Skip to content

FORM AND CONTENT IN KANT AND WITTGENSTEIN

Nabokov uses this notion of world-creation over and over again . . . [a] metaphor . . . which goes back to Kant and is parasitic on the disastrous Kantian distinction between form and content -Richard Rorty (1989, p. 168) 

Introduction

Why is this distinction so disastrous? Is it disastrous everywhere, or just in philosophy? Everywhere in philosophy or just in epistemology (Rorty's prime target in Philosophy and the Mirror of Nature)? Rorty puts forward his views on the distinction and the role it plays in Kant's philosophy as follows: Kant . . . [identified] the central issue of epistemology as the relations between two equally real but irreducibly distinct sorts of representations- "formal ones" (concepts) and "material ones" (intuitions) . . . [in addition] by taking everything we say to be about something we have "constituted," he made it possible for epistemology to be thought of as a foundational science, an armchair discipline capable of discovering the "formal" (or, in later versions, "structural," "phenomenological," "grammatical," "logical," or "conceptual") characteristics of any area of human life. ([ first brackets mine], 1979, pp. 138-9) In a related passage Rorty remarks: When we begin our philosophical meditations we do not, as C.S. Lewis and Strawson suggest, inevitably stumble across the intuition/concept distinction [when we begin to analyse experience]. Rather we would not know what counted as experience much less being philosophical about it unless we had mastered that distinction. (my brackets, 1979, p. 152) Now, as those in the trade know, we master this distinction by studying the responses of the British empiricists to the puzzle set by the Cartesian premise: a premise which states that we do not know the world directly but instead we know it-if we know it at all-through a veil of perception: we know it in terms of ideas, impressions, or using Kant's expression, objects of representation. The British Empiricists thought that objects of representation were combinations of sensible ideas (touches, tastes, smells, etc.), which were presented to us as already combined in the form of qualitied objects in space and time. Locke and Berkeley speculated about the provenance of these ready-made objects of thought, but Kant hypothesised that objects of representation-qualitied objects in space and time-were not given 'ready-made' as qualitied objects, but were constituted as such through the transcendental machinery of the mind. He argued that the rules which governed this machinery-the rules for synthesising objects of representation out of intuitions-could be deduced by considering the conditions of a possible experience. The conclusion he drew was that the given-what he called the manifold of intuition-had to be processed (given a form) so that this given (the content of our intuition) could manifest itself in our experience. The form that emerged was the various qualities of objects (of representation) "yellow," "hard," "oily-smelling," etc. The operations of the Transcendental machinery involved, in essence, 'imaging' the manifold of intuition-the raw material yielded by our sensibility-in a subjective time-frame. It is only if an intuition can be imaged and made to last-'spread out' so to speak -that it can become an object of awareness and manifest its phenomenological quality. [The metaphor I have employed of the transcendental machinery being used to 'spread out' the intuition in a subjective time-frame should not be taken to imply that, for Kant, the end result of this 'spreading out' ought to be regarded as revealing some intrinsic character of the sensation which is there previous to its being 'spread out' in time. Thus, though 'yellow' is what the correlated intuition 'looks like' when it is spread out in time, this quality is something which is somehow created when the intuition is processed or synthesised. Intuitions are not themselves, e.g., yellow. They simply yield the yellow effect when they are fed into the transcendental machinery. As Kant says "colours ... cannot rightly be regarded as properties of things but only as changes in the subject, changes which may indeed be different for different men." (1933, B 45 ) ] These distinct modal characteristics, (sights, sounds, etc.,) become qualities of a single object of representations insofar as experience teaches us to locate them all at one place and time (i.e., as qualities of a particuar object). The phenomenological result of temporalizing (and spatializing) intuitions (the result of giving a form to the content delivered by the senses) are the various qualities of an object of representation. Bearing this in mind, recall that Rorty's objection to Kant's transcendental psychology is expressed by him in the form of a question: Why should we think that sensibility in its original receptivity presents us with a manifold, a manifold which, however, "cannot be represented as a manifold" until the understanding has used concepts to synthesise it? We cannot introspect and see that it does, because we are never conscious of unsynthesised intuitions nor of concepts apart from their application to intuitions. (1979, pp. 152-3) This is an odd objection since it simply points out some straightforward consequences of Kant's transcendental approach to epistemology. When we adopt Kant's approach, we are assumed to be, by definition, unable to experience anything but the end product of the process whereby experience itself is synthesised. So why should it be regarded as an objection to cite the fact that we are not ever conscious of the original unsynthesised material? This is an unsatisfactory criticism of Kant since it, in effect, begs the question: it supposes that Kant has given no theory to explain why 'sensibility in its original receptivity presents us with a manifold [a content], a manifold which, however, "cannot be represented as a manifold" until the understanding has used concepts [a form] to synthesise it. . . etc)'. Rorty here ignores Kant's explanation of how our experience of objects originates in the transcendental processing of intuitions and thus ignores the theoretical basis for Kant's distinction between form and content. This argument is to be found in Section 14 of the Critique of Pure Reason (1933, B125-129) in which Kant puts the case for his Copernican revolution. In this section Kant argues that approaching the problem of a priori knowledge without the form-content distinction-without using the transcendental approach-has been tried by Locke and Hume and found wanting. It has been found wanting in that the problem of justifying synthetic a priori propositions cannot be solved without the critical turn that Kant's transcendental approach to the question embodies. Rorty's general line in Philosophy and the Mirror of Nature-that such attempts (transcendental or otherwise) to provide foundations for synthetic a priori propositions (understood as attempts to do 'epistemology')are moments in a dialectic which cannot finally be resolved-cannot be considered in this context. However, I do want to consider the legitimacy of Rorty's objection to this distinction from another angle in order to see whether the results of such an examination would serve to support Rorty's assessment of the disastrous character of Kant's form-content distinction. A First Look at Form and Content: The English Department For many, their very first working encounter with an abstraction is when they are asked to distinguish between form and content in a Literature class. Students may be asked, for example, how the sonnet form contributes to the effect of a given poem. They learn that the abstracted form can be filled with different contents, new wine in old bottles. Later, this same distinction may evolve into the distinction between the style and the substance of a given passage. In the course of these investigations, it becomes apparent that the same message or content may be delivered in a variety of styles with varying aesthetic effects, and that the subtleties of the interaction between style and substance are at the heart of the study of Literature. A first exercise in this connection may be to distinguish the content from the form by attempting to construe the meaning of a chosen passage. In order to accomplish this task, we find ourselves attempting to render in conventional language the thoughts which we believe to be expressed in the unconventional language which characterizes a particular passage of prose or poetry. We learn to call the difference between the conventional and unconventional language "style". At the same time we learn to refer to what is the same between them-the meaning or message of the passage-as content. We learn that the aesthetic effect of literature depends upon style and that the thought conveyed by the conventional rendering of a passage is in some subtle way enhanced when it is expressed using unconventional forms. The lesson for our purposes is that the style/substance (form/content) distinction is only ever applied (in the English Department) to unconventional language. Thus, I cannot take the conventional form of the thought and highlight its form in turn, by translating the thoughts expressed in the conventional form into some even simpler form. With the conventional form we have already reached bedrock, the point at which the form/content (or style/substance) distinction cannot be applied to sentences. (To do so would be, in effect, to ask for the conventional form of a conventional form.) To ask this question would constitute a category mistake and to do such a thing in earnest would never occur to a member of the English Department. The Distinction in the Philosophy Department In philosophy, the form/content distinction is encountered in many guises, most commonly (in this century) when we distinguish between a sentence and the proposition it expresses. On the basis of this distinction, it has been assumed that the same proposition can be expressed in many different sentences, whether they are conventional or unconventional. The delight in abstraction natural to philosophers-coupled with the fact that half the fun in any good dialectic is to be literal-minded-results, almost inevitably, in the question arising as to what a proposition actually is, this entity that different sentences have in common. One natural answer is to say that all the sentences which express the same proposition-and thus have the same content-do so in virtue of having the same meaning. And thus a proposition is identified as the meaning of these sentences. This in turn sets us wondering what the meaning of a proposition can be. Presumably the meaning is the content of the proposition, but what then is the form of a proposition? Enter Frege, Russell, the early Wittgenstein, all searching for the form of a proposition (its logical form) and later-repudiating the entire project-the Wittgenstein of the Investigations, denying that such a distinction is relevant to a theory of meaning understood as use. Following this line of thought it would appear that the modern development of the theory of meaning owes its origin to a mis-application of the form/content distinction. If the lesson taught in the English Department is any guide, the trouble begins when we make the category mistake of searching after meanings by applying a form/content distinction to a conventional sentence. Such a sentence would not tempt the English faculty to a search after its meaning since its meaning-because conventionally expressed-is already clear: all you have to do to know the meaning of a conventional sentence is to know the conventions, the rules for its use, and these are embodied in the skills we possess as native speakers. In the English Department, Wittgenstein's maxim "meaning is use" has always been implicitly accepted. Their interest is directed to unconventional uses of language which provoke questions as to what semantic effect the writer has conveyed by various departures from the norm. A nice joke in this context: Hemingway's answer to the question "Why did the chicken cross the road?": "To die. (pause ) In the rain." Conclusions from the comparison with respect to the Rorty/Kant nexus This look at the English Department's use of the form-content distinction seems remote from Kant's use of the distinction because Kant's philosophy-understood as epistemology-straddles two modern disciplines, viz., philosophy of language (or theory of meaning) and cognitive science (understood as the discipline which attempts to explain how our subjective experience arises from the activities of our brains). The straddling is apparent in Kant's two fundamental theoretical terms. This first of these terms is concept:. He identifies concepts as rules, but not rules for the use of words. Rather, concepts are rules governing the mental activity we call judging. These rules are either empirically derived (like the rule for picking out the horses from the cows) or they are a priori rules which are used by the mind to construct those qualitied objects in space and time (objects of representation) which we have to be able to experience before we can pick out horses and cows in the first place. Moreover, the a priori rules we follow are not followed in a vacuum: there is a content which is subjected to these rules and Kant calls this content intuition (this is Kant's other basic theoretical term). If he had been doing cognitive science he might have referred to intuitions (the sensible content provided by the sense organs)-on the computer analogy-as zeros and ones (or bits of uninterpreted data) requiring some software, some rules, before this content could manifest itself on the computer screen in the form of representations. Because of this straddling, it is not easy to see the connection between Kant's construal of the relationship between form and content and the relationship which is mooted in the comparison between what goes on in the English department and the Philosophy department when the form and content of sentences is discussed. The reason for this difficulty is that Kant, in effect, had not made the linguistic turn: he had not come to regard judgements to be essentially reducible to linguistic behaviour. From his point of view, every act of judgment was always a mental activity: the bringing of an intuition under a concept. This amounted to an act of recognition by the intellect (where the judgments were empirical (cows and horses) or (where the concepts were a priori ) an act of constitution (of objects of representation) by the transcendental machinery of the mind. So Kant has no theory of meaning, only a theory of judgment, and the comparison (just discussed between the English and Philosophy departments) is irrelevant to his use of the form-content distinction. To find a comparison which is closer to the form-content distinction as used by Kant consider the distinction at work in the Physics Department. Form and Content in the Physics Department When it comes to the epistemological question of distinguishing form from content in perceptions (or mental representations or ideas or impressions or phenomena), what goes on in the Physics Department is instructive. Starting with the natural desire to distinguish the various components which make up things in the world-the desire to analyse-Physics very quickly gets to simple things which have a form and some sort of stuff which instantiates that form. This stuff is matter. Later it turns out to be equivalent to energy and everything that exists is simply a manifestation (or form) of energy. Form plus content make up the furniture of the world and the business of Physics is to describe the rules in accordance with which energy can be transformed from one form into another. The Philosophy Department applies this same distinction to a realm similar to the physicists' world, namely, the mental representation of the world that we variously describe using terms like appearances, phenomena, perceptions, ideas., etc. Early epistemologists thought that this world of appearance was a simple copy of the world investigated by physics, but as the investigation of objects of representation continued, doubts arose. Attempts to analyse perceptions into simple ideas did yield a vague analogue of the discovery of matter (or energy) in physics: some fundamental mental stuff- consciousness perhaps-which manifested itself in various forms, e.g., sights, tastes, smells, etc. However, it became clear almost immediately that these forms of consciousness could not be transformed into each other in accordance with laws. And when the other technique of physical analysis was attempted, namely, dividing matter into smaller and smaller pieces, it proved impossible to distinguish form from content as far as the atoms of the various sense modalities were concerned. Hume's doctrine of sensible minima (1978, p. 27) makes this point nicely. It is not possible to apply the form-content distinction to phenomena: this is due to the fact that the atoms of sense are not infinitely divisible-at a certain lower limit they simply disappear if they are reduced in size-and there is no point at which form can be separated from content (no way to split the sensible atom). In order to apply the form/content distinction to phenomena, a radical new approach was required. Kant changed everything when he discovered a way to distinguish between the form and the content of phenomena. He offered a theory in which the individual appearances which together make up the items which we experience were the end- result of a special kind of process in which the raw material of sense perception (the matter of our ideas which he called "sensation" or "intuition") was transcendentally organized in accordance with a set of rules (a priori concepts) so that the given (sensation) could manifest itself to us in a form in which it could be experienced. This parallels the model from the physics department in which energy itself (= the physical counterpart of the matter of sensation) can only exist (=be known/ experienced, cognized) in various forms. By analogy, the matter of sensation can only be experienced when subject to a transcendental synthesis yielding qualities: colours, sounds, smells, etc. Kant's concept/intuition version of the form/matter distinction was dynamite. It carried with it the assumption that the process (effected by transcendental psychological machinery) of bringing an intuition under a concept was creative: it produced the world of experience. (At this point Rorty's remark in the opening quotation about world creation being parasitic on the from/content distinction in Kant becomes clear.) This end-result (experience) was not a simple copy of the outside world: instead the process presented in an accessible form, the raw information or data supplied by the senses. The whole theory assumed that this raw material was not capable of being experienced without being synthesised (by the transcendental machinery) so as to then become manifest as qualities (of objects of representation). Reconsidering Rorty's objection Now what can we say about Kant's presumption that we are in possession of such a manifold of sensible intuition-the content of sensation-to be synthesised; something Rorty calls into question when he says: We cannot introspect and see [this process at work], because we are never conscious of unsynthesised intuitions nor of concepts apart from their application to intuitions (Ibid, 152). Kant's presumption (that there is a given composed of unsynthesised intuitions) is based on a fact which he only brings to light in the Anticipations of Perception. There he points out that all real experiences (versus our experiences of the products of the empirical imagination) have an intensive magnitude which varies continuously. A good way to understand this is to think of the frequency and the amplitude of a sine wave on an oscilloscope as reflecting the objective content of a sound wave. This objective content is re- presented and, in a certain sense, preserved in the shape of the sine wave. In the same way, our experience of a sound has two features which preserve the objective content of the sound wave. The first feature-the fact the the particular sound has a certain loudness-reflects (and in this sense, preserves) the particular amplitude of the sine wave, while the relative pitch of the sound reflects the second feature of sine wave, viz.,its particular frequency . These facts about the intensive magnitudes of qualia signal, as it were, the quantity in the qualia. They signal the objective content of the given (supplied by the unsynthesised sensible intuition) which is preserved as quantitative features of the qualitative phenomena that we synthesise. Thus, according to Kant, experience is not constituted solely by the mind: only its form is so constituted: its content remains as an indication of its givenness and is reflected in the intensive magnitude of qualia. When this is taken into account it provides a way of relieving the pressure exerted by Rorty's objection: thus, in the sense explained, we are (contra Rorty) "conscious of of unsynthesised intuitions" insofar as we are conscious of the quantity in the the qualia, viz., the content apparent to us in the intensive magnitude(s) of any quality, but of course we are not consicous of this objective feature of unsynthesised intutions (or anything else) until it is revealed as a feature (e.g., loudness, pitch) of a quality of an object of representation (created through the operations of the transcendental machinery). The radical nature of Kant's form-content distinction According to this transcendental tale of the synthesis of the qualities which characterize objects of representations, this re-presentation of the world's effect upon our senses could no longer be thought of naively as a straightforward reproduction in a mental medium of something in the world which exists in a physical medium. Instead, this end result of the transcendental process became a new kind of thing-a sui generis medium for representing the world that Kant dubbed experience. The radical nature of Kant's theory lay in the fact that-in the case of experience-there was no handy analogue, as there was in the case of an appearance, that could be used to understand the relationship between experience and the thing of which it was an experience. Thus, in the case of Locke, where the idea is thought of as an appearance, the relationship between the idea and the thing of which it was an idea was thought of on the analogy of a portrait and the person who posed for it. This analogy never actually worked for ideas thought of as appearances, for, as Berkeley pointed out, an idea can be like nothing but another idea. However, for all that, the analogy was there as an implicit assumption. Locke's discussion of primary and secondary qualities makes this obvious when he speaks in the Essay of the ideas of primary qualities of bodies as being "resemblances of them, and [in his maintaining that] their patterns do really exist in the bodies themselves ..." (1947, p. 47 [Book 2. Chapter 8.¶15]). No such analogy was available to explain the relationship between a Kantian object of representation and that which it represented. Experience, so far as its form was concerned, had to be regarded as sui generis and its relationship to its ground could not be explicated. The only thing you knew about experience was that it represented something real outside the mind in that an element within it, viz., its intensive magnitude-was not something which could be attributed to the synthetic activities which accounted for its form. Conclusions drawn from this comparison with Physics We have seen that the form-content distinction borrowed from the Physics Department was not applicable to phenomenal objects as conceived by the early empiricists. In order to apply the form/content distinction to phenomena we have to regard them, with Kant, as constructions. The content will then be the given (the uninterpreted deliverances of sense: the sensible manifold) and the qualitative results of following the rules for constructing phenomena (via the transcendental machinery) will constitute the form. This move ended the traditional epistemological debate basd on the question: do our ideas resemble their causes in the world? The new answer, according to Kant, would be: the content of these ideas originates outside of us-its signature as a real input is its intensive magnitude - the brightness of the yellow colour. But the form which this given takes, its appearance as a bright yellow (and more generally, its spatial and temporal features)-originates with us. An observation on the realism/anti-realism debate If this view is correct then the whole realism/anti-realism debate which stutters on endlessly is no more than the continuing reverberation of Kant's application of the form/content distinction when applied to the notion of an idea as naively conceived by the empiricists.. Realists are the Lockeians hoping that (with regard to primary qualities at least) "the patterns [present in ideas] do really exist in the bodies themselves" and anti-realists are Kantians who have overlooked the significance of the Anticipations of Perception passage. In this passage Kant emphasised the fact that our representations of reality-in so far as they have intensive magnitudes-preserve an element in the given which reflects an independent reality. Anti-realism cannot be true because we do not create the intensive magnitude-the content-of the qualia. Concluding assessments: One: On the legitimacy of the form/content distinction in a theory of meaning The lesson of the comparison with the English department is clear: we cannot apply the form-content distinction in the philosophy department if this amounts to the suggestion that it makes sense to ask what a conventional sentence means. To appreciate this we have to recognize that the form/content distinction is only applicable when you can describe the concrete relationship between the form and the content i.e., when you can actually separate the two things and consider them on their own. I can actually specify the rules that describe the form or style of a given example of Shakespeare's prose. These rules, when followed, produce unconventional styles of expression. This unconventional way of speaking gives a semantic "spin" to the content expressed in such utterances. Somehow, "To be or not to be, that is the question", has a more profound content than its conventional "equivalent": viz., "the question of whether or not a person should commit suicide is of supreme importance" and that apparent lack of semantic equivalence is why style or form is so interesting to the English Department However, as I said before, it would be pointless to write down the rules that characterize the style (form) of conventional language in order to see how conventional language gets its special meaning. There is no semantic spin imparted by conventional language: it means just what it says; ergo: there is no distinction between its form and its content. From the point of view of the English department this fact is unremarkable: from the point of view of the philosophy department the idea that a conventional sentence means just what it says could only seem mysterious-a fact requiring an explanation-because some philosophers are still applying the form/content distinction to conventional expressions. Thus for the early Wittgenstein, the problem of how the content (equivalent to the "meaning" or "sense") of a proposition somehow managed to represent the content of a state of affairs-a situation in logical space-was solved when it was seen that both had the same logical form: " That is how a proposition is attached to reality: it reaches right out to it" (2.1511 Tractatus) via our projection of its logical form.. However, The ultimate dependence of the theory of meaning on this metaphor of reaching right out to it and of projection (and the whole idea of the proposition as "picture") signals that the theory is in trouble. The trouble stems ultimately from (the category) mistake of applying the form-content distinction in the wrong field, of imagining, for example, that conventional prose has both a form and a content and that these can be distinguished. Is there a way to rescue the form-content distinction as developed by Kant in its transcendental employment? So far as its empirical employment is concerned, Rorty is right in saying that there is no way to apply it to phenomena. Rorty points out that we do not introspect the content-the sensible manifold-therefore, there is no content to be organized. ("We cannot introspect and see [the manifold of sensibility], because we are never conscious of unsynthesised intuitions nor of concepts apart from their application to intuitions") Rorty's point about the negative results of introspection can be reinforced by considering a topic examined by Wittgenstein in connection with his views concerning philosophical psychology. The things we see under normal conditions might be thought to be subject to a form-content distinction if we were- literally-seeing through a glass darkly (and hence seeing unconventionally). But under normal conditions we cannot apply this distinction. Wittgenstein's explanation for this-using Kant's language-is that the experience of seeing something (thought of as process in which we bring a content (intuitions) under a concept (rule or form) is not a process within which we can actually distinguish these two elements: the concept and the intuition. Thus we cannot distinguish (in the case of the familiar duck/rabbit picture puzzle) the purely perceptual (intuitional) content which is at one moment interpreted as a duck and the next as a rabbit. Because we cannot, the theory which holds this empirical distinction to be a possibility must be mistaken. The origin of the mistake is quite straightforward: we are used to classifying items in accordances with rules, e.g., in biology we classify via phylums, classes, genera, species, etc., but the items that get classified-the qualitied things extended in space and time-are not themselves kinds of things: "thing" things. We can not get at these items empirically and distinguish their form and their content: we cannot introspect the as yet uninterpreted intuition-which is then to be interpreted under different concepts, e.g., duck, rabbit. So there is no doubt that, as far as the empirical application of the distinction is concerned, Rorty is right. However, for all that, his criticism is irrelevant: Kant's theory is a theory about the transcendental application of the form-content distinction. This consideration comes out quite sharply when you consider Kant's views regarding the creation of phenomena (objects of representation) in accordance with the a priori forms or categories. These categories all turn out to be aspects of time-determination. But if you ask what it is that supplies the content to be determined-that which is to be 'spread out' in time-it becomes clear that the content which gets determined is itself without any qualitative character at all: simply because we only know it as having a character once it has be determined in time. It is as if the input from the senses-the content or matter of experience-were a non-entity until-suddenly-it takes on a phenomenal form and its content then becomes visible, audible, tangible, However, this magical interpretation of Kant's theory-the "world creation" interpretation- ignores the point made earlier, viz., that all qualia have an intensive magnitude which we can recognize (in Kant's language) as "the real in sensation" (1933, A 168). The manifestation of intensive magnitudes in our experience of qualia emphasizes the fact that our experience has a content: we only provide the form of experience, we do not create the content which our experience represents to us. This consideration of the Anticipations of Perception passage shows that Rorty's complaint against Kant's use of the form-content distinction misses the mark when considered in relation to the transcendental employment of the form-content distinction. Summing up: assessing the justice of Rorty's' remark Where, following the later Wittgenstein, we see meaning as use, there is no room whatsoever for the form/content distinction to take hold: It does not make sense to talk of a "form of use" or a "content of use". There is, therefore, no temptation to make some sort of category mistake and apply this distinction to linguistic practice as there was when it was imagined that sentences expressed propositions and that these propositions or their meanings were the content of these sentences. Where meaning is simply use, then the manifestations of meaning, viz., linguistic abilities or skills, are simply not the sorts of things which could have both a form and a content. They are just behaviours that we have learned. In this area it is fair to say that Rorty has it right: as far as theory of meaning is concerned: "the . . . Kantian distinction between form and content" is disastrous once we have made the linguistic turn.. The idea that meaning is use-at a stroke-makes the application of the form/content distinction impossible. For example: the question "What is the content of using a particular sentence?" sounds unintelligible. Compare: "What does that (unconventional) sentence mean?" when it is equivalent to the question: "Under what circumstances would you use that sentence?": Here the answer would be: "Under the circumstances in which you would use this conventional sentence." This construal reveals the way in which talk about qualia can lead to talk about the content of experience. Thus, for this unconventional sentence: "Notice the greenness of this green (pointing to some grass)", substitute: "That grass is an unusual green in this light (i.e., seen in this unconventional lighting)". Here there is no temptation to say that the two sentences are both expressions of the same proposition-that they express the same content in different forms-with the implied problem of what that content actually is. Instead, we can see that both sentences can be put to the same use and that this is the source of their equivalence in meaning, not some common content related to some private reference to the intrinsic character of the qualia in question (the greenness of the green) to which they both make reference. Nevertheless, we might still imagine that in every case we are always directly aware of the qualitative content of our experience in terms of, e.g., the greenness of green; or (in music) the peculiarly b flatness of b flat; or the intrinsic silkiness of the touch of silk. We have a strong feeling here that we know what is being distinguished as the content of this experience, as if somehow we could intuit the content of the experience directly (privately) though without perhaps being able to articulate it any further. Now this thought leads into familiar territory: if this content cannot be articulated, it means that the content of our experience-understood as sensations or qualia-is known directly only to us. But (as in Wittgenstein's famous 'beetle in the box' example) to regard such private knowledge as knowledge must be an illusion since we cannot characterize it except by abstracting the content from the form verbally, i.e., through talk of the greenness of green: we cannot state publicly what the rules for applying this concept are (but more on the notion of whether the 'greenness of green' makes sense in a moment). To sum up: The form-content distinction will not work within the field of empirical psychology. The duck/rabbit example makes this clear: I just cannot experience any separation of form and content when I experience this figure first under the one aspect duck and then under the other, rabbit., There is no neutral content:the-figure-as-not-yet-brought-under-a-concept that I can somehow experience before I bring this intuition (to use Kant's term for content) under one or other of the contesting concepts. Thus the upshot, so far as empirical psychology is concerned, is that form and content cannot be separated and, again, Rorty-following Wittgenstein-is right about the disastrous nature of the form-content distinction which leads to a view of the mind as interpreting some neutral sensible content in accordance with a concept. in Rorty's words "We cannot introspect and see that it does, because we are never conscious of unsynthesised intuitions nor of concepts apart from their application to intuitions." The value of the transcendental form/content distinction On the cognitive science side, the distinction, considered in its transcendental application, still seems useful or at least intriguing. Would it be possible to redeploy Kant's answer to the problem set by epistemology-not as a solution to the problem of how synthetic a priori judgments are possible-but instead as an early attempt at solving the most interesting problem of cognitive science: viz., how consciousness is related to brain activities? The idea would be that we need to look for brain structures which could perform some of the functions that Kant's transcendental machinery performs. e.g., producing a subjective spatial and temporal framework, supplying a transcendental unity of apperception, a transcendental imagination, and so on.1 The Kantian machinery even, as it stands, sheds some light on the qualia problem. For example, intensive magnitudes are something which Frank Jackson's Mary might accurately predict. Before she is exposed to light of a given wavelength she might be able to predict with confidence that the amplitude of the light waves will make her experience of the qualia, in part, an experience of a particular "brightness" (or whatever we choose to call the factor correlated with the amplitude feature of the stimulus) which will vary continuously along a continuum of "brightness". Further, she will predict that the frequency of the light waves will make her experience of the qualia, in part, an experience of a particular "hue" (or whatever we choose to call the feature correlated with the frequency feature of the stimulus) which will vary continuously along a continuum of "hues". Doubtless, the form which this hue continuum takes will come as a surprise. However, this new knowledge of the intrinsic form of hues-her particular rainbow-will, as a matter of fact, be something which Mary will never be able to communicate to the interested parties anyway. The conversation will go: "Well, Mary, what do you think of this portion of the spectrum; we call it green." To which Mary will reply: "Wow" meaning: "I could never have predicted that." But then the parties will not go on to chat about the particular form of green-the greenness of green: that element will be contributed by the unique character of Mary's transcendental machinery and cannot be communicated: as Kant says: "[Colours (strictly hues) ] . . . cannot be regarded as properties of things, but only as changes in the subject, changes which may, indeed be different for different men" (1933 B 45). Mary included. References Hume, David 1978: A Treatise Concerning Human Nature. Oxford: Oxford University Press. Kitcher, P. 1990: Kant's Transcendental Psychology. Oxford: Oxford Univesity Press. Kant, Immanuel. 1933: Critique of Pure Reason.Translated by N. Kemp Smith. London: Macmillan. Locke, John 1947: An Essay Concerning Human Understanding. London: Dent and Sons. Rorty, Richard 1979: Philosophy and the Mirror of Nature. Princeton, New Jersey: Princeton University Press. --1989: Contingency, Irony and Solidarity. Cambridge: Cambridge University Press. 1Cf. Kitcher: Besides commonality of subject matter, there is a further relation between transcendental psychology and empirical psychology. In defense of applying the "psychology" label to their work, Newell and Simon (Human Problem Solving, Englewood Cliffs, New Jersey.: Prentice Hall, 1972 ) argue that highly abstract normative analyses of the requirements of knowledge are relevant to experimental work aimed at identifying psychological mechanisms. The same point applies to transcendental psychology. If we discover that certain cognitive tasks require the synthesis of diverse representations, then we have an abstract description of mechanisms that can be sought and further described through empirical investigation. Kant sees this relation clearly: "What I called applied logic [empirical psychology (P. K. p 26)] ... is a representation of the understanding and of the rules of its necessary employment in concreto, that it, under the accidental subjective conditions which may hinder or help its employment. It treats of attention and its impediments and consequences, of the source of error, of the state of doubt, hesitation, and conviction, etc. (A54-55/ B78-9)" (1990, p. 25-26) COMMENTS AND QUESTIONS WELCOME AT: david.ward@stonebow.otago.ac.nz