STEFANIE GRÜNE | Blinde Anschauung. Die Rolle von Begriffen in Kants Theorie sinnlicher Synthesis | Klostermann 2009
By Thomas Land
The doctrine of synthesis Kant puts forth in the Transcendental Analytic of the Critique of Pure Reason is notoriously difficult to understand. One problem is that it appears to conflict with the teaching advanced in the Transcendental Aesthetic. In particular, while the Aesthetic can seem to say that intuitions are independent of any contribution of the understanding, the upshot of the doctrine of synthesis appears to be that intuitions depend on the application of concepts in judgement. The recent debate between conceptualist and nonconceptualist interpretations of Kant’s theory of intuition illustrates this difficulty: Whereas nonconceptualists argue that the apparent teaching of the Aesthetic represents Kant’s considered view, conceptualists deny this and claim that intuitions depend not just on sensibility, but also on the application of concepts in judgement.
The thesis of Stefanie Grüne’s book, Blinde Anschauung. Die Rolle von Begriffen in Kants Theorie der sinnlichen Synthesis, is that neither side is right (see also précis here). Pace the nonconceptualist position, Grüne argues, intuitions do depend on a contribution of the understanding. However, in contrast to the standard conceptualist position, this contribution does not consist in the application of concepts in judgement. Rather, it consists in the act of sensible synthesis, which is distinct from, and independent of, judgement.
Central to the position Grüne articulates is a distinction between two ways in which, according to Grüne’s reading of Kant, one may be said to possess a concept: One may possess a concept F as a clear concept or one may possess F as an obscure concept. Both are ways of possessing the same concept, F, but they are distinct because they consist in the possession of different capacities. According to Grüne, to possess F as a clear concept is to have the capacity to apply F in judgements of the form Fa (in the basic case). By contrast, to possess F as an obscure concept is to possess the capacity for sensible synthesis in accordance with F, where this comes to roughly the following: Let an F have essential properties G, H, and I. Then sensible synthesis in accordance with F is the act of grasping sensibly given instances of G, H, and I as constituting a unity. Thus, if the concept of a lemon is the concept of something that is round, yellow, and sour-tasting, then to sensibly synthesize in accordance with the concept of a lemon is to grasp instances of roundness, yellowness and sourness that are jointly given in sensibility as constituting a unity.
Importantly, this kind of grasp does not require that one recognize these instances as instances of roundness, yellowness or sourness; or as an instance of ‘lemon’, for that matter. All that is required is that one experience sensibly given instances of these concepts, which one may not be able to classify in any particular way, as belonging together. According to Grüne, whether or not one experiences these features as belonging together makes a difference to the phenomenal character of the experience.
The general shape of the position Grüne defends seems to me exactly right. Specifically, I agree that, for Kant, intuitions depend on sensible synthesis, rather than being (i) exclusively a matter of sensibility, or (ii) depending on the application of concepts in judgement. I agree, further, that sensible synthesis depends on concepts and therefore on the understanding. However, the way in which Grüne spells out what this claim comes to seems to me mistaken in several respects. Unfortunately, I will not be able here to discuss all relevant aspects of Grüne’s nuanced and well-argued position. Instead I will raise two objections, which address points that seem to me central to Grüne’s position. Both objections focus on the categories. I will thus give no consideration here to Grüne’s detailed and illuminating discussion of the role of empirical concepts in sensible synthesis. The first objection is that Grüne’s account of what it is to have the capacity to apply categories in judgement is mistaken. The second objection is that her claim that one can possess the capacity for sensible synthesis in accordance with the categories without possessing the capacity to apply categories in judgement (and, possibly, without possessing the capacity for judgement at all; see p. 202n.17) is false.
I now turn to my first objection. Two central features of Kant’s account of the categories are the following:
(i) the categories originate purely in the understanding
(ii) the categories are concepts of the “synthetic unity of the manifold in intuition in general” (A79/B105).
I will argue that consideration of these two features shows that the primary application of the categories is in judgement, but does not require possession of the categories as clear concepts.
According to Grüne’s Kant, one can possess a concept in one of three ways: as an obscure, clear, or distinct concept (see p. 71ff). Since possessing a concept F as a distinct concept can be regarded as a particular case of possessing F as a clear concept, obscure and clear concept-possession are the only two relevant options. Clearly, possession of the categories as obscure concepts (as Grüne understands this notion) does not account for the application of categories in judgement. However, the capacity for applying a category in judgement in the way I have in mind does not amount to possession of the category as a clear concept, either. That is, the primary way in which a category is applied in judgement, according to Kant, is not in the way in which F is applied in Fa. In more Kantian terms, the primary way in which the categories are applied in judgement is not by figuring among the concepts that constitute the matter of a judgement. For instance, in a judgement of the form ‘S is P’, a category typically figures neither as subject-concept nor as predicate-concept. If this is right, then neither Grüne’s obscure concept possession nor her clear concept possession capture what it is, for Kant, to possess the categories.
What, then, does possession of the categories come to, on my reading of Kant? I will argue that for Kant the categories pertain to the form of a judgement, whereas Grüne’s notion of possessing a clear concept pertains only to the matter of a judgement. To make the case for this, I will begin by considering the first of the two central features of Kant’s account of the categories just mentioned. This will yield an initial, preliminary sense in which the categories pertain to the form of a judgement rather than its matter. I will then go on to consider also the second of the two features. This will yield a richer account of the form of a judgement and, accordingly, a richer account of the way in which the categories are applied in judgement.
To say that the categories have their origin (with regard to their matter) in the understanding—equivalently, that they are pure concepts of the understanding—must mean that possession of the categories is tied to possession of the faculty of understanding itself. Now, the understanding is characterized by Kant as, at bottom, a capacity for judgement. So the connection at issue is between the capacity for judgement, on the one hand, and the categories, on the other. Why should there be such a connection? A plausible suggestion would be the following: If any judgement whatsoever somehow involved an application of the categories, then one could not possess the capacity to judge without possessing the categories. If this is so, then the categories presumably pertain to a structural feature of judgement. The idea is this: If making a judgement requires a grasp of the basic structure of judgements, and if grasping this structure is to apply the categories (in the sense at issue here), then it is clear why the categories are implicated in any exercise of the capacity to judge.
Frege’s concepts ‘concept’ and ‘object’ play a role of this sort, and a comparison will be helpful. For Frege, these two concepts articulate the structure of an atomic proposition. To understand an atomic proposition (and, thus, to understand any proposition), one must grasp this structure; that is, one must grasp that the proposition is composed of the sense of an object-expression and the sense of a concept-expression. However, on pain of a regress, this kind of grasp cannot consist in entertaining a proposition of the form ‘Fx is a concept’. Rather, it must be displayed, for instance, in one’s understanding of the relations that Fa bears to various other propositions, e.g., Fb, Fc, Ga, and Ha. Nonetheless, since the concepts ‘concept’ and ‘object’ describe precisely the structure that one must grasp in order to understand the proposition, it makes sense to say that a grasp of this structure shows mastery of these concepts. And if this is right, then there is a sense here in which a thinker’s mastery of the Fregean concepts ‘concept’ and ‘object’ is at work in every act of grasping a proposition, even though neither of these concepts explicitly figures in the proposition (in the way ‘Fx’ figures in ‘Fa’).
Kant’s categories have the same kind of structural role, I think. We should therefore recognize an analogous sense in which a thinker deploys the categories in every judgement, even if they do not figure in its explicit content. However, Kant’s account of the structure of a judgement is different from Frege’s account of the structure of a proposition. Accordingly, so is his account of what one must grasp in order to understand a judgement. In Kant’s account, we can distinguish two layers. These pertain, respectively, to the forms catalogued in the Table of Judgement and to the way in which a judgement relates to objects. I will discuss them in turn.
First, consider the table of elementary logical forms of judgement (the Table of Judgement
). Bracketing the modal forms, which have a special status, all of these forms characterize different aspects of the way in which the elements constituting what Kant calls the matter of a judgement—the concepts it contains—are related to one another. For instance, they pertain to whether the predicate-concept is affirmed or denied of the subject-concept; whether it is affirmed or denied of all or only part of the subject-concept’s extension; whether the affirmation or denial depends on a condition internal to the subject-concept (categorical) or external to it (hypothetical) etc. For short, the elementary logical forms of judgement are forms of predication. Now, the fact that to each elementary logical form of judgement there corresponds a category suggests, at a first approximation, that the categories are concepts of forms of predication. Following the analogy with Frege, we might say that to understand a judgement one must grasp the forms of predication exemplified in it, and that to have this grasp is to possess (indeed, to apply) the relevant categories. If this is right, then there is a clear sense in which a category is applied in a judgement even when it does not figure in it as one of the concepts that constitute its matter. Thus, in the basic case, the category of substance would be applied in a judgement of the form ‘S is P’ even though the judgement is neither a case of ‘Substances are P’ nor a case of ‘S is a substance’. The fact that the content of the unschematized category of substance is specified by Kant as ‘first subject of predication’ might be taken to support this suggestion.
However, the categories are not only concepts of modes of predication. They are also characterized by Kant as concepts of an object in general (see e.g. B128) and as concepts of “the synthetic unity of the manifold in intuition in general” (A79/B105). Obviously, a story about what it is to possess the categories has to take this into account. This brings me, then, to the second of the two layers in Kant’s account of the categories that I distinguished above, the layer that concerns the way in which a judgement relates to objects.
Since intuitions are the representations of sensibility and sensibility is a distinct capacity from the understanding, it should seem surprising that concepts that originate purely in the understanding are characterized in terms of certain features of sensible representations. For why should the concepts that one possesses in virtue of possessing the capacity to make judgements have anything to do with characteristics of intuitions, which are, after all, the acts of a different capacity? If judgement were, as such, dependent on intuition, then this question would receive a fairly straightforward answer. For in that case the account of the categories given so far could be extended as follows: To understand a judgement, it is not only necessary to grasp the logical structure of the judgement (as I have sketched that so far). One must also grasp the way in which judgement depends on intuition. Accordingly, possession of the categories supplies not only a grasp of the logical structure of judgement, but also a grasp of its dependence on intuition. While I will not be able fully to establish that this is Kant’s view here, the following considerations should confer at least some plausibility on the claim.
First, consider that, in the first half of the B-Deduction, Kant repeatedly connects the categories with the intuition-dependence of finite understanding (see B135, B138ff., B145). He does this by way of contrasting finite and infinite understanding. In the case of the infinite understanding, the act of thinking provides for the actuality of its objects. For this reason, the acts of the infinite understanding are cognitions. By contrast, a finite understanding cannot on its own provide for the actuality (or even the real possibility) of its objects. For this, it is dependent on a different kind of capacity, viz. sensibility. Only a sensible capacity can furnish representations of instances of the understanding’s concepts. Only a suitable relation to sensibility, therefore, can ensure that the understanding’s concepts have objective reality. But this means that a finite understanding is a capacity for cognition only if its acts are suitably related to actualizations of a sensible capacity. We might put this point by saying that, qua capacity for cognition, the capacity to judge is intuition-dependent.
Next, consider the following well-known passage:
Understanding is, generally speaking, the faculty of cognitions. These consist in the determinate relation of given representations to an object. An object, however, is that in the concept of which the manifold of a given intuition is united. […] Consequently the unity of consciousness is that which alone constitutes the relation of representations to an object, thus their objective validity, and consequently is that which makes them into cognitions […]. (B137)
Here, the intuition-dependence of judgement is made explicit. The passage begins by characterizing the understanding as a capacity for cognition. Since the understanding is, fundamentally, a capacity for judgement, a judgement, in the sense at issue here, is by its nature a cognition. The passage then continues to connect the concept of cognition with the concept of being related to an object, and the latter notion with intuition. Finally, it is claimed that relation to an object is constituted by unity of consciousness. The upshot is that the understanding is a capacity for cognition (Erkenntnisvermögen) only to the extent that its acts bear the right kind of relation to intuition. Bearing this relation to intuition, however, is dependent on what Kant here calls unity of consciousness. More precisely, the representations of the understanding relate to objects only if they bear the right kind of relation to intuition; and this relation involves a certain kind of unity of intuition. The categories, Kant says elsewhere, are concepts of just this kind of unity of intuition (see e.g., B144).
The last point is important and needs spelling out. The cognitive role of intuitions, I said, is to furnish representations of instances of the concepts employed in judgements. Part of the idea is that an intuition can serve as a ground for connecting two concepts in judgement because both concepts are co-instantiated in the object given in that intuition. But if this is right, then intuition must have a kind of structure that somehow mirrors the structure of judgement, though it does so in a sensible rather than discursive manner. The categories are concepts of this structure of intuitions. They are concepts of the kind of structure (or unity, as Kant likes to say) that an intuition must minimally possess in order for it to be a representation of the kind of thing that we represent in judgements exhibiting the logical structure articulated in Kant’s table of logical forms of judgement; that is, for it to be the representation of an object.
Now, if judgement essentially depends on intuition and if this dependence requires that intuition exhibit a certain kind of structure, then we can say that the categories are concepts of the structure that intuition must exhibit if the understanding is to be a capacity for cognition. Extending the account of the categories already sketched, we can say, further, that to understand a judgement, one must grasp that the judgement in effect claims that there are possible intuitions which exhibit a certain kind of structure. To have this kind of grasp is to possess, and to apply in judgement, the categories.
If all of this is right, it shows that the categories are applied in judgement not in the way in which one applies a concept when one possesses it as a clear concept. They do not figure among the matter of a judgement, that is, the concepts it combines. Instead, they are concepts whose content captures certain formal-structural features of judgement. One applies categories in grasping these formal features of judgement. As I have tried to show, these features include the intuition-dependence of judgement. It is for this reason that Kant makes reference to intuition in his definition of the categories: “They are concepts of an object in general, by means of which its intuition is regarded as determined with regard to one of the logical functions for judgements” (B128).
I now turn to my second objection, which is that Grüne’s thesis that one can possess the categories as obscure concepts without possessing the categories as clear concepts implies a falsehood. In Grüne’s view, to have the capacity to apply categories in judgement is to possess the categories as clear concepts. And to possess the categories as obscure concepts is to have the capacity for sensible synthesis in accordance with the categories. Therefore, her thesis implies that one can have the capacity for sensible synthesis in accordance with the categories without having the capacity to apply the categories in judgement. But this, I think, is false. While it is true that the capacity for sensible synthesis in accordance with the categories does not depend on possessing the categories as clear concepts in Grüne’s sense—that is, on being able to make judgements in which the categories figure explicitly, i.e. figure in what Kant calls the matter of a judgement—this capacity does depend on the capacity to apply categories in judgement in the way that has been the topic of the discussion so far (and which I believe Kant is prepared to regard as a case of obscure concept possession).
The argument for this dependence is as follows: First, the application of categories in sensible synthesis consists in grasping the structure, or form, of an intuition. Second, this structure cannot be specified independently of the structure of judgement.
The idea is that to grasp the structure of an intuition is to understand that the intuition is the representation of an object; and to understand, further, that an object is something that theoretical cognitive judgements are about. So what one understands in grasping the structure of an intuition is that an intuition presents one with a subject of possible judgements. Let me elaborate.
I argued above that Kant’s thesis that the categories originate purely in the understanding should be taken to entail that possession of these concepts consists in the ability to grasp the logical structure of a judgement. This ability is exercised in judging, so we can say that to apply the categories in judgement is to grasp the structure, or form, of a judgement. As Grüne, I think, agrees, the same kind of account should be given, mutatis mutandis, of the application of the categories in sensible synthesis. Here, too, possession of the categories consists in the ability to grasp the structure, or form, of a representation. One applies the categories in sensible synthesis, then, when one has an intuition and grasps its structure.
But what is this structure? For present purposes, all that needs to be said is that it is the structure on account of which an intuition is a sensible representation of an object. Similarly, the kind of grasp of the structure of an intuition that one has in virtue of sensible synthesis can be specified by saying that one takes the intuition to be the representation of an object. And now my point is that what it is to be the representation of an object cannot be understood independently of the notion of judgement. In a nutshell, to be an object is to be something that is appropriately represented through judgement. So in taking an intuition to be a representation of an object, one regards the object of one’s representation as a possible content of judgement.
In support of this claim, consider first some textual evidence. In several places, Kant explains what the categories are by reference to the notion of judgement. More precisely, he claims that a category is a concept of a certain kind of unity, and the only specification he gives of this unity is to refer to what he calls the logical functions of judgement. Thus, consider once again the definition of the categories at B128:
[The categories] are concepts of an object in general, by means of which its intuition is regarded as determined with regard to one of the logical functions for judgements (B128).
I will assume that a logical function for judging is, roughly, a capacity for having representations exhibiting the logical forms catalogued in the Table of Judgement. If this is right, then what this passage says is that for a sensible manifold to exhibit categorial unity is for it to be determined in respect of a logical form of judgement. I take this to mean that the intuition presents an object that would be appropriately represented in judgement through a representation having this logical form. That is, if one wanted to make a judgement about the content of this intuition, the judgement would have to exhibit the relevant logical form. It would have to be, e.g., singular, affirmative, categorical etc. If this is right, then the passage implies that the unity for which sensible synthesis in accordance with the categories is responsible cannot be specified without any reference to judgement.
In addition to the textual evidence, the following consideration may serve to lend some plausibility to my proposal. An intuition is a representation of an object if and only if it exhibits categorial unity. For spatio-temporal objects, categorial unity is the unity described by the Pure Principles of the Understanding. So to say that an intuitional manifold exhibits categorial unity is to say that it exhibits the unity described by the Pure Principles. This unity, however, is itself described in terms that make reference to judgement. Consider, e.g., the Second Analogy. It says that all change occurs in accordance with causal laws. For Kant, the content of the category of cause and effect can be put like this: A is a cause of B just in case, if A is the case, then, necessarily, B is also the case. In this specification of the content of the category, the hypothetical form of judgement is employed. This suggests that one cannot grasp the content of the category independently of mastering the correlated form of judgement. If this is right, it follows that one cannot grasp an intuition as having categorial unity without possessing the capacity to apply categories in judgement (which, as I have argued, amounts to the capacity to make judgements).
Consider also what Kant says about the category of substance in the continuation of the passage from B128 just quoted:
Through the category of substance however, if I bring the concept of a body under it, it is determined that its empirical intuition in experience must always be considered as subject, never as mere predicate; and likewise with all the other categories (B129).
This formulation is offered as an example of what it means to regard the intuition of an object as determined in respect of one of the logical functions of judgement. The point of the passage can be put by saying that, in bringing an intuition of a body under the category of substance, the intuition is regarded as exhibiting the kind of unity that the representation of a subject of predication has. More precisely, the intuition exhibits the sensible version—the way of being in time—of the unity of a subject of predication. Again, the unity that a representation has in virtue of being synthesized in accordance with a category is characterized in terms that make reference to the capacity to judge.
To make a solid case for my proposal would, of course, take more work. But I have given some evidence for thinking that the role of the categories in sensible synthesis is to provide a thinker with a grasp of the object-presenting character of intuition. If this grasp involves regarding what is given in intuition as a possible topic for judgement, then one cannot possess the capacity for sensible synthesis in accordance with the categories independently of the capacity to make judgements. But according to my earlier argument, this entails that one cannot possess the former capacity without having the capacity to apply categories in judgement.
At bottom, then, my objection to Grüne’s account of sensible synthesis in accordance with the categories is that it completely severs the connection with judgement that the categories seem to me to have. In raising this objection, I have disregarded many of the more specific aspects of her account. There is much that is of interest here, and much that is worth debating. I have disregarded it in this discussion merely because I chose to use the limited space available to focus on a single topic that seems to me of central importance to Kant’s position and to Grüne’s account of it.
 Note that Grüne calls the position that is commonly associated with the label ‘conceptualism’ the “urteilstheoretische Lesart” and reserves the label ‘Konzeptualismus’ for her own position. While this is arguably a more suitable use of ‘conceptualism’, the alternative use of the terminology has become so entrenched that it seemed advisable to stick with it. So, as I use ‘conceptualism’ here, the position encompasses both Grüne’s urteilstheoretische Lesart and her Konzeptualismus.
 Grüne’s own account of sensible synthesis is not in terms of essential properties and their instances, but in terms of conceptual marks, on the one hand, and corresponding intuitive marks, on the other. I do not find the case for Kant’s acceptance of intuitive marks (largely due to Houston Smit) convincing, so I prefer not to use this terminology.
 The categories are explicitly mentioned only in the last of these three passages. The first two talk instead about synthesis and the unity of apperception. It is clear, however, that, due to the close connection between synthesis, apperception and the categories, the upshot is the same.
 More precisely, a judgement by its nature purports to be, but may fail to be, a cognition.—Note that the type of judgement under discussion here is theoretical cognitive judgement. Kant recognizes other types of judgement (e.g. aesthetic judgements, which are not cognitive, and practical judgements, which are cognitive, but not theoretical), to which the considerations adduced here may not apply.
 Different aspects of this unity can be distinguished in such a way that the contrast between intuition and experience (in Kant’s technical sense) can be brought to bear. But contrary to Grüne, I do not think this warrants saying that Kant has two different notions of objectivity, one for intuitions and one for experience (see p. 227).
© 2014, Thomas Land
Thomas Land received his Ph.D. in philosophy from the University of Chicago in 2010. He is currently Donnelley Research Fellow at Corpus Christi College, University of Cambridge. His primary research interest is in Kant, analytic Kantianism, and German Idealism, on which he has published various articles and essays.