by treating them as such that we are capable of encountering and, on occasion, knowing them. The appearance, as we have seen, is a composite of form and matter, but Kant now goes on to focus exclusively on the formal side of intuition. This central emphasis of the ‘Transcendental Aesthetic’ has encouraged the impression that his account of mind is wholly constitutive and not receptive. Form counts as pure insofar as to achieve it we abstract all the features that arise from sensation, that is the material side of the appearance. These include impenetrability, hardness and colour.57 Additionally, we must abstract the features that arise from the understanding – for instance, substance, force and divisibility.58 What is left is characteristic only of pure intuition, that is, the aesthetic form of appearance, namely, extension and figure. Pure intuition, Kant says, ‘is to be found in the mind a priori as a mere form of sensibility’.59 This formal residue of every appearance is what Kant puts down to self-affection, as we saw in the previous section.60 Thus form cannot be derived from sensation; nevertheless, we have seen that there would be no appearance in the first place, were we not affected through sensibility by the object. Form is not imposed on the matter of sensation, but is the framework within which the latter is represented. An appearance arises from two sources, one extra-mental and the other mental. The appearance differs from the thing in itself and from the merely phenomenal illusion in that it is relational, that is, it arises from a relation between object and the mind. We have already seen in a provisional way in our discussion of the Copernican revolution that mind initiates the forms not only of intuition but also of understanding.61 Later we will 104


29/3/07

8:12 AM


Formal Idealism and the Aesthetic Condition of Experience see that the combination of a plurality of modes of knowledge or faculties is necessary for Kant’s account of experience. In the ‘Transcendental Aesthetic’, Kant chiefly investigates the principles of a priori sensibility, that is, the aesthetic form of appearances, which he presently announces are only two, namely space and time.62 But the first few paragraphs of the Critique have revealed that these mental forms are only one side of the story of intuition and that appearances arise only insofar as we are affected, in a non-technical sense, by an empirical object. The forms of intuition are representational forms through which we take up what empirically affects us.

V The complex identity of receptivity In the first paragraph of the ‘Transcendental Aesthetic’ Kant identifies sensibility as ‘the capacity (receptivity) for receiving representations through the way in which we are affected by objects’.63 I have construed this as the view that we have representations only insofar as we are affected through our capacity of sensibility. This affect gives rise to a sensation. But, as we have also seen, this is only one side of the story of receptivity. Our intuitions – that is, our capacity to stand in an immediate relation to objects – also require an a priori element, that is, a form that makes possible the ordering of the manifold of matter. Receptivity arises out of an affect, but the latter can only count as such – that is, affect us – if it can also take on a certain order and this requires a pure intuition. The forms of intuition make possible that we take up something and hold it together in an empirical form.64 The aesthetic side of our experience is (philosophically) revealed as a point of contact between objects and mind.65 A full account of the structure of intuition must include not only its formal conditions, but also the way in which form makes possible access to a material given. Without this crucial move concealed in the initial paragraphs of the first Critique, Kant’s formal idealism would have fallen back into material idealism. The form of space makes possible that there are empirical spaces, that is, that these can be taken up by us as representations. In the first and third propositions of the ‘Metaphysical Exposition’ of space, Kant rules out that space counts as an empirical or general concept, respectively.66 Having entertained and rejected one possible account for the status of space, in the second proposition he claims that space is the pure intuition underlying all our outer intuitions.67 Space, as pure form of intuition, counts as the condition of the possibility of 105


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology appearances.68 Thus space and time as pure intuitions are the formal conditions that make possible that we have empirical intuitions, that is, that we can be affected by an object through sensibility so as to give rise to a sensation. But this is to say that pure intuitions are necessary but insufficient conditions of empirical spaces and times, not that they generate the latter. The necessity of form does not undermine the status of intuition as given. Rather, it reveals the complex structure of our receptivity necessary for taking up any affect. Form is simply the ordering process through which we take up the given. Intuition arises out of a material and a formal source; first, our sensibility as the capacity to be affected; and second, the pure intuitions of space and time. These are the two sides of our capacity of sensibility or receptivity, revealing how affection and the possibility of representation are inseparable from one another. Receptivity or sensibility is thus, on further investigation, twofold. It comprises, first, the capacity for being affected, which gives rise to the material element in intuitions, and, second, pure intuitions or forms. These are two sides of one coin. We could not be affected were we not capable of taking up the given within a form or order. In the next chapter we will examine how experience or knowledge arises out of a combination of understanding and intuition. We have now learnt that even the intuitive component of experience is not a simple event, being directed both toward the subject and toward an indeterminate object. It is because of this two-sided character of receptivity that Kant is not an impositionalist. It is also because of this that transcendental idealism qualifies as transcendental, that is, as a philosophical reflection on the conditions of experience. The conditions are ‘of’ experience with the genitive force of belonging to experience, in contrast to the suggestion that they are an independent set of rules applied to experience. The dually oriented structure of receptivity is what allows our minds to get a grip on the world. Admittedly, Kant’s failure to fully develop his account of the affective side of experience and, at the same time, adequately explain the complex structure of receptivity are responsible for many of the misunderstandings that have arisen. While Kant and his interpreters have tended to characterise sensibility as passive in contrast to the activity of the understanding, we can now see that this is misleading. Sensibility is a capacity, albeit a curious one insofar as it is the ability to receive what is given in experience. It is not active because it is the mode in which we are affected 106


29/3/07

8:12 AM


Formal Idealism and the Aesthetic Condition of Experience by things in the world. But no more is it merely passive, first and most obviously because it requires a formative ability. The second of these features has led many readers to conclude that sensibility does not really count as receptive. However, sensibility is a capacity to open ourselves to things in the world. This entails that we anticipate or prepare for the event of the material given, not that we create or impose the conditions of that event. Intuitive form is nothing other than the anticipation of the event of the affect of the material given.69 Kant’s account of sensibility shares some of the force of the phenomenologist’s notion of passive synthesis.70 The latter is a pre-reflective capacity to make sense of things in the world.71 Kantian sensibility is not pre-reflective, however, and is always combined with some level of reflection. Nevertheless there is a parallel insofar as both passive synthesis and receptivity inhabit a middle ground between activity and passivity.72 At this stage, I simply wish to establish the point of departure for the reading of the relation between form and matter that I will develop in later chapters, especially in Chapter 6 where I examine the overall structure of the objective deduction. In what sense has the approach adumbrated here escaped the circle of representation? I fully agree with Buchdahl, Allison and Longuenesse that any object we experience is a representation. But I also have some sympathy with the insistence by Pippin, Guyer and Strawson that our experience must be of empirical objects and not just of representations. My solution has been to argue that representations are our mode of access to things. It is not that we experience representations instead of things. It is rather that we experience things as representations, not through the latter but in them. The interpretation I have developed in this chapter allows me to suggest provisionally a way in which we might construe the relationship between empirical realism and transcendental idealism. The forms of experience are our modes of representation that give us access to an empirically real world. Transcendental idealism secures the empirical reality of the world and does not threaten it. No more is empirical realism a cheap substitute for transcendental realism, which, being incapable of proving its premises, cannot secure what it tantalisingly promises, namely, knowledge. It is not so much that Kant offers empirical realism because this is all that can be derived from transcendental idealism, as that transcendental idealism is the strategy he adopts to safeguard the empirical realism to which he is ultimately committed. Empirical realism is the only realism compatible with the 107


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology relation in which we stand to an extra-mental world, the form of which arises from our minds.

Notes 1. An affect is a sensory event, viewed as something that happens or is given to a receptive subject. 2. Or, in the case of temporal sensibility, to be aware of ourselves as other, that is, over time. 3. Added in the second edition of Critique of Pure Reason, at B 519. 4. In Chapter 6 we will see that Kant uses the expression ‘appearances themselves’, Erscheinungen selbst, at A 190, B 235, to express that appearances are capable of bearing objective validity. 5. B 519 (my additions). The German is: ‘In manchen Fällen scheint es ratsam zu sein, sich lieber dieser als der obgenannten Ausdrücke zu bedienen, um alle Mißdeutung zu verhüten’. 6. In the mind of God, though, not merely in our minds. See Berkeley ‘A Treatise Concerning the Principles of Human Knowledge’ in Philosophical Works ed. Ayers, pp. 116–17. 7. B 274. For a discussion of the systematic position of the ‘Refutation’ and its contribution to a possible spatial schematism, see Chapter 6, pp. 234–5. 8. Though both are defensible only within the context of experience and not as purely theoretical or conceptual theses. They require intuition and not just concepts. 9. My analysis of the early paragraphs of the ‘Transcendental Aesthetic’ later in this chapter will show that the position of the Refutation cannot be seen as a divergence from Kant’s main project. 10. Although not on the absolute given, as all three concede the need for reflective mediation of the given. 11. Preface to second edition of CPR, B xvi. Strictly, the reversal applies not only to objectively valid cognition or knowledge, but all cognition or Erkenntnisse. 12. See Hume, A Treatise of Human Nature, p. 11, on the role of fancy in relations of cause and effect. See also p. 86 on the role of belief or assent. 13. See, for instance, my discussion in Chapter 6, pp. 238–41 of the principle of hypothetical necessity, which is a precursor for the reflective principle of the systematicity of empirical nature. 14. B xvi/xvii (my additions). 15. B xvii. 16. See discussion of the first edition ‘Transcendental Deduction’ in Chapter 4, pp. 122–7. 17. B xvii. 18. B xxi. 108


29/3/07

8:12 AM


Formal Idealism and the Aesthetic Condition of Experience 19. See also B xxv/xxvi. In Chapter 2 we saw that there is a problematic role for things-in-themselves even within the epistemic domain insofar as thought of them facilitates a focus on undetermined matter. 20. A 20, B 34 (my addition). See discussion in following section. 21. See Chapters 6 and particularly Chapter 7, pp. 249–55. 22. See final section of this chapter, pp. 105–8. 23. A 51, B 75. 24. See final section of this chapter, pp. 105–8. 25. Later I will argue that these are not all that are necessary. See Chapter 7, pp. 249–55. 26. A 20, B 34. See my discussion of the initial paragraphs of the ‘Transcendental Aesthetic’ below. It will emerge that appearances may also count as the ‘not yet fully determined’ and not merely the ‘indeterminate’. 27. In the Preface to first edition of CPR, A xii, Kant characterises the task of the Critique thus: ‘I do not mean by this a critique of books and systems, but of the faculty of reason in general, in respect of all knowledge after which it may strive independently of all experience’. Clearly the last phrase refers to a priori knowledge. My argument will be that a priori knowledge must nevertheless anticipate the possibility of empirical knowledge, the full account of which requires more than the arguments of the Critique. 28. See Chapter 6 for a hierarchical reading of the ‘Transcendental Analytic’. 29. See Chapter 1, p. 18, for a discussion of Guyer. 30. B 519. 31. A 35, B 51. See also A 37, B 54. 32. A 30, B 45. 33. B 67/8. 34. A 42, B 59. 35. B 72. 36. See Chapter 2, pp. 65–7, on Allison’s reading of the transcendental significance of affection. 37. Gerold Prauss’s interpretation is in the same vein. He argues that affection is empirical; however, empirical affection cannot be grasped at the level of empirical experience nor by the empirical sciences. Only transcendental reflection can explain the empirical affect on a non-empirical subject in a ‘non-empirical science of the empirical’. See Prauss, Kant und das Problem der Dinge an sich, especially pp. 213–27. My own formalist account shares with Prauss the conviction that the subjective structure of the empirical is not empirical, but transcendental. 38. See Chapter 8, p. 304, where I argue that aesthetic apprehension makes possible a glimpse of the event as event. 39. A 19, B 33. 109


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 40. 41. 42. 43. 44. 45. 46. 47.

48.

49.

50. 51. 52. 53. 54. 55.

56. 57. 58. 59. 60. 61. 62. 63. 64. 65.

A 19, B 33. A 19, B 33. A19, B 33 (my translation). A 19, B 33. A 20, B 34. A 19, B 33. McDowell, Mind and World. See, for instance, pp. 13 and 23, which seem to emphasise only the first of these relations. A 89, B 122; my translation of ‘ohne daß sie sich notwendig auf Funktionen des Verstandes beziehen müssen’. He reinforces this at A 89/90, B 122–3. See my discussion in Chapter 4, pp. 137–47, of B 160 where Kant appears to rule out the possibility of undetermined intuitions. See also Chapter 5, pp. 175–6, on the relation between aesthetic judgement and cognition in general. Sarah Gibbons holds a similar view. See Kant’s Theory of Imagination, p. 18, note 6, where she distinguishes between the possibility and the actuality of conceptualisation. This distinction is first introduced in the Critique of Judgment at AA 179. I first mentioned this in my discussion of Pippin in Chapter 1, p. 15. A 19/20, B 34: ‘the effect . . . so far as we are affected by it, is sensation’. A 20, B 34. A 20, B 34. A 20, B 34. In later chapters I will argue that a priori form anticipates instantiation in the a posteriori. See Chapter 6, pp. 221–9, and 7, pp. 249–55. But although matter is always experienced as in some sense formed, it is not necessarily experienced as determined by form. See the discussion of aesthetic form in Chapter 8. It is because in an aesthetic judgement form and matter stand in relation and yet are not unified in a cognition that the relation between form and matter is displayed. At least not epistemically. A 20/1, B 35. A 21, B 35. A 21, B35 (my translation). B 67/8. See discussion of forms of understanding in Chapter 4, pp. 113–16. A21–2, B 35–6. A 19, B 33 (my translation). See discussion of synthesis of apprehension in intuition in Chapter 4, pp. 122–4. Later I will argue that aesthetic judgement provides a phenomenological revelation of the aesthetic or sensible dimension of knowledge. 110


29/3/07

8:12 AM


Formal Idealism and the Aesthetic Condition of Experience 66. I will not discuss here the remainder of the content of the 3rd proposition, which taken in conjunction with the 4th establishes that space as a whole is the horizon of all particular spaces. See my ‘Kant’s Phenomenological Reduction?’ for the claim that the forms of intuition are best understood as horizons of sensory experience. 67. A 24, B 38. I will leave discussion of the ‘Transcendental Exposition’ until Chapter 4. 68. A 24, B 39. 69. See Chapter 6, pp. 221–9, where I bring out that the principles of understanding are anticipatory of a material given in space and time. 70. See Edmund Husserl, Analyses concerning passive and active synthesis, especially pp. 508–12, Hu 408–11. 71. Volume 11 of Husserl’s collected writings is devoted to this theme. A more accessible version is to be found in Merleau-Ponty’s discussion of ‘general synthesis’ in the Phenomenology of Perception. See, for instance, p. 428 (French edition, p. 489). See also Heidegger’s discussion of Sein lassen or ‘letting-be’ in Being and Time; for instance, H354. 72. This ground that is not a ground can be grasped through the trope of the middle voice, a notion that has been fruitfully explored by John Llewelyn. See, for instance, his ‘Heidegger’s Kant and the Middle Voice’, and the ‘Preface’ to The HypoCritical Imagination.

111


29/3/07

8:12 AM


4

The Deep Structure of Synthesis

In the previous chapter I examined the receptive condition of knowledge and its source in sensibility. This faculty is able to take up the object given in experience because of its introduction of intuitive forms into the manifold or empirical matter. I have claimed that there are three conditions for knowledge. First, there must be something given to us; second, we must be receptive to that given; and third, we must be capable of unifying the given under a concept. In the last chapter I discussed the first and second of these conditions. I now turn to the third, the condition of conceptual determination. Knowledge as opposed to intuition is only possible if we can unify and thus identify what Kant provisionally called the object in the initial paragraphs of the Critique of Pure Reason. In the first section of this chapter I discuss the other side of Kant’s dualism and argue that it is best understood as reflective form. I also argue that, on closer examination, dualism requires a plural iteration of the operations of the mind as the synthesis of affectivity and conceptualisation requires a third term, imagination. The combination of these different elements of experience counts as synthesis. In the second section, I discuss the faculty talk that I have employed in my initial characterisation of synthesis. I suggest that the faculties need be seen neither as psychological constructs nor as curious entities. Kant’s faculty talk allows for reflection on a complex model of mind in which only the combination of distinct orientations gives rise to the structure or form of experience. Within his epistemology Kant principally distinguishes sensibility from understanding. At a more general level of analysis of human existence, he distinguishes the cognitive, the moral and the aesthetic. At each level of analysis Kant portrays the human subject as operating with a plurality of different orientations. Faculty talk is a way of capturing that plural constitution. In the third section of this chapter I turn to Kant’s account of determining judgement in the two editions of the ‘Transcendental Deduction’. Kant attempts to validate the claim that the concepts of the understanding are applicable to all empirical objects. I argue that 112


29/3/07

8:12 AM


The Deep Structure of Synthesis both editions reveal a commitment to a plurality of faculties working in conjunction with one another and that the imagination plays a vital role in making this possible. However, in both editions Kant’s account reveals the hegemony of the understanding and not a free cooperation among a plurality of faculties. In the final section of this chapter I argue that the synthetic activity arising from a combination of a plurality of faculties is examined directly, not merely presupposed, in Kant’s account of aesthetic judgement. The latter counts as synthetic insofar as its characteristic ‘harmony of the faculties’ reveals the synthesis in process necessary for any judgement.

I Reflective form and the plural conditions of synthesis If the ‘object’ given in experience is to be more to us than mere intuition, then we must be capable of taking it up in such a way that we can identify or determine its appearance in space and time. In order to do this, we need not only intuitions but also concepts. Just as intuitions are traced back to the human capacity for sensibility, concepts are ultimately traced back to the human capacity for understanding. If we did not have the latter, we would be incapable of any level of experience beyond that of sensation. Appearances can only count as objects for us in the strict sense, that is, as identifiable particulars, insofar as we are able to distinguish them as falling under concepts. What was termed an object in the ‘Transcendental Aesthetic’ in fact fulfilled only one of the conditions for the latter, namely, that there is something given to us, not merely constructed by our minds. Kant’s considered account of objects is that the proper use of the term is epistemic and refers to the result of the synthesis of a concept with an intuition. Our experience of objects, thus, is dependent on the exercise not only of the faculty of sensibility, but also on that of the understanding. In the previous chapter a close link emerged between form and mind. It is we who introduce form into experience. A priori forms are of two types: first, the forms of intuition – namely space and time – and, second, the forms of understanding, that is, the categories. Sensibility has been identified as the source of the forms within which any empirical intuition arises. Now I will examine understanding as the source of the forms that make conceptualisation possible. The form of experience arises from the combination of these two elements of experience. 113


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology Dualism expresses a distinction within subjectivity between our receptivity to the outside world and our capacity for conceptualisation that makes possible unification of what is given to us in intuition. The most famous statement of this position is: ‘Thoughts without content are empty, intuitions without concepts are blind’.1 Our dual makeup means that we are reflective beings, for our thinking only arises in relation to a world not of our own making. This means that our peculiar mode of thinking is that of synthesising, that is, we make sense of this world through the combination of receptivity and understanding. Dualism thus requires a relation between concepts and intuitions and this entails a third term, the imagination, which makes possible their synthesis. This is to say that our mental activity is plurally and not merely dually iterated. Later in this chapter I will show that in the heart of Kant’s epistemology the operation of a third term is essential. While the plural constitution of human experience only becomes explicit in the Critique of Judgement, at all stages of his account the faculty of imagination is the mediating faculty that makes unification possible. This is true even though it is also the case that the lineage of imagination is often put in question, as it is sometimes allied with intuition and often with understanding. Although at a surface level dualism suggests a fragmented psyche, at a deeper level it must be recognised as entailing the cooperation of a plurality of mental orientations that constitute the subjective condition of experience. At first glance dualism names the complexity within a subject who is both receptive and reflective. But as a receptive subject necessarily stands in relation to a world of things that are given, dualism also reveals the relation in which we stand to a world external to our minds. A receptive subject is not the author of its experience and thus dualism reveals the human condition of finitude. The result, however, is not an impasse between subject and object when dualism is interpreted as entailing a pluralist relation to an extra-mental world. The third or relational term makes possible not only unity within a subject, but also access to an external world insofar as we are capable of taking up the given in a synthesis and achieving cognition of it. This is in contrast to the two-term relation of the impositionalist model, which were it true of Kant’s epistemology would mean that he was committed to the view that the understanding projected its forms upon an external world. Impositionalism gives a reductionist interpretation of dualism, according to which dualism results in the dominance of mind. But dualism is neither committed to a simplistic 114


29/3/07

8:12 AM


The Deep Structure of Synthesis relation among the faculties nor between mind and world, and for good reason. The initial triumphalism of mind would fall back into a subjectivism based on understanding alone where instead of encountering the world, we would be faced only with our own reflection. Contrastively, if the philosopher rests his or her account on the faculty of intuition as the sole effective source of knowledge, this version of monism would establish only that we are capable of having impressions of the world, not that we can know objects. In the previous chapter I developed the idea that mind initiates form. By this I mean that the initial possibility of ordering the world comes from the mind, but experience only becomes possible insofar as mental forms are set in relation to possible experience. This requires that the forms be rearticulated so that they are capable of ‘anticipating’ empirical experience, as I will argue later.2 The limitation on the role of form was revealed by my discussion of Kant’s distinction between formal and material idealism. Our mental activity is principally that of ordering and making sense and we do so by generating structures that allow us to take up the contents of the world we find ourselves within. Kant characterises the conceptual side of experience as a capacity for ‘spontaneity’. Understanding is capable of spontaneously or actively introducing a priori forms of unification, that is, categories into experience. These forms are not to be found in the given nor can they be supplied by the power of receptivity, sensibility. Unification arises from the human capacity for thought and, in particular, from our capacity for understanding. However, it is important to recognise the limitation Kant places on our ability to generate the conceptual form of experience. The latter can only give rise to the form of experience insofar as it is combined with the aesthetic form arising from our capacity for receptivity. For this reason I believe it is important to temper the characterisation of our power of conceptualisation as spontaneous, justified by its status as the source of priori forms, with a recognition of its reflective status.3 This is particularly important when considering the application of a priori forms. The forms introduced spontaneously only gain validity in that they take up content that they themselves cannot deliver. We do not have an intuitive intellect that would directly grasp the inner essence of things. Our consciousness is necessarily a synthesising one. We work at making sense of the world and have no immediate or final access to the truth of things as they are in themselves. The latter would require an operation of an intuitive understanding independent 115


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology of any other faculties and for finite beings such as we are this is simply not an option. We lack the royal road to truth and must always struggle for an ideal, which, necessarily, is absent. But this does not make us indifferent to truth. We do not simply accept that what appears to be the case, is the case. Instead we seek out ever more layers of appearance, tending towards an infinite and unrealisable idea of total revelation. This goal must, however, function strictly as a regulative idea of reason if we are not to fall back into a new form of dogmatism that would arise from assuming that we can achieve absolute truth. To do so, would be to ignore the necessarily reflective and synthesising nature of our consciousness.4 While the position just adumbrated may appear to go beyond Kant’s commitment to the certainty of knowledge, his claim for apodeictic certainty refers only to the synthetic a priori judgements that provide the minimal framework, not the content, of experience.5

II The form of experience and faculty talk We can make sense of Kant’s faculty talk in the following way. Our consciousness is plurally oriented and comprises a number of different capacities that must be distinguished and coordinated, that is, synthesised within experience. The responsive, yet spontaneous human mind necessarily stands in a complex relation with the world. It is open to the world and yet it is not wholly determined by it. The thesis of dualism is expressed by a faculty theory that draws out the various different orientations we take within experience. At a more general level of analysis than the epistemological one within which we are currently operating, Kant is committed to the view that human experience cannot be analysed using one singular explanatory model, for instance, a mechanical one of natural entities, a teleological one of natural purposes, or a moral one. Instead his account of human experience requires recognition of a combination of different orientations. Human beings are neither wholly rational as angels perhaps would be, nor wholly sensory as Kant holds animals are. Rather, we are the kind of beings who combine rationality with sensible experience. We do so through the mediation of a capacity arising from a third faculty, imagination.6 Many commentators, understandably, have cavilled at this apparatus and the psychological foundationalism it seems to entail.7 In my view, the faculty theory is neither psychological nor does it provide an autonomous foundation for experience, but rather is an attempt, 116


29/3/07

8:12 AM


The Deep Structure of Synthesis sometimes clumsily handled, to account for the way in which a complex mental life makes experience possible. In order to avoid unnecessary reification, we should say that synthesis is the activity of the faculties cooperating with one another and that the faculties are nothing other than the elements or poles of that cooperation. The faculties need not be taken as predating, or as substrates of, our experience. They are not even in principle independent of experience, for they bear validity only insofar as they are at work within particular experiences of the world. However, this does not mean that they are simply superstructural or fictional. Kant’s talk about faculties expresses the dynamic activity of mind and provides a way of addressing the latter’s necessary moments. In order to do this he is forced to talk as if there were poles or constituent elements in this activity, while this is not strictly true. It is necessary to identify specifiable tendencies within our mental lives. We orientate ourselves to the world in a number of distinguishable ways and an adequate philosophy of mind must account for these differences. We know what it is like to attempt to achieve knowledge about a state of affairs, or to respond to it morally or imaginatively. Human experience would be unimaginable without these necessary possibilities, which although intertwined with one another are also distinguishable. Faculty talk makes possible a philosophical reflection on the necessary elements of our mental activity, allowing certain primary orientations to be individuated, while allowing them to be treated as necessary structures of possibility, not as substantial mental entities. This is in contrast both to the deflationary account that would treat faculties as merely possible, yet cannot explain why they are not just some capacities among others, and, on the other hand, the reificationary account that can only establish the individual characteristics of the faculties at the cost of positing them as substantial. The fear of psychologism has led many philosophers to excise the necessary role for mental activity in Kant’s epistemology, leaving such concerns to the psychologists. Yet Kant believes that objective knowledge arises from subjective sources. An advantage of Kant’s use of faculty talk in his epistemological argument is that it allows him to focus on the differentiation and interaction within mind, without cutting off mental activity from the world. I agree with what I will argue is his position, namely, that establishing a relation between subjectivity and objectivity is necessary if knowledge is to be shown to be possible. I also agree that the subjective process by which knowledge emerges should be recognised as depending on the ability to combine 117


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology a plurality of different orientations or perspectives. Whether it is necessary to characterise the subjective side of this relation as comprising faculties is debatable, but that Kant was committed to such a model is undeniable. For Kant the particular faculties of understanding, sensibility and reason are the sources of knowledge, perception and speculation, both moral and scientific.8 However it is only through the interaction of the faculties that anything like thought or experience emerges. Knowledge, for instance, has a primary connection to the understanding, but cannot arise unless the latter takes up the sensible given to which we are receptive in intuition and indeed, as I will argue later, Kant holds that knowledge also requires that we exercise reflective judgement.9 The need for cooperation among our faculties is first introduced as a theme in the Critique of Judgement under the title of ‘the harmony of the faculties’ although there is plenty of evidence, which I will endeavour to uncover, that the pluralist model was already operative as a presupposition in the first Critique.10 All of this underlines the pervasiveness of reflection within human experience in the Kantian account. Reflection is the thinking activity of a consciousness that is also receptive. When Kant talks of the combination of sensibility and conceptualisation within experience, he is not, of course, intending that there are two sorts of experience that are then forged together. Instead his point is that all human experience necessarily displays both a sensory and a conceptual dimension. This is what McDowell means by saying that spontaneity goes all the way down.11 There is no sensible impression that is wholly detached from the possibility, at least, of reflection on it. This is true even when the power of expression breaks down. Faced with the highest peak in Europe and the expanses of snow surrounding it, I may say ‘there are simply no words for this!’ I may search for an expression that at least points to my lack of a description, saying perhaps, ‘this is sublime’. But I cannot escape the power of reflection. My sense of awe is ineliminably connected to the inadequacy of my power of words and draws on a contrast to the latter. Correspondingly, any exercise of my power of reflection or spontaneity stands in some relation, however distant, to sensible experience. Receptivity goes all the way up, just as spontaneity goes all the way down.12 Kant expresses this complex structure using the model of a plurality of faculties. While worries about the reintroduction of foundationalism – now psychological – should not be underestimated, the importance of faculty talk is in providing Kant with a way of expressing the complex 118


29/3/07

8:12 AM


The Deep Structure of Synthesis identity of the form of experience. Orientations, irreducible to one another, cooperatively give rise to a structure without which experience would not be possible. Kant captures this in talking about distinct but related forms of experience that jointly give rise to the transcendental form of experience. It is thus that faculty talk allows him to develop a complex philosophy of mind. Mind is not reconstructed as a singular entity, facing an undifferentiated world, but, rather, a complex and internally relating activity facing a diverse world – the manifold of sensibility. The unification that the mind effects is necessary for the unifications in our experience of the world, for it is principally our capacity of synthesis that makes any unity in the world possible. The model or notion of mind and world that arises from all of this is one of a constant dynamic process and yet one in which unity or sense, though no final unity or sense, can and must arise. Experience and knowledge are projects or tasks which comprise the ongoing unification or determination of the indeterminate, but determinable, which we take up through the combination of sensibility and understanding. There is, however, a danger that the account of the pluralism of the faculties I have sketched and will defend in the chapters that follow appears both blandly moderate and implausible as a representation of Kant’s position. What is the advantage of a cooperation of a plurality of the faculties in contrast to the dominance of one singular mental capacity? And, is not Kant, in the end, of the view that moral reason is the dominant force in our experience of the world – a position his own account of orientation in thinking reinforces?13 In contrast to this prima facie persuasive view of Kant’s commitments, I will defend an interpretation of the relation of the faculties, not simply as a cooperation of indifference, but rather a dynamic interaction that requires the distinctiveness of the elements in the relation. While one faculty is dominant in each of the legislations covered by the three Critiques, the effectiveness of the principle governing each requires a plurality of mental orientations. Even moral law cannot operate by independent Diktat. In Kant’s philosophy of history we find that any hope we may have for the emergence of a kingdom of ends on earth, rests on our ability to organise our relations with other moral agents according to political strategies, which are, necessarily, amoral.14 Kant’s system continually relies on a cooperative model of mental activity, though not one of mere equivalence. This only becomes explicit for his readers and for Kant himself in his articulation of the ‘harmony of the faculties’ in the Critique of Judgement. 119


29/3/07

8:12 AM



III How the legitimation of the categories requires the cooperation of a plurality of faculties The categories are ‘concepts of an object in general, by means of which the intuition of an object is regarded as determined in respect of one of the logical functions of judgement’.15 It is thus clear that the conceptual side of Kant’s account of knowledge necessarily stands in relation to the affective or aesthetic side that I was concerned with in the previous chapter. The relation in which concepts stand to intuitions is distinctive of transcendental logic, in contrast to general logic which abstracts from all content.16 The ‘Table of Judgements’ belongs to general logic, but nevertheless provides the most general shape of thought characteristic also of transcendental logic. Judgements are functions of unity in our representations.17 Transcendental logic, in contrast, will establish the functions of unity or synthesis among objects.18 Thus Kant identifies the most general functions of judgement and uses these as his touchstone for discovering the most general ways of thinking of objects, namely, the categories.19 Kant warns that it is very tempting to use the categories ‘by themselves and even beyond the limits of experience, which can alone yield the matter (objects) to which those pure concepts of understanding can be applied’.20 By doing so, the understanding makes a material and not a properly formal use of its principles.21 Kant’s point is that within transcendental logic, a logic that is concerned with the a priori conditions of empirical experience, the understanding can only be used in conjunction with sensibility, that is, in application to matter or objects that are as yet undetermined. This is what he aims to establish in the ‘Transcendental Deduction of the Categories’; any other use of the categories would count as dialectical. The ‘Transcendental Deduction’ is the centrepiece of the logical or conceptual side of Kant’s account of knowledge. In it he seeks to prove the validity of the categories of the understanding. This entails showing that the categories or subjective conditions of thought ‘furnish conditions of the possibility of all knowledge of objects’.22 He remarks that the forms of intuition do not need a deduction. This is because, although they are a priori, they must necessarily relate to objects and therefore bear objective validity.23 This supports my insistence that sensibility is two-sided.24 Establishing the objective validity of the categories requires a specific argument because, Kant says, appearances can arise ‘independently of functions of understanding’.25 Kant’s point is that everything we experience must arise 120


29/3/07

8:12 AM


The Deep Structure of Synthesis within the forms of space and time and therefore counts as appearance. But he also makes clear that not all appearances are subject to the unifying power of understanding. It sounds as if Kant’s commitment to the dual structure of experience has collapsed. On other occasions he appears to suggest quite the opposite, namely, that all intuitions must be unified by the understanding.26 This, too, would threaten dualism. Kant should have said that while not all appearances are determined by the understanding, they must stand in some relation to our reflective power of understanding and thus are determinable, if not determined.27 I will not provide here a continuous reading of the two editions of the ‘Deduction’. There are many excellent accounts available and it would not be possible to rehearse Kant’s arguments in full or assess them.28 Instead I will examine the ways in which both editions reveal the pluralist relation among the faculties I have been adumbrating. Nevertheless, because Kant is ambivalent about the role played by the imagination and this leads to much disagreement among his interpreters, I will need to go into some detail even on this issue. I will show how an insufficiently articulated account of the relation between the faculties contributes to Kant’s prevarication as to whether or not knowledge exhausts the field of possible experience. (i) The A ‘Deduction’ and The Identity of Productive Imagination Following an introductory section retained in the B edition, the presentation of the ‘Transcendental Deduction of the Categories’ in the A edition is divided into two further parts. In Section 2 Kant analyses what he presents as three syntheses, while in Section 3 the same syntheses are treated as three aspects of one systematic account.29 It may seem most easy to support a non-impositionalist reading starting from the A edition ‘Deduction’. This is the position of Heidegger, for whom the A edition has the potential to become an ontological account insofar as the imagination affords access to things and not just representations.30 However, I will argue that both the A and B ‘Deductions’ provide an ambivalent account of the relationship between imagination and understanding. In both editions the relations between the faculties and the central mediating role of the imagination have to be teased out, while ultimately the understanding is dominant. In short, it is not, as Heidegger claims, the case that the A ‘Deduction’ reveals the imagination as the root of sensibility 121


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology and understanding, while the B ‘Deduction’ makes the understanding the final source of knowledge. Both versions require that imagination is the necessary mediator between the faculties, while at times compromising the identity of imagination in favour of understanding. In Section 2 of the A edition Kant presents his position through an analysis of three syntheses. Commentators have often suspected that this version of his argument falls into the trap of offering a faculty psychology, as he identifies each of the syntheses with an operation of the mind. It should be remembered, however, that Kant’s analysis of mental activity, which is to be found in both editions, is normative, as Longuenesse points out, in that it operates at the level of the structure of the mind and not its contents.31 Kant’s account in Section 2, in particular, has also been criticised for giving the impression that intuition is a prior operation, detached from that of the understanding. I will show, however, that in Section 2 Kant avoids this trap as well, although I will suggest that he leaves room for a level of integrity within intuition that falls short of the full unity required for knowledge. The synthesis of apprehension in intuition establishes coherence within the sensory given in that the temporal manifold of intuition is ‘run through, and held together’.32 If an intuition is a representation for consciousness, then the impressions out of which it is made up must be unified.33 This occurs over time. Kant is not concerned here with re-identification over time, or even with our capacity for maintaining attention to an object over an extended period of time. It is rather that he is insisting that our consciousness of something in the present requires that we are able to hold together temporal moments as the content of our consciousness. Kant says that without this synthesis the representations of space and time would not be possible.34 If it later transpires that understanding already has a role to play in the synthesis of apprehension, this might seem to count as persuasive evidence for Longuenesse’s synthesis speciosa thesis.35 However, I will show that there is scope for resisting the conclusion that the forms of intuition are a covert operation of the understanding. There are grounds for arguing that the synthesis of apprehension entails more than the forms of intuition tout court. Whereas the manifold of intuition is given in the forms of space and time, that manifold is only ‘held together’ through a synthesis in apprehension. If this distinction goes through, then we can argue that even if the understanding plays a role within the synthesis of sense, the former is excluded from the framework conditions of the latter. The forms of 122


29/3/07

8:12 AM


The Deep Structure of Synthesis intuition turn out to be independent of the understanding, even if the synthesis in apprehension is not. Potential additional evidence for a distinction between the forms of intuition and the manifold appearing within those forms arises in Kant’s occasional mention of a synopsis of sense that coincides with the manifold in intuition. He says on one occasion that a synthesis must correspond to this synopsis.36 This sounds as if the manifold in intuition coheres (in some sense that has not yet been identified) prior to a synthesis of apprehension. The suspicion is reinforced by Kant’s immediately going on to distinguish between receptivity and spontaneity as the sources of knowledge, suggesting that the first corresponds to synopsis, while only the second counts as a synthesis.37 This may correspond to a distinction Kant makes in the second edition of the ‘Deduction’ where he says that the forms of intuition give only a manifold, whereas formal intuition gives unity of representation.38 We could, then, conclude that the manifold in intuition arises as a synopsis, whereas a synthesis is required if there is to be unity of that manifold. Thus the idea would be that the forms of intuition, independently of understanding, introduce within the manifold a coherence of sorts that does not yet count as unity. However, this suggestion must remain speculative and stands in tension with an earlier passage in the A edition where Kant appears to identify the synopsis of sense with the first of the three syntheses necessary for knowledge.39 While it is possible that Kant meant to distinguish a way in which the manifold holds together prior to synthesis, the evidence is at best inconclusive and I will not investigate the potential nature of such a synopsis any further here. Even if the evidence for a distinction between synopsis and synthesis is not persuasive, it is still possible to argue that in his account of the synthesis of apprehension Kant is concerned with space and time as experienced, that is, strictly, with representations in space and time and not with the pure forms of experience. While the first paragraph reiterates the general finding of the ‘Transcendental Aesthetic’ – namely, that all appearances belong to inner sense – the second paragraph turns to the conditions of the unification of the sensory given.40 Thus when Kant says that the representations of space and time arise from the synthesis of apprehension, he is explaining the first stage or condition of the process of unification of the given in space and time and is no longer addressing the initial aesthetic condition of experience, namely, that the given must arise within the singular context of space and time as the forms of intuition. So even if this first synthesis 123


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology makes possible representations in space and time, it does not make possible the forms of intuition.41 The latter remain distinctive conditions of the unification of experience. This reading is reinforced by Kant’s account of apprehension in Section 26 of the B ‘Deduction’, as I will argue later in this chapter.42 But there is a further and independent way in which Longuenesse’s interpretation can be answered. If I can show that the synthesis of apprehension in intuition is not grounded in the understanding, while still standing in a necessary relation to the latter, then we need not conclude that the synthesis of apprehension results from a covert exercise of understanding. The account of the synthesis of reproduction in imagination is particularly resistant to clarification. Kant moves between two levels of argument. At the outset of his discussion the second synthesis appears to be merely supplementary to apprehension. If we are to re-identify an object over time, then we must be able to presuppose that objects are subject to a law of association. His examples reveal that he is thinking about the re-identification of empirical objects or events, for instance cinnabar, the human form and the weather conditions on the longest day of the year. Association, be it causal or merely reproductive as is under discussion here, is not simply the result of the customary association in our minds, but must also must be found in objects. Kant says that this is the case because objects, as appearances, are subject to a rule of association.43 Only in the second Analogy of Experience will Kant present his considered account of how a rule grounds causal association in objects. Kant’s response to Hume in the A ‘Deduction’ seems to be, at best, indirect. He is not principally concerned with causal association, but with the re-identification of one image over time. But now Kant moves to a different level of analysis, for he says: But if I were always to drop out of thought the preceding representations (the first parts of the line, the antecedent parts of the time period, or the units in the order represented), and did not reproduce them while advancing to those that follow, a complete representation would never be obtained . . .44

At first he was concerned with the problem of re-identification over time, but now he turns to the retention of consciousness of a phenomenon within the present. ‘Reproducibility of appearances’ now refers to a curious sort of reproduction, for at a deeper level Kant considers how we hold onto an object of consciousness within the 124


29/3/07

8:12 AM


The Deep Structure of Synthesis extended present.45 Kant’s examples reveal that he has turned his attention to the internal structure of our time-consciousness of a particular phenomenon, rather than the problem of its re-identification over time. He says that we could not draw a line in thought, think of the time within one twenty-four-hour period or even represent a particular number if we could not retain earlier impressions in attending to such phenomena. It is this shift in levels that explains Kant’s comment that ‘the synthesis of apprehension is thus inseparably bound up with the synthesis of reproduction’.46 Examination of our ability to retain consciousness of something within the extended present reveals that we can only run through and hold together the manifold in intuition in that we are capable of reproducing or retaining a content of consciousness as the same thing. While our intuition allows us to take in something given, the retention of that given over time requires the faculty of imagination. Imagination is our capacity for representing objects even when they are not present to consciousness. In this case the object is present to consciousness and the imagination holds together our fleeting impressions (Eindrücke). The synthesis of apprehension is thus necessarily connected with the synthesis of reproduction, without which there would be no unity in intuition. If the second synthesis were simply concerned with re-identification over gaps in time and not retention within an extended present, the activity of imagination would be merely supplementary and not necessarily connected with the operation of intuition. An impression can only qualify as an image insofar as it is retained over time; this requires we have not only a capacity of intuition, but also imagination.47 If a complete representation and ‘even the purest and most elementary representations of space and time’ require an operation of the imagination, then must we conclude that the first and second syntheses are not distinct from one another?48 It is not necessary, however, that we reduce apprehension and imagination to one function, in expectation perhaps of their subsequent reduction to a higher power of understanding. Kant’s point need only be that having a complete representation (and perhaps even an incomplete or indeterminate one) requires a complex operation of more than one mental capacity. He is not therefore committed to the view that imagination is, on consideration, capable of replacing our capacity for apprehension. Such a conclusion would in any case be at odds with his rejection of material idealism, for it would entail that imagination supplied the affective material of experience and did not have to rely on external, though 125


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology empirical, matter. All we need conclude is that the synthesis of reproduction extends the range of the synthesis of apprehension, by making us capable of sustaining the affective given in consciousness beyond a merely fleeting impression. What is the unity to which Kant refers in his account of the synthesis of apprehension? So far it seems only to entail that the manifold can be run through and held together. This could count as a rather low level of integrity. However, Kant has now claimed that a complete representation requires an operation of the imagination. Does this development in his account mean that sense achieves no holding together whatsoever or only that a higher degree of integrity is required if we are to focus on an impression as an image? The text at least seems to leave open the possibility that the latter could be the case, for Kant does not yet claim that objects that are not known are nothing to us. We will see that Kant closes down this possibility in later versions of his argument, arguing that even apprehension requires unification under a concept.49 In my consideration of the synthesis of recognition in a concept, I will focus exclusively on the relation in which it stands to the previous syntheses. Kant now announces that a third synthesis is necessary if the reproductive activity of imagination is to give rise to knowledge: ‘If we were not conscious that what we think is the same as what we thought a moment before, all reproduction in the series of representations would be useless.’50 Reproduction requires consciousness of unity if it is to be of any use and we can only suppose that the use in question is epistemic. Unity – Kant slips in without anything much in the way of explanation – entails that we have a concept. He immediately provides the missing link by defining a concept as ‘unitary consciousness’.51 A concept simply is a combination of the manifold in time, which as he has just argued entails that it is reproduced or retained. We can now return to Kant’s mention of unity in intuition in his account of the first synthesis. Only under a concept is a fully-fledged unity of intuition achieved, although this rests on the first necessary condition of unity, namely that the manifold is run through and held together in intuition. We cannot be sure if Kant intended to leave open the possibility that some intuitions may not be determined by concepts.52 But we can be certain that knowledge as a unity of our intuitions over time requires not only understanding, but also the contributions of intuition and imagination. We can thus address the question: are the three faculties best captured as reducible to one complex operation of the Vermogen zu urteilen or as three distinct 126


29/3/07

8:12 AM


The Deep Structure of Synthesis orientations working in conjunction? In Section 2 there seems to be little reason for coming to Longuenesse’s radical conclusion. The forms of intuition are presupposed by, rather than being the primary topic of, the synthesis of apprehension; and while understanding works in conjunction with imagination to give rise to a form or image, this is best understood as a complex operation linking our receptive and reflective capacities. Kant begins Section 3 with a complex commitment. Initially he announces that the syntheses that were examined in distinction from each other in the previous section will now be treated in systematic interconnection.53 He then adds a statement that appears to supply strong evidence for the distinctiveness of the faculties: ‘There are three subjective sources of knowledge upon which rests the possibility of experience in general and of knowledge of its objects – sense, imagination, and apperception.’54 While there is only one integrated synthesis, it arises from three distinctive sources. A cooperation of a plurality of faculties is thus necessary The cooperative status of the faculties is reinforced later in the section when Kant says that understanding is the unity of apperception, exercised in relation to the synthesis of imagination. He also says that the categories of the understanding relate to objects only by means of both intuition and the faculty he now presents as operating on the latter, namely, imagination.55 The faculties must work in conjunction with one another if experience is to arise. Interpreting the first of these claims as entailing that imagination is a covert operation of the understanding simply does not capture the intricacy of the overall picture, where Kant sometimes traces imagination back to intuition and sometimes to understanding. Only a relational model of the faculties can account for this complexity. Kant now introduces the notion of a productive synthesis of imagination that, in contrast to the reproductive imagination, is a priori.56 Productive synthesis is identified as the active faculty for synthesis or combination and is immediately directed to perceptions or apprehensions.57 Not only is imagination identified as one of the three fundamental sources of experience, but Kant also suggests that it may be first among equals: ‘Thus the principle of the necessary unity of pure (productive) synthesis of imagination, prior to apperception, is the ground of the possibility of all knowledge, especially of experience.’58 The supposed priority of the imagination supports Heidegger’s reading that the imagination is the hidden root that joins intuition and apperception.59 127


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology However Kant’s distinction between reproductive and productive imagination eventually puts in jeopardy the supposed autonomy of the latter. Imagination only counts as transcendental and productive if it stands in relation to the original unity of apperception; only as such is imagination the pure form of all knowledge.60 Reproductive imagination alone cannot achieve the unity or affinity of appearances that arises from the principle of the unity of apperception.61 Kant goes on to say that affinity ‘is a necessary consequence of a synthesis in imagination which is grounded a priori on rules’.62 It is thus imagination exercised in conjunction with apperception and, indeed, acting according to the rules of the latter that counts as productive.63 While this account does not necessarily threaten the claim that apperception has to cooperate with imagination, it raises considerable doubts about the view that the latter is in any sense prior to intuition and apperception. Having initially established the productive role of imagination, Kant announces he will show the necessary connection between the categories of the understanding and appearances ‘starting from below’, that is, starting from empirical appearances.64 As was already established in the ‘Transcendental Aesthetic’, what is first given to us counts as appearance. When we are conscious of the given in intuition, this qualifies as perception. Kant adds in parentheses that consciousness is a necessary condition of knowledge, a claim that is clearly in line with his dualist perspective for which knowledge arises from a combination of an aesthetic given with reflective consciousness. However the precise statement Kant offers is that if appearances do not stand in ‘relation to a consciousness that is at least possible, appearance could never be for us an object of knowledge, and so would be nothing to us’.65 The latter phrase seems to rule out a possibility he previously asserted, namely, that there could be an awareness or intuition of things, without our achieving knowledge of them. It could be argued that here he restricts the range of consciousness to coincide only with the level of awareness that qualifies as knowledge, leaving open that there could be another level of awareness that falls short of the unity that corresponds to apperception. Nevertheless the suggestion that an appearance would be nothing for us if it were not an object of knowledge tells against this. I believe there is sufficient ambiguity in Kant’s account to support the view that he has not adequately clarified the levels at which objects function in his account. At this stage the problem has not clearly emerged and, consequently, is impossible to dissipate. 128


29/3/07

8:12 AM


The Deep Structure of Synthesis The problem of the exclusion of not-yet determined objects is more explicit at the outset of Section 3 where he says: ‘Intuitions are nothing to us, and do not in the least concern us if they cannot be taken up into consciousness, in which they may participate either directly or indirectly. In this way alone is any knowledge possible.’66 However, while, appearing to rule out unsynthesised intuitions, this earlier passage leaves a loophole by introducing a distinction between direct and indirect consciousness of things. This leaves open the possibility that not all intuitions give rise to the direct or determinate awareness necessary for knowledge, while all representations in principle or indirectly stand in relation to the categories and so could be the objects of knowledge. This can explain Kant’s subsequent claim that representations must ‘at least be capable’ of being connected in a unitary consciousness.67 While we are only in principle capable of connecting all our representations within a unified system of consciousness, those that are so connected will qualify as knowledge. But Kant is not attentive to the limitations of knowledge in the central argument of the A ‘Deduction’, where, as we have seen, all appearances count as perceptions and thus are subject to apperception.68 Whereas I argued that, in Section 2, Kant left open the possibility that there could be a low level of integrity among representations without their necessarily attaining the unity afforded by apperception, Kant now seems to rule this out by presenting the relation between the three elements of synthesis as one of entailment. Appearances contain a manifold that must be combined. This combination ‘such as they [appearances] cannot have in sense itself’ arises from the imagination.69 Thus, what was first presented as the synthesis of apprehension in intuition is now announced as the work of an ‘active faculty’, that is, imagination in its productive guise.70 Kant goes on to declare that the representation of the form of an image requires that the imagination is able to reproduce the combination it has achieved.71 It should be noted that the form of an image referred to here clearly denotes the form of an empirical appearance and not the pure forms of space and time. Therefore there is no implication, in this statement at least, that the forms of intuition arise from the productive imagination. Even in Section 2 the holding together of impressions over an extended moment in time required an operation of the imagination. Knowledge only arises insofar as the form or image conforms to a rule, the empirical rule of reproduction.72 This subjective rule only allows us to grasp objects in that it, in turn, has an objective ground in the affinity of appearances, which requires an operation of apperception.73 Thus any 129


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology perception whatsoever requires the application of the categories to intuitions. The three syntheses appear to be firmly bound to one another, leaving no room for a distinction between different levels of consciousness or unity. But, in fact, there is a potential gap between the second and third stages of this new argument. Kant could have said that only intuitive representations giving rise to knowledge are unified by understanding. He could then have distinguished between determined intuitions – for which he could have retained the title ‘perceptions’ – and underdetermined intuitions, which require only intuition and imagination, while falling short of full determination under a concept. If Kant had not started out by remarking, almost in passing, that an object would be nothing for us at all were it not to give rise to knowledge, his main argument need not have come to such a radical conclusion. Despite Kant’s withdrawal of a certain level of autonomy from both apprehension and imagination, we need not necessarily conclude that this implies that imagination and intuition are covert activities of the understanding even in strictly cognitive activity. Admittedly, the cooperation among the faculties is not one of full reciprocity, for in the cognitive case, the understanding is dominant in its role of introducing unity into the manifold of intuition. However, as I have shown, it can only achieve this through the cooperation of intuition and imagination, thus the model is one of hegemony rather than of imposition. The understanding must work through the intercession of the faculties of intuition and imagination. And, thus, while the account of the productive synthesis of imagination in Section 3 of the A ‘Deduction’ identifies the third faculty as exercising rules that ultimately are derived from the understanding, we need not conclude that apperception simply imposes order on sensory appearances. In Chapter 1, I argued in response to Paul Guyer that the centrality of apperception in Kant’s accounts of affinity and necessity did not necessarily entail impositionalism, because form initiates, but is not the sufficient condition of, order. Now it has emerged that form arises not as a unitary activity on the part of one faculty, but as a cooperation of a plurality of faculties. Apperception achieves affinity and necessity only as a complex operation, one dimension of which is receptivity to empirical objects given to sensibility. While I have argued that imagination is not primary, no more is it merely subsidiary to understanding, being instead the mediating faculty par excellence. A passage, often remarked on, tells against concluding that imagination is merely an instrument of apperception: 130


29/3/07

8:12 AM


The Deep Structure of Synthesis A pure imagination, which conditions all a priori knowledge, is thus one of the fundamental faculties of the human soul. By its means we bring the manifold of intuition on the one side, into connection with the condition of the necessary unity of pure apperception on the other. The two extremes, namely sensibility and understanding, must stand in necessary connection with one another through the mediation of this transcendental function of the imagination, because otherwise the former, though indeed yielding appearances, would supply no objects of empirical knowledge, and consequently no experience.74

We can make sense of the apparent tension in which this stands to other passages, where Kant appears to reduce imagination to an exercise of the understanding only insofar as we recognise the relational status of the faculties. In this passage Kant uses faculty talk to insist on the dualist perspective, that is, there are at least two fundamental dimensions of experience, namely sensibility and understanding. His insistence that these must be set in relation to one another necessitates the introduction of a third term, the imagination, which also counts as one of the fundamental faculties of the human soul. This latter claim is prima facie evidence for the claim that there must be a cooperation among a plurality of faculties.75 Imagination is not only the third faculty but is also the relational faculty par excellence in that it is our capacity for mediating between different orientations. If there were not something given to the mind in sensible form, then there would be no knowledge; just as if there were no mental capacity for unification, there would be no knowledge. The extra-mental given and mental formative power are both necessary for knowledge and are in no way reducible to one another. The two poles of dualism are not simply opposed, or indeed two aspects of one faculty, but instead must be related and for this a third term is necessary. The impositionalist account cannot accommodate the intricate examination of the determination of the empirical by these a priori forms. It is bound to find the elaboration of the roles of the faculties an idle wheel. To grasp the complexity of Kant’s position, we need a relational and dynamic account of formalism. Kant’s account, while recognising the distinctiveness of receptivity and understanding, at times undermines this by the emphasis he puts on the role of spontaneity within synthesis. Had he emphasised the reflective status of consciousness, commentators might have been less inclined to conclude that he is committed to impositionalism. While spontaneity – that is, the mind’s capacity to initiate order at the formal level of experience – is a necessary feature of his account, Kant 131


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology sometimes sounds as if he is saying that the mind introduces order as a pure act. When this is combined with his tendency in the first Critique to leave insufficient space for non-cognitive awareness of the world, Kant can appear to have a very top-down account of experience. However, I have shown that even in the face of this, the A ‘Deduction’ reveals his commitment to the need for a cooperation of a plurality of faculties. This provides at least the beginnings of a more comprehensive account of experience, where mind must be receptive if it is to achieve knowledge and non-cognitive apprehension has a place. (ii) The B ‘Deduction’: Figurative Synthesis and a Renewed Attention to the Mode of Intuition We have already seen how the imagination facilitates the relation of understanding and intuition in the A ‘Deduction’. I intend to show that the B ‘Deduction’ also reveals the mediating power of imagination through the new notion of figurative synthesis, which Kant contrasts to intellectual synthesis. On closer examination, we will see that it is not so much that there are two different syntheses as that there are two levels of analysis of what is necessary if the categories are to be synthesised with intuitions or objects.76 Synthesis is analysed first as intellectual, and second as figurative. I will argue that the contrast between the former’s two-term and the latter’s three-term relation establishes the plural conditions of the epistemic task. There has been much debate about the B ‘Deduction’, which has generated perhaps more literature than any other section of the Critique. My limited task here will be that of assessing how pluralist or reductionist an account Kant gives of the relation between the faculties. I agree with Allison and Henrich that there are two parts to the ‘Deduction’ as presented in the B Edition and that these two parts culminate in Sections 21 and 26, respectively. Allison’s interpretation, with which I am in general agreement, hangs on a terminological distinction between Objekt and Gegenstand where the former stands for a logical object and the latter for an object of experience. As is now well established, Allison’s interpretation is that the first part of the ‘Deduction’ is concerned with establishing only that the categories apply to logical objects, while the second establishes that they apply to objects of experience. The general claim that the ‘Deduction’ gradually works from a more logical position towards a more experiential one is, I believe, convincing. My reading differs from Allison’s 132


29/3/07

8:12 AM


The Deep Structure of Synthesis insofar as I draw out the subjective side of the story. The two stages reveal not only two different analyses of the object, but also two stages of analysis of the subjective conditions of knowledge. My point is not to replace an objective account with a subjective one, but rather to show how the two sides are mutually implied by each other. Kant’s question, as I read it, is: how can we philosophically reconstruct the activity of the subjective modes of knowledge or faculties so as to establish how they can give rise to objective knowledge? Or, how can a subject know an object? In agreement with Longuenesse and Allison, I take the first part of the B ‘Deduction’ to be a development of the ‘Metaphysical Deduction of the Categories’ from the ‘Table of Judgements’.77 What Kant calls intellectual synthesis recapitulates the forms of thought, which in general logic function in abstraction from application to objects.78 Those same forms of thought are now transformed into categories that make possible the unity in our sensible intuitions. Mere forms of thought become categories when they are set in relation to the content of experience, namely, intuitions. Thus, one more element is added to the logical monism of the ‘Metaphysical Deduction’, where thought operates in isolation.79 The first step towards a pluralism of faculties has been taken. Anticipating a more complex account, Kant reveals on several occasions that the emphasis in this part of the text is one-sidedly oriented towards the understanding in its guise as the ‘original unity of apperception’. He reminds us that there is another side to the story. For instance, in Section 17 he states: Insofar as the manifold representations of intuitions are given to us, they are subject to the former of these two principles [the forms of space and time]; insofar as they must allow of being combined in one consciousness, they are subject to the latter [original synthetic unity of apperception].80

Thus Kant is well aware that the presentation of his argument results in emphasising one side of dualism over the other, temporarily at least. This same passage poses a question I raised in my discussion of the A ‘Deduction’. In what sense is anything given to us necessarily unified or combined by apperception? If this were the case then it would be impossible for us to have sensations without also conceptually identifying the latter. As we have seen, Kant states in both editions that not all appearances are conceptually determined, but we have also seen that he is inconsistent on this issue. The situation is 133


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology exacerbated by the fact that, whereas the imagination acts as intermediary in the A edition, here concepts operate directly on intuitions. This would encourage the impositionalist reading in that concepts would immediately apply to sensations. There would be no gap between mind and world, no task for mind to undertake in the world other than one already guaranteed success. This danger is diminished in the second part of the B ‘Deduction’ by the introduction of figurative synthesis, while the formalist interpretation allows us to read the passage under consideration as saying that it is only if the manifold is taken up in such a way that it is unified so as to give rise to knowledge that it is subject to apperception. This would leave open the possibility that there may be different modes of experience such as the aesthetic where determinate unity is not achieved, nor indeed sought. In the conclusion of the first part of the ‘Deduction’, Kant remarks that he has abstracted from the mode in which the manifold is given, although he has not abstracted from the fact that the manifold is given ‘prior to the synthesis of understanding and independently of it’.81 This last phrase reinforces the argument that synthesis arises out of a cooperation of distinct but connected faculties. As we have seen, even in the account of intellectual synthesis, concepts stand in relation to intuition. But we can now make sense of the distinction between the two versions of synthesis, as he goes on to say: ‘How this takes place, remains here undetermined’.82 Kant’s point is that in the first part of the deduction he has simply taken the relation between understanding (in its guise as apperception) and intuition for granted. He merely states that intuitions must be subsumed under concepts if knowledge is to arise. Now he is going to identify the process through which the relation between the two faculties is possible and this will require focusing on the combination of the distinctive form or mode of intuition with the forms of understanding or categories. While in the first part of the B ‘Deduction’ Kant progresses to a two-term model of knowledge comprising apperception on one side and sensibility on the other, this account is inadequate on two interconnected grounds. The polarised account presented as intellectual synthesis is incapable of showing how categories can be applied within experience and, at the same time, of showing how two seemingly opposed mental faculties can be coordinated in the interests of knowledge. Indeed these two failures converge, for it is only if Kant can show the possibility of the cooperation of the faculties, that he can vindicate his claim that we have access to an empirical given. The relation between the subjective conditions of experience must be 134


29/3/07

8:12 AM


The Deep Structure of Synthesis examined if the objective validity of the categories is to be established. This is not because subjective form is imposed on external matter, but rather because one of those faculties, sensibility, is nothing other than our capacity to take in the empirical given, as we saw in the previous chapter. The other subjective condition, the understanding, must be able to unify the given if knowledge is to be achieved. This entails a cooperation of the faculties. In the two-term account of mind, the only possible scenario, given the Copernican revolution’s insistence on the a priori element in knowledge, is that apperception imposes form on the sensible given. But this would mean that what counts as knowledge would be no more than the projection of our subjective impressions and not a grasp of an extra-mental world. It is because this would not be a satisfactory conclusion to his epistemic project that Kant goes on to develop the argument of the deduction so as to focus on a relation of form and matter and not simply an imposition of one on the other. In the second part of the B ‘Deduction’, the relation between understanding and intuition is brought into focus, for knowledge arises not only from the forms of thought, which Kant now retrospectively suggests were the topic of the first part of the ‘Deduction’.83 The unity of apperception in itself knows nothing ‘but merely combines and arranges the material [der Stoff] of knowledge, that is, the intuition, which must be given to it by the object’.84 The intuition that must be given is not merely the formal condition of intuition, that is, the pure intuitions of space and time. This would give rise only to mathematical concepts that do not themselves count as knowledge. Knowledge requires an empirical intuition, with its material content or sensation.85 Apperception thus provides merely the form for arranging a material given that must be given. Only thus do the categories achieve objective reality.86 If pure concepts are to be combined with empirical intuitions so as to give rise to knowledge, synthesis must be more than merely intellectual.87 Figurative synthesis replaces a two-term relation with a threeterm one. Sensibility and understanding are now joined by imagination, which is not simply a third party but the very ground of the possibility of mediation between the other two mental orientations. The transcendental imagination makes possible the application of concepts to empirical intuitions that is necessary if knowledge is to arise. Nevertheless, the status of the imagination as third faculty is difficult to express. As mediator, it is hybrid in status. Kant claims both that imagination belongs to sensibility and that it is an action of 135


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology understanding on sensibility.88 Just because the latter claim comes later in the account of imagination’s origin, it might seem that it trumps the former statement of allegiance to sensibility. Kant suggests that figurative synthesis, which he also calls the transcendental synthesis of imagination, is comparable to the role of the productive imagination, first mentioned in the A ‘Deduction’.89 As we saw, the latter is closely allied with the understanding. Leaving aside the lineage of the imagination for the moment, what is critical about both accounts is that imagination emerges as the capacity to generate the form of an image.90 This explains the relevance of Kant’s comment that imagination is the ‘faculty of representing in intuition an object that is not itself present’.91 The form is the representation of an object that can be attended to even when its material correlate is not presented in intuition. Imagination is thus the linking term between the mere impressions of intuition and the concepts of understanding. Grasping the form of an object is the first step towards knowledge and this is achieved by the imagination.92 We have two alternatives in response to the account of the relationship in which imagination stands to the other faculties. Either we tidy up Kant’s account and conclude that only some of his statements express his considered view. We may then conclude that the imagination is not really a third faculty but only an outpost of the understanding. This would lead to problems later on insofar as aesthetic judgement rests on a use of the imagination that is free from the rules of understanding. Moreover in the strictly epistemic context it condemns Kant’s epistemology to the pyrrhic victory of impositionalism.93 Alternatively, we can see Kant’s various characterisations of the relations in which the different faculties stand to one another as revealing a struggle which he has not yet the means of resolving. We can then recognise that imagination is a problematic third term, which, at times, is an ally of sensibility and, at others, of understanding. It is this very status as mediator par excellence that leads to the problem of identifying it in a definitive way. 94 Kant will not find a methodological position that is sufficiently subtle for expressing its problematic status before the third critique. What is undeniable is that from Section 21 through to Section 24, Kant insists that apperception is the framework within which the material given must be ordered. This requires what he calls figurative synthesis, which mediates between the faculties of sensibility and understanding. While the status of the imagination is difficult to pin 136


29/3/07

8:12 AM


The Deep Structure of Synthesis down, it is clear that its function is to make possible the application of the formal conditions of experience to empirical matter. (iii) The Re-identification of the Synthesis of Apprehension in Section 26 The introduction of figurative synthesis in Section 24 is evidence for Kant’s continuing commitment to the view that knowledge arises from a plurality of faculties. However, the culmination of his argument in Section 26 once again invites a more reductionist interpretation. This is one of the most difficult passages and particularly challenging for my reading. What we can call broadly the proof of the ‘Deduction’ can be reconstructed in four steps with a prefatory definition: Def: Synthesis of apprehension is ‘that combination of the manifold in an empirical intuition, whereby perception, that is, empirical consciousness of the intuition (as appearance) is possible’.95

1. The synthesis of apprehension must always conform to the forms of sensible intuition. 2. Space and time are not just forms, but are also intuitions containing a manifold. As this is the case, their representation requires ‘the determination of the unity of this manifold’.96 At this point Kant refers us to the ‘Transcendental Aesthetic’ and adds a now famous footnote.97 3. Thus, unity of the synthesis of the manifold and therefore also combination ‘to which everything that is to be represented as determined in space or in time must conform’ is the a priori condition of the synthesis of apprehension.98 He adds that this condition of unity is given ‘not indeed in, but with these intuitions’. 4. This unity is in an original consciousness in accordance with the categories ‘insofar as the combination is applied to our sensible intuition’.99 Thus Kant is able to conclude: All synthesis, therefore, even that which renders perception possible, is subject to the categories; and since experience is knowledge by means of connected perceptions, the categories are conditions of the possibility of experience, and are therefore valid a priori for all objects of experience.100

This proof offers a recasting of the argument of the A ‘Deduction’, addressing the problem of how the given in intuition is determined in time. The main difference is that there is no direct correlate for the 137


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology synthesis of reproduction, or, indeed, for the productive synthesis of imagination. In the conclusion to the B ‘Deduction’, Kant fails even to integrate the account of figurative synthesis he has just introduced in Section 24. The Definition makes clear, were there ever any doubt on the issue, that Kant’s aim is to establish the applicability of the categories at the empirical level. The first step establishes that all objects of empirical experience appear within the horizon of space–time or the forms of sensible intuition. This may count as a correlate for the synopsis of sense that was mentioned in the A edition. However Kant does not suggest that the forms of intuition provide a distinguishable way in which the manifold has a degree of coherence short of conceptual unity. What he does claim is that the forms of intuition are a first condition of the determination of empirical experience. He states that the synthesis of apprehension must always conform to these forms, because otherwise synthesis would be impossible. While this does not rule out the possibility that the forms of space and time may later be trumped by the categories of understanding, at first sight this is strong evidence for resisting the view that the forms of intuition can be traced back to an operation of the understanding, however covert. In the second step Kant distinguishes between the forms of intuition and experienced or empirical intuitions thus reinforcing the distinctiveness of both. 101 Space and time are either the conditions of experience, or they are experienced. Insofar as they are experienced they are not merely formal: they have content. Empirical intuitions mark out identifiable segments of the infinity of space–time. Whereas we could never intuit the whole of space and time, we can have intuitions of individual spatio-temporal objects. But if intuitions are to be identifiable as having content, then they must have some degree of integrity. This entails that they are unified. Now while empirical intuitions, which I discussed in Chapter 3, must display some unity, it is not clear that Kant should have gone so far as to claim that they require ‘the determination of this unity’ as he now does.102 We have seen that Kant denied this on two occasions at the outset of the ‘Deduction’ and in both editions.103 I have also suggested that the version of the deduction presented in Section 2 of the A ‘Deduction’ leaves open the possibility that there could be two levels of unity, where the first requires only that intuition conform to the general form of the categories. However, in the second step of the conclusion to the B ‘Deduction’, Kant implies that all apprehension is determinately unified. While it is clear that knowledge requires deter138


29/3/07

8:12 AM


The Deep Structure of Synthesis mination, Kant is in danger of suggesting that all sensory apprehension is cognitive. We can now turn to the especially difficult footnote at B 160 where Kant expands Step 2 of his account of the possibility of empirical knowledge. I have already established that in this second stage of his argument, Kant is concerned with intuitions as experienced, rather than merely with the forms of intuition that were the subject of the first stage. In the footnote he reintroduces geometric intuitions, because he is intent on revising the account he gave in the ‘Transcendental Aesthetic’, where he provided only a very provisional account of experienced intuitions. The exception was in his discussion of geometric forms. The ‘Transcendental Exposition’ provided a defence of the interpretation of space and time as forms of intuition established in the ‘Metaphysical Exposition’. Kant argued that such an interpretation underwrites the apodeictic status of geometry. Retrospectively, we can see that geometric forms, albeit pure or formal, are experienced intuitions. They are distinct from the forms of intuition discussed in the Metaphysical Exposition, which are best understood as the framework within which experience arises and never directly experienced as such.104 In line with the development between Step 1 and Step 2, Kant now establishes that the geometric shapes under consideration in the ‘Aesthetic’ are formal intuitions and not merely forms of intuition. He must hope that the reintroduction of this example will clarify his general point about experienced intuitions and in particular empirical ones, the conditions of which this proof is intended to clarify. In the ‘Transcendental Aesthetic’, Kant takes the unity of empirical intuitions for granted, treating the latter as if they arose from the power of intuition alone. Intuitions are treated as if they were identifiable and capable of being experienced and yet no account is given of the conditions of their unification. They are examined only with respect to their affective side and not as to the possibility of taking up affect in consciousness. Kant does not explain the reason for this at any length in the B ‘Deduction’. One reason we can suggest is that his primary concern in the ‘Transcendental Aesthetic’ was with the forms of intuition, not with empirical intuitions directly. His intention at the outset of the Critique was to establish that space and time must count as pure intuitions, derived neither from experience nor from concepts. I believe this is why he says that he did not supply an account of the unity of intuitions in the ‘Transcendental Aesthetic’, where he was intent only on establishing that intuitions arise from a source that 139


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology ‘precede[s] any concept’.105 The Aesthetic establishes an element of experienced intuitions, but not their full structure. Having conceded this omission, he goes on to say that in fact their unity ‘presupposes a synthesis which does not belong to the senses but through which all concepts of space and time first become possible’.106 Longuenesse has taken this statement as crucial evidence for her claim that the forms of space and time are a first operation of the understanding on sensibility.107 The first question we must address is this: is the synthesis in question the synthesis of apprehension in intuition, or the synthesis of imagination, which in the A ‘Deduction’ turned out to be necessarily bound up with the combination of intuition? Here Kant mentions only intuition, but he is concerned with intuitions as represented and this can only be achieved through the figurative or productive power of imagination. In the footnote at B 160–1, Kant elides the distinction between intuition and imagination, presumably on the grounds that the former cannot count as a representation unless its content is taken up by the faculty of imagination so as to give rise to a figure or form. This would then explain why Kant says that the unity in intuition entails a synthesis that does not belong to the senses. The synthesis in question is imaginative and not necessarily intellectual. Now we know from the first edition that if knowledge is to arise, imagination must stand in a necessary relation to understanding. But Kant did not need to conclude that the synthesis of apprehension is already an operation of the understanding. Unity in intuition can require more than the senses, without that entailing that the understanding alone is the source of unity. Admittedly, the full story of the unification of intuitions as representations will require the understanding, but there is no need to make the imagination an outpost of intellect. A cooperative model of the relation between the faculties will account for the complexity of Kant’s account in the footnote. It is true, though, that Kant’s omission of any mention of imagination at this stage of his argument encourages the conclusion that understanding operates immediately on intuition. When Kant goes on to say that ‘all concepts of space and time first become possible’ through this synthesis, Longuenesse concludes that this means the forms of intuition are only possible through the synthesis of apprehension, which is now identified as an operation of understanding.108 The possibility that the forms of intuition are outposts of the understanding is the second issue we need to address with regard to the phrase under consideration. But this runs against Kant’s 140


29/3/07

8:12 AM


The Deep Structure of Synthesis recently repeated distinction between forms of intuition and formal intuitions. The concepts of space and time in question are determinations of the infinite whole of space and time. Determinations of space and time, as experienced intuitions, are perceptions and entail unification. They require a synthesis beyond that of sensibility, but the forms of space and time do not. We may suspect that Kant is sometimes insufficiently careful and refers to space and time as concepts when he should have referred to them as intuitions or, at least, used a neutral term that left their status open. However, in this passage he has just used the term concept to refer to an operation of the understanding, so it seems likely that he sustains that usage, as he always should. I, therefore, conclude that the phrase under consideration tells us that all representations in space and time, not the forms of space and time, become possible through a synthesis requiring not only intuition, but also imagination.109 The final sentence of the note begins with the claim that space and time are ‘first given as intuitions’ insofar as understanding determines sensibility.110 Now for Longuenesse’s reading this mention of the understanding merely renders explicit what was assumed all along, namely, the operation of the understanding on intuition. How, according to my alternative reading, can I explain the introduction of the understanding at this stage of Kant’s account? If representations in space and time are to give rise to knowledge, the establishment of which possibility is the goal of the overall argument, then they must be subject not only to a synthesis of imagination, but also must be determined by a concept of the understanding. Only thus will space and time qualify not only as the forms of intuition, but also as given to us as determinate intuitions. As we have seen, at this stage of his argument Kant does not leave a space for undetermined intuitions. So his initial phrasing of Step 2, with its early mention of determination, is echoed by the footnote’s moving rather abruptly from the operation of the imagination in making possible the form of an object to the determination of that form by the understanding. While this raises a problem about the relation between apprehension in general and knowledge, the distinctiveness of the forms of intuition is not at risk. In my view Kant simply moves too quickly at this stage of his argument. All he needs say, as he did in Section 2 of the A edition, is that the synthesis of apprehension must be bound up with a function of the understanding, through the mediating role of the synthesis of reproduction in imagination. However he suppresses the mediate stage and threatens to transform an indirect relation in which 141


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology understanding relates to apprehension into a direct one. By failing to account for the role of imagination, which he had established as the centrepiece of figurative synthesis in Section 24, he collapses the complexity of his position into a flattened account that invites an impositionalist interpretation. My interpretation so far requires no straining of Kant’s sense, although I do criticise the speed with which he moves from understanding to sensibility and his requirement that all apprehension must be unified. In the second clause of the last sentence of the note my reading is faced with a final sticking point. Kant now says – as if it followed naturally from his claim that the synthesis of apprehension is the means by which understanding determines sensibility – the unity in question ‘belongs to space and time, and not to the concept of the understanding’.111 But why should this be if, as I am arguing, he is talking not about forms of intuition, but rather about formal intuitions, which are necessarily hybrid in origin? Longuenesse is in a position to account for this peculiar move in Kant’s argument through her distinction between an implicit and an explicit operation of the understanding on intuition. The first is only preparatory for determination and thus could count as belonging to intuition. In answering this question we should remember that the second step of the proof to which this note is appended is concerned with unity in intuition. So far Kant has argued that this requires not only intuition, but also a further synthesis, which I have argued should be identified as that of imagination. Full determination of intuition so as to give rise to knowledge would require the additional input of understanding, in line with Kant’s final goal of showing that even empirical intuition is governed by the categories. But he has not yet got so far in his argument. The introduction of the unifying role of understanding does not occur until Step 4, although it is prepared for in Step 3. In Step 2 there had been no mention of the understanding, prior to the final sentence of the note, where we must suspect that Kant jumps ahead of himself. My suspicion is that having prematurely introduced what is as yet only a goal, that is, the claim that the understanding finally gives rise to the determination of representations, he steps back to the matter in hand. His argument at this stage corresponds to the relation between the first and second syntheses of the A ‘Deduction’, namely to the way in which intuition is ‘run through and held together’ in intuition in conjunction with imagination. Kant remains committed to dualism and thus protects the distinc142


29/3/07

8:12 AM


The Deep Structure of Synthesis tion between the sensory and conceptual conditions of experience. The synthesis of apprehension in intuition falls under the first of these conditions. And although, as he later says, it must conform to the synthesis of apperception, this is not to say that the two conditions of knowledge are identical.112 If Kant had not suppressed and rather had explored the mediating role of imagination, pivoted between the sensible and intellectual conditions of experience, he could have made his point without inviting the conclusion that he is reducing intuition to an operation of understanding. Admittedly, later in the main text he says that the imagination and understanding amount to ‘one and the same spontaneity’, but this need not be read as denying their distinctive roles and only as insisting on the necessity of the cooperation of the faculties.113 Thus I take the final clause of the last sentence in the note at B 160 and its claim that unity ‘belongs to space and time, and not to the concept of the understanding’ to express a continuing commitment to the plural sources of experience, however clumsily presented. Even in the B ‘Deduction’, Kant resists reducing the synthesis of apprehension to a mere operation of the understanding. Another factor that may well have influenced Kant’s conclusion to the footnote is the leading example he has used to elaborate the status of representations in space and time. Geometric figures, while they do not coincide with the framework of space are also unlike empirical intuitions in that they express only a pure synthesis of space. Determinate empirical intuitions are subsumed under a concept, but what is the role of a concept for a geometric figure? The concept of a triangle does not subsume or explain the figure: it names something that can only be apprehended in intuition or imagination. Clearly, in the case of geometric figures the unity of the formal intuition does indeed belong to space and time and not to a concept of the understanding. Kant now resumes his argument in the main text. Step Three states that the unity of the manifold is the condition of the synthesis of apprehension. He explicitly equates unity with combination, whereas in the A edition a possibility remained that combination arose from intuition and imagination and unity only from a concept. However, even in the B ‘Deduction’ the third step of Kant’s account leaves open the possibility that he is claiming only a conditional necessity, that is, unity is necessary only for those representations that qualify as determinate and amount to knowledge. But by the same stroke, the potential distinction between combination and 143


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology unity is annulled by Kant’s conclusion that unity is necessary for the synthesis of apprehension and not simply for synthesis under a concept. And if there is no way in which intuitions hold together short of displaying a determinate unity, then we must conclude that all our apprehension amounts to cognition. If Kant wanted to maintain the possibility that there could be intuitions that are not (yet) determined, he would have to leave open the possibility that there were representations that did not yet display the full level of unity arising from subsumption under a category. In the footnote, as we have seen, Kant at best fails to articulate this possibility and is in grave danger of suppressing it entirely. We need not, however, conclude that this is his definitive intent, for at the conclusion of the third step he remarks that the unity of apprehension is given ‘not indeed in, but with these intuitions’.114 This is evidence for Kant’s continuing commitment to the distinctiveness of the aesthetic conditions of experience. What is less clear is whether the phrase once more opens up the possibility that some apprehension may not be cognitive. My suggested extension of his argument is as follows. Unity operates on intuition, but the former is not constitutive of the latter. Intuitions are distinct not just because they arise from a specifically aesthetic formal ground, the forms of intuition, but because even empirical intuitions are only unified insofar as the understanding operates on apprehension, which though necessarily bound up with the understanding, is not simply identical with the latter. And if unity arises from a distinctive faculty, then it is at least in principle possible that we could have intuitions that were not unified under a concept. While this possibility is left open by the text under consideration – and we have seen that on other occasions Kant explicitly espouses it – at this stage he makes no restriction on the range of knowledge. The fourth step of Kant’s proof states that unity in appearances coincides with unity in apperception.115 Understanding now, by right, enters the argument explicitly for the first time. Although Step Two rather precipitately referred to determinate unity, the need for unity of apprehension was only established in Step Three. It is not until Step Four that unity is traced to the understanding. But even now, the understanding cannot operate in isolation. The unity of consciousness would not be achieved, if understanding were not directed to an object, that is, applied to our sensible intuition.116 Were this not the case, we know from earlier arguments of the ‘Deduction’, apperception would merely give rise to forms of judgement and not to categories. The subjective 144


29/3/07

8:12 AM


The Deep Structure of Synthesis forms of the understanding only attain knowledge when they grasp appearing objects. This requires that the unity achieved by understanding is applied to sensible intuition and thus unity arises from a complex operation of the faculties. Kant completes the proof of the ‘Deduction’ by concluding that the unity of empirical objects can be nothing other than the unity of the combination of the manifold arising from the categories of the understanding.117 As I have argued, this is a foreshortened version of what is a complex account. While the logic of his argument certainly requires that the synthesis of apprehension must stand in some relation to the understanding, he need not have concluded that the combination of empirical intuition is simply an operation of the understanding. And Kant could have allowed that some empirical intuitions have a level of integrity, while failing to be wholly unified under a concept. Such underdetermined intuitions would still stand in a potential relation to the understanding.118 These would include sensory affects that we could, but do not yet, take up in reflection. But they would also include judgements of taste where both intuitions and concepts are in play, yet the first is not determined under the second. Kant need only claim that knowledge requires the full level of unification by the understanding. In collapsing the distinctiveness of levels between the syntheses of apprehension and recognition, while suppressing the mediating role of imagination, he is at risk of making his technical sense of experience as knowledge coincide with the only possible way in which we engage with the world. This would result in cognitive reductionism and invite an impositionalist interpretation of his epistemology. If he is to avoid this, he must separate out the component parts of knowledge and, by the same stroke, show how experience short of knowledge is possible. We will see that he attempts the first of these in the development of his argument that follows on from the ‘Deduction’, and indirectly addresses the second issue in the Critique of Judgement. It is only in the third critique that the aesthetic – that is, sensory – dimension of Kant’s epistemology is finally safe-guarded against the reductionist tendencies that we have seen emerge in the B ‘Deduction’. Kant’s account of apprehension has shifted since Section 2 of the A edition where the synthesis of apprehension was presented as a distinctive synthesis from that of recognition in a concept.119 Nevertheless even there we saw that unity in apprehension finally arises not just from intuition, but also from the cooperation of imagination and understanding.120 This extended account is not 145


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology inconsistent in substance with Section 3 of the A edition where apprehension is still associated with intuition, although it is explicitly identified as an action of the imagination on intuition.121 Nor is there any necessary shift of position in the final stages of the B ‘Deduction’, where apprehension is presented as standing in conformity with the synthesis of recognition in concepts.122 But Kant is at risk of collapsing the necessary relation between two elements of synthesis into an identity. This is because his omission of any account of imagination leaves the impression that the unity arising from the understanding operates immediately within the synthesis of apprehension. By the same omission, he covers over the possibility that there could be a non-cognitive apprehension. In the Critique of Judgement another possibility emerges, for aesthetic judgement counts as an indeterminate and thus non-cognitive apprehension of a sensory object. This alternative account of apprehension arises from a free operation of the imagination, the cognitively oriented exercise of which has been suppressed in Section 26. At the same time as Kant has difficulty establishing the limits of knowledge in the ‘Deductions’, he struggles to express the combination of faculties that is necessary if knowledge is to arise. As I have tried to show, his relative inattention to the subjective side of his story contributes to the difficulties that have emerged in his account of knowledge. At times it sounds as if he is reducing imagination to an operation of the understanding. But even in Section 26 of the B edition, more complex formulations show this not to be the case: ‘Now it is imagination that connects the manifold of sensible intuition; and imagination is dependent for the unity of its intellectual synthesis upon the understanding, and for the manifoldness of its apprehension upon sensibility.’123 What this reveals is that imagination is a mediator. It is, indeed, identical to understanding when it unifies intuition: ‘It is one and the same spontaneity, which in the one case, under the title of imagination, and in the other case under the title of understanding, brings combination into the manifold of intuition.’124 But it also belongs to sensibility in that the imagination makes available intuitions for concepts.125 Kant’s struggle to finally identify the role of imagination is due to his lack, as yet, of a sufficiently sophisticated methodological approach for capturing the plurality of the faculties and the process of mediation among those faculties. His account of experience is at risk of being overly epistemic, not because he relies on a faculty theory, but because his account of the relation in which the faculties stand to 146


29/3/07

8:12 AM


The Deep Structure of Synthesis each other and to possible experience is as yet insufficiently developed. (iv) Heidegger’s Account of Imagination as the Root of Sensibility and Understanding Heidegger’s Kant and the Problem of Metaphysics, while flawed, is one of the most insightful works into the lasting significance of Kant’s Critique of Pure Reason.126 His interpretation inspires readers to look beyond the surface of Kant’s account and search out what remains unstated, but is made evident in the letter of the text.127 For Heidegger, Kant’s project is not one of epistemology, but is rather an attempt to provide a fundamental ontology in the style of his own Being and Time. Such an ontology is not a science of beings or entities, but rather of the ‘Being’ that is prior to all experienced or ontic beings.128 Kant, according to this account, is not simply concerned with establishing the epistemic conditions of experience, but with showing how human beings are capable of transcending experienced beings towards their ontological ground, which he calls Being. In drawing out this potential in Kant’s position, Heidegger focuses on the role played by the imagination. Heidegger’s great insight is in drawing attention to the intermediary role of the imagination between sensibility and understanding. His greater insight is to insist that imagination is not simply part of the apparatus of faculty theory, but is the point of access to human beings’ finitude and their transcendence. Imagination sets us in relation to something other than ourselves and makes possible our transcendence of mere experience; at the same time, it reveals that we are finite beings. Transcendence and finitude are ineliminably linked for human existence or Dasein. For Heidegger, the transcendental imagination is linked to our receptivity or sensibility and, at a deeper level, is the source of our sensible capacity of intuition. As I have brought out in my reading of the A edition ‘Deduction’, receptivity is bound up with the capacity for retention that arises from our capacity for imagination. But, as we saw, imagination not only cooperates with intuition but also with understanding. Heidegger holds that imagination is the linking term between the two extremes, which by virtue of its mediation can be understood as inseparable from one another.129 Unlike most contemporary interpreters of Kant, Heidegger prefers the A edition ‘Deduction’ because of the importance it gives to the 147


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology role of imagination. He does not accept the usual contrast between a ‘psychological’ and a ‘logical’ account: It should be noted, in truth, that the laying of the foundation is no more ‘psychological’ in the first edition than it is ‘logical’ in the second. On the contrary, both are transcendental, i.e., necessarily ‘objective’ as well as ‘subjective’.130

Heidegger rightly puts in question the assumption that the emphasis on the understanding in the B edition qualifies that account as more ‘logical’ than the earlier account. But it is this emphasis on the understanding that, in his view, puts at risk the ontological orientation of Kant’s project. The understanding can only determine objects insofar as it stands in a (complex) identity with imagination.131 This position, established in the third Section of the A edition is forfeited, in Heidegger’s view, in 1787: ‘In the second edition of the Critique of Pure Reason the transcendental imagination, as it was described in the vigorous [leidenschaftlichen] language of the first edition, is thrust aside and transformed – to the benefit of the understanding.’132 The imagination still has a role to play, but Heidegger, in a move that prepares for Longuenesse’s reading, considers that Kant’s account of the synthesis speciosa entails a co-opting of the imagination by the understanding: he shows by this expression [synthesis speciosa] that the transcendental imagination has lost its former autonomy. It receives this name only because in it the understanding is referred to sensibility and without this reference would be synthesis intellectualis.133

Heidegger says that Kant strikes out two passages presenting imagination as the third fundamental faculty.134 The imagination is even reidentified as a ‘function of the understanding’ in a handwritten note found on the margin of Kant’s copy of the Critique, whereas the published text in both editions reads that the former is a ‘fundamental function of the soul’.135 While I have shown how the culmination of the B ‘Deduction’ in Section 26 suppresses discussion of the imagination, I have also argued that Kant’s account of figurative synthesis establishes the mediating role of the imagination. I have further argued that Section 3 of the A edition does not protect the position of the imagination in the way Heidegger suggests, but my disagreement with his account goes deeper than this and is more difficult to tease out. Heidegger’s concern for the ‘autonomy’ of the imagination may be misplaced and reveals more about his own concern for origins than it does about Kant’s faculty theory. Once we recognise that the role of 148


29/3/07

8:12 AM


The Deep Structure of Synthesis the imagination is principally one of mediation, as I have argued in previous sections, it is unsurprising that its autonomy or identity will frequently be in question. As the middle term between the principal faculties of intuition and understanding, it is both necessary and liable to inclusion in one or other pole.136 Heidegger is inclined to make the middle term into the origin. This is especially evident when he announces that the imagination is the ‘condition of the possibility of all the faculties’.137 For Heidegger, imagination is not equidistant between intuition and understanding and leans more towards the first than the second.138 He insists that intuition is the primary term for knowledge, for it is what allows us to take up an affect.139 If imagination is to hold the primordial position he claims for it, it is hardly surprising that the mediating faculty is more frequently shown in its systematic relation to intuition than to that of understanding. Heidegger is so committed to his view of imagination as a further elaboration of intuition that he goes so far as to say that transcendental imagination is time. Time is the medium within which all intuition affects us and is thus the horizon within which we encounter beings. Heidegger particularly focuses on Kant’s characterisation of time as ‘self-affection of the self’ and concludes that, ultimately, time and not space is the source of affection.140 ‘Time as pure intuition is in one the formative act of intuiting and what is intuited therein’.141 In taking this position Heidegger remains true to the letter of Kant’s Schematism chapter in which the necessity of the temporalisation, not the spatialisation, of the categories is analysed. However there are grounds for thinking that Kant’s account is insufficient, even within his own terms. I have argued in the previous chapter that the initial paragraphs of the ‘Transcendental Aesthetic’ reveal the event of the material affect of objects. Later I will argue that while his project of establishing the subjective conditions of objective objects must start with a temporal schematism, it cannot stop there.142 In order to show that mental forms grasp the material given he needs to show how the categories are combined with the spatial form of intuition as our mode of access to outer things. And while Heidegger goes on to say that space is ultimately grounded in time, thus establishing that human existence is spatial, this is not good enough for Kant’s purposes. Kant must show that the different, although intertwined, trajectories of time and space establish the condition of the possibility of the subject’s encounter with an external object. This is exactly what he tries to establish in the ‘Refutation of Idealism’.143 149


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology My present concern is with the way in which Heidegger presents the relation between the faculties. He sets up the perspective from which we can see the imagination as the relation between two extremes.144 But he falls into treating the third faculty not, as he says, as ‘only a mediating, intermediate faculty’, but rather as ‘the primordially unifying element’.145 The transcendental imagination is the ‘root’ that joins the two extremes, but it now appears that the joining in question is not so much a process, but rather a source or origin. 146 In his interpretation of the Schematism we can find a further aspect of his ontologising of the faculties while he seeks, legitimately in my view, to establish that they serve as points of access to things in the world. Whereas Kant says that the schema is a rule for the generation of images, Heidegger interprets it as a pure image.147 Even though he adds that it is the possibility of a pure image that is in question, his account encourages the view that the schemata are pure entities rather than processes.148 There can be no doubt that the unity Heidegger is seeking to establish is not static, for it is nothing other than the flowing of time towards the future and from the past towards the present. The imagination is only origin as temporal synthesis. But the tension between seeing imagination as mediator and as origin leads to Heidegger’s continually complicating its identity and the relations in which it stands to the other faculties. The complexity of his account is not wholly justified by the intricacy of the relation he is trying to express. Unnecessary complexity leading to obfuscation arises from his temptation to establish the imagination as more than merely intermediary. Heidegger’s complex presentation goes hand-in-hand with an oversimplification of the role of imagination. He seeks to make imagination first among equals, when the situation is rather that the imagination is the mediator par excellence, as I have argued in previous sections. In slipping into his preferred lexicon of ‘primordiality’ and ‘origins’, Heidegger does not go far enough in his examination of the relational status of the imagination. If he is to capture Kant’s concern with how we are affected by something given in experience, he must show how our capacity for affection allows us not only to self-affect, but also to be affected by something other than ourselves.149 Heidegger believes he can achieve this through an analysis of the temporality of human existence, or Dasein. However as Heidegger’s consideration of the primacy of temporality over spatiality already reveals, an account of self-affection simply cannot establish the possibility that Kant is aiming to legitimate, namely, our being affected by objects external to ourselves. This would require not only 150


29/3/07

8:12 AM


The Deep Structure of Synthesis time, but also space.150 Revealingly, when Heidegger cites Kant’s comment that ‘space and time must affect the concept’, he goes on to talk exclusively of time.151 The great potential of Heidegger’s interpretation of Kant has been taken further in the work of the contemporary philosopher John Llewelyn, who has revealed the role of imagination as an ongoing process, unashamed of its merely intermediary status.152 Heidegger’s own shortcomings are, I believe, related to his failure to address the role played by the imagination in the Critique of Judgement where Kant examines the faculty of judgement as such.153 Kant’s account of reflective judgement is of an operation of mediation between intuition and understanding. In the next section, I will examine how this account sheds light on Kant’s earlier analysis of cognitive judgement.

IV Synthesis as the activity of judgement As I have already suggested earlier in this chapter, I believe that there are resources in the Critique of Judgement for resolving a problem that has emerged in Kant’s epistemology, namely his uneasy identification of the role of imagination. I have suggested that the imagination is best understood as mediating between the other faculties. This perspective, while already presupposed in the first Critique, is at the same time compromised by judgement’s being linked not only to sensibility, but also to understanding. In this final section I will argue that aesthetic judgement reveals the synthetic process that is necessary for any judgement. In the course of my analysis of determining judgement, I have shown how Kant displays a further ambivalence about the limits of knowledge. At times, he suggests that all apprehension amounts to knowledge, while at others he states clearly that not all intuitions are unified under the understanding. A deeper analysis of the relation between the faculties, afforded by the Critique of Judgement, will allow us to show the importance of his retaining the possibility that there are unsynthesised intuitions. In Chapter 8 I will argue that the Critique of Judgement also allows us to resolve this problem. Aesthetic judgement is not cognitive simply because it does not aim at a determination of the object. It counts strictly as reflective; first, insofar as it aims only at an ideal unification in experience that is never actually achieved; and, second, in that we are thrown back on our own subjective activity. Nevertheless, in the course of his analysis of aesthetic judgement, Kant has much to say that illuminates his 151


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology epistemological stance. Most importantly, Kant suggests that aesthetic judgements are based on a harmony of the faculties that is necessary for ‘cognition in general’ (Erkenntnis überhaupt).154 Although the latter claim suggests a direct and problematic lineage between cognitive determinative and aesthetic reflective forms of judgement, I will argue that they stand in an indirect, yet systematic relation. Aesthetic judgements are reflective in that they throw us back on the subjective activity that makes them possible, namely, the harmony of the faculties. My suggestion will be that they also indirectly reveal the less harmonious (but, on inspection, cooperative) activity necessary for any cognition whatsoever. This activity is the subjective process that makes synthetic conclusion possible. Kant’s claim that aesthetic judgements qualify as synthetic has been roundly criticised, as it seems to suggest that aesthetic judgements fall under the rubric of cognitive synthetic judgements. I will argue that aesthetic judgements are synthetic in the sense that they display the synthetic process necessary for any cognitive judgement whatsoever. Aesthetic judgements are syntheses in action, with the important caveat that they do not aim at a cognitive conclusion. The cognitive activity of judgement – which we have just been examining in the ‘Deduction’ – is a means to an end, whereas in the aesthetic case synthetic activity is an end in itself. (i) A Re-reading of Kant’s Account of Synthesis in the Critique of Judgement The clearest statement of the synthetic status of reflective judgements comes in the first Book of the third Critique: We can readily see that judgments of taste are synthetic; for they go beyond the concept of the object, and even beyond the intuition of the object, and add as a predicate to this intuition something that is not even cognition: namely [a] feeling of pleasure (or displeasure). And yet, that these judgments are, or want to be considered, a priori judgments as regards the demand that everyone assent, a demand they make despite the fact that their predicate (of one’s own pleasure [as] connected with the presentation) is empirical, is also already implicit in the expressions used to make that claim. Hence this problem of the critique of judgment is part of the general problem of transcendental philosophy: How are synthetic judgments possible a priori?155

We have already seen in the previous section that figurative synthesis entails going beyond the mere concept of an object and seeking the intuition for that concept. A concept alone would not achieve knowl152


29/3/07

8:12 AM


The Deep Structure of Synthesis edge. Kant now says that judgements of taste share this general feature of synthesis, in that they, too, go beyond the mere concept of an object. But in this case the concept does not determine an intuition, so what does Kant mean? Moreover, for Kant, aesthetic appreciation does not start from a concept, but rather from a sensory response to something given in experience. One way in which we can make sense of his first claim is to say that an aesthetic judgement goes beyond the concept insofar as it abstracts from the conceptual determination of an object and focuses instead on the feeling we have in response to it. In finding something aesthetic, we may be well aware of what it is, but in taking pleasure in it, we abstract from considerations of identification and concentrate on the pleasurable effect it has on us. What is really crucial for the status of aesthetic judgements as synthetic is that they go beyond the intuition of the object, in that we have a feeling of the reflective harmony of our faculties that is inseparable from the intuition, qualifying the latter as aesthetic. An aesthetic judgement starts from an empirical intuition of an object, but in this case the intuition is not determined by a concept. We aim towards a concept that would explain the given intuition, but all that is available is an indeterminate idea of beauty. It is thus not so much that these judgements go beyond the concept as that they stop short of determination by the latter.156 Synthesis in general links a concept or an intuition with something further. In cognitive judgement, the concept is linked with an intuition or vice versa, but in aesthetic judgement there is a gap between the intuition and a possible explanatory concept. The concept of beauty that we attribute to this object is simply too general to determine the empirical object with which we are faced. Yet the gap between concept and intuition is not a void, for we experience an aesthetic feeling that serves as a link between the particular thing under our gaze and the indeterminate predicate ‘beautiful’. In an aesthetic judgement we become aware of the movement or activity between the two poles of a given object and a subjective reflection that would in other circumstances lead to conceptual determination. We are aware of the synthetic process that creates a link across the gap identified by dualism. We are able to remain at the level of the process because of the focus afforded by a feeling of pleasure. The feeling allows us to remain at the imaginative level of attention that in cognition was resolved into a unitary consciousness. In the aesthetic case, we stop short of unity or determination and direct our attention to the feeling. This has the result that we can become aware of 153


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology the mediating role of imagination that was necessary for, but concealed in, cognition. While aesthetic synthesis does not subsume the intuition under a concept as in the cognitive case, the intuition bears a universal value from its being associated with a very particular kind of feeling. The peculiarity of the feeling characteristic of these judgements is that it is not merely private, despite the fact that the feeling is mine and arises in response to an empirical object. Aesthetic feeling raises a claim for universal validity.157 When I say that a particular object is beautiful, I implicitly appeal to the agreement of all other judging subjects. Kant concludes the passage under consideration with the claim that the third critique contributes to the general problem of transcendental philosophy, which he identifies with the question: How are synthetic judgments possible a priori? This is because aesthetic judgements belong to the class of judgements that are under consideration in transcendental philosophy. This species of judgements is identified as synthetic, insofar as they go beyond the intuition and the concept. They also qualify as a priori in that they are universally valid for everyone. But now we are faced with two questions. Is Kant right in claiming that aesthetic judgements count as synthetic within the logic of his own argument? And, if so, does the a priori status of aesthetic judgements simply add a new species of synthetic a priori judgements, or does it illuminate the account of synthesis as a whole? My answer, which will occupy the remainder of this book, involves showing that aesthetic judgements are synthetic in that they reveal the process of synthesis that is at work in all judgements. I will now examine two passages that will provide us with a clue to the identity of synthesis in process and the relation in which it stands to determinate synthesis.158 In the published Introduction to the Critique of Judgment, Kant makes a connection between aesthetic judgement and judgment (or ‘cognition’) in general. In an aesthetic judgement, which counts as empirical and singular, there is a purposive harmony or fit between the object and the mental activity arising in response to that object: For the basis of this pleasure is found in the universal, though subjective, conditions of reflective judgments, namely, the purposive harmony of an object (whether a product of nature or of art) with the mutual relation of the cognitive powers (imagination and understanding) that are required for every empirical cognition.159

154


29/3/07

8:12 AM


The Deep Structure of Synthesis Pluhar’s translation illuminates something that is understated in the original German text. In using the expression ‘the mutual relation of the cognitive powers’ Pluhar brings out how any empirical judgement requires a cooperation of the faculties.160 An aesthetic reflective judgement is distinguished by the presentation of the object being purposive or conducive to the mental activity that arises in response to it. There is a harmony between the object and our response to it. Later Kant will say that the mental response itself counts as a harmony of the faculties.161 These two harmonies are indeed corollaries of one another. Our faculties spontaneously cooperate with one another because the object is propitious for our judgement. However, in this passage Kant also claims that the purposive harmony is – in some way yet to be established – related to a necessary condition of all empirical cognition. What is at stake in Kant’s claim that an aesthetic object stands in a purposive harmony with the mutual relation of the faculties necessary for all empirical cognitions? He surely cannot be saying that a purposive harmony between object and mind is the necessary condition of knowledge? The key to making sense of this passage is to understand that it is only the mutual relation between the faculties and not the purposive harmony between the object and the faculties that is characteristic of empirical judgements in general. Empirical judgements necessarily display some mutual relation of the faculties in that the cognitive synthesis of concept and intuition requires a cooperation between understanding and intuition. We can conclude that the play or harmony of the faculties that arises when we encounter an object we judge to be beautiful, is a heightened version of the pervasive coordination of the faculties necessary for any judgement, that is, any experience at all. In this harmony imagination stands in for intuition insofar as the former provides the conditions for the combination of the latter and thus for its formal status.162 In the harmony of the faculties imagination and understanding harmonise freely with one another. Thus the mutual relation necessary for cognition becomes apparent only in an aesthetic judgement where a specific cognitive goal is not in view. Unfortunately for the persuasiveness of a series of Kant’s arguments, he does not deploy a clear distinction between a mutual relation and a harmony of the faculties and this lack becomes more acute in the body of the Critique of Aesthetic Judgement.163 The second passage that will serve as a clue to the relation in which synthesis in process stands to determinate synthesis comes in the 155


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology ‘General Comment’, at the end of the ‘Analytic of the Beautiful’. Kant remarks that aesthetic judgements display free lawfulness: ‘the object may offer it [the imagination] just the sort of form in the combination of its manifold as the imagination, if it were left to itself [and] free, would design in harmony with the understanding’s lawfulness in general. . .’164 In a determining judgement, synthesis is governed by a rule or law introduced by the understanding. An aesthetic judgement is prompted by an object that has a form that encourages the imagination to combine the manifold in a way that is conducive to the unifying activity of understanding, that is its unification of the manifold under a law. But in this case there is no law, so the imagination’s combination counts as a free lawfulness. Kant concedes that such a notion appears contradictory.165 However, he goes on to resolve this apparent paradox by arguing that in an aesthetic judgement the imagination is capable of following the understanding without being compelled by it. The combination that would usually prepare for unification under a concept in this case stops short of its resolution. This counts as free lawfulness. We can explain this with reference to the reading I have given of the ‘Transcendental Deduction’. In a determinate cognitive judgement, the faculties cooperate when the understanding gives the rule to intuition through the intermediary of the imagination. In a reflective aesthetic judgement, the faculties cooperate without any rule being applied, however the relation in which the faculties stand to one another is so harmonious that it is as if a rule had been applied. Aesthetic judgement thus mimics cognition’s lawfulness, but does so without a law operating. This is why the aesthetic harmony of the faculties is capable of revealing the subjective structure of synthesis, that is, its synthetic activity. While determining judgements aim at syntheses as results, aesthetic judgements reveal the process of synthesis that otherwise would only be available to us in a philosophical reconstruction such as that offered by the Critique of Judgement. For the moment, I have not explained how aesthetic judgements could exhibit this general structure of cognition; I will only consider this question in the final chapter. (ii) Makkreel’s Rejection of the Synthetic Status of Aesthetic judgement It has often been concluded that the account of synthetic a priori judgements is restricted to the first Critique and its concern with 156


29/3/07

8:12 AM


The Deep Structure of Synthesis objective knowledge. Rudolf Makkreel provides one of the most systematic rejections of the claim that aesthetic judgements count as synthetic.166 First, Makkreel argues that the role of the imagination in the figurative synthesis of the first Critique is that of the handmaiden of the understanding.167 He concedes that the imagination brings some of its own role to bear in the determination of sensibility, insofar as it provides the schemata necessary to figurative synthesis.168 However, despite this concession, his reading of the ‘Deduction’ of the first Critique is impositionalist and this is the fundamental reason why he denies that aesthetic imagination is synthetic.169 Given the restricted and subservient role he apportions to the imagination within figurative synthesis, it is perhaps unsurprising that he insists on such a deep cleft between cognitive experience and what he prefers to call ‘aesthetic consciousness’.170 If synthesis is always determinative and is also central to knowledge, then Makkreel would be right to conclude that otherwise aesthetic apprehension would be reduced to a subset of cognition.171 However, I have shown that we can at least make sense of the idea of a synthesis that is not determinative. Makkreel’s reading of the B ‘Deduction’s’ account of figurative synthesis runs along the familiar lines of the impositionalist interpretation. However, he pauses at the following claim, coming, as it does, so late in the development of Kant’s account: Synthesis in general, as we shall hereafter see, is the mere result of the power of imagination, a blind but indispensable function of the soul, without which we should have no knowledge whatsoever, but of which we are scarcely ever conscious.172

He comments that this was an ‘oversight on Kant’s part’.173 In a textual correction never incorporated into a published edition, Kant amends the passage thus: ‘Synthesis in general, as we shall hereafter see, is the mere result of the power of the imagination, a function of the understanding.’174

This unpublished remark seems at first sight to cancel the distinctiveness of the faculties, resulting in there being no need for a cooperation between understanding and imagination. But should we take this comment as the decisive move in a struggle to express the relation between the faculties, evident throughout the published text? Heidegger, while also noting this correction, remarks that Kant nevertheless left unaltered the subsequent Schematism 157


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology chapter, in which the imagination is presented as the source of transcendental synthesis.175 In any case, discovering another statement in favour of the primacy of the understanding does not solve the problem, but rather reopens it. Makkreel’s assumes this unpublished remark represents Kant’s final view on the matter and is thus conclusive. This is questionable as an interpretive strategy. We cannot know this was Kant’s last thought on the matter and, even if it was, the last opinion an author expresses is not always most representative of his or her position as a whole. A further problem arises from the fact that this is not the only occasion on which Kant identifies synthesis with the imagination, making it unlikely that this particular passage counts as a mere oversight.176 Moreover, even if we did take the amendment as decisive, we would still be faced with the question of how we should interpret it. Is the imagination a function of the understanding in the sense that it is merely an exercise of the latter, or is it that imagination sometimes functions in cooperation with sensibility and sometimes with understanding, as other passages suggest? Makkreel has further reasons for denying that aesthetic judgements qualify as synthetic. He remarks that the term synthesis is never used in the accounts of aesthetic apprehension and comprehension.177 He dismisses the idea that this might be due to a wish to avoid selfrepetition, saying that Kant never hesitates to repeat himself on other occasions. This is, however, hardly a conclusive claim. As we have seen above, Kant not only announces that aesthetic judgements are synthetic, but also makes it clear that he holds them to be central to the general transcendental project of establishing the possibility of synthetic judgements. Whether he would have been willing to repeat himself or not is difficult to ascertain. Sometimes Kant is repetitive, while at other times he fails to state the obvious: there is no a priori rule for deciding which is the case here. All we can be sure of is that Kant declares that aesthetic judgements are synthetic. Makkreel’s most interesting suggestion is that reflective judgements are synthetic ‘only in form’. Aesthetic judgements: are not synthetic in the objective sense applicable to cognitive judgements, in which we add to the concept of an object a concept of one of its attributes. Instead of claiming something about the objective properties of an object, the judgment of taste discloses something about our own subjective state of mind in apprehending the form of an object.178

I agree with this and conclude that it is quite appropriate to characterise aesthetic judgements as synthetic in that they display the subjective 158


29/3/07

8:12 AM


The Deep Structure of Synthesis conditions of synthesis in general. Makkreel is prevented from coming to this conclusion because he generally treats aesthetic judgements as if they were concerned only with the subject and not with the relation between subject and object. The subjective bias of Makkreel’s interpretation is picked up by Ameriks, unsurprisingly given his own objectivist reading of Kant’s aesthetics.179 Interestingly, Makkreel does not wholly rule out that there is some relation to an object in an aesthetic judgement. Towards the end of the second section of his book, he retrospectively claims that objects were at issue and corrects his previous subjectivist bias by saying that aesthetic, like teleological, judgements refer to objects but are nevertheless not directly cognitive.180 From this perspective, it would have been possible for Makkreel to reach the conclusion that aesthetic judgements are not only formally or subjectively synthetic, but that they reveal the relation between subject and object. It would then have been a short step to the position that I will defend, namely that aesthetic judgements are capable of contributing to the general project of transcendental philosophy, namely, the project of showing how synthetic a priori judgements are possible. But this is exactly what Makkreel denies.181 I agree that aesthetic judgements are not synthetic in the same sense as is cognition. The question is: do the former reveal the deeper structure of the synthesis that is the subjective condition of cognition? This is what I will try to establish in the chapters that follow. Makkreel prefers to insist that aesthetic judgements are post-categorial, that is, that they expand rather than reveal the conditions of the possibility of cognition.182 For this reason he is not inclined to concede that aesthetic judgements reveal the deeper structure of synthesis. In conclusion of this discussion, I wish to draw attention to a particularly suggestive connection made by Makkreel. In discussion of the way in which the concept of life binds together the extremely unwieldy text of the Critique of Judgement, he suggests that this concept refers to the power to move, and that this, at its deepest level, is grounded in the life of the mind.183 Despite my disagreement with the role he apportions to synthesis, his suggestion could have initiated a very illuminating way of addressing what I am calling synthesis in process. Revealingly, Makkreel says that considering life as the power to move ‘allows us to interpret the aesthetic feeling of life as a transcendental point of unity for both the active power of the understanding and the receptivity of sense’.184 This, he believes, suggests a way of ‘mitigating Kant’s dualism of understanding and sense’.185 159


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology While I agree with the way in which Makkreel develops Kant’s suggestive, but elliptical comment on life, I have argued in this chapter that dualism is always more dynamic and more pluralist than his interpretation would suggest. At a deeper level of analysis, understanding and sense are to be seen, not as opposed, but as standing in necessary relation to one another. In order that they can relate, the intermediary capacity for imagination is required. Thus dualism is unveiled as entailing a pluralist model of mind and a dynamic relation between the faculties. Aesthetic judgements of beauty, in particular, facilitate an indirect access to the cooperation of the faculties as a ‘feeling of life’ of the mind.186

Conclusion In this chapter I hope to have established that, for Kant, synthesis relies on a cooperation of the subjective faculties or synthesis in process. This is the deep structure of any synthetic achievement. A priori synthesis, as analysed in the ‘Transcendental Deductions’ requires the activity of a plurality of orientations, even though the role apportioned to imagination is ambiguous. I have also argued that aesthetic judgement properly counts as synthetic and contributes to the general project of establishing the possibility of a priori synthetic judgements, just as Kant claims. This is because the play of the faculties is nothing other than an instance of the cooperation of the faculties necessary for any judgement, free from any cognitive conclusion. In the next chapter I will argue that the deep structure of synthesis is best understood as the subjective side of the deduction of the categories.

Notes 1. CPR, A 51, B 75, at the outset of the ‘Transcendental Logic’. 2. See Chapter 6 on the anticipatory character of knowledge. 3. Recognition of the reflective status of the understanding leaves open the possibility for comprehending the role of concepts in relation to aesthetic judgement. See Chapters 7, pp. 260–9, and 8, pp. 280–90. Longuenesse has very helpfully highlighted the role of reflection throughout Kant’s account of experience (see Chapter 2, p. 77). 4. For a discussion of the ‘task’ of knowledge, see Chapter 7, pp. 252–3. 5. See my account of empirical knowledge in Chapter 7, pp. 249–55. 6. This is especially evident in aesthetic judgement, which, Kant believes, is peculiar to human beings. See CJ, AA 210. 160


29/3/07

8:12 AM


The Deep Structure of Synthesis 7. Notable exceptions include Patricia Kitcher’s Kant’s Transcendental Psychology and Onora O’Neill’s ‘Transcendental Synthesis and Developmental Psychology’. 8. Later we will see the importance of the further faculties of reflective judgement and imagination. Both of these are ‘linking’ or relational faculties without which experience would not be possible. 9. I argue this in Chapter 7, pp. 249–55. 10. CJ, AA 218. 11. See mention of McDowell’s Mind and World in Chapter 3, p. 101. 12. See Chapter 3, p. 101, where I have already made both claims. 13. See ‘What is Orientation in Thinking?’, AA, VIII, pp. 131–47. This was first published in 1786 and thus predates the Critique of Judgement’s development of the idea of a ‘harmony of the faculties’. 14. I come back to this issue in a discussion of the ‘Dialectic of Taste’ in Chapter 8. 15. B 128. 16. A 55, B 80. 17. A 69, B 93/4. 18. A 77, B 102–3. 19. This is the highly criticised ‘Metaphysical Deduction’ recently defended by Béatrice Longuenesse. See Kant and the Capacity to Judge, pp. 73–80. 20. A 63, B 87–8. 21. A 63, B 87–8. 22. A 89/90, B 122. 23. A 89, B 121–2. 24. As I argued in Chapter 3, pp. 105–8. 25. A 89–90, B 122. 26. See discussion of A 116 below. See also discussion of B 160–1 below. 27. On this question, see Chapter 3, p. 102. 28. See Allison, KTI, pp. 133–5; and Henrich, ‘The Proof-Structure’. 29. A 115. 30. See discussion below. 31. See discussion of Longuenesse in Chapter 2, pp. 70–1. Patricia Kitcher, in Kant’s Transcendental Psychology, suggests that ‘normative and factual claims commingle in transcendental philosophy’ (p. 21). 32. A 99. 33. A 99; ‘Impressions’ translates Eindrucke. 34. A 99. 35. See discussion of Longuenesse in Chapter 2, pp. 72–8. 36. A 97. 37. A 97. 38. B 160 note. 39. A 94. 161


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 40. A 98–9. 41. A related problem arises towards the end of the account of the synthesis of reproduction in imagination where Kant says that the ‘purest and most elementary representations of space and time’ require the operation of imagination (A 102). I believe that this should be understood in parallel with the problematic footnote at B 160–1 where, as I read it, Kant says that pure geometric forms, not the forms of space and time, require a synthesis of imagination. See discussion below, pp. 139–43. 42. See discussion below, p. 138. 43. A 100. 44. A 102. 45. This is what phenomenologists sometimes call the ‘living present’. 46. A 102. 47. Kant remarks that it is because reproduction is necessarily bound up with apprehension, one of the transcendental acts of the mind, that reproductive imagination also qualifies as transcendental (A 102). He withdraws this status in Section 3. See below, p. 127. 48. A 102. 49. See discussions of Section 3 of the A edition and of B 160–1 in the second edition below, pp. 127–30 and pp. 145–6. 50. A 103. 51. A 103. 52. See discussion in Chapter 3, p. 102. 53. A 115. 54. A 115. 55. A 119. 56. A 118. This is in contrast to his earlier claim that the reproductive synthesis of imagination counts as transcendental. See A 102 . 57. A 120. 58. A 118 (my emphasis). 59. See discussion of Heidegger below, pp. 147–51. 60. A 118. 61. A 121–2. 62. A 123. 63. A 123. 64. A 119/120. 65. A 120. 66. A 116. 67. A 116; ‘wenigstens müssen verknüpft werden können . . .’ 68. A 120. 69. A 120 (my addition). Compare the very similar claim in the footnote at B 160–1 discussed below, p. 137. 70. A 120. 162


29/3/07

8:12 AM


The Deep Structure of Synthesis 71. 72. 73. 74.

75.

76. 77. 78.

79.

80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92.

93. 94. 95. 96. 97.

A 121. A 121. A 121–3. A 124. See, also, discussion of an almost identical claim at A 78, B 123, in my discussion of Makkreel in the final section of this chapter, pp. 157–8. See, however, the next section for a discussion of Makkreel’s assessment of an emendation to the B edition where Kant seems to withdraw imagination’s status as a fundamental faculty. I argue, however, that this withdrawal cannot be taken as decisive. Sarah Gibbons also thinks the distinction is of level, not of type of synthesis. See Kant’s Theory of Imagination, p. 40. See discussions of Allison and Longuenesse in Chapter 2, p. 64 and p. 70. B 151. Kant also calls this combination through the understanding in contrast to figurative synthesis, which he associates with the imagination. This distinction suggests that the understanding must cooperate with imagination if knowledge is to arise. This, however, can only count as an abstraction from a more concrete level of experience, if we take seriously Kant’s claim that thought always refers to intuition (A 19, B 33). See discussion in Chapter 3, p. 101. B 136/7. B 144–5. B 145. B 150. B 145. B 147. B 150. B 151. B 151–2. B 152. A 120. The link between imagination and form is also assumed in Kant’s account of figurative synthesis in Section 24 of the B edition. B 151. This is also why the form of the object is so important in aesthetic judgement’s non-cognitive exhibition of the general possibility of cognition. See Chapter 8, pp. 280–90. See Chapter 2, pp. 72–8 on Longuenesse for both these issues. See my ‘Taste as Productive Mimesis’. B 160. B 160. B 160–1. I will return to the footnote later, restricting my reconstruction for now to the main text. 163


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 98. 99. 100. 101.

102. 103. 104.

105. 106. 107. 108. 109.

110. 111. 112. 113. 114. 115. 116. 117. 118.

119. 120. 121. 122. 123. 124. 125.

126.

127.

B 161. B 161. B 161. This distinction is repeated on several occasions in Section 26. See, for instance, the beginning of the note at B 160. See also his example of the apprehension of a house at B 162. See, again, B 164. B 160. A 89–91, B 122–3. In my ‘Kant’s Phenomenological Reduction?’, I argue that the forms of space and time are analogous in status to a phenomenological ‘horizon’ (Hughes 2007). B 161. B 161. See discussion of this point in Chapter 2, pp. 72–8. B 161. I argued that the distinction between forms of representation and the representations in space and time was also in force in the synthesis of apprehension in the A ‘Deduction’. B 161 (Kemp Smith’s emphasis). B 161. B 162 note. B 162 note. See below, pp. 151–6. B 161. B 161. B 161. B 161. As I argued in Chapter 3, pp. 101–2. Later I discuss how aesthetic judgements relate to cognition in general and yet are not cognitive in status. A 98–100. A 100–3. A 120. B 162 note. B 164. B 162 note. B 151. In the last section we saw that Kant says imagination belongs to sensibility just as he says that it is an action of the understanding. Heidegger, Kant and the Problem of Metaphysics (= KPM); Heidegger, Kant und das Problem der Metaphysik. For an incisively critical, while sympathetic account of Heidegger’s reading of Kant, see John Llewelyn, ‘Heidegger’s Kant and the Middle Voice’. Heidegger, KPM, p. 206; references to German edition in brackets (p. 195). 164


29/3/07

8:12 AM


The Deep Structure of Synthesis 128. At its deepest level, Being is temporal and human beings are beings for whom time is at issue. To consider things ontologically and not merely ontically is to consider them in their temporal givenness and to understand our own finitude in respect to that givenness. 129. It is tempting to think that Merleau-Ponty’s notion of a chiasm – that is, a relation between two intertwined and ultimately inseparable beings – would have been helpful for the complex set of relations Heidegger tries to express. See Merleau-Ponty, The Visible and the Invisible, Ch. 4 ‘The Intertwining – The Chiasm’. 130. Heidegger, KPM, p. 175 (p. 164). 131. See, for instance, Heidegger, KPM, p. 252 (p. 236), where Heidegger says that time, which has now been identified with imagination, ‘is essentially one with pure apperception’. 132. Ibid., p. 167 (p. 155). Leidenschaftlichen could also be translated as ‘passionate’. 133. Ibid., p. 170 (p. 159). 134. Ibid., p. 167 (p. 156). He refers to CPR, A 94 and A 115. In both passages Kant says there are three subjective sources of experience. It must be admitted that the suppression of these passages is significant. 135. Heidegger, KPM, p. 168 (p. 156). See A 78, B 103. Heidegger gives the reference as Nachträge XLI. Makkreel also comments on this alteration as I discuss below. He refers to AA XXIII, 45. 136. I discuss a related question about the ‘heautonomy’ of aesthetic judgement in my ‘Taste as Productive Mimesis’ (Hughes 2006). 137. Heidegger, KPM, p. 154 (p. 148). 138. Ibid., pp. 135, 136, 137 (pp. 123, 124, 125). 139. Ibid., p. 28 (p. 21). ‘Cognition is primarily intuition’. Heidegger’s point is that intuition stands in immediate relation to objects. 140. See discussion of self-affection in Chapter 3, p. 97. 141. Heidegger, KPM, p. 180 (p. 169). 142. See Chapter 6, pp. 229–37. 143. Again, see Chapter 6, pp. 234–5. 144. See, for instance, Heidegger, KPM, p. 86 (p. 78). 145. Ibid., p. 202 (my emphasis) (p. 190). 146. Ibid., p. 152 (p. 140). 147. Ibid., pp. 108–9 (p. 100). Pippin’s interpretation is rather similar to this. See Chapter 1, p. 14. 148. Ibid., p. 109 (p. 100). 149. See my discussion of affection in Chapter 3, pp. 105–8. 150. Heidegger sees this problem and intends to address it in On Time and Being. As Llewelyn remarks, there are already many spatialising elements in Being and Time; see his ‘Heidegger’s Kant and the Middle Voice’, p. 115. 165


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 151. Heidegger, KPM, p. 193; German edition, p. 182; where he refers to CPR, A 77, B 102. 152. See not only Llewelyn’s ‘Heidegger’s Kant and the Middle Voice’, but also his The HypoCritical Imagination, especially the Prologue; and The Middle Voice of Ecological Conscience: a chiasmic reading of responsibility in the neighbourhood of Levinas, Heidegger and others, especially the Preface and Ch. 10, ‘Something like the middle voice’. 153. Heidegger remarks that he cannot discuss the Critique of Judgement; see KPM, p. 167; German edition, pp. 155–6. 154. CJ, AA 218. Although Kant’s claim clearly covers all cognition and not merely that which gives rise to objective knowledge, our concern here will be entirely with the latter. 155. CJ, AA 288–9, Section 36, ‘On the Problem of a Deduction of Judgments of Taste’. 156. My concern in this discussion is to provisionally identify the primary structure of aesthetic judgements and to establish the relation in which they stand to cognition. Judgements of taste also go beyond the concept in the sense that they say more than any particular conceptual determination could express. But this over-determination of sense arises from the lack of determinacy characteristic of any aesthetic judgement. It is because an aesthetic judgement rests only on the activity of judgement in general that it is open to a range of different determinations. Makkreel suggests that aesthetic judgement is postdeterminative, rather than pre-determinative; see the discussion in the next section. Although I cannot argue the case here, my view is that aesthetic judgements can be post-determinative just because they are, primarily, pre-determinative. 157. CJ, AA 211–12. 158. For a related discussion, see Howard Caygill, Art of Judgement, pp. 318–20, on formative activity and judgement. 159. CJ, AA 191. The following is the crucial passage in the original version: ‘mit dem Verhältnis der Erkenntnisvermogen unter sich, die zu jedem empirischen Erkenntnis erfordert wird (der Einbildungskraft und des Verstandes)’. 160. Makkreel also makes a helpful contrast between the ‘accord’ necessary for any cognition and the ‘agreement’ characteristic only of aesthetic judgements. See Makkreel, Imagination and Interpretation in Kant, p. 62. However, I disagree with his claim that it is only in the aesthetic case that the faculties must adapt to one another. 161. CJ, AA 217/18. 162. See discussion of the ‘Transcendental Deduction’, especially in the A edition, above. 163. I will discuss the problems this gives rise to in Sections 9 and 21, in particular. See Chapter 5, pp. 174–6 and p. 188. 166


29/3/07

8:12 AM


The Deep Structure of Synthesis 164. CJ, AA 240/1. Additions to passages from the Critique of Judgement in square brackets here and elsewhere are Pluhar’s, unless otherwise signalled. 165. AA 241. 166. See Makkreel, Imagination and Interpretation in Kant (= IIK), pp. 47–51. See also his ‘Response to Guenter Zoeller’. 167. Makkreel, IIK, p. 30. 168. Ibid., p. 30, where he says that imagination introduces its own formative power. Consistently, Makkreel suggests that the schemata are necessary for the determination of a concept. See, for instance, p. 124 and 129. Sometimes he additionally talks as if schemata were already at issue in the account of figurative synthesis in the B ‘Deduction’. See IIK, p. 31. This position bears some similarity to Longuenesse’s suggestion that the schemata come before the categories. She means, of course, that the schemata are prior to the explicit determination of intuition by concepts. See Chapter 2, pp. 72–8. 169. See especially IIK, pp. 42 and 46, where he identifies figurative synthesis with ‘definite objective cognition’. 170. See IIK, pp. 4, 45, 47–9, 52, 67, 73, 92, 106, 118. See also, ‘Response to Guenter Zoeller’ p. 278. 171. Though my reconstruction of the Deductions shows that he would not be right to conclude that Kant’s position is simply impositionalist. 172. CPR, A 78, B 103. 173. Makkreel, IIK, p. 29. 174. Ibid., p. 29. He refers to XXIII, 45. 175. Heidegger, KPM, p. 168 (p. 156). 176. See, for instance, CPR, A 124; also in the A ‘Deduction’ and discussed earlier in this chapter. 177. Makkreel, IIK, p. 48. 178. Ibid., p. 48. See also ‘Response’, p. 277. 179. See K. Ameriks, ‘Rudolf A Makkreel, Imagination and Interpretation in Kant’ (= ‘Review’), pp. 232 and 233, referring to Makkreel, IIK, pp. 53 and 63 respectively. For Ameriks’ own position, see ‘Kant and the Objectivity of Taste’, and ‘How to Save Kant’s Deduction of Taste’. 180. Makkreel, IIK, p. 99. 181. As does Allison, KTI, p. 172. 182. See, for instance, Makkreel, IIK, p. 47. See also p. 57. The contrast between the pre- and the post-categorial is used by Ameriks, ‘Review’, p. 230, referring to IIK, p. 45. Ameriks’ remarks on a pre-categorial reading Makkreel attributes to Guyer and to an objection raised by Meerboote, namely that such a position leads to the conclusion that all objects are beautiful. (I discuss this problem in Chapter 8, pp. 284–90.) See Makkreel, IIK, p. 49; Guyer, Kant and the Claims of 167


29/3/07

8:12 AM



183.

184. 185. 186.

Taste, pp. 86 ff. Meerboote, ‘Reflection on Beauty’, in Essays in Kant’s Aesthetics, ed. Ted Cohen and Paul Guyer, p. 79. Ameriks then argues for an alternative and, he believes, more successful argument for the pre-categorial significance of taste. The general conditions of cognition are made use of by, but are not sufficient conditions of, taste. See Ameriks, ‘Review’, pp. 233–4; Ameriks, ‘How to Save Kant’s Deduction of Taste’, pp. 295–302. I agree with Ameriks that a pre- and a post-categorical account of aesthetic judgement need not conflict. However my account of the pre-cognitive role of aesthetic judgement is quite different from his. I will argue in the next chapter that it is the relation in which we stand to the object, rather than the object per se, that is central for aesthetic judgement. See especially Makkreel, IIK, pp. 105–7. See also Douglas Burnham, Kant’s Philosophies of Judgement, pp. 149 ff., see especially p. 172. Aesthetic judgement is schematising without a concept and this is possible because a free sensible lawfulness can be felt. Later Burnham goes on to investigate the way in which feeling reveals the structure of life. Makkreel, IIK, p. 106. Ibid., p. 106. CJ, AA 204. My turn of phrase here allies Kant’s phrase with Hannah Arendt’s expression ‘the life of the mind’; see Hannah Arendt, The Life of the Mind.

168


29/3/07

8:12 AM


5

The Completion of the Subjective Deduction in the Deductions of the Critique of Judgement

In the previous chapter I suggested that aesthetic judgement reveals the synthesis in process necessary to all judgements. In this chapter I argue that synthesis in process is best understood as the subjective side of the deduction, often referred to as ‘the subjective deduction’. In the first section (pp. 170–6) I discuss Kant’s distinction between subjective and objective deductions and insist that these are two sides of the deduction, rather than two separate deductions. The subjective side of the deduction is the cooperation of the faculties, or synthesis, necessary for any judgement. However at this stage of his presentation, Kant is hesitant, although not entirely negative about the significance of the faculties for his epistemology. I suggest that the positive presentation of the subjective deduction is provided in the Critique of Judgement. I take as a clue Kant’s statement that all objectively valid judgements are also subjectively valid. I then show how aesthetic judgements – bearers of subjective universality – are not solely a class of judgements set apart, but are internally linked to the structure of judgements already analysed in the first Critique. In the second section (pp. 177–89) I discuss the claim in Section 21 of the third Critique that aesthetic judgement contributes to a nonsceptical epistemology by uncovering a principle of common sense. I assess the degree to which Kant is successful in avoiding the conclusion that cognition is dependent on aesthetic judgement. In contrast to the interpretation of Henry Allison, I suggest that this section counts as a first attempt at providing a deduction of aesthetic judgement insofar as it offers a completion of the subjective deduction. Allison denies that Section 21 is concerned with aesthetic judgement and I discuss his interpretation in the third section of this chapter (pp. 189–93). Although Kant does not as yet provide a convincing deduction, I cannot accept Allison’s view that Section 21 has merely epistemic resonance. I agree, however, that cognition entails a feeling and that sensus communis has a cognitive application. 169


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology In the fourth section of this chapter (pp. 193–201), I turn to the structure of the official deduction. I argue that the distinctive contribution of the latter is its focus on the ‘power of judgement as such’ and show how the latter is only revealed directly in judgements of taste. However, I also suggest that the failure of this version of the deduction is that there is no explanation of how aesthetic judgement relates to the subjective conditions of cognition.

I The subjective and objective deductions On first sight, it may seem that a central passage in the first Critique runs totally against the emphasis I have put on the cooperation of the faculties and the account of synthesis I have built upon it. In the Preface to the A edition, Kant distinguishes the subjective and objective deductions. The subjective deduction is the account of the role of the faculties in cognition and, as we will see, is given at best an ancillary role. But we will see that it is not so much that there are two deductions – one subjective and one objective – but that there are two sides to cognition. On one side there is the determination of the object by the understanding, while on the other is the cooperation of the faculties necessary for that determination. Recognising this is the first step to showing how the aesthetic play of the faculties reveals the necessary condition for knowledge of objects in the world. The ‘subjective deduction’ is often taken to refer either to the A ‘Deduction’ in general or more precisely to its second section, where Kant outlines the three syntheses necessary for knowledge.1 Admittedly, Kant makes the distinction between the subjective and objective deductions only in the Preface to the A edition, but this distinction reveals a general structure of his method. I will now examine the problematic status of the ‘subjective deduction’ in its first and only explicit appearance in the text and show how Kant holds out a tentative promise of its having a more positive role. While this alternative role can be discerned implicitly throughout the arguments of the first Critique, it is not until the third Critique that its full potential is realised. In the Preface to the first edition of the Critique of Pure Reason, Kant says that the subjective deduction is of great importance, but not essential for the main argument which is supplied by the objective deduction. Kant says that there are two sides to the enquiry ‘for exploring the faculty which we entitle understanding, and for determining the rules and limits of its employment’.2 He refers, almost in 170


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement passing, to these two sides as the objective deduction and the subjective deduction, saying of them: The one refers to the objects of pure understanding, and is intended to expound and render intelligible the objective validity of its a priori concepts. It is therefore essential to my purposes. The other seeks to investigate the pure understanding itself, its possibility and the cognitive faculties upon which it rests; and so deals with it in its subjective aspect.3

The objective deduction establishes that the categories are valid in respect of the objects that provide the content of the judgements we make using our power for unification, the understanding. The subjective deduction reflects back on that capacity. Kant elaborates that such a reflection would concern the faculties upon which the understanding rests. Why does he not simply refer to the capacity of understanding alone? It is because, as we saw in the last chapter, the power of understanding is only able to achieve unification of objects insofar as it cooperates with the faculties of intuition and imagination. For his immediate purposes, Kant announces the relative lack of importance of the subjective side of the deduction: Although this latter exposition is of great importance for my chief purpose, it does not form an essential part of it. For the chief question is always simply this: – what and how much can the understanding and reason know apart from all experience? not: – how is the faculty of thought itself possible?4

Kant refers us forward to the ‘Transition to the Deduction’ at A 92–3 for a discussion of the objective deduction. This passage is repeated in the B edition and serves as an introduction to both editions of the ‘Deduction’ in providing a short account of their shared purpose.5 The argument is almost identical to the account of the Copernican revolution presented in the Preface to the B edition and discussed in Chapter 3, pp. 89–95.6 Just as in the latter passage, in the ‘Transition’ Kant says that either the object makes the representation possible or vice versa and opts for the latter option, but now he goes on to clarify something that was not stated in the account of the Copernican revolution, namely, that this does not entail that the representation produces the existence of the object and only that it determines the latter.7 We can conclude that insofar as the representation does not ‘produce’ the object, Kant distinguishes his formal idealism from material idealism. As I also argued in Chapter 3, pp. 100–5, representations take up the affect given in sensibility and thus we are not condemned to the circle of representation. 171


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology In the ‘Transition’, Kant anticipates the ‘Deduction’ concluding that: The objective validity of the categories as a priori concepts rests, therefore, on the fact that, so far as the form of thought is concerned, through them alone does experience become possible. They relate of necessity and a priori to objects of experience, for the reason that only by means of them can any object whatsoever of experience be thought.8

The objective deduction qualifies as objective insofar as it establishes the application of the categories to all objects of experience. But are the subjective and objective sides so distinct from one another? What is to be proven by the objective deduction is that a subjective faculty, the understanding, is capable of introducing order into the objective world. Experience becomes possible through a form of thought. This shows that the objective deduction is dependent on the operation of a subjective faculty, the understanding. Moreover, the cooperation of faculties already observed in the initial statement of the Copernican revolution is still required in the ‘Transition’, for Kant insists that there must be two conditions under which the knowledge of an object is possible. There must not only be a concept, but also an intuition.9 In the previous chapter, I discussed the way in which faculty talk is used to draw out the character of figurative synthesis in the B ‘Deduction’. The latter form of synthesis prepares for the application of categories to empirical objects and its distinctiveness lies in being presented as a three-term relation among the faculties, in contrast to the two-term intellectual synthesis. The faculty talk is not restricted to Section Two of the A ‘Deduction’, although it is undoubtedly highlighted there. Throughout the ‘Analytic’ of the first Critique, and in both editions, Kant develops his account of the rules and limits of the understanding – that is, of the objective deduction – using a vocabulary of the cooperation between the faculties of intuition and understanding. I have given a case study of this in my interpretation of the B ‘Deduction’ and close examination of the ‘Schematism’ and of the ‘Principles’ will reveal that Kant continually presents his argument in terms of a relation between understanding and intuition.10 Often commentators regard Kant’s faculty talk as, at best, a stylistic irrelevance and, at worst, a serious mistake, resting on a covert metaphysical agenda. Kant himself remarks that the subjective side of his undertaking is of great importance and, against the grain of his initial characterisation of the situation, I will argue that it can be 172


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement established as a necessary condition of cognition and thus of the objective deduction. It is tantalising that even in the initial statement in the A Preface where Kant characterises the subjective deduction as not essential, he suggests that he intends to give a further defence of its status. He first remarks that it may seem ‘hypothetical in character’ insofar as it seems to involve ‘the search for the cause of a given effect’.11 This would involve trying to establish how the faculty of thought itself is possible, in contrast to restricting investigations to the question ‘what and how much can the understanding and reason know apart from all experience?’12 Thus the danger is that the subjective deduction appears to go beyond the limitations of the transcendental analysis of knowledge in search of a cause for thought. This would be to seek a transcendent explanation of experience, whereas Kant insists that it is only valid to seek out the transcendental structures of experience. He is worried that the subjective deduction risks seeking a genetic explanation for thinking, which would involve deriving the latter from a metaphysical or psychological foundation. Kant goes on to say that he will show elsewhere that the subjective deduction is not hypothetical in character.13 I take this as showing he believes that it is possible to investigate the subjective conditions of knowledge on the basis of a properly transcendental method. But it is unclear how – or, indeed, where – he intended to return to this question. My suggestion is that, whatever his intentions at the time of writing the Preface to the first Critique, the argument of the third Critique, with its establishment of the synthetic activity of the faculties as a pre-condition for the possibility of any cognition, fulfils the promise made earlier. The faculty talk is no longer at risk of straying into a search for the transcendent origin of thought and is given a transcendental role insofar as the activity of thinking is revealed as providing the subjective, though universal and necessary, form of experience. As such, the subjective deduction is no longer simply an inessential, though important vehicle for the expression of the possibility of knowledge and has become part of the internal structure of the ‘chief purpose’, that is, of the objective deduction. At this stage the subjective and objective trajectories truly become ‘two sides’ of one argument. If we are to determine ‘how much understanding and reason can know apart from all experience’ we cannot ignore the activity of thinking, which on deeper investigation is revealed as a cooperation of the faculties necessary for any judgement. 173


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology If we now turn to Section 8 of the third Critique, we find a clue to the link between objective and subjective validity: Now a judgment that is universally valid objectively is always subjectively so too, i.e., if the judgment is valid for everything contained under a given concept, then it is also valid for everyone who presents an object by means of this concept. But if a judgment has subjective – i.e., aesthetic – universal validity, which does not rest on a concept, we cannot infer that it also has logical universal validity, because such judgments do not deal with the object [itself] at all. That is precisely why the aesthetic universality we attribute to a judgment must be of a special kind; for although it does not connect the predicate of beauty with the concept of the object, considered in its entire logical sphere, yet it extends that predicate over the entire sphere of judging persons.14

All objective cognitive judgements are valid objectively insofar as they determine the object by means of a concept. They state, for instance, that such and such a thing is an x. They thus achieve a determining or explanatory synthesis vis à vis the object. But in doing so, they also bear subjective universality because they must be valid for all judging persons. Grasping the object simply means to represent reality in a way that is valid for everyone. Thus the objective deduction necessarily has a subjective dimension in commanding the agreement of all judging subjects. In contrast, aesthetic judgements do not determine the object.15 They are, however, subjectively valid and thus valid for all judging persons. In Section 9 Kant develops his account of subjective validity, arguing that aesthetic judgements display the ‘merely subjective determining basis’ of all cognitions.16 They are characterised by a ‘relation between the presentational powers insofar as they refer a given presentation to cognition in general’.17 He calls this a ‘free play’ in which the cognitive powers harmonise.18 Kant concludes that this relation must hold for everyone and thus counts as universally communicable in that it is the subjective condition of cognition.19 Thus the subjective validity of aesthetic judgements arises from their being based on a relation between the faculties necessary for any judgement whatsoever. The subjective validity that was established as necessary for any objective judgement in Section 8 is revealed in the following section as the subjective cooperation of the faculties on which aesthetic judgements are exclusively based. In Section 9, Kant’s account of the relationship between aesthetic judgement and cognition reveals a problem that haunts his account on many occasions. How can it be that aesthetic judgement is distinct 174


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement from cognition and yet displays the necessary subjective conditions of cognition in general? If cognition were founded on the specifically harmonious play of the faculties, then it would count as aesthetic. We can assume that Kant does not want to argue this. From the outset of the first Critique he is committed to the distinctiveness of cognitive and aesthetic principles, while he comes latterly to recognise their systematic connection. However, the establishment of that systematic connection causes him many problems. For the moment he is at risk of suggesting that the harmony of the faculties characteristic of aesthetic judgement is necessary for judgements giving rise to knowledge. There are two possible amendments that could solve the local problem in Section 9. Either ‘cognition in general’ is taken to refer to something distinct from the general case of cognition, or, the harmony of the faculties is exemplary of, but not identical to, the subjective grounds of cognition. In the first case ‘cognition in general’ would be taken to refer to the general possibility of cognition displayed in aesthetic judgements, despite the fact that they are not cognitive. This general possibility would stop short of actual cognition, establishing only its form. Aesthetic judgements would display the form of cognition, that is, its subjective conditions, but not its actuality. In an aesthetic judgement the faculties are prompted to cooperate in a way that is particularly conducive for cognition, but no knowledge is achieved because our mental activity is so free that it stops short of the determination that is the objective condition of cognition. It is as if we were engaged in a cognitive exercise, but our exploration is so free and open-ended that cognition will not be achieved. The other option is to retain the natural understanding of ‘cognition in general’ as referring to all cognition and fine-tune Kant’s argument so as to say that aesthetic judgement is characterised by a harmonious relation of the faculties, which is a heightened expression of the mutual relation of the faculties necessary for any cognition.20 As in the previous interpretative option, aesthetic judgement displays the form of cognition, but is not included within the range of the latter.21 The two interpretations converge insofar as they establish that aesthetic and cognitive judgements are systematically connected, but distinct from one another. This is surely what Kant intends, although he has not yet established the apparatus to express his position. We will see that in later discussions he makes progress in articulating his intricate position, while never quite pinning it down. For the purposes of 175


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology our current discussion both amendments allow the same conclusion, namely, that aesthetic judgement is subjectively valid insofar as it reveals the subjective activity necessary for any cognition whatsoever. We saw in the last chapter that aesthetic judgement is characterised by a double harmony insofar as the thing we deem beautiful is purposive for our mental activity, while the latter is characterised by a harmonious play. While this does not amount to a determinate grasp of reality, as in the cognitive case, we nevertheless feel that it is appropriate to attribute such a harmony to all other judging subjects. The subjective validity of the aesthetic judgement is based on the latter’s being grounded in the subjective activity displayed by all judging subjects, which makes possible our response to an object in any judgement. Thus the subjective side of the deduction – that is, the cooperation of a plurality of the faculties – and the relation in which that subjective side of judgement stands to the objective side, is brought to the fore in aesthetic judgement. It is this link to the general form of cognition that establishes the universal validity of what would otherwise be merely subjective. Aesthetic judgements reveal the subjective side of the deduction, as first outlined in the Preface to the A edition of the first Critique. Moreover, they do so in such a way as to show that the subjective deduction is not to be understood hypothetically or genetically, but rather as part of the formal structure of validity in general, a validity that always has two sides, that of the subjective, but universal structure of judgement and that of the latter’s application to an intentional object. Put in this way, we can see that the subjective side of the deduction is an essential component of establishing objectivity. Thus we can justify the suggestion I made in the previous chapter that aesthetic judgements contribute to the general transcendental project of establishing the possibility of a priori synthetic judgements, for they reveal the synthetic activity of the faculties necessary for any cognition. Although there is a systematic link between aesthetic and cognitive judgement insofar as they both rely on a subjective process of synthesis and both can claim subjective validity, they are distinct. In the absence of objective validity, subjective validity functions not just as one side of a larger story, but as a constitutive principle in its own right. It is for this reason that it becomes possible to inspect the subjective side of the transcendental project in a way that was not possible when the principal concern was with the determination of knowledge per se. However Kant only achieves this distinction in the official version of the deduction, as we will see. 176


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement

II The anti-sceptical force of aesthetic judgement Kant does not, however, believe that he has provided the definitive account of the relation between aesthetic and cognitive judgement in Section 9 of the Critique of Judgement. In the fourth Moment of the ‘Analytic of the Beautiful’, he provides an argument for the position he has already sketched in the second Moment. In Section 21, Kant claims that aesthetic judgement is based on a principle that is necessary for any non-sceptical logic or epistemology. While this may appear to be a remarkable claim, it was in fact entailed in nuce in the earlier position we have just examined. What is clear is that he cannot mean that the harmony of the faculties characteristic of aesthetic judgement is a necessary condition of determining cognitive synthesis. This would indeed qualify aesthetic judgement as necessary for resisting scepticism, but would at a stroke render otiose the distinction between determining and reflective judgement.22 In the light of my reading of the ‘subjective deduction’, we can now make sense of his claim, explaining how judgements that are strictly subjective in their universality nevertheless have a role to play in what must be a cognitive argument. We will now discover the helpfulness of the clue discovered in Section 8, namely, that there are necessary subjective conditions for every objective judgement. The link Kant makes between aesthetic judgements and antisceptical arguments is significant for a theme to which we will return. It seems that it would be too simple to take at face value the many occasions on which Kant says that aesthetic judgements have nothing to do with the object. In the passage from Section 8 cited in the last section, he says first that ‘such judgements do not deal with the object at all’. However, he then goes on to say that they do ‘not connect the predicate of beauty with the concept of the object, considered in its entire logical sphere’.23 This second, more careful statement leaves open the possibility that aesthetic judgements relate to the object, while not determining it under a concept. If aesthetic judgements contribute to anti-sceptical arguments, we can say, at the very least, that they stand in some relation to objects insofar as they reveal the necessary, though insufficient, subjective conditions for the possibility of a successful objective judgement. The general goal of the fourth moment of the ‘Analytic of Taste’ is to establish that the liking we feel for beautiful objects is not merely contingent, but necessary. In Sections 18 and 20, Kant argues that if a pleasure is necessarily associated with judgements of taste, then they 177


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology must be determined by a subjective principle that grounds their universal validity.24 This is the principle of common sense presupposed in all aesthetic judgement. Common sense is not to be confused with outer sense, as it is ‘the effect arising from the free play of our cognitive powers’ (die Wirkung aus dem freien Spiel unserer Erkenntniskräfte).25 We might be tempted to conclude from this that common sense is the empirical effect of a priori activity. But Kant cannot mean this, for common sense is the principle on which aesthetic judgements are based. As such, it is a transcendental principle of experience and is distinct from the other principles we have examined so far, strictly in that it functions as a feeling. For a clarification of Kant’s elaboration of common sense we need to look to Section 40 where Kant characterises it as ‘an effect that mere reflection has on the mind’.26 Common sense is a feeling that arises from the activity of the mind and acts as the principle governing aesthetic judgements, that is, it supplies the standard for the latter. Insofar as aesthetic judgements presuppose common sense, they rest on our ability to respond to objects with a free play of the mind that gives rise to a feeling of pleasure. In responding aesthetically to things we judge them on the basis of a free play of the faculties. The feeling of common sense is this play of the faculties affecting our minds so as we become conscious of it, at least implicitly. In Section 20, common sense is linked with aesthetic judgement, but later it emerges that it has a wider application. Common sense is, thus, an ability to coordinate a plurality of distinct mental operations, but, for Kant, this description is not restricted to the mental activity of a solitary judging subject. Judgements based on common sense are necessarily communicable to other judging subjects, as we will see in what follows. Kant thus establishes the inter-subjectivity of mind in the course of his analysis of judgements of beauty. The title of Section 21 asks ‘[w]hether we have a basis for presupposing a common sense’.27 Given the preparation for this question in the discussion of aesthetic common sense in Section 20, it would be very unnatural if Kant now restricted his discussion, as Allison thinks he does, solely to cognitive judgements.28 Admittedly Kant fails to make an adequate distinction between cognitive and non-cognitive judgements, as we will see. This, however, results not in a wholly epistemic proof as Allison suggests, but in an epistemic bias in the way Kant connects aesthetic to cognitive judgement. As I read this section, the problem Kant now poses is this: how can the aesthetic principle introduced in the previous section count as subjectively necessary for 178


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement experience? He attempts to answer this affirmatively by showing that common sense is a necessary subjective presupposition of ‘cognition in general’. The argument is most naturally divided into seven steps.29 After a promising start where he refers to judgements in general, he restricts the scope of his argument to cognitive judgements. This restriction operates as the effective premise of the rest of his argument and results in a progression from cognition, via aesthetic judgements, to common sense, culminating in the claim that common sense counts as a presupposition of cognition. As Henry Allison has recently put in question the idea that this section is concerned with aesthetic judgement at all, I must reconstruct its stages in some detail. My aim is to show how Kant’s account provides an extended, although flawed, argument for the claim he made in Section 9, namely, that aesthetic judgements display the subjectively valid structure necessary for all cognitive judgements. The first step refers to judgements in general, not only determining ones or cognitions. Nevertheless, insofar as Kant focuses on the problem of grasping objects, he seems to be concerned principally with judgements giving rise to knowledge of objects. 1. Cognitions (Erkenntnisse), judgements and the conviction accompanying them must be communicable. Otherwise they would amount to no more than a subjective play of the faculties and there would be no guarantee that they grasp their objects, just as scepticism suggests.30 Kant’s claim is that if cognitions are to count as objectively valid – and thus give rise to knowledge – then their truth must be communicable, that is, they must be acknowledgeable by all judging subjects. Something that is known is not simply the opinion of one or more subjects. Kant thus begins his argument by restating the connection between what, in Section 8, he called objective and subjective validity. Objective validity requires subjective validity, that is, the agreement of all judging subjects. While admitting the plausibility of Kant’s claim for the inter-subjective communicability of knowledge, Guyer fails to find a successful defence for this thesis at any stage in the critical opus.31 Kant’s position in the ‘Transcendental Deduction’ of the Critique of Pure Reason is one of methodological solipsism in Guyer’s view and he sees this as standing in tension with the development of Kant’s argument in Section 21.32 His charge that Kant fails to provide a convincing argument for the inter-subjective status of cognition is a fair one, for if any 179


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology is provided, it is at best indirect. Nevertheless, while it is true that the ‘Deductions’ – and even more relevantly the ‘Analogies’ – distinguish between subjective and objective order without any explicit reference to inter-subjectivity, Kant surely held as unquestionable that his analysis was of mind in general and not of an independent subjective psyche.33 And if it is necessary for any mind to apply the same epistemic rules, then, in principle at least, a space is left for communication between minds at the transcendental level. Admittedly, this potential is insufficiently developed in Kant’s epistemological account of the status of mind and we must conclude that the premise of Kant’s argument in Section 21 is not subjected to proof because it counts as an unquestioned and undefended axiom. But, while not proving the validity of this fundamental presupposition, the account of communicability in the third Critique reveals Kant’s commitment to it, not only in aesthetic judgements, but for cognition in general. The second stage of Kant’s argument deepens the analysis of the subjective side of cognition. He moves from the subjective conviction characteristic of any successful claim to knowledge, to the subjective grounds that make that conviction possible. Notably he now speaks only of cognitions and not of judgements in general. 2. If claims to knowledge are communicable, then the attunement or proportion of the faculties, as the subjective condition of those claims, must also be communicable.34 The reference to the proportion or relation of the faculties shows that this is the real locus of the subjective side of the deduction. A valid claim to knowledge and its attendant conviction are communicable because they are based on a cooperation among the faculties and this, too, must be shown to be universally valid. As Guyer remarks, this does not simply repeat the subjective aspect of the first step, but extends the latter, introducing the ‘“subjective condition” of knowledge’.35 But knowledge can only be based on subjective capacities that are shared by all judging subjects and thus cannot be merely psychological as Guyer concludes.36 Cognitions must display a universal validity, complementary to that of the objective side of the deduction of cognitive claims. The subjective condition of cognition that Kant now calls the attunement or proportion of the faculties is what, in the previous chapter, I identified as the cooperation or mutual relation of the faculties necessary for any cognition.37 The next two sentences of Kant’s argument correspond with two distinct steps. First, he claims: 180


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement 3. This attunement actually takes place when an object induces the imagination to combine the sensory manifold and the imagination in turn induces the understanding to provide unity for this manifold in concepts.38 Kant now makes the crucial claim that the subjective conditions necessary for objective validity do, in fact, hold. Our faculties actually cooperate in the way that allows them to grasp an object on condition of a dual operation. Kant’s account here recalls the two stages of the Copernican revolution.39 It is also very similar to the account of synthesis given in the A ‘Deduction’.40 On the one hand, the object affects us in such a way that intuition in collaboration with imagination combines the sensible given. On the other hand, understanding unifies what has been ‘run through and held together’ by the imagination.41 The third step provides an anatomy of the proportion or cooperation among the faculties that the second step claimed was necessary for any cognitive judgement. There is no reason to suspect that Kant is, as yet, speaking of the special proportion characteristic of aesthetic judgement. This account of the attunement or proportion of the faculties counts as the subjective side of the deduction. The determination of the object only occurs insofar as there is a cooperation of the cognitive faculties. Thus, the subjective side of the deduction is not merely hypothetical but necessary for cognition. The objective deduction is only concluded when its subjective side is taken into consideration. However, we have not yet heard the full story. 4. There is however variation in the proportion of this attunement, ‘depending on what difference there is among the objects that are given’.42 This is the crucial distinction that makes possible the introduction of aesthetic judgements in the fifth step and for this reason I identify it as a distinct step, in contrast to Allison.43 Not all cooperation among the faculties is identical. But how can there be a difference of proportion among cases in which the understanding unifies the manifold? We can concede that the content of judgements about objects vary according to their referent, but does it make any sense to say that the formal subjective conditions of cognition vary? Surely any cognitive judgement will display the same proportion or relation of the faculties, because otherwise they would not count as subjectively universal? 181


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology But Kant is not trying to establish a diversity of subjective conditions among cognitive judgements, the formal conditions of which are consistent. He is, rather, preparing the ground for establishing the distinctiveness of a different set of judgements – aesthetic judgements – while maintaining their systematic connection to cognitive judgements. Aesthetic judgements, as a class, display a different attunement that will be described in the fifth step. Admittedly, the difficulty of establishing that this is Kant’s intent is compounded by the fact that the third step appears to refer only to cognitive judgements characterised by unification. This encourages the conclusion that Kant is now making a distinction within the range of determining judgements. As the special attunement he is about to identify is characteristic of aesthetic judgements, this would mean that the latter are a subset of judgements displaying unity. But aesthetic judgements qualify as non-cognitive insofar as they are not determined, that is, unified. They remain strictly reflective. What Kant should have said was that all judgements entail that the manifold is taken up through a combination of imagination and understanding, but there are, nevertheless, some that do not attain unity. In these judgements there is a qualitatively different relation among the faculties that nevertheless reveals the general subjective condition of cognition. He would then be in a position to say that the peculiar subjective proportion of the non-cognitive aesthetic judgement is capable of revealing the mutual relation necessary for any cognitive judgement. The problem about the range of the third step can be traced back to the second step, and even to the elaboration of the first step. While Kant starts his putative deduction of common sense from the perspective of the range of judgements in general, he quickly suggests that he is concerned only with cognitive judgements. A problem arises when he needs to distinguish the species of aesthetic judgements from cognitions in the fifth step. Kant should have maintained the position that his argument concerns judgements in general, within which he could have distinguished determining and aesthetic judgements. This would have resulted in the following amended argument: [Step 1] All judgements entail universality of communicability and this is true even of those that bear objective validity. [Step 2] Universality of communicability entails that the subjective conditions of judgements must be based on a cooperation of the faculties, characteristic of all judging subjects. 182


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement [Step 3a] This attunement of the faculties does in fact take place when the manifold of sensation is unified in a cognitive judgement. [Step 3b] But a subjectively universal relation between the faculties is also displayed by aesthetic judgements, which appear to follow a law or rule, while not giving rise to determinate unity.44 [Step 4] The particular relation between the faculties in any judgement depends on variation in the objects under consideration. This would allow him to proceed to the next stage of his argument, namely, [Step 5] Only aesthetic judgements display the particular proportion of the faculties that qualifies as a harmony. I suspect that Kant thought he could move directly from cognitions to aesthetic judgements because he was committed to the view that the latter rest on the subjective structure of ‘cognition in general’. In the course of my discussion of Step 4, I have touched on an important move that once again puts the impositionalist reading of Kant under pressure. Kant distinguishes not only between different forms of judgement, but also says that the difference in attunement between judgements rests on a difference among the objects. Different objects give rise to different subjective affects and thus have a role to play in the content of cognition. But this does not mean that the object qualifies as a pure given. The object can only give rise to an affect insofar as it sets the subjective faculties in motion. What we finally know is the object as it has been taken up through our powers of representation. There is no epistemic thing-in-itself, but only an object that stands in relation to a judging subject. Nevertheless, formal idealism requires that the material existence of the object supplies the content for cognitive judgements. Form takes up content and is not simply an imposition on the latter. I have also argued that the variation in proportion in question is between the forms of cognitive and aesthetic judgement. If this distinction arises from a difference among objects, then it strongly suggests that, for Kant, there is something in the object that gives rise to aesthetic pleasure and that only some objects will give rise to judgements of taste.45 Kant now moves to discuss the special case of aesthetic judgements: 5. But there must be one attunement ‘in which this inner relation is most conducive to the (mutual) quickening of the two mental 183


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology powers with a view to cognition (of given objects) in general: and the only way this attunement can be determined is by feeling (rather than by concepts)’.46 Following the adjustment I have suggested, namely that Kant should have consistently referred to judgements in general and not just to those giving rise to cognition, we can now make sense of the distinction expressed in this step. Judgements determined by a feeling of common sense are aesthetic and are characterised by a cooperation of the faculties that is highly congenial to cognition. If this reconstruction is right, then it might strike us as paradoxical that no knowledge arises. Surely, if the activity of the faculties is most conducive to cognition, cognition will result! The answer lies in Kant’s claim that the attunement is conducive to ‘cognition in general’ (Erkenntnis überhaupt). This is the same phrase that was used in the problematic account from Section 9 we have already considered in the previous section and for which I have suggested two possible interpretations. The crux of both options is that aesthetic judgement displays the formal conditions of cognition, while falling short of actual knowledge. Guyer, as I have already noted, concludes that aesthetic judgement can only have a psychological status due to a problem he finds in Kant’s account of proportion. How can it be that all cognitions display a proportion or attunement between the faculties and yet aesthetic judgements display a different attunement?47 If all judgements display the same attunement of the faculties, this would render impossible a distinction between cognition and taste, resulting in the conclusion that all things are beautiful. On the other hand, Guyer argues, if the difference between proportions is stressed, then it would mean that aesthetic judgements are not linked to the transcendental ground of cognition, and rather count as merely psychological variations on our capacity to judge.48 But the key to answering Guyer lies in challenging his reading of attunement as implying the unification of the manifold. Following the adjustment to Step 3 that I have suggested, and making use of the terms I introduced in the previous chapter, the attunement of the faculties is the activity of synthesis that results only in some cases in unification of the manifold. If we distinguish between synthesis as activity and synthesis as result, we can make sense of the distinction between different proportions of judgement without breaking the link between aesthetics and cognition. In an aesthetic judgement we are 184


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement aware of the sort of synthetic activity that would, under other circumstances, give rise to a determination or unification of the manifold. However, on this occasion a determinate unification does not arise because of certain qualities of the object and the way in which we respond to it. Guyer assumes that either aesthetic judgement can display an optimal attunement, in which case in both forms of judgement the proportion must always be the same, or taste fulfils no such role in relation to cognition.49 But why should this be so? This would mean that if cognition and aesthetic judgement are linked, then they must have an identical subjective formal structure. For Kant, the formal structure of a judgement is the subject of a transcendental investigation. It was necessary to write three Critiques and not just one, because the formal structures of cognition, morality and aesthetics are not the same as one another. Nevertheless, the three main orientations in human life are connected, not only empirically insofar as we connect all three in living our everyday lives, but also because there are systematic connections between the distinct forms by which we introduce order into our lives. Aesthetic judgement’s form is distinguished insofar as aesthetic judgement displays a different attunement of the faculties from the relation required for cognition. Despite my conviction that the text can be reconstructed so as to make sense, I have argued that it does not work as it stands. I am particularly uneasy with the stipulative style in which Step 5 is expressed. Kant says that there ‘must’ be an attunement most conducive to cognition. Why, we might ask, must there be such an attunement? There is no reason in the world why there must be, unless we believe, for instance, that there is a metaphysical order of things that dictates that beauty and knowledge are, perhaps at some higher level, one. Kant’s presentation in Step 5 should have followed that of Step 3 where he makes a claim about the actuality and not merely the possibility, of cognitions. He should have said that there are, in fact, experiences – namely aesthetic ones – in which we are aware of a particularly conducive relation among the faculties. This is a fact of experience from which transcendental analysis begins. At the end of this stage of his argument, Kant says that, in these special cases, the relation of the faculties is determined by a feeling. Why must the attunement be determined at all, especially given that it counts as free? The answer for this comes in the first sentence of Step 6, which builds on Step 2’s axiom that the subjective condition of cognition – the proportion of the faculties – must bear a universal validity 185


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology comparable to objective validity and, indeed, necessary for the latter. Both the objective and subjective conditions of knowledge must be communicable. Any judgement insofar as it is universally valid must be governed or determined by a principle. If this were not the case, my judgement would simply be for me and not for all judging subjects, as all objective judgements and all subjectively valid judgements must be (see Steps 1 and 2). Insofar as they bear subjective universal validity, aesthetic judgements are determined by a feeling. Therefore, Kant now extends his account of the universal validity of the subjective conditions of knowledge and insists that feeling too must be communicable. The next move in Kant’s account may appear to contain two steps: 6. Additionally, this attunement and the feeling arising from it must be universally communicable. The communicability of a feeling presupposes common sense.50 However I have already shown how the first sentence follows from Step 2. Thus, there is only one genuinely new piece of information.51 The second sentence, as stated above, could mean one of two things. Either the claim is that if feelings are acknowledgeable by all, this can only be because there is a shared common sense on which those feelings are based. Or it could mean that common sense is the only feeling that bears universal communicability. In the first case feeling would be based on a further principle, the status of which would have to be established. But Kant has already said in Step 5 that attunement is determined by a feeling. We must therefore conclude that the subjective principle that determines aesthetic judgements is a feeling.52 Aesthetic judgements presuppose common sense, so is the latter aesthetic? It certainly is tempting to conclude so. We have already seen that, in Section 20, common sense was identified as ‘the effect arising from the free play of our cognitive powers’, a distinctively aesthetic characterisation.53 Section 21, as I have reconstructed it, has progressed from a limitation of the argument to a consideration of cognitive judgements to the special case of aesthetic judgements. It is only at this stage that Kant introduces common sense as the ground of the latter, so it seems natural to conclude that common sense, too, is aesthetic. Now if the feeling characteristic of aesthetic judgement were based on a further principle of common sense, it would be arguable that the latter need not be aesthetic. However, I have already argued that common sense is a feeling. Does this mean that the reading of Section 21 developed here must conclude that common sense is aesthetic in status? 186


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement Kant does not see fit to clarify his position and we are left with the now familiar problem of an insufficiently articulated account of the relation between cognitive and aesthetic judgements. As a result, some commentators have concluded that here he makes cognition dependent on an aesthetic ground, whereas, going in the opposite interpretative direction, Allison has argued that Section 21 is not concerned with the relation between aesthetics and knowledge at all. In contrast to both these options, I believe that Kant is trying to establish that aesthetic judgements reveal a broader characteristic of ‘cognition in general’, that is, the subjective conditions of cognition that establish the subjective validity of both cognitive judgements and of aesthetic judgements. All judgements are based on common sense, that is, the ability to coordinate a plurality of faculties exercised by a community of judging subjects. This is the form of cognition, as I suggested in my discussion above of the two possible connotations of ‘cognition in general’. If Kant is to make his deduction of taste as presented in Section 21 persuasive, he requires, in addition to the adjustments I have suggested above, a distinction between two levels at which common sense operates. Such a distinction can be discovered in the official ‘Deduction of Taste’, although even there interpretative work is required in order to draw it out.54 Only in aesthetic judgement does the subjective condition of all cognition count as a principle in its own right, that is, as a special principle of reflection.55 We could say that this is a subjective principle of cognition, but it is preferable, I think, to say that the aesthetic principle reveals the subjective side of all cognition. The principle of common sense reveals the capacity for common sense displayed in all our judgements. Thus all judgements are based on a capacity for cooperation by the faculties that is written large and exemplified in what is later given the Latin tag of sensus communis, that is, the principle of aesthetic judgement. The reconstruction of how aesthetic judgement exhibits the form or possibility of cognition will be explored in the final chapter of this book. Kant concludes his argument, claiming that common sense is necessary for epistemologies that reject scepticism. As I have suggested, his view should be that the principle of, although not the general ability for, common sense is aesthetic. This would mean that aesthetic judgements have a part to play in the project of establishing the legitimacy of claims to knowledge. They do so, however, only in an indirect fashion, the nature of which will only be clarified at a later stage of my account. 187


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 7. We do have a basis (and not a merely psychological one) for assuming such a common sense as the necessary condition of the universal communicability of our cognition. This must be presupposed by any logic and any non-sceptical account of knowledge.56 Kant believes he has achieved the task he set himself in Section 20. Judgements of taste successfully appeal to the necessary agreement of all judging subjects insofar as they are based on common sense. And, as I have rearticulated his argument, Kant has shown that it is valid to presuppose common sense as the principle for aesthetic judgements because the latter express the subjective condition of all judgements. Thus construed, Kant’s claims that common sense is required for any account of knowledge committed to providing an alternative to scepticism is not implausible. It is merely the culmination of the steps that have come before. Common sense arises as a particular attunement of the faculties in aesthetic judgement, but, despite its particularity, it bears a general significance insofar as it is the relation between the faculties ‘most conducive’ to ‘cognition in general’. Common sense thus reveals the cooperation or mutual relation of the faculties that is necessary if the subjective conditions of claims to knowledge are to be validated. And we have seen that this is also necessary if the objective conditions are to be legitimated. This is exactly what any non-sceptical account of knowledge and, in particular, Kant’s transcendental project sets out to achieve. A successful account of knowledge will need to show that the faculties cooperate in such a way as to be capable of taking up something extra-mental given in experience. Aesthetic judgement displays a common sense that counts as a heightened example of the mutual relation necessary for any cognition. The problem remains, however, that Kant has not yet established the status of common sense. Steps 5 and 6 seem to suggest it is aesthetic, while Step 7 leads us to believe common sense is cognitive. I have suggested a solution that would resolve this tension, but in Section 21 at least Kant provides no such distinction between common sense as a general cognitive capacity and sensus communis as a principle. Moreover, the narrowing of range from judgements in general in Step 1, to cognitive judgements in Step 2, makes the later transition to taste extremely awkward. Kant has not yet arrived at an account that is capable of expressing both the systematic connection and distinctiveness of cognition and taste. 188


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement What is, however, clear from Section 21 is that the objective validity of cognitive judgements is dependent on the establishment of their subjective validity. The subjective activity of the faculties is necessary for objective validity.

III Allison’s Epistemic Reading of Section 21 Allison concludes that while it is normally assumed that Section 21 supplies a first attempt at the deduction, it should be read strictly epistemically and is concerned only with sensus communis as a precondition for knowledge and thus does not attempt to establish that there is an aesthetic presupposition of all cognition. 57 The problems we have found in the presentation of Kant’s argument in Section 21 explain why Allison opts for a wholly epistemic reading. Moreover, it is only much later that Kant presents what he explicitly calls the ‘Deduction of Taste’. Nevertheless, the cost is that what should have been the core of the fourth Moment of taste becomes a diversion. And if Kant has not been concerned with showing that judgements of taste are based on a principle of common sense, we would have to conclude that the fourth Moment has not even attempted to establish that judgements of taste exhibit subjective necessity or make a valid appeal for universal agreement. This is because the appeal to universal assent concerning taste, which Kant believes it is the object of the fourth Moment to establish, arises from aesthetic judgements being based on the subjective conditions of cognition in general. If Section 21 is not concerned with taste, then Kant has not given us any reason to believe that the latter makes any such appeal to universal agreement. My account diverges most radically from Allison’s at the step I have identified as Step 5, where Kant introduces the idea of an ‘optimal attunement’.58 My reconstruction of this step is very close to Allison’s version, the first and relevant part of which reads: Nevertheless, there must be one optimal attunement, that is, one in which the inner relation is most conducive to the mutual quickening of the cognitive faculties with a view to cognition in general . . .59

Allison argues that if, as is normally the case, this is taken to refer to aesthetic judgement, then it would ‘provide a transcendental grounding for taste by linking it directly to the conditions of cognition’.60 Allison says that there are at least two problems with such an argument in addition to the fact that it ‘does not answer the skeptic’.61 189


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology Allison now discusses these two problems, which he considers to be major ones. Firstly, he raises the question: why should the common sense that must be presupposed as a condition of the communicability of cognition have anything to do with the one that supposedly must be presupposed as a condition of taste? This, he says, is particularly implausible given Kant’s critique of Baumgarten’s conflation of cognition and beauty at the outset of the first Critique.62 The second problem is, he says, the converse of the first. If we read Section 21 as a deduction of the principle of taste, ‘then it follows that the aesthetic common sense or taste must itself be presupposed as a condition of cognition’.63 This is impossible, Allison continues: if one keeps in mind that the common sense at issue in the case of taste is the effect of the free play of the cognitive faculties. There is simply no way in which a feeling resulting from the noncognitive condition of free play could serve as a condition of cognition.64

Allison concludes that to read Section 21 aesthetically renders its conclusion not merely unconvincing, but incoherent. Allison is quite right to ask if there must be a connection between the communicability of cognition and that of taste. However, the way in which he pursues this problem is not very persuasive. Surely it is clear that Kant shifted position from the stance he took in his comment on Baumgarten early in the first Critique, which, in any case, he revised significantly in the second edition. The original version of the footnote argued that a critique of taste is empirical and psychological and thus does not qualify for the title ‘aesthetic’, which should be restricted to the science of sensibility.65 In 1787, Kant drew back from his former position, saying only that taste is not based on determinate laws and suggesting that a critique of taste could perhaps count as ‘aesthetic’. The very writing of the Critique of Judgement is testament to his abandoning the view that the aesthetic, in the sense of taste, has merely empirical status, though he retains his commitment to the view that the principles of taste are not determinate. And while he still insists that cognition and beauty should not be conflated, he is no longer of the opinion that beauty lies outside the range of transcendental investigation. Consequently, the content of Section 21 is unlikely to be illuminated by a remark made before the author realised the possibility, or indeed the necessity, of writing the book within which its arguments appear. But it is even more odd that Allison sees the footnote in the first Critique as relevant for this particular discussion, for he later supports the view that aesthetic judgement reveals 190


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement the subjective conditions of cognition.66 Agreed, Kant owes an explanation of how the principle of common sense can apply to both aesthetic and cognitive judgement, but there can be no doubt that he has moved far beyond the position he espoused in the footnote about Baumgarten. The only real problem that Allison raises is that if Kant suggests that common sense is aesthetic in status, insurmountable difficulties would result for his claim that it is also the basis for cognition. I have already suggested a solution to this problem in the previous section, namely, that common sense reveals the subjective side of cognition but only qualifies as a principle in the aesthetic case. This amendment to the text of Section 21 would allow Kant to avoid making an aesthetic principle the ground of cognition. While this entails going beyond Kant’s own argument, it does I think, show that there is no need to diverge from Kant’s trajectory so radically as does Allison in his reading of Section 21. While Allison may be more faithful to the integrity of the section than I have been, he renders its place within the wider text highly problematic. This removes a great deal of the plausibility from his epistemic reading, which I will now examine in its own right. The sticking point, as Allison realises, for his alternative epistemic reading is how one is to construe the ‘optimal attunement that can be determined only by feeling’ without any reference to the feeling proper to taste.67 He suggests that we can understand this by reference to the peculiar talent for judgement ‘which can be practised only and cannot be taught’.68 He suggests that it follows from Kant’s characterisation of judgement ‘that the subsumability of an intuition under a concept must be immediately seen, that is, “felt” ’.69 Allison goes on to suggest that this feeling for the attunement of the faculties is nothing other than the feeling we have for our capacity to judge. If this feeling and its object are to qualify as universally communicable in the manner necessary for a non-sceptical position, then, Allison argues, Kant is right to claim that we must presuppose an epistemic and non-psychological common sense.70 Allison concedes a weak relevance of Section 21 for the broader aesthetic argument of the fourth Moment in that, construed epistemically, it has the additional merit of showing that the sensus communis of taste, from the discussion of which Kant has digressed in Section 21, is at least not incoherent in principle.71 If I have shown that the aesthetic reading need not fall into the trap Allison outlines, there is no reason why my account cannot be compatible with many 191


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology aspects of his general interpretative approach. In particular, the link Allison makes between the feeling of common sense and the capacity for judgement is an interesting and important one.72 I agree that for Kant judging in general requires some feeling of the fit between the capacities for intuition and understanding.73 In contrast to Allison, I see no reason to conclude that there is more than one feeling of judgement or sensus communis, while agreeing that feeling has a role not only in aesthetic judgement, but also, tacitly at least, in cognition. Common sense is at work in cognition, but as its subjective condition and not as a principle. The subsumption effected by judgement is possible because of the latter faculty’s capacity for coordinating a plurality of faculties. This cooperation of the faculties is normally unremarked on by us, but in aesthetic judgement the usually invisible subjective side of cognition becomes available to us in abstraction from its normal entailments. In this regard, the aesthetic principle of judgement uncovers the wit of judgement, insofar as the latter is a capacity for connection among the faculties.74 At the point at which common sense becomes a principle determining (aesthetic) judgement, it also becomes available for our reflection. It is not so much that there are two distinct feelings, but rather two different relations in which we stand to the feeling that arises as ‘an effect mere reflection has on the mind’ in the aesthetic case.75 Cognition necessarily entails a balancing or harmonising of understanding and intuition through the intermediary role of the imagination. To balance the faculties in this way is to be aware, although only implicitly as our attention lies elsewhere, of an activity, a potential tension and a possible balance within cognition. I agree that it makes sense to say that this is a sort of feeling. But it is so subliminal, so unremarked as to be almost invisible to us when we are concerned to arrive at some end of cognition.76 In the aesthetic case we are not dealing with a wholly different sort of feeling, but rather can become aware of the feeling of the activity of judgement in a way that cognition simply does not allow.77 The cooperation or activity of mediation necessary for cognition is now a focus in its own right. In taste the subjective conditions of judgement are held up for inspection in a way that transfigures their operation within cognition. Taste is only achieved once these subjective conditions are reflected on in their own right and this gives rise to an explicit feeling or an effect. But this also reveals that in cognition there is an aesthetic dimension, namely, something must be given in intuition if it is to be taken 192


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement up by the understanding. This process requires a cooperation of the faculties, which only becomes available to us in experience at a second level at which Kant uses the term aesthetic, that is, within aesthetic judgements. These reveal the subjective side of the deduction and at the same time the necessary sensory dimension of our knowledge. These two senses of aesthetic are at play in the title of this book.

IV The arguments of the ‘Deduction’ If I am right and Kant already offers an argument for the universal communicability of taste in Section 21, then why does the ‘Deduction’ of the principle of taste only come much later in the text? Allison is in a position to answer this, because he reads the earlier account as concerned only with epistemic common sense. In order to make sense of the structure of Kant’s account of aesthetic judgement, Allison re-positions a distinction deployed by Kant in the Critique of Pure Reason. Kant distinguishes between the investigation of matters of fact, and the establishment of the legitimacy of those facts, by referring to two distinct questions, namely, quid facti? and quid iuris?78 Transcendental philosophy is concerned only with the latter question, insofar as it seeks to legitimate the claims to knowledge that we do, in fact, make. Allison suggests that this distinction can be used to illuminate the relationship between the ‘Analytic of the Beautiful’ and the ‘Deduction’ finally supplied in Section 38. While the four Moments analyse the characteristics of a pure judgement of taste, the ‘Deduction’, he argues, addresses ‘whether a judgment that meets the conditions of purity can make a rightful demand on the agreement of others’.79 In other words the ‘Analytic’ establishes what a pure judgement of taste would be like, whereas the ‘Deduction’ establishes its possibility. This is a helpful heuristic device for reading the extremely complex structure of the third Critique but is rather neater than a reading of the text supports. Moreover, a distinction Kant uses to mark the limits of the transcendental project is resituated within the transcendental analysis of judgements of beauty. Allison concedes that the fourth Moment, centred on Section 21, adds nothing to the content of the judgement, although he argues that it has a bearing on the content insofar as it aims to unite the three earlier elements in the idea of a sensus communis.80 Establishing the role played by the fourth Moment leads Allison to argue against Guyer, for whom the second Moment’s claim for the 193


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology subjective universality of judgements of taste already counts as a first attempt at a deduction, while the fourth is yet another attempt.81 Allison replies that we should distinguish the second Moment’s claim for the non-private status of aesthetic judgements from the fourth Moment’s concern with the claim we make on others to agree with the latter.82 While Allison is right to draw out the distinctiveness of the second and fourth Moments, I share Guyer’s sense that the official ‘Deduction of Taste’ in Section 38 emerges out of a series of attempts that are not wholly distinct from one another. The text would suggest that it was not of the utmost importance for Kant that one stage of his argument be wholly distinct from its preparation at a previous stage. For Kant, the fact that the fourth Moment concludes the identification of the character of taste does not rule out that it might additionally function as an attempted deduction of the validity of pure judgements of taste. But if, as I have suggested, Section 21 is a first attempt at a deduction of taste, then why does Kant feel the need to refine his account further in Section 38? While Kant may not argue in a series of welldefined and distinct steps, no more does he merely repeat himself without aiming at moving his position forward. How does the official deduction deepen the account of the subjective conditions of ‘cognition in general’ that was first raised in Section 9 and returned to in Section 21? The new element in Kant’s account is that of the power of judgement as such. While Kant previously talked of the conditions for judgement in general, he now takes as his theme the power that makes judgements possible. It is the link between this power and taste that will finally provide the latter’s legitimation. The first stage in Kant’s final attempt at establishing the legitimacy of judgements of taste begins in Section 35. The title of this section establishes the stakes: ‘The Principle of Taste is the Subjective Principle of the Power of Judgment as Such’.83 The judgement of taste has as its basis ‘only the subjective formal condition of a judgment as such’, that is, ‘the very ability to judge. i.e., the power of judgement’.84 This statement leaves open the possibility that taste is founded on a principle of judgement that is not aesthetic, as Allison suggests. Kant goes on to say: When we use this power of judgment in regard to a presentation by which an object is given, then it requires that there be a harmony between two presentational powers, imagination (for the intuition and the combination of its manifold) and understanding (for the concept that is the presentation [Vorstellung] of the unity of this combination).85 194


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement We might suspect that here we re-encounter the ambiguity that haunted Kant’s earlier attempts at providing a legitimation of taste. For while Kant seems to be talking specifically about aesthetic judgements insofar as he refers to a harmony between the faculties, he could be talking about judgements in general. If he were, he would once again be guilty of introducing a harmony of the faculties when he should have spoken only of the subjective conditions of non-aesthetic cognition.86 However, the preceding and succeeding discussion is exclusively concerned with taste and Kant’s intent is to establish the principle that governs the latter. The mention of harmony in this passage is associated only with aesthetic judgements, as it always should be. But if this is right, then the passage also suggests that the power of judgement exclusively give rise to aesthetic judgements, whereas does it not also give rise to a range of other species of judgements? The solution to this problem lies in the insight that a judgement of taste is characterised by the power of judgement being exercised in and for itself, in contrast to the cognitive case where judgement is the means to a cognitive conclusion. Kant immediately goes on to argue that, as a judgement of taste is not based on a concept of an object – that is, does not subsume an intuition under a concept – then it ‘can consist only in the subsumption of the very imagination under the condition for the understanding to proceed in general from intuition to concepts’.87 He calls this ‘schematising without a concept’ and says that this counts as a subsumption of the ‘power of intuitions or exhibitions (the imagination) under the power of concepts (the understanding)’.88 This, he concludes, is what he means by a harmony of the faculties.89 The exercise of the power of judgement in its own right is peculiarly allied to judgements of taste insofar as they have as their ground the power of judgement per se. Kant clarifies his position further in Section 36 insofar as he says that in a judgement of taste the pure power of judgement is ‘subjectively, object to itself as well as law to itself’.90 We can understand this claim in the following way. In judgements of taste we have a rare opportunity to reflect on judgement itself, or more strictly, on the subjective conditions of its operation, just because the cooperation of the faculties becomes visible. If this is right, then it is not puzzling that it is at this point that Kant claims that judgements of taste are part of the general problem of transcendental philosophy. However, his argument is much more modest than it need be. He merely claims that as aesthetic judgements are synthetic then they belong to the general project of establishing the possibility of synthetic judgements a 195


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology priori.91 He could have gone much further and argued that aesthetic judgements do not just belong to the class of synthetic a priori judgements, but that they also have a special status among the latter insofar as they offer a reflection on the very activity of judgement itself.92 This is surely where his account of the relation between judgements of taste and the activity of judgement as a power leads. Judgements of taste would thus qualify as transcendental judgements par excellence. Section 37 appears to specify further what is to be deduced. In Section 21 the principle of common sense was identified with a feeling of a certain proportion among the faculties. Kant now says that it is not the pleasure, but the universal validity of this pleasure that counts as a universal rule for the power of judgement.93 This sounds as though Kant has withdrawn the claim that the principle of taste is a feeling, in favour of a rule prior to feeling. Yet, elsewhere he consistently characterises aesthetic judgements as based on a feeling: only this marks them out as aesthetic. In his discussion Kant shifts between two levels of talking about liking.94 At the empirical level we experience a ‘liking’ (Wohlgefallen) for a particular object. This liking is based on a feeling of pleasure (Gefühl der Lust) in the free play of the faculties. It is tempting to conclude that when Kant says that not pleasure, but the universal validity of pleasure, is a universal rule for the power of judgement, he is excluding my liking for a particular empirical object and not the feeling of reflective activity or proportion among my faculties that is the ground of that liking. The feeling of a proportion among my faculties, in contrast to my particular liking for something, could then still count as a rule for aesthetic judgement. But the problem with this solution is that Kant consistently distinguishes between Wohlgefallen, which arises at the empirical level and Lust, which arises only from the reflective cooperation of the faculties.95 I can only conclude that while it is indeed the feeling, and not some prior rule, that is universal, the former is universal only insofar as it operates as a rule. Kant’s intention in Section 37 is to stress the peculiar status of the feeling that, as common sense, is the ground of the possibility of judgements of taste. Feeling operates as a rule because it expresses the subjective conditions of judgement and not because it expresses a preference for a particular object. So, in Section 36 when Kant mentions both ‘a feeling of pleasure [Lust] (or displeasure) and a liking [Wohlgefallen] that accompanies the objects’ presentation’, he must mean that I like this object, but I do not expect others to like it in exactly the same way as I do. 96 My 196


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement liking is unavoidably empirical and thus is, in part, determined by my own empirical history and projects. What I expect of others is that the object I appreciate will encourage a freedom of mind that I discover in myself as a necessary concomitant of my empirical liking, insofar as the latter qualifies as aesthetic. It is this feeling of aesthetic freedom that I expect others to share. Thus what is to be deduced in Section 38 remains the principle of common sense, which is a feeling of a reflective proportion in response to a given empirical object. The ‘Deduction’ proper offered in Section 38 pulls together the strands of the account developed by Kant in the previous three sections. In a judgement of taste, the liking we have for an object arises from ‘our mere judging of the form of the object’. 97 As I will argue in the final chapter, this must be the spatio-temporal form of the object, which in some – but only some – cases, gives rise to aesthetic pleasure.98 The liking (Wohlgefallen) for the object is for its subjective purposiveness with the power of judgement.99 This is the ‘mere judging prior to any concept’ identified as the power of judgement in Section 35.100 Although Kant does not as yet explicitly say so in Section 38, we already know from Section 35 that the judgement of taste rests on the subjective conditions of judgement, which are, as a cooperation of the faculties, the power of judgement as such. And now we see that these subjective conditions are not wholly detached from the object that would, in the cognitive case, be determined by them. This will allow Kant to establish the grounds for a coherent account of the relation in which taste stands to cognition, while maintaining the distinctiveness of their respective status. Whereas in a cognitive judgement the understanding would determine the form of an object under a concept, in an aesthetic judgement the power of judgement operates in response to that same form, without determining it. Kant’s next step explicitly reintroduces the idea that the power of judgement is directed only to the subjective conditions of any employment of judgement. On the basis of this characteristic he concludes that taste is based on a power that can be presupposed of all people. Were a judgement of taste determined by a particular content of sense or by a concept of the understanding, it would not be based on a general condition of all judging subjects. This subjective condition is required for ‘possible cognition as such’.101 Thus, as we already discovered in Section 35, the exercise of the power of judgement is the cooperation of the faculties that is necessary for all judgements. In the judgement of taste, however, this cooperation is experienced as available for our reflection and counts as a free play or harmony. 197


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology Kant concludes that the harmony between the presentation or the form of the object and the subjective conditions of cognition is valid for everyone.102 Judgements of taste are proven to bear the subjective universality first claimed for them in Section 8 insofar as they are based on the subjective coordination of the faculties that is necessary for any cognition whatsoever.103 Kant goes on to say that this is equivalent to the claim that when we judge an object of sense and are aware through a feeling of pleasure of its subjective purposiveness for our judgement, we are justified in requiring that the pleasure we feel is shared by all.104 The pleasure that has been deduced as universally valid is that taken in the harmony between the form of the object and the activity of our subjective faculties, not our empirical liking for a particular object. Strictly speaking I can only call on others to display the form of aesthetic judgements, that is, the ability to judge on the basis of a free play of the faculties.105 Thus the pleasure arising from the exercise of the power of judgement in a judgement of taste is proven to be universally valid insofar as it rests on the subjective conditions of ‘cognition in general’. My main disagreement with Allison concerns the status of the principle of taste deduced in Section 38.106 Allison suggests that taste is grounded on a principle that is ultimately not aesthetic, that is, the principle of the subjective activity of judgement, whereas I hold that the aesthetic principle of judgement is the only pure expression of the subjective activity of the faculty of judgement. Allison’s point is made especially clear at an earlier point when he is concerned with the relationship between the systematicity of empirical nature and aesthetic judgement: the true relationship between formal or logical purposiveness and taste is not that the former is itself the principle of the latter; it is rather that the principle licensing the former (the conditions of a reflective use of judgment) is identical to the principle underlying the latter.107

The principle of judgement is the ground both for the logical purposiveness of nature and for taste. Allison is concerned to avoid a double danger in Kant’s account. On some occasions it sounds as if an aesthetic principle of taste is dependent on empirical systematicity.108 On others it looks as if taste provides a principle for cognition.109 Allison’s alternative epistemic reading of Section 21 was intended to rebut the second horn of this dilemma, that is, of making cognition in general dependent on taste. Allison’s solution is to argue that in the deduction proper, Kant’s position is that taste is based on a principle 198


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement that expresses the subjective condition of judgement. This principle is the ground for taste, but it would appear the principle itself is not aesthetic. This is Allison’s strategy for showing how taste can be based on the subjective activity of judgement and thus displays the subjective conditions of cognition, while avoiding the conclusion that Kant suggests that taste is the foundation for cognition. My aesthetic reading starts from the fact that Kant at no stage commits himself to there being a principle of judgement distinct from that of taste. I agree that the principle of taste displays the subjective conditions of judgement or ‘cognition in general’. However, I read Kant as holding that taste rests on the subjective conditions of judgement, which, only in the aesthetic case are taken in isolation and count as a principle. There is no further non-aesthetic principle of judgement underlying taste. The principle of taste simply is the expression of the activity of judgement, which only operates as a principle in the aesthetic domain. Elsewhere those same subjective conditions operate as the subjective side of cognition, but never as a principle in their own right. My strategy of interpretation has a further advantage. I will be able to make sense of a connection that Allison dismisses out of hand, that is, Kant’s claims that aesthetic judgement qualifies as an exposition of the reflective principle of the purposiveness of empirical nature. Allison’s insistence that taste is based on a non-aesthetic principle entails that the former cannot make any direct contribution to Kant’s account of the order of empirical nature.110 The primary difficulty with my account is that it could be thought to imply that an aesthetic principle is necessary for cognition. If the ‘Deduction’ deduces an aesthetic principle of taste, then Kant has only established that the latter entails the subjective conditions of cognition. But he has not explained how aesthetic judgements entail these conditions in a distinctive way from cognitive judgements. We have seen that he has prepared for such an account by focusing on the power of judgement and by introducing, although not developing, the suggestion that aesthetic judgement allows for a reflection on that power. But as his argument stands, Kant has not yet established the full systematic relation between cognition and aesthetic judgement. My additional argument will be that aesthetic judgements are exemplary of the subjective activity of judgement necessary for cognition in general. Kant gives the clue to this solution of his problem only in a provisional way in the third Critique. My solution has an additional benefit in that, insofar as we establish a direct connection between aesthetic judgement and empirical 199


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology cognition, we can restore the materiality of the aesthetic object. Allison is inclined to insist on aesthetic judgement’s reliance on the subjective conditions of cognition to the exclusion of an adequate account of the aesthetic object. Allison insists that the judgement of taste is concerned with the ‘object qua represented’.111 This is surely right, for what comes to the fore in aesthetic apprehension is the presentation of the object. However, the negative part of Allison’s comment risks reinforcing a prevailing and rather simplistic view that if a judgement of taste is not objective, then it has nothing to do with the object. He says that the aesthetic judgement is not directed at ‘the inherent nature of such an object, not even considered as phenomenon’.112 Our liking for something beautiful is most definitely not directed at a thing in itself, nor indeed at qualities that are viewed as inherent, in the sense that they are taken apart from the relation to the subject. However to say that taste is not concerned with the object as phenomenon surely suggests that Allison agrees with those interpreters who insist that Kant brought about a subjective turn in his aesthetics. Moreover, the contrast Allison proposes places him in danger of obfuscating the relation in which representations stand to objects. An object is always a represented object and a representation is always intentionally directed to an object. This is true in Kant’s epistemology, and in his aesthetics. To be concerned with the form or representation of the object is not to be unconcerned with the object, but rather to turn our attention to certain primordial features of its objecthood. I have argued that in an aesthetic judgement it is the relation between subject and object that is potentially brought to our attention. We saw in Chapter 2 (pp. 64–8) that Allison failed to give a sufficiently direct and robust account of the relation in which represented objects stand to representations and that this leads to an underestimation of the role of affect in knowledge. Allison’s account of aesthetic judgement is similarly biased towards the subjective side of experience, although he concedes a role for objects, just as he did for matter in the epistemic case. The problem, as in the epistemic case, is that he does not sufficiently examine the relation between subject and object that is brought out in Kant’s examination of aesthetic judgement. In the final chapter, I will argue that taste allows for a reflection on the appearing of the appearance. Perhaps suspecting that, even at this late stage, he has not found the best way of establishing the universal validity of taste, Kant recasts the main argument in a footnote at the end of Section 38, saying that two conditions are necessary if we are to be justified in claiming universal 200


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement assent for an aesthetic judgement. 113 First, in aesthetic judgement the faculties of all judging subjects must stand in the ‘same relation’ as they do in cognition. This, he says, must be the case because otherwise people could communicate neither about their presentations (Vorstellungen) nor about cognition. Second, only this relation – and hence ‘the formal condition of the power of judgement’ – must determine aesthetic judgements. While mistakes could be made as to the formality of our judgement, this is a problem only of application.114 This restatement confirms that the relation of the faculties is the formal condition of all judgement, necessary for any cognition or judgement whatsoever. It still, however, does not resolve the problem of how aesthetic judgement rests on the relation necessary for ‘cognition in general’ without counting as a variety of the latter.

Conclusion In this chapter I have shown how the cooperation of the faculties, identified as synthesis in process in the previous chapter, counts as the subjective side of the deduction. In particular, I have argued that in Section 21 of the third Critique, Kant seeks to establish that the synthetic process displayed in aesthetic judgement is necessary for the legitimation of claims to objective knowledge. This attempt fails, however, due to an insufficiently finely-tuned argument and, not least, the lack of a distinction between two levels at which sensus communis operates. The official ‘Deduction of Taste’ is much more successful in establishing the necessary subjective conditions of cognition in general, but even it fails to show the nature of the relation in which the latter stand to aesthetic judgement. In Chapter 7 I will argue that aesthetic judgement is best understood as standing in an exemplary relation to cognition. I will also show how we can make sense of Kant’s apparently ill-advised claim that the ‘Analytic of Taste’ offers an exposition and a deduction of the principle of the purposiveness of nature. First, however, I turn to examine the objective side of the deduction in more detail.

Notes 1. See Makreel, IIK, for instance, p. 9 and also pp. 80–1; Ameriks, ‘Review’, p. 229. Kemp Smith suggests the same location for the subjective deduction in Commentary, p. 236. However, he also suggests that the subjective deduction has a wider relevance, especially in the B 201


29/3/07

8:12 AM



2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

16. 17. 18. 19. 20.

21. 22. 23. 24.

25. 26.

edition ‘Deduction’. See pp. 234–45. See also Sarah Gibbons, Kant’s Theory of Imagination, who argues for the general significance of the subjective deduction in revealing the subjective conditions of judgement and the relation in which this stands to aesthetic judgement. See, for instance, pp. 52, 79 and 83. Patricia Kitcher’s Kant’s Transcendental Psychology offers a sustained argument for the relevance of what she calls the ‘dark side’ of the Critique and points to the ambivalence of Kant’s retreat from the subjective side of the deduction (p. 13). CPR, A xvi. A xvi/xvii. A xvii. B 124–6. Preface to B Edition, B xvi/xvii. Compare B 519, discussed in Chapter 3, pp. 87–9. A 93, B 126. A 92/3, B125. See Chapters 4, pp. 132–47, and 6, pp. 213–17, respectively. A xvii. A xvii. A xvii. CJ, AA 215. Indeed Kant says that they have nothing to do with the object at all. This is, however, much too strong a statement of his position. On p. 177 I discuss Kant’s revision of his statement. CJ, AA 217. AA 217. AA 217–18. AA 218. See the related discussion in Chapter 4, pp. 154–5 where I noted that in the Introduction Kant avoided the suggestion that cognition is based on an aesthetic condition. As Pluhar’s translation brings out, the mutual relation of the faculties is necessary for all empirical cognition, but distinct from the harmony of the faculties characteristic of aesthetic judgements. Later I argue that this counts as a ‘contrapuntal exemplarity’. See Chapter 8, pp. 296–9. CJ, AA 179. See Chapter 1, p. 15. CJ, AA 215. AA 236 and 238. In Section 21, Kant intends to give a ‘proof’ for the validity of this principle. If this is right, it complicates the picture suggested by Allison when he says that the analysis of the four moments of taste contribute only to the question of quid facti. See pp. 193–4. AA 238 (Pluhar’s translation). AA 295. 202


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement 27. AA 238. 28. See pp. 189–93. 29. Allison and Guyer also divide the argument into seven steps but carve it up slightly differently from each other and from the reconstruction offered here. See Guyer, Kant and the Claims of Taste (= KCT) pp. 279–97; and Allison, Kant’s Theory of Taste (= KTT), pp. 150–1. 30. CJ, AA 238. 31. Guyer, KCT, pp. 288–94. 32. Ibid., pp. 288–9. 33. Kant’s real error is that his commitment to the community of judging subjects is too strong. He lacks any serious account of the interaction and dissonance among subjects. I therefore agree with Gardner’s comment that Kant is not committed to inter-subjectivity in a strong sense where empirical reality would be constituted by inter-subjective interaction. Gardner, Kant and the Critique of Pure Reason, p. 280. However, inter-subjectivity becomes less of an automatic presupposition and more of a task in the account of aesthetic judgement where we ‘require’ agreement from others. AA 214. 34. CJ, AA 238. 35. Guyer, KCT, pp. 284–5. 36. Ibid., pp. 296–7. His conclusion arises from his analysis of Kant’s account of ‘proportion’ to which we will return. 37. CJ, AA 191. 38. AA 238. 39. See Chapter 3, pp. 90–1, and also the discussion of the ‘Transition to the Deduction’ in the previous section of this chapter, pp. 171–2. 40. See previous chapter, pp. 122–7. 41. CPR, A 99. 42. CJ, AA 238. 43. See Allison, KTT, p. 150. 44. These two sub-stages should properly count as two distinct steps, but I retain the existing numeration for reasons of comparison. 45. See Chapter 8, pp. 284–90, for a discussion of the charge that Kant’s aesthetics results in the view that all objects are beautiful. 46. CJ, AA 238–9; The German for ‘mutual relation’ here is einer durch die andere. Pluhar uses the same English phrase at AA 191 to translate mit dem Verhältnis der Erkenntnisvermogen unter sich. This stage of my reconstruction correlates with Allison’s Step 5 and is the crucial one for his and any other reconstruction of Section 21, as he recognises. See discussion in next section, pp. 189–93. 47. Guyer, KCT, pp. 294–7. 48. Ibid., p. 297. 49. Ibid., p. 295. 50. CJ, AA 239. 203


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 51. This is in contrast to Allison who treats the two sentences, as stated here, as two steps. Allison, KTT, p. 51. 52. Or, perhaps, the form of a feeling. Any particular aesthetic pleasure shares the transcendental form of a play of the faculties. Its content would be specific to the particular experience, which may be expressed as a ‘liking’. See discussion below, pp. 196–7. 53. CJ, AA 238; cited above, p. 178. 54. See discussion of the ‘Deduction of Taste’, and in particular of Section 35, below, pp. 194–5. 55. CJ, AA 212′, where Kant says that determining judgement ‘requires no special principle by which to reflect’. 56. CJ, AA 239. 57. As examples of the usual reading, he refers principally to Guyer’s KCT and Anthony Savile’s Aesthetic Reconstructions, both of which are critical of Kant’s attempt to link aesthetics to cognition via sensus communis. Allison, KTT, pp. 145 and 153. However, on p. 145 he also refers to Ameriks, ‘How to Save Kant’s Deduction of Taste’, pp. 295–302. 58. In Allison’s reconstruction, this is Step 4. See Allison, KTT, pp. 151 and 152. 59. Ibid., p. 151. 60. Ibid., p. 152. 61. Ibid., p. 152. 62. Ibid., pp. 152–3. Allison is referring to CPR, A 21, B 35–6. 63. Allison, KTT, p. 153. 64. Ibid., p. 153; Allison’s emphasis. 65. CPR, A 21. 66. See discussion of his interpretation of the ‘Deduction of Taste’ in the next section, pp. 198–9. 67. Allison, KTT, p. 154. 68. Ibid., p. 154. Allison refers to CPR, A 133, B 172. 69. Allison, KTT, p. 154. 70. Ibid., p. 155. 71. Ibid., p. 155. 72. The most sustained account of a link between feeling and judgement is offered by Jean-François Lyotard in his Lessons on the Analytic of the Sublime. See, for instance, p. 9 (French edition p. 22): ‘In the analytic of taste, sensation no longer has any cognitive finality; it no longer gives any information about an object but only about the “subject” itself.’ Sensation translates la sensation, which in French could have the connotation either of sensory impression or of feeling. Thus ‘sensation’ here can stand for Gefühl in German. See also Makreel, IIK, pp. 103–7, for a discussion of ‘a feeling of the life of the mind’. See G. Zoeller, ‘Makkreel on Imagination and Interpretation in Kant’, 204


29/3/07

8:12 AM


Subjective Deduction in the Critique of Judgement

73.

74.

75. 76. 77.

78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88.

89. 90. 91. 92. 93. 94.

95.

pp. 266–75; see p. 272. See also Makkreel, ‘Response to Guenter Zoeller’, pp. 276–80, p. 279. See also an earlier discussion of the relation between judgement and feeling as the faculty for discrimination in Allison, KTT, p. 70. Allison’s references are to CJ, AA 177–8 and AA 207′–8′. See A. Bauemler, Das Irrationalitätsproblem. See particularly pp. 141–66 on the relation between imagination and wit, our capacity for connection. CJ, AA 295. This is linked to the general level of purposiveness discussed in Chapter 7, pp. 255–7. Sometimes, however, we glimpse the activity of reflection at the limits of our thought, especially whenever we are in a particularly reflective mood. This may count as a feeling of the measure or balance of thinking. Aesthetic judgement gives us a less uncanny, more reliable, although also more indirect access to the same phenomenon. See Chapter 8, pp. 297–8, for a discussion of a reflective feeling of the process of judgement. CPR, A 84, B 116. Allison, KTT, p. 67. Ibid., p. 144. See also p. 157. Guyer, KCT, pp. 161–4. Allison, KTT, p. 104. CJ, AA 286. AA 287. AA 287. He was previously guilty of this conflation in Sections 9 and 21. AA 287. AA 287. In Chapter 7, I will argue that aesthetic judgements display a general level of purposiveness between mind and nature that expresses the possibility of synthesis. I believe that Kant’s comments in Section 35 can be made sense of in terms of ‘general purposiveness’. AA 287. He also reintroduces the notion of free lawfulness from the General Note at this point. See AA 240–1. AA 288. Compare discussion of free lawfulness in Chapter 4, pp. 155–6. AA 289. I claimed this in Chapter 4, p. 154. AA 289. Kant uses the term Wohlgefallen in a broad sense in Section 5, AA 209 ff., to encompass three sorts of liking: appetitive, moral and aesthetic. He then narrows down his usage, reserving the same term as ‘free liking’ strictly for aesthetic judgement alone. AA 210. See AA 288 in Section 36. 205


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 96. AA 288. 97. AA 289. 98. Contrast Allison who denies that the form in question is spatiotemporal. Allison, KTT, p. 175. 99. My account of Section 38 so far relates to the first sentence of both Pluhar’s translation and of the German text. 100. CJ, AA 287. 101. AA 290. The reconstruction offered in this paragraph corresponds to the second sentence of Pluhar’s translation and stops at the colon in the second sentence of the original German version. 102. See Chapter 4, p. 155, on the two harmonies characteristic of aesthetic judgement. 103. This is an account of the third sentence of Section 38 and of the conclusion of the second sentence in German. 104. This is an account of the final sentence of Section 38, both in Pluhar’s translation and in the original. 105. Allison argues that the ‘Deduction’ concerns only a ‘pure’ judgement of taste and not actual judgements. Allison, KTT, pp. 177–9. 106. For Allison’s reconstruction of the ‘Deduction’, see ibid., pp. 175–6. 107. Ibid., p. 64. 108. This is the case in the Introductions to the third Critique. See Chapter 7, p. 248. 109. See discussion of Section 21 above pp. 186–8. 110. See Chapter 7, pp. 260–8. 111. Allison, KTT, p. 119. 112. Ibid., p. 119. 113. CJ, AA 290. 114. Kant shows himself remarkably sanguine on this point also at AA 216.

206


29/3/07

8:12 AM


6

A Priori Knowledge as the Anticipation of a Material Given and the Need for a Spatial Schematism

In this chapter I return to the objective side of Kant’s epistemological project. In Chapter 4 I discussed the ‘Transcendental Deduction’, which, it is often thought, provides the whole of the objective deduction. My aim is to show that the legitimation of the categories requires not only the whole of the ‘Analytic’, but also recognition of the aesthetic dimension of his epistemological argument. It is unavoidable that in our investigation of the relation between Kant’s accounts of cognitive and aesthetic judgement, we are forced to proceed in a zigzag. The reciprocal relation between cognition and aesthetic I am in the course of defending simply cannot be traced out in a linear fashion. As many commentators have recognised, Kant’s hopes for the conclusiveness of the ‘Transcendental Deduction’ turned out to be rather premature. In both the A Preface and in the ‘Transition to the Deduction’ Kant claimed that he would establish the objective validity of the categories insofar as they relate to objects of experience a priori and necessarily.1 Meanwhile, in the B edition ‘Deduction’, Kant claims that the categories apply to all perception, to the possibility of experience and therefore to all objects of experience.2 In later sections of the ‘Analytic’, he reveals that further articulation of the categories is necessary if they are to qualify as the form of experience for all empirical objects. My account of the structure of Kant’s extended legitimation of the categories owes much to Buchdahl’s reading in which every element of the ‘Analytic’ has a part to play.3 However, my reading brings out elements essential to the relation between Kant’s epistemology and aesthetics, not to be found in Buchdahl or in the accounts given by others who also resist the temptation to assume that the full deduction of the categories is supplied in the ‘Transcendental Deduction’.4 My aim in this chapter is to give a clear and systematic view of the 207


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology hierarchical stages through which Kant advances his argument in the later stages of the ‘Analytic’ in the light of my insistence on the affective dimension of Kant’s epistemology. I will discuss how the principles of the understanding are not merely an application of the categories, but rather a further and aesthetically charged articulation that is necessary for the legitimation of the latter. At the same time, I will establish that within the ‘Principles’ chapter there is a development from formal conditions toward a material given. Moreover, I will argue that the systematicity of the principles is a necessary condition of their legitimacy.5 Drawing on the account of ‘synthesis in process’ I established in Chapter 4, I will show how the arguments of the ‘Principles’ count as analyses of figurative synthesis and argue that Kant’s account reveals that the subjective side of the deduction is a necessary condition of objective determination. Most importantly, throughout this chapter a priori knowledge emerges as a project of determination, that is, as an anticipation of application to a material given. In particular, and, relatively late in the development of his thought, Kant comes to realise the centrality of space for establishing the distinction between formal and material idealism. I argue that it makes sense to speak here of a ‘spatial schematism’. In the first section, I assess the progress achieved so far in providing a deduction of the categories in the ‘Transcendental Deduction’, before turning to examine the extent to which the ‘Schematism’ chapter offers new input. It is more difficult to ascertain the role of the principles, which are often taken to offer merely an application of the categories. In the third section, I argue that the ‘Principles’ chapter offers a further aesthetic development of the pure concepts of understanding. They extend the transcendental account in order to establish that the categories apply to what Kant now refers to as actual or possible experience.6 It would thus appear that he believes this was not achieved either in the ‘Deduction’ or in the ‘Schematism’. In the fourth section, I argue that the development within the system of principles from those that are merely a priori to those that are regulative counts as a further articulation of the structure of the applicability of the categories to existing objects. The significance of this development turns out to be that objects are now considered as entailing a material given in space and time. However, even now we will find that Kant further hones his account of the categories in order to show that they qualify as the form of experience and so bear objective validity. The ‘Transcendental Deduction’ was thus far from being the end of the story. 208


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism In the fifth section, I examine the possibility that Kant may have offered the beginnings of a spatial schematism, complementary to the official temporal schematism in the ‘Principles’ chapter. This is a development of Kant’s thesis in the ‘Metaphysical Exposition’ of space in the ‘Transcendental Aesthetic’. In the final section, I examine Kant’s final attempt to establish the objective validity of the categories in the ‘Analytic’. He introduces a new articulation of causal necessity and three associated ‘propositions’. I suggest that this late addition is evidence for a continuing worry that he has not yet proven what he claimed much earlier in the ‘Transcendental Deduction’, in particular, because the given is only now considered as material. In the next chapter we will see that the same concern gives rise in the Critique of Judgement not only to a further principle, but to a distinctive reflective level of analysis of experience.

I The task of the ‘Transcendental Deduction’ In this section I build on the hierarchical reading established by Buchdahl, emphasising the issue arising from the ‘Deduction’ as to whether or not the transcendental form of knowledge is sufficient for establishing the possibility of empirical experience. In particular, I uncover a problem in Kant’s account as to whether the ‘Deduction’ already contains the conditions of possibility for singular empirical objects. The discussion here sets the scene for my claim that a priori knowledge is anticipatory of a material given. In the ‘Transition to the Deduction’ Kant states that the aim of the ‘Deduction’ is to show that not only the forms of intuition, but also the categories of the understanding count as a priori conditions of the possibility of experience.7 More precisely, he says that pure concepts are the conditions under which anything can be thought as an ‘object in general’.8 As it stands, this claim would establish only intellectual synthesis and set a minimal goal for the ‘Deduction’. But a few lines later, Kant claims: ‘through them [the categories] alone does experience become possible’.9 The ‘Deduction’ aims to prove not only that the categories are necessary for thinking objects, but also for experiencing them. Even this stronger claim does not entail that the categories generate experience and only that they count as its necessary conditions insofar as they provide its reflective form. Kant says that it is only insofar as categories supply the form of thinking experience that they make the latter possible.10 209


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology Kant has already made clear that the categories must be combined with an aesthetic component, namely the forms of intuition.11 So we know that although the categories must provide not only the form of thinking, but also the form of thinking the experience of objects, the form of actual experience cannot be established entirely at the level of thought. But the question is: can the possibility of actual experience be established at the transcendental level of analysis? And even if this is possible, is the apparatus of the ‘Deduction’ sufficient for doing so? Significantly, at the outset of the A ‘Deduction’ Kant says that if an a priori concept did not relate to experience, it would qualify only as a logical form and not as a concept ‘through which something is thought’.12 Is the experience in question empirical experience and, if so, in what sense? It is clear that, for Kant, experience is empirical and for this reason it may be thought that the question is meaningless. But the issue is whether or not Kant’s account of experience establishes its empirical status from the outset. While Kant was always committed to the empirical status of experience, his philosophical reconstruction of it began at a very abstract or general level. He only gradually worked towards providing a legitimation of experience qua empirical. The evidence for this is that it is only fairly late in his account that the givenness distinctive of the empirical comes to play a central role in his account of the applicability of the categories. Undoubtedly, empirical givenness is presupposed from the outset of his account, as I argued in Chapter 3, pp. 100–5, but in the initial accounts of the categories this aesthetic dimension of his epistemology is underplayed in favour of an emphasis on the purely formal elements of knowledge. For this reason it is necessary to stress that the goal of Kant’s account is to legitimate the application of the categories within empirical experience. Kant immediately goes on to say that although the categories, as a priori, have no empirical content, they must relate to experience.13 The contrast between form and content in this statement already strongly suggests that the experience in question is empirical. Kant concludes that the categories only have ‘objective reality’ if they are the conditions of a possible experience.14 So although Kant abstracts from everything empirical in the a priori analysis of the categories, the deduction will have to show how empirical objects, and not just a mere object in general, finally fall under the conditions of the categories. Or he will at least have to show that an object in general expresses the form of empirical objects. It is not clear at this stage that Kant recognises how much more articulation of his theory this may require. 210


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism It certainly appears on occasion that Kant believes that the ‘Transcendental Deduction’ supplies the full conditions of conceptual determination for all empirical objects.15 When he says that categories ‘prescribe laws a priori to appearances, and therefore to nature’, it sounds as if intuitions are not only universally but also wholly determined by the understanding.16 But Kant suggests this is not the case towards the end of the ‘Transcendental Deduction’, when he says: ‘Particular laws, because they concern empirically determined appearances, cannot be wholly derived [from the categories], although they nevertheless all stand under them.’17 In saying this, Kant concedes that there is a gap between transcendental structure and empirical experience, although he does not yet recognise that there may be a problem in bridging that gap.18 The categories do not complete the task of determination at the empirical level, for there must be empirical determination in addition to the a priori determination arising from the categories. This limitation on the ‘Deduction’ suggests two possibilities. First, it could be the case that the full formal structure of any singular empirical object is provided at the a priori level, leaving only the specific relations between objects to be determined at an empirical level. Second, it could be argued that the full determination of any singular empirical object must take account of the special laws that govern the relations in which it stands to other objects. This view would draw on the fact that an object of experience is distinguished by its standing in relation to other objects, as established by its falling under the category of relation and by the argument of the ‘Analogies of Experience’.19 While Kant does not even raise this question, he now announces that the full a priori specification of experience is not yet complete: How they [the categories] make experience possible, and what are the principles of the possibility of experience that they supply in their application to appearances, will be shown more fully in the following chapter on the transcendental employment of the faculty of judgment.20

The contents of both the ‘Schematism’ and of the ‘Principles’ are part of the account of how experience is possible: they thus still operate at the transcendental level. All that the ‘Transcendental Deduction’ has done is lay out the structure of experience at the pure a priori level. This is what Kant means when he concludes the B ‘Deduction’ by saying that he has established the principles of the possibility of experience insofar as they determine appearances in space and time in general.21 211


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology If the categories do not fully determine empirical experience, then it is at least in principle possible that we could have relatively indeterminate experiences. Intuitions alone do not qualify as experiences and anything of which we are aware stands in some relationship to the categories and thus bears a minimum determination. But not all intuitions are fully determined by the categories, which underdetermine empirical experience.22 This leaves open the possibility that there could be a gap between the transcendental order supplied by mind and the world of given objects. We can draw three conclusions from this. First, the mind does not simply impose order on the world and instead has to take further considerations into account in its empirical operation. Second, a gap is left for underdetermined intuitions, in particular those that are the starting point of aesthetic judgements. Third, the picture that emerges is one of knowledge as not simply a fait accompli and entailing a project. We have to go out into the world and examine particular empirical affects if we are to achieve empirical knowledge. We have a safety net insofar as we can rely on the a priori framework of the categories and of space and time; however, these formal conditions will not supply us with the full empirical determination of objects. Indeed, as I have already suggested, they do not even supply us with the full transcendental determination of empirical experience. Knowledge emerges out of an exploration of the world rather than as a projection of mental order onto a featureless world. I have shown how Kant concedes that the a priori categories cannot supply the full account of the determination of empirical knowledge, which also requires an empirical determination. I have also shown that he promises that the ‘Schematism’ and ‘Principles’ will extend the transcendental story insofar as they give a greater elaboration of how the categories make experience possible. A full account of the way in which the form of experience initiated by the mind is finally applicable within empirical experience will entail an examination of these later parts of the ‘Analytic’. If the aim of the ‘Transcendental Deduction’ is to supply the proof that the categories apply to all empirical intuitions, then it already looks as if, at the end of the B ‘Deduction’, Kant realises that there is still work to be done. In the remaining sections of this chapter, I will reconstruct the further stages of the transcendental determination of empirical objects through an analysis of the ‘Schematism’ and ‘Principles’, while in the next chapter I will examine the conclusion of his account of the determination of empirical experience in the two Introductions to the Critique of Judgement. 212


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism

II The third thing: a depth analysis of figurative synthesis In the reading of the ‘Schematism’ that follows, I bring out the relational status of the schema, allowing Kant to address how figurative synthesis, already adumbrated in the ‘Deduction’, is first possible. I suggest that the best way of grasping the possibility of figuration is as ‘synthesis in process’, first introduced in Chapter 4, and that this recognition helps explain the ambiguous status of the schema. In the ‘Analytic of Principles’, which comprises both the ‘Schematism’ and the ‘System of all Principles of Pure Understanding’, Kant begins by saying that he will now provide a ‘canon’ for judgement, that is, instructions for the application of the categories to appearances.23 It sounds as if the schemata and principles will merely serve as an instrumental implementation of a proof that is already complete. However, examination of the significance of a canon for Kant’s method reveals a more complex story. Kant uses the term ‘canon’ on various occasions to express a limitation on the critical project, and in particular its restriction to the empirical domain.24 It is only, however, late in the Critique that he supplies a definition: ‘I understand by a canon the sum-total of the a priori principles of the correct employment of certain faculties of knowledge.’25 If the ‘Analytic of Principles’, supplies a canon, then this could mean it is simply a working through of the rules already established in the ‘Transcendental Deduction’. But this does not seem to be Kant’s intent, for he goes on to say that the ‘Transcendental Analytic’ qualifies as the canon of the pure understanding.26 This suggests that the ‘Transcendental Deduction’ alone does not supply the principles of experience, although the ‘Analytic’ as a whole does do so. And if the categories of the understanding are not established as the principles of experience in the ‘Deduction’, then it would appear that there is still work for the ‘Analytic of Principles’ in showing that the categories apply to all possible experience. If Kant were to have claimed that the categories simply are the principles of experience, then he would be in danger of falling into an impositionalist position, where the form of thought coincides with the form of objects. But this is not Kant’s position. At the beginning of the ‘Analytic of Principles’, he says that concepts would be mere logical forms and not pure concepts of the understanding if they did not ‘formulate by means of universal but sufficient marks [Kennzeichen] the conditions under which objects can be given in harmony with these concepts’.27 The distinction between logical 213


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology forms and categories strongly echoes the initial paragraphs of the A ‘Deduction’ and it would appear that, in Kant’s view, his initial goal is not yet achieved.28 In the account of figurative synthesis in the B ‘Deduction’, Kant argued that the categories apply to objects only insofar as the imagination mediates between concept and intuition. Now he announces that a further stage is necessary. What are these marks that will establish the conditions of applicability of the categories? Kant answers this question by immediately promising that in his next chapter – that is, in the ‘Schematism’ – he will discuss the sensible condition under which the categories can be employed.29 We must conclude that the marks that are necessary to distinguish a mere logical form from a principle of the understanding are the schemata. At the outset of the ‘Schematism’ chapter, Kant introduces the rather odd idea that the representation of the object must be homogeneous with its concept. He then presents as equivalent the reverse, namely that ‘the concept must contain something which is represented in the object that is to be subsumed under it’.30 While the first formulation makes it sound as if the object must conform to the concept, the second suggests that the concept must conform to the object. As it stands this could lead us to conclude that Kant does not see that the two claims are contraries; neither does he appear to be aware of the tension in which the second formulation stands with the Copernican revolution. However, I believe that the combination of these two positions arises from his attempt to express the relation in which the concept stands to the object or underdetermined intuition. The concept is nothing other than the condition for the possibility of the unification of something, that is, the given in intuition. On the other hand, the intuition can only be known insofar as it is capable of being unified under a concept. The process of determination entails at least three conditions, as we saw in Chapter 3, p. 94. Something has to be given in intuition, our sensibility has to be capable of taking up that given as an affect, and our understanding must be capable of unifying the affect. If any of these three conditions failed, knowledge would not arise. But, surely, the need to unify an intuition under a concept was already satisfactorily dealt with in the ‘Deduction’; at least in the second part of the B edition, where Kant developed his account of figurative synthesis? Imagination was the third thing that made possible the necessary unification of intuition and concept and showed how the categories were not only a form of thought, but also the form of experience. So is something genuinely new added in the ‘Schematism’ chapter? 214


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism Whereas in the ‘Deduction’ Kant established the necessary role of the transcendental synthesis of imagination, he now makes the additional claim that the imagination must produce a third thing, or transcendental schema, that makes possible the application of concepts to intuitions or objects.31 This addition will allow for the resolution of the problem, which he clearly thinks is not yet answered, namely ‘How, then, is the subsumption of intuitions under pure concepts, the application, of a category to appearances, possible?’32 Kant remarks that this question makes a transcendental doctrine of judgement necessary, that is, the ‘Deduction’ must be supplemented by the ‘Schematism’ if the applicability of the categories is to be established. But if it is not yet established that the categories apply to all objects, then the ‘Deduction’ did not yet validate pure concepts as supplying the form of experience even at an a priori level. Although the ‘Deduction’ revealed that figurative synthesis was necessary, the ‘Schematism’ is required if we are to understand how it is possible.33 The schema counts as a mediating representation insofar as it is both intellectual and sensible.34 This dual status allows it to establish the relation between pure concept and appearance, that is, the application of a category to an empirical intuition. The schema must also be pure, that is, it must have no empirical content. The third thing qualifies as a ‘universal but sufficient mark’ insofar as it is capable of applying the form of thought to all intuitions.35 As all of this must occur at the a priori level, the schema makes possible the synthesis of the pure form of the understanding with the forms of intuition. Nevertheless, Kant’s aim is to show how a category is capable of synthesising an empirical intuition or appearance, through an operation on the pure form of intuition. He consistently states that the task of the ‘Schematism’ is to show how an appearance can be rendered homogeneous with a pure concept of the understanding.36 The mediating representation in question is temporal. Time is the form of all intuitions. Yet the determination of time depends on the unification afforded by the category. The schema can now be identified as the transcendental determination of time that allows the application of the categories to appearances.37 This is the mark that makes possible the application of categories. We will presently find that the determination of time, in this context, is not a determinate temporal intuition, but rather the temporal elaboration of a category so that it can take up an intuition, which is necessarily temporally formed. In the ‘Schematism’ the process of time has been introduced into the categories; a temporal conclusion has not been achieved. 215


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology Towards the end of the ‘Schematism’, Kant says that ‘[t]he schemata are the true and sole conditions under which these concepts obtain relation to objects and so possess significance [Bedeutung]’.38 This confirms the view that the significance of the categories, arising as it does from their applicability within experience, is not secured in the ‘Transcendental Deduction’ alone. Kant sometimes characterises the schema as if it were a mental object; for instance, when he speaks not only of a third thing, but also of a monogram or a transcendental product of the imagination.39 At other times the schema counts as a mental process, namely, the transcendental determination of time, ‘a universal procedure of imagination in providing an image for a concept’.40 This tension is resolved in his characterisation of it as ‘a rule of synthesis of the imagination’.41 The schema is the temporal elaboration of a category, so that the latter can be applied within the field of appearances. The schema allows the concept to synthesise a given object. So, although the claim that the schema of a triangle ‘can exist nowhere but in thought’, and that it is a rule for pure figures in space, may make it sound as though the ‘Schematism’ is not yet concerned with empirical appearances, this is not the case.42 The schemata are pure operations of the imagination that introduce figure or design into empirical appearances, thus rendering them knowable: they are not pure images, as Heidegger suggests.43 Does this genuinely add something new to the account of figurative synthesis in the ‘Transcendental Deduction’? On the occasions when Kant characterises the schema as a third thing, it appears that a new element has been introduced. However, on further analysis the schema is nothing other than the operation of figurative synthesis. Kant now emphasises the process through which the categories are applicable within experience. The process no longer appears to be an immediate or automatic result of the combination of understanding and intuition through the mediation of imagination. The imagination must generate a mark, that is, the concept must be temporally articulated. It is not so much that in the ‘Schematism’ Kant introduces a new element, as that he carries through his analysis of figurative synthesis at greater depth, revealing that it is a temporal process through which mental operations are capable of taking up an object, albeit at the pure a priori level.44 This is synthesis in process, which, I argued in Chapter 4 (pp. 151–6) is the condition of any cognitive judgement and yet is only held up for our inspection in aesthetic judgement. 216


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism The schema, which is characterised as a transcendental product of the imagination, is also described as the schematism of the understanding.45 The problematic status of figurative synthesis that we already discovered in the official ‘Deduction’, in both its versions, is still evident. We can now see, however, that schematism is of the understanding strictly in the sense that a schema operates on the understanding in order to achieve the temporal determination of the categories. This clarification does not undermine Kant’s remark that the schematism also belongs to the imagination, for it is the latter that makes possible the temporal elaboration of the pure concepts of understanding. The complex cooperation out of which schematism arises may have been one of the motivations for Kant’s famous remark that it is: ‘an art concealed in the depths of the human soul, whose real modes of activity nature is hardly likely ever to allow us to discover, and to have open to our gaze’.46 The ‘Schematism’ chapter further emphasises the way in which the faculties must cooperate if the subsumption of an object under a category is to be possible, thus deepening our understanding of how the subjective side of the deduction contributes to the objective deduction. This is easily lost sight of, however, because Kant is primarily concerned with establishing the possibility of the determination of an object. I have emphasised that the ‘Schematism’ chapter is already concerned with the possibility of the application of the categories to empirical appearances. Kant’s aim is to establish how empirical intuitions fall under the categories. However, as I will argue in what follows, the full story of synthesis is not yet complete, neither at the a priori nor at the empirical levels. The schematism operates as a figurative synthesis at the pure a priori level between a pure category of the understanding and the pure form of intuition, time. Kant’s aim is to show how a schema facilitates the application of a category to an appearance, but he has not yet got so far. In the ‘Principles’ chapter, a potential spatial schematism emerges as a complement to the official temporal schematism, while in the Introductions to the Critique of Judgement, Kant introduces a further principle of reflective judgment, which I will argue is necessary for the full articulation of synthesis at the empirical level.

III The principles and possible empirical knowledge in general It might seem, nevertheless, that the applicability of the categories to intuitions is now established, especially if we return to the passage 217


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology immediately preceding the ‘Schematism’ where Kant announced the need for a universal, but sufficient mark.47 In this passage he claimed that, in the transcendental doctrine of judgement, he would provide: first, the sensible condition under which the categories can be employed; and, second, he would lay out ‘the synthetic judgements which under these conditions follow a priori from pure concepts of the understanding’.48 The ‘Schematism’ achieved the first of these tasks, while it might appear that the ‘Principles’ chapter need only spell out the temporally determined categories. Nevertheless, at the outset of the ‘System of all principles of pure understanding’, Kant announces that the principles require a proof.49 Why should this be so if the principles are simply an application of the temporalised categories? Although the table of categories is ‘the natural and safe guide’, it would appear that the argument is not yet complete.50 We will find that the new development entails laying out the systematic connection of a priori judgements, a step that is now revealed as necessary for establishing that the categories apply universally within experience. Importantly, for my continuing interest in showing that the objective deduction has a subjective side, Kant remarks that the required proof can only be carried out by examining the ‘subjective sources of the possibility of knowledge of an object in general’.51 This should not, by now, be surprising, for Kant is engaged on an a priori analysis of the form of experience. This form requires a mental input and can only be understood as arising from a subjective process. However, Kant could have made his position more clear; for although the form of experience arises from subjective activity or synthesis in process, the latter is always intentionally oriented towards something given in experience. Kant establishes this at the outset of the ‘Transcendental Aesthetic’, as I discussed in Chapter 3. In consequence, each of the stages in the B ‘Deduction’ and the ‘Schematism’ are marked by a repeated insistence on how a (subjective) category must be shown to be capable of synthesising an object.52 Kant’s a priori analysis is of a process initiated by the mind, but always exercised on an extra-mental given. We will see that the ‘Principles’ take this process of establishing a fit between mental form and the given a step further. In ‘The Highest Principle of all Synthetic Judgements’, Kant continues to fine-tune the problem of the application of the categories. The problem now is: how is an a priori synthetic judgement possible? An analytic judgement requires only the analysis of a concept in order 218


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism to draw out an implication that it already contains. A synthetic judgement requires us to go beyond what is merely thought in the concept and to link the latter with an intuition.53 He says this requires a third thing, which is the medium for all synthetic judgements. Just as in the ‘Schematism’, the third thing is primarily understood as time, but he goes on to suggest that the synthesis of representations requires the operation of all three faculties of intuition, imagination and understanding.54 This explains his earlier claim that the proof of the principles requires a turn to the subjective sources of knowledge. But how is an examination of the faculties to help in establishing the application of a pure concept to a pure intuition and show how an a priori synthetic judgement is possible? Kant now returns to the objective side of his argument: If knowledge is to have objective reality, that is, to relate to an object, and is to acquire meaning and significance [Bedeutung und Sinn] in respect to it, the object must be capable of being in some manner given. Otherwise the concepts are empty; through them we have indeed thought, but in this thinking we have really known nothing; we have merely played with representations.55

The thought or representation must take up something given to it in empirical experience, as I discussed in Chapter 3. Even space and time are ‘without objective validity, senseless and meaningless [ohne Sinn und Bedeutung]’ if they are not applied to objects of experience.56 This is because they, as the forms of intuition, only make receptivity possible insofar as something is given within them. Kant’s ongoing project of establishing that the forms of thought are capable of taking up the empirical given, now leads him to develop his account of the aesthetic or sensory side of experience beyond its formal conditions in order to address what is given in intuition. The ‘Principles’ thus provide further confirmation of the complex identity of intuition already established in the B ‘Deduction’.57 If a concept is to have objective reality and significance this entails not merely the temporal determination of the category, but the further requirement that the ‘representation through which the object is thought relates to actual or possible experience’.58 While Kant has had the latter as his goal since the beginning of the ‘Deduction’, it would appear that he has not yet arrived at his destination. It is not, of course, that Kant has abandoned his commitment to the temporality of experience, but rather that the full determination of empirical experience requires more than temporal schematism. 219


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology Experience, for Kant, as we already know, is inextricably linked with knowledge. He now says that experience is only possible as a synthetic unity of appearances.59 Knowledge does not arise if the objects that affect us are not related to one another in a systematic fashion. In order for this to be possible, there must be principles that establish the form of experience, that is, its most general order.60 Kant now intends to provide an account of the principles that order experience at its most general level. The principles are elaborations of the categories, and thus he will finally be in a position to show that the categories are not merely forms of thought, but also provide the form of experience. At the conclusion of the principles, he claims: ‘We then assert that the conditions of the possibility of experience in general are likewise conditions of the possibility of the objects of experience, and that for this reason they have objective validity in a synthetic a priori judgment.’61 Thus, although at the end of the B ‘Deduction’ Kant purported to have established that the categories are the principles of the possibility of experience, it now transpires that this result was only achieved at the most general level. 62 While, supposedly, he was already concerned with the objects of experience, we now discover that he was operating only at a provisional level. Kant now intends to go further in his analysis of the possibility of experience insofar as he takes into consideration not only the form of experience, but also the given to which the latter necessarily stands in relation. We can now return to consider Kant’s claim that the latest stage of his argument cannot be carried out in an objective fashion. He went on immediately to explain that this is because the principles are the foundation for knowledge.63 He must mean that because knowledge arises out of a subjective activity, it rests on the operation of our subjective powers. But clearly, this does not entail that those subjective faculties do not relate to objects. Indeed, both in the ‘Schematism’ and in the ‘Principles’, we have seen that Kant’s aim is to show how concepts can relate to objects, and, latterly, not only to the formal conditions of sensibility, but also to what is given through them. The subjective conditions are the conditions of our access to objects given in experience. The explanation of this process must start from the subject and not from the object, because it is knowledge of objects that is in question. This, and not impositionalism, is the deeper significance of the Copernican turn in philosophy. The goal is to show how the subjective conditions reach their goal, that is, take up the object given in experience. 220


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism Towards the end of the introductory account of the principles in Section 2, Kant clarifies that the systematic unity must be of empirical experience. Only empirical synthesis will give reality to a priori synthesis.64 The categories must be shown to apply to empirical objects. Kant thus acknowledges that although empirical experience has always been the goal of his epistemology, he has not yet provided the full analysis of its possibility. The truth of a priori synthesis arises only insofar as it provides the form of empirical experience.65 Kant concludes the section saying that synthetic a priori judgements are only possible insofar as the subjective formal conditions of experience relate ‘to a possible empirical knowledge in general’.66 We must conclude that the ‘Principles’ chapter will show that the objective reality of the categories is finally achieved by establishing how they apply at the empirical level. But the last stage of the argument of the ‘Analytic’, or the canon of understanding, still operates at the level of the possibility of the objects of experience. Although the ‘Principles’ chapter is more empirically oriented than earlier sections of Kant’s account, it still falls short of the specific empirical determinations mentioned in the ‘Transcendental Deduction’. The new element, in what has emerged as a hierarchical determination of empirical experience, is the focus on the givenness of intuition.67 We saw in the ‘Deduction’ proper that Kant assumed that the categories would supply the general form of experience. Nothing at the level of experience, it would appear, interferes with the framework arising from the forms of thought, which are merely supplemented by special empirical laws. However, in the ‘Schematism’ and ‘Principles’, Kant concedes that if the categories are to legislate within empirical experience, the pure concepts must be further articulated, first temporally and second as a system of principles oriented to the possibility of empirical experience as given.

IV The mathematical and dynamic principles as further articulations of the a priori in anticipation of experience If schematism transforms the categories into temporalised concepts, then the principles establish the latter as rules for the synthesis of objects given in experience. These rules comprise a system that finally provides the transcendental system of experience. Only under these rules is judgement capable of determining appearances and this is possible only insofar as the principles are exercised as a system, that is, they jointly make possible the determination of objects of experience 221


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology at the transcendental level. I will show that a progression toward a material given can be uncovered not just in, but within the ‘Principles’ chapter. Kant distinguishes four principles, or sets of principles, ultimately derived from the ‘Table of Categories’. The first two sets count as mathematical insofar as they are the unconditionally necessary a priori conditions of any possible experience. They are concerned with the ‘mere intuition of an appearance in general’.68 The ‘Axioms’ and ‘Anticipations’ are concerned only with the form of experience. However we will see that, in comparison with the ‘Schematism’, they are more closely associated with the given in intuition because they stand in systematic relation to the ‘Analogies’ and ‘Postulates’. The ‘Axioms of Intuition’ establish that all intuitions have extensive magnitude, that is, they are necessarily extended in space and time. This extension becomes knowable when it is subject to a rule of unification arising from understanding, allowing their appearance in space to be constructed through an exercise of figurative synthesis.69 Kant is here concerned with how synthetic activity generates the form of appearance, that is, of something given in intuition. The ‘Anticipations of Perception’ establish that ‘in all appearances, the real that is an object of sensation has intensive magnitude, that is, a degree’.70 Appearances are characterised not only by extension in space, but also by an affect in sensation. This principle also counts as mathematical insofar as it is necessary for intuitions in general. Appearances, Kant goes on to explain, contain not only the form of intuition, but also ‘the matter for some object in general’ or the ‘real of sensation’.71 Our representation of this is merely subjective, yet it makes us conscious that we are affected by an object in general.72 Kant here reiterates the position he first laid out at the beginning of the ‘Transcendental Aesthetic’, namely, that experience arises from the affect of an object given in experience. This is the force of the distinction between his position and material idealism, for which there would be no material given. Any appearance arises from an affect, which puts us in relation to a given material object. But the given object counts only as an object in general because an affect is precisely what is not yet determined. The affect is merely a sensation, that is, the material in sensation.73 What that sensation will be cannot be anticipated at the a priori level. We can only await its advent, for the matter in experience must be given in experience.74 There is, nevertheless, something that can be anticipated a priori, namely, that all perceptions will have some degree of reality – that is, there will be a 222


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism degree of sensation or affect.75 This anticipation is the formal condition of our empirical apprehension of the qualitative or material aspects of objects. Kant refers not only to the material given in general, as we have already seen, but also to those aspects of an object that he considers empirical and have traditionally been referred to as secondary qualities.76 Our anticipation supplies a ground for the productive imagination’s syntheses of intensive magnitudes, which count as flowing insofar as they are temporal.77 Figurative or productive synthesis begins from a rule, or principle, sanctioning its anticipation of a material reality given in experience. Only the possibility of these syntheses is known – more accurately, objectively cognised – a priori. The syntheses of the material given can only occur in experience. Kant concludes the ‘Anticipations’ with: ‘Everything else has to be left to experience.’78 In passing, Kant comments on the anticipatory character of knowledge in general: ‘All knowledge [Erkenntnis] by means of which I am enabled to know and determine a priori what belongs to empirical knowledge [empirische Erkenntnis] may be entitled an anticipation . . .’79 We can now reinforce the significance of this statement for Kant’s formal idealism. A priori knowledge as comprehended by material idealism would not count as an anticipation simply because it would entail the a priori construction not only of the principle of knowledge, but also of its content. The principles Kant is in the course of developing give merely the form of experience and only have objective reality and significance insofar as they take up something given in experience. A priori knowledge anticipates the experience of which it is a form. This is the added emphasis that Kant introduces in the ‘Principles’ chapter and it supports my earlier claim that the formal structure introduced by mind initiates rather than imposes order.80 To say that the principles anticipate experience is to say that their validity requires their applicability, that is, the possibility of application, within empirical experience. This is not, however, to say that empirical experience can validate its own formal principles. This would make no sense from a Kantian position. Application of the principles does not establish their legitimation, which requires, strictly, the establishment of their applicability at an empirical level. Despite signalling the anticipatory status of a priori knowledge (or cognition) as it is understood from the perspective of formal idealism, Kant distinguishes the peculiarly anticipatory character of the relation in which we stand to matter given in sensation. Nothing can 223


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology be determined beyond the possibility that there is some material affect, bearing some degree of intensity or other. This characterisation marks a further level of discrimination within the mathematical principles, which, as we have seen, count as apodeictic insofar as they concern only the necessary intuitive form of experience.81 The ‘Axioms of Intuition’ establish that we can know a priori that a spatial intuition can be constructed as an extensive magnitude. The a priori analysis of spatial construction gives the formal conditions for empirical syntheses. However no overt role for empirical experience is developed in the account of the axioms. In contrast, a priori syntheses of intensive magnitudes cannot operate without anticipating the empirical. The second mathematical principle is not just a principle or form for experience, it is a form of experience, in the sense that it belongs to the latter and cannot even be expressed at the formal level without reference to experience. The second two sets of principles count as dynamic and are equally a priori and necessary, but, only, Kant adds ‘under the condition of empirical thought in some experience, therefore only mediately and indirectly’.82 Their necessity is thus strictly in anticipation of their application to empirical experience. Kant finally approaches the goal of empirical application that has structured the composition of the ‘Analytic’ as a whole. The dynamic principles comprise the ‘Analogies of Experience’ and the ‘Postulates of Empirical Thought in General’. I will not attempt to enter into the detail of Kant’s elaborations of these, focusing only on the way in which they reveal the further determination of the a priori in the interests of the establishing the latter’s applicability at the empirical level. I will thus comment only selectively on the principle of the analogies as a whole, on the second analogy and on the general structure of Kant’s account of the postulates. The terms in which the general principle of the analogies is presented in the B edition reveals a progress beyond the intuitions legislated for by the axioms and the appearances that were the subject of the anticipations. Kant now speaks of perceptions.83 The dynamic principles are distinguished from the mathematical principles in that they deal with the existence, and not just the possible mental construction in intuition, of appearances84. To consider an object as existing is to grasp its givenness, in addition to its aesthetic and reflective formal properties. And while the anticipations established merely the form of a material given, that is, the anticipation that there would be some material given, the analogies are characterised as dynamic principles insofar as they are concerned with the possibility of actual given objects. 224


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism If we are to consider an appearance as an object – that is, as existing – we must necessarily conceive of it as standing in relations with other objects. ‘Experience is possible only through the representation of a necessary connection of perceptions’.85 A mere mental representation could stand alone for our imagination; for instance, the dagger that Macbeth saw before him. In contrast, a real dagger would necessarily stand in certain temporal and spatial relations to other perceptions, and our perceptions of one object would be affected by changes in other objects. For the moment, Kant speaks only of temporal relations but, later, he will retrospectively declare that the relations in which objects stand to one another are also spatial.86 The principle of the analogies thus can only be expressed in relation to, or in anticipation of, possible experience where such relations will be found. The three sub-principles of the analogies – the permanence of substance, succession in time and reciprocal coexistence – only count as regulative principles.87 Whereas the extensive form of an appearance, and the possibility (though not, as we have seen, the actuality) of its intensive quality, can be constructed, the existence of an object cannot. The analogies of perception do not allow us to construct the existence of a further object standing in relation to one that we presently perceive.88 However, we possess a rule that allows us to search out that missing element in the interests of achieving unity in experience.89 It would appear, then, that the unity of experience as a systematic set of relations between appearances is something that we have an a priori right to search out, but not to assert in advance of experience.90 Knowledge – at least, in its guise as a priori cognition – is now firmly established as a task and not merely as a fait accompli. In this respect, the ‘Analogies’ develop the anticipatory story of Kant’s account of intensive magnitude. Kant now makes a claim that sounds odd. He says that, the principles in general and the analogies in particular ‘have significance and validity only as principles of the empirical, not of the transcendental, employment of understanding’.91 By now, we must take his point that the principles only have significance insofar as they are applicable to empirical experience. But can he really intend that they do not count as transcendental? What Kant must mean here is that the principles are a priori in status and thus count as part of the transcendental form of experience, but they must be applicable to empirical experience and would have no validity were they considered only as mere forms of thought, that is, of the understanding as a merely subjective 225


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology process. The categories alone, as examined in the ‘Transcendental Deduction’, are not yet the principles of empirical experience.92 Kant adds that only a schematised category qualifies as a principle.93 While this claim leaves open whether or not the principles count as a further development of the categories, it is at least clear that the ‘Schematism’ chapter is necessary for the latter to qualify as the form of experience. The analysis of the principles I have given in this section suggests strongly that they, too, are needed if the deduction of the categories as the form of empirical experience is to be completed. The principles establish the application of the categories, not only to the form of intuitions but also to existing objects. We must be careful, however, to remember that even the existence or givenness of objects is, as yet, considered only at a general level. The second analogy is central not only for the establishment of a necessary connection between perceptions – that is, for the analogies as a whole – but also for the system of principles. In this discussion, I will merely highlight the way in which Kant’s account of the law of causality pushes forward his account of the application of the principles within experience. As is well known, in the second ‘Analogy’ Kant seeks to provide the rule for distinguishing objective from merely subjective succession in time. The sequence of cause and effect, where the former determines the latter, cannot simply occur in intuition, but is an operation of figurative synthesis.94 Objective succession must capture something in the object and is, thus, not merely a subjective representation. This requires a concept of the understanding.95 Kant is well aware that the term ‘object’ is problematic and that its referent is a moveable feast: Everything, every representation even, insofar as we are conscious of it, may be entitled object [Objekt]. But it is a question for deeper enquiry what the word ‘object’ ought to signify in respect of appearances when these are viewed not insofar as they are (as representations) objects, but only insofar as they stand for an object.96

Representations must ‘stand for’ objects if they are to qualify as expressing objective succession in time. We cannot, in experience at least, distinguish appearances, that is, appearing objects from the representations or apprehensions we have of them.97 But despite the fact that appearances always stand in some relation to our cognitive capacity, Kant nevertheless insists that it is necessary to establish ‘what sort of a connection in time belongs to the manifold in the appearances themselves [an der Erscheinungen selbst]’.98 This curious expression is 226


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism his way of expressing that while appearances are necessarily for us, they are nevertheless extra-mental insofar as they arise from a material given that is taken up by our sensibility and understanding. Kant’s terminological struggle to express the way in which representations give access to objects is reminiscent of Merleau-Ponty’s wonderfully evocative and seemingly contradictory expression: ‘in-itself-for-us’.99 An objective representation has to fulfil the requirement of something having been given, which is the first condition of knowledge, as I argued in Chapter 3. The rule of causal succession makes possible that we can distinguish between mere representations in the imagination and appearances themselves, which are accessed through the synthesis of imagination.100 Kant is referring to the complex activity of the imagination through which something given in intuition is taken up and made available for unification by the understanding. Access to substances, which was the topic of the first ‘Analogy’, is only possible insofar as the faculties are used in cooperation with one another in response to a material given. The law of causality finally supplies the criterion for distinguishing a mere concept of the understanding from a principle. A principle achieves a figurative synthesis through which our subjective faculties take up something given in experience. The subjective conditions are necessary but insufficient conditions of achieving objective knowledge, as I argued in the previous chapter. It is, however, tempting to conclude that objective succession arises from nothing other than a subjective rule: ‘If we enquire what new character relation to an object confers upon our representations, what dignity they thereby acquire, we find that it results only in subjecting representations to a rule . . .’101 Kant’s point is that we can never escape from our representations in order to establish some ‘mysterious kind of objective reality’.102 Grasping objective reality is indeed the goal of our cognitive engagement with the world, but all we have is a rule that allows us to make a distinction between subjective and objective succession. We cannot simply abandon ourselves to the circle of representations, nor can we venture out beyond those representations. Our position is irrevocably that of seeing objects in our representations and only thus of being capable of knowing rather than imagining things in the world.103 All we have as security in our cognitive project is a law of causal connection on which rest our figurative syntheses of the empirical given. Only the form of alteration is known a priori.104 Everything else – for example, ‘how anything can be altered, and how it should be 227


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology possible that upon one state in a given moment an opposite state may follow in the next moment’ – has to wait for experience.105 Highlighting the anticipatory status of his epistemological stance, Kant concludes the second ‘Analogy’ by saying a priori knowledge is simply the anticipation of our own apprehension, the formal conditions of which alone count as a priori.106 I understand Kant’s use of ‘anticipation’ here as suggesting that the validity of the formal conditions of experience – and even their meaning and sense – requires that they are implemented within the material given. Formal idealism anticipates material content, but it does not pre-empt the latter. Consequently, a priori synthetic syntheses do not impose form on matter. The status of the postulates of empirical thought in general is not only regulative but also subjective. Although Kant says that they ‘concern possible experience and its synthetic unity’, they achieve only a subjective synthesis in contrast to the objective synthesis of perception achieved by the analogies.107 This is due to the fact that ‘they add to the concept of a thing (of something real), of which otherwise they say nothing, the cognitive faculty from which it springs and in which it has its seat’.108 We can make sense of the status of the postulates through the distinction between the subjective and objective sides of the deduction. The last group of principles refers us back to the cognitive faculties that are the first condition of experiencing objects. As we have seen, the principles in general are only capable of a subjective proof insofar as they are the foundation for knowledge of objects.109 Knowledge is a possible relation in which a subject stands to a world and the proof of its possibility necessarily entails going back to its subjective sources.110 The principles, as a whole, elaborate the way in which a subject takes up something given in experience. The postulates in particular establish the ways in which an object can be epistemically significant for a subject, namely as possible, actual and necessary. The last set of principles thus particularly focus on the relation in which the object stands to the subject Given the internal dependence between the subjective and objective sides of the deduction I have argued for, it should not by now be surprising, though it is striking, that in the heart of the ‘Postulates’, which supposedly achieve only a subjective synthesis, Kant provides his most trenchant statement of formal idealism and its commitment to empirical objects. Something is actual if it is ‘bound up with the material conditions of experience’.111 The formal and universal conditions of experience give rise to what is possible and what is necessary, which 228


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism are the subject of the first and third postulates respectively. But if something is actual then it arises from a material condition given in experience. This is the subject matter of the second ‘Postulate’ and of the ‘Refutation of Idealism’. For my present purposes, I simply want to emphasise that Kant’s critique of Descartes’ material idealism is specifically of his predecessor’s claim that certainty can arise from a mere concept. Kant insists that even self-knowledge requires an aesthetic element, that is, a temporal intuition and a spatial consciousness of outer objects. Knowledge entails a cooperation of a plurality of subjective orientations in response to a material given. In this section, I have shown how the ‘Principles’ chapter extends the project of establishing the objective validity of the categories, by showing that their temporally schematised form must anticipate the given in experience. This entails recasting the formal conditions of intuition as anticipations of the empirical material given, that is, empirical objects, as I first argued in Chapter 3 (p. 107). This finally establishes the relational and anticipatory character of Kant’s formal idealism. In the next section, I will consider if a full account of the transcendental structure of the relation between form and matter requires Kant’s establishing the role of space in the determination of empirical objects.

V Spatial Schematism? At the outset of the ‘Principles’, and in both editions, Kant insists on the necessity of space for the objective validity of pure synthetic judgements. Were it not for space, our a priori knowledge would count as ‘nothing but a playing with a mere figment of the brain’.112 Immediately prior to this claim he argues that it is actual or possible experience that gives ‘objective reality to all our a priori modes of knowledge’.113 We must, then, conclude that the possibility of experience is established only when the categories are subjected to a figurative synthesis that takes not only time but also space into account. Kant goes on to claim that while experience depends on ‘a priori principles of its form’, the objective reality of those conditions can only be shown in experience.114 This reinforces the central thesis of the ‘Principles’ chapter, namely that form must be applicable in experience. Kant says: ‘Apart from this relation synthetic a priori principles are completely impossible’.115 This is because there would be ‘no third something, that is, no object, in which the synthetic unity can exhibit the objective reality of its concepts’.116 Kant once again 229


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology uses the notion of a third thing to express the requirement that the forms of thought apply within experience. The third thing was first identified as temporal determination. 117 Next it was identified as the temporal whole of experience arising out of the cooperation of the faculties of intuition, imagination and understanding.118 Kant’s point seems to be that the synthesis of a category with an empirical intuition can only arise in relation to the possibility of experience, which we have now discovered must be understood in spatial terms. This development coincides with the claim that the third thing, which is the distinctive characteristic of figurative synthesis, is the object of experience. Synthesis arising from the productive imagination requires a bridging term, the most explicit example of which is the temporal schema. It now appears that the full elaboration of figurative synthesis requires the spatial articulation of the categories. But Kant’s account does not simply add articulation in respect of the pure form of space to that of time. He says we can know much about space in general, and the figures constructed in it, by productive imagination at the a priori level alone, but only insofar as space is considered as ‘a condition of the appearances which constitute the material for outer experience’ can we achieve knowledge and secure the objective validity of the synthesis grounded on the categories.119 If the categories are to qualify as principles, space has to be considered not merely as a form of intuition, but as the condition of our receptivity to a material given.120 Does this mean that the principles supply a spatial schematism? It might be argued that it is not appropriate to suggest that the spatial articulation of the categories necessary for their application to materially given or existent appearances qualifies as a schema. A schema counts as a pure synthesis and is a transcendental product of the imagination: which concerns the determination of inner sense in general according to conditions of its form (time), in respect of all representations, so far as these representations are to be connected a priori in one concept in conformity with the unity of apperception . . .121

Admittedly, Kant consistently characterises the schema only in temporal terms.122 However, I take it that the schema’s principal role is to establish the possibility of figurative synthesis. If a category were not temporally articulated, an intuition could not be subsumed under it. And now it has emerged that a priori knowledge would count as a fiction, were it not related to space.123 If figurative synthesis is to 230


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism achieve objective knowledge then the categories must be not only temporally but also spatially articulated.124 The ‘Schematism’ establishes the possibility of figurative synthesis and thus counts as transcendental, that is, as part of the necessary structure of experience. In the ‘Principles’ chapter we discover that the full articulation of the transcendental structure of experience requires that figurative synthesis is not only temporal, but also spatial. In the concluding sections of the ‘Analytic’, Kant argues not that any actual empirical experience requires a spatial synthesis, but rather that the general form of experience is spatial. He thus re-establishes space as a transcendental condition of experience, something that was first outlined in the ‘Transcendental Aesthetic’. Space must be presupposed if objects are to be established as ‘outside me’ and ‘outside and alongside one another’ in space.125 Only such objects qualify as objects of knowledge for Kant. Time is the form of all intuitions without exception, but the knowledge that Kant seeks to establish is of intuitions that are also extended in space. If spatial iteration of the categories is a necessary condition of the figurative synthesis that gives rise to knowledge, then it makes sense to suggest that a spatial schematism appears in the ‘Principles’ chapter. Crucially, though, and as I have argued, space must be understood not only at the formal level, but also as bound up with the material given in appearances. For this reason, the spatially figurative synthesis is not and cannot be purely formal. This may seem to close the discussion on whether or not it makes sense to speak of a spatial schematism. But I think rather, that the form of experience has now emerged as entailing an anticipation of a material given. Admittedly, this development cannot establish the need for a spatial schema in the narrow terms of the ‘Schematism’ chapter. However, if a schema can be principally understood as the formal or transcendental iteration of a concept through figurative synthesis, what Kant now reveals is that the latter must be effected spatially, and not only temporally, if knowledge of extra-mental objects is to be achieved. The purely formal level of analysis adopted in the ‘Schematism’ chapter has been replaced by an account that is more attuned to the possibility of empirical experience. Despite a forceful beginning, the role of space slips out of view in the specific arguments of the first three principles, although, as we will see, there are some traces of its importance. What remains a continual focus in both editions is experience as actual or possible. This, as we have seen, is the new third thing that will make knowledge and the validation of the categories possible and I have suggested that it 231


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology is closely associated with space. It appears that the role of space in establishing the possibility of experience, although initially realised by Kant as evidenced by our discussion above, only reappears later in his argument and especially in the B edition. This is revealed by the increasingly explicit claims about space in the ‘Postulates’, especially in the B edition and in the ‘General Note’.126 The ‘Axioms’ are concerned with the construction of concepts in space. Geometry, or the mathematics of space, is based ‘upon the successive synthesis of the productive imagination in the generation of figures’.127 Concepts alone are insufficient for geometric insight, which always requires construction in space. In drawing a line, I must draw it in thought, ‘generating from a point all its parts one after another’.128 This process of spatial synthesis is necessary for a determinate geometric intuition. This is synthesis in process within cognition. It is to be contrasted to aesthetic synthetic process that is not directed to any determinate cognition, but reveals the synthetic process, the cooperation of the faculties, necessary for any knowledge whatsoever.129 Synthesis of spaces and times is necessary for all knowledge of outer experience.130 But does space now qualify as a constituent part of what Kant says is the third something, ‘medium’, or ‘whole’, within which figurative synthesis is possible?131 Kant’s initial account states that there is only one such framework, which is time.132 But although there is no denying that time is a necessary condition of synthesis, surely the account as it stands is insufficient. Clearly, extensive magnitudes must be predicated of outer objects and can be presented only in the form of outer objects: that is, space. It would appear, therefore, that Kant should have revised his account of the medium that is necessary for any possible experience and established that this is spatiotemporal and not just temporal. This is, in any case, surely evident from the fact that the knowledge he seeks to legitimate is of outer things, the form of which is spatial. Kant makes some progress in the direction of a spatial articulation of figurative synthesis in the ‘Axioms’, but his account remains strictly at the level of the formal construction of space and makes no mention of space as a condition of material appearances. The rule for figurative synthesis introduced in the ‘Axioms’ is the spatial analogue to the temporal schematism, for it operates at the a priori level. It establishes that it is only if we are capable of spatially schematising a concept, of constructing it in space, that our concepts can qualify as applying to really existing objects. 232


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism The ‘Anticipations’ add nothing explicitly to the elaboration of the spatial schematism, concerned as they are with the intensive quality of any appearance. The anticipations of perception are flowing magnitudes insofar as they are temporal in status. But, as any intensive magnitude will be presented only in relation to an extensive magnitude, at the level of experience, we can conclude that they must finally relate to space. Intensive magnitude is a characteristic of appearances, which the ‘Axioms’ establish as spatial.133 Even more importantly, Kant now insists on the insufficiency of a purely formal account of the legitimation of the categories. Form alone can only anticipate what must be given empirically, as we saw in the previous section. The material given is the necessary counterpart to the formal principle of quality, and empirical matter can only be understood in spatial terms. The first and second ‘Analogies’ do not prima facie discuss space. They are concerned with the temporal determination of an appearance, in accordance with the ‘Schematism’. While the first ‘Analogy’ concerns the temporal determination of a singular appearing object as a substance, the second turns to the temporal relation in which a plurality of substances stand to one another. But how can an appearance that is not merely a subjective representation be conceived in purely temporal terms? We will see that Kant revises his account in the ‘Postulates’, and first implicitly and then explicitly suggests that the analogies must be understood as bearing a spatial significance. The third ‘Analogy’ seems to promise an explicit role for space, for it is concerned with the necessary whole or ‘community’ within which all appearances coexist. Moreover, in the second edition statement of the general principle of the third analogy this whole is identified as space.134 Surely, here at last Kant must explicitly return to the commitment he made to the necessity of the spatial dimension of experience at the outset of the ‘Principles’ in both editions. How could the interaction of a plurality of substances be understood without reference to their coexistence in space? Kant undercuts this expectation by distinguishing between two senses in which community can be understood. First, there is communio or communio spatii, which Kant considers to count only as ‘local community’.135 He contrasts this to commercio by which he means a ‘dynamical community’ on which local community is dependent.136 Thus, spatial community – if this is what he means by local community – is based on a community of substances. If so, spatial determination would be only an after-effect of a primary community among substances. However, this does not seem to be consistent with the 233


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology internal logic of his account. While a singular substance was analysed in purely temporal terms in the first ‘Analogy’, even in a commercium appearances ‘stand outside each other and yet in connection’ and surely this can only be comprehended as a spatial relation?137 Proliferating the difficulties involved in trying to clarify Kant’s terms, he also speaks of a ‘community (communio) of apperception’.138 Apperception is a subjective community insofar as it is a unity of appearances within the subject. It must stand in a reciprocal influence with a ‘real community (commercium) of substances’.139 In the context of the general strategy of the ‘Principles’ we have uncovered so far, this sounds as if subjective unification of representations must refer to a unity in the appearances themselves. This supports my insistence on the relation between the subjective and objective sides of the deduction. And surely it would be plausible to conclude that such an objective unity would have to be in space, the form of outer intuitions. But we have to wait until the ‘General Note’ at the conclusion of the ‘Postulates’ for confirmation of this suspicion.140 The first ‘Postulate’ is concerned directly with the construction of a concept in space as the condition for the possibility of an empirical object.141 But if this is so, then it calls for a re-evaluation of the role of space in the first, second and third analogies. Kant says that the concepts of substance, causality and community would contain only ‘a merely arbitrary synthesis’ if they were not related to experience in general.142 I have already argued that Kant’s introduction of actual or possible experience should be understood in spatial terms. We can conclude that if a category is to qualify as the form of experience then it must be constructible in space. Substance, causal succession and the reciprocal community of substances must be spatially articulated. Nevertheless, we can only infer this from Kant’s account. The role of space in the second ‘Postulate’ comes to the fore in the ‘Refutation of Idealism’. Even self-knowledge must stand in relation to outer objects in space. The ‘Refutation’ starts from what Descartes considered indubitable, the Cogito, and argues that the latter is only certain insofar as we are immediately and not mediately, as Descartes held, conscious of outer objects. Thus, unless we doubt the Cogito, we can be confident that we are capable of knowing outer objects.143 If our outer sense were merely a product of imagination, inner intuition would also be annulled! If the internal activity of reflection need not be doubted – and even Descartes did not doubt that – then there is no need to doubt the evidence of outer sense. I think, therefore, there are things in the world. 234


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism Prior to this striking argument, Kant declares the primacy of the given object in relation to the concept. The rule that establishes the actuality of appearances requires not that we construct the concept, but that we recognise that a ‘perception can precede the concept’.144 The ‘Refutation’ then shows how this precedence can only be the case if we accept the material existence of appearances in space. The second ‘Postulate’ establishes the primacy of perception over concepts in Kant’s formal idealism. Kant is committed to the view that the formal structure must anticipate the possibility of application in space. Kant sometimes suggests that perception [Wahrnehmung] coincides with intuition, as he does at this stage of his account of the second postulate.145 But he sometimes makes the further suggestion that all perceptions are combined with a concept, or, as he sometimes says, with consciousness.146 In the passage just cited, he suggests that a perception coincides with an intuition that is not yet determined by a concept. We need not conclude that the intuition is wholly unrelated to a concept, for the intuition anticipates the possibility of conceptualisation. But this also supports the suggestion that an intuition could, under other circumstances, escape determination by a concept.147 The third ‘Postulate’ is concerned with necessity, but strictly with material, that is, causal and not logical necessity.148 In the second ‘Analogy’ Kant had already argued that we cannot know the existence of things as necessary, but we can know that their state is so, insofar as they stand in a causal relation to some other appearance.149 Kant does not explicitly mention space in the final ‘Postulate’, but as he has already insisted in the second ‘Postulate’ that space is the condition of the material existence of objects, we can conclude that material necessity arises only in space. Thus, the rearticulation of the concept of causality as spatial, inferred by the first ‘Postulate’, coincides with the third ‘Postulate’s’ characterisation of causal relation as equivalent to material necessity. The ‘General Note’ was added only in the B edition. The development that we have already seen in the ‘Postulates’ is now stated very clearly and counts as a rehearsal of the content of the ‘Principles’ chapter as a whole. Only in relation to an intuition can something be shown to be permanent, alterable through causality and coexistent in a community with other permanent things or substances.150 The concept of substance must be understood as a principle for the possibility of experience, if the category is to achieve objective reality.151 And this entails that the latter will refer to our 235


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology ‘knowledge of an object given in empirical intuition’.152 What was only inferred in the first and third ‘Postulates’ is now explicitly stated. It emerges that in order for the concept to relate to objects of experience, substances must be recognised as spatially existing appearances or objects. The permanent is not to be found in inner sense.153 Causal relations, properly understood, hold between spatially existing substances.154 Finally, community is firmly established as spatial, whereas we found that its status was not entirely clear in the third ‘Analogy’.155 What I am calling the spatial schematism of the categories is thus completed. If we are in any doubt that the original goal of the ‘Transcendental Deduction’ has finally been achieved, we need only turn to the text: But it is an even more noteworthy fact, that in order to understand the possibility of things in conformity with the categories, and so to demonstrate the objective reality of the latter, we need, not merely intuitions, but intuitions that are in all cases outer intuitions.156

By the end of the ‘Postulates’, even in the A edition, it was becoming clearer to Kant that the possibility of experience, and thus the validity of the categories, requires the applicability of the latter in space. By the time of adding the ‘General Note’ to the B edition, he was adamant on the point. What we find at the conclusion of the second edition of the ‘Principles’ is not only the elaboration of the categories as the principles for the possibility of experience and for appearances themselves as existing, but also a spatial schematism. The latter is distinguished from the official schematism, not only in referring to space–time and not simply to time but also insofar as turning to consider space brings out the dual face of the aesthetic element of experience. Space is the form of experience that allows us to take up the empirical given. And the account of figurative synthesis concludes with its spatial iteration, because it is principally space that allows for bridging the gap between receptivity and the reflective forms or categories. An understanding of the complex role played by space makes possible a comprehension of how the forms of intuition and the categories are capable of relating to an empirical object. In achieving this, the often repeated goal of the ‘Deduction’ is now also legitimated, that is, the claim that the categories of the understanding apply to empirical objects. The final sentence of the ‘General Note’ sums up Kant’s judgement on the matter: 236


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism The final outcome of this whole section is therefore this: all principles of the pure understanding are nothing more than principles a priori of the possibility of experience, and to experience alone do all a priori synthetic propositions relate – indeed, their possibility itself rests entirely on this relation.157

The spatial schematism of the principles is not merely an addition to the temporal schematism. Whereas the latter was able to operate wholly at the level of experience in general, albeit in anticipation of application to an object in general, the spatial schematism reveals a necessary preoccupation with the material and empirical given in experience. This is hardly surprising as space is the form of outer intuitions, that is, of objects given to us, which necessarily stand beyond, though accessible to, the mind. Nevertheless, the spatial schematism incorporates the official version and reinforces the temporal iteration of experience. We have had to excavate Kant’s text and draw out hidden connections in order to uncover the spatial schematism offered by the ‘Principles’ chapter. We have also seen that the evidence for this is not continuous and only comes centre-stage in the ‘Refutation’ and ‘General Remark’. However I believe there is sufficient evidence in both editions to conclude that Kant gradually came to the conclusion that if the categories are to anticipate empirical application, in the way that is necessary for their objective validity, they must be spatially articulated.158 And I further believe that it is helpful to think of this as a spatial schematism. Before moving on to the final developments in Kant’s analysis of the a priori form of empirical experience, it is important to emphasise the level at which he is operating. Even though he has moved a great distance from the initial statement contained in the ‘Transcendental Deduction’ and is now concerned with the material givenness or existence of objects, he has not strayed from the analysis of a priori determining judgement. He has, however, shown that transcendental analysis operates in anticipation of a material given. As he says of the ‘Anticipations of Perception’ in particular: ‘This anticipation of perception must always, however, appear somewhat strange to anyone trained in transcendental reflection, and to any student of nature who by such teaching has been trained to circumspection.’159 Despite the strangeness of the thesis, if Kant were not committed to the anticipatory status of a priori knowledge, his position would be condemned to material idealism. 237


29/3/07

8:12 AM



VI Approaching empirical experience: material necessity and the possibility of a gap between transcendental order and empirical experience We have seen that the objective validity of the categories will only be established insofar as they are shown to be applicable within empirical experience. Kant has argued that this requires a temporal and, I have suggested, a spatial schematism. We have also seen that the ‘Principles’ chapter offers only a subjective proof and yet this establishes the possibility of objective experience. Throughout his account of the principles, Kant highlights the cooperation of the faculties necessary for the figurative synthesis of objects given in time and space. The ‘Principles’ chapter displays the two sides of the deduction working in collaboration with one another so as to give rise to knowledge of empirical objects. In my account of the ‘Transcendental Deduction’ in Chapter 4, I insisted that it requires a cooperation of the faculties, although I conceded that there is a hegemony in favour of the understanding. Examination of the text of the ‘Principles’ chapter shows it to be consistent with the ‘Deduction’ in this regard. The principles are the principles of pure understanding, yet we have seen that the articulation of the concepts of the understanding, qualifying them as the form of objects of empirical experience, requires a complex account of the relation in which the understanding stands to the faculties of intuition and imagination. 160 The concepts must not only be temporally but, I have argued, also spatially schematised insofar as the understanding must be supplemented by intuition through the exercise of the figurative synthesis of imagination. When judgement is exercised in the interests of knowledge, the understanding is always determining at the end of the day, but it can only succeed in its task if it operates in cooperation with a plurality of faculties. The complexity of the relation in which the understanding stands to intuition and imagination was not yet established fully in the ‘Transcendental Deduction’ and needs the deeper investigation of the structure of figurative synthesis provided by the full ‘Analytic’ or canon. We have come a long way since Kant first claimed that the categories provide the form of empirical experience in the ‘Transcendental Deduction’. But are we there yet? We have seen that the ‘Schematism’ and ‘Principles’ chapters provide a successive determination of the categories in anticipation of application within empirical experience. Even at the conclusion of the ‘Principles’, however, Kant holds that 238


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism he has only established the form of empirical experience in general. Empirical determinations, as he stated in the ‘Deduction’, are not yet in the frame. Nevertheless, with the conclusion of the ‘Analytic’ we would appear to have before us the fully articulated version of the a priori form of empirical experience. We have seen that Kant’s confidence in his ability to articulate the form of empirical experience at a purely a priori level is tempered by his continuing attempts to fine-tune that form to render it capable of anticipating empirical application. The late discussion of what he calls the principle of hypothetical necessity in the third ‘Postulate’ is particularly instructive in this regard.161 Previously, in the ‘Axioms’, Kant had simply dismissed the ‘idle objections’ that sensible objects might not conform to their construction in space, on the grounds that this would undermine the objective validity of space, mathematics and geometry.162 The very possibility of a failure of fit between categories and sensible objects was thus dismissed out of hand on the grounds of the pernicious results that would ensue. But this is, of course, to beg the question. In contrast, in the third ‘Postulate’, Kant considers the question somewhat more seriously, although his conclusion is no less confident. In a development of the second ‘Analogy’, Kant addresses the question of the material, rather than merely formal necessary connection between objects. Consistently with his previous account, he concludes that we can know things only as states following on other states and not as isolated existing objects.163 The principal shift in the ‘Postulates’ is in the emphasis put on material necessity and, simultaneously, on the subjective sources of any causal judgement.164 Kant now refers to the principle of hypothetical necessity, which he says is necessary for the very possibility of nature.165 This is not so much a new principle as a further specification of the law of causation. He immediately proceeds to introduce four propositions, although he gives little sense of the relation in which these stand to the general principle. A materially necessary connection between objects is secured by the additional presuppositions that there will be no hiatus and no leaps in nature; that nothing happens without a cause; and that necessity in nature is conditioned and therefore intelligible.166 These jointly: allow of nothing in the empirical synthesis which may do violence or detriment to the understanding and to the continuous connection of all appearances – that is, to the unity of the concepts of the understanding. For in the understanding alone is possible the unity of experience, in which all perceptions must have their place.167 239


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology Kant’s aim is to establish the systematic unity of empirical nature so as to make the fit between the latter and the system of the categories secure. The difference between the treatment here and in the ‘Axioms’ is that while Kant insists in both on the legislation of the categories, he now finds it necessary to further elaborate the principles in order to guarantee the empirical, now recognised as identical with the material, applicability of the categories.168 The introduction not only of a material and not merely formal articulation of causal necessity, but also of four supplementary propositions, suggests he is not yet quite satisfied that he has established that the a priori formal structure coincides with the form of empirical experience. But if this is true, he suppresses his anxiety quickly. The principle of hypothetical necessity legislates from its a priori position that there will be nothing in empirical syntheses that fails to fit with the principles of understanding. His painstaking and repeated arguments that the principles can only apply insofar as they anticipate experience is abruptly short-circuited. In the next chapter we will find that, on further reflection, Kant realises that the form of empirical experience cannot simply be established by the system of the principles. In the introductions to the Critique of Judgement he introduces a distinct new principle, though a reflective not a determinative one, of the purposiveness of nature for judgement.169 This is also a priori insofar as it arises from the pure subjective activity of judgement and so is not to be confused with the empirical determinations Kant discussed in the ‘Deduction’. The remit of this further a priori principle is to secure that empirical nature fits with our judgements. Kant says that the new principle establishes empirical systematicity, in contrast to the systematicity of the principles. Thus this principle establishes a further link in the chain between a priori form and empirical givenness. The attempted Diktat expressed as the principle of hypothetical necessity has not yet established that the categories provide the form of empirical experience. This will only be achieved with the addition of a principle that delves even deeper into the relation between subjective powers and empirical given in order to show how knowledge is possible. It is also arguable that Kant is only finally in position to establish the validity of the categories when he recognises the limitation of the formal structure and accepts that there must be a gap between the transcendental and empirical levels of analyses. This is the deeper significance of the subjective status of reflective judgement. We can only seek order in the empirical world and, in advance of experience, we cannot 240


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism be apodeictically certain that objects will in fact display the order we attribute to them.

Conclusion In this chapter I have shown how synthesis in process is at work throughout the ‘Analytic’ and have argued that this is the subjective side of the objective deduction as a whole. I have also brought out the anticipatory status of a priori knowledge insofar as it stands in a necessary relation to a material given. However, the ‘Analytic’ stays strictly at the level of the formal conditions of the possibility of experience in general. Kant finally comes to the view that if he is to show that any actual empirical experience necessarily falls under the rules of the categories of the understanding, a further stage of analysis is required. He offers this final stage of his epistemological argument in the account of the reflective principle of the systematicity of empirical nature in the ‘Introductions’ to the Critique of Judgement, which will be addressed in the next chapter.

Notes 1. CPR, A xvi/xvii and A 93, B 126 respectively. See discussion of the initial characterisation of the task of the ‘Deduction’ in Chapter 5, p. 172. 2. B 161–2. See Chapter 4, pp. 143–5. 3. See discussion of Buchdahl’s MPS in Chapter 2, p. 49. 4. Allison, for instance, holds that the ‘Schematism’ chapter is necessary for the full account of the deduction. He also investigates how the ‘Principles’ contribute to Kant’s core argument. See KTI(1983), Part 3. 5. This is important for forging a link to the discussion in the next chapter of the principle of the purposiveness of nature, which I will argue is the completion of Kant’s epistemological argument. 6. A 156, B 195. The relevant phrase is: Erfahrung (es sei wirkliche oder doch mögliche). 7. A 93, B 125–6. 8. A 93, B 125. 9. A 93, B 126. 10. A 93, B 126. 11. A 92, B 125. 12. A 95. 13. A 95. 14. A 95. 241


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 15. See, for instance, B 161–2. 16. B 163. 17. B 165. I have altered Kemp Smith’s translation, which makes Kant’s point opaque. Kant’s own expression is very clear on this issue. Kant makes the same point in the A ‘Deduction’ at A 127/8. 18. The gap only becomes fully apparent to him in the Critique of Judgement. See discussion of this development in Chapter 7, pp. 249–50. 19. In Chapter 7, I will argue for this position (see pp. 254–5). 20. B 167. 21. B 168/9. 22. I discussed indeterminate intuitions in Chapter 4, pp. 130 and 145. 23. A 132, B 171. 24. See CPR, A 12, B 26; A 61, B 85; A 63, B 88. 25. A 796, B 824. 26. A 796, B 824. See also A 63, B 88. 27. A 136, B 175. 28. A95. See previous section above, p. 210. 29. A 136, B 175. 30. A 137, B 176. 31. A 138, B 177. 32. A 138, B 177. 33. Gibbons also sees the Schematism as addressing how intuition can be schematised with understanding. See Kant’s Theory of Imagination, pp. 70–1. 34. A 138, B 177. 35. A 136, B 175. 36. A 137, B 176; A 138, B 177; A 139, B 178. 37. A 139, B 178. 38. A 146, B 185. 39. A 138, B 177; A 142, B 181. 40. A 140, B 179–80. 41. A 141, B 180. 42. A 141, B 180. 43. See discussion of Heidegger in Chapter 4, p. 150. 44. And while greater depth is achieved in the ‘Schematism’, we have seen in Chapter 4, (pp. 151–60) that only aesthetic judgement offers a full depth analysis of synthesis. 45. CPR, A 142, B 181; A 141, B 180. 46. A 141, B 180–1. 47. A 136, B 175. 48. A 136, B 175. 49. A 148/9, B 188. 50. A 148, B 187. 242


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism 51. A 149, B 188. 52. Indeed, the intentional direction of concepts was already established in the A ‘Deduction’. 53. Kant talks of linking two concepts at A 155, B 194, but he must mean that the second is an empirical concept that stands for something given in intuition. 54. A 155, B 194. 55. A 155, B 194–5. 56. A 156, B 195. Kemp Smith is not consistent in his translation of these terms, using both significance and meaning to translate Bedeutung (A 146, B 185; A 155, B 194–5, respectively) and both sense and significance to translate Sinn (A 156, B 195; A 155, B 194–5, respectively). But Kant’s intention, which does not require a modern technical use of these terms, is unambiguous. Both categories and pure intuitions are mere forms of thought if they are not applied to empirical intuitions. On the contrast between the role of meaning for Kant and modern analytical ‘theories of meaning’, see Gardner, Kant and the Critique of Pure Reason, pp. 288–9. 57. CPR, B 160–1. See discussion in Chapter 4, p. 138. 58. A 156, B 195. The claim that categories have objective reality only in relation to intuitions comes at the outset of the A ‘Deduction’ at A 95. See also A 155, B 194–5. We saw at A 146, B 185 that categories only have significance (Bedeutung) insofar as they are schematised. 59. A 156, B 195. 60. A 156/7, B 196. 61. A 158, B 197. 62. B 168/9. 63. A 149, B 188. 64. See A 157, B 196. Experience is empirical synthesis. 65. A 157/8, B 196/7. 66. A 158, B 197. 67. Clearly, my reading of Kant as laying out the conditions of objectively valid cognition – or knowledge – in a hierarchical fashion is distinct from the claim that the resultant account of those conditions establishes them as anticipatory of the possibility of experience. Nevertheless, Kant’s style of presentation reinforces the strictly formal status of transcendental cognition. 68. A 160, B 199. 69. A 162–3, B 202–3. 70. A 166, B 207. 71. A 166, B 207. 72. A 166, B 207/8. 73. See discussion of sensation in Chapter 3, pp. 102–3. 74. A 167, B 209. 243


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90.

91. 92.

93. 94. 95. 96.

97. 98. 99.

100.

A 168, B 210. A 175, B 217. A 170, B 211/2. A 176, B 218. A 166, B 208. See discussion of Guyer in Chapter 1, pp. 21–2. A 160, B 199. A 160, B 199/200. B 218 A 160, B 199; A 178, B 220. B 218. See discussion in next section, pp. 229–37. A 179, B 222. A 179/80, B 222. A 180, B 222. See Chapter 1 (pp. 21–2) for my interpretation of A 125/6 where Kant says that nature should [soll] be a unity. At A 181, B 224 he says this unity can be thought only through a schema, not a concept. The initial account in the A ‘Deduction’ has been revealed to be merely the beginning of a more detailed account. A 180/1, B 223. We could also interpret Kant’s denial of the transcendental status of the principles in the following way. He is just about to insist that the Analogies are not concerned with things in themselves. He may simply have used ‘transcendental’, when ‘transcendent’ would have been more appropriate. Kant does not always adequately distinguish between these two terms, as has often been remarked. In any case, the result of both interpretations is the same. The principles are transcendental in status, as Kant makes clear in the Introductions to the Critique of Judgement. See Chapter 7, p. 249. CPR, A 181, B 223. B 233–4. B 234. Note the similarity to the tripartite argument of the A ‘Deduction’ and the part played by the faculties. A 189/90, B 234/5. Kant goes on to use the term Gegenstand to refer to objects of experience, supporting Allison’s reading of the B ‘Deduction’. See discussion of the latter in Chapter 4, pp. 132–3. See my discussion of representation in Chapter 3. A 190, B 235. Merleau-Ponty, Phenomenology of Perception, p. 322 (French edition p. 372) ‘un véritable en-soi-pour-nous’. However, Merleau-Ponty’s phrase has a perceptual rather than a cognitive connotation. I discuss this in my ‘Kant’s Phenomenological Reduction?’. CPR, A 190, B 235. 244


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism 101. A 197, B 242. 102. A 197, B 242. 103. See discussion of the status of representations in Chapter 3, pp. 96–100. 104. A 207, B 252. 105. A 206/7, B 252. 106. A 210, B 255/6: ‘Wir antizipieren nur unsere eigene Apprehension . . .’ 107. A 219, B 267. 108. A 233/4, B 286. 109. A 149, B 188. 110. A 149, B 188. 111. A 218, B 266. 112. CPR, A 157, B 196. 113. A 156, B 195. 114. A 156, B 195. 115. A 157, B 196; my emphasis. 116. A 157, B 196. 117. See discussion of this expression in the section on ‘Schematism’ above, pp. 213–17. 118. A 155, B 194; discussed in the previous section, p. 219. 119. A 157, B 196. This mention of the materiality of appearances predates Kant’s introduction of their existence. However, I understand this early comment as promissory for his later account where he will establish not only the formal conditions, but also the material givenness of objects. 120. See my discussion of the two-sided structure of receptivity at the end of Chapter 3, pp. 105–8. 121. A 142, B 181. 122. See Gardner, Kant and the Critique of Pure Reason, p. 169, on why time is the key to transcendental schematism. Gardner says this is because time is ‘the most general unifying condition of intuitions and concepts; all sensible objects are intuited in time, and all conceptual activity stands under the condition of self-consciousness, the objects of which are temporal.’ This is undoubtedly correct. However, my argument is that, as Kant intends to establish the validity of the categories to all objects of experience and the latter are necessarily intuited in space, then it is arguable that he needs to provide not only a temporal, but also a spatial schematism. See also Makkreel, IIK, pp. 31–2. Makkreel argues that despite the emergence of what he sees as a role for space in the Schematism, time is not dependent on space and ‘their relation is conceived as reciprocal’. 123. See above; A 157, B 196. 124. In Chapter 8 (pp. 292–3 ) I will argue that this is also true of aesthetic synthesis and despite apparent counter-examples, such as music. 245


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 125. A 23, B 38. 126. Guyer is one of the few commentators who comments on the role played by space in the ‘Principles’ chapter. However, he sees this late development in Kant’s position as leading away from transcendental idealism, which Guyer believes inevitably leads to impositionalism. See Guyer, KCK, especially Ch. 16, for the role of space within transcendental idealism. See also pp. 288–9 for the claim that Kant develops an account of space that entails the failure of transcendental idealism. 127. CPR, A 163, B 204. 128. A 162/3, B 203. 129. See Chapter 4, pp. 151–6, on synthesis in process. 130. A 165/6, B 206. 131. A 155, B 194. 132. A 155, B 194. 133. A 166, B 207. 134. B 256. 135. A 213, B 260. 136. A 213, B 260. 137. A 215, B 261/2. We have already seen that objects’ standing outside of one another was shown to require space in the ‘Transcendental Aesthetic’ at A 23, B 38. 138. A 214, B 261. 139. A 214/15, B 261. 140. B 292–3. 141. A 221, B 268. See also A 224, B 271 on space as the formal a priori condition of outer experiences. 142. A 221–2, B 268–9. 143. B 276 footnote. 144. A 225, B 272/3. 145. See also A 115. 146. See A 120, B 160. 147. See discussion of undetermined intuitions and, in particular, aesthetic intuitions in Chapters 3 (p. 102) and 4 (p. 145). 148. A 226/7, B 279. 149. A 179/80, B 222. 150. B 288. 151. B 288. 152. B 289. 153. B 292. 154. B 292. 155. B 292–3. 156. B 291. 157. B 294. 246


29/3/07

8:12 AM


A Priori Knowledge, Material Given and Spatial Schematism 158. In the Metaphysical Foundations of Natural Science (= MFNS), first published in 1786, Kant analyses matter as the movable in space. See Michael Friedman’s ‘Introduction’ to the English translation, p. xv. As Friedman comments, in the second edition of the Principles chapter, Kant claims the categories must refer not only to intuitions, but also to outer intuitions. See Friedmann, pp. xx–xxii. In contrast, though, the MFNS operates at a metaphysical level insofar as it is concerned with the possibility of the further a priori determination of empirically given objects. See CPR, AA 181. The Principles chapter, meanwhile, remains at the transcendental level in being concerned only with the possibility of the empirically given in general. I believe we can make sense of the apparent coincidence between the ‘Principles’ and MFNS as to the need for external intuition by making use of my suggestion that Kant’s method is anticipatory of empirical experience. The ‘Principles’ chapter anticipates the metaphysical analysis of moving objects in space treated in 1786. I have also shown that there is evidence of the need for a spatial schematism even in the first edition of CPR. 159. CPR, A 175, B 217. 160. A 148, B 187. 161. A 228, B 280. 162. A 165/6, B 206/7. 163. A 226–8, B 279–280. 164. A 226/7, B 279; A 233–4, B286, respectively. 165. A 228, B 280. 166. A 228/9, B 280/1. 167. A 229/30, B 282. 168. The new emphasis on the material status of causal relation is reinforced by Kant’s saying that ‘without material [ohne Stoff] nothing whatsoever can be thought’ (A 232, B 284). 169. Indeed an intermediate stage is offered in the ‘Regulative Employment of the Ideas of Pure Reason’, where the systematicity of nature is treated as an idea of reason. However, Kant’s considered position is that the systematicity of empirical nature requires a principle of judgement.

247


29/3/07

8:12 AM


7

Empirical Systematicity and its Relation to Aesthetic Judgement

We have found that the ‘Deduction’ of aesthetic judgement reveals that the latter is based on the subjective conditions of cognition.1 In Chapter 5 I argued that the principle of taste or common sense is aesthetic in status, insofar as it counts as the principle of the faculty of judgement as such. Kant makes no mention of any other principle of judgement other than taste, so I disagreed with Allison’s suggestion that taste is grounded on a further principle, namely, the principle of judgement in its subjective employment. In my view, the principle of taste is the only principle that expresses the autonomous use of judgement. My solution appears to give rise to a worrying result, namely, that cognition is grounded on an aesthetic principle. There are strong reasons, both independent and internal to Kant’s philosophy, for resisting such a conclusion. A related problem arises for my interpretation in the two Introductions to the Critique of Judgement insofar as Kant introduces a principle of judgement that does not at first sight count as aesthetic. This is the principle of the purposiveness of nature and is the basis for our presupposition of systematicity across the range of empirical laws, thus making possible empirical judgements. Kant suggests that the ‘Analytic of Aesthetic Judgement’ will serve as an exposition and deduction of this principle, thus it would appear that taste is grounded in the principle for empirical systematicity.2 We are thereby faced with the reverse of the problem in the ‘Deduction’. That Kant should have courted two contrary positions, both leading to conclusions that would eliminate the distinctiveness of the principles grounding the first and third Critiques, should make us pause for thought. I will show how both problems can be resolved, once the exemplary status of taste is established. In the first section of this chapter, I examine the role played by the principle of the purposiveness of nature. Although reflective judgement has often been viewed as an optional supplement to Kant’s 248


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement account of determining judgement, I argue that it is necessary for the completion of the latter’s task of establishing the possibility of empirical knowledge. Kant often characterises the principle of the purposiveness of nature as identical to the presupposition that there is systematicity across the range of empirical laws. In the second section, I suggest that we can distinguish between the general question of a fit between mind and empirical nature and the more specific claim for systematicity. I argue that the general level of purposiveness makes possible the completion of the Copernican turn at the empirical level. In contrast, the concept of empirical systematicity is an instrumental application of this general principle. In the third section, I discuss the dual direction of both the purposiveness of nature and the purposiveness of judgement, insofar as they are directed to subject and object. I further suggest that they are expressions of one and the same principle of the purposiveness of nature for our judgement. Judgement is the faculty that allows sensibility to be taken up by the understanding. It is thus that judgement makes possible a relation between subjectivity and objects in the world. Finally, in the fourth section, I turn to Kant’s problematic claim that the ‘Analytic of Aesthetic Judgement’ will provide an exposition, and even a deduction, of the purposiveness of nature. I draw on the distinctions established in the earlier sections in arguing that aesthetic judgements are exemplary of the possibility of empirical synthesis.

I Understanding goes out into the empirical world In this section I will argue that formal reflective judgement of nature is the necessary supplement to the ‘Analytic’ of the first Critique insofar as it makes possible the application of the categorical system within the empirical world.3 Retrospectively, in the two Introductions to the third Critique, Kant reveals that the first Critique established only the transcendental structure of experience. He suggests that empirical experience might have been so varied that we would have been incapable of grasping [fassen] it through the categorial system alone.4 There could have been a wholly consistent mental order that did not tally with the world out there. Having raised this problem at the beginning of the Critique of Judgement, Kant answers it by saying that the categorical system of 249


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology the first Critique is supplemented by a further system founded on the principle of the purposiveness of nature for our judgement. This principle establishes that empirical nature is organised as if it were a system oriented towards a purpose. However, there is no purpose or final cause and, thus, the principle counts as subjective or reflective. Within the broad range of reflective purposiveness that is the topic of the Critique of Judgement, Kant distinguishes between formal purposiveness and real or objective purposiveness. Formal purposiveness is the systematicity reflection finds in empirical nature viewed as a whole and is subjective in status.5 Objective purposiveness is teleological and is to be found either in the internal organisation of an object viewed as organic, or insofar as an object is viewed as standing in relation to the final purpose of nature.6 Teleological judgement is subjective in the broad sense that it rests on judgement’s principle of purposiveness and is not determinative of objects. However, while formal purposiveness arises in relation to intuition, the real purposiveness of nature is judged in respect of concepts.7 Teleological judgement is objective in the weak sense that it applies a subjective principle in conjunction with a concept in order to make sense of objects. While formal purposiveness is necessary for any empirical cognition whatsoever, objective purposiveness is necessary only for objects insofar as we view them as natural purposes. Kant suggests that the teleological order in nature offers a ‘logical presentation of the purposiveness of nature’, allowing us to get a grasp on the more indeterminate formal purposiveness.8 Teleological presentation counts as logical because it requires a concept of understanding or an idea of reason. The problem formal purposiveness addresses is what Allison refers to as the possibility of ‘empirical chaos’, which would arise were the principles of the first Critique incapable of getting a foothold in the empirical world.9 Kant solves the problem thus: Now this principle can only be the following: since universal natural laws have their basis in understanding, which prescribes them to nature (though only according to the universal concept of it as a nature), the particular empirical laws must, as regards what the universal laws have left undetermined in them, be viewed in terms of such a unity as [they would have] if they too had been given by an understanding (even though not ours) so as to assist our cognitive powers by making possible a system of experience in terms of particular natural laws.10

The principles of the understanding establish the most general order of nature, but this underdetermines the order that arises over the range of 250


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement empirical laws. Purely intelligible order arising principally – though, as I have argued, not exclusively – from the understanding is not sufficient for establishing order at the empirical level. The strictly mechanical order of the first Critique must be supplemented by a purposive order that treats empirical nature as a system for our judgement. I remarked in the previous chapter that in the ‘Transcendental Deduction’, Kant sets a limit on the reach of the categories. Special laws are laws that specify the law-like relations between empirical objects. Thus Kant makes clear that although empirical laws fall under the categories, which supply the general structure of experience, the detail of empirical experience is not derived from its formal structure.11 Whereas in the first Critique Kant conceded a distinction between transcendental order and empirical laws, he now introduces a further transcendental principle for those empirical laws. While we might momentarily be tempted to think that the introduction of the reflective principle of systematicity finally supplies the empirical correlate to the universal formal order of the categories, we cannot sustain such an opinion. The new principle of systematicity operates formally insofar as it expresses the general organisation of empirical nature and not the particular contents of the latter. Nevertheless, the reflective principle is more attuned to the specificity of empirical nature than were the categories in their initial transcendental form. Kant has added a further layer to his hierarchical reconstruction of the conditions for the possibility of experience. In so doing he has narrowed the gap between formal structure and empirical given. The formal structure now finally supplies an order at the level of the diversity of forms in which the material given arises. The new principle of order is transcendental insofar as it only establishes the applicability of the categories and not their actual application.12 The latter requirement would annul the gap between form and content that Buchdahl rightly establishes as characteristic of transcendental status.13 The system of the laws of the understanding requires the system of reflective judgement in order to complete its epistemic task at the empirical level. Correspondingly, the second system takes the first as its model and ultimately depends on the system of understanding as its ground: ‘Now it is clear that reflective judgement, by its nature, cannot undertake to classify all of nature in terms of its empirical variety unless it presupposes that nature itself makes its transcendental laws specific in terms of some principle.’14 This mutual dependency of the two systems can be simply explained by saying that the second is the specification at the empirical level of the first: 251


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology The principle by which we reflect on given objects of nature is this: that for all natural things concepts can be found that are determined empirically. This means that we can always presuppose nature’s products to have a form that is possible in terms of universal laws which we can cognize.15

But this is to say specification is not automatic and requires a principle in addition to those analysed in the ‘Analytic’ of the first Critique. As Buchdahl argues, the first system gives only the form of the second, but not the latter’s validating grounds.16 If the categorical system is to be applied at the empirical level, we must additionally presuppose that concepts are possible for all empirical objects. Kant expresses this development in his position in a variety of ways. Without the principle of reflective judgement, understanding would find it impossible ‘to discover in nature an order it could grasp’.17 Only by means of this supplementary principle ‘can we make progress in using our understanding in experience and arrive at cognition’.18 Finally, without it ‘understanding could not find its way about in nature’.19 Thus, there are two systematicities and both count as transcendental insofar as they supply the framework conditions of experience.20 However, whereas one is categorial and arises out of the unifying power of understanding, the other is reflective and arises from the power of judgement, which, as we have already seen, can be characterised as mediating. In this case, reflective judgement mediates between the transcendental system of the categories and empirical nature. I have argued elsewhere that despite the variety of descriptions Kant uses to capture the status of the principle of the purposiveness of nature – including heuristic, assumption etc. – it is most helpful to understand the latter as a presupposition in the hermeneutic sense of the term. 21 We cannot make sense of empirical nature unless we presuppose that it displays an order accessible to our minds. This presupposition is a condition for the possibility of experiencing empirical nature, thus it is in no sense optional or psychological as the terms ‘assumption’ and ‘heuristic’ might suggest. The presupposition of the purposiveness of nature is formal insofar as it is the structure within which we take in the empirical world in our intuition of it, but it does not guarantee that on every occasion the world will make sense in terms of our mental structures.22 It simply secures the possibility of that success. In this way, the principle of the systematicity of nature is not simply a projection – as Vaihinger, 252


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement Buchdahl and Pippin suggest – but rather is an anticipation that makes possible an exploration of the world or an orientation within it.23 We presuppose that it is possible to make sense of the complex data of the world, but we do not yet know how we can do so in any particular case. The cognitive encouragement given us by judgement does not encourage the epistemically lazy attitude that the order of nature is already established by our minds, but rather encourages us to take up the task of exploring the world with at least the hope that we can make sense of it. As we will see later, aesthetic experiences are particularly important in fostering this hope. The reflective task of making sense of the world is one that permeates the great majority of our cognitive tasks. It is as relevant to everyday low-level puzzles as it is to the most abstruse theoretical scientific problems. At both extremes, while the answer is not supplied a priori, what is transcendentally assured is that there is some sense in seeking one. The sceptic may still find that too much assurance is assumed by this account. However, we should remember that this is a transcendental argument that starts from our everyday practice in the world. We do in fact operate as if we can make sense of the world. Kant’s transcendental analysis of the principle of the purposiveness of nature seeks to make sense of that practice and to show how it can be legitimated. The deduction of a principle of systematicity aims to legitimate our practice of judgement by revealing a necessary condition for the possibility of our experience. Kant aims to achieve this by displaying how this principle is grounded in the subjective conditions of our cognition.24 In effect, Kant intends to show that the cognitive hope we display in everyday experience can be legitimated, not to disprove the radical doubts of the sceptic in his or her own terms. Kant makes a number of different claims for the principle of reflective judgement. As Allison summarises: In fact, in various places in the Introductions, Kant suggests that the principle of the purposiveness of nature is necessary for the formation of empirical concepts, the classification of ‘natural forms’ into genera and species, the unification of empirical laws into a system (theory construction), the formulation of empirical laws in the first place, and the attribution of necessity to such laws.25

Allison resolves the diversity of functions by saying that, in all cases, reflective judgement is concerned to find universals for given particulars. He goes on to argue that as the search for empirical concepts is inseparable from the search for empirical laws, and the latter entails 253


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology a hierarchical organisation of such laws, then the attempts to establish the possibility of empirical concepts and the systematic organisation of empirical laws ‘are best seen as two poles of a quest for the conditions of the empirical knowledge of nature qua empirical’.26 This is a very helpful synthesis of Kant’s position. Allison’s account leaves open the question raised in the previous chapter in the discussion of Kant’s proviso about special laws in the B Deduction. Is Kant’s position that the singular empirical object can be determined at the categorical level, while only special laws cannot? Now that Kant has introduced a finer tuned account of the formal structure of experience, he could merely be saying that our consideration of the empirical laws of nature requires an additional reflective principle, while the principles are sufficient conditions for the determination of an individual empirical object. In my view, Kant should have said that the principle of the purposiveness of nature is necessary for the possibility of any empirical judgement. The form of empirical judgements is their organisation in a system. For Kant, the necessity characteristic of laws entails unification within a system, should that be at the level of the transcendental unity of apperception or with respect to empirical order. Laws operate within a framework of other laws and empirical concepts must be understood within that framework.27 Thus empirical laws and concepts, the classification of natural forms into species and genera and the establishment of a system of empirical laws all depend on the principle that nature is specifiable as a system that harmonises with the categorial system of the understanding. Were we not capable of presupposing this principle, none of these would be possible. In short, order at the empirical level requires the principle of the purposiveness of nature. And if we were not able to presuppose order in empirical nature, we could not even conceive of making a judgement about singular empirical objects. Many interpreters, including Buchdahl, have interpreted Kant’s account of reflective judgement as referring only to the heuristic system of natural laws, the condition of scientific investigation of nature.28 However, Kant’s claim that the transcendental system can only be applied at the empirical level through the mediation of reflective judgement shows that Kant’s account of empirical systematicity bears a further significance. If knowledge of a given empirical phenomenon is to be possible, the categories of the understanding must be applied at the empirical level. If understanding – the faculty giving rise to the categories – is to find its way about in empirical nature, a 254


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement presupposition of empirical systematicity is required. Kant’s emphasis on law-likeness may seem to suggest that the cognition of particular empirical objects is not under consideration, but, for Kant, cognition is only possible within a systematic framework. The categories of the understanding operate conjointly and give rise to a system of principles. The principles operate as a system in establishing the possibility of an object in general. Correspondingly, an empirical object is only determinable within the context of a system of empirical laws. Knowledge of a particular empirical object arises within a context of lawful relations in which it stands to other objects. The principle establishing the latter is thus a necessary condition of the application of the categories to a given empirical object. The very possibility of empirical synthesis requires that there be an order in empirical nature that can only be secured by the principle of empirical systematicity. Kant says that universal natural laws are sufficient for the coherence of objects in terms of their genus ‘as natural things as such’.29 This encourages the view that transcendental systematicity suffices, at least for particular natural beings. However, he goes on to say that the principles ‘fail to provide them with specific coherence in terms of the particular natural beings they are’.30 What Kant must mean is that while transcendental systematicity establishes the determination of an object in general, it does not yet establish the complete grounds for the determination of a specific empirical object. Kant concludes that an a priori principle of judgement is required in order to find a unity in accordance with law within empirical contingency.31 His point clearly is that were we not able to do so, empirical cognition could not arise.

II A distinction between general purposiveness and systematicity In the previous section, I have argued that the empirical application of the categories is only possible on presupposition of the purposiveness of nature for our judgement. The principle of purposiveness establishes, although strictly reflectively, that we can treat nature as a system at the empirical level. This presupposition is necessary if we are to establish knowledge not only of the lawful connections of empirical nature, but even of the empirical objects that can only be experienced as part of a wider system. Towards the end of his discussion of the two ‘Introductions’, Allison mentions a suggestion by Klaus Düsing, namely, that we 255


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology should distinguish between a more general and a more specific level of the purposiveness of nature.32 Only purposiveness at a general level would be the basis for aesthetic judgement. Allison remarks that Düsing gives no sense of how even a more general sense of purposiveness could ‘license’ particular claims of taste.33 I now intend to suggest a way in which we can identify a general sense of the purposiveness of nature in contrast to a more specific sense of the systematicity of nature’s empirical laws.34 This distinction is one within the formal purposiveness of nature, which Kant has argued is strictly reflective or subjective in status. The significance of this description is not, however, that purposiveness has no role to play in our knowledge of objects. As I have argued, it is necessary for knowledge of objects at the empirical level. The point is that this reflective principle does not determine objects and rather throws us back on the relation in which the subject stands to an appearing thing. The general sense of purposiveness concerns the accessibility of the empirical object to our subjective faculties that cooperate when we make a judgement. The more specific expression of the principle of purposiveness, that there is systematicity across the empirical laws of nature, counts as an instrument or device that facilitates the general project of establishing a fit between mind and empirical nature. Both levels are necessary for the completion of the objective Deduction, which requires that the categories are capable of determining empirical objects. There must be a fit between mind and empirical nature if cognition is to be possible and establishing a hierarchical order of laws is the means to securing this. There are occasions on which Kant’s presentation of the order in empirical nature appears to imply just such a distinction between a more general question of the fit between mind and world and a more specific device for bringing it about. For instance, early in the ‘First Introduction’ we are told that if judgement is to be distinguished from the other faculties by its own concept or rule then this will be one of ‘things of nature insofar as nature conforms to our power of judgement’.35 Kant then goes on to say ‘the only concept we could form of this character is that [nature’s] arrangement conforms to the ability we have to subsume the particular laws, which are given, under more universal laws, even though these are not given’.36 We might say that the principle of nature, insofar as it conforms to our power of judgement, regards nature understood at the level of general purposiveness – that is, in terms of the possibility of empirical cognition as such – 256


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement while our way of grasping this fit requires that we devise a concept of the systematicity of nature in its empirical laws. Meanwhile, in the second Introduction Kant says that the principle of reflective judgement is ‘the basis for the unity of all empirical principles under higher though still empirical principles’.37 This, too, could suggest a distinction between levels at which purposiveness operates, insofar as the power of judgement is the basis for the hierarchy of laws. The purposiveness of nature at its general level takes up the task of the Copernican revolution. If knowledge claims are to be legitimated, it must be established that concepts are capable of applying not only to intuitions in general, but, in particular, to empirical intuitions. This development of Kant’s argument was established in the Principles.38 It now transpires that understanding cannot go out into nature, that is, the concepts are not applicable at the empirical level, unless there is a further reflective principle of judgement. The final legitimation of the validity of the categories, first attempted in the ‘Transcendental Deduction’, is only brought to fruition in the Introductions to the third Critique. Kant most frequently presents the purposiveness of nature for our judgement in its more specific sense as the systematicity of nature in its empirical laws. As Allison says, if there is a distinction between two levels of purposiveness, there is also a continual slippage between them.39 Allison says this makes the task of establishing a link between the purposiveness of nature and aesthetic judgements extremely difficult. However, I have suggested that it is possible to distinguish the general question of purposiveness from its specification, even if the former cannot be fully stated in isolation from the latter. The distinction between the two levels of purposiveness is crucial for making sense of the link Kant draws between the purposiveness of nature and judgements of taste. But a further clarification is necessary if I am to establish this.

III Purposiveness of Judgement and the Dual Direction of Reflective Judgement We need to establish the relation between the purposiveness of nature and the purposiveness of judgement. Kant only rarely mentions the purposiveness of judgement, preferring to talk of the purposiveness of nature. While this sounds like a quite different notion, perhaps even a contrary one, I will argue that the two phrases express the same idea, i.e. that of ‘a formal purposiveness of nature for judgement’.40 257


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology Purposiveness is the relation in which empirical nature stands to our faculties, insofar as they cooperate in judging a given thing. This is the general sense of purposiveness that I established above. Judgement, in both its determinant and reflective modes, is the faculty that facilitates a relation between mind and nature, a task for which it qualifies in that it operates as a mediator between different faculties, primarily between intuition – for which in the third Critique Kant often substitutes imagination – and understanding. It is, to a great extent, the relation in which judgement stands to intuition and its orientation towards the given in experience that qualifies the former as fostering not merely the relation between subjective faculties, but also between mind and world. In that judgement relies on imagination, which is necessarily bound up with the synthesis of apprehension, this, too, qualifies judgement as intentionally directed towards an empirical object.41 Allison rightly insists that the principle of the purposiveness of nature is heautonomous, that is, in exercising it, judgement legislates only to itself and not to nature per se. This principle allows us to make sense of the diversity of nature for our own purposes, while falling short of claiming that nature is so structured in itself. However, despite this stipulation, Allison still proceeds as if the purposiveness of nature were principally to do with nature and not with judgement itself, especially when he tries to prise what he calls ‘logical purposiveness’ apart from judgements of taste. This is justifiable on the grounds that Kant repeatedly refers to this principle as that of the systematicity of nature in its empirical laws. Even if the principle is heautonomous, it still refers to our judgement of nature and does not constitute a mode of self-reflection on the part of the subject, at least not directly, rather making possible our orientation in the empirical world. Nevertheless, there appears to be a tension in the status of the principle of purposiveness insofar as it counts as a self-legislation and yet makes possible our empirical judgement of nature, This tension can be resolved insofar as we recognise the relational status of the principle of the purposiveness of nature. By this, I mean that the latter principle is concerned with the purposive relation between mind and nature, a relation without which there would be no possibility of experience. The principle points in two directions: towards the subject and towards the object. While its intentional direction towards objects is emphasised in the Introductions’ discussions of the systematicity of nature, its subjective direction is entailed by its reflective or subjective status. Reflective judgement in all its 258


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement forms counts as a mediation between the subjective and objective dimensions of experience. Allison believes that the legitimacy of aesthetic judgement can only be shown by decoupling the principle of the purposiveness of nature from the strictly subjective purposiveness characteristic of aesthetic judgement.42 He concludes his initial discussion of the issue saying that what he refers to as logical purposiveness, and I prefer to call the formal purposiveness of empirical nature, is not the principle on which taste is founded. Instead they are both founded on a further principle, that is, the principle of judgement.43 I agree that both forms of judgement are grounded in the activity of judgement, but, as I have already argued in Chapter 5 (pp. 194–6), there is only one pure expression of the latter, namely, in aesthetic judgements. In order to show how the two species of formal reflective judgement are related, I need to bring out the subjective side of the formal purposiveness of empirical nature. The analysis of reflective judgement at its most general level reveals the subjective conditions that are necessary for knowledge of empirical objects. In other words, reflective judgement rests on the subjective cooperation of the faculties necessary for cognition. In the First Introduction, in the section significantly entitled ‘On the Technic of Judgment as the Basis of the Idea of a Technic of Nature’, Kant first comments on how the purposiveness characteristic of reflective judgement in general relates to an object. This arises in one of two ways: we perceive purposiveness in our power of judgement insofar as it merely reflects on a given object, whether it reflects on the object’s empirical intuition so as to bring it to some concept or other (which concept this is being indeterminate), or on the empirical concept itself so as to bring the laws it contains under common principles.44

The first alternative refers to the formal purposiveness of nature, which reflects only on the intuition of an object and thus counts as formal, as we saw above.45 In the second case, reflective judgement reflects on an empirical concept in its systematic or instrumental use in order to facilitate unification of empirical nature in a teleological ordering. In both cases, the faculty of judgement reflects purposively on a given object or on an empirical concept. This establishes the intentional direction of reflective judgement towards an object given in intuition, or at least towards a concept that orders objects teleologically. 259


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology Only now does Kant go on to address the subjective side of the exercise of the power of judgement in response to empirical nature: But when we merely reflect on a perception we are not dealing with a determinate concept, but are dealing only with the general rule for reflecting on a perception for the sake of understanding, as a power of concepts. Clearly, then, in a merely reflective judgment imagination and understanding are considered as they must relate in general in the power of judgment, as compared with how they actually relate in the case of a given perception.46

Kant’s point is that the technic or purposiveness of nature arises out of a technic or purposiveness of judgement. A technic is a procedure by which judgement orients itself with respect to natural objects, which it views as if their possibility rested on art.47 Such a technique is necessary due to the absence of direction from another faculty. In establishing the relation between the two technics, Kant turns his attention away from teleological judgement and addresses ‘merely reflective judgement’, where the power of judgement is exercised in its own right without reliance on understanding or reason.48 The formal purposiveness of nature throws us back on the general condition of judgement, that is, the cooperation of the faculties of imagination and understanding. This is the subjective direction of the purposiveness of nature. Without explanation, Kant moves from a consideration of the general subjective conditions of judgement of the purposiveness of nature directly to a discussion of the harmony of the faculties, characteristic of aesthetic reflective judgements.49 It almost appears as if Kant is suggesting that the immediately preceding discussion of the formal purposiveness of nature referred to aesthetic judgement. But this surely cannot be the case, as that discussion is directed to the purposiveness of nature and not to aesthetic judgements. Moreover, if Kant simply equated these, it would lead him to the implausible conclusion that one of the general conditions of empirical cognition is a harmony of the faculties. We have seen this problem emerge on many occasions. In the next section, I will finally establish the means for resolving it.

IV The link between the purposiveness of nature and aesthetic judgement finally revealed The evidence for Kant’s belief that there is a necessary relation between purposiveness and aesthetic judgements is to be found in 260


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement both editions of the Introduction.50 Admittedly, the claims are less apparent in the main body of the text, although they are not entirely absent.51 The most strident example of his claims come in the First Introduction where he says that the ‘Analytic of Beauty’ will provide ‘an exposition and then the deduction of the concept of a purposiveness of nature’.52 He also says that the proof that aesthetic judgements refer intuitions to an idea of the lawfulness of nature will be in the ‘treatise itself’, that is, in the Critique of Aesthetic Judgement.53 While not quite so dramatically couched, Kant’s commitment to the link between the two species of formal reflective judgement is also to be found in the second Introduction, where he says that natural beauty is ‘the exhibition of the concept of formal (merely subjective) purposiveness’.54 I have argued that formal purposiveness is nothing other than the purposiveness of nature for our judgement. The question I now address is whether Kant’s claims should really be construed as entailing that aesthetic judgements have, as their ground, the principle of the systematicity of nature. Allison concedes there is a connection between the principle of judgement, which is the ground for aesthetic judgement, and the principle of the purposiveness of nature. But he insists that this relation is merely analogical insofar as the deduction of a principle of systematicity is a propadeutic, or establishes the possibility that there is a principle for aesthetic judgement. This analogical connection is based on the heautonomous status of both principles, that is, their entailing a self-legislation on the part of judgement.55 But Allison does not further investigate their systematic connection. While conceding that Kant tries to systematically connect the two principles, Allison concludes that the attempt failed.56 In contrast to Allison, Makkreel is happy to connect aesthetic judgements with the purposiveness of nature, but he insists that both are merely post-cognitive.57 Aesthetic judgements encourage us to think that we could have success in our enquiries into empirical nature and in particular into the ordering of the latter according to genera and species. The latter activity is post-cognitive, Makkreel concludes, because he identifies cognition strictly with the ‘Analytic’ of the first Critique.58 I have argued that cognition fully elaborated should be understood as empirical and, as such, requires the principle of reflective judgement.59 So although the latter is post-categorial, it is not post-cognitive. Moreover, aesthetic judgements reveal both the pre- and post-categorial conditions of cognition. The harmony of the faculties refers us to the initial conditions of cognition and not 261


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology merely to its successful application at the empirical level, as Makreel suggests.60 Aesthetic judgements exhibit the initial conditions of cognition, that is, the relation or synthesising activity of the faculties in a particular empirical application of those conditions. My solution for making sense of Kant’s claims about the link between systematicity and aesthetic judgements is as follows: we can distinguish between two levels of the purposiveness of nature: first, as the fit in general between mind and nature; and second, as the more particular systematicity of empirical nature. But it is only the more general level of purposiveness of nature for judgement that is directly exhibited in an aesthetic judgement. The exemplary role of the latter is only possible because both the purposiveness of nature and the purposiveness of judgement characteristic of aesthetic judgements hold a relational status, mediating between subject and object. A singular aesthetic instance shows that objects in the world are, at least in principle, capable of being taken up by our mental faculties, and that the forms arising from those faculties are applicable at the empirical level. The general fit between mind and world is thus established as possible. With this in view, we can make sense of the passage from the first Introduction considered at the end of the previous section in which Kant puzzlingly seems to equate the formal purposiveness of nature or technic of nature with aesthetic judgements.61 Aesthetic judgements are the only judgements in which the power of judgement is exercised without relying on the understanding or reason. Thus only aesthetic judgements reveal the principle of judgement, which expresses the general level of the purposiveness of nature for our judgement. Teleological judgements rely explicitly on a concept of what an object is, that is, its natural purpose. Even judgements aimed at empirical cognition without any teleological implication, while based on a strictly formal principle, make use of the more specific notion of a hierarchy of laws. Their cognitive orientation masks the activity of the power of judgement that is only revealed in aesthetic judgements. While the philosopher can analyse reflective judgements of nature so as to reveal the cooperation of the faculties on which they are based, only aesthetic judgement gives insight into the synthesising activity of the power of judgement as such at the experiential level.62 We can find the beginnings of a claim that aesthetic judgement counts as exemplary for empirical cognition in the second Introduction to the Critique of Judgement. Kant says that although we once found pleasure in an ability to divide nature into genera and species, we no longer experience such pleasure.63 We tend to conflate 262


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement this ordering capacity with ‘mere cognition’ for the reason that even the most everyday experience would be impossible without such ordering [der gemeinste Erfahrung ohne sie nicht möglich sein würde].64 This must mean that we tend to conflate the empirical ordering of nature, which at one stage in our development led to pleasure, with the transcendental structure of cognition, which does not. Kant suggests that we need something that helps us focus on the purposiveness of nature.65 In the absence of a characteristic pleasure in purposiveness, we lack awareness of the latter’s distinctive status. The title of the next section reinforces this insight, for Kant now turns to what he calls ‘the aesthetic presentation of the purposiveness of nature’.66 While he does not say in so many words that aesthetic presentation counts as exemplary of the reflective principle, it is clear that this is the function it plays. Aesthetic judgements are singular instances of something that is generally invisible to us. The beautiful is, as Kant says in the fourth Moment of the ‘Analytic of Taste’, ‘an example of a universal rule that we are unable to state’.67 In the Introduction, this rule is identified as the principle of the purposiveness of nature for judgement at its formal level. At this level the principle can only be exhibited in an exemplary fashion in a particular case: it cannot be stated in a general proposition. My claim is that taste is exemplary only of the general level of formal purposiveness.68 The possibility of cognition is encapsulated in one aesthetic instance, insofar as this particular object, while not under our epistemic scrutiny, nevertheless reveals itself as open to the subjective cognitive activity necessary for cognition. It thus displays the subjective side of knowledge. But we cannot generalise from this case to conclude that all nature is like this, as would be necessary if beauty were to exhibit systematicity. First, all objects are not beautiful and, second, there are phenomena that defy our cognitive ambitions.69 At best, aesthetic judgement can suggest but not prove that it is as if nature were systematic for our judgement. Thus, aesthetic judgements are not based on the systematicity of nature for our judgement, as the latter is strictly heuristic for the furthering of our cognitive projects. But now we need to look more closely at the stronger claims, namely, that the ‘Analytic of Beauty’ counts as an exposition, a deduction and an exhibition of the purposiveness of nature and that this is proved in the Critique of Aesthetic Judgement.70 Aesthetic judgements have as their ground the principle of judgement, which rests on the subjective conditions of cognition. The harmony such judgements display is a special case of the cooperation of the faculties necessary 263


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology for any cognition.71 Such cooperation is thus the necessary subjective condition of a successful synthesis at the empirical level.72 Aesthetic judgements thus serve as examples of the possibility of the synthesis of an empirical intuition under a concept, although they do so without a determinate concept being in view. Kant says that aesthetic judgements display a schematism not so much of an intuition under a concept, but of the power of imagination under that of understanding.73 The subsumption of one power under another is what allows ‘the understanding to proceed in general from intuition to concepts’.74 Kant shifts register from the schematism of an intuition under a concept, to a schematism of one faculty under another because he is now concerned with the general possibility of synthesis, that is, of the cooperation of the faculties that counts as the subjective side of the Deduction. But aesthetic judgements do not simply display the subjective side of the Deduction. If they refer to the general possibility of synthesis, then they also show how our subjective capacities are capable of taking up a given empirical object. Kant says that in an aesthetic judgement the pleasure arises from the object being commensurate with the cognitive powers.75 This counts as a ‘purposiveness of the object with regard to the subject’s cognitive powers’.76 It would appear that the objective orientation of Sections 13 and 14 of the ‘Analytic of Beauty’ does not amount to a slippage from the subjective orientation of the previous concentration on the harmony of the faculties.77 Aesthetic judgements reveal the relation between subject and object necessary for cognition in general. In the second Introduction Kant reiterates a distinction he made in the first version, where, as we have already seen, he distinguishes between the formal purposiveness of nature and teleological purposiveness. I cite only his revised comment on the former: When an object is given in experience, there are two ways in which we can present purposiveness in it. We can present it on a merely subjective basis: as the harmony of the form of the object (the form that is [manifested] in the apprehension (apprehensio) of the object prior to any concept), with the cognitive powers – i.e., the harmony required in general to unite an intuition with concepts so as to produce a cognition.78

It is clear in this passage, as it was in the first Introduction, that formal purposiveness rests on a relation between the faculties that is necessary for cognition in general.79 As so often before, Kant fails to make an adequate distinction between the latter relation and the specifically 264


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement harmonious subjective relation characteristic of aesthetic judgements. However, unlike the first version, where Kant did not adequately explain the transition to a discussion of aesthetic judgement, he now gives a strong hint as to how the relation between the latter and formal purposiveness should be construed: ‘Hence we may regard natural beauty as the exhibition of the concept of formal (merely subjective) purposiveness, and may regard natural purposes as the exhibition of the concept of a real (objective) purposiveness . . .’80 In the face of the indeterminacy of the purposiveness of nature for our judgement he has already discussed, aesthetic judgements serve as a presentation of the fit between our subjective faculties and a given object.81 They thus reveal the hinge between the subjective and objective sides of the Deduction. Kant speaks of an exhibition, and my clarification is that this must be identified as an exemplary exhibition.82 Aesthetic judgement’s presentation of a harmony between intuition and concept is an exemplary exhibition of formal purposiveness. On several occasions, Kant claims that aesthetic judgements reflect on judgement’s capacity for synthesising intuitions under concepts: ‘For this apprehension of forms by the imagination could never occur if reflective judgement did not compare them, even if unintentionally, at least with its ability [in general] to refer intuitions to concepts.’83 Aesthetic appreciation of objects rests on an implicit awareness of our ability for synthesis. And, crucially, we can see that in aesthetic judgement the synthesising activity takes place at the empirical level because it arises in response to a particular given object. Something counts as beautiful insofar as we experience: only a harmony in reflection, whose a priori conditions are valid universally, between the presentation of the object and the lawfulness in the empirical use in general of the subject’s power of judgement (this lawfulness being the unity between imagination and understanding).84

Aesthetic judgements display not only the cooperation of the faculties characteristic of the transcendental schematism (and which give the formal conditions for all experience), but also the completion of the Copernican turn in a particular empirical application. The initial conditions of cognition are combined with the necessary supplement of the fit between mind and empirical nature as the conditions of cognition in general. This is the form of empirical synthesis. It is thus that Kant, in conclusion, suggests that this indicates ‘a purposiveness of objects in relation to the subject’s reflective power of judgement, in accordance with the concept of nature’.85 265


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology The general fit between mind and world is not merely inferred: it is experienced in a particular phenomenon. In contrast, the systematicity of nature in its empirical laws could only be inferred from a particular instance. Thus, aesthetic judgements, far from being grounded in empirical systematicity, indirectly support the latter by exhibiting the initial conditions of cognition in an empirical application. The singularity of the object under inspection reveals the general possibility of cognition, but it does not do so as an explicit proof or demonstration.86 What we get is a snapshot or intimation of the general purposiveness between mind and world. A singular judgement about this particular instance does not reveal the order of nature in general. But it does show us that this object, at least, is conducive to cognition.87 This encourages us in a cognitive hope that nature in general may also fit with our subjective faculties. Moreover, it reminds us that experience offers us evidence for such a fit. This supports the project of cognition and reveals the educative influence of aesthetics.88 But it is still the case that the purposive fit between mind and world can only be proven, if it can be proven at all, in the course of experience itself. It is often put in question and sometimes we are faced with the contrary insight that there is no harmony between mind and world. This is only to be expected, for the fit between mind and nature is strictly a presupposition, and a priori knowledge is a task, not a mere fait accompli. Having the capacity to introduce formal structure into the material given is only a necessary and not a sufficient condition of experience. A priori cognition requires a material given and in this sense counts as an anticipation, not simply an achievement. It is crucial that aesthetic synthesis remains incomplete or indeterminate and that no concept completes the synthesis.89 Only thus is the very process of synthesis necessary for cognition in general available for reflection and it is this that gives rise to the pleasure characteristic of an aesthetic judgement. The peculiar capacity of aesthetic judgements for presenting the possibility of cognition must be a fragile one and, as such, the presentation counts as a deduction of the principle of purposiveness only insofar as it offers an exemplary exhibition of the possibility of cognition. The specific nature of this exhibition will be explored in the next chapter. It could be objected, however, that while the a priori conditions of cognition are anticipatory and that even though the formal condition of empirical knowledge – that is, the principle of reflective judgement – also so qualifies, it is not the case that empirical knowledge per se 266


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement is anticipatory. This would result in a belated recognition that knowledge at the empirical level, at least, is simply an achievement and not a task. Even then, I would have established that Kant’s formalism is not impositionalist. However, my further point is that any particular claim for empirical knowledge necessarily arises as a synthetic process and that this coming-to-know marks even the conclusion of a cognitive process. While I do not pretend that this is an innocent interpretation of Kant, I do think that his account opens up an understanding of knowledge as a task. Our experience of objects in the world necessarily arises as a network of inter-related claims, the totality of which we cannot grasp. While we can have sufficient certainty about the reliability of individual claims, they are therefore revisable in the light of their wider context. This is not to say that they are merely provisional, but only that we must be able to review them. It is because empirical knowledge arises within an anticipatory or initiatory formal framework – both a priori and empirical – that any particular empirical judgement is marked by its openness to the possibility of revision. This is what I mean by my claim that knowledge is anticipatory at the empirical level. And aesthetic judgement has a role to play in this development in my account. Empirical knowledge makes a claim to a fit between mind and object that would establish its legitimacy. But this claim is an anticipation of a proof that cannot be displayed in the judgement itself; the fit can only be demonstrated in an exemplary fashion by an aesthetic judgement of taste. Aesthetic harmony of the faculties makes almost visible a trace of the general systematicity that is the necessary background to any empirical knowledge claim. Knowledge (Erkenntnis) is, as Nietzsche says, not ‘a bed to rest on, or the way to such a bed’.90 Whether or not I have succeeded in convincing my readers that even empirical knowledge is anticipatory, we can conclude that aesthetic judgement is not grounded in systematicity. If it is grounded in the general level of purposiveness, then it is strictly in the sense that it is a particular exhibition of the possibility of the latter. And the relation is reciprocal, for the ground, i.e. the fit between mind and empirical nature, does not become apparent except in its exhibition in a particular instance. Thus aesthetic judgement is the ratio cognoscendi of the completion of the Copernican turn, while the fit entailed by the latter is the ratio essendi of that exemplary exhibition. But as the ground is exhibited in a singular instance, it can only count as a possibility or as a hypothetical ground that encourages our acting as if there were a purposiveness of nature for our judgement. 267


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology We can now establish the significance of these two sides of judgement for the objective and subjective deductions.91 The principle of reflective judgement allows for the completion of the objective deduction insofar as it allows ‘understanding to go out into [empirical] nature’, but it also deepens our grasp of the subjective side of the deduction as the necessary cooperation of the faculties in the process of synthesis. Indeed, it shows how the two sides require one another. Formal reflective judgement achieves the completion of the objective deduction by making possible the synthesis of understanding and intuition at the empirical level. Our response to a beautiful object, where our understanding and our intuition freely harmonise, exemplifies the general condition of cognition. Additionally, Kant says that judgement supplies a bridge between reason and understanding.92 Despite the overt insistence on a harmony between faculties in Kant’s characterisation of aesthetic reflective judgement, it is easy to lose sight of the importance of his ‘faculty-talk’. We might, as in the narrowly cognitive terms of the first Critique, be inclined to see it as, at best, merely a device for expressing the way in which we make sense of the formative activity of the mind in the face of the natural world. But in the third Critique, Kant’s pervasive rhetorical trope emerges as the vehicle for the examination of the relation between subject and object that is the starting point for transcendental idealism’s account of experience.

Conclusion In this chapter I have shown that there is a way of linking the systematicity of empirical nature to aesthetic judgement via a more general conception of purposiveness that refers us back to the general project of the Copernican revolution. I have argued that grasping the duality within purposiveness, that is, its relational status, reveals how it is necessary for Kant’s epistemological project. General purposiveness is never directly proven, but it is exhibited and as if in a snapshot in an experience of beauty. In an aesthetic judgement we experience the subjective conditions of cognition as the cooperation of the faculties, that is, the subjective side of the Deduction, while at the same time seeing how a particular empirical object could be determined by the categories of the understanding. Yet what we actually experience is not the determination of an intuition under a concept, but the free play between the faculties of intuition (or imagination) and 268


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement understanding, a harmony that is conducive to the possibility of ‘cognition in general’. Thus the subjective and the objective sides of the Deduction are exhibited not so as to give rise to an actual empirical cognition, but rather by showing how empirical cognition in general is possible in one singular and ungeneralisable case. The synthesising activity of the faculties is the key to the way in which reflective judgement relates or mediates both subjectively and between subject and object. If I am right, then the principle of judgement in its full form is the principle of the purposiveness of nature for judgement, understood at its general level as the condition of the possibility of empirical synthesis. This is not as implausible as it may at first appear. Judgement is the mediating capacity that allows for the synthesis of concepts with intuitions. Ultimately Kant seeks to show that concepts apply to empirical intuitions. In the Critique of Pure Reason, judgement operates directly under the jurisdiction of understanding, but this is because the analysis has not yet reached the level of the empirical given and concerns intuition only as the form of its givenness in space and time. We saw this in the previous chapter. In his account of taste, judgement emerges as a faculty in its own right and no longer operates in the interests of the understanding, as it did in the first Critique. Judgement is our ability to make sense of objects in the world through the formal structures our minds initiate: it makes possible a relation between mind and world. An aesthetic judgement provides an epiphany of judgement as a power in action in the world, in particular as our purposive capacity for taking up nature.93 As such, an aesthetic judgement is an exemplary exhibition of the principle of reflective judgement. The principle of judgement is only expressed in the exercise of aesthetic judgement. It is not that there is a deeper principle of judgement that grounds aesthetic judgement, but that only as aesthetic judgement does the faculty of judgement operate autonomously, that is, as a principle. This makes it a curious principle, not one that is an independent foundation for experience, but rather one that depends for its identification on a mode of experience.94 The activity of judgement would not exist as a principle – in contrast to its role as the subjective side of all cognition – were there not aesthetic judgements that displayed it. This principle is constitutive for taste, but strictly exemplary for cognition and experience in general. In the final chapter, I will explore the character of taste’s exemplarity for cognition. 269


29/3/07

8:12 AM



Notes 1. I had already argued this in my article ‘The Technic of Nature: What is Involved in Judging?’; see especially pp. 184, 186–7. This was first presented at a colloquium on the Critique of Judgement at Cérisy-la-Salle in 1993. See, also, Allison KTT, p. 169 and my discussion of the ‘Deduction’ in Chapter 5 (pp. 193–201). 2. CJ, AA 251′. See discussion in final section of this chapter, pp. 260–8. 3. For a discussion of the difficulty of establishing Kant’s view on the relation between empirical systematicity and the determination of empirical knowledge, see my ‘Technic of Nature’, p. 180; for a positive conclusion on the matter, see p. 187. For a perceptive comment on the gap between transcendental principles and their empirical instantiation, see Rodolphe Gasché, The Idea of Form Rethinking Kant’s Aesthetics, p. 39. Although reflective judgement allows for the applicability of the categories, the latter never coincide with their application. 4. CJ, AA 209′, AA 179–80. 5. AA 186 ff; 211′ ff. 6. Kant calls these intrinsic and relative teleological judgements, respectively. For the distinction between these, see AA 378. 7. AA 232′. 8. The phrase is at AA 192, in the title of Section VIII. The confirmation that teleological judgement counts as logical comes at AA 193. This is one reason why I prefer to use ‘formal purposiveness’ to refer to the order that is necessary for any empirical nature whatsoever. Allison, in contrast, principally uses ‘logical purposiveness’ for the order of empirical nature. See Allison, KTT, pp. 6, 32–3, 169, 354–5 note 11. The other reason for my preference is that we will see that both judgements concerning the systematicity of nature and aesthetic judgements display formal purposiveness according to Kant. Retaining this convergence at the level of nomenclature makes it easier to go on to make sense of the substantive link between these two types of reflective judgement. Admittedly Kant’s own usage is not consistent and he also refers to the formal purposiveness of nature as logical. See, for instance, CJ AA 219′. 9. Allison, KTT, pp. 38–9. 10. CJ, AA 180. 11. CPR, B 165; see also A 127/8. 12. At CJ, AA 181/2, Kant says that the principle of the purposiveness of nature is transcendental insofar as it concerns ‘only the pure concept of objects of possible empirical cognition in general and contains nothing empirical’. This is contrasted at AA 181 with a metaphysical principle that establishes the conditions for further determination of the empirically given at the a priori level. The latter is the realm of the Metaphysical Foundations of Natural Science. The principle of pur270


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement

13. 14. 15. 16. 17. 18. 19. 20.

21. 22. 23. 24.

25. 26. 27.

28. 29.

posiveness and the metaphysical principles of MFNS, while occupying distinctive roles in the broader hierarchy of Kant’s philosophical investigation of the a priori structure of cognition, both count as further determinations of the transcendental principles of the ‘Principles’ chapter of CPR. See Chapter 2, pp. 53–4. CJ, AA 215′. CJ, AA 211′/12′. Buchdahl, MPS, pp. 501/2. See Chapter 2, pp. 53–4. CJ, AA 185. AA 186. AA 193. See AA 181/2. At 241′, both aesthetic and teleological reflective judgement count as a priori, although the latter does not count as ‘pure’ for only the former is grounded – we must add ‘directly’ – on the principle of judgement. See 243′. We can conclude that only aesthetic judgement is transcendental, being directly founded on one of the higher cognitive powers. Teleological judgement is only indirectly founded on the same principle, being also dependent on the principle of reason. See my ‘Technic of Nature’, pp. 179–80. CJ, AA 232′. See Chapter 6 for a discussion of the anticipatory status of knowledge. CJ, AA 182. Importantly in this passage while insisting that a transcendental deduction entails the establishment of the basis for such judgements in the a priori sources of cognition, he also insists that we should not try to follow ‘the psychological route’. The distinction is based on the latter being concerned with how we actually judge, while the transcendental deduction is concerned only with how we ‘ought to’ judge, i.e. with the formal or structural norms of judgement. Allison, KTI, pp. 30–1. Guyer also emphasises the range of roles played by the principle. Allison, KTI, p. 31. Philip Kitcher argues that explanations of empirical phenomena necessarily require a process of integration within a unified system. See ‘Projecting the Order of Nature’, p. 213. See also pp. 225 and 231. Kitcher’s reconstruction of the link Kant makes between empirical explanation and unification in a system of laws is surely correct. His suggestion that causal laws ‘imply generalisations that legislate for unactualized possibilities’ (p. 219) is, moreover, congenial to my interpretation of form as an anticipation of empirical instantiation. However, Kitcher’s aim of discarding ‘Kant’s apriorist lapses’ (p. 232) clearly is not. See discussion above in Chapter 2, pp. 59–60. CJ, AA 183. 271


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 30. AA 183. 31. AA 183/4. 32. Allison, KTT, p. 63. Allison treats the formal or logical purposiveness of nature as identical to the systematicity of empirical nature. In this, he agrees with Christel Fricke, see Kants Theorie des reinen Geschmacksurteils, p. 109. However, unlike Allison Fricke sees a way of making sense of the link between aesthetic judgements and systematicity. She argues that the purposiveness without purpose displayed in aesthetic judgements reveals the general conditions of the application of schemata at the empirical level. See p. 115. Düsing gives a similar account of the relation in which purposiveness without purpose stands to empirical synthesis in his Die Teleologie in Kants Weltbegriff, see p. 81. I agree that aesthetic judgements display the conditions of empirical synthesis, while insisting with Düsing against Fricke that it is necessary to identify a general level of purposiveness. Meanwhile in contrast to Düsing, it is my view that purposiveness is principally an aesthetic, rather than a teleological, concept. 33. Ibid., p. 63. 34. For a more extended version of this argument, see Hughes 2006a. 35. CJ, AA 202′. 36. AA 202. 37. AA 180. 38. See Chapter 6, pp. 221–9. 39. Allison, KTT, p. 63. 40. CJ, AA 232/3′. 41. See Chapter 4 (pp. 124–6), for an account of the relation between intuition and imagination that explains why the latter can stand for the former. 42. See Allison, KTT, p. 365, referring to Christel Fricke, Kants Theorie des reinen Geschmacksurteils, pp. 109–11. Fricke holds that the purposiveness discussed in the third moment of the Analytic is the same notion of systematic organisation introduced in the Introductions. Fricke concludes that Kant’s position, as it stands, is incoherent. Allison insists that the two discussions do not deal with the same issue. 43. Allison, KTT, p. 64. See Chapter 5 (p. 198) where I cited this passage. 44. CJ, AA 220′. 45. AA 232′. 46. AA 220′. 47. Howard Caygill, Art of Judgement, p. 316. See also Howard Caygill, A Kant Dictionary, pp. 387–8. 48. Though, as Kant says, for the sake of understanding, which (as we have seen) could not operate at the empirical level without the power of judgement. See pp. 249–55. 49. AA 221′. 272


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement 50. See my ‘Technic of Nature’ p. 184, where I claim that the two forms of reflective judgement are different modes of access to the same principle of judgement. This is close to Allison’s position (see Allison, KTT, p. 64), although he would disagree with my further claim that the purposiveness of nature reveals the subjective conditions of cognition. 51. See my discussion of the ‘Dialectic of Taste’ in Chapter 8 (pp. 299–302). At AA 246, Kant states: ‘Independent natural beauty reveals to us a technic of nature that allows us to present nature as a system in terms of laws whose principle we do not find anywhere in our understanding: the principle of a purposiveness directed to our use of judgment as regards appearances.’ This strongly supports those readings, such as my own and Fricke’s, that insist on the continuity between discussions of purposiveness in the Introductions and the main body of the text. 52. CJ, AA 251′. Kant’s claim that the Analytic will provide a deduction of taste could be seen as evidence for the position that Section 21 counts as a first attempt at a deduction pace Allison. See Chapter 5 (pp. 189–94). 53. AA 247′. 54. AA 193. 55. See above, p. 258. 56. For a more positive account of Kant’s attempt to connect aesthetic judgement and the purposiveness of nature, see Douglas Burnham, Kant’s Philosophies of Judgement, p. 161. Referring to CJ, AA 300, Burnham brings out how beauty counts as a trace (Spur) or hint (Wink) of purposive action. See also Burnham, p. 164, where he says that beauty ‘is not merely the ‘image’ but the symbol of the empirical system of laws, of the immanent lawfulness of nature’. 57. See Ameriks’ helpful review of Imagination and Interpretation in Kant. 58. Moreover, Makkreel reads cognition as imposing form on matter. See discussion in Chapter 4, p. 157. 59. See Chapter 6 and Chapter 7, pp. 249–55. 60. See discussion in Chapter 4, p. 159. 61. CJ, AA 220–1′. 62. See Chapter 4 (pp. 151–6) on synthetic process and my discussion of the significance of the introduction of the power of judgement in Chapter 5 (pp. 194–201). 63. CJ, AA 187. 64. AA 187. Kant’s phrasing here is rather confusing, compounded by his referring to three feminine nouns, Faßlichkeit, Einheit and Lust. However, I take it he must be saying that our being able to grasp nature and its unity in its division into genera and species, not the pleasure once associated with this ability, is necessary for experience. This interpretation is supported not only by the fact that he could hardly claim that something that is no longer the case is necessary, but also that he goes 273


29/3/07

8:12 AM



65.

66.

67.

68.

on to say that what is necessary is easily conflated with mere cognition. As he has already said that the latter does not give rise to pleasure, the evidence seems overwhelming. Pluhar’s translation seems to suggest that it is the pleasure that is necessary. AA 187–8. Kant refers here to the purposiveness of nature for our understanding. He means the purposiveness that is necessary when understanding goes out into empirical nature. CJ, AA 188, in the title of Section VII. At AA 192, he offers another exhibition of the indeterminate purposiveness of nature, that is, logical or teleological presentations. However, those are not presentations of formal purposiveness, while aesthetic judgement is. AA 237. In the first Critique examples were deemed necessary for those lacking in the natural talent of judgement. They encourage our using rules as formulae rather than as principles (A 134, B 173/4). In the third Critique the artistic genius is exemplary for other geniuses (AA 318). This shows that exemplarity has taken on a more positive role. This new development is, however, more indirectly related to the status of taste than the two passages I refer to here. See, nevertheless, my discussion of a possible parallel in Hughes, ‘Taste as Productive Mimesis’. Moreover, examples have a role to play in the development of Kant’s aesthetic argument. See, for instance, the case of Antiparos, which I discuss in the next chapter. Nevertheless, the exemplarity I argue for here concerns the relation between cognitive and aesthetic judgement only. Both Düsing and Fricke also see aesthetic judgements as exemplary of empirical cognition as I mentioned in a note earlier in this chapter, in which I also noted the distinction between their positions and my own. In her article ‘Lawfulness without a Law’, Hannah Ginsborg argues that aesthetic judgements are exemplary of rule-following. See pp. 59 ff. As for Düsing and Fricke, aesthetic reflective judgements reveal ‘a condition of the very possibility of empirical concepts’ (see p. 66). For Ginsborg, however, this is because they are exemplary of how our perceptual synthesis ought to be (ibid.), Generally, I think her approach is much more promising than those readings that focus on the subjectivity of aesthetic judgement and seek to distinguish the latter from cognition. However, I am not sure how helpful it is to say that aesthetic judgements reveal how perceptual judgements should respond to the object (see pp. 70 and 73). Ginsborg’s account suggests that perception in an epistemic judgement should aspire to the quality of perception characteristic of an aesthetic judgement. But, surely, this cannot be the case for the two types of judgement are distinct from one another. I prefer to say that aesthetic judgements are exemplary of the possibility of cognitive judgements, insofar as they display the subjective conditions of knowledge and the relation in which they stand to the object. 274


29/3/07

8:12 AM


Empirical Systematicity and Aesthetic Judgement 69. On the first issue, see Chapter 8 (pp. 284–90). I intend to argue elsewhere that the sublime is a symbol of the limitation on our will to know. 70. CJ, AA 251′, 193 and 247′, respectively. See above, p. 261. 71. See Chapter 4 (pp. 152–6) where I first suggested this. 72. See pp. 255–7. 73. AA 287. See Chapter 5, p. 195. 74. AA 287. See Chapter 4 (pp. 151–6) on synthetic process. 75. AA 189–90. 76. AA 190. 77. Guyer, ‘Formalism and the Theory of Expression in Kant’s Aesthetics’. I discuss this problem in Chapter 8 (pp. 280–4). 78. AA 192. 79. Compare AA 220′. See discussion on pp. 259–60. 80. AA 193. I have removed Pluhar’s emphasis, which is not in the original text. 81. AA 187–8. 82. In general, exhibition (Darstellung) consists in placing beside a concept an intuition corresponding to it, see AA 192. See Pluhar’s note to AA 233. In this case the concept does not subsume the intuition under it. It is for this reason I speak of an exemplary exhibition. 83. AA 190. See also AA 249′. 84. AA 190. See also AA 192. 85. AA 192. He says, specifically, that judgements of the sublime do not display this characteristic of judgements of taste. 86. See Chapter 8, pp. 284 and 295. 87. This reveals that aesthetic experience is educative in the sense that it encourages our cognitive exploration of the world. But it is not didactic, because the insight and encouragement it gives is only ever exemplary. We are encouraged to pursue ‘cognition in general’, but we are given no rules for that task. 88. Although I am mostly here concerned with the pre-cognitive status of aesthetic judgement, this is not incompatible with the position that aesthetic judgement also plays a post-cognitive role in encouraging the development of our knowledge. One way in which Kant suggests this is in saying that an aesthetic judgement involves the ‘quickening’ of our cognitive powers; see AA 222. Makkreel’s account is extremely helpful in this regard. See discussion in Chapter 4, pp. 158–60. 89. See Chapter 4 (pp. 151–6) on synthesis in process. See Chapter 8 (p. 283) on indeterminacy of aesthetic form. 90. Nietzsche, The Gay Science, Book 4, Aph. 324. 91. See extensive discussion of this in Chapter 5. 92. The bridge between understanding and reason would have to be strictly propadeutic. That is, the harmony between intuition and understanding in aesthetic judgement can serve as a symbol for the possibility of the 275


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology application of moral purposes within the empirical world. However, in no sense can beauty serve as a guarantee for moral reason’s success at the empirical level. Just as we have cognitive hope, so can we also have moral hope. The latter, however, is more fragile even than the former. 93. The epiphany is, of course, perceptual and not divine. 94. Although this characteristic is striking, it is not idiosyncratic. All transcendental principles ultimately rest on the possibility of experience; the principle of judgement makes explicit what would otherwise remain implicit.

276


29/3/07

8:12 AM


8


While the argument of this book is restricted to showing the way in which cognitive judgement is dependent on a process of cooperation that is only explicitly expressed in aesthetic judgement, I believe the picture of aesthetics that has emerged has a much wider significance. Reason in general is a plural not a unitary process, for we need to think with and against our own thought, in addition to recognising the limitations of thinking. Any logic that views itself as selfcontained, as not open to another position that could introduce a new and necessary perspective, risks falling into dogmatism. Thinking that is aware of its own relational status is open to what stands outside itself. The relation in which aesthetic judgement stands to an aesthetic object is exemplary for the openness of thinking that is required in cognitive, moral and political thinking. In this chapter I look ahead to this wider perspective, while drawing to a conclusion the interpretation of Kant that I have developed in previous chapters. I argued in the previous chapter that aesthetic judgement provides an exemplary exhibition of a general purposiveness between our subjective capacities and the empirical world of objects. This aesthetic exhibition establishes the possibility of synthesis of an empirical concept with a given empirical intuition, necessary for any cognition whatsoever. In conclusion I now need to delineate more closely how exactly aesthetic judgement is exemplary of cognitive synthesis without counting as a species of the latter. How is it that a mode of attention to the world that is not aimed at achieving knowledge, can, nevertheless, illuminate the process of cognition? Finding an answer to this question is important for both technical and non-technical reasons. My reading of the relation between the roles played by judgement in the first and third Critiques requires an account of the specific identity of aesthetic apprehension and the general significance it bears for cognition. However, I also want to establish that Kant’s account of aesthetic judgement reveals an 277


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology existential insight into the deeper structure of judgement in general and allows us to reflect on the way in which our experience depends on a combination of a plurality of different orientations. This is what I have identified in earlier chapters as the subjective side of the deduction, which we can now see is necessary for establishing the applicability of the categories within empirical experience. Focusing on cognitive judgement – or indeed on morality – alone would risk encouraging a view of human beings as oriented toward one overarching goal. I hope I have shown sufficiently by now that Kant’s aesthetics presents human judging agents as necessarily coordinating a plurality of orientations. Judgement is the ability to connect, not only with respect to a particular instance, but also, more fundamentally, at the level of the thinking process. If we were not able to coordinate sensory receptivity with a reflective capacity for unifying and identifying what we take in through our senses, we could not experience anything at all. Neither could we think about anything, for our capacity for thought always stands in relation to receptivity. Aesthetic judgement reveals the human condition of finding ourselves in a world we make sense of through a combination of orientations. The perspective I have presented in earlier chapters prepares for a positive characterisation of aesthetic apprehension as directed to the appearing of the appearance. All objects are appearances for Kant, but only some exhibit the way in which an object can appear for human beings who are not merely sensory but also reflective. Aesthetic phenomena reveal the relation in which the subject stands to the object and vice versa. My relational interpretation of aesthetic judgement allows me to reconstrue what might otherwise seem like a slide from a subjective to an objective determination of form in the third moment of the ‘Analytic of Beauty’. In my account of the appearing of the appearance, I implicitly make use of phenomenological insights from authors including Husserl and Merleau-Ponty. Having suggested how we can understand the role of aesthetic apprehension within experience, I turn to address a technical problem to which my account seems to give rise. If aesthetic objects are characterised by their appearing for us and all perceptual objects are appearances, then how can aesthetic form be distinguished from perceptual form in general? I address this by arguing that while aesthetic judgement begins with the spatio-temporal presentation of an object, only some objects qualify as aesthetic. Unlike objects in general, the perceptual form of aesthetic things throws us back on the subjective 278


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition activity that is necessary for any judgement whatsoever. Thus I need not arrive at the conclusion often considered the unavoidable corollary of Kant’s formalism, namely, that all objects are beautiful. I make particular use of Kant’s example of the grotto of Antiparos to explore the particularity of a merely mechanical form that is also aesthetic. The connection I draw between perceptual form, in general, and the appearing of that appearance now requires me to address another issue. It is usually accepted that Kant restricts his aesthetics to considerations of the spatial and temporal features of aesthetic phenomena, to the exclusion of their qualitative features. I suggest that although this conclusion need not necessarily be drawn from his account of aesthetic apprehension, he is nonetheless committed to what I call the ‘primacy of presentation’. One strength of the way I interpret aesthetic apprehension as entailing an implicit reflection on the appearing of the appearance, is in highlighting the anticipatory nature of the relation in which the subject stands to the object in cognition in general. Aesthetic judgement makes us aware at one and the same time that judgement requires not only a combination of different mental orientations, but is also dependent on something beyond our minds. In the aesthetic judgement there is a heightened insight into the anticipatory structure of a priori knowledge, due to the specific character of aesthetic judgement as singular. The result is that there are no rules for beauty and we cannot know in advance of experience what will or will not give rise to aesthetic pleasure. We might still suspect, however, that aesthetic apprehension is a species of cognition. I argue against this, that aesthetic judgement displays what I call contrapuntal exemplarity. Aesthetic appreciation is exemplary by contrast with cognition. Throughout my account I have emphasised the way in which aesthetic judgement requires a combination of different mental orientations. But what if aesthetic judgement itself were dependent on a higher order of reason, on which thinking ultimately depends? If the higher order of reason did not display plurality and, instead, displayed an orientation to a unitary end – for instance, a moral end – the pivotal role I have assigned to aesthetics would be undermined. In this case, the pluralist activity that I have argued is necessary for judging in general would turn out to be merely subsidiary to a thinking oriented to a unitary end. Now, while I do not deny that judging strives for unification and that this is a condition of making sense of the world, I do resist the idea that unity is determining for judging. This is the significance of Kant’s distinction between reflective and 279


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology determining judgement for thinking in general. We seek out unity and sense, but we do so out of a genuine engagement with a multiplicity of orientations to the world. Unity is an ideal goal or orientation, not something we can impose upon the world. A textual development towards the end of the Critique of Aesthetic Judgement requires I address this problem. In the ‘Dialectic of Taste’, Kant seems to suggest that aesthetic judgement is ultimately based on morality. I show, however, that for Kant the plurality of mental activity is conserved even at the supersensible level and that morality serves only as an ideal, not a determining, end for aesthetic judgement. Finally, I return to the question that has haunted formalism from the outset. How can the subjective faculties make possible access to an objective given? Drawing on the account of formalism I have developed in earlier chapters, I conclude that aesthetic judgement reveals formation as synthesis in process, taking up the material given and revealing the event of affection as an event. My interpretation thus finally makes sense of Kant’s claim that his method is one of formal and not material idealism. Formal idealism is characterised by a dynamic relation between mind and things in the world, their ultimate context.

I The appearing of appearance is aesthetic form In the third moment of the ‘Analytic of Beauty’ Kant claims that when we judge something to be beautiful, we are concerned only with its form. ‘A pure judgment of taste is one that is not influenced by charm or emotion (though these may be connected with a liking for the beautiful), and whose determining basis is therefore merely the purposiveness of the form.’1 Many have taken this to raise a problem for his account insofar as he seems to have shifted from a position where aesthetic judgements are based on the form of purposiveness in a judgement to one where they are based on the purposive form of the object.2 In the first case beauty arises solely from the harmonious activity of our faculties, whereas in the second case its source lies in something given to us in experience. Such a switch in position would fly in the face of many of Kant’s most fundamental commitments, especially the transfer of impetus from object to subject commonly held to be expressed by the Copernican revolution. It would also render incomprehensible Kant’s regular characterisation of aesthetic judgement as ‘subjective’ and, more precisely, as based on a subjective principle. 280


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition It is hardly surprising in the face of these and other considerations that the great majority of Kant’s readers have concluded that the development of his argument in Sections 13 and 14 must be taken with a large pinch of salt. Allison, for one, concludes that the aesthetic form of the object is not equivalent to its spatio-temporal form.3 Nevertheless we have already seen that the emphasis Kant puts on the object in the third Moment is not an isolated instance, for in the official deduction Kant says that in a pure judgement of taste ‘our liking for the object is connected with our mere judging of the form of the object’.4 In the previous chapter, I argued that aesthetic judgement is capable of displaying the possibility of empirical synthesis and thus of showing that cognition is, at least in principle, possible. For Kant, objectively valid cognition is knowledge of objects in space and time.5 If aesthetic judgement is capable of exhibiting the possibility of knowledge, then that exhibition must itself be in space and time. It is for this reason that design or spatio-temporal form is inextricably caught up with aesthetic form. Now, how can this be so, without entailing the problems I have just raised? My approach starts from an insistence on the relational status of aesthetic judgement, by which I mean that the latter reveals how a subject has access to an external world of objects. As we have seen in previous chapters, our capacity for standing in relation to something given in experience rests on sensibility, our capacity for being affected. However, if we are to become conscious of a given affect, the imagination is also crucial as it must hold together the manifold in such a way that it appears to us as a figure. The account of the epistemic role played by the imagination is most developed in the ‘Schematism’. If the imagination did not mediate between sensibility and understanding, the latter would not be able to subsume an intuition under a concept. The mediating role of imagination is crucial for the pluralist model of mind at work in Kant’s epistemology and in his system in general. The imagination allows a plurality of faculties to remain distinctive orientations to the world, while standing in ongoing relations with one another. In aesthetic judgement the mediating role of the imagination and the figure to which it gives rise are held up for phenomenological inspection, that is, for reflection within experience. This is in contrast to the philosophical analysis of the same phenomena presented in the ‘Transcendental Deduction’ and ‘Schematism’. In the aesthetic case, imagination does not deliver the schematised intuition for the purposes of cognition. It stops short, 281


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology while still standing in relation to the formal conditions of cognition in general, that is, the mutual relation of the faculties that counts as the subjective side of the deduction. The imagination generates a figure that holds together the manifold in intuition as if it were about to be unified under a concept and yet the subsumption does not occur. Aesthetic apprehension displays a free lawfulness that prepares for and yet does not conform to a rule of the understanding. The relational or mediating role of imagination is thus constitutive for aesthetic judgement, while in the epistemic case the third faculty operates under the hegemony of understanding.6 The relational status of aesthetic judgement is the key for understanding the mutual dependency of the form of purposiveness and the purposiveness of form. In an aesthetic judgement the subject becomes reflectively aware of the relation in which he or she stands to an object, which would under other circumstances be an object of knowledge. I am not, of course, arguing that there is something about the object in-itself that causes an aesthetic response, but rather that the aesthetic object encountered in experience prompts a heightened relation between subject and object and thereby reveals the intentional structure of judgement. The object has a role to play, but only insofar as it is for us or stands in relation to us as an appearance. For Kant all objects are appearances, but aesthetic objects appear in such a way as to highlight their appearing. The purposiveness of form is not an alternative to the form of purposiveness, from Kant’s perspective. Some objects prompt a free play of our faculties insofar as their appearance resists cognitive conclusion. Due to this resistance we remain preoccupied with such things at the level of their appearance. We can only sustain an attention to their appearance insofar as our cognitive powers remain in a free or open relation. We have to be capable of a high degree of mental mobility and an ability to tolerate a degree of inconclusiveness if we are to focus on the mere appearance of the object. Certain objects facilitate this freedom of mind allowing us to focus on the appearing of the appearance, rather than on the latter as a mode of access to an object that can be known, valued or acquired. Such contemplation reveals the general relation between subject and object, that is, the form of cognition in general. This is to say that aesthetic judgement is subjective in the strict sense that it is based on the activity of the cognitive powers of the subject, exercised in abstraction from any ulterior purpose. The aesthetic subject, however, is not detached from the aesthetic phenomenon and only from the object determined for cognition. In 282


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition earlier discussions we have seen that the term ‘object’ poses problems for Kant. Strictly speaking, an object is the result of the synthesis of an intuition under a concept. But he also uses the term in a more general sense to express something given to the mind in experience. We saw that in the ‘Transcendental Aesthetic’ he talks of ‘[t]he effect of an object [Gegenstand] upon the faculty of representation’, even when he is referring to mere sensation.7 He also says that it is possible to experience intuitions without conceptual determination.8 This suggests that we can be affected by phenomena that do not strictly count as objects, that is, that are not fully determined. We can now see the importance of a distinction I have made on several occasions. In the aesthetic sphere Kant must leave room for phenomena that are not, or at least not presently, considered as cognitive objects. An aesthetic judgement is characterised by its reflective status. We are aware of the appearance as appearance and not as the appearance of a known object, but we still stand in relation to something given to our minds. The object in the aesthetic case is the undetermined or underdetermined referent of an indeterminate or underdetermining judgement, which arises as synthesis in process. This would not be possible if all intuitions were determined rather than merely determinable by the categories. Certain objects present themselves in such a way that we respond with a free play of the faculties. But these objects could not have an aesthetic affect on us, were we not capable of exercising such a free form of judgement.9 While it is true that the object prompts us, it is also true that the imagination transfigures the empirical givenness of the object so that appearance as appearance can be apprehended. The relation between subject and object is reciprocal in the aesthetic case. This is what it means for there to be a harmony between an object and our judging power. Appearances in general are appearing objects and we usually treat them as such. Insofar as we do so, representations are our means for experiencing objects that are represented. But in the aesthetic case we focus on the appearance and are not currently concerned with the existence of what appears. This contrast is not intended to suggest that what appears stands behind the appearance, but rather that the appearance is our mode of access to the appearing thing. When we look at the appearance for itself we no longer treat the phenomenon as a point of access to something else: we reflect on the point of access or relation in which we stand to the thing. Thus an aesthetic appearance reveals the relation in which representations stand to objects: we 283


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology experience objects in – not through – our representations of them. We do not have representations, in contrast to experiencing objects; representations are nothing other than the subjective conditions of our cognition of objects. Our focus in the aesthetic case on the appearing of the phenomenon for us makes possible a reflection on the process of mental activity as it takes up something given in experience. Something is received by intuition, held together by imagination and, potentially, unified by understanding. Aesthetic form thus displays the characteristics of the first and second elements of cognitive synthesis (as outlined in the A edition ‘Deduction’) and the preparatory relation in which they stand to the third and properly cognitive element.10 This process that is the precondition of any experience, but usually invisible, is now rendered visible. Aesthetic judgement thus makes the affective condition of experience available for reflection.11 The appearing of the appearance that is revealed in aesthetic judgement is not mere semblance (Schein), but rather the relational status of the object.12 An aesthetically pleasing thing brings to the fore the ontological status of appearances in general, as objects that are available for our cognitive faculties. The aesthetic phenomenon affects us and we feel a pleasure in it which is focused on the presentation of the thing. We are aware of how an object affects us, but only at the level of representation per se. Thus the possibility of representation, not the functional effectiveness of any particular representation, becomes available as a topic in the aesthetic case. Aesthetic judgement thus reveals phenomenologically what I have previously argued philosophically.13 Our powers of representation, the faculties, do not constitute a screen between ourselves and the world: they are our modes of discovering the world. And the world faces us as a task and not as revelation. The aesthetic judgement presents us with a perceptual epiphany or sudden manifestation of the meaningfulness of that task in that one particular thing stands out as being in harmony with our subjective faculties. This is the purposiveness of form of the aesthetic object, which is identified only insofar as our response to it counts as a harmony of the faculties, the form of purposiveness.

II Not all perceptual forms are aesthetic forms If an object prompts our power of judgement so that it is exercised purely as an aesthetic appreciation, then how are we to distinguish 284


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition beautiful objects from any other empirical object? If it is the perceptual form of the object that is in question, why do not all empirical objects give rise to aesthetic judgements, as they all have perceptual forms?14 In this section I examine the identity of aesthetic form and discuss how it can be distinguished from other perceptual forms.15 Breaking the link between perception and aesthetics would lead to two connected problems regarding judgement’s place in the critical system and consequently for its status as a distinctive form of judgement. These technical problems would also put in question the validity of the phenomenological description of aesthetic apprehension that I have developed through an interpretation of Kant. The transcendental status of aesthetic judgement and the reason it merits a Critique devoted principally to it, arises from its place within the system of cognitive powers. This place is earned by the status of judgement as contributing to the general project of establishing the possibility of a priori synthetic knowledge.16 It is thus that it contributes to the project of ‘cognition in general’. It is difficult to see how it could do so unless aesthetic judgement stands in some relation to the object and our capacity for synthesis of objects at the empirical level. While the first problem regards judgement’s systematic status within the range of cognitive faculties, the second concerns the specific relation in which judgement stands to reason and morality. While this topic cannot be addressed in any detail here, it is important to position my account of the relation in which aesthetic judgement stands to cognition within a wider perspective of reason in general.17 Judgement, considered as a power in its own right, secures the status of the third Critique insofar as it acts as mediator between cognition and morality. Aesthetic judgement in particular reveals that moral purpose is, at least in principle, applicable within the domain of nature. Judgements of beauty achieve this by giving a particular, though indeterminate, presentation of an idea within the phenomenal realm.18 But aesthetic judgements are only capable of doing so, and only tacitly, because we can presuppose a fit between mind and nature, that is, there is a purposiveness of nature for our judgement.19 I have argued that aesthetic judgements present, in an exemplary fashion, the possibility of empirical synthesis. Beauty thus serves as a concrete presentation of the relation within which mind stands to nature and gives hope that even our higher faculty of reason might successfully be applied within the empirical domain. If aesthetic and perceptual form were not intrinsically linked, this would not be possible. 285


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology But what is the form of the object that is at issue in aesthetic judgement? Is it the form of intuition? Allison argues in Kant’s Transcendental Idealism that the form of intuition is distinct from the formal intuition characteristic of geometry, and which is, in fact, a hybrid of intuition and understanding. He contends that the ‘Transcendental Aesthetic’ is principally concerned with the form of the intuited, while conceding that the latter can be traced back to the form of intuiting.20 Thus, following Allison’s account, if the form of the aesthetic object coincided with the form of intuition, it would count as the form of the intuited. If an aesthetic judgement were prompted merely by the indeterminate form in intuition, then it would entail responding to an intuition in isolation from any conceptual activity. This would conflict with Kant’s dualism, that is, his commitment to a plurality of mental activity characteristic of all experience, falling back into an empiricist position with an attendant myth of the given.21 Intuitive form can only be part of the story, not least because of the analytical status of the argument in the ‘Transcendental Aesthetic’. The form of an object arises from the cooperation of intuition with understanding. Consequently, aesthetic awareness of form is more than a mere affect, being a reflection or judgement. Were this not the case it could hardly give insight into the conditions – subjective or otherwise – of cognition, which necessarily entails the combination of an intuition with a concept arising from understanding. While we have seen that aesthetic judgement counts as formal insofar as it is a reflection on the purposiveness in intuition, Kant more carefully says that the reflection in question is on intuition ‘to bring it to some concept or other’.22 The perceptual form at issue in an aesthetic judgement is not to be understood as the form of the intuited taken in isolation from our power of concepts, but rather as the experiential form that arises from the combined operation of the form of intuiting and the form of conceptualising, the understanding. The beautiful object has a perceptual form that like every other perceptual form arises from a cooperation of intuition with understanding.23 However, in this case the cooperation is brought to our attention because the phenomenon resists determination – that is, cognitive conclusion. Insofar as a particular phenomenon invites reflection on the subjective conditions of cognition and yet resists cognitive conclusion, it exhibits the purposiveness of nature for our judgement. We become aware of the formative process by which the mind stands in relation to objects in the world and in which judgement plays a principal role. 286


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition Thus we are aware of the perceptual form of the object as arising from a process involving both intuition and understanding. As we saw in the last chapter, Kant says that aesthetic judgement involves a schematisation of the faculties. What he means is that whereas in determinate judgement a concept subsumes an intuition, in aesthetic judgement we become aware of the subjective necessary conditions of that result. We become aware of this in a singular judgement about a particular object and thus the exhibition of the purposiveness of nature is not direct and can only be exemplary. As we saw in the last section, aesthetic judgement pauses at the point where imagination has prepared for cognition, but has not yet achieved it. The imagination is primary here because only it can sustain the delicate balance that stops short of epistemic resolution. Imagination both links and holds apart intuition and understanding.24 The form of a beautiful object is the correlate to this imaginative activity and displays a level of integrity and a predisposition to cognitive resolution without any knowledge actually arising. The pleasure necessarily associated with the recognition of beauty sustains the freedom of the mind and the quality of apprehension that are the prerequisites of aesthetic attention. The imagination is not only the agent of this balance, but also its subject. In an aesthetic judgement we can become aware of the mediating role of imagination that is the precondition for any experience whatsoever. The object that comes to be judged as beautiful has a form, just as does any other object, as a result of the cooperative activity of our faculties. However, in this particular case, the form of the object is conducive to a play of the faculties. Something about the object gives rise to a particular proportion of the faculties. Kant makes this point in Section 21, where he argues that a cooperation of the faculties occurs when an object induces the imagination and the proportion of the faculties that ensues depends on what he calls a difference between objects.25 Only certain objects give rise to a proportion of the faculties that counts not just as cooperative, but also as harmonious. What we discover at the level of perception in such cases prompts us to attend further not only to the object but also to the formative process that was the condition of its coming to be. Not every object has a perceptual form that is conducive to aesthetic reflection on the subjective conditions of cognition. The process of formation, that is, the relation between mind and object, is displayed in the spatiotemporal form of these objects insofar as the latter prompt our reflection on the cognitive process in general. 287


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology A consideration of Kant’s example of the grotto of Antiparos will help us clarify the relation between perceptual and aesthetic form. Shortly before the end of the Critique of Aesthetic Judgement there is a discussion of free formations of nature or the process through which fluids take on form. When water freezes, it becomes solid through crystallisation. The switch between a fluid form and a solid one is not a gradual transition but rather occurs ‘as it were by a leap’.26 This transition is strictly mechanical, that is, we need only the laws of mechanical nature to describe its occurrence and yet some such transitions give rise to forms that we consider beautiful: Many such mineral crystallizations, e.g., spars, hematite, and aragonite, often result in exceedingly beautiful shapes, such shapes as art might invent; and the halo in the grotto of Antiparos is merely the product of water seeping through layers of gypsum.27

The shape emerging from crystallisation ‘varies in accordance with what difference in kind there is in the matter (Materien), but is exactly the same wherever the matter is the same’.28 This comment echoes his earlier remark on the significance of the difference between objects for the distinctiveness of the proportion characteristic of aesthetic judgement.29 In this passage Kant reinforces his initial point saying that different snow-figures – that is, forms – arise ‘depending on what the particular mixture of air is at the time’.30 He believes this can be accounted for by the theory of ether. Leaving aside his attachment to a scientific theory no longer tenable, this passage is significant for the distinctiveness of aesthetic form. As in the initial passages of the ‘Transcendental Aesthetic’ Kant is committed to a material given, and as in Section 21 of the third Critique that given can be differentiated.31 The shape or form of free formations differs insofar as there are differences in matter. But is this not to say that transcendent matter shapes our conception of the form, a position that would run contrary to the Copernican revolution’s shift of perspective from material given to formal conditions as the condition of experience? Surely, Kant’s transcendental idealism entails that all differences in matter come from the mind as the initiator of form? It would appear, however, that – at least in the case of free formations – form is not merely imposed on the material given but arises in relation to it. I have been arguing throughout this book that mental form is a mode of access to the material given. The idea developed in Chapter 3 is that matter is only ever experienced empirically, yet our experience arises from a condition that is not yet 288


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition determined. This is a relational or dynamic account of the form– matter relation. The specific significance of the relation between form and matter for our purposes here is that it allows us to see how only some perceptual forms qualify as aesthetic. Kant believes that merely mechanical phenomena can be beautiful and that only some, arising from certain mechanical circumstances, will be so. It is contingent which ones are beautiful. Indeed, Kant makes it a defining characteristic of beauty that it arises contingently. Phenomena that display purpose do not qualify as beautiful. To say that beautiful objects are characterised by ‘purposiveness without purpose’ is to say that though they are merely contingent, it is as if they arose as the result of design. Some perceptual forms just are beautiful. This is an ineliminable feature of our experience of the world. The contingency of beauty means that we have to await its event. This event will typically arise as something that stands out from the normal range of perceptual forms. It is difficult to see how beauty could qualify as an exemplary exhibition if it were not differentiated in this way. A direct pleasure arising from the appearance itself and with no additional agenda – or, at least, without one that is determining – is the necessary mark of a perceptual form that qualifies as aesthetic. We cannot tell in advance whether pleasure will arise or not. Once a beautiful object appears, a pleasure necessarily arises in response to it.32 We recognise it as beautiful through this feeling of pleasure. But there is no way of knowing in advance of experience what will give rise to aesthetic pleasure. Only a retrospective and phenomenological analysis reveals the necessity of the pleasure associated with a particular object. Thus, the distinctiveness of aesthetic from other perceptual forms must also await experience. In his discussion of Antiparos, Kant further brings out the contingent distinctiveness of beautiful things, suggesting that it is as if the stalactites had been formed by human intervention, that is, by art, even though we know this is impossible. In fact it is because we know that they came about naturally that their similarity to artworks gives rise to a pleasure. They are so beautiful that we can hardly warrant they arose without some purpose or other. The analogy with art brings out once again that the distinctive pleasure arising from beautiful phenomena is marked by the appearance, but only the appearance, of their being motivated by a purpose. While a number of objects may strike us as not yet determined as to their identity or purpose, only some do so in such a way that, at 289


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology the same time, they appear to be conducive to our cognitive activity in general. These exceptional objects or phenomena encourage rather than thwart reflective activity. And yet this reflective activity does not lead to a cognitive resolution. Moreover it is only in an aesthetic mode of attention that our toleration of the suspension of a solution is accompanied by a pleasure. The hiatus in cognition is the result of the simultaneous holding apart and relating of our capacities for taking in the world in intuition and reflecting on it with our understanding.33 Our feeling of pleasure in the phenomenon encourages our capacity to sustain an open-minded attitude.

III The primacy of presentation An additional problem that arises from the link Kant makes between perceptual and aesthetic form must be faced before returning to a further exploration of the appearing of the appearance. If aesthetic form is a species of perceptual form, then what of other non-formal features of phenomena? Are colour and tone, or indeed the sheer materiality of things, not candidates for aesthetic appreciation? In addressing this important problem, I develop Kant’s account in a way I believe is compatible with the spirit and logic of his systematic account. While I agree with Allison that Kant’s account of aesthetic form should be more inclusive, I do not agree that we have to break the link between perceptual and aesthetic form in order to achieve this.34 My approach would be to insist that any aesthetic phenomenon must be presented in space and time and thus has some perceptual form, although that form may well be indeterminate or ill-defined. Even the most inchoate mass of stuff would have to be presented for our attention and this would entail that it has some spatio-temporal extension in space and time, some affect on our senses and stands in some relation to other things and to our cognitive faculties.35 And while a particular aesthetic response may be principally concerned with qualitative rather than quantitative features, it must still present features such as colour and tone in space and time. This is what I call the ‘primacy of presentation’. Aesthetic phenomena, whatever their focus, draw our attention to the appearing of the appearance and to this extent highlight the formative process through which the object comes to figure as a phenomenon for the subject. Contrary to the conclusion Kant sometimes, although not consistently, draws, quantitatively formal qualities are not exhaustive of the range of aesthetic 290


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition possibilities. Nevertheless, some presentation in space and time is unavoidable. But it may be objected that although presentation in space and time is a necessary condition of aesthetic apprehension, spatio-temporal form is not determining for aesthetic affect. Extension in space and time could be an initiating necessary condition and yet remain external to the specifically aesthetic quality of an apprehension. Although I cannot develop my idea of the primacy of presentation extensively here, I would argue that the aesthetic status of a judgement requires that presentation in space and time become part of the elaboration of the aesthetic phenomenon.36 Space and time are co-opted into the specifically aesthetic quality of the beautiful. The spatial extension of the canvas is no longer merely a material condition of there being a painting: the canvas becomes the spatial field within which the aesthetic phenomenon appears. And I think it is unimaginable that there could be an aesthetic phenomenon without such a co-opting and transfiguration of space and time. Artworks and natural aesthetic phenomena not only can, but must make us experience space and time differently. If aesthetic form is understood as the appearing of the appearance and not as perceptual form in general, design need not be set in opposition to features such as colour. As Merleau-Ponty says in ‘Eye and Mind’: It is simply a matter of freeing the line, of revivifying its constituting power; and we are not faced with a contradiction when we see it reappear and triumph in painters like Klee or Matisse, who more than anyone believed in color. For henceforth, as Klee said, the line no longer imitates the visible; it ‘renders visible’; it is the blueprint of a genesis of things. Perhaps no one before Klee had ‘let a line muse’.37

Space and time at a deeper level are the conditions of the possibility of the appearance of things and in aesthetic phenomena these conditions become part of the topic rather than merely an initial condition. This is the case, even if the topic is now predominately the phenomenon of colour. Paul Klee’s watercolour painting ‘Southern Gardens (Tunisian Gardens)’ invites us to see the way in which line and colour can cooperate so that each renders visible the other.38 This and others painted in the same period present a patchwork of colours expressive of the Mediterranean ambience of North Africa, representing, it would appear, irregular plots of land intermittently planted with palm 291


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology trees.39 The lines, as the limit points of the patches of colour, are not determinate boundaries, but are defined by the patches of colour they demarcate. The vibrant colours rely on the indeterminately drawn lines for the visibility of their appearing to us as patches of colour. Design and colour are mutually enhancing complementaries and not contraries. In this aesthetic example, form and content cooperate in a way that is exemplary of the more general relation between form and matter. The insistence I have placed on the formative process rather than the formed object goes hand-in-hand with the way in which I have emphasised the relation in which form stands to matter. Form is anticipatory of synthesis with matter. Kant’s formal idealism entails that form stands in a necessary reciprocity with material. If Kant had developed his account of formal idealism more explicitly, he would have been in a position to show that attention to aesthetic form need not be seen as excluding concern for the material in aesthetic phenomena. If form stands in a dynamic relation with matter, then aesthetic response is not only to the formal conditions of experience, but also potentially to its material possibilities. Appreciation of a display of colour in nature or in art, for instance, can reveal the relation in which the judging subject stands to a given phenomenon, and the dependence of that relation on the combination of sensory affect and reflective response. Unfortunately for the destiny of Kant’s philosophical project, his account of the reciprocity of form and matter was insufficiently developed. There are a few exceptions, however, and we have seen one in his discussion of the grotto of Antiparos. I have argued that the aesthetic phenomenon reveals the possibility of empirical synthesis in an exemplary fashion insofar as a particular empirical object is peculiarly accessible to our faculties and their activity of synthesis.40 But are there aesthetic phenomena that do not appear within space and time? Any aesthetic phenomenon appears in time. A painting takes time to look at. A sculpture requires that we take several perspectives on a plastic form and this requires time. But do all aesthetic phenomena appear in space? Does a sonata or a prelude occupy a position in space? I believe that it is arguable that they do, insofar as hearing a sound, and especially a musical sound, entails that we are within a certain range of it. And although spatial position is not the primary content of an auditory aesthetic phenomenon, it is not only a necessary condition of it, but is co-opted within the aesthetic acoustic experience. The particular aesthetic effect of a sound makes use of the spatial relation in which we stand to that sound. 292


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition Nevertheless, might there not be aesthetic phenomena that stand in no relation whatsoever to the senses? Examples of hidden or invisible works of art to be found in ancient cultures and also among contemporary artworks, raise an important question. For instance, in the Neolithic passage tomb at Newgrange outside Dublin intricate designs have been placed on the upper side of the ceiling, buried into the surrounding mound.41 Jochen Gerz, a contemporary artist, orchestrated a work where the names of 2,146 Jewish cemeteries in use until 1933 in Germany were inscribed on the underside of the same number of paving stones in front of Saarbrücken castle, the seat of the provincial parliament.42 Although this started as an illegal artistic action, the authorities later sanctioned it and the Schloßplatz was renamed Platz des unsichtbaren Mahnmals (‘The Square of the Invisible Monument’). If the artwork is positioned so as not to be seen, heard or otherwise accessed by the senses, must we not conclude that there is no sensory appearance and thus no spatio-temporal form? Although I believe such examples force us to extend our perspective to the limits of sensory experience, any non-sensible phenomenon will stand in some relation to a possible sensory experience in imagination. Here imagination is the surrogate for the senses. Are not these artistic phenomena, however, instances of the sublime, a topic that has not yet been addressed? I believe that this may, indeed, be the best way of accounting for the aesthetic affect achieved by hiding something visual, or even just of suggesting that something visual is hidden. However, the role of the sublime in Kant’s analysis of aesthetic apprehension can only be understood as a contrastive move, always standing in relation to the beautiful. In the experience of the sublime the logic of the senses breaks down, but the lack of a sensory resolution for a phenomenon or, more strictly, a mental state, can only be felt as a lack. The imagination, failing to find a harmony between our understanding and our senses, turns to reason and our capacity for thinking the infinite. But the sublime does not occupy the rational perspective; it turns towards it through the intermediary power of imagination. Imagination unavoidably straddles the middle ground between the world of the senses and the perspective of pure intellect. Thus the sublime cannot escape a reference to the senses, even though the relation is a negative or contrastive one.43 While these remarks are merely suggestive of how I would address a number of very important issues that arise from the strong link I draw between aesthetic appreciation and perception, I hope that they 293


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology will be helpful in showing how Kant’s aesthetics could be further developed. Kant gives us a rich starting point that allows for the recognition of the centrality of the aesthetic within experience: we can critically build on this in ways that Kant himself could not have imagined.

IV The Anticipation of the Aesthetic Object Aesthetic judgement reveals not only the combination of a plurality of faculties necessary for any cognitive judgement, but also the relation in which the subject stands to an object. The aesthetic thing encourages a judgement that is exemplary of and yet stops short of cognition. Beautiful things give rise to a pleasure that arises contingently and yet is necessarily connected with the phenomenon under consideration. I have argued for this account of aesthetic judgement in the first two sections of this chapter. I will now develop my account to show how aesthetic form reveals the dependence of judgement on the givenness of an object in experience. We have seen that in Kant’s account of sensible affect at the outset of the ‘Transcendental Aesthetic’, it emerged that we are capable of being affected through our power of sensibility, which makes possible receptivity to the given. Only if such an affect has taken place, is there anything to take up in sensibility and determine under a concept.44 When we find something beautiful we dwell on our intuition of it in association with a feeling. Our capacity for understanding has some part to play, as Kant tentatively suggests at the outset of the ‘Analytic of Beauty’.45 However, the concept that is evoked is a merely indeterminate one of beauty or purposiveness without purpose. Intuition is oriented towards the possibility of conceptual determination, but necessarily falls short of an epistemic conclusion. This allows us to become aware of the givenness of the object. It is exactly because of the failure to achieve cognitive determination, that the aesthetic judgement makes us aware of this, while at the same time it is capable of revealing the structure of ‘cognition in general’. The failure or weakness of aesthetic judgement from the cognitive perspective is the necessary condition of its success in acting as a dual reflection on an object and our activity of thought, without resolving either into the other. Intuiting while aiming at a concept that is always absent is a form of anticipation, not of knowledge of the object, but rather of the possibilities presented by the object. We anticipate these possibilities by 294


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition exploring the object in our perception of it, envisaging various possible aspects or developments of the phenomenal presentation. This is the way in which it makes sense to talk about aesthetic apprehension as an anticipation of the object. But not only do we anticipate the aesthetic object once we are faced with it, we can do nothing but anticipate it in advance. The aesthetic event is contingent. We are always in the position of awaiting the aesthetic object and anticipating how it might (next) appear. The contingency of the event of the beautiful – both within and prior to aesthetic apprehension – allows us, while in the aesthetic mode, to glean an awareness of how every experiential judgement is an anticipation of something that must be given to finite beings such as us.46 Once again, the aesthetic offers a heightened example of a general condition of experience. The contingency of the aesthetic event has further important ramifications for the relation between theory and aesthetic judgement. We cannot define in advance which properties the object must have in order that it qualifies as beautiful. Even ruling out that it could be geometric – or regular – simply does not capture the way in which beauty must be discovered by us and cannot be achieved by following a formula.47 We have to await the event of beauty. However, might it not be objected that cognition also awaits the given and we can nevertheless specify the properties that are characteristic of an object? Indeed I have argued that a priori knowledge is best understood as an anticipation of the given. Yet Kant presents the categories of the understanding as the form of an object in general, that is as the necessary descriptors of quantity, quality, relation and modality. The givenness of the object does not, it would seem, exclude the specification of the necessary characteristics of an object. Why should this not also be the case for the aesthetic object? But, as beauty gives rise to a free play of the faculties there is no objective criterion for aesthetic judgement and no rules that prescribe how or what an aesthetic phenomenon must be. This means that the aesthetic phenomenon must be an event our response to which cannot be determined in advance. In contrast, in the epistemic case rules are given in advance. Even here, though, a specific cognition requires that we await experience to see if anything conforms to those rules. As a result of the lack of a determining basis for aesthetic judgement, it would be futile to seek out instances of beauty in the way that we seek knowledge. It is true that we can put ourselves in a position where we are likely to experience beauty, be that by going to a spectacular 295


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology natural environment or to a gallery where we know there to be works of great aesthetic interest. But we cannot guarantee that just because we are in a propitious location that we will experience aesthetic feeling. Indeed the harder we try to control the result, the more likely we are to fail. Beauty cannot be taken up as a goal, in contrast to cognition.48 In the aesthetic case, theory can only be a reflective anticipation or a reconstruction of the given. There is no catalogue of rules or properties of the beautiful. This also means that we cannot determine theoretically which perceptual forms will count as beautiful and which will not. We can be confident that beautiful objects will stand out from the norm and that they will give rise to the mental freedom characteristic of aesthetic judgement. Aesthetic judgement and perceptual form are inextricable from one another, but they are not identical. Only experience will reveal those perceptual forms that qualify as aesthetic forms and those that do not.

V Contrapuntal Exemplarity Aesthetic response makes use of the subjective conditions of cognition in the absence of a cognitive conclusion. Removing the end of cognitive activity transforms the latter into something quite different. So while we might suspect that Kant’s account renders aesthetic judgement merely a component part of cognition, this is not the case.49 Aesthetic judgement facilitates a reflection on the possibility of cognition, but this is only possible because the latter does not complete the cognitive synthesis. The lack of a cognitive end is not to be understood as failure, but rather as a resistance to what would otherwise be natural, namely, to bring a synthetic activity to some conclusion. The aesthetic standpoint interrupts cognition in such a way that not cognition, but its possibility becomes available for inspection. Aesthetic judgement reveals the form of judgement. This is why Kant remarks at the outset of the ‘Analytic of Beauty’ that he has used the ‘Table of Judgements’ to guide his investigation.50 We can draw out his comment concluding that his investigation is not, however, based on the ‘Metaphysical Deduction’, but rather uncovers the conditions of the possibility of the latter, that is, the faculty of judgement per se. It is even less the case that the analysis of beauty could have been based on the ‘Table of Categories’, because aesthetic judgements are not derived from the conditions of determinate judgement. They are, however, derived from the general form of judgement and display this capacity as a pure activity. This form is the basis not only of 296


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition judgements in general, but also of our possible knowledge of objects, as the ‘Transcendental Deduction’ first argued. An aesthetic judgement stands in relation to a phenomenon as appearing and thus is concerned with the presentation of a thing in space and time. The intentional object of a judgement of beauty reveals the relation between subject and object that makes judgements of quantity, quality, relation and modality possible. Recognition of the relational quality of aesthetic apprehension reveals why mere resistance to cognitive determination would not suffice.51 The beautiful invites our cognitive powers and suggests that we may come to a happy resolution of our cognitive quest, yet at the same time, ultimately resists any determinate resolution of it. Only some objects combine this quality of inviting cognitive activity while resisting a final resolution of that synthetic process. Aesthetic and cognitive judgements illuminate one another by contrast, as in a contrapuntal keyboard composition where the left and the right hands develop contrastive independent melodies, creating a complex and reciprocally reinforcing harmony.52 We have seen that reflection is a condition of the possibility of cognition insofar as it is necessary for empirical cognition, and that the harmonious reflection characteristic of aesthetic judgement reveals the mutual relation of the capacities necessary for any cognition whatsoever. This is the reflective process that is the subjective side of synthesis. Aesthetic judgement allows us to become aware of our general capacity for reflection. Now the question is: if cognition is internally structured so as to depend on reflective activity, is it conceivable that we could exercise our cognitive capacity in the absence of a capacity for reflecting on the possibility of cognition? And if there are alternative possible species of reflection on our cognitive power within experience, do they share the structural form of aesthetic judgement?53 I will merely suggest that for finite beings such as ourselves, an absence of the possibility of self-reflection, even at this highly general level, is unimaginable. I am sure that our grasp of the possibility of cognition in general is always indistinct and rarely recognised as such, but, although I cannot argue the case here, we must surely have the capacity for such an uncanny awareness of the possibility of thinking as a process. I will offer a provisional answer to the second question by considering the possibility that the power of judgement could be exercised in a non-aesthetic guise. Kant claims that the only pure exercise of judgement is to be found in aesthetic judgement. But is this right? Surely it is at least possible 297


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology that we become aware of the activity of judgement on other occasions. For instance, in thinking about a philosophical problem, we can become aware, not so much of the problem, but of the process of thought through which we approach that problem. We focus not on the contents of thought, but on the very process of thinking and become aware of the appearing of thinking to itself. We will almost certainly struggle to capture this fleeting glimpse of what appears ineffable. Such moments are probably brief and, as they tend to occur when we are deeply involved with a problem we wish to solve, we most likely move on to the work in hand. But surely what we are aware of in such cases is the play of different mental orientations in thinking. Our reflective powers cooperate freely without being anchored in a goal. And what is striking about such an insight is that our minds do not merely wander from content to content, but remain at the level of the form of thinking as an activity. But does such an insight count as aesthetic? If not, we have discovered at least one pure expression of judgement other than the aesthetic case. I would venture to suggest that such a relation to the thinking process is aesthetic in the restricted sense that we are concerned only with the appearing of thinking to itself. Leaving aside the difficult – perhaps unanswerable – question of the status of such introspection, I would suggest that even if it is impossible to conclude definitively that Kant is right in claiming that aesthetic judgement is the only exercise of pure judgement in and for itself, it is at least recognisable as one such exercise. Aesthetic judgement is characterised by its lack of a determinate end and indeed even by the lack of a desire for such an end. This is not to say that aesthetic judgement must be wholly void of any cognitive content. This is a point that Kant did not sufficiently develop. What is necessary is only that an aesthetic phenomenon cannot be resolved into a cognitive conclusion. Thus, it is quite feasible for us to respond to a natural or artistic phenomenon aesthetically, while knowing certain things about it. I can know which mountains I am looking at, where they are relative to my own position and that they are dangerous to traverse during winter. I do not have to forget this information, although I may abstract from it, in appreciating their beauty. Indeed some information – for instance, that they are dangerous – might contribute in an indirect way to my aesthetic pleasure. All that is necessary is that the cognitive information does not determine my pleasure. This is especially important when we consider artworks that are informed by scientific enquiry – be it a work by Leonardo or a modern installation.54 298


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition In responding aesthetically, our powers of reflection are at play in a dual sense. At the first level we are using our power of thought in response to a given affect. This is the characteristic cooperation necessary for any judgement whatsoever, including any cognitive judgement. At a second level, we reflect on our reflective and affective activity. But this second level is not a meta-level where we stand outside and look on. Aesthetic reflection only arises in that we are engaged with an empirical given object. This is reflection that is necessarily intertwined with experience. Even in the speculative case I just developed, where the pure exercise of judgement arises within the context of a philosophical project, it only arises insofar as we are already engaged with something else.55 Reflection on reflection can only be glimpsed at the limits of our thought. For this reason it is always liminal and fleeting. Aesthetic phenomena command our attention and in so doing allow us to glimpse the usually unfocused on process of attention. Aesthetic phenomena – both natural and artistic – allow us to experience the process of thinking necessary for any experience whatsoever, but only in the midst of our involvement with a particular empirical given.56

VI The supersensible harmony of the faculties I have argued that aesthetic judgement reveals the necessary cooperation of a plurality of capacities as the subjective condition of cognition. I have also argued that, for Kant, this reveals the way in which we have access to extra-mental objects in the world. A further hurdle for my insistence on a pluralist model of mental activity is introduced at the end of the ‘Analytic of Beauty’. In the ‘Dialectic’ of the Critique of Aesthetic Judgement Kant introduces the idea that there is a supersensible harmony of the faculties on which aesthetic harmony is ultimately grounded. As the supersensible is usually associated with our moral capacity, does this mean that the plurality revealed by aesthetic judgement is ultimately grounded in our moral capacity? If this were so, a pluralist model of mind would ultimately resolve into a complex monism. In contrast to the impositionalist model within which understanding is the dominant mental faculty, this alternative model would be one in which practical reason presides.57 This result would correspond with the prima facie plausible interpretation of Kant’s position as entailing the primacy of moral reason in orienting the relation in which we stand to the world. Whereas it may appear that the Dialectic makes aesthetic judgement dependent on the moral supersensible, Kant persistently refers 299


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology to the idea of a purposiveness of nature for judgement as the supersensible specifically related to taste.58 The supersensible is tripartite and consists of three ideas that are ultimately derived from the three higher faculties of understanding, reflective judgement and reason. The three ideas are those of nature in general, the purposiveness of nature for our judgement and our moral freedom.59 Thus when Kant ultimately traces the validity of aesthetic judgement to the supersensible, we need not conclude that he is committed to the view that the aesthetic is derived from the moral, for the aesthetic also has a place in the supersensible. The harmony of the faculties experienced in aesthetic judgements refers back to the cooperation between a plurality of faculties at the deepest level of our rational being. The supersensible structure of reason is not simply moral, but plurally constituted by three cooperating faculties giving rise to distinctive, but compatible principles. Nevertheless the priority given to the moral within the supersensible seems to put the autonomy of aesthetic judgement at risk. The morally good is the intelligible that taste has in view . . . for it is with this intelligible that even our higher cognitive powers harmonize, and without this intelligible contradictions would continually arise from the contrast between the nature of these powers and the claims that taste makes.60

Although the mere relation between the aesthetic and the supersensible does not commit Kant to the view that the former is grounded in morality, it would appear that his account of the internal dynamics of the supersensible finally leads to that conclusion. But what exactly is Kant’s committment here? He says that the higher cognitive powers harmonise with the morally good and only thus are intelligible contradictions avoided. But this is not to say that the aesthetic supersensible is grounded in the moral, but rather that the former aims at the latter as a goal. Only insofar as the plurally constituted mind is capable of aiming at the morally good as an ideal, can we hope our faculties will not lead us towards mutually incompatible goals. My interpretation of the ideal status of morality takes inspiration from Kant’s philosophy of history, where he argues that the only hope for an eventual realisation of our moral goals within the world – and thus for a harmony between morality and nature – is that a moral end is adopted as the ideal goal by which we orient our practice, where the latter is understood in a broader sense than that of the purely practical or moral.61 Taking the ideal status of 300


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition Kant’s historical teleology as a clue, we can conclude that ideally, but only ideally, the faculties in their diversity tend towards a moral ideal. This does not mean that the aesthetic can be derived from the moral, but rather that the moral idea is only approached through a cooperation of different orientations and aesthetic harmony is preparatory for moral harmony. Schiller develops Kant’s position, arguing that the aesthetic is the step that makes the moral possible within experience.62 Aesthetic judgement is a bridge between reason and understanding, in that taste prepares for the realisation of morality within experience. Beauty signals the possibility of a link between reason and nature in particular aesthetic cases. The achievement of a moral world or kingdom of ends would require a generalisation of this result and for Kant this remains strictly an ideal.63 Kant goes on to remark: And because the subject has this possibility within him, while outside [him] there is also the possibility that nature will harmonize with it, judgment finds itself referred to something that is both in the subject himself and outside him, something that is neither nature nor freedom and yet is linked with the basis of freedom, the supersensible, in which the theoretical and the practical power are in an unknown manner combined and joined into a unity.64

It is difficult to give a clear interpretation of the text at this point, but we can at least be confident that the bridging role played by aesthetics depends on the way in which taste makes possible a relation between the freedom within us and the nature that lies outside. Aesthetic judgement gives an exemplary exhibition of the possibility in principle that our rational projects could be in tune with external nature. If aesthetic judgements did not reveal the purposiveness of nature for our judgement, they would be incapable of providing a transition between reason in its narrow, or moral, role and understanding, the faculty that makes possible unification of objects within mechanical nature. I can now reinforce the link made in the last chapter between the purposiveness of nature and aesthetic judgement. Aesthetic judgement is capable of linking the theoretical and practical orientations just because of the possibility that nature will harmonise with taste. This possibility of the purposiveness of nature in general is actualised in the purposiveness of beautiful objects. It is thus that aesthetic judgement is capable of making the link between what is subjective or ‘within’ and what is objective or ‘outside’. 301


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology Taste is the exemplary exhibition not only of the relation between understanding and intuition, but of the system of reason in its broad sense. Aesthetic harmony captures the dynamic cooperation characteristic of all rational activity, even at the supersensible level. The dynamic is now oriented towards an ideal resolution, which can never be achieved within experience.65 For beings such as ourselves, dualism or, as we can now see, more precisely, the plural orientation of our mental makeup cannot be overcome. What counts as knowledge for us starts with something given to us in appearance and only brought to determinate form through a combination of the faculties. Aesthetic judgement itself arises out of a complex combination of faculties and is, at the same time, a specific exemplary exhibition of the general makeup of our minds and the relation in which we stand to the world within which we find ourselves.

VI Form and World Aesthetic judgement is a response to an empirical object and simultaneously a revelation of the conditions of possibility of that response. It is the only point in Kant’s system where the transcendental and the empirical almost coincide, yet they remain distinct for we are not faced with a concrete presentation of the general framework.66 The beautiful gives us a glimpse of the general form of empirical synthesis crystallised in a singular ungeneralisable instance. And although our awareness of the beautiful object need not be merely formal, the resultant insight into cognition remains formal as it is only through the harmony between the form of the object and the form of our mental activity that this epiphany arises. In what follows I will suggest that this formal insight reveals the dynamic relation between form and matter, but this is not to suggest that we have access to matter per se. The aesthetic phenomenon reveals the formal conditions of access to materiality; any other considerations of materiality (for instance, colour or tone) are aspects of our aesthetic appreciation of the particular thing at the phenomenal level. Aesthetic judgement finally shows how our minds are capable of getting at something outside of ourselves in the world by revealing the process of cooperation that makes possible receptivity to the given. Although this cooperation arises from our subjective faculties, it is not a privatively subjective activity. It begins with our capacity to be affected by something outside us through sensibility. However, this affect is nothing for us if it is not combined with reflection. As we have 302


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition seen, in the first Critique Kant sometimes suggests that all intuitions must be subject to the determining power of understanding. I have argued that even in these passages we should interpret Kant’s position as entailing that it must be possible for all intuitions to be taken up by the understanding, but not that they are in fact necessarily subject to determining synthesis. This leaves open the option that we could have intuitions of which we are aware and yet which do not give rise to knowledge. I now intend to show how this is crucial for the relation between form and matter characteristic of formal idealism. At the outset of the ‘Transcendental Aesthetic’ of the Critique of Pure Reason, Kant all too briefly established that our experience begins with an affect on our sensibility. I argued in Chapter 3 that this affect must be understood as arising from an empirical material object. In Chapter 6 I argued that Kant’s account of the legitimation of the categories culminates in the Principles’ claim that they apply to what he calls experience in general and finally, in the Analogies, to the material givenness or existence of objects. While at first Kant is satisfied to show that the categories must be combined with the form of intuition, he concludes the ‘Analytic’ by requiring that they must apply to empirically given things. Most notably in the ‘Refutation of Idealism’, he asserts that these are spatial things. What, if anything, has the Critique of Aesthetic Judgement to add to this hierarchical account of the conditions of knowledge? The aesthetic object is an empirically given object. It is one that is particularly conducive to the activity of our faculties necessary for any cognition whatsoever. As an empirical object, it is a composite of form and matter. In Kant’s account of the grotto at Antiparos it transpires that the material conditions of this beautiful phenomenon play a role in the emergence of its form. This is surely suggestive for the general case of beauty. What is distinctive about a beautiful form in contrast to other perceptual forms is that the beautiful thing invites our attention and a response, that is, our synthetic activity. The combination of material affect and form that is the condition of any experience whatsoever is, in this case, particularly congenial to our cognitive activity. This thing is not only easy on the eye, but also conducive to our reflection: the aesthetic event is distinguished by the one necessarily leading to the other. Kant, unfortunately, focuses disproportionately on the formal qualities of the phenomenon, but his account does not preclude a more dynamic relation between form and matter. Indeed, as I have argued in previous chapters, his formal idealism requires a reciprocal 303


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology relation between form and matter in which both play distinct though necessary roles. The aesthetic phenomenon displays our access to an object that is formed by our mental activity and yet is only constructed out of a material given. This is the ‘appearance itself’ of the second Analogy, which is neither a thing in itself nor a mere representation. But in the aesthetic case, the objectivity of the thing is not the issue. The beautiful thing could under other circumstances be considered as an independently existing object displaying extensive and intensive magnitude and standing in relation to other things. But in this case we are not interested in considerations of objectivity. We suspend the goal of knowledge in a way that is comparable to the phenomenological epoché.67 We are concerned not so much with achieving cognition as with observing something, which indirectly gives us access to the subjective activity that makes cognition possible. And aesthetic judgement does not merely reveal a harmony of the faculties that are the subjective conditions of cognition: it also shows how those subjective faculties access the empirically given object. Our reflection on the beautiful thing is intertwined with a sensory apprehension of the qualities of the thing. Aesthetic reflection only arises in conjunction with aesthetic apprehension. In this way aesthetic judgement not only displays the twin conditions of perception, but also deepens our awareness of how they cooperate with each other in response to a material given. An aesthetic phenomenon not only turns our attention to the form of the appearing thing, but at the same time reveals that form as our mode of access to a material given. The event of intentional consciousness – that is, of form taking up matter – is displayed as an event. I argued in Chapter 3 that the event of affection is only experienced in relation to reflection. When an intuition is determined by a concept, affect arises only as sensation, that is, as an inseparable element of our experience of an empirical object. I have suggested in this case affect should be viewed as result. But if the intuition is not determined by a concept, the affective element becomes available for inspection in distinction not from our capacity for reflection, but from a determining concept. In this case, the material dimension in empirical experience becomes available as the affective moment or the event of affectivity. A merely indeterminate intuition would not achieve this. It is only an indeterminate intuition that in harmonising with the conditions of cognition in general gives rise to a pleasure that is capable of revealing the affective component of empirical experience. It does so by showing how a material given is taken up by our subjective formative capacities. 304


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition Aesthetic judgement finally reveals the structure of experience as understood by formal idealism. In an aesthetic judgement I am concerned with the form of the object. This form arises first as the form of the intuited, an organisation in space and time. We saw in Chapter 3 that intuitive form arises insofar as our sensibility, the form of intuiting, introduces an order that allows us to take up empirical matter. This complex operation counts as ‘affect’. There must be a material given, thus preventing Kant from falling into material idealism. But the form of the object also arises from our capacity for reflection, the understanding. The latter makes the unification of the affect under a concept possible. In aesthetic judgement there is no unification of the phenomenon, but the imagination nevertheless holds together the given in such a way that would under other circumstances prepare for determination, that is, cognition. We thus have a capacity for making sense of things prior to determination under a concept.68 The combination of the affect and a process of unification, falling short of unity, give rise to the form of the object, though not yet a determinate form or formal intuition.69 Aesthetic form is the crystallisation of the process of synthetic activity, balanced between sensory affect and conceptual determination. As such it is neither one nor the other and symbolises their process of cooperation. Aesthetic judgement’s focus on the form of the object throws us back on the synthetic activity, now experienced in a heightened form as a ‘harmony of the faculties’. It is thus that the form of an object we find beautiful is able to uncover the possibility of cognition in general, that is, how it is that we are able to have access to the world of things through the exercise of our formal subjective capacities. Form, or the formal subjective faculties that make forms possible, opens up our access to the empirical material world.70 Form is not imposed on the world: it is the form of the world. But, in this case, the world cannot be an alien collection of things in themselves: the world is a world of things that stand in relation to us. Our environment is this relation.

Notes 1. CJ, AA 223, in Section 13. In Section 14 he develops this idea further. 2. See, for instance: Guyer, KCT, pp. 220–3 and ‘Formalism and Theory of Expression’, pp. 59–61. 3. Allison, KTT, p. 175. 4. CJ, AA 289. See discussion of the ‘Deduction of Taste’ in Chapter 5 (p. 197). 305


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 5. See Chapter 6 (pp. 229–37) on the implicit spatial schematism in the ‘Principles’. 6. Earlier chapters establish each of the component parts of this account. 7. CPR, A 19/20, B 34. 8. A 89–90, B 122–3. 9. Just as they could not have any affect on us at all, had we not a power of sensibility or receptivity. 10. They are preparatory, not only because they are prior in order of presentation, but because they provide the form for an apprehension, which would, if it were determined, count as knowledge. As it does not so qualify, the apprehension counts as indeterminate or underdetermined. 11. See the final section of this chapter (pp. 302–5)for further discussion of how the relationship between form and matter is revealed by aesthetic judgement. 12. This is much closer to what Merleau-Ponty calls the ‘in-itself-for-us’, than are cognitive ‘appearances themselves’ discussed in Chapter 6 (pp. 226–7). See Merleau-Ponty, Phenomenology of Perception, p. 322, French edition, p. 372: un véritable en-soi-pour-nous. 13. See Chapter 3 (pp. 96–105) on the status of representations as our point of access to objects. 14. For a discussion of the problem arising from the link Kant makes between perceptual and aesthetic form, see Paul Guyer, Kant and the Claims of Taste, especially p. 226; Donald Crawford, Kant’s Aesthetic Theory, especially pp. 96–110. See also John H. Zammito, The Genesis of Kant’s Critique of Judgment, pp. 119–21; and Henry Allison, Kant’s Theory of Taste, pp. 132–5. 15. It is perhaps surprising that Allison accepts Guyer’s conclusion that there should be no link between perceptual and aesthetic form, given the distinctiveness of their interpretative approaches. Nevertheless Allison’s agreement with Guyer about perceptual form arises from an independent conviction that aesthetic judgement is grounded in the subjective and not the objective conditions of cognition. See Chapter 5 (pp. 198–9) on Allison’s interpretation of the ‘Deduction of Taste’. 16. See Chapter 4 (pp. 151–60) where I argue that aesthetic judgement qualifies as synthetic. 17. See also the discussion of the relationship between the faculties at the supersensible level on pp. 299–302 below. 18. In the first instance, the idea is aesthetic, but Kant holds that the latter can serve as a symbol of a moral idea. See CJ, AA 351–4. 19. See Chapter 7, pp. 257–68. 20. Allison, KTI, p. 97. See discussion of Allison’s account of the forms of intuition in Chapter 2 (p. 69). Contrast Longuenesse’s reading of the 306


29/3/07

8:12 AM



21. 22. 23.

24. 25. 26. 27.

28. 29. 30. 31. 32.

33.

34. 35.

36.

37.

‘Transcendental Aesthetic’ as already directed at formal intuition also discussed in Chapter 2, p. 73. For a discussion of this issue, see McDowell, Mind and World, for instance, pp. 5–9. See CJ, AA 232′ and AA 220′ respectively. This precision was discussed in the previous chapter. Even an undetermined, not merely an underdetermined, intuition would stand in a potential relationship with our capacity for reflection. This is necessitated by Kant’s dualism. See Chapter 3 (pp. 92–4). Sarah Gibbons makes a similar observation in her Kant’s Theory of Imagination, p. 29. CJ, AA 238. See discussion in Chapter 5 (pp. 181–3). These two claims count as Steps 3 and 4 respectively of my reconstruction. CJ, AA 348. AA 349. In his aesthetic appreciation of mineral forms and crystallisation, Kant anticipated a popular theme in German Romanticism. See, for instance, Novalis’ use of chemical metaphors in his ‘Miscellaneous Writings’, 95, reprinted in Kathleen M. Wheeler (ed.), German aesthetic and literary criticism, p. 92. Later Stendhal uses the idea of crystallisation to explain the formation of feelings, in particular, love. See De l’amour, I, Ch. 1, p. 5. AA 348. AA 238; discussed in Chapter 5 (p. 181). AA 349. See Chapter 3 (pp. 87–9) on the materiality of the given and Chapter 5 (p. 181) on the differentiation among the given. This is the sense in which the liking for an aesthetic object is a necessary one, as Kant argues in the fourth moment of the ‘Analytic of Taste’. See AA 236–7. Ian Hamilton Finlay’s bridge comprised of two identically inscribed planks at ‘Little Sparta’ expresses this hiatus in an aesthetic form. See the discussion in the Introduction (pp. 1–2). Allison, KTT, especially pp. 135–8. This is to say, that it will stand in some relation to all four categories and to their correlates in the Axioms, Anticipations, Analogies and Postulates. A discussion with Michael Podro helped me clarify this issue. In a related but distinct context, see Podro, Depiction, p. 13, where he talks of ‘the recruitment of the paint surface’ as a way of sustaining recognition of the subject of the painting. Merleau-Ponty, ‘Eye and Mind’, The Merleau-Ponty Aesthetics Reader, p. 143. The final phrase comes from Michaux’s Aventures de lignes. The French is ‘laissé rêver une ligne’: L’Oeil et L’Esprit, p. 74. 307


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology 38. Paul Klee, ‘Southern Gardens (Tunisian Gardens)’, 1919, watercolour and pen on paper, 24.4 x 18.7/ 18.9 cm, The Berggruen Klee Collection, Metropolitan Museum of Art, Paul Klee: catalogue raisonné, vol. 3 (1919–22), p. 86. 39. See also ‘Tunisian Gardens’ (1919), ibid., vol. 3, p. 87 and ‘View from St. Germain (Tunis) Looking Inland 1’ (1918), ibid., vol. 2 (1913–18), p. 440. 40. See Chapter 7 (pp. 260–9). 41. See Claire O’Kelly, ‘Passage-grave art in the Boyne Valley’, particularly pp. 363–4. 42. Jochen Gerz, with the participation of students from Hochschule für Bildende Kunst, Saarbrücken ‘2146 Steine – Mahnmahl gegen Rassismus’ (‘2,146 stones, monument against racism’), 1993, Saarbrücken, Germany. See Stewart Martin, ‘A New World Art? Documenting Documenta 11’, particularly p. 12. 43. I will argue in my Afterword that beauty stands in a reciprocally necessary relation to the possibility of the breakdown of harmony and order, expressed most explicitly in experiences of the sublime. 44. These are the general conditions of cognition. See Chapters 3 (p. 94) and 4 (pp. 120–47). 45. CJ, AA 203: ‘ . . . rather, we use imagination (perhaps in connection with understanding) . . .’ 46. I come back to this in the final section, pp. 302–5. 47. I thus derive a rather different conclusion from his position than does Kant himself. See AA 242–3, where he suggests that regularity would not qualify as worthy of aesthetic appreciation. 48. Although setting cognition as a goal does not result in its end being completely within our control. 49. See Avner Baz’s article, ‘Kant’s Principle of Purposiveness and the Missing Point of (Aesthetic) Judgements’, for the view that aesthetic judgement is, for Kant, a variety of cognition. 50. CJ, AA 203. 51. Gasché appears to suggest this in The Idea of Form, p. 16. 52. An example would be the last prelude of the first book of Bach’s WellTempered Klavier. 53. Clearly within philosophy our cognitive power can be analysed. But my question concerns reflection on the possibility of cognition at a phenomenological level. 54. One such contemporary artwork is Richard Wilson’s ‘20:50’, 1987 (Saatchi Gallery, London). The title of this work concerns the specific density of recycled engine oil, which floods a gallery space. The aesthetic affect of this work, which has a dramatic reflective quality, is not diminished by the information the title supplies about the material properties of oil. 308


29/3/07

8:12 AM


Aesthetic Judgement’s Exemplary Exhibition of Cognition 55. It could also arise in the context of scientific, for instance, mathematical reflection or even alongside a practical concern. 56. Kant expresses some hesitation about the purity of an aesthetic judgement in response to an artwork. However the artwork still plays a central role in his account of aesthetic pleasure. 57. In this section I am only concerned with the significance of the relation between the aesthetic play of the faculties and supersensible harmony for my position that aesthetic judgement reveals a plurality of orientations, which is necessary for experience. 58. See in particular CJ, AA 340, 347–8, 350, 351, 353. 59. AA 346. 60. AA 353. 61. See Kant, ‘Perpetual Peace’ (On History), ‘Appendix I: On the Opposition (Mißhelligkeit) between Morality and Politics with respect to Perpetual Peace’ and ‘Appendix II: Of the Harmony (Einhelligkeit) which the Transcendental Concept of Public Right establishes between Morality and Politics’, AA Band VIII, 370–86. Harmony only arises in the face of the distinction between the political and moral orientations and through the adoption of the latter as an ideal and thus, necessarily absent, goal for the former. Kant focuses exclusively on how political practice should be oriented towards the moral, but I believe his account has a wider relevance for experience in general. 62. Friedrich Schiller, On the Aesthetic Education of Man. See, for instance, p. 131, Nineteenth letter, paragraph 6: ‘beauty can become a means of leading man from matter to form, from feeling to law, from a limited to an absolute existence’. 63. See ‘The End of All Things’, in Kant, On History, pp. 69–84. AA Band VIII 325–39. 64. CJ, AA 353 (Pluhar’s addition). 65. This is the message of Kant’s essay, ‘The End of All Things’. 66. Kant does not claim that the aesthetic phenomenon counts as an Idee in Hegel’s sense. The sensory given expresses the universal conditions of cognition, but only indirectly. 67. I argue elsewhere that Kant’s account of aesthetic judgement shares a project with Merleau-Ponty’s version of the epoché, which operates in the interests of uncovering the relation between subject and object. Although Husserl’s classical version aims to achieve this, arguably even in his later works he remains trapped in the perspective of the subject. See my ‘Kant’s Phenomenological Reduction?’. 68. See the discussion of a potentially pre-determinative unity or synopsis in Section 2 of the A Deduction in Chapter 4 (p. 123). 69. For Kant, unity entails determination under a concept. I am highlighting the way in which unity arises out of a process of unification. When this process arises without reaching a conclusion, then we ‘hold 309


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology together’ sensory impressions. This is sufficient for the figurative synthesis of an aesthetic form, but not for a formal intuition. 70. I am not suggesting, however, that we have direct access to the world as such. Our access is to things that necessarily appear against the backdrop of the world. See discussion in my ‘Kant’s Phenomenological Reduction?’.

310


29/3/07

8:12 AM


Afterword

I would like to add a few further comments on the character of harmony in aesthetic and cognitive judgements. I will also tentatively suggest a way in which judgements of the sublime have a significance not only for judgements of beauty, but also for Kant’s epistemological project. It is easy to conclude that Kant’s account of aesthetic judgement underestimates the extent to which the disharmonious plays a role in our experience. In twentieth century and contemporary art it would be fair to say that the disharmonious holds priority over the harmonious. This raises questions about the continuing relevance not only of Kant’s aesthetics but also of his theory of knowledge. For if I am right in arguing that aesthetic judgement presents an exemplary exhibition of cognition in general, then it might appear that Kant’s account of the cognitive relation between mind and world suggests much too unproblematic a ‘fit’ between the subject and the object. The importance of this is that were Kant’s position to amount to the view that the mind and world simply stand in harmony with one another, he would be in severe danger of falling back into something approaching a ‘pre-established harmony’. I will leave to one side the question of whether this would amount to the mind imposing order on the world or vice versa. In either case Kant would be guilty of falling back into the sort of metaphysics he was intent on avoiding. Were Kant committed to the view that mind and world immediately harmonise with one another, this would run counter to his conviction that we are governed by moral goals, that although imperfectly exercised by us, nevertheless count as categorical. Kant’s insistence that morality is an idea that is not to be found in the mechanically ordered world of nature, and yet which can be introduced into that world as an ideal goal, simply would make no sense were our minds and the world already standing in harmony with one another. Our capacity for moral reason allows us to go beyond the finite world of objects in thought. A possible disharmony between mind and nature is thus equally necessary in the moral sphere, as 311


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology is a possible harmony in the cognitive domain. The disharmony between the senses and reason is expressed aesthetically in the sublime, especially in its dynamic mode. The only question I will address here is that of the epistemic relevance of the claim that Kant is committed to a harmony between mind and world. Kant’s formal idealism has emerged as comprising a series of stages of determination. The subjective capacities are the sources of formal structures that allow us to take up given objects in the world. But our subjective faculties do not stand in harmony with the world; they are simply the conditions that make possible our knowledge of the world. Knowledge is achieved when the subjective representation of the world matches the objective state of affairs. Kant has revealed that subjective representation and objective states of affairs necessarily imply one another, and if they did not, he argues, we would have no chance of knowing the latter. This, broadly, is the message of the Copernican Revolution. But Kant does not merely state dogmatically that objects fit with our cognitive powers, he shows how this is possible through a process of synthesis. This is the work of the ‘Transcendental Deduction’, the ‘Schematism’ and the ‘Principles’ chapter. Knowledge a priori has emerged as a task, which I have suggested marks even the achievements of empirical knowledge. The formal structures supplied by the mind anticipate the material given so as to make knowledge possible. When and if knowledge is achieved, then what we can call an epistemic harmony can be glimpsed. But this knowledge will be of a particular empirical state of affairs, situated among a variety of such states, some of which have and some of which have not yet been determined. Even a firm piece of knowledge will only be partial because our knowledge of the world always occurs as a system, as the Analogies revealed. And as the certainty or harmony we have achieved occurs within a context that has not been fully determined, then the system of knowledge is never experienced as a totalised harmony. In Sartre’s terms there are totalisations, but not totalities.1 The possibility of a harmony of mind and world may be a working presupposition of the cognitive project, but it has to be combined with our need to discriminate and differentiate, tendencies that work against the goal of totality but in the interests of knowledge.2 But what of the aesthetic case where it appears that Kant is committed not only to a harmony between the faculties but also one between mind and world? It is necessary to look more closely at what Kant means by ‘harmony’ in aesthetic judgement. As I have argued, 312


29/3/07

8:12 AM


Afterword aesthetic judgement reveals the synthetic activity of judgement in general. Imagination holds a phenomenon together as if it were exercising a rule of the understanding, that is, as if it were determining an object. But the thing is not viewed as an object of knowledge. It is apprehended through a free or open combination of different orientations to the world, intuition or apprehension, understanding or a broadly reflective power and imagination. This ‘harmony of the faculties’ may appear, for lexical reasons, to signal a rather comfortable accommodation within the subject and between subject and world. But the meaning of ‘harmony’ is not self-evident and has to be gleaned from the way in which it is used in the text. Harmony is a ‘play’ of the faculties. Now as children know very well, play is a serious and varied activity. When the faculties are in free play, they do not merely meld into one another. To say that there is a play is to say that there is an exchange and an engagement, even a communication, while insisting that all this activity occurs through the mediation of imagination. We have already seen that even in the cognitive case the imagination is the mediating faculty par excellence. But in the aesthetic case, the imagination comes into its own. Play is activity that embraces its lack of a final resolution and thus stays in the domain of the imagination. A harmony of the faculties is not an identity relation, no more is it a bland compromise. It is rather a tonic exchange where the lawfulness of the understanding combines with the receptivity of intuition without either side taking priority over the other. This harmony is one within which there is difference and yet communication. An aesthetic harmony already has gaps within it, even before Kant moves on from the beautiful to the sublime. The pleasure of an aesthetic phenomenon is of something that holds together even though it is not bound together. And this is very important for the cognitive resonance of aesthetic harmony. Aesthetic judgement is exemplary of the possibility of cognition insofar as it displays synthesis in process and not synthesis as conclusion. The process or project of knowledge is anticipated in the aesthetic instance, revealing the viability of aiming at cognition, but giving us no grounds for epistemic complacence. The harmony of the faculties is a heightened case of the cooperation of the faculties necessary for any cognition whatsoever. Cooperation depends on the distinctiveness of the two terms that, in this case, work together. If sensibility and understanding genuinely cooperate with one another then it must at least be possible that their joint purpose could break down. This would be true both in the case 313


29/3/07

8:12 AM


Kant’s Aesthetic Epistemology of the failure of a cognitive aim and when the aim is other than cognitive. For Kant, morality entails the possible overcoming of the senses by reason. This discord of the harmonies is displayed in moral agency and in our aesthetic feeling of the sublime. Thus, the harmony of the faculties is only one pole of a range of possible relations in which the faculties can stand to one another and in which mind can stand to the world. At the other pole is the discord characteristic of the sublime. And neither of these poles is pure. Harmony, as we have seen, entails gaps and disharmony is never wholly discordant with our experience. Between these extremes is the diversity of proportional relations in which the faculties stand to one another. Judgements of beauty and of the sublime symbolise the range between harmony and disharmony within which our experience emerges. The sublime is not simply external to the beautiful: the two species of aesthetic judgement stand in an internal relation to one another, as I will now suggest. The sublime is the moment when the senses fail to make sense of the world and in this moment the faculties of imagination and understanding stand in disharmony with one another.3 The possibility of such a disharmony is the shadow of disruption within the beautiful. Harmony is an achievement, our pleasure in which requires an awareness that this result is not automatic. A harmony between the faculties and between mind and world – and we have seen that, for Kant, these are necessary corollaries of one another – can only emerge against the background of a possible disharmony. Indeed complex harmonies employ disharmony, just as Kant remarks that the artist is able to transform an ugly scene into a beautiful artistic presentation. Thus the sublime is not a mere parergon to the main account of aesthetic judgement, although it may count as such in a deeper sense, where the possibility of disruption is necessary to the possibility of presentation.4 Ian Hamilton Finlay’s bridge at Little Sparta is not, I think, an instance of the Kantian sublime. Our senses are not thwarted: they amplify our reflection on a perceptual phenomenon. However, the unity that we achieve emerges in and through difference. The difference within harmony marks the necessary possibility of the sublime that haunts the beautiful.

Notes 1.

I am thinking of the comment in his Critique of Dialectical Reason: ‘la totalité – contrairement à ce que l’on pourrait croire – n’est qu’un 314


29/3/07

8:12 AM


Afterword

2.

3.

4.

principe régulateur de la totalisation . . .’ (Sartre 1960, p. 138); ‘totality, despite what one might think, is only a regulative principle of totalisation . . .’ (Sartre 1976, p. 46). Unity or affinity is only achieved through seeking both homogeneity and variety. See ‘The Regulative Employment of the Ideas of Pure Reason’, CPR, A 657, B 685. For the purposes of this discussion, I focus only on the first disruptive moment of the sublime and not on Kant’s resolution of the sensory aporia in a turn to our capacity for moral reason. Allison’s characterisation of the sublime as a parergon is not at all similar to Derrida’s consideration of the same topic. Allison refers to Derrida’s notion of the parergon in KTT, on p. 8. Derrida discusses this notion in The Truth in Painting; see, in particular, pp. 37–82.

315


29/3/07

8:12 AM


Bibliography

Allison, H. (1983), Kant’s Transcendental Idealism, New Haven, CT, and London: Yale University Press (2nd rev. enlarged edn, 2004). Allison, H. (1996), Idealism and Freedom: Essays on Kant’s Theoretical and Practical Philosophy, Cambridge: Cambridge University Press. Allison, H. (2001), Kant’s Theory of Taste, Cambridge: Cambridge University Press. Ameriks, K. (1982), ‘Recent Work on Kant’s Theoretical Philosophy’, American Philosophical Quarterly vol. 19, no. 1, January: 1–24. Ameriks, K. (1982), ‘How to Save Kant’s Deduction of Taste’, Journal of Value Inquiry 16: 295–302. Ameriks, K. (1983), ‘Kant and Guyer on Apperception’, Archiv für Geschichte der Philosophie vol. 65: 174–86. Ameriks, K. (1983), ‘Kant and the Objectivity of Taste’, British Journal of Aesthetics vol. 23, no. 1, Winter: 3–17. Ameriks, K. (1992), ‘Rudolf A. Makkreel, Imagination and Interpretation in Kant’, Man and World 25: 227–34 [= ‘Review’]. Arendt, H. (1978), The Life of the Mind, Orlando, FL: Harcourt Brace and Company. Bauemler, A. [1923] (1967), Das Irrationalitätsproblem in der Ästhetik und Logik des 18. Jahrhunderts bis zur Kritik der Urteilskraft, Tübingen: Max Niemeyer Verlag. Baz, A. (2005), ‘Kant’s Principle of Purposiveness and the Missing Point of (Aesthetic) Judgements’, Kantian Review vol. 10: 1–32. Berkeley, G. [1710] (1996) ‘A Treatise Concerning the Principles of Human Knowledge’ in M. R. Ayers (ed.), Philosophical Works Including the Works on Vision, London: Everyman. Bird, G. [1962] (1973), Kant’s Theory of Knowledge, New York: Humanities Press. Buchdahl, G. [1969] (1988), Metaphysics and the Philosophy of Science: The Classical Origins Descartes to Kant, Lanham, MD, and London: University Press of America. Buchdahl, G. (1992), Kant and the Dynamics of Reason, Oxford: Blackwell. Burnham, D. (2004), Kant’s Philosophies of Judgement, Edinburgh: Edinburgh University Press. Caygill, H. (1989), Art of Judgement, Oxford: Blackwell. Caygill, H. (1995), A Kant Dictionary, Oxford: Blackwell. 316


29/3/07

8:12 AM


Bibliography Crawford, D. (1974), Kant’s Aesthetic Theory, Madison, WI: The University of Wisconsin Press. Derrida, J. (1987), The Truth in Painting, trans. G. Bennington and I. McLeod, Chicago and London: University of Chicago Press. Dickens, C. [1837] (1953), The Pickwick Papers, London and Glasgow: Collins. Düsing, K. (1968), Die Teleologie in Kants Weltbegriff (Kantstudien Ergänzungshefte, 96), Bonn: H. Bouvier & Co. Verlag. Fricke, C. (1990), Kants Theorie des reinen Geschmacksurteils, Berlin and New York: Walter de Gruyter. Gardner, S. (1999), Kant and the Critique of Pure Reason, London: Routledge. Gasché, R. (2003), The Idea of Form Rethinking Kant’s Aesthetics, Stanford, CA: Stanford University Press. Gibbons, S. (1994), Kant’s Theory of Imagination, Oxford: Clarendon Press. Ginsborg, H. (1997), ‘Lawfulness without a Law: Kant on the Free Play of Imagination and Understanding’, Philosophical Topics vol. 25, no. 1, Spring: 37–81. Guyer, P. (1977), ‘Formalism and the Theory of Expression in Kant’s Aesthetics’, Kant-Studien 68: 46–70. Guyer, P. (1979), ‘Review of Identität und Objectivität’, Journal of Philosophy vol. 76, no. 2: 151–67. Guyer, P. (1979), Kant and the Claims of Taste, Cambridge, MA, and London: Harvard University Press. Guyer, P. (1980), ‘Kant on Apperception and A Priori Synthesis’, American Philosophical Quarterly vol. 17, no. 3: 205–12. Guyer, P. (1987), Kant and the Claims of Knowledge, Cambridge: Cambridge University Press. Heidegger, M. (1962), Kant and the Problem of Metaphysics, Bloomington, IN: Indiana University Press. Heidegger, M. (1972), On Time and Being, trans. J. Stambaugh, New York: Harper & Row. Originally published in 1969 as Zur Sache des Denkens, Tübingen: Max Niemeyer Verlag. Heidegger, M. [1929] (1973), Kant und das Problem der Metaphysik, Frankfurt am Main: Klostermann. Heidegger¸ M. (1978), Being and Time, trans. J. Macquarrie and E. Robinson, Oxford: Basil Blackwell. Heidegger, M. [1927] (1986), Sein und Zeit, Tübingen: Max Niemeyer Verlag. Henrich, D. (1969), ‘The Proof-Structure of Kant’s Transcendental Deduction’, Review of Metaphysics 22: 640–59. Henrich, D. (1989), ‘The Identity of the Subject in the Transcendental Deduction’, in E. Schaper and W. Vossenkuhl (eds), Reading Kant: New 317


29/3/07

8:12 AM


Bibliography Perspectives on Transcendental Arguments and Critical Philosophy, Oxford: Basil Blackwell, pp. 250–80. Henrich, D. (1992), Aesthetic Judgement and the Moral Image of the World, Stanford, CA: Stanford University Press. Henrich, D. (1994), ‘Identity and Objectivity: An Inquiry into Kant’s Transcendental Deduction’, in R. Velkley (ed.), The Unity of Reason, Cambridge, MA: Harvard University Press. Originally published in 1976 as Identität und Objectivität, Heidelberg: Carl Winter Universitätsverlag. Hughes, F. (1998), ‘The Technic of Nature: What is Involved in Judging?’, in H. Parret (ed.), Kants Ästhetik/Kant’s Aesthetics/L’esthétique de Kant, Berlin/New York: de Gruyter, 176–91. Hughes, F. (2006a), ‘On Aesthetic Judgement and Our Relation to Nature: Kant’s Concept of Purposiveness’, Inquiry vol. 49, no. 6, December 547–72. Hughes, F. (2006b), ‘Taste as Productive Mimesis’, Journal for the British Society of Phenomenology vol. 37, no. 3, October: 308–25. Hughes, F. (2007), ‘Kant’s Phenomenological Reduction?’, Études phénoménologiques vol. XXII, nos 43–4, January: 157–85. Hume, D. [1739–40] (1978), A Treatise of Human Nature, 2nd edn, with text revised and notes by P. H. Nidditch, Oxford: Clarendon Press. Husserl, E. (2001), Analyses Concerning Passive and Active Synthesis: Lectures on Transcendental Logic, trans. A. J. Steinbock, Dordrecht: Kluwer, mainly drawn from Husserliana [=Hu] Band XI, Analysen zur passiven Synthesis. Husserl, E. (1950), Cartesian Meditations, trans. D. Cairns, The Hague: Nijhoff; Husserliana [=Hu] Band I. Kemp Smith, N. [1918] (1992), Commentary to Kant’s Critique of Pure Reason, Atlantic Highlands, NJ: Humanities Press. Kitcher, P. (1990), Kant’s Transcendental Psychology, Oxford: Oxford University Press. Kitcher, P. (1986), ‘Projecting the Order of Nature’, in R. E. Butts (ed.), Kant’s Philosophy of Physical Science, Dordrecht: D. Reidel. The Paul Klee Foundation (ed.) (1998–2004), Paul Klee: catalogue raisonné, London: Thames and Hudson. Llewelyn, J. (1982), ‘Heidegger’s Kant and the Middle Voice’, in R. Bernasconi and D. Wood (eds), Time and Metaphysics, Warwick: Parousia Press. Llewelyn, J. (1991), The Middle Voice of Ecological Conscience: A Chiasmic Reading of Responsibility in the Neighbourhood of Levinas, Heidegger and Others, London: Macmillan. Llewelyn, J. (2000), The HypoCritical Imagination, London: Routledge. Locke, J. [1689] (1975), An Essay Concerning Human Understanding, ed. P. H. Nidditch, Oxford: Oxford University Press. Longuenesse, B. (1998a), ‘The Divisions of the Transcendental Logic and the Leading Thread’, in G. Mohr und M. Willaschek (eds), Immanuel Kant: Kritik der reinen Vernunft, Berlin: Akademie Verlag. 318


29/3/07

8:12 AM


Bibliography Longuenesse, B. (1998b), Kant and the Capacity to Judge. Sensibility and Discursivity in the Transcendental Analytic of the Critique of Pure Reason, trans. C. T. Wolfe, Princeton, NJ: Princeton University Press. Lyotard, J-F. (1994), Lessons on the Analytic of the Sublime, trans. E. Rottenberg, Stanford, CA: Stanford University Press. Lyotard, J-F. (1991), Leçons sur l’Analytique du sublime, Paris: Éditions Galilée. McDowell, J. (1994), Mind and World, Cambridge, MA: Harvard University Press. Makkreel, R. (1990) Imagination and Interpretation in Kant, Chicago: University of Chicago Press. Makkreel, R. (1992), ‘Response to Guenter Zoeller’, Philosophy Today Fall: 276–80. Martin, S. (2003), ‘A New World Art? Documenting Documenta 11’, Radical Philosophy 122, Nov/Dec: 7–19. Meerboote, R. (1992), ‘Reflection on Beauty’, in T. Cohen and P. Guyer (eds), Essays in Kant’s Aesthetics, Chicago: University of Chicago Press. Melnick, A. (1973), Kant’s Analogies of Experience, Chicago: University of Chicago Press. Merleau-Ponty, M. (1945), Phénoménologie de la perception, Paris: Gallimard. Merleau-Ponty, M. (1962), Phenomenology of Perception, trans. C. Smith, London: Routledge. Merleau-Ponty, M. (1968), The Visible and the Invisible, Evanston, IL: Northwestern University Press. Merleau-Ponty, M. (1964), Le visible et l’invisible, Paris: Gallimard. Merleau-Ponty, M. (1993), The Merleau-Ponty Aesthetics Reader: Philosophy and Painting, ed. G. A. Johnson, Evanston, IL: Northwestern University Press. Merleau-Ponty, M. (1964), L’Oeil et L’Esprit, Paris: Gallimard. Nietzsche, F. (1974), The Gay Science, trans. W. Kaufmann, New York: Random House. Nietzsche, F. [1887] (1988), Die fröhliche Wissenschaft, Kritische Studienausgabe, Band 3, G. Colli and M. Montinari (eds), Berlin/New York: De Gruyter. O’Kelly, C. (1973), ‘Passage-grave Art in the Boyne Valley’, in Proceedings of the Prehistoric Society vol. 39: 354–82. O’Neill, O. (1984), ‘Transcendental Synthesis and Developmental Psychology’, Kant-Studien 75, Heft 2: 149–67. Pippin, R. (1979), ‘Kant on Empirical Concepts’, Studies in the History of the Philosophy of Science vol. 10: 1–19. Pippin, R. (1982), Kant’s Theory of Form, New Haven, CT, and London: Yale University Press. 319


29/3/07

8:12 AM


Bibliography Pippin, R. (1995), ‘Avoiding German Idealism: Kant and the Reflective Judgement Problem’, Proceedings of the International Kant Congress vol. 8: 977–97. Pippin, R. (1996), ‘The Significance of Taste: Kant, Aesthetic and Reflective Judgement’, Journal of the History of Philosophy vol. 34, no. 4, October: 549–69. Podro, M. (1998), Depiction, New Haven, CT, and London: Yale University Press. Prauss, G. (1989), Kant und das Problem der Dinge an sich, Bonn: Bouvier Verlag. Sartre, Jean-Paul (1960), Critique de la raison dialectique, Paris: Gallimard. Sartre, Jean-Paul (1976), Critique of Dialectical Reason, trans Alan Sheridan-Smith, ed. Jonathan Ree, London: New Left Books. Savile, A. (1987), Aesthetic Reconstructions: The Seminal Writings of Lessing, Kant and Schiller, Oxford: Basil Blackwell. Schaper, E. and Vossenkuhl, W. (1989), Reading Kant: New Perspectives on Transcendental Arguments and Critical Philosophy, Oxford: Basil Blackwell. Schiller, F. [1795] (1967), On the Aesthetic Education of Man, ed. Elizabeth M. Wilkinson and L. A. Willoughby, Oxford: Clarendon Press. Stendhal, (1891), De l’amour, Paris: Calmann Lévy. Strawson, P. (1966), The Bounds of Sense, London and New York: Routledge. Vaihinger, H. (1918), Die Philosophie des Als Ob, Leipzig: Felix Meiner. Vaihinger, H. (1935), The Philosophy of ‘As if ’, trans. C. K. Ogden, London: Kegan Paul, Trench, Trubner & Co. Ltd. Vaihinger, H. [1876] (1976), Kommentar zu Kants Kritik der reinen Vernunft: zum hundertjährigen Jubiläum derselben, New York: Garland. Walker, R. C. S. (1978), A Selective Bibliography on Kant, 2nd edn, Oxford: University of Oxford. Wheeler, K. M. (ed.) (1984), German Aesthetic and Literary Criticism: The Romantic Ironists and Goethe, Cambridge: Cambridge University Press. Zammito, J. H. (1992), The Genesis of Kant’s Critique of Judgment, Chicago: The University of Chicago Press. Zoeller, G. (1992), ‘Makkreel on Imagination and Interpretation in Kant’, Philosophy Today Fall: 266–75.

320


29/3/07

8:12 AM


Index

aesthetic formalism see formalism aesthetics aesthetic form see form aesthetic ideas see ideas definitions of ‘aesthetic’, 3–4, 190–3 affect, 55, 57, 61, 65–7, 69–72, 74–5, 78, 86–7, 95–107, 139, 149–51, 171, 183, 200, 211, 214, 222–4, 281, 290, 292, 294, 299, 302–5 definition of, 108n Allison, H., 25–6, 32–3, 38, 49, 61–9, 73, 75–8, 86, 89, 98–9, 107, 132–3, 169, 178–9, 189–94, 198–200, 253–9, 261, 281, 286 Ameriks, K., 30, 43n, 46n, 159, 167n–8n, 201n, 204n, 273n appearance (definition of), 92 appearances themselves, 226–7, 234, 236 appearing of the appearance, 278–84, 290–1 apperception, 20, 26, 29, 32–4, 127–9, 134–6, 144, 234 unity of, 12, 18, 20, 24, 29, 30, 34, 127–9, 133, 135, 143–4, 254 Baumgarten, 190–1 beauty the beautiful, 6, 263, 280, 291, 293, 295–7, 302, 313–14 beautiful object, 268, 284–7, 289, 296, 301–4 judgements of beauty, 160, 178, 193, 285, 297, 311, 314 Berkeley, 65, 87–8 Buchdahl, G., 38, 49–61, 62, 64, 67–8, 71, 78, 86, 89, 107, 207, 209, 251–2 Burnham, D., 168n, 273n

canon, 213, 221, 238 circle of representation, 49–78, 96–100, 102, 107, 171, 227 ‘cognition in general’, 76, 152, 154, 174–5, 179–80, 183–4, 187–9, 194, 198–9, 201, 264–6, 269, 279, 282, 285, 294, 297, 304–5, 311 cognitive formalism see formalism common sense, 169, 178–9, 182, 184, 186–93, 196–7, 201, 248 concepts see ideas constructivism see impositionalism Copernican revolution, 8, 38, 58, 62–3, 65, 86, 89–90, 92–7, 100, 104, 135, 171–2, 181, 214, 220, 249, 257, 265, 267–8, 280, 288, 312 deduction of aesthetic judgement, 169, 193–4, 197–9, 248 objective, 5, 107, 169–74, 181, 256, 268 subjective, 5, 68, 169–201, 268 transcendental, 5, 20–5, 27–30, 33, 62, 64, 68–9, 74, 91, 101, 112, 120–48, 152, 156–7, 160, 170–2, 176, 179–80, 207–21, 226, 236–40, 251, 257, 264–5, 268–9, 281, 297, 312 Descartes, 88, 229, 234 design, 216, 281, 289, 291, 292 dualism, 11, 34–6, 38, 59, 69–70, 73, 112, 114, 116, 121, 128, 131, 142, 153, 160, 286, 302 Düsing, K., 255–6 empirical guidedness, 10–15, 17, 39 event (of affection), 99, 103–4, 106–7, 280, 289, 295, 303–4

321


29/3/07

8:12 AM


Index exemplary exhibition, 265–7, 269, 277–305 faculties faculty talk, 10, 13, 35, 37–9, 68–71, 112, 116–19, 131, 172–3, 268 harmony of, 113, 118–19, 152, 155–6, 175, 177, 183, 195, 260–3, 267–8, 284, 299–302, 304–5, 312–14 mutual relation (or cooperation) of, 113, 117–51, 155, 160, 169–70, 172–6, 180–2, 184, 187–8, 192–3, 195–7, 201, 227, 230, 232, 238, 259–60, 262–5, 268, 282, 287, 299–300, 302, 313 plurality of, 78, 113, 118–19, 127, 130–2, 137, 146, 176, 178, 187, 192, 238, 281, 294 Fichte, 57 Finlay, I. H., 1–2, 3, 6, 314 form, 11, 24–6, 49, 52–6, 59–61, 63, 65–7, 69, 70–5, 77, 86, 92–5, 97–100, 104, 103–7, 113, 115, 130, 135, 137, 228–9, 239, 285–9, 292, 302–5; see also matter aesthetic, 71, 93, 104–5, 115, 278, 280–92, 294, 305 of matter, 56, 61, 98 of thought, 32, 34, 70–2, 172, 213–15 formalism, 5, 8–39, 49–78, 86, 97, 131, 267, 279–80 aesthetic formalism, 4, 93–5 Fricke, C., 272n–4n Gardner, S., 203n, 243n, 245 Gibbons, S., 110n, 163n, 202n, 242n, 307n Ginsborg, H., 274n grotto of Antiparos, 279, 288–9, 292, 303 Gurwitsch, A., 68–9 Guyer, P., 8–13, 17–27, 29–30, 34, 36, 38–9, 60, 71, 88, 95, 99, 107, 130, 179–80, 184–5, 193–4

Hegel, 17 Heidegger, 121–2, 127, 147–51, 157, 216 Henrich, D., 8, 10, 25–35, 39, 132 hidden art, 293 Hume, 65, 89 Husserl, 50, 54, 278 idealism formal, 5, 14, 19, 24, 62–3, 78, 86–108, 115, 171, 183, 223, 228–9, 235, 280, 292, 303, 305, 312 material, 5, 86–9, 96–7, 99, 104–5, 115, 125, 171, 223, 229, 237, 280, 305 ideas difference from concepts, 12–13, 15 imagination, 72, 76–7, 121–2, 125, 127, 130–6, 140, 146–50, 156–7, 214–15, 293, 313 free lawfulness of, 76, 156, 282 productive, 76, 127–30, 136, 138, 140, 223, 230 synthetic power of, 75, 116, 127, 138, 140–4, 146–51, 156–8, 160, 215, 223, 227, 258, 264, 281–2, 287, 293, 313 transcendental synthesis of see synthesis, figurative impositionalism, 5, 8–30, 33–6, 39, 60, 63, 71, 77–8, 86–91, 93–4, 97, 100, 106, 114, 121, 130–1, 134, 136, 142, 145, 183, 220 intuition a priori, 13 form of, 59, 73–6, 91, 96–8, 104, 107, 113, 120, 122–3, 127, 139–42, 144, 222, 236, 286 formal, 12, 73–4, 123, 139, 141–3, 286, 305 judgement aesthetic, 1–6, 76–7, 102, 113, 136, 151–9, 169, 174–201, 207, 212, 216, 249, 256–7, 259–62, 264–9, 277–305, 311–14; deduction of aesthetic judgement see deduction

322


29/3/07

8:12 AM


Index of beauty see beauty cognitive, 77–8, 151–3, 156, 174, 176–9, 181–3, 186–8, 191, 197, 199, 207, 277–8, 297, 299, 311 constitutive, 55, 60 determining, 15, 73, 77, 102, 112, 151, 156, 177, 182, 237, 249, 258, 280–1 empirical, 15, 49, 155, 248, 254, 258, 267 power of, 170, 194–9, 201, 252, 256–60, 262, 265, 284, 297 reflective, 15–16, 77, 102, 118, 151–2, 155–6, 158, 177, 217, 240, 248–9, 251–4, 257–62, 266, 268–9, 280–1 of taste, 3, 145, 153, 170, 177, 183, 188–9, 193–8, 200, 257–8, 267, 281 teleological, 159, 250, 260, 262 Kemp Smith, N., 11 Kitcher, P., 68 Klee, P., 291 knowledge anticipatory status of, 207–41, 266–7, 279, 295 relation to cognition, 6n–7n Leibniz, 28, 31 Llewelyn, J., 111n, 151, 164n–6n Locke, 38 Longuenesse, B., 49, 64, 69–78, 86, 89, 107, 122, 124, 127, 133, 140–2, 148

objective deduction see deduction O’Neill, O., 161 orientation in empirical world, 252–3, 258 plurality of different mental orientations, 38, 78, 92, 112, 114, 116, 118–19, 160, 229, 278–81, 298, 301–2, 313 phenomenology, 54, 107, 278–85, 304 Pippin, R., 8–17, 20, 38–9, 49, 59, 60–2, 64, 71, 88, 99, 107 principles analogies, 50, 64, 124, 180, 211, 222, 224–8, 233–6, 239, 303–4, 312 anticipations, 222–5, 233, 237 axioms, 222, 224, 232–3, 239, 240 categorical, 51, 249, 252, 254 constitutive, 51, 176, 269 dynamic, 221, 224 mathematical, 221–4 postulates, 222, 224, 228–9, 232–6, 239 regulative, 51, 208, 225 purposiveness, 6, 198–9, 201, 240, 248–61, 265–8, 277, 280, 282, 284–6, 300–1; see also judgement (aesthetic and reflective) formal, 250, 252, 256, 259–61, 264–5 general sense of, 256, 258 objective, 250 of judgement, 249, 257–60, 262 of nature, 198, 201, 240, 248–50, 252–69

McDowell, J., 101, 118 Makkreel, R., 156–60, 261–2 matter, 14, 25, 54–6, 59, 60–1, 65–7, 75, 77, 86, 98, 99–100, 103–5, 107, 137, 228–9, 288–9, 292, 302–4; see also form Merleau-Ponty, M., 227, 278, 291 music, 292

receptivity, 2, 56, 59, 61, 75, 86–7, 93–4, 96, 100, 104–7, 114, 118, 147 reflective judgement see judgement ‘Refutation of Idealism’, 17, 88, 149, 229, 234–5, 237, 303

object affective status of, 66–7, 96–106 Kant’s understanding of, 58–9, 65, 90–2, 97, 100–2

‘Schematism’, 13, 14, 33, 64, 149–50, 157, 172, 208, 211–22, 226, 229–38, 264–5, 281, 312 spatial, 207–9, 217, 229–38

323


29/3/07

8:12 AM


Index Schiller, 301 sensus communis see common sense Strawson, P., 8, 10, 13, 35–9, 60, 68, 71, 88, 100, 107 subjective deduction see deduction sublime, 6, 293, 311–14 synthesis a priori synthesis, 11–13, 18–19, 21–2, 23, 24, 27–8, 30, 160, 221 apprehension in intuition (synthesis of), 73, 122–9, 137–47 empirical, 11, 20, 221, 249, 255, 265, 269, 281, 285, 292, 302 figurative, 132–8, 142, 148, 152, 157, 172, 213–7, 222–3, 226–7, 229–32, 236, 238 in process, 113, 154–5, 159–60, 169, 201, 208, 213, 216, 218, 232, 241, 280, 283, 313 intellectual, 132–4, 146, 172, 209

recognition (synthesis of), 126, 146 reproduction in imagination (synthesis of), 20, 124–7, 137–8, 141 spatial, 141–2, 230–2 synthetic power of imagination see imagination temporal, 34, 141–2, 150, 231–2 systematicity, 6, 198, 240–1 empirical, 2, 48–69 transcendental, 251–5 transcendental deduction see deduction Trendelenburg, A., 18, 95 Vaihinger, H., 11 Walker, R., 30

324


29/3/07

8:12 AM



29/3/07

8:12 AM



29/3/07

8:12 AM



29/3/07

8:12 AM


Kant's Aesthetic Epistemology: Form and World

Kant's Aesthetic Epistemology: Form and World

Merit, Aesthetic and Ethical

Kants Welt. Eine Biographie

Epistemology and Developmental Psychology

Mainstream and Formal Epistemology

Realism Antirealism and Epistemology

Epistemology and the Social

Aesthetic Creation

Aesthetic Creation

Aesthetic Essays

Aesthetic Creation

Aesthetic Computing

Mathematics, form and function

The Geopolitical Aesthetic: Cinema and Space in the World System

Pure Pragmatics and Epistemology

Argumentation and Social Epistemology

Epistemology, Realism and Truth

Mainstream and Formal Epistemology

Kant and Contemporary Epistemology

House Form and Culture

Goodman, Forgery, and the Aesthetic

Epistemology and Emotions (Ashgate Epistemology and Mind Series)

Romanticism and Form

Marxism and Form

Logic, Form and Grammar

Mathematics: Form and Function

Marxism and Form

Aesthetic Experience

Aesthetic Communication

Kants Critique of Pure Reason

Kant's Aesthetic Epistemology: Form and World