The Transcendental Deduction of the Categories (Version A)

by Philip McPherson Rudisill

Last Edited 4/29/2017

This is still very much in process and will remain so until serious editing takes place and as I continue to muse about all this. This is a very difficult section of Kant's writing and it is taking a real effort to arrive at clarity. The author's translation of the Deduction, along with comments, can be found here.

The question with regard to the Deduction has to do with the propriety or authority of the understanding to provide objects which can unify a manifold of appearances and in that way also show the reality of the provided object. In the pure viewing/Anschauung it was easy to show how the appearance/Erscheinung must be consistent with the conditions of our at-look/an-schauung/viewing. There we discovered how geometry provides its own object by means of a construction in space, in mid air before our eyes even. And we saw that it was only in this way that we could come to a recognition that, for example, any two sides of any and every triangle are greater than the third. And we then saw how this holds also for any appearances which might appear to us. And since all objects in the viewing/intuition (via time and space) must conform to time and space (as to the possibility of appearing to us at all), and since time and space are our own (and perhaps even peculiar) way of looking at appearances, all objects appearing in the senses are necessarily conformed to the conditions of space and time. Now we turn to see how it is that the understanding and its categories might have application to these same appearances.

There seem to be two directions in the Deduction which unite in the category of understanding.

Beginning with the Apperception.

On the one hand we can look at the conditions of the transcendental unity of apperception (consciousness of self). We begin with a pure, original, identical and transcendental apperception as a capacity for consciousness of self. The conditions for an actual recognition of self are the conditions for the unification of a diversity or manifold* which is given as a contingent singularity via the viewing, i.e., a spectral object (such as a face in the cloud or a tree). In order to have a manifold leading to this synthesis and unification we must perceive an object.

* This "manifold" is a technical term. Click here for a further and more detail discussion of this term.

In the pure viewing* of space we apprehend a triangle by connecting three lines by all their end points. In the empirical viewing we have an unidentified object in general, an object = X, and in order to perceive this we must peruse it in a state of mind focused on taking it (the manifold) all in, i.e., we must apprehend it, and then, as in the pantomimic description of a triangle in mid air, we must keep the perused or apprehended manifold in mind the entire time.

* See Kant and the meaning of the Anschauung. Earlier I used the term "envisagement for Anschauung.

When we take on the consciousness called "paying attention", we differentiate time and begin a span of time and then, ever mindful of that, we accumulate the relevant manifold, e.g., the sides of a triangle, and then we must ignore all that comes afterwards in order to focus on the manifold via our imagination. We do a start, an accumulation of data, and then a stop in time. This is a pure viewing/intuition/Anschauung, for there is nothing in the synopsis of the appearances/Erscheinungen to suggest a start or continuation or stop. Pauses in the tracing in mid air are just pauses, and not starts and stops, although we see the pauses in that way. That is our way of looking which is devised and instituted by the productive imagination in terms of this apprehension and reproduction or retention.

At this stage in the synthesis that is underway we have achieved to what I consider to be perceptions. When I perceive an object I don't just stare at it blankly; I rather take it in (and this can often be done without moving the eye balls, which is certainly a work of the imagination) and keep it in my imagination all together as a single thing (reproduce the manifold). A related next step is to associate the manifold by providing a rule which requires the elements of the manifold as it appears in the empirical viewing. An example of this, I think, might be the association of the legs and top of a table, namely that the legs are between the top and the floor.* This is then adequate for the subjective identification, namely it will serve me, although I cannot require it of others nor can I even require it for the future for myself, for there is no more necessity here at this point in time than in the face in the cloud. I spy a configuration of a manifold and can call it a table, but cannot in any way necessitate the manifold so that it must be so configured.**

* This goes beyond merely a rote remembrance and reproduction of the manifold of the table, for this is an ordering of the manifold via a rule which (subjectively) requires it, i.e., a table is a top in the air and legs going from the top to the floor.

** Once I necessitate the legs and top of the table by the concept of table (to be considered below) then I see this as not only sufficient for me, but even binding on me and all others, i.e., as universally objective.

Kant refers to such as Judgments of Perceptions in his Prolegomena. Another association Kant mentions there might be noting that when the sun shines on a stone the stone grows warmer (and which means no more than thus far in the perception this has occurred; and while it will obviously be expected in the future via Hume's Law of Association, we would have no reason to be surprised if it did not). Again this holds for us subjectively and entirely without the least necessity, being no more than the application of a descriptive rule to a repeated perception.

Let us pause to note that if our consciousness of self were based on nothing more than the subjective perception, we would be able to be conscious of a manifold and even to associate it, but then that particular subjective perception would lead nowhere and another perception would be nothing more than the contents of that particular manifold, i.e., no connection between the two perceptions. And so there would be no consciousness of self which is objective, but merely the unity that describes a manifold via a rule, e.g., legs between the top and the floor (regarding a table), or when the sun shines on the stone, the stone grows warm. We could in the case of this example be conscious of the top and legs together and see that they were joined by the rule in a single consciousness. But this is a subjective union which is entirely empirical and thus entirely contingent and not what is meant by necessity, nor would there be any relationship between the diverse perceptions.

For there to be an objective consciousness of self we must not merely join a manifold by means of a rule, but moreover we must bind it necessarily by means of laws. For example, once we conceive of a table as an elevated surface convenient for human use, we necessitate the manifold and see that the legs are not only between the top and the floor, but elevate the top to the desire level and, given gravity, have to be in that position.*

* This is still contingent since it is empirical, for we might find a table on its side. But the concept and recognition would force the question: why? And the explanation would have to be forthcoming, for else it remains a puzzle, e.g., the table might be broken or is being painted or discarded, etc. And so it is only my means of the necessitation of the viewing (of top and then legs elevating the top for human convenience) that I could ever consider a table on its side or upside down as curious and "out of whack".

Now this concept of the object provides the unity of the manifold which is accumulated by the apperception in paying attention; this identical and original apperception is unified in the recognition of the object via the concept. And this holds true of all objects, i.e., all are unified and recognized via a concept. But it is not just that the apperception is identical with regard to all the manifold of a given perception, but is identical with regard to all perceptions. And this leads us to experience. Experience is the unity of all the perceptions in accordance with a category (as the form of rules of the understanding). Each perception then is not only a synthesis via apprehension, reproduction and association and recognition (by means of the concept, e.g., the table), but is consistent and cohesive with all perceptions, those already made and those yet to come.* **

* This latter phrase alludes to the affinity of the appearances to be introduced below.

** See Quick Take on Kant's Perceptions and the affinity of all appearances.

As a consequence of this argument we can conclude that there is a necessary referral of all appearances to the categories (as the form of apperceptual unity), for it is only by means of the categories that we can recognize an object, and indeed even perceive an object.

Beginning with the Appearance (Erscheinung).

Now we want to look at all this from the other direction starting with the appearance. We have an empirical viewing of a spectral manifold and we apprehend and reproduce and associate it. Now the association, to the extent it is aimed at the unity of the viewing in general, is the work of the productive imagination.*

* And so in the synthesis of the perceptions we are working with the identical consciousness of self and a total unification must be made with all perceptions in a single experience, and not just in the immediate manifold, which the subjective perception (in the association and rule) can hold together in one consciousness.

There are three possible ways to consider the appearances as a whole. They are all either part and parcel of a single nature, or some of them are ruled by laws of nature and the others are things on their own (and not subject to any rule that is universal), or else all are things on their own (and subject to no universal rules). But since the only reason for unifying a manifold objectively (instead of remaining with the subjective and contingent perception) is for the provision of necessity; if I sought to unify a manifold, i.e., supply necessity, and if I thought of either of the last two states of the appearances above (some or all are not subject to laws), and if I presumed to provide a necessity via the concept of an object, that necessity could never materialize; for if even only one of appearances were not subject to laws of a single nature, and instead were a thing on its own, then that one might very well be the immediate manifold which I intent to unify via a concept of necessity, and so then, of course, there would and could be no necessity; and hence also no objective united and identical consciousness of self.*

* An interesting parallel arises in the theology of John Wesley of 18th century England. He maintained against the Calvinist that if God did not love all people, if there were even one person whom God did not love, then that one could be Wesley and that would mean that the "witness of the spirit" regarding his own approach to moral perfection could be false, i.e., not a single person could have the peace which were promised by the gospel account.

As a consequence of all this we see two things: The very perception that Hume used to construct his system of skepticism is not possible except under the presupposition of the categories. And likewise we know a priori that all appearances are part and parcel of a single nature (and are assumed as such) and thus any manifold can be unified and can be expected to be unified.* ** ***

* Kant mentions that this connection of all appearances (the affinity) is close or remote. The blowing of the Noon whistle and the invariable movement of the train shortly after are connected as disparate and coincidental effects of a single cause, namely the zenith of the sun, but this is a remote connection. A closer connection would be the sun light causing the stone to grow warm.

** We can now see that the object = X is actually the Transcendental Object = X = a single nature encompassing all objects, and for which reason we are able to expect that all possible perceptions will cohere together in a single experience, i.e., knowledge of this single nature necessarily assumed by the productive imagination (and required by the makeup of connective human understanding) on behalf of unity and the continuing identity of the consciousness of self.

*** By assuming the affinity of all appearances, i.e., that they are all together ruled by laws of a single nature, we are first able to recognize that the appearances are appearances (in the brainarium) and not things on their own, e.g., we can distinguish the rainbow (as a appearance which can only be seen, but not located, in space) from the rain drops (as real objects which can be both seen and located in space).

Thus we see the accomplishment of the deduction as the only, and thus necessary, way that we come to unity of apperception (called experience). We supply an object, i.e., the Transcendental Object = X = a single nature (the affinity), and by virtue of that have the experience that first enables us to recognize ourselves as an identical consciousness. So we justify the application of the categories, i.e., supplying the form of a nature, to the appearances. We supply an object (nature of universal laws) and are able to confirm the legitimacy of that object through the experience that arises as a result.

In this way we justify the provision of an object by the understanding, for here there is a validation and certification in the recognition of an actual empirical object.

* And it is just this that, we will later discover, is absent when pure reason dreams up its objects (Kant's "Ideas") and concludes that they exist necessarily, e.g., free will and God. But because no object is given in pure reason (contrary to understanding and experience), reason will stumble in a vain attempt to speak out beyond the concepts of experience, and will wander about in a circle and will stumble in self inflicted illusions.

Now before we get into a critique of pure reason itself we need to see how these categories of understanding, these pure concepts of unification of a manifold, are actually applied with regard to their specific orientation, i.e., quantity, quality, relation and modality.

The most important point here, I think, is the affinity of the appearances, for it is only via the assumption of a nature connecting all appearances via laws, that we are on the look out for hints of rules and connection, e.g., patterns and coincidences, and that we come to apprehend and finally unify, any manifold of the appearances. Hence this affinity even leads first to the unified self itself and makes it manifest and not just a potential. But it is no coincidence that these objects that we are searching for in the appearances (objects for the appearances to represent) are of the same nature as what we have assumed of all the appearances in the first place (both being based on the same categories), namely they are found together in space and time and bound together in time and space as a single nature.

And this ties in with the conception of the pure, original and transcendental apperception. This is a framework or potentiality for combining diverse consciousness into a single consciousness. And so the affinity is seen as an expression of a single object = nature = law-determined appearances, and this singularity and identity represents the apperception. We have the capacity for unification of diverse consciousnesses in a single consciousness in accordance with laws (per the categories) and the assumption regarding the appearances is that they can all be unified. Thus equipped with the pure apperception we look upon all the appearances as a single manifold called nature and which we then gradually unify in objects and achieve to objective perceptions (or recognitions). For this reason we can expect all our perceptions to coincide with each other and all fit together in a grand whole called experience (of that single nature).

Exposition of the Object

What do we mean with “object of representations”? We have the appearance and we have the concept of an object which calls for and requires a certain manifold in a appearance. This object which is different from both the recognition and the appearance is only a something = X which we are able to see by adding the concept in thought to the viewing in the appearance, such that then the appearance represents that the object denoted by the concept.*

* When we look at a table, what do we see? This depends upon the viewing as to what this thing (the table) represents. Is it a table? Is it wood? Is that what we are looking at? Is it teak wood? Is it a table, i.e., top elevated by legs for human convenience? Is it an article of manufacture? Is it a polished finish? All this depends upon the viewing and which is determined by the rule/object. For more on this see especially Kant and the meaning of the Anschauung.

And so we conceive of and put together a world full of spectral objects and which we acknowledge to be objectively in time and space, e.g., tables and stars. And we recognize that these objects are truly and objectively as represented to us in space and time. This is the realm of human knowledge or recognition. It is called the world of experience.

But we should not forget that there may be other objects which could be fashioned and recognized, either in a different viewing from space and time, or independently of any sensitive viewing altogether and subject only to an intellectual viewing/Anschauung alone, and which we cannot imagine, and which is meaningless for humans. In this way we can better appreciate that the appearances are looked upon as representations of a something which, necessarily, remains unknown, and for us we utilize the something = X to provide ourselves with a something, the object, for the appearances to represent and avoid having to consider them as things on their own.* And there is no reason to think that other objects might not be forthcoming if it were possible to consider the appearances in some way besides time and space. But for time and space this is the only way an object can appear, i.e., as represented in and by the appearance.

* The appearances either represent some object, or else they have to be treated as things on their own which, as I guess, is the take of the animals.

Hence the objects that we recognize in space and time via the categories are recognized as actual and objective, but we cannot expect any validation (and thus no use of the categories) except within the confines of experience. This is borne out when Kant gets into the Dialectic of pure reason.


For an expanded version of a gloss by the author on the first (A) version of the Deduction which was published in Kant-Studien in 1996 see "Circles in the Air".

See the actual wording of the author's translation of the first (A) version of the Deductions

Return to the essay Kant in a Nutshell.