Transcendental Deduction

Dr William Large

Today we come to the most difficult part of The Critique of Pure Reason. Indeed, there are some commentators who have made whole careers through simply discussing this part of the book.  Even Kant himself was so concerned about the difficulty of the transcendental deduction that he completely rewrote it for the second edition. We cannot hope, therefore, to be able to grasp and understand every argument and detail of this section, but we should be able to get the general aim and show how it fits into our own understanding of transcendental idealism.

Remember it is Kant’s hypothesis that it is not our knowledge that must conform to objects, but rather objects that must conform to our knowledge.  Why is this so?  For the opposite is not a tenable position.  If we claim that the object is utterly exterior to us, then how can we show that it is possible to know it at all?  We saw that Descartes could only solve this problem through the existence of God, who would ensure that our representations, how the object appears to us, would be the same as the object is in itself.  Kant’s solution to the problem, however, does not need to turn to theology for assistance.  Rather he simply puts forward the hypothesis that since it is we who relate to the object, then it is we who will determine what the object is.  This means that what the object is in itself is simply unknowable by us, and all we can know our appearances.  We need to be very clear here that what Kant means by ‘appearances’ is not what is usually meant by this word in philosophy, namely illusion.  For to interpret appearance as illusion is to understand it in opposition to essence, the thing in itself, but if appearance is the only thing we can know, then we cannot compare it to another kind of knowledge in comparison to which our knowledge of appearances would be said to be deficient.

If we only know appearances, then the question of objectivity becomes quite different.  It is no longer the problem, which has been the age-old philosophical problem, of knowing the objective essence of the object as opposed to our subjective impressions of it, but whether our experience of appearance has any coherence or order.  In Kant’s vocabulary, whether our experience involves any necessity.  It is clear that this necessity cannot be found in experience itself, for our experience is contingent and open.  This leads us to the paradox that in order to understand experience, we need to go beyond experience.  This is precisely the trap for philosophy, for in this murky area it is very tempting to invent all kinds of mythical concepts, since the limiting factor of experience has been left behind.

Kant’s solution, first of all, is to stay with the appearances themselves, so as not to be lead astray.  Thus, in the transcendental aesthetic, we begin with immediate experiences, or what Kant calls ‘intuitions’.  We recognise that every one of our experience takes place in space and time, but Kant has to show, in order to defend his position of transcendental idealism, that they can neither be real, nor merely concepts.  In other words, space and time, although not empirical, organise our experience of the world.  This would mean, against common sense, that space and time are not something outside ourselves, nor merely ‘words’, but an element in the subjective constitution of appearance, but nonetheless objectively valid since they are true for every experience what so ever.

Our experiences, however, for Kant are not merely immediate representation to the mind, but they are also discursive.  He means by this that we do not merely perceive something through our senses rather we also experience it through a concept.  Thus, we never just see the house rather we see the house as a ‘house’.  For this reason, knowledge for Kant is intuition plus concepts, but he equally wants to show that just as space and time constitute our experience prior to any actual experience, so too does our conceptual understanding.  The difficulty here is that is clear that space and time are given to us immediately in intuition and that they determine how objects appear to us, but it is not so clear that concepts do.  It is certain that there are concepts in our mind, but why aren’t these concept simply in our mind?  How do we know that they have any necessary connection to appearances in the way space and time does?  It is the purpose of the transcendental deduction to show that our conceptual understanding is just as important to the constitution of the object as the form of our intuition, in other words that concepts, at least the pure concepts of the understanding, are not merely descriptive, but constitutive.

In the metaphysical deduction, Kant’s argument goes from the logical use of concepts to the necessity of the categories.  In other words, we cannot understand the logical form of categorical judgement (x is F) without positing category of substance.  In the transcendental deduction we start with the relation between the subject and the object itself.  Now in ordinary view of things, we would think that all we need to make sense of is the terms of the relation itself: subject, object.  The most difficult philosophical question, however, is how it is that the subject has any relation to the object at all.  In other words what is the bridge between the subject and the object.  Kant’s rejection of transcendental realism is that we cannot see how there could be a bridge between them at all.

Now this bridge has two sides that coincide with the two element of the relation that need to be joined.  Thus, there is a subjective side of the transcendental deduction and an objective side.  The clue to finding out what these two sides do is in the function of the concept.  We remember that in the metaphysical deduction, Kant describes this as the capacity of concepts to organise our experiences by giving them a unity and coherence.  This question now is whether there is a conceptual unity that is even prior to our experience and which makes this experience possible, just as space and time make possible our empirical intuitions.

This means that we are looking for a subjective unity and objective unity that is prior to our actual experience.  Another way of looking at this question is what ensures that appearance in general, rather than this or that appearance has wholeness and integrity, and is not simply a jumble and mess of sensations and impression.  For I know that my experience is ordered.  Kant would say that first of all that I can say my experiences.  Thus no matter what experience I am having it is always related back to an ‘I’ (what this ‘I’ will the object of a latter part of The Critique of Pure Reason, but it itself cannot be an object, otherwise we would fall into an infinite regress).  It is this ‘I’ that I can refer back every experience, which gives unity to my representations.  We should be careful not to confuse this ‘I’ with the empirical consciousness itself; that is to say the individual consciousness of you or I, rather it is the general form of consciousness itself which ensures that intuition hold together in a minimal unity without which there would be no object at all to experience:

The synthetic unity of consciousness is, therefore, an objective condition of all knowledge.  It is not merely a condition that I myself require in knowing an object, but is a condition under which every intuition must stand in order to become an object for me.  (B137).
Just as there is a subjective pole to the unity of experience, so there is an objective pole.  Here we are speaking of not just this or that object, but the object in general - in other words, the concept of the object.  The argument is that without this general concept, the experience of an actual object would be impossible.  Again like the transcendental self, this transcendental object gives unity to our experience.  It is important, however, not to confuse the transcendental object x with the thing in itself.  Kant is not saying that behind every object there lies the general object that gives unity to it.  So, for example, every particular object would be the actual manifestation of one and the same substance.  Rather the general object only has a organising function in terms of our knowledge, but without which our experience would be impossible, without which, in other words, there would be no object at all for us.

Both the subjective pole and the objective pole of human knowledge are mutually dependent upon one another.  Thus without the unity of experience there would be no unity of the self, and without the unity of self there would be no unity of experience.  This is why it is the transcendental unity of apperception and the transcendental object x which act as the bridge between the subject and the object.  Not of course the object as it is in itself, but the object as appearance.  It is also demonstrate that without conceptuality; that is to say, without unity, there would be no experience of any object at all.  Thus conceptual conditions for human knowledge, and not merely the condition for valid knowledge of the object, but equally the possibility of that object at all: 

The a priori conditions of a possible experience in general are at the same time conditions of the possibility of objects of experience. (A111)
Both the necessity of a minimal self and a minimal object demonstrates the truth of transcendental idealism, for in their absence no experience of an actual object would be possible, but at the same time neither can be found in experience. It follows, therefore, that the general form of subjectivity must constitute experience in advance, and this general form is the pure intuitions of space and time, and the pure concepts of the understanding that organise our experience into a unity without which there would be only chaos and anarchy.

back to WL menu page