Kant's advocacy of transcendental idealism has always been regarded as detrimental to the plausibility of his system. Despite Kant's many attempts to clarify the meaning of 'idealism,' it seems a drastic measure to accept that space and time, together with the things in them, are ideal. This paper topic asks you to analyze one of Kant's efforts to show that his idealism is not so bad as it seems, and that one must recognize that it is also a kind of realism.
The issue is whether time is real. Kant had claimed that time is empirically real, which indicates "its objective validity in respect of all objects which allow of ever being given to our senses" (BA35/B52, immediately preceding the target passage). What he denied is that time is "absolutely" real, that it belongs to things "as their condition or property, independently of any reference to the form of our sensible intuition" (A36/B52). The objects for which time is a condition are appearances, so time "cannot be ascribed to the objects in themselves (apart from their relation to our intuition) in the way either of subsistence or inherence" (ibid). Because time is not a thing or property of a thing in itself, it is ideal. In the same section as the target passage, Kant stated that the doctrine of the absolute reality of time conflicts with "the principles of experience itself" (A39/B56).
In the passage from the Dissertation, Kant had stated that time is not "objective and real," but instead a necessary subjective condition for co-ordinating sensible things. Lambert interpreted this as meaning that time "is only a helpful device for human representation," and we may infer from this that he took Kant to be denying all reality to time. Lambert simply gave a consideration in favor of the reality of time: that it is a condition of change. Since it is indisputable that there is change in our representation, time is real. In the Critique, however, Kant had made the distinction noted above between two kinds of reality, which allowed him to grant that time is empirically real. "Thus empirical reality has to be allowed to time, as the condition of all our experiences; on our theory it is only its absolute reality that has to be denied" (A37/B54).
Now Kant claimed that the objection just stated is directed "against this theory, which admits the emprical reality of time, but denies its absolute and transcendental reality" (A36/B53). In its original form, the objection was to an earlier theory, not the theory of the Critique. But in the context of the distinction between kinds of reality, the argument, Kant thought the argument could be overcome easily, so long as the doctrine of the empirical reality of time is at all plausible.
I believe that the objection can be redirected against Kant's view as stated in the Critique. Time is said to be something real, in that "I really have the representation of time and of my determinations in it" (A37/B53). From this it follows that I really represent myself as something existing in time. It seems plausible to suppose that I have different representations of myself: schematically, I have one representation of myself as being in a certain state (thinking of playing golf) at t1, and I have another representation of myself as being in a different state (thinking of a drink at the 19th hole) at t2. So to represent myself as being in different states at different times, I must have different representations. But this requires that there be change myself from one act of representing to another. And change occurs in time; in this Kant agreed with Lambert. So my act of representation of myself as in time must itself be in time.
This conclusion seems inconsistent with the ideality of time. In the case given we do "abstract from the subjective conditions of sensible intuition" because we are talking about what is needed for these conditions to be applied in the first place. Granted, we do not intuit ourselves as that which represents itself as in time, but we know that in order to represent anything as an object of inner sense, we, as the subject of representation, must be changeable and thus in time.
The argument is quite simple. Our representations really change, and change takes place in time only. So time is real as a condition of changes that really take place. It does not matter whether objects in space exist or not, for even an idealist admits that changing representations are real.
As noted above, Kant granted that time is real but qualified its reality as being empirical only. The objection claims that our representations change, and Kant interpreted this to mean that we intuit ourselves as having different representations at different times. Time is real "as the mode of representation of myself as object" (A37/B54). Since we can only represent ourselves through inner intuition, and since time is the form of inner intuition, the only way we can represent ourselves as subject to time is as an object of inner intuition, an appearance. If we could intuit ourselves in some other way (as a thing in itself), we would not represent ourselves as changing or consequently in time at all.
Is there a response to the deeper objection noted above? The best hint I can find is in the footnote: "I can indeed say that my representations follow one another; but this is only to say that we are conscious of them as in a time-sequence, that is, in conformity with the form of inner sense." This seems to indicate that because the higher-order description of ourselves as representing ourselves as objects in time is also a representation of ourselves, its object (our representing self) is also subject to time. Thus I first represent myself as at t1 thinking of playing golf, and later represent myself as at t2 thinking of drinking in the club bar. The claim would be that any representation of myself, no matter how high-level is ultimately a representation of myself as an object of inner intuition, and hence as subject to time.
I am not satisfied with the response I have constructed on Kant's behalf. Although I grant that it makes some sense that to say that all self-consciousness is through inner intuition, I also think that on Kant's theory, self-conscousness is not the only mode of self-knowledge. In doing transcendental philosophy, Kant made many claims about the subject of representation, e.g. that it has sensibility and understanding, that its intuition is inner and outer, that the form of inner intuition is time and that of outer intution is space. The point at issue, that there is change in the way we represent ourselves, seems to fall in with this group of claims. It is a higher-order description of the way an intuiting subject must be if it is to intuit itself. But if this is right, then Lambert's objection applies. The succession of self-representations would have to be a temporal succession, and time is more than empirically real.