First, I would like to congratulate Professor De Boer for writing such a stimulating and thought-provoking book. In addition to incorporating secondary literature in English, French and German, she brings a tremendous diversity of primary literature to bear on her revisionist interpretation of Kant’s Critique of Pure Reason as a book that seeks not to destroy the metaphysics of the Wolffian tradition but to pave the way to its reform.
It is a testament to the creativity of a work on Kant that it forces even long-time readers of the Critique to see many of its central passages in a new light, and to the argumentative rigour of such a work when that light illuminates its subject more fully than the old. By that measure, at least in terms of my own understanding of Kant, Kant’s Reform of Metaphysics (De Boer Reference De Boer2020; references throughout simply by page number) must be reckoned a great success. I will, of course, draw attention to places where I question either De Boer’s general reading or the path by which she comes to it. But on the whole, I found this to be a lively, carefully argued and meticulous book.
As I have already mentioned, the principal thesis of De Boer’s book is that the Critique of Pure Reason is not an attempt to destroy the metaphysics of the Wolffian tradition but to pave the way to its reform. De Boer thus takes seriously Kant’s frequent claims that the Critique is a propaedeutic to metaphysics but also his claim that the metaphysics developed in the wake of the Critique will ‘follow the strict method of the famous Wolff, the greatest of all dogmatic philosophers’ (Bxxxvi; translations throughout from Kant Reference Kant1998). I must admit to having always read this as a bit of hollow praise designed to satisfy Kant’s Wolffian readers. Having now read Kant’s 1790 letter to Kästner, which De Boer cites in chapters 1 and 8, I am far more inclined to take these remarks seriously. Indeed, these comments, combined with the remaining correspondence she considers in chapter 8, make clear that Kant was not only interested in developing a reformed version of metaphysics in 1781 but also that he remained interested in doing so up through the end of the 1790s.
As an initial point of criticism, however, it seems to me that Kant’s 1799 open letter to Fichte presents more of a challenge to De Boer’s characterization of Kant’s continued interest in articulating his reformed metaphysics than she suggests (p. 251). As Zweig translates the relevant portion of this letter, Kant writes:
I must remark here that the assumption that I intended to publish only a propaedeutic to transcendental philosophy and not the actual system of this philosophy is incomprehensible to me. Such an intention could never have occurred to me, since I took the completeness of pure philosophy within the Critique of Pure Reason to be the best indication of the truth of that work. (12: 370–1; in Kant Reference Kant1999)
Thus, it appears in 1799 that Kant regards the Critique of Pure Reason as the system of transcendental philosophy and, presumably, that this system is complete, contrary to his many letters in the 1780s and early 1790s that attest to the contrary.
As De Boer notes, however, Zweig’s translation omits the words ‘gepriesen habe’ from the second sentence of the passage, which I now quote in German:
Es hat mir eine solche Absicht nie in Gedanken kommen können, da ich selbst das vollendete Ganze der reinen Philosophie in der Crit. der r. V. für das beste Merkmal der Wahrheit derselben gepriesen habe. (12: 371)
Taking these omitted words into consideration, De Boer translates this sentence as follows:
Such an intention could never have occurred to me, since I myself, in the Critique of Pure Reason, have lauded (gepriesen habe) the completed whole of pure philosophy as the best indication of the truth of this philosophy (derselben). (p. 252; second interpolation mine)
Translated in this way, the passage does not indicate that the Critique of Pure Reason itself is the ‘completed whole of pure philosophy’ but rather that it is merely the place in which Kant ‘lauded the complete whole of pure philosophy’. As De Boer puts it:
The preposition ‘in’ pertains not to what is actually contained in the Critique of Pure Reason, but to what is recommended in it. Read in this way, there can be no doubt that the passage concerns Kant’s long-term intention to publish a metaphysical system, not the question as to what he took himself to have achieved. (p. 252)
On this reading, Kant’s letter is indeed consistent with De Boer’s view that Kant ‘never changed his mind as regards his intention to elaborate a reformed version of Wolff and Baumgarten’s metaphysical treatises’ (p. 250).
In the broader context of the letter, however, it seems to me that what Kant ‘took himself to have achieved’ is precisely what is at issue. Kant’s letter responds to the anonymous review of Johann Gottlieb Buhle’s Sketch of Transcendental Philosophy published in Erlangen’s Literary Magazine. In his review, the reviewer expresses surprise at Buhle’s insistence that he ‘has no difficulty understanding that Kant, as the inventor of the idea of transcendental philosophy, also realized it in its complete totality (ganzer Umfang)’ (Buhle Reference Buhle1798: 49; quoted in Anonymous 1799: 58). The source of the reviewer’s surprise is that Kant himself, the reviewer insists, regards the Critique of Pure Reason as ‘a propaedeutic to transcendental philosophy but not a system of this philosophy itself’ (Anonymous 1799: 58).
Kant references this assertion in the first sentence of the passage from his letter to Fichte, which I now quote in my own translation, which differs little from De Boer’s and Zweig’s: ‘the presumption to claim I intended merely to deliver a propaedeutic to transcendental philosophy and not a system of this philosophy itself is incomprehensible to me’ (12: 370). The ‘presumption’ (Anmaßung) in question is the reviewer’s claim that the Critique is ‘a propaedeutic to transcendental philosophy but not a system of this philosophy itself’. Indeed, Kant’s use of indirect speech makes it clear that he has the precise wording of the reviewer’s claim in mind. Thus, the question at issue in this part of Kant’s letter is not what is ‘recommended’ in the Critique, as De Boer (p. 252) suggests, but what the Critique actually accomplishes.
There is also a grammatical ambiguity in the second sentence of the passage that I think is best resolved in favour of the reading De Boer wants to resist. What strikes me as ambiguous here is the antecedent of derselben. On De Boer’s translation, this antecedent is the ‘pure philosophy’ the ‘complete whole’ of which is ‘lauded’ in the Critique. Since one can laud something in a work that is distinct from the work itself, De Boer, as we have seen, also interprets the second ‘in’ to indicate not ‘what is actually contained in the Critique of Pure Reason’ but ‘what is recommended in it’ (p. 252).
However, another grammatically possible and indeed more proximate antecedent of derselben is ‘the Critique of Pure Reason’. On this reading, which Zweig also prefers, the ‘completed whole of pure philosophy in the Critique of Pure Reason’ would be ‘lauded’ as the ‘best indication of the truth of the Critique of Pure Reason’ or, as Zweig phrases it, ‘the best indication of the truth of that work’ (my emphasis). I therefore propose the following translation:
Such an intention could never have occurred to me, since I myself have lauded the completed whole of pure philosophy in the Critique of Pure Reason as the best indication of the truth of the Critique of Pure Reason.
Interpreted in this way, this passage from Kant’s letter suggests that, in 1799 at least, he wanted the public to regard the Critique of Pure Reason as both the propaedeutic to his system of transcendental philosophy and the system itself.
Thus, instead of attempting to reinterpret the content of Kant’s letter on this point, I think it would be more advantageous from De Boer’s perspective to undercut its significance for her overall interpretation in a different way. The reviewer of Buhle’s book goes on to describe Kant as the ‘first teacher of transcendental philosophy’ but Fichte as ‘indisputably’ the ‘first transcendental philosopher’ who ‘realized the plan sketched in the Critique and systematically developed the transcendental idealism to which Kant had gestured’ (Anonymous 1799: 61). It thus seems reasonable to interpret Kant’s insistence in his open letter that the Critique is both a propaedeutic to a system of transcendental philosophy and the system itself as an attempt to wrest control of his philosophical legacy from the hands of Fichte and those, like the reviewer of Buhle’s book, who clearly supported him. That this attempt is at odds with Kant’s privately expressed desire in his letters from the 1780s and early 1790s as well as in the unpublished drafts of the prize essay to develop his projected metaphysical system does not imply, it seems to me, that Kant’s belief in the importance of developing this system had changed.
Another, more general, question I have concerns the role of pure sensibility in Kant’s projected system of metaphysics as De Boer understands it. In chapter 8, she divides this system into a metaphysics of nature and a metaphysics of morals. She then divides the former, which is her focus, into transcendental philosophy, which is the part of Kant’s projected system that is begun but not finished in the Critique of Pure Reason itself, and rational physiology, which is the part of the system that is only announced in the Critique (pp, 216–22).
Transcendental philosophy, as I understand De Boer’s position, consists in the examination of a priori concepts and principles of the cognition of objects as such. And I think she makes a persuasive case in chapter 3 that, at this level of generality, ‘transcendental philosophy’ can denote Kant’s specific way of approaching these issues as well as the Wolffian ontology and the ‘transcendental philosophy of the ancients’, which Kant wants either to reject or reform (pp. 87–8).
Specifically with respect to Kant’s arguments in the Critique, however, she takes the Metaphysical Deduction and the Analytic of Principles to belong to transcendental philosophy, while the Transcendental Deduction, Schematism and Phenomena/Noumena chapters are relegated to the aspect of transcendental cognition she terms ‘transcendental critique’ (pp. 91–6). The Critique, however, does not completely develop transcendental philosophy, primarily because it gives no account of the predicables, i.e. the a priori concepts that are derived from the categories of the pure understanding (pp. 219, 237).
Since the Analytic of Principles uses the schematized categories to demonstrate the principles of possible experience and since the schematization of the categories itself involves pure time, it would therefore seem that the content of transcendental philosophy includes at least this aspect of pure sensibility. However, there are passages in De Boer’s book that suggest the contrary. In her discussion of the Inaugural Dissertation in chapter 2, for example, she writes that both this work and the Critique ‘defend the position that metaphysics ought to purge itself of any concepts and judgments gained by sensibility in order to resolve its internal conflicts’ (p. 57).Footnote 1 Similarly, in her discussion of Kant’s two-pronged critique of metaphysics in chapter 3, she writes that the Critique requires ‘scientific metaphysics’ to, among other things, ‘shed its dependence on pure sensibility’ (p. 68). Finally, De Boer emphasizes in a number of places (e.g. pp. 82, 93, 99) that transcendental philosophy concerns the ‘pure intellect’, and one of the primary senses in which the intellect is pure is that it is non-sensible and that its products (i.e. the categories) are independent of sensibility.Footnote 2
The role of sensibility is also difficult to pin down in rational physiology, the branch of the metaphysics of nature that on De Boer’s view is not developed at all in the first Critique. Kant divides rational physiology into immanent and transcendent branches, and De Boer is clear that the two disciplines that are part of the latter branch (rational cosmology and rational theology) both abstract from space and time (pp. 220, 242–, 245). So if sensibility enters rational physiology, it is not through its transcendent part.
With respect to the immanent part of rational physiology, however, there is some ambiguity in De Boer’s account. This part of rational physiology contains two disciplines, rational physics and rational psychology. De Boer is clear that the former involves sensibility in two ways. First, rational physics presupposes the empirical concept of matter but only insofar as it treats ‘the pure concepts and principles that allow the human mind to obtain knowledge of material bodies’ (p. 235). These concepts and principles are contrasted with those of ontology or transcendental philosophy, which allow the human mind to obtain knowledge of objects ‘as such’ (ibid.). Second, De Boer suggests that at least some of the principles of rational physics are synthetic a priori, which would entail that rational physics also involves pure sensibility. In particular, De Boer writes that the Metaphysical Foundations of Natural Science ‘arguably instantiates, albeit in a provisional manner, Kant’s projected rational physics’ (p. 237). Of that book, she subsequently writes:
The metaphysical principles treated in the various chapters, for their part, appear to be – warranted – synthetic a priori judgments even though Kant does not use the term in this context: these principles arguably determine, independently of experience, how something can be an object of mathematical physics in the first place. (p. 237)
De Boer claims, however, that rational psychology, in contrast to rational physics, involves no synthetic a priori principles and also that its object, the ‘I think’, is produced by ‘completely abstracting from sensible determination, that is, by isolating the mere act of apperception’ (p. 239). So it would appear that, unlike rational physics, rational psychology does not rely on an empirical concept of its object.
There are, it seems to me, two problems with laying out the distinction between rational physics and rational psychology in this way. First, Kant is clear in the Architectonic that rational psychology takes as its object not merely the concept of a ‘thinking being’ but the ‘empirically inner representation “I think”’ (A848/B876; my emphasis).Footnote 3 So it would seem that rational psychology draws on the empirical in much the same way rational physics does, namely, by taking ‘from experience nothing more than what is necessary to give ourselves an object’ (A848/B876).
Second, if the synthetic a priori judgements of the Metaphysical Foundations belong to rational physics understood as a part of Kant’s reformed metaphysics, it would seem that they must also draw – contrary to De Boer and Pollok (Reference Pollok2001: 81), whom she approvingly cites – on the synthetic a priori principles articulated in the Analytic of Principles (p. 236). The reason for this is that, on De Boer’s view, the Analytic of Principles belongs to ontology or transcendental philosophy, which is the discipline that treats the pure concepts and principles that allow the mind to obtain knowledge of objects ‘as such’ (p. 235). If the synthetic a priori principles of the Analytic of Principles are necessary for the mind to obtain knowledge of objects as such, they must also be necessary for the mind to obtain knowledge of objects of rational physics.
Finally, De Boer begins her book by quoting Mendelssohn’s often-cited reference to ‘the all-crushing Kant’ in the preface to his Morning Hours (p. 1). Her overall intent, as I have already mentioned, is to undermine this characterization, both by drawing attention to Kant’s continued interest in developing a reformed system of metaphysics and by developing an outline of that system based on the Architectonic and Kant’s metaphysics lectures, correspondence and Reflexionen.
However, I think it is worth asking whether one can accept De Boer’s characterization of Kant’s overall aims and of the outline his reformed system of metaphysics would have taken while still maintaining that, at least with respect to Wolffian metaphysics, Kant remains the alles Zermalmender. In certain respects, Kant’s insistence that he wants to identify the faults of the old metaphysics to usher in the new is similar to Hume’s comments in the first Enquiry that the ‘justest and most plausible objection’ against metaphysics is that it is ‘not properly a science’ and that we must ‘cultivate the true metaphysics’ in order ‘to destroy the false and adulterate’ (Hume Reference Hume, Selby-Bigge and Nidditch1975: 1.11–12). What Hume means by ‘true metaphysics’, of course, is unlikely to be accepted as metaphysics by those who defend what he describes as its ‘false and adulterous’ cousins. While an exploration of this suggestion is beyond the scope of my comments, one might similarly wonder whether Mendelssohn would recognize Kant’s reformed metaphysics as metaphysics or whether what De Boer characterizes as reform would seem to him more like revolution.