Willaschek, Marcus. 2018. Kant on the Sources of Metaphysics: The Dialectic of Pure Reason. Cambridge: Cambridge University Press.

Marcus Willaschek has written an excellent book on Kant’s account of reason as the source of metaphysical speculation in the Transcendental Dialectic of the Critique of Pure Reason (CPR). There are insights on every page and it will be essential reading for Kant scholars, especially but not only those who work on the theoretical philosophy. Willaschek’s writing and presentation make for an exceptionally clear, accessible read, so the book will also be useful for students. It will be of interest to those working on the history of metaphysics and metametaphysics more generally, and it may also be of interest to contemporary metaphysicians and metametaphysicians. The book should become a standard in its field.

In this short review I briefly introduce the topic of the book, its core structure and content, and some selected points of interest.

The Critique of Pure Reason is an investigation into the nature, scope, and limits of pure reason. Such an investigation is necessary, according to Kant, because there is a problem. As he puts it in the opening sentence of the A-edition preface:

Human reason has the peculiar fate in one species of its cognitions that it is burdened with questions which it cannot dismiss, since they are given to it as problems by the nature of reason itself, but which it also cannot answer, since they transcend every capacity of human reason. (CPR Avii)

Kant’s claims about the limits of pure reason are well-known, well-studied, and have been generally well-received. Many have found devastating his attack on traditional speculative metaphysics concerning the soul, the world, and God. The same cannot be said of Kant’s claims about the source and inevitability of metaphysical speculation as arising from the nature of reason itself. Willaschek’s book puts these latter claims front and centre. His aim is to develop a novel and detailed interpretation of them, as well as a partial defence.

Willaschek calls Kant’s account of how metaphysical speculation arises naturally and inevitably from the nature of reason the Rational Sources Account. It consists of three distinct theses (e.g. pp. 5, 157):

RS1. Rational reflection on empirical questions necessarily raises metaphysical questions about “the unconditioned.”

RS2. Rational reflection (by “pure reason”) on these metaphysical questions necessarily leads to metaphysical answers that appear to be rationally warranted.

RS3. The rational principles that lead from empirical to metaphysical questions and from there to metaphysical answers are principles of “universal human reason”; that is, they belong to rational thinking as such.

According to Willaschek, we can see this account operating at four different “levels” in the Critique, each roughly corresponding to one of the four main parts of the Transcendental Dialectic: the Introduction, on the transition from the logical to the real or transcendental use of reason; Book One, on the system of the transcendental ideas of the soul, the world, and God; Book Two, on the dialectical inferences of reason that purport to provide knowledge of the soul, the world, and God; and the Appendix, on the legitimate regulative use of reason’s principles in contrast to their illegitimate constitutive use. On Willaschek’s reading, Kant lays out the general framework of his Rational Sources Account in the Introduction to the Transcendental Dialectic, before applying this framework and filling in its details in the parts that follow. It is this structure that provides the organizing principle of the book. It divides into two main parts. Part I, chapters 1–5, offers a detailed interpretation of the “first level,” general framework of the Rational Sources Account, while Part II, chapters 6–9, moves to the subsequent three levels in which this framework is applied and fleshed out. There is a very useful general introduction, as well as a fascinating, and I hope promissory, postscript on Kant’s practical metaphysics.

Chapter 1 gives a terrific overview of Kant’s conceptions of reason and metaphysics by way of background and stage-setting. Chapter 2 then concerns the logical use of reason and the Logical Maxim “to find the unconditioned for the conditioned cognitions of the understanding” (CPR A307/B364). Willaschek argues, controversially but forcefully, that the logical use of reason aims at comprehending the systematic unity of nature, not merely the hierarchical ordering of cognitions according to generality, and he gives a detailed account of the content of the Logical Maxim, arguing in particular that it concerns both inferential and epistemic conditioning. The chapter closes with a partial defence of Kant’s claim, as Willaschek sees it, that the Logical Maxim is a legitimate regulative principle of universal human reason. It “normatively guides the way rational beings (qua scientists) organize their body of cognitions” and is “valid for rational beings as such” (p. 65). This was one of the few parts of the book that I found somewhat unsatisfying, if only because I found myself wanting more.

Willaschek focuses on two issues that he thinks might make Kant’s account seem problematic. First, Kant’s foundationalist conception of epistemic justification (in the scientific context) and his view that genuinely scientific knowledge must be certain. Second, the tension between the sheer demandingness of the Logical Maxim and the idea that it binds reasoners per se. In the first case, Willaschek simply points out that the supposedly outmoded aspects of Kant’s conception of science can be detached from the basic idea that there is a rational requirement to look for general principles from which specific cognitions can be derived. In the second case, we are told that “the Logical Maxim does hold for everyone, but vacuously so for most, since a necessary condition of its making substantive requirements on us and our cognitive activity is not satisfied in most cases” (p. 70). For the Logical Maxim is a hypothetical rather than a categorical imperative, so that “we are rationally required to pursue [systematic unity in the unconditioned] only when doing so is morally permitted and pragmatically feasible” (p. 64), and the important point is that, precisely because of the demandingness of the Logical Maxim, doing so will only very rarely be pragmatically feasible. In the first case, one might worry that Kant’s view has been defended only by stripping it of anything especially Kantian. In the second case, one might worry that such a move makes the Logical Maxim rather too hypothetical, to the point that metaphysical speculation starts to seem somewhat less than inevitable. But Willaschek makes a number of philosophically and exegetically interesting points here and what he says is good as far as it goes. I’m just not sure it goes far enough.

What I thought was missing at this point was any general account of the more foundational issue of what it really means for reasoners as such to be “normatively guided” by a principle like the Logical Maxim, for it to be a “valid rational requirement.” There are a number of deep issues here, and a number of very different ways to cash out such claims. Kant of course has interesting, controversial things to say. But it wasn’t entirely clear to what extent the Rational Sources Account depends on a uniquely Kantian conception of reason. This matters for what is required to defend it. In any case, I would certainly have welcomed seeing Willaschek’s expertise brought to bear on the matter.

Chapter 3 moves from the logical to the real use of reason and from the Logical Maxim to the Supreme Principle: “when the condition is given, then the whole series of conditions subordinated one to the other, which is itself unconditioned, is also given” (CPR 307–308/B364). Willaschek argues for an ontological reading of what Kant means by “given” in the Supreme Principle—when the conditioned exists, so too must the unconditioned totality of its conditions exist—and he defends philosophically and textually sophisticated accounts of the real conditioning relation, the unconditioned, and the relation of the Supreme Principle to the Principle of Sufficient Reason. This is a complex, tightly argued chapter that well repays the close critical scrutiny it requires.

Chapters 4–5 conclude Part I by giving an original and powerful account of the transition from the Logical Maxim to the Supreme Principle. Chapter 4 focuses on a situated textual analysis of what Willaschek calls the Transition Passage (A307–308/B263), a short, one-sentence paragraph on which Willaschek relies heavily and out of which he teases a lot. Chapter 5 lays out the core philosophical account of the transition that we will see play out at different levels in the chapters that follow. Crucially, Willaschek understands this transition as involving two stages, first the transition from the Logical Maxim to the regulative use of the Supreme Principle, and second the transition from the regulative to the constitutive use of the Supreme Principle. Only the first stage is rationally necessary, in a nutshell because the Logical Maxim, concerning as it does on Willaschek’s reading the systematic unity of nature, necessarily presupposes the regulative use of the Supreme Principle, which recall concerns the existence of the unconditioned. The second stage, by contrast, merely appears rationally necessary under the (supposedly natural but ultimately spurious) assumption of transcendental realism.

Willaschek gives an intriguing account of exactly how transcendental realism is the “key” to transcendental illusion. Starting with Kant’s basic definition of transcendental realism as the view that empirical objects are identical to things in themselves, he argues—via a discussion of noumena and the intuitive intellect—that transcendental realism ultimately comes down to the view that “[t]here is a necessary correspondence between the principles of reason and the principles of reality” (p. 144). From there he proposes that such a view can plausibly be thought a tacit background assumption of everyday rational thinking or common sense, and that this explains the way in which there is a natural tendency, that will forever assert its pull, towards transcendental illusion.

Willaschek gives bivalence as an example of a principle of reason that, on his reading, Kant thinks it would be a mistake to treat as a principle of reality (p. 149). Willaschek’s reasoning here seems based on a misreading of the Antinomies. The problem is not that Kant doesn’t reject the principle that Willaschek takes him to reject, namely:

Bivalencew. Of the two cosmological claims “The world is finite in magnitude” and “The world is infinite in magnitude,” precisely one is true and one is false.

Kant does of course reject Bivalencew. But his whole point in the Antinomies, it seems to me, is that claims like Bivalencew don’t really follow from the principle of bivalence because the propositions in question, such as those concerning the magnitude of the world, are not really contradictories. They only appear so under some false presupposition. This is clear, I think, from Kant’s example of a body that has no smell and thus smells neither good nor not good (A503–504/B531–532). He says that the analogy holds for the other antinomies (A505/B533), the difference being that while the (apparently but not really contradictory) propositions of the mathematical antinomies can both be false, the (apparently but not really contradictory) propositions of the dynamical antinomies can both be true. Nowhere does Kant suggest that there are propositions here that are something other than either true or false.

Now, it may well be that doubts about bivalence somehow follow from Kant’s transcendental idealism. But that won’t help Willaschek. He later suggests (chapter 9) that it’s a philosophical benefit of the Rational Sources Account that it depends only on a rejection of transcendental realism and not on the acceptance of transcendental idealism. This is coherent, he argues contra Kant, because the two views are contraries rather than contradictories: transcendental realism says that there is a necessary correspondence between the principles of reason and the principles of reality, while transcendental idealism insists that this correspondence only holds for appearances, not things in themselves. One might well deny that there is any such correspondence. Put to one side the plausibility of attributing to Kant such a basic misunderstanding of the relation between transcendental realism and transcendental idealism, especially in a context in which he is so acutely attendant to scope ambiguities and the contrary/contradictory distinction. If Kant presupposes realm-spanning principles of reason in diagnosing and rejecting transcendental realism, that puts pressure on Willaschek’s interpretation of the doctrine.

The discussion of “levels” two through four in Part II goes by much more quickly than that of the first level in Part I, as the first level has provided the general template that is then applied and fleshed out at the three subsequent levels. This shift in gear between Part I and Part II enhances rather than detracts from the book, which thereby manages to bring out superbly the often elusive structural similarities between Kant’s treatment of the different areas of traditional speculative metaphysics in this long, labyrinthine part of the Critique.

Chapter 6 concerns the “second level” transcendental ideas of the soul, the world, and God, and how they are supposed to arise necessarily from rational reflection. They do not, according to Willaschek (and as he admits Kant seems to suggest), arise from the mere forms of rational inferences, but rather from rational inferences about specific subject matters in psychology, cosmology, and theology. Their derivation or metaphysical deduction, then, does not itself take place until the “third level,” that of the dialectical inferences of reason treated in the Paralogisms, the Antinomies, and the Ideal. This level is the concern of chapters 7–8, with chapter 8 also including Willaschek’s account of the “fourth level” regulative-constitutive transition treated in the Appendix. Finally, chapter 9 rounds out the discussion by relating the Rational Sources Account—Kant’s account of reason as the source of inevitable metaphysical speculation—back to Kant’s more famous account of the limits of reason, i.e. his critique of traditional speculative metaphysics. It is here that Willaschek argues that Kant’s general diagnosis of what goes wrong in trying to gain knowledge of the unconditioned or supersensible, namely the tacit assumption of transcendental realism, is independent of any commitment to transcendental idealism, since the two views are contraries rather than contradictories. The concerns I noted above notwithstanding, this is another particularly excellent chapter. It, along with parts of chapters 5 and 7, will be of special interest to those working on Kant’s signature doctrine.

In addition to the general introduction, Part I and Part II each have their own introductions and conclusions, and the same is more or less true of each of the nine individual chapters. All this signposting is welcome. It aids comprehension and it makes the book eminently usable for scholars and students alike, as does the fact that the book is extremely well situated in the literature with extensive references throughout. Willaschek’s book is an extremely welcome addition to the literature on the Transcendental Dialectic and Kant’s metametaphysics more generally.