2. Bird's Account of the Transcendental Aesthetic

Bird's discussion of Kant's Transcendental Aesthetic takes place in Part II of TRK. The first section of this part is simply titled ‘Introduction to the Aesthetic’ and offers an overview of the background to Bird's reading of Kant. To begin with, he outlines the task of Kant's Transcendental Aesthetic. Then, he summarizes the main claims of the Aesthetic. Subsequently, he briefly describes the most important debates in the literature. Finally, he presents his own interpretation and explains in what Kant's major contribution to philosophy consists. In the remaining part of the ‘Introduction to the Aesthetic’, he offers an overview of chapters 6, 7, 8, 9 and 10 that constitute Part II of the book. Let me first briefly introduce the background to Bird's account.

Concerning the aim of the Transcendental Aesthetic, Kant's task is to isolate a priori elements in sensibility. The interest in these elements, Bird notes, is not psychological, but metaphysical. In other words, they are not investigated as conscious events, which reflect on the psychological faculties necessary for the adult, human experience. On the contrary, they are regarded descriptively as fundamental conditions of experience (44). The a priori elements Kant identifies in sensibility are space and time, which are forms or conditions of sensible intuition. He first shows that space and time are a priori ways in which objects are presented to our senses. Then, he outlines his own theory of perception and experience. The central tenet of this theory is that space, time and appearances are transcendentally ideal and empirically real (99).

Concerning the controversies in the literature, Bird notes that interpreters have usually disagreed in their interpretations of how Kant shows that space and time are a priori, and of his theory of perception (in particular the theory of transcendental ideality and empirical reality). One general interpretative direction regards Kant simply as a supporter of traditional idealism. A second trend maintains that he merely appeals to a common-sense view of what we perceive. Two questions follow from these debates and Bird purports to answer both. The first question concerns the extent to which Kant shares the framework of traditional idealism. The second examines the role of the appeal to common sense as an antidote to scepticism. Bird's answers are that Kant neither supports traditional idealism nor tries to answer an idealist scepticism about the external world (99).

Before moving on to an outline of his argument, Bird mentions some of the complexities of Kant's argument. As I have mentioned, these are discussed in order to bring to light the main respects in which the significance of Kant's argument is still missing from current interpretations and, hence, in order to make this significance explicit. Obviously, Kant's argument is much more complex than the consideration of traditional idealism and scepticism. Moreover, Kant adds complexity to the earlier vocabularies of ‘impressions’ and ‘ideas’. These come to include now terms such as ‘intuition’, ‘sensation’ or ‘appearance’ (100). Kant offers a preliminary account of mathematics. He considers disagreements between Leibnizian and Newtonian doctrines over the character of space and time in physics. He adjudicates upon these through an account which is supposed to be entirely original.

Hence, in Bird's view, Kant attempts to distance his own account from a series of alternative doctrines (including Leibniz's, empiricism and Berkeley's) and, importantly, from traditional idealism. A clear indication of the distinction between his view and traditional idealism is given by the rejection of the epistemic priority the latter attaches to inner experience over outer. By contrast, Kant claims that there is no such priority. Both inner and outer sense have an immediate relation to their objects. Moreover, space, time and appearance, on Kant's account, have an unquestioned empirical reality.

Having presented Kant's task in the Transcendental Aesthetic, his main claims, the debates in the literature and the alternative, better interpretation which can solve the debates, Bird goes on to summarize the content of chapters 6–10. In the next section, I will focus on the summary Bird provides, in order to place the discussion of this paper in the appropriate context and to elaborate further on the motivation behind the choice of this particular topic.

3. Revolutionary Interpretation and Mind-Dependence

Although the Transcendental Aesthetic is one of the shortest parts of the first Critique, in the previous section we have seen the impressive range of topics it raises. Bird's discussion, as will become clearer in a moment, touches on the most important of these topics, the issue of the mind-dependence of space and time being one of them. The starting point of Bird's discussion is, not surprisingly, a descriptive introduction to Kant's discussion in the Aesthetic.Footnote 2 This offers an overview of the arguments Kant offers and of the main topics he considers.

Two main issues are discussed next.Footnote 3 The first one is the so-called ‘blindness’ problem. This is generated by Kant's famous claim, at A51/B75, that ‘thoughts without content are empty; intuitions without concepts are blind’. But, as mentioned above, the Transcendental Aesthetic has the task of identifying the a priori elements of sensibility alone. Hence, it is unclear how we can identify such a priori elements, if the condition is that we focus on a priori intuitions alone, that is, without any admixture of concepts from the understanding.Footnote 4

The second issue concerns the receptivity of the senses, in particular, the character of the causes of our sensory responses. Traditionalist commentators, Bird notes, see Kant as committed to a causal relationship between things in themselves and our senses. This is in conflict with Kant's claim that we have no knowledge of things in themselves.Footnote 5 Once it becomes clear that Kant's task of identifying the a priori elements of sensibility is not made impossible by the blindness problem and that his account of the receptivity of the senses is not in conflict with an agnosticism about things in themselves, questions can be raised about Kant's arguments for space and time as such elements of sensibility.Footnote 6

One issue here concerns the role of mathematics in Kant's arguments for space and time. Although simply by reference to the Transcendental Aesthetic this issue cannot be adjudicated upon, Bird suggests an approach to a related interpretative controversy. According to Kant, mathematics contains synthetic a priori truths, and it is precisely because empiricism cannot account for such truths that he offers his transcendental philosophy. Some commentators think that Kant attributes the synthetic character of mathematics to inferences, whereas others, to axioms. One major difference between these two positions is that the former can link this interpretation to the suggestion that Kant regards mathematical inferences as synthetic because he does not have access to the equivalent logical, but analytic, apparatus of polyadic quantification.Footnote 7

The next issueFootnote 8 concerns the mind-dependence of space and time as a priori and intuitive. Kant insists that space and time are ‘ideas’, ‘in the mind’ and ‘subjective’. Yet Bird argues that this mind-dependence is not a feature which is shared with traditional idealism, but, on the contrary, distinguishes Kant's position from traditional idealism.Footnote 9 Although distinct from traditional empiricism, Kant's position should also not be assimilated to a traditional realism that is committed to the existence of things in themselves. The distinction between Kant's position and transcendental realism is the last most important topic of this part of Bird's book.Footnote 10

Any of the issues related to Kant's Transcendental Aesthetic and discussed by Bird would merit examination. The reasons why I will focus on the question of mind-dependence are the following. The discussion of the blindness problem and that of a causal account of receptivity are meant to reject some important objections which would considerably weaken Kant's account. Together with this, the examination of Kant's arguments for space and time as a priori intuitions reconstructs Kant's view in such a way that his conclusions become at least plausible and free from some well-known problems.

By contrast, the problem of the mind-dependence of space and time relies on the assumption that Kant's account in the Transcendental Aesthetic is at least plausible. Secondly, however, the issue of mind-dependence is a more general issue, which can be raised not only about space and time as necessary conditions of experience, but also about the objects of this experience more generally – about appearances or empirical objects. Finally, the last issue Bird discusses in Part II (whether Kant's account is not in fact a version of traditional realism) is not raised as often as that of mind-dependence and that of a traditional idealist commitment of Kant's account. So mind-dependence is a crucial issue for the question of the traditionalist vs. revolutionary interpretations of Kant.Footnote 11 In the next section, I will begin by presenting a related debate between Henry Allison and Paul Guyer, a debate to which Bird also makes reference.

4. A Debate on Ideality and Subjectivity

The debate between Guyer and Allison concerns the relation between the claims that space and time are not properties of things in themselves, on the one hand, and, on the other, that space and time are subjective. These two claims are two of the conclusions that Kant draws from his metaphysical and transcendental expositions of space and time. Thus, Kant says: ‘Space represents no property whatever of any things in themselves, nor does it represent things in themselves in their relation to one another’ (A26/B42). He also adds: ‘Space is nothing but the mere form of all appearances of outer sense; i.e., it is the subjective condition of sensibility under which alone outer intuition is possible for us’ (A26/B42).

According to Allison, Kant derives the first claim from the second (Reference Allison1983: 102; see also 104). The argument runs as follows. First, and restricting the argument to space, space is a mere form of appearances. This is the second conclusion cited above. As such, space pertains to objects only in virtue of our manner of representing them. Hence, space cannot pertain to objects as these are independent of our manner of representing them. But, in this case, space cannot pertain to things in themselves. This is the first conclusion above. It is derived from the second conclusion by a clarification of what an a priori form of sensibility involves.Footnote 12

By contrast, Guyer purports to show that Kant's inference is from the first claim to the second, and ‘not vice versa’ (Reference Guyer1987: 354). According to Guyer, Kant's textsFootnote 13 clearly indicate that he takes the first claim to be the premise of an argument which demonstrates the second claim as its conclusion. Moreover, he thinks the premise of this argument is grounded in the necessary truth of the contents of our knowledge about space (1987: 354–5). Finally, he takes Kant to argue from the premise to the conclusion on the basis of an exclusive disjunction:

What Guyer calls ‘the problem about necessity’ is the ground of the premise of the putative argument (the premise that space cannot be a property of things in themselves). So the conclusion follows: space can only be in some way subjective.

Allison's and Guyer's interpretations start from completely different aims. Allison hopes to show that the transcendental ideality of space and time can be demonstrated by Kant starting from the materials provided in the Transcendental Aesthetic but independently from assumptions concerning the nature of mathematics (1983: 114). By contrast, Guyer aims to identify a contradiction or tension between Kant's transcendental method and the necessary truth of the contents of our knowledge about space (Reference Guyer1987: 355).

This indicates that we might in fact be dealing here with only an apparent debate: whereas Allison is concerned with a reconstruction of Kant's argument, which would be able to make the points Kant should be able to make, Guyer is concerned to capture what Kant actually says and to evaluate his arguments. In this sense, therefore, we need not regard Guyer's comments as the beginning of a debate. Even if we accepted that the exchange amounts to a debate, we would still be able to question the extent to which such a debate has in fact something to do with Kant.

The two Kantian claims which Allison and Guyer try to organize as parts of an argument are, as I have already mentioned above, conclusions of Kant's discussion in the Metaphysical and Transcendental Expositions. Hence, one point raised by Bird in relation to this debate is precisely whether it is legitimate to attempt to adjudicate upon the disagreement concerning the role each of the two claims is supposed to play in Kant's argument (168). In other words, the claims are more likely to be conclusions which are drawn independently from Kant's arguments in the Expositions, rather than claims which depend on each other.Footnote 14 Both claims could be derived independently from the earlier claims demonstrated in the Expositions that space and time are a priori and intuitive (168).

Irrespective of how we decide to understand the nature of this debate, there seems to be an implicit traditionalist reading of Kant in the view that space is either real or subjective. The next step must therefore be to reject this reading and to show in what ways Kant's account differs from traditional idealism. I will begin this in the next section.

5. Traditional Idealism

One thing one can immediately note when considering the exclusive disjunction above (space is either subjective or real) is that it can function as an implicit assumption in both Allison's and Guyer's arguments. Allison can be regarded as demonstrating that space is subjective and, hence, as excluding its status as a ‘real’ property of things in themselves. By contrast, Guyer can be seen as showing that space is not real and, hence, as asserting its subjective character.

This implicit exclusive disjunction is regarded by Bird as an assumption, specific for traditional idealism, of a dualism: ‘space is either real or else merely a form of representation, either independent of our senses or else a subjective mental state’ (171). This traditional dualism, he thinks, is underpinned by the assumption, noticeable also at work in Guyer's argument, that only reference to things in themselves would allow space to be real. This is, of course, problematic, since, for Kant, space and time are empirically real.Footnote 15 Yet, according to Bird, traditional idealism assumes the mind-dependence of space and time without questioning the sense in which we talk about ‘dependence’ or ‘independence’ in such a case. The assumption that the notion of mind-independence unequivocally refers to things in themselves is, for Bird, a clear symptom of traditionalism.

Bird adds three preliminary conclusions. First, he thinks that it is now possible to explain Kant's use, in the Aesthetic, of two notions that his Critical philosophy seems able to distinguish, but which had been identified in the Dissertation. Thus, Bird says, Kant's use of the notion of ‘things in themselves’ and that of ‘things themselves’ suggests Kant's earlier view that the senses only yield cognition of appearances, not of the things they are appearances of. Such variability in the use of the notions, Bird adds, encourages the identification of ordinary spatial objects and of things in themselves. This is then what leads to Guyer's conclusion that, since space cannot be a property of things in themselves, it is not real (since it is not a property of ordinary spatial objects either).

According to Bird, the confusion can be clarified not only by keeping the distinction between things themselves and things in themselves operative, but also by emphasizing the distinction between ‘being derivable from’ and ‘belonging to’ objects (172). He links this conclusion to the central argument in the Aesthetic (concerning the a priori status of space and time), which, he adds, shows in fact that, as a priori, space and time cannot be derivable either from things in themselves or from empirical objects. Before going on to the second conclusion formulated by Bird, I would like to spell out in more detail these claims. To begin with, this is an example of Kant's uses of ‘things in themselves’ and ‘things themselves’:

In this passage, Kant states first that space cannot be a property or determination of things in themselves. He explains this further with the claim that space does not belong to things in themselves or to objects themselves. This is because any determination belonging to such an object or thing could only be derived from the experience of things/objects and, hence, could not be cognized a priori.

If we read Kant as claiming that objects themselves are things in themselves which cannot have space as one of their properties, then we must conclude also that space is not a property of empirical objects either. This suggests that space is merely subjective, an illusion, rather than a feature of the ordinary objects of our experience. By contrast, if we distinguish between objects themselves, as empirical objects of our ordinary experience, and things in themselves, independent from our experience, then Kant's argument can be understood in the following way.

Kant can claim that space is derivable neither from things in themselves nor from objects themselves, since it is an a priori form of intuition. Moreover, as an a priori form of intuition, space cannot belong to things in themselves, since the latter are objects independent from us, including from our intuition. Yet, this leaves it open whether space can belong to objects themselves, that is, to the objects of our ordinary experience. Furthermore, it is clear Kant thinks space does belong to empirical objects, since he claims space (and time) are empirically real.

The second conclusion drawn by Bird concerns the notion of mind-dependence. As we have seen, on Bird's account of Kant's Aesthetic, space is real and, therefore, in some sense, mind-independent; at the same time, however, as a form of intuition, space is also mind-dependent. Hence, space is at the same time mind-dependent and mind-independent. Consequently, the notion of mind-dependence needs to be clarified.

The third conclusion formulated by Bird points to the need for a criterion which can settle the extent to which Kant makes the traditional idealist assumptions that are presupposed by Guyer's argument. He takes Kant's General Comments on Transcendental Aesthetic to be the criterion which can settle this issue: If Kant acknowledges there the traditionalist idealist dualism, Guyer's position is vindicated; if not, it is rejected (173). In the next section, I will focus on these issues.

6. Mind-Dependence

Bird acknowledges that, in places, the General Comments encourage a traditionalist dualism. Yet he insists that such a reading will need to ignore complexities of Kant's account, which are required by his apparatus (181). In particular, the traditionalist dualism he has in mind consists of transcendent objects (things in themselves) and mental states (ideas). This simple distinction however cannot account for space and time. On Kant's account, as we have seen, space and time belong to objects of experience, even though they are not derivable from them. They are not derivable from them because such a derivation would require as a preliminary stage the experiencing of these objects. Yet space and time are both a priori forms of intuition.

It follows that space and time are neither simply mental states nor properties of things in themselves. As properties of things in themselves, they should be derivable from these transcendent objects. Yet, they cannot be derivable from things in themselves in the same way and for the same reason as they cannot be derivable from objects of experience. But as mental states, they could not belong to objects of experience.

What are the implications of this more complex account of the idea of a mind-dependent world? Bird's conclusion is that any such notion (similar notions would be those of idea, of subjectivity or of ‘being in the mind’) has an empirical and a transcendental use, which have to be clearly distinguished, identified and consistently employed (187). Accordingly, on Bird's account, there are three ways to make the contrast between transcendental and empirical idealism in relation to space and time. First, as forms or modes of representing through intuition, space and time are universal and necessary conditions in which we represent in sensibility anything – whether inner or outer. They are not inner experiences in consciousness in the manner in which traditional idealism takes ideas to be. They can be features of inner and outer objects the existence of which is not doubted. Space and time are certainly mind-dependent in a certain sense, but not in that of empirical idealism, where ideas are not real.

The second way in which the mind-dependence of space and time can be clarified is by deriving this feature from their a priori character. Space and time are mind-dependent, because they cannot derive from experience, although they may characterize and belong to objects of experience (188). At the transcendental level, there is an exclusive choice between what is subjective (related to the mind and our cognitive powers) and what is objective (derivable from things which are independent from mind and our cognitive powers). Once space and time are demonstrated as a priori, they can only be subjective in the transcendental sense. They are not subjective in the empirical sense, because they are constitutive of, and therefore belong to, both inner and outer objects.

A third way in which empirical and transcendental idealism can be contrasted is by reference to the notion of an outer object. Empirically outer objects are spatial objects, which can be contrasted with empirical ideas, which are inner experiences. Transcendentally outer objects are things in themselves, things which are ‘outer’ in the sense that they are independent from mind and our cognitive powers. Transcendentally ‘inner’ objects or appearances are all objects of possible experience – whether empirically inner or outer. Hence, when Kant talks about space and time as subjective, he does not mean that they are empirical ideas, inner experiences. After all, he takes space and time to be universal and necessary conditions of both inner and outer experience.

The discussion so far clarifies the sense in which, for Kant, space and time are mind-dependent, and distinguishes clearly between Kant's transcendental idealism and the subjective idealism along the lines of which Kant has been interpreted by traditionalists. Before concluding, I would like to look at one final issue, which seems to me to move the debate one step further. The issue is the following: even if we accept that, on Kant's account, space and time are empirically real and that Kant's transcendental idealism is a version of idealism quite distinct from traditional, subjective idealism, it still seems the case that, for Kant, empirical reality is a ‘second-best’ and not the ‘ “real” reality’ (175). Transcendental reality is ‘real’ reality, since, by definition, it does not depend on mind and our cognitive powers. How real could empirical reality be when it depends on such cognitive powers and vanishes with them?

Bird's response is that, for Kant, the genuine article is not the absolute reality of things in themselves, but empirical reality. Absolute reality is a philosophers’ myth about which we should be sceptical. We should be sceptical about this, because we know nothing of things in themselves. Our knowledge or cognition needs both intuitions and concepts, but things in themselves cannot be intuited, since, by definition, they refer to the way things are independently from our senses. Things in themselves are only accessible to us through reason (177).

As I have mentioned, Kant only denies that spatio-temporal properties are properties of things in themselves, since, on his account, these properties are tied to our sensory apparatus, which, by definition, is absent in the case of things in themselves. What is unclear, however, is the extent to which being sceptical about something means denying its ontological status. It seems clear that, if they exist, things in themselves are real in a sense in which empirical objects are not. The latter are mind-dependent in a sense in which, if they exist, the former are not. As long as we are not sure whether things in themselves exist or not, we cannot deny the possibility of a reality which is ‘stronger’ than empirical reality.

In other words, what I am suggesting is that we can respond to the objection that our cognition is not cognition of the ‘genuine article’ only by denying the existence of things in themselves. This seems to be confirmed by Bird's claim at 184:

One can wonder here on what basis Kant would reject the underlying reality of things in themselves. Since this is a type of object for which sensible intuitions are not possible, all we can do is to ‘present it through all the predicates that are already contained in the presupposition that the object has as a property nothing belonging to sensible intuition’ (B149; emphasis in the original). We can say that things in themselves are not extended or in space, but we cannot reject their reality. By assuming that Kant rejects the reality of things in themselves, Bird answers the worry that, in Kant's account, the genuinely real things are the things in themselves, rather than the empirical objects, but he does this at the price of introducing an inconsistency in Kant's account.

Now, my objection evaporates if we understand Bird as claiming that Kant's rejection is not of the existence of underlying things in themselves, but of a claim to this existence.Footnote 16 We are not entitled to such a claim for the same reason that we are not entitled to a claim which would deny the existence of things in themselves – by definition such things are not objects of possible cognition. This however leaves the issue of the reality of things in themselves untouched. Under the assumption that they exist, these become ontologically the genuine article, whereas empirical reality is not.

Nevertheless, the question is whether the implication concerning the absolute reality of things in themselves affects in any way Kant's account. Does Kant claim that empirical reality is absolute? The answer is evidently negative, and can simply be derived from Kant's claims concerning the mind-dependence of space and time. The possibility of the existence of things in themselves, which Kant's account leaves open, and the implication of a superior ontological status for things in themselves over empirical objects can therefore always lead to sceptical questions concerning the reality of empirical objects. Yet, as Bird notes repeatedly in TRK, Kant is not engaged in a normative project aimed at responding to sceptical challenges concerning the reality of the external world.