Cognizing Coexistence: Perceptions and their Synthetic Unity in Kant’s 3rd Analogy

§1 The Order-Indifference of Our Perceptions

In the 3^rd Analogy, Kant claims that I cognize that things ‘exist at one and the same time’ by combining my successive perceptions of them in an order-indifferent way (A211/B258). Or:

Order-Indifference: To cognize that states of substances that I successively perceive coexist, I must synthesize my perceptions according to the rule that I can perceive the states in any order. (Relatedly, I will say a sequence of perceptions is ‘order-indifferent’ if they have been synthesized in this fashion.)⁵

Kant’s claim is widely acknowledged in the literature on the 3^rd Analogy, as is the parallel claim from the 2^nd Analogy. So, I take the claim to be relatively uncontroversial. In what follows, I will show that existing interpretations of the role that this claim plays in Kant’s argument fail to do it justice, before turning to give an alternative interpretation. In this section, I want to explain what Kant means by ‘perception’ (‘Wahrnehmung’).

Kant seems to use the term ‘perception’ in different senses at different points in the Analogies. Sometimes, as Kant uses the term, it picks out something that does not yet involve cognition of any temporal determinations of what one perceives—for instance, when Kant writes that ‘through the mere perception the objective relation of the appearances that are succeeding one another remains undetermined’ (B234). But sometimes it does involve cognition of temporal determinations—for instance, when Kant writes that ‘[a]lteration can be perceived only in substances, and arising or perishing per se cannot be a possible perception unless it concerns merely a determination of that which persists’ (A188/B231; B225, B233, and A198/B243). On this usage, I can perceive alterations of a substance, and alteration is an objective relation of the appearances that are succeeding. This apparently conflicts with the prior claim. What, then, does Kant mean by ‘perception’?

Recent work by Clinton Tolley helps answer this question. According to Tolley, Kant’s argument in the Principles is developmental in that it progresses from intuitions (Axioms) to perceptions (Anticipations) to experience (Analogies) (Tolley 2019). He has argued that, for Kant, perception in the first instance is consciousness of either a sensation or an intuition (A119–20). So, perceptions add consciousness to intuition—in particular, Tolley argues, consciousness of the manifold contained in the intuition (B202–3; Tolley 2016: 86; 2018: §3). Tolley also thinks that perception is not yet cognition of an object (not yet experience, in Kant’s technical sense of that term), but merely consciousness of a representation—to arrive at cognition of an object requires further cognitive processing.⁶

Recognizing, with Tolley, that the Principles develop from perception to experience helps us understand Kant’s shifting use of the term ‘perception.’ In the Analogies, Kant’s explicit topic is experience (they are the ‘Analogies of Experience’) rather than perception. Experience, Kant says, is the determination of an object through a synthesis of perceptions (B219). So, Kant’s topic in the Analogies is a synthesis of perceptions: specifically, the synthesis that determines the temporality of the objects perceived. So synthesized, what we perceive is an object, and our perception is cognition of it. Kant evidently thinks it appropriate to refer both to what is synthesized and to the synthesis as perceptions. We can, following Kant, call the first sense of perception, which involves no objective time determination, ‘mere perception’; we can, adding to Kant’s terminology, call the synthesis of perceptions, which involves objective time determination, ‘enriched perceptions.’ Enriched perceptions make up experience.

Using this terminological distinction, we can make sense of Kant’s apparently conflicting claims about perception. ‘Mere perception’ does not determine the objective relation of successively perceived appearances; in the sense of ‘mere perception,’ I do not perceive alterations. But, in the sense of enriched perception, I do perceive alteration. Conforming to the enriched side of Kant’s usage, unless otherwise noted in what follows when I talk of ‘perception’ I will mean perception as synthesized (enriched perception).

This essay at disambiguation naturally raises the question: How does the synthesis of perceptions relate to perception, enriched perception to mere perception? Is the synthesis of perceptions necessary for (mere) perception or logically posterior to it?

Tolley takes Kant to hold that the capacity to merely perceive and its acts are independent of the capacity for enriched perception; he also holds that the capacity for mere perception and its acts are independent of the activity of the understanding (Tolley 2013). So, on his view, a being could merely perceive appearances without being able to perceive appearances in an enriched way, and I can merely perceive appearances in a way that involves no objective time determinations. To arrive at objective time-determinations, I must in some way augment my acts of mere perception, perhaps by interpreting the acts of mere perception through an activity of the understanding; the augmentation in question brings into play another capacity (or perhaps many different capacities), the possession of which is not implied by the possession of the capacity for mere perception.

Tolley might be right—but the fact that the Principles develop does not show he is. For we must ask how Kant thinks about the development. One might take Tolley’s view, or one might instead argue that Kant builds up an essentially inseparable sequence of capacities, showing developmentally that each capacity depends on the one that follows it.⁷ On this view, to which we shall return in §3, mere perception is dependent for its possibility upon enriched perception.

The readings of the 3rd Analogy that I consider in §2 all agree with Tolley in taking it that mere perception is not dependent for its possibility on enriched perception. I will argue that these interpretations make the argument of the 3rd Analogy mysterious and then suggest (§3) that we can make the argument of the 3rd Analogy unmysterious if we adopt the view that mere perceptions depend upon the synthetic unity that connects those perceptions with one another (enriched perceptions).

§2 A Puzzle about Order-Indifference

In this section, I will consider three interpretations of the role played by Order-Indifference and show that all of them face significant problems. In §3, I will suggest a way to avoid these problems.

§3 Perceiving Coexistence

So far, we have considered three views of the role of Order-Indifference in the 3^rd Analogy. One view fails to acknowledge Kant’s claim that Order-Indifference is the way in which we cognize that things coexist with one another. The other two views attempt to explain the claim but fail in the attempt. The criterial theory claims that Kant invokes Order-Indifference as our criterion for determining whether things coexist, and the interpreting theory claims that Order-Indifference is the act by which we interpret the sequence of perceptions and thereby determine that the objects perceived coexist. Both views fail for what is at root the same reason: the data available prior to cognizing coexistence is insufficient to justify the claim or interpretation that what we perceive coexists.

For the criterial theory, Order-Indifference is meant to serve as evidence by means of which we cognize that things coexist. But we cannot use Order-Indifference as evidence for the claim that things coexist because we can only cognize that our perceptions are order-indifferent if we already know that the things we perceive coexist. For the interpreting theory, Order-Indifference is meant to be our interpretation of the data available to us from mere perception. But we cannot justify combining our perceptions in an order-indifferent way, and therein interpret what we perceive to coexist, on the basis of what is made available by mere perception. How might we avoid the problem that plagues both the criterial and interpreting theories?

The solution, I suggest, is simple, though it has far-reaching consequences. In the B-edition version of the principle of the 3^rd Analogy, Kant says that substances ‘can be perceived in space as simultaneous’ (B256). According to the interpretations we have considered so far, Kant means that substances can be perceived as simultaneous only after the understanding acts on the mere perceptions of what is in space. These views fail: mere perceptions do not provide enough for the understanding to produce the enriched perception of coexistence rather than the enriched perception of objective succession. On that diagnosis, the solution must be that Kant thinks that, in the fundamental case, an act of perception is an act of enriched rather than mere perception.

Let me clarify what I mean. In §1, I followed Tolley’s suggestion that what explains Kant’s ambiguous use of the term ‘perception’ is that the course of the Principles is developmental. But I also noted that to say that the Principles are developmental is not yet to say how the stages of that development relate to one another. On one view, the view one finds in Tolley, experience arises from acts of a capacity of (mere) perception that are augmented by the acts of other capacities, eventually bringing in the understanding. Key to this picture is that the capacity to (merely) perceive is not dependent for its activity on the subsequent capacities—one could have the capacity to (merely) perceive regardless of whether one had the further capacities. But one might instead think that the development is a development of different, increasingly richer aspects of what is involved in the fundamental case of an act of perception, such that the capacity to perceive is dependent upon the other capacities and its defining act is an act that involves the activity of those other capacities (including, most significantly for our purposes, the understanding).

Both the criterial and interpreting theories hold that, for Kant, the capacity to merely perceive is separable from the capacity to perceive in an enriched sense, such that we arrive at acts of the second capacity only through acts of the understanding on the results of the acts of the first capacity: I must determine that the things I perceive coexist on the basis of first perceiving them in a way that involves no objective time determination. In the face of the problem with this view, the natural suggestion is that there is no initial stage prior to the temporal determination of coexistence. The capacity to merely perceive is not a self-standing capacity: it is a partial description of the capacity to perceive in an enriched sense.

Does my view entail that, for Kant, there are no acts of mere perception? No, it does not, and that is why I refer above to ‘the fundamental case’ of an act of the capacity to perceive. For perhaps I can merely perceive—that is, perhaps it is possible that I engage in acts of mere perception, or acts which involve no objective time determination of the successive manifolds that I perceive. I may, for instance, associate my perception of the moon with my perception of the sun in a way which involves no determination of their coexistence (or succession). The point I am making does not exclude the possibility of that. My point, rather, is that such acts are derivative or parasitic upon the more fundamental or capacity-defining act of enriched perception. I can, perhaps, merely perceive the sun and the moon—but only if I can determine that they coexist. Such, anyway, is the view that I am attributing to Kant.¹⁷

On the view I am suggesting, when I perceive the earth (before perceiving the moon), I am cognizant that the earth exists at a time at which other things, like the moon, might exist. In first perceiving the earth, I do not thereby know what other things coexist with it—but I can know, on the basis of what I go on to perceive, what coexists with it. According to Kant, that is possible only if, in perceiving the earth, I am cognizant that I might have perceived whatever else exists at the time at which I perceived the earth (had I been positioned differently—for instance, had my head been turned). In other words, in perceiving the earth, I am in a position to judge that it coexists with an indefinite number of other things, and the capacity I exercise in perceiving the earth is a capacity to cognize what other things coexist with the earth.¹⁸ (Kant’s argument in the rest of the 3^rd Analogy is meant to show that the capacity to perceive is a capacity to cognize coexistence only if the things that coexist causally interact with one another.)

I have said that my capacity to perceive is the capacity to perceive coexistence: I can perceive the earth only because I can perceive other substances as coexisting with it. But, one may wonder, how can I determine in a given case whether the states I have successively perceived coexist with or succeed one another? For surely one of the main items on Kant’s agenda in the 2^nd and 3^rd Analogies is to explain the difference between successive perceptions of a house and successive perceptions of a moving ship. How does my appeal to the capacity to perceive coexistence help explain the difference between those two cases?

There is clearly a sense in which Kant is trying to explain how I can know of one set of successive perceptions that they are perceptions of a change and of another that they are perceptions of coexisting states of substances. There are different interpretations of what this might mean, however. First, one might think that Kant means to explain how, given some set of successive perceptions, I can subsequently determine whether they are perceptions of change or of coexistence. Second, one might think that Kant starts from the claim that I can perceive that things are changing and that things coexist, and is asking how I can do that. I am arguing for the second interpretation. But, if the second interpretation is right, then the objection falls to the ground: Kant takes for granted that I can perceive that things coexist with one another, and specifies conditions for the possibility of that. There is a perceptually available difference between successive perceptions of a house and successive perceptions of a moving ship. Kant is giving an account of the conditions of possibility of that difference.

The criterial and interpreting theories both hold that Kant is trying to explain the basis on which I determine that the states coexist in the one case and the basis on which I determine that they succeed one another in the other case. That’s not quite right, on my view, for that assumes that the only answer Kant can offer appeals to features of what is available to us in mere perception. (From this vantage point, it does not matter whether those features offer an epistemological, psychological, metaphysical or putatively transcendental answer to the ‘On the basis of what?’ question.) I do not believe that Kant asks this ‘On the basis of what?’ question. And, I think, if one were to press him to explain on what basis I perceive the states of the house to coexist while I perceive the states of the moving ship to succeed one another, his answer would be: because the states of the house do coexist, while the states of the moving ship do not. At this point, one would be well within one’s rights to ask Kant to defend that answer (against, e.g., an appeal to subjective habits of association), and that is where (on the criterial and interpreting theories) the appeal to features of mere perception would come in. But my suggestion is that Kant is not trying to answer this question—he instead articulates the conditions of possibility of the representations of temporal determination (of coexistence and succession) which the ‘On the basis of what?’ question takes for granted. And his argument in the 3^rd Analogy is that the perceptual representation of the coexistence of states of substances depends upon the objective determination that those substances causally interact with one another.¹⁹ (Again, I have not sought to defend his argument in the 3^rd Analogy; I have instead tried to explain the starting point for his argument.)

Before connecting my interpretation to larger debates about Kant’s project in the Critique, I want to clarify it in two further respects. First, note that I mean ‘perceive coexistence’ to imply knowingly perceiving coexistence. All agree that Kant is talking about how I cognize that things coexist on the basis of perception. To invoke a kind of perception that involves perceiving coexistence unknowingly would just reraise the problem that besets the criterial and interpreting theories: on the basis of what do I come to know that the things I perceive coexist?

Further, to say that I knowingly perceive coexistence means that perceiving puts me in the position to judge that states of distinct substances coexist. This does not mean that perceptions are judgments, or even that perceiving involves judging: the way the categories inform perception does not turn perception into a judgment. I am not, in knowingly perceiving the moon and the earth as coexisting, exercising the capacity to judge, since I am not judging that they coexist—rather, I am exercising the capacity to perceive, a capacity that puts me in the position to make judgments about the moon and the sun that involve the (implicit or explicit) recognition that they coexist.²⁰

My solution to the puzzle about Order-Indifference involves attributing a version of what is typically called ‘conceptualism’ to Kant. ‘Conceptualism’ as applied to Kant is, roughly, the view that the understanding operates in and not just on sensibility. As I noted in my introduction, the primary philosophical motivation for this view is that it is required to enable judgment to have both content (or be about the world) and to be sensitive to reason. The idea is that if sensibility is independent of the understanding, then acts of sensibility will be blind and acts of the understanding will be empty. I will follow Jim Conant in calling the view the ‘transformative conception of human mindedness,’ with the idea being that the operation of the understanding in sensibility transforms the very nature of sensibility, so that it is a distinctively rational form of sensibility.²¹ The term ‘transformative’ better highlights the philosophical aspirations of the view, as I understand it: the idea is not to reduce sensibility to the understanding (intuitions to concepts) but to articulate some kind of inseparable unity of the two.

I will not rehearse the voluminous debate about whether Kant’s view is transformative or non-conceptualist, as my topic is the 3^rd Analogy. But my point is relevant to the debate because my solution to the puzzle about Order-Indifference attributes a transformative view to Kant. Our capacity to merely perceive, once fully in view, is the capacity to perceive coexistence because it presupposes the rule, supplied by the understanding, to combine successive perceptions of coexisting things in an order-indifferent way. (Note that my topic is Kant’s account of perception (in the 3^rd Analogy) and not intuition. Typically, defenders of the transformative view of Kant target intuition. While what I say is in the spirit of that view, it is logically independent of it: what I say is consistent with the view that intuition is independent of the activity of the understanding. This should make my view more tolerable to defenders of non-conceptualism who draw a sharp distinction between intuition and perception.)

It may surprise that I group Longuenesse with the non-conceptualists. Certainly, she embraces a central aspect of the transformative view when she defends the claim that perceptions are as such informed by the relational categories of the Analogies. Thus, she would deny the non-conceptualist claim that I can perceive in a way that is independent of the understanding. But she denies that mere perception puts me in a position to judge that the objects I perceive fall under any of the relational categories. For that judgment, I must first interpret what I perceive.²² As I argued in §2.3, it is unclear how we can become cognizant of the coexistence of what we perceive if we proceed in this way. And that is to say that, with respect to the objective time determination of coexistence, perception is blind and the understanding is empty—precisely because with respect to that determination the understanding acts on and not in mere perception. Thus, while Longuenesse’s view is transformative overall, her reading of the 3^rd Analogy betrays the philosophical aspiration of the transformative interpretation.²³

I am proposing that we can solve the puzzle about Order-Indifference if we credit Kant with the view that we perceive coexistence, such that the fundamental act of perception is an act of enriched perception. I have said that this is a natural suggestion, given the problem that besets the criterial and interpreting theories. I have not offered a decisive argument showing that it is the only way to explain Order-Indifference. Perhaps one can explain Order-Indifference in a different way. Certainly, people attracted to non-conceptualism, or criticisms of the transformative view, are not likely to be persuaded to abandon their view by any arguments about the 3^rd Analogy. I contend, however, that my argument shows that prominent versions of non-conceptualist readings of the 3^rd Analogy fail. I thus invite non-conceptualists to find other solutions to the puzzle about Order-Indifference.

§4 Conclusion

Kant opens the 3^rd Analogy by claiming that we can cognize coexistence by synthesizing our successive perceptions according to the rule that the order of the succession is objectively insignificant. One immediately wants to know: how do the successive perceptions relate to the synthesis of those perceptions? A common way of reading the Analogies takes it that the successive perceptions are independent of and prior to their synthesis, such that the problem posed in the Analogies is a ‘problem of time-determination’ (Watkins 2005: 188): how can I cognize the temporal relation between things on the basis of perceptions that are always successive?

I have argued that the influential versions of this strategy put forth by Strawson, Watkins, Longuenesse and Allison all face problems. The interpretation put forth by Watkins fails to do justice to Kant’s claim that the synthesis of our perceptions is the way in which we cognize that things coexist. And the interpretations put forth by Strawson, Longuenesse, and Allison fail because the successive perceptions do not provide enough material to license either a conclusion or an interpretation of coexistence.

This suggests that, for the 3^rd Analogy, we need to understand the relation between the successive perceptions and the synthesis of perceptions differently. Rather than construing the first as independent of the second, we should instead see the second as already operative in the first. Our perception of an object already involves the rule for combining that perception with perceptions of other objects, the rule that enables us to cognize the coexistence of those objects. Thus, upon successively perceiving coexisting things, we are in a position to judge that those things coexist without needing to further work up the successive perceptions. Perhaps one can understand how the successive perceptions relate to their synthesis differently. In that case, more work is necessary to explain that different relation. However, if my reading convinces, then the 3^rd Analogy gives further support to transformative (or conceptualist) readings of the Critique.²⁴