Time-determination
In the Refutation of Idealism, Kant puts forward the claim that every time-determination requires a permanent, without providing any explanation or justification. Such an explanation can, however, be found in the First Analogy where it is argued that permanence is a necessary presupposition of the possibility of experience. Kant's argument starts with the claim that time cannot be perceived. He then argues that owing to this "there must be found in the objects of perception, that is, in the appearances, the substratum which represents time in general" (B225). But why do we need a representation of time in the field of appearances? While it is the case that only in time can we represent relations of time, it need not be the case that the instantiated time-relations in any way represent or exhibit their temporality. Accordingly, no representation of time in perception need be given. The whole problem that we cannot perceive time would be solved if there were a representation of time in perception as we could then perceive, not time itself, but its representation. Instead, we have to use a representation of time that is produced by the imagination. That time cannot be perceived is irrelevant to occurrences being related in time as it only matters with regard to the mechanism by which we have to determine these time-relations. Hence the fact that time cannot be perceived has only epistemological implications and the metaphysical status of time-relations is not affected by it. Because we do not perceive time, we cannot identify occurrences by their objective temporal places, by reference to perceived time. Instead, we have to identify them by reference to other occurrences. While reference to other occurrences is required, insofar as one must determine what coexists, precedes or succeeds, there does not seem to be any need for a permanent.
Can anyone think of a way of saving Kant's claim?
Can anyone think of a way of saving Kant's claim?
posted by Bader at 2:41 pm
9 Comments:
Ralf, I was wondering if you would clarify your concern. You ask:
[W]hy do we need a representation of time in the field of appearances?
Isn’t this a premise (an unsupported premise) of the Transcendental Aesthetic? Do you mean to question it, or is your concern something else?
I have other questions, but some of them may be answered by your response to this one.
Hi Andrew,
Isn’t this a premise (an unsupported premise) of the Transcendental Aesthetic? Do you mean to question it, or is your concern something else?
I did not think that this was a premise (whether supported or not) of the Transcendental Aesthetic. As far as I know, Kant claims that all appearances are in time, that time is the form of inner sense. But from this it does not follow that we have a representation of time in the field of appearances; any such claim rather than being a premise is something that he might be taken to argue for in the First Analogy. However, I cannot see how to make his argument work. I would rather be inclined to say that the field of appearances is a temporal field. Appearances are represented as being in time, rather than there being in appearances something that represents time.
I see. You mean to question Kant’s opening moves in the First Analogy. It’s one thing to say that (i) we represent time (and that objects must appear in time); it’s another thing to say that (ii) there appears to be something among the appearances that itself “represents” time. Right? When you speak of “a representation of time in the field of appearances,” this can look a little like (i).
I’ll think some more, then, about your original post. As you no doubt know, you’re in good company about being puzzled by Kant’s argument here. A lot of commentators have given up on the opening passages of the First Analogy and turn to A188/B231, thinking that a better argument is to be found there. Personally, I’m inclined to think that, for better or worse, Kant’s real argument is in the first two paragraphs. I have some stuff on this that I haven't looked at in a while. This will be a good opportunity to reexamine it.
More soon….
I see. You mean to question Kant’s opening moves in the First Analogy. It’s one thing to say that (i) we represent time (and that objects must appear in time); it’s another thing to say that (ii) there appears to be something among the appearances that itself “represents” time. Right? When you speak of “a representation of time in the field of appearances,” this can look a little like (i).
That's absolutely right. I am concerned with claim (ii).
Personally, I’m inclined to think that, for better or worse, Kant’s real argument is in the first two paragraphs.
I agree with you that the beginning of the First Analogy is very important and generally think that his argument is quite instructive. I just do not think that it helps us much with Kant's claims in the Refutation. So, my worry is partly a general puzzlement about the First Analogy, but mostly about the extent to which it can help us with the argument in the Refutation.
More soon….
Looking forward to it.
I wonder whether Kant thinks roughly this. Time, like space, is structural or formal. Consequently, we can have no sensation of time, which would be required for direct perception. But it is a doctrine from the Aesthetic that we do, in fact, (sensibly) connect to time in experience (i.e., it is no mere delusion). So Kant concludes that there must be a proxy for time in the content of experience, something to which we connect cognitively: something non-formal that bears the relevant property of time, i.e., the property of being permanent. The only possible thing is substance. Since substances are not formal or structural, we can have sensations of them, and thus perceive them.
This reading needs further defense than I shall provide here; moreover, I am doubtful of the success of the argument. But it at least addresses your initial question. You write further, however:
While it is the case that only in time can we represent relations of time, it need not be the case that the instantiated time-relations in any way represent or exhibit their temporality.
I don’t quite understand the worry; but I think that Kant’s response would be that to explain our experience of time—in particular, our representation of its all-encompassing-ness (I don’t say “permanence,” since we would not say that time is permanent)—we must find something non-formal among objects of experience that instantiates the corresponding temporal property.
But it is a doctrine from the Aesthetic that we do, in fact, (sensibly) connect to time in experience (i.e., it is no mere delusion).
Could you elaborate on what you mean by 'connecting to time in experience'? According to the Aesthetic, our experience is structured in time, but this is because the time is the form of inner sense and has been imposed on the manifold of intuition. It is not because time is somehow sensibly perceived by us. Temporality is our contribution, rather than being something known to us as a result of the receptivity of sense.
So Kant concludes that there must be a proxy for time in the content of experience, something to which we connect cognitively: something non-formal that bears the relevant property of time, i.e., the property of being permanent.
This is slightly problematic since further below you say that time is not permanent.
The only possible thing is substance. Since substances are not formal or structural, we can have sensations of them, and thus perceive them.
I have a few worries about this claim. If we want a proxy for time, then it is not sufficient that we simply have sensations of the substances, that we perceive them. Instead, we must somehow perceive their temporality, their permanence. I find this aspect rather problematic. Secondly, Kant somehow seems to think that we need this permanent for time-determination. So, the need for a proxy on the basis of ensuring that time is not a mere delusion, does not yet provide us with the need of the proxy for time-determination.
I don’t quite understand the worry; but I think that Kant’s response would be that to explain our experience of time—in particular, our representation of its all-encompassing-ness (I don’t say “permanence,” since we would not say that time is permanent)—we must find something non-formal among objects of experience that instantiates the corresponding temporal property.
Why can we not just explain the all-encompassingness by saying that it is a form of intuition? Space is also all-encompassing - does this mean we need a spatial analogue of 'substance', not permanent but spatially all-encompassing?
Could you elaborate on what you mean by 'connecting to time in experience'? According to the Aesthetic, our experience is structured in time, but this is because the time is the form of inner sense and has been imposed on the manifold of intuition. It is not because time is somehow sensibly perceived by us. Temporality is our contribution, rather than being something known to us as a result of the receptivity of sense.
Well, as I construe transcendental idealism, this isn’t right. We have pure intuitions of space and time. Through intuitions, we relate immediately to objects. Pure intuitions are intuitions. We connect to space and time through them.
I would not expect to convince you of this. Let me just note the following. First, you can disagree entirely with me on the idealism and still accept my characterization of the argument of the First Analogy if you are willing to hold that Kant’s idealism is separable from his argument in the First Analogy. This latter point is exactly Paul Guyer’s view. I think it is also P.F. Strawson’s view. Of course, the only good reason to do this is if my proposal does a good job of accounting for Kant’s argument, which is another matter. But since as it stands you regard Kant’s argument to fail pretty badly, perhaps this proposal is no worse off for you.
This is slightly problematic since further below you say that time is not permanent.
Ah. I said:
So Kant concludes that there must be a proxy for time in the content of experience, something to which we connect cognitively: something non-formal that bears the relevant property of time, i.e., the property of being permanent.
I did not intend the “of” to imply possession. It would have been clearer had I said “relevant temporal property.”
I have a few worries about this claim. If we want a proxy for time, then it is not sufficient that we simply have sensations of the substances, that we perceive them. Instead, we must somehow perceive their temporality, their permanence. I find this aspect rather problematic.
I concur entirely with the requirements that you mention. You don’t explain what the problem is, however. As I see it—although this claim needs defense that I won’t try to provide here—we learn over the course of the Transcendental Deduction that pure intuitions structure empirical intuitions just so that we perceive objects’ categoriality. This includes their substantivality (which would include their being determinations of a substance), which for Kant entails permanence (A144/B183). So we perceive permanence.
Secondly, Kant somehow seems to think that we need this permanent for time-determination. So, the need for a proxy on the basis of ensuring that time is not a mere delusion, does not yet provide us with the need of the proxy for time-determination.
I worry that this sort of point has become an article of faith in Kant interpretation. What leads you to say that Kant thinks that we need the permanent for time-determination? And what exactly do you mean by “time-determination”? Also, it is not my position that we need a proxy for time to insure that time is no mere delusion. I take it for granted (that Kant took it for granted) that time is no mere delusion. The question is: what conditions must be in place for this assumption to hold?
Why can we not just explain the all-encompassingness by saying that it is a form of intuition? Space is also all-encompassing - does this mean we need a spatial analogue of 'substance', not permanent but spatially all-encompassing?
As I take you to understand a form of intuition, this would suffice to account for the phenomenology of time. But I do not see the First Analogy as concerned only with the phenomenology. Kant does not simply assume that it seems to us as though everything is in time; he assumes that everything (that we can experience!) is in time and that we connect to its temporality. His question is: what makes this possible?
As for space: it would be no objection to anything that I have said if Kant did not provide for a spatial analogue. He is free to elaborate on what conditions on experience he wants. But the fact is that in the Third Analogy he does something along these lines. I’m not going to touch that argument, partly because I can’t make it work. But here is what he says:
The word “community” is ambiguous in our language, and can mean either communion or commercium. We use it in the latter sense, as a dynamical community, without which even the local community (communion spatii) could never be empirically cognized (A213/B260).
Kant thinks that we would not experience objects in space did we not experience them (or the substances of which they are determinations) to be interacting. But we cannot experience empty space; so this as much a condition on our experience of space.
Hi Andrew, thanks for the comments.
Well, as I construe transcendental idealism, this isn’t right. We have pure intuitions of space and time. Through intuitions, we relate immediately to objects. Pure intuitions are intuitions. We connect to space and time through them.
I do not agree with this. It is much better to characterise space and time as forms of intuitions, rather than as intuitions themselves. Though we can have pure intuitions of spaces and times, in the sense that we can construct representations of them, we cannot have pure intuition of Space and Time, i.e. the all-encompassing spatial and temporal frameworks that Allison classifies as pre-intuited: “This one unbounded space is ‘pre-intuited’ in the sense that it is given together with every determinate space as the latter’s ‘horizon’, without itself being actually intuited as an object.” (Allison, p. 113). It is these all-encompassing frameworks that are at issue. Thus any representation of space and time is (i) not a representation of Space and Time, and (ii) is not an object known to us as a result of our receptivity, but a result of our spontaneous construction.
if you are willing to hold that Kant’s idealism is separable from his argument in the First Analogy. This latter point is exactly Paul Guyer’s view. I think it is also P.F. Strawson’s view.
In general I agree with Allison that the Analogies (and most of Kant's other arguments) cannot be successfully separated from transcendental idealism.
I did not intend the “of” to imply possession. It would have been clearer had I said “relevant temporal property.”
I see.
we learn over the course of the Transcendental Deduction that pure intuitions structure empirical intuitions just so that we perceive objects’ categoriality. This includes their substantivality (which would include their being determinations of a substance), which for Kant entails permanence (A144/B183). So we perceive permanence.
I actually think that there might be a clash between the argument in the B-Deduction and the claim in the First Analogy. In the B-Deduction he says that we require a representation of time that is produced by the imagination in order for there to be perception in the first place. He then argues that the synthesis required for the construction of this representation of time involves the categories which implies that perception itself is subject to the categories. If time is somehow perceived as required by the Refutation and the First Analogy, then this argument in the B-Deduction no longer works since we no longer need to construct this representation of time, but can simply acquire it by means of perception.
Moreover, I do not think that it is adequate to say that we perceive objects' categoriality. We do so with respect to the constitutive, but not the regulative categories. Kant is pretty much a Humean about our perception of causation and substance. If we were to perceive the categoriality of objects which includes subsantivality, then it seems that there would be no real need for the Analogies anymore.
I worry that this sort of point has become an article of faith in Kant interpretation. What leads you to say that Kant thinks that we need the permanent for time-determination?
In the Refutation Kant says that: "All determination in time presupposes something permanent in perception." (B275)
And what exactly do you mean by “time-determination”?
What exactly 'time-determination' means and what it involves is a difficult issue. Roughly, I would say that to determine something in time is to give it a definite place in time, to fix its temporal location and duration.
As I take you to understand a form of intuition, this would suffice to account for the phenomenology of time. But I do not see the First Analogy as concerned only with the phenomenology. Kant does not simply assume that it seems to us as though everything is in time; he assumes that everything (that we can experience!) is in time and that we connect to its temporality. His question is: what makes this possible?
Phenomenology seems to be sufficient if the ideality of time and space is accepted. As Kant says in the Fourth Paralogism: "space is itself nothing but mere representation, and therefore nothing in it can count as real save only what is represented in it" (A374). The same applies to time. All it is for something to exist in time is for it to be represented in time. No appeal to permanence is required here to explain how this possible; we simply need to appeal to the ideality of space and time.
I do not agree with this. It is much better to characterise space and time as forms of intuitions, rather than as intuitions themselves.
Actually, I don’t dispute this, although we’ll disagree about what “forms of intuition” amounts to. My point is that we represent (and connect to cognitively) space and time through (pure) intuitions. I realize, however, that you don’t agree with this, either.
You say that we can have pure intuitions of spaces and times, but not of Space and Time. I accept that some such distinction is in order when interpreting Kant, but plainly Kant thinks that Space and Time are at issue in the Aesthetic (e.g., “Different times are only parts of one and the same time” (A31-2/B47; my emphasis)), and it is of S/space and T/time so construed that he argues we have pure intuitions (e.g., “[t]he original representation of space is an a priori intuition” (B40)).
In general I agree with Allison that the Analogies (and most of Kant's other arguments) cannot be successfully separated from transcendental idealism.
Here, too, I agree. I think that the idealism—properly understood—is essential.
Moreover, I do not think that it is adequate to say that we perceive objects' categoriality. We do so with respect to the constitutive, but not the regulative categories. Kant is pretty much a Humean about our perception of causation and substance. If we were to perceive the categoriality of objects which includes subsantivality, then it seems that there would be no real need for the Analogies anymore.
I don’t see why you make this last claim—certainly not with the Second and Third Analogies. The Deduction does not establish that every event has a cause or that substances are in thoroughgoing interaction. I admit that with the First Analogy this may seem less obvious. But as I read the First Analogy, its claim is not simply that there are substances (which are permanent by definition) but that everything either is, or is a determination of, a substance, and this is more than the Deduction could reasonably be said to establish.
As for what you say about Kant’s Humean tendencies, I deny it. Roughly, he agrees with Hume’s conclusions on the assumption of Hume’s limited conception of what experience is. But he contests just this conception. At B291 he says that we are given, through intuition, that which corresponds to the concept of substance, i.e., “something that persists.” He also implies (although I don’t like the way he puts this) that we “exhibit alteration as the intuition corresponding to the concept of causality.”
In the Refutation Kant says that: "All determination in time presupposes something permanent in perception." (B275)
What exactly 'time-determination' means and what it involves is a difficult issue. Roughly, I would say that to determine something in time is to give it a definite place in time, to fix its temporal location and duration.
I see that my worry was very poorly phrased, so please allow me to rephrase it, especially now that I have your answer to what you take time determination to be. I think many understand time determination to be roughly what you say it is. It is determining that x occurred before or after y or at the same time; perhaps it is also dating a particular state or event. Often tied to this is a gloss on what the Analogies are doing: giving us rules to determine the times of states and events. I believe that there is no reason to think that the First Analogy is at all concerned with this. I suspect that you would be sympathetic to this claim. I also think, though I won’t speculate about your beliefs here, that there is no reason to think that the Second or Third Analogy is concerned with time determination, so construed. The Analogies, like all of the Principles’ sections, are trying to establish certain synthetic a priori claims; they are not providing us rules to follow to order events in time.
I read the passage at B275 that you cite as follows. Yes, to experience time, we must experience something permanent—this was the argument of the First Analogy. It follows trivially that we could not cognize (truly) things to be in various locations in time if we failed to experience something permanent. I think that this is Kant’s point at B275. I don’t think that he intends anything more than this.
Phenomenology seems to be sufficient if the ideality of time and space is accepted.
This is another reason why I’m disinclined toward more idealistic interpretations of transcendental idealism. I don’t think that Kant can be satisfied with this.
Post a Comment
<< Home