It is hard to imagine an interpretation of Kant’s Critical theory that is new, informative, charitable and persuasive, but Lucy Allais has pulled it off. In Manifest Reality she offers a ‘moderate metaphysical interpretation’ of transcendental idealism. Her main goal is to argue that Kant is committed to the existence of things in themselves as grounds of appearances, having their own natures, unknowable to us. In defending her view she criticizes alternative accounts, including noumenalism, which takes things in themselves as intelligibilia, and deflationary theories, which interpret them as mere abstractions from empirical objects, without an existential commitment. Her second aim is to explain empirical realism by a relational account of perception according to which appearances are ‘essentially manifest’. In intuition subjects directly perceive publicly available objects, which, although representations ‘in us’, are not merely mental states of perceivers. She thus rejects phenomenalist and other readings assuming subjects directly know only their own mental states. Her defence includes a careful evaluation of competing interpretations of Kant, as well as more general theories of cognition. Her mastery of the Kant literature allows her to distinguish her account from others in an informative way. In short, she makes an exceptionally clear and strong case for a charitable defence of Kant.
The text is organized into three parts: part I lays out her interpretation of Kant’s idealism; part II develops her analysis of appearance and intuition through a relational theory of perception; and finally part III examines the interplay between idealism and realism, including the non-relational nature of things in themselves, and the connections between Kant’s transcendental idealism and his defence of metaphysics in the Transcendental Deduction. While her analysis touches on many topics, it focuses on the passages and issues most relevant to establishing her conclusions. Here I will discuss only her main theses and targets.
Part I consists of four chapters on transcendental idealism. Chapter 1 introduces the ‘moderate metaphysical view’ Allais proposes to defend Kant’s existential commitment to things in themselves as grounds of appearances. Here she briefly sketches the main competing interpretations. These include noumenalism (things in themselves are mere intelligibilia), the deflationism of Allison and Prauss which denies that Kant has an existential commitment to things in themselves as grounds of appearance, and the phenomenalism of Van Cleve and others according to which appearances are constructions from mental states of perceivers.
In chapter 2 Allais tackles the phenomenalist reading, that appearances are nothing more than combinations of or constructions from mental states. There are variants of phenomenalism, among them a ‘two world’ view that distinguishes spatio-temporal phenomena from intelligible noumena. There is also the intentionalist or adverbial strain interpreting appearances as a kind of act rather than object. Allais offers seven reasons to reject phenomenalism. The strongest is Kant’s repeated claim that the notion of appearance logically implies the existence of things that appear (Bxxvi ff., A251–2, Prolegomena 5: 315). Others include Kant’s insistence that empirical objects are perceived in one public space (and so cannot be merely objects ‘in the mind’) as well as his differences with Berkeley on the status of space and secondary qualities. She also cites the Transcendental Analytic conclusions that empirical reality consists of persisting substances that can exist unperceived and stand in causal connections, including theoretical entities such as magnetic force. Finally, the Refutation of Idealism and Fourth Paralogism argue that knowledge of inner states presupposes direct acquaintance with permanent spatial objects. Phenomenalists such as Van Cleve recognize the logical force of these views, but claim that Kant is simply inconsistent. Since Allais’s account conforms to the texts and renders Kant’s idealism coherent, there is no need to foist phenomenalism on him.
Chapter 3 criticizes the noumenalist reading of things in themselves as non-sensible intelligibilia such as Leibnizian monads. Allais’s main arguments include Kant’s claims that appearances are the same entities as things in themselves and that mere concepts alone cannot yield knowledge of any kind of object. The unknowability of things in themselves and Kant’s rejection of intellectual intuition also appear sufficient to kill this reading. But additional evidence includes the Phenomena and Noumena argument that the positive notion of noumenon has no legitimate use, and Kant’s rejection of Leibnizian monads in the Amphiboly section. Allais also mentions the ‘bare empirical realism’ propounded by Bird and Hanna, which interprets claims about things in themselves as committed to the concept rather than existence of things in themselves. She notes that many of her objections to deflationary views discussed in the next chapter apply here. The last section tackles the objection to taking things in themselves as numerically identical to appearances on the grounds that they have contradictory natures. Allais rightly argues that, since we have no determinate concepts of things in themselves, there is no basis for individuating them. Kant’s view that the things grounding appearances have their own nature, unknowable to us, does not commit us to any positive claims that could give rise to contradictions.
The subject of chapter 4 is a more significant target, the deflationary reading or ‘two aspect’ view of Allison, Prauss and others. It maintains that Kant’s only existential commitment is to appearances; the thing in itself is merely a way of considering them in abstraction from the conditions of cognition. This position is motivated by viewing transcendental idealism as a ‘methodological standpoint’ concerned with epistemic conditions of cognition. The mistake is defining epistemic conditions as necessarily based on the subjective structure of the mind rather than conditions that could also be real properties of objects. In effect, Allais thinks the deflationary view reads transcendental idealism into the notion of an epistemic condition: we cannot know things in themselves because the idea has no content. On her view (defended in part III), Kant’s order of reasoning is the reverse: he concludes that the forms of intuition are merely subjective and ideal because they are a priori; it follows that we can only know things as they appear to us in intuition. Allais also criticizes Allison’s claim that the discursivity thesis supports transcendental idealism, objecting that Kant’s treatment of transcendent metaphysics in the Dialectic requires him to recognize the existence of things as the grounds of appearance. Like phenomenalism, the deflationary reading in effect rejects Kant’s explicit claims that things in themselves must exist with natures that are unknowable to us.
Part II contains the most original part of Allais’s treatment, her argument that appearances should be understood as ‘essentially manifest’ based on a relational theory of perception. Although appearances are mind-dependent, they are not merely states of perceivers. Chapter 5 lays the foundation by explaining the relational theory in which the object is a component of the perception. Accordingly, perception is transparent since subjects are directly conscious of objects, and experienced qualities are presented as states of objects rather than of the subject. She claims this theory can meet the standard objection based on non-veridical perception by denying that subjectively indiscernible states are metaphysically identical, and discusses different strategies for dealing with such cases. The one she favours involves relativizing the appearance to the perceptual context, so that the same object can be presented differently based on the conditions of perception. Allais then develops the essentially manifest view through a comparison with different theories of colour, including various forms of objectivism and dispositional accounts. Like colours, the essentially manifest features of intuited objects are relational qualities dependent on objects, subjects and the conditions of perception.
Chapter 6 continues the analysis by generalizing the secondary quality account of colour to all experienced qualities. Bemoaning the translation of Vorstellung as ‘representation’ rather than ‘presentation’, she interprets the statement that appearances are ‘mere representations’ to mean their properties belong only to the object as perceptually presented; that they are ‘in us’ entails they are mind-dependent in the relational sense, but not merely inner states of subjects. The criterion of empirical reality is being a possible object of experience: this means existing in space and time, being material, and standing in causal relations with other objects. Thus theoretical entities such as magnetic forces can be empirically real since they are causally connected to perceivable objects.
In chapter 7 Allais argues that intuition without concepts produces direct awareness of singular objects. Here she offers a welcome corrective to the standard conceptualist reading that denies intuition alone can present unified particulars rather than mere collections of sensory data. The key is distinguishing the unity of intuition from that provided by concepts. Allais argues that for Kant mere intuition is of bounded particulars distinct from oneself and other individuals, for example, the kind of awareness characteristic of animal consciousness. This presupposes a ‘binding’ of the manifold into spatio-temporal bundles (by association or other processes) prior to synthesis by concepts. What synthetic combination adds is ‘relation to an object’, namely, the organization required to think an object as a whole made up of parts. Thus intuition alone gives us awareness of individuals, but not the thought of them as complexes of parts. Allais’s knock-down argument is that the successive synthesis by categories requires some perceptual unit on which to operate; because intuition alone gives no ‘smallest’ parts to be combined, intuition must produce perceptual wholes to be synthesized.
Chapter 8 completes Allais’s analysis of intuition and concept by relating the Aesthetic and Analytic arguments to Kant’s idealism and theory of synthetic a priori knowledge. On her view Kant establishes his idealism in the Aesthetic where he argues that space and time are a priori forms of intuition (and thus presupposed in our perceptions of particulars). Because knowledge of things in themselves could not be a priori (since human intuition is sensible), space and time must be merely forms of appearances and not features of things in themselves. Thus the Aesthetic establishes that space and time and the objects given in them are transcendentally ideal. It also shows that synthetic a priori knowledge of geometry necessarily applies to objects of experience. By contrast, the point of the Analytic is not to establish idealism, but that empirical cognition (which Kant assumes) presupposes pure concepts and metaphysical principles. (She elaborates on this in chapter 11.) Here she argues that the Analytic assumes the ideality of spatio-temporal objects. Thus the standard interpretation that sees idealism as following from the view that the mind ‘makes’ objects is not only too simple but circular, since it assumes the ideality of forms of intuition to justify idealism in general. Allais agrees that Kant uses idealism to explain how we have synthetic a priori knowledge of metaphysics. But this cannot be his justification of idealism, since metaphysics is suspect and requires a separate epistemological justification.
The final part consists of four chapters bringing together her results on Kant’s realism and idealism. Chapter 9 defends Kant’s repeated claims that the concept of appearance logically presupposes the thing existing independently, contrary to some anti-realist readings. For one thing, his resolutions to the Antinomies require an existential commitment to things in themselves, since empirical reality is not a complete, determinate totality. She also contrasts Kant with anti-realists who require existence claims to be verification-dependent. Rather than embracing a verification theory of meaning (since transcendent metaphysics is not literally meaningless), Kant argues that knowledge is experience-dependent.
Chapter 10 looks more closely at Kant’s view that things in themselves have intrinsic, non-relational natures unknowable to us. The Critique spells this out in the Amphiboly of the Concepts of Reflection, where Kant criticizes Leibniz’s monadology. The problem is to separate the Leibnizian views Kant accepts from those he rejects. Allais argues convincingly that Kant accepts the Leibnizian premises that appearances are relational and must be grounded in something non-relational, but rejects Leibniz’s conclusion that the grounding must be intelligible monadic substances. Based on this analysis she criticizes Langton’s interpretation of things in themselves as having causally inert intrinsic natures, for lacking textual support and ignoring the distinction between logical and real possibility.
Chapter 11 develops Allais’s view that the Deduction arguments are epistemological rather than intended to support idealism. As she indicated in chapter 7, Kant wants to justify a priori concepts and principles as conditions of referential thought, necessary to apply empirical concepts. Justifying empirical concepts is a problem for any theory taking thought to be spontaneous and general, whether or not its objects are ideal. I found her treatment of categories as ways to avoid ‘gerrymandered’ and ‘deviant’ objects that lack unified causal continuity especially enlightening. Chapter 12 returns to the relation between idealism, established in the Aesthetic, and the possibility of metaphysics, defended in the Analytic. Again she argues that idealism makes metaphysics possible, but is not justified by it. She ends the book with a brief, provocative interpretation of Kant’s resolution of the conflict between free will and determinism based on the account developed throughout the text.
I confess to being sympathetic to Allais’s treatment at the outset. Based on my work on Kant’s incongruent counterparts arguments, which I saw as the source of his theory of sensibility and resulting idealism, I have always read Kant as making two bare-bones assumptions: first, that something exists having its own intrinsic, non-relational nature, and second, that knowledge is a relation between a subject and an object. These imply nothing about the nature of existing things or the source and extent of knowledge, and would seem acceptable to all theories of knowledge. Kant’s aim is to draw the right lines concerning what we can know based on an accurate appraisal of our subjective capacities. Despite my antecedent agreement with Allais, I learned a great deal from her analysis. Especially illuminating are her account of appearances as ‘essentially manifest’, and her explanation of the relations between Kant’s idealism and his defence of metaphysics. While there may be minor points of contention, she has succeeded in presenting a strong, coherent and well-argued defence of Kant’s idealism.
