Every student of Kant’s thought is familiar with his reflective assertion that the Transcendental Deduction in the first Critique cost him more labor than any other part. The centrality of that presentation for his entire argument coupled with its sheer conceptual difficulty lay behind Kant’s efforts and the attention he afforded to elaborate it correctly. Countless students from his time to our own have themselves spent an inordinate amount of time attempting to follow the train of thought Kant pursued in it. Not just have disagreements surfaced on the actual steps Kant took in the Deduction, but differing opinions arose on how he could best achieve his intended end. We know that even in his own day Kant’s contemporaries were puzzled by the Transcendental Deduction as it appeared in the Critique’s first edition. Kant took the criticism to heart, and in a second edition of the work he gave a completely new version of the argument. Regrettably, what many thought should be the clearest of all presentations in the Critique – owing to its centrality – has been viewed by able scholars over the ensuing decades, nay centuries, as puzzling and obscure, but above all as inconclusive. Whereas many have ventured opinions on the success of Kant’s endeavor, few, if any, have concluded that he succeeded in achieving whatever it is that he had set out to do.
Numerous attempts have been made even in recent years to do what Kant himself seemingly was unable to accomplish, namely to give a clear account of Kant’s argument in the Deduction, quite apart from whether it succeeds or not. Even if we lay aside the presentations in the form of journal articles, the number of book-length studies alone is surprising, even astonishing – and this just in the English language. Fortunately, we have two eminently lucid expositions of Kant’s Transcendental Deduction, namely Guyer’s Kant and the Claims of Knowledge and Allison’s Kant’s Transcendental Deduction, though as the titles suggest the latter is more pointedly directed at explaining Kant’s text than the former. Both, however, are works that no serious student of Kant can afford to ignore. Certainly, there are considerable differences between the two books in their conclusions and their manner of executing their respective projects. Nevertheless, regardless of one’s familiarity with Kant’s text, whether a graduate student trying to understand Kant’s problem for the first time or an accomplished Kant-scholar, both books offer much clarification and many insights. Moreover, both books make ample use of Kant’s writings from his so-called “Silent Decade” and thus attempt to trace the evolution of Kant’s problem in the Deduction from his early “pre-Critical” writings. Now, we have Laywine’s dense contribution to the literature.
Alison Laywine is one of the few scholars who already in 1995 undertook an examination of Kant’s pre-Critical works in considerable depth with the hope of shedding light on the basic tenets of his Critical writings and positions. In that previous work, Kant’s Early Metaphysics and the Origins of the Critical Philosophy, Laywine told us that an understanding of the first Critique’s dichotomy between the faculties of sensibility and the understanding requires an understanding of his earlier position, why Kant adopted it at the time, and what led him to alter it, assuming, of course, that he did. Implicit in Laywine’s train of thought here is that such knowledge of Kant’s development is necessary in order to arrive at Laywine’s understanding of Kant. She and so many others who make similar claims take no account of the fact that others with a different understanding of Kant may not feel the need to turn to Kant’s pre-Critical writings. Given her position regarding just what the Deduction seeks to achieve, her argument for an examination of Kant’s early writings surely would include that he retained the kernel of his early metaphysics but reinterpreted and adapted it to fit those aspects of his philosophy that changed over the intervening years.
In any case, Laywine claims – and not without good reason – that Kant came to realize the inadequacies of his 1770 Inaugural Dissertation and tried during the subsequent decade or so to investigate and establish the limits of human cognition and the role of the respective faculties he delineated within those limits. He argued that what could be known under the conditions of one faculty could not be known under the conditions of the other. Yet unlike in 1781, the year of the first edition of the Critique of Pure Reason, Kant held earlier in 1770 that the understanding through its concepts can cognize things in themselves, whereas things conceived through the senses are representations of things merely as they appear to us. He drew this conclusion from his contention that space and time are formal conditions of our sensibility. The strict separation of the faculties would go on to pose significant issues for him, and the attempt to resolve them led him to his mature system.
Laywine’s newest book essentially takes up where she left off previously, namely with a discussion of a set of Kant’s papers now known collectively as the Duisburg Nachlaβ, written most likely in 1775, and thus just about half-way between the Inaugural Dissertation and the first Critique. Short of paper, Kant often wrote notes on whatever paper was at hand including in the blank areas on letters he received. The so-called Duisburg Nachlaβ is such a set of jottings written in the blank areas of a letter bearing a date. Based on that fact, we know approximately when the note was scribbled. For those interested in the development of Kant’s mature positions, these notes are of great importance, since we have little else by which we can see the evolution of his thought. Although Laywine acknowledges Guyer’s treatment of the Nachlaβ in his own book, she charges him with neglecting Kant’s early metaphysics and with resisting the idealism that she sees present in the Nachlaβ. She, on the other hand, recognizes “some kind of idealism” (19) in them, but in doing so she in effect begs us to ask of her what kind of idealism is it that she sees. We get an answer or, rather, to use her own words, “some kind of” answer, further on in her detailed exposition of the Nachlaβ, where she says that the idealism is not transcendental idealism, but an idealism based on the idea that the cognizing subject intuits oneself directly through an intellectual insight and “bodies” – presumably meaning all else besides the subject – only insofar as they affect me. In this she explicitly sees herself as diverging from Guyer’s reading of the Nachlaβ, according to which Kant’s writing hopes to provide a theory of the a priori conditions of empirical knowledge but without thereby establishing an unbridgeable chasm between the world as cognized and the world as it is apart from our human cognition, i.e., as it is “in itself.” Even if one were to agree with Guyer in his reading of the Nachlaβ, one must wonder along with Laywine what Guyer meant by realism, particularly if he had Kant’s own definition in mind. Kant claimed, after all, to be an empirical realist even while espousing his transcendental idealism. Thus, the onus falls here on Guyer to clarify his position and interpretation – or at least it does on Laywine’s reading of Guyer.
However, even if we agree that Guyer failed to provide such a clarification, this does not absolve Laywine from providing her own account of Kant’s idealism. She writes that based on her reading of the A-Deduction, i.e., the Deduction as found in the 1781 Critique, Kant was an idealist about objects taken as objects of knowledge, but this need not mean that he was a skeptic concerning the external world (137). Clearly in 1781 Kant did not offer, and presumably therefore saw no need to offer, a special refutation of idealism. To Laywine’s thinking, Kant must have been as surprised as anyone that his sheer confidence concerning externality could be questioned. That critics charged him with Berkeleyan idealism forced Kant to add a distinct “Refutation of Idealism” in the 1787 second edition of the Critique. The question, then, is whether Laywine’s understanding of Kant’s idealism as presented in 1781 stands scrutiny given her premise that Kant took for granted the existence of the external world at the time. To be sure, Laywine finds nothing in Kant’s idealism that would disturb his confidence in externality. However, for us the question is whether Laywine’s confidence is itself misplaced. Are there not ample grounds in the 1781 Critique to think that Kant must have recognized the significance of the problem? And since he did not offer a refutation in the first Critique, is it not possible that he was still searching for one? Much, then, depends on the nature of Kant’s 1781 position, to which Laywine writes she will return in §3d (147-150) of the second chapter of her book. Indeed, Laywine there does somewhat return to the issue, albeit only in a footnote, in which she stresses her disagreement again with Guyer’s treatment of Kant’s idealism in the Nachlaβ. Unfortunately, determining Laywine’s own position requires an understanding of Guyer’s in order then to set Laywine’s against it. Certainly, this can be done, but the procedure requires the reader to make the necessary inference. The burden Laywine thrusts upon her readership is not aided by her assertion that Kant’s idealism in 1775 is “something like” (149f) the idealism that Guyer attributes to Kant in 1781. If the two positions are merely “like” each other, then in what way are they different?
Regrettably, Laywine, by her own admission, states that although she will take into account both versions of the Deduction – the A-Deduction and the substantially revised version in the second edition, or B-Deduction – her focus throughout her treatment is on the latter. This may be understandable given the argument she develops, but it does significantly narrow her potential audience, who may want an understanding of the Deduction on the whole, and not just her particular argument. In this respect, Henry Allison’s earlier work, in patiently dealing with both versions of the Deduction, succeeds far better and is far more accessible to a general reader. True, the B-Deduction may be, as Laywine writes, “more perspicuous” (14), but it is for that very reason, then, that one might well expect her to devote more attention to the first version. To be sure, she does not wholly dispense with the first edition tout court. She does draw instructive parallels between passages in the two versions of the Deduction, but she often finds the A-Deduction more convoluted and the argument more ambiguous than in the B-Deduction. An understanding of Laywine’s discussion here is again helped by the juxtaposition of her understanding with that of Allison’s. Laywine states (14) that a second reason for concentrating on the B-Deduction is to show that certain “infelicities” some have seen in the A-Deduction concerning metaphysics are not mistakes that are “ironed out” in the B-Deduction. On the contrary, Laywine believes that they are essential to the Deduction as such.
Laywine’s devotion to the Nachlaβ in the context of the present work concludes with the assertion that it more or less shows that Kant recognized already in the middle of his “Silent Decade” the need for a deduction of the categories, regardless of their number, of the understanding, i.e., for an, in effect, legal argument substantiating the claim that pure concepts in the understanding apply to appearances and do so a priori. Even stronger, Laywine holds that Kant did provide such a deduction already in the Nachlaβ at least in outline. This can hardly come as a surprise to readers familiar with the vast secondary literature. For example, Wolfgang Carl in a paper “Kant’s First Drafts of the Deduction of the Categories” originally read in 1987 to an audience at Stanford University, but published in 1989, noted that Kant had already drafted several versions of the deduction before the 1781 Critique.
In turning to the Deduction itself in the B-Deduction, one of Laywine’s first concerns is understanding what Kant meant by such terms and expressions as “manifold” and “synthetic unity of apperception.” She turns again to Kant’s pre-Critical writings for clarification as to how he employed the word “manifold” previously, hopefully, thereby, throwing light on his use of the word in the Critique. Laywine writes that for Kant every intuition has a manifold, including a priori intuitions. This would seem incontestable, particularly since Kant himself clearly makes that assertion, for example at B160. Whatever the case, Laywine suggests that not everyone recognized this, Dieter Henrich being one.
Particularly since Henrich’s 1969 article on the proof-structure of the B-Deduction, virtually every commentator on Kant’s first Critique has had something to say on whether the B-Deduction presents two distinct steps or two distinct arguments for a single conclusion. The Deduction, in Laywine’s words, is to show that the categories or pure concepts of the understanding “are the formal conditions of thought in the same way that the pure intuitions of space and time are the formal conditions of sensibility” (13). The problem, so to speak, is that the argument of the B-Deduction extends through §26 of the Critique, but the conclusion offered there does not appear to be substantially different from that already presented in §20, where Kant writes that “the manifold in a given intuition also necessarily stands under categories,” the categories being nothing other than the functions for judging (B143). Only by standing under categories is the unity of the sensible manifold possible. Thus, does the first step of the B-Deduction end with the quite brief §20 (B143) and then resume with a second step at §22 or does §22 advance another distinct argument? If what we have here are two arguments for one conclusion, it looks as though Kant was saying you should pick the one that you like best (209). On the other hand, if the B-Deduction consists of two steps, how do the two steps differ?
Laywine writes that she “prefers” to think that the B-Deduction consists of two steps rather than two arguments (209). Stating that one of the two options is preferred is hardly a ringing endorsement of a choice. We can only hope that now having completed her inquiry Laywine is firmly convinced that she made the correct decision. She does realize, though, that having made her choice she must now say what the second step offers to the argument beyond what the first affirmed. Kant provides the answer, or as Laywine herself calls it “an important clue” to the answer, at the beginning of §26. She then quotes one sentence from that section – in her own translation – after which she remarks that its significance lies in its announcement that the “final step” of the Deduction will be an attempt to account for how nature is possible (210). Are we, therefore, to conclude that what sets the “second step” apart from the first is that it finally answers how nature is possible? But, then, the first step, contrary to its appearance, cannot have reached the same conclusion as the second step. Laywine’s answer lies in understanding that Kant uses the word “nature” in two different senses, a material and a formal sense. Laywine finds the material sense given explicitly in Prolegomena §36, where Kant writes that nature is the totality of all appearances, and the formal sense concerns appearances governed by laws so that they form a unified whole. Stated in such terms what is at issue appears simple enough. But interpreting the intricate and puzzling B-Deduction through the lens of the Prolegomena asks us to assume that the twofold sense given in the Prolegomena accurately reflected Kant’s thinking at the time of the Prolegomena – and not one offered for the sake of simplicity alone – but also that Kant continued to maintain the same stance in 1787. That the B-Deduction does indeed hold to the twofold sense of nature is a major task of Laywine’s treatise.
Laywine sees her proposal for understanding the B-Deduction as unique. Henrich, for one, wrote that what Laywine sees as the respective sections of the B-Deduction constituting the two steps in fact do not actually come to the same conclusion. For Henrich, Kant presents two different arguments with two different conclusions. Others have proposed variations. In Laywine’s estimation, Hans Wagner came close to her own by recognizing that the second step focuses on empirical intuition and perception, whereas the first step deals only with intuition as such. While this train of thought leads him to recognize the importance of the question how nature is possible, he failed to exploit this insight. In focusing on how perception is possible, Wagner, in Laywine’s estimation, overlooked accounting for how perceptions can be connected. Certainly, she correctly remarks that without such an explanation of connected perceptions there can be no explanation of how nature is possible. But we may ask of Laywine what more needs to be added to Wagner’s argument to produce her own. The answer is both easy and hard. That is, it can easily be stated as that we must notice the cosmological aspect of the second step, the contribution of a cosmology of experience makes to the completion of the B-Deduction argument.
We know that Laywine puts much weight on this conception of a “cosmology of experience.” After all, it features prominently in her book’s subtitle, and she mentions the expression many times in her text. She tells us on page 87 that such a cosmology – she also calls it a “metaphysics” (3f) – treats experience “as a unified whole of appearances … and tries to establish its conditions of possibility by showing that its unity comes from laws legislated to appearances by the understanding through its categories.” This treatment of experience, allegedly, is conspicuously absent from the first step of the B-Deduction. The emphasis here is on the word “unity,” for it allows us to characterize the world as a whole. Laywine claims that the germ of this conception of a cosmology of experience “informs” Kant’s account of human sensibility already in the Inaugural Dissertation and again is revealed in the Nachlaβ and even in the A-Deduction. Of course, if it is as easy as Laywine says, it is hard to see why the answer to Henrich’s challenge appears explicitly only on page 214, but in her defense she did prepare much of the needed groundwork up to this point. In her own estimation, the hard part of her argument is to make an understandable presentation of what a cosmology of experience will contribute to the Deduction.
Even by her own reckoning, having reached §26 of the B-Deduction, the final section of the argument, Laywine, by her own opinion, has still not been able to clinch the required proof. According to her, to say, as Kant does, that the categories are valid for all objects of experience in B161 is not the same as invoking universal laws to make the unity of appearances possible. But is it? Laywine continues, holding that if we are to speak of universal laws of nature, they must either stem from God, which Kant has dismissed for his purposes here, or from the categories. Kant at B163 wrote that the “categories are concepts that prescribe laws a priori to appearances, thus to nature as the sum total of all appearances. … Here is the solution to this riddle.” Can we say that with this Laywine’s task is complete?
Hermann Cohen, in his own all-too-brief commentary from 1907, found, like Laywine, that these words contained the transcendental question. He found the answer – or so he says – in the distinction between objects and things in themselves, a distinction that Laywine fortunately does not invoke. But she does, like Cohen, find that the solution she seeks lies in the understanding’s self-activity and the imagination. Cohen writes, “Here we see, however, how the imagination and the connection that it creates between sensibility and the understanding make the resolution more plausible. … Thus by means of the imagination nature, as the sum total of appearances, becomes nature, ‘as the original ground of its necessary lawfulness’” (63, B165). Certainly, while there are many nuances that distinguish Cohen’s endeavor from that of Laywine, there is a distinct similarity that Laywine does not so much as mention.
There are certainly many positive points to Laywine’s treatment of the B-Deduction. Not one of them, however, is that her investigation is largely ahistorical. In this too, though, she is not unique. Despite all the scholarship over the more than two centuries that preceded her work, she makes little reference to it. Does she think that her formulation of and solution to the central problem of the Deduction is original as compared to all that has gone before? If it is unique, that would be a stunning claim in light of all of the nineteenth century German scholarship alone, not counting those from the twentieth, some of which she does briefly mentions. But if it is not, in what way does her treatment contribute to what has gone before? Or is she, in effect, saying that each of us should attempt for ourselves unaided by the past to square the circle? Frederick Beiser has remarked with regard to contemporary Anglophone scholarship on German idealism that many of the champions of a normative interpretation of it do not realize that their reading was worked out with greater sophistication and subtlety long ago (10). Can we not draw a parallel comparison to the core message in Laywine’s reading of the B-Deduction – apart from the many details she provides – but referring instead to Cohen’s interpretation?
Undoubtedly, Laywine’s treatment of the Duisburg Nachlaβ in relation to the B-Deduction is a valuable addition to contemporary English-language Kant scholarship. It is this rather than her actual treatment of the Deduction that sets her book apart. It is not an easy read for the casual or beginning student of the first Critique, who might come upon it looking for guidance. But those already quite familiar with the B-Deduction will find a number of observations they will have likely overlooked previously.
Beiser, Frederick C. 2009. « Normativity in Neo-Kantianism: Its Rise and Fall. » International Journal of Philosophical Studies, vol. 17 (1): 9-27.
Cohen, Hermann. 1907. Kommentar zu Immanuel Kants Kritik der reinen Vernunft. Leipzig: Verlag der Dürr’schen Buchhandlung .
A Guide to Kant’s Psychologism (2019) is presented as a more accessible and to-the-point delivery of the interpretive theses Waxman lays out in Kant’s Model of the Mind (1991), Kant and the Empiricists: Understanding Understanding (2005), and Kant’s Anatomy of the Intelligent Mind (2014). This comparatively compact 351-page book promises a unique angle on Kant’s theoretical philosophy for a range of philosophical and scientific audiences. The work is original both in its iconoclastic style and its thesis, which defends Kant’s ‘psychologism’ and interprets the titular empiricist philosophers as precursors thereof. Kant’s account of objective representation in terms of the interrelations between sensation, pure intuition, logic and concepts is argued to be firmly rooted in problems brought into the spotlight by the empiricists, such as animal consciousness and multimodal sensory perception. The book thus combines a historical sensitivity to the genealogy of Kant’s philosophy with the systematic ambition of a new interpretation, not merely of one isolated aspect of Kantian theory of mind but of the way its various doctrines fit together, from pure intuition to apperception to judgment.
The book is organized into two parts. After an introductory chapter preparing the reader for a radical departure from what Waxman presents as an anti-psychologistic consensus in Kant interpretation, Chapters 2-4 chronologically introduce key thinkers from the empiricist tradition and their contributions to the book’s central concept of psychologism. In chapter 5, Waxman uses Wittgenstein to illustrate Hume’s conventionalism, which in Waxman’s view Kant targets no less than rationalist platonism. Chapters 6-10 guide the reader through Kant’s theoretical philosophy. The sequence of chapters, “The Kantian Cogito” (6), “The Logical I” (7), “The Aesthetic I” (8), “The Objective I” (9), and “The I of Nature” (10) is prescribed less by the text of the Critique of Pure Reason than by the conceptual layers of Waxman’s reconstruction. While regularly referring Kant’s insights back to their empiricist lineage, the progression from the ‘I think’ to the objectivity of physical nature also points to Cartesian and Leibnizian influences in Kant’s treatment of logical universality. The chapters on Kant additionally argue that the doctrines of pure intuition and logical form, interpreted as elements of an ‘a priori psychologism’, can accommodate post-Kantian scientific developments in logic, geometry, mathematics and physics. The concluding chapter assesses platonism and conventionalism as the only possible routes of refutation of Kant’s psychologism, as well as indicating how Waxman’s interpretation may be illuminating for contemporary study of the mind.
One of the most original features of the book is that it makes the compatibility of Kantian doctrines with subsequent scientific advances a matter of first importance: “naturalistic theories like Kant’s and those developed by his British empiricist forebears were intentionally crafted to leave open a place for future science on which philosophy can never impinge” (12). The scope of the book is not only to radically overhaul received opinion on Kant’s methodology, and the relation of his ideas to the sciences, but to defend a biologically plausible version of Kant’s account of logical form. In the remainder of this review I will address in turn: (1) Waxman’s definition of ‘psychologism’ and how his use of this concept situates him with regard to other interpreters; (2) his reading of Kant, with a focus on the logical forms of judgment according to the psychologistic approach; and (3) some questions emerging at the interface of Waxman’s naturalistic reading of Kant and the sciences, especially neuroscience and evolutionary biology.
Waxman stresses that Kant’s ability to resolve skepticism about the objective purport of (some of) our representations is rooted in his radicalization of the British empiricists’ methods to accommodate logical universality, modality, and relational concepts. Thus, the book’s overarching thesis about the relationship between its titular figures is that Kant continues what Locke, Berkeley and Hume started, expressed as their common adherence to psychologism. Waxman has in mind a quite specific interpretation of this term:
[T]he task of psychologism is to explicate meanings, with special emphasis on identifying psychological ingredients essential to notions that, in language, are free of any tincture of psychological content. There can therefore be no expectation that the psychological contents adduced as essential to the meaning of familiar notions will themselves be familiar. (147)
Waxman’s compelling take on the psychologistic philosopher’s undermining of rationalist metaphysics is thus that psychologistic elucidation makes us distrust the appearances of the natural language of metaphysics, thereby overturning realist intuitions that the qualities we discern in appearances are properties of mind-independent objects (33f). Psychologism in this sense, Waxman is clear, is not to be confused with the fallacy of explicating the non-psychological psychologically (305). According to Waxman’s use of the term, psychologism means giving contents which are in fact psychological their due explication in terms of mental representations. Kant’s account of how we are capable of “cognitive representation of sense-divide transcending external objects” (264) equips empirical psychologism with the resources to explicate the objectivity and logical universality of our representations, where this is possible. Hence, “any representational content that neither empirical psychologism nor conventionalism can explicate, a priori psychologism can, and what the latter cannot explicate, nothing can” (147).
Waxman introduces his interpretation as a novel defence of Kant’s theoretical philosophy which rejects two pervasive trends in its reception: an anti-psychologistic consensus, and a more general obsolescence consensus that due to revolutions in logic, geometry, and physics, Kant’s philosophy is “a once formidable structure long since reduced to ruin, fit only for piecemeal salvage” (24ff). The latter engenders attempts to clear the respectable theory behind Kant’s project of its psychologistic methods, resulting in a range of approaches to ‘updating Kant’ e.g. by explaining conscious representation using anything from post-Fregean mathematical logic to Chomskyian linguistics, or Roger Penrose’s quantum theory of consciousness (15). Against these kinds of salvage attempts, Waxman argues that a properly psychologistic account of Kant’s theory of objective representation reveals its compatibility with subsequent developments in logic, geometry, physics, as well as neuroscience and evolutionary biology.
Kant scholars may suspect that the anti-psychologistic consensus is by no means universal: for instance, Andrew Brook’s Kant and the Mind (1994), and Patricia Kitcher’s Kant’s Transcendental Psychology (1990) (see also Falkenstein and Easton 1997) make similar critical points about the anachronism of understanding Kantian theory of cognition ‘anti-psychologistically’. For example, “Kant clearly held that his ‘logic’ of the mind is part of what we would now call psychology” (Brook 1994, 6). Waxman takes note of these authors in a footnote, nonetheless signalling his substantive interpretive differences from Kitcher and Brook as well as Lorne Falkenstein and Scott Edgar. The problem, Waxman explains, is that these authors refer to ‘psychology’, which is concerned with “whether and how a representation comes to be in us (empirically, innately, etc.)” rather than ‘psychologism’, which is concerned with “how the representation itself comes to be” (17n). This distinction, in my view, is not as straightforward as Waxman needs it to be in order to distance himself from other interpreters as far as he claims to.
Waxman himself says in the aforementioned footnote that he agrees with Beatrice Longuenesse’s (1998) work on Kant in its essentials, despite her not using the term psychologism (16). This fosters doubts about exactly what defines Waxman’s psychologism as an interpretive approach. If we stipulate that an anti-psychologistic reading foregoes any reference to mental activity in explicating the meanings of objective representations, psychologistic approaches still come in stronger and weaker flavours, which come down to how the genesis of “the representation itself” is construed.
If ‘psychological’ is interpreted weakly, any view that endorses Kant’s references to mental activities such as synthesis and apperception as non-negotiable parts of the story about objective representation, not wholly translatable into an analytic argument, is “identifying psychological ingredients” (147) in discourse which does not overtly suggest them. I believe most interpreters today would uphold this kind of psychological account. Any particular view on what informs Kant’s psychology need not be empiricist in orientation, as one may instead emphasize the scholastic heritage of Kant’s references to mental acts (e.g. Sellars 1967). On stronger versions, e.g. Kitcher (1990), Kant’s ‘psychologism’ qua account of the genesis of objective representations is a ‘proto-cognitivist’ theory which can be re-formulated in light of contemporary experimental methods and results. Waxman’s qualifies as a weakly psychologistic interpretation, in my view, since nowhere in the book does ‘psychologism’ imply ‘experimental psychology’. In the first chapter and the conclusion, Waxman seems to reject mainly Neo-Kantian attempts to place Kant in the service of twentieth century analytic semantics, philosophy of mind and epistemology—explicitly confronting Strawson (1966) in the conclusion (306), but probably also thinking of authors such as Evans (1982) McDowell (1994), and Cassam (1997). However, they present their work as critical reconstructions and selective adaptations of Kantian ideas. Given that Waxman accuses Strawson’s interpretation of Kant of reverting to a kind of platonism, it would have been interesting to hear more about where Waxman agrees and disagrees with existing criticisms of these Neo-Kantian projects, including McDowell’s (1994) ‘naturalized platonism’.
Let us consider the specifics of Waxman’s use of the term ‘psychologism’. Psychologism is presented as a mode of conceptual explication concerned with the psychological content of the meanings of traditionally metaphysical notions, rather than the psychological mechanisms underlying conceptual meanings. Consciousness turns out to be the key ‘psychological ingredient’ reference to which constitutes a psychologistic account. Waxman stresses that all representation requires some degree of consciousness, but that for Kant and his predecessors consciousness is graded from ‘dark’ to ‘bright’ to ‘clear’ (41) and is hence ubiquitous, reaching down to the sensory representations of basic organisms such as molluscs. Another notable feature of Waxman’s psychologism is its gloss on the normative dimension of Kant’s inquiry into our transcendental justification for applying concepts:
instead of proceeding by defining notions in terms of other notions without regard to whether there is, or even can be, any corresponding conscious representation, psychologism [has] the express aim of seeking out such representations. If the search reveals that a notion owes none of the contents indispensable to its meaning to consciousness, then its scope of application is nowhere limited by it (10).
Conversely, if the notion “can be shown to be beholden to consciousness for any of the ingredients essential to its meaning—ingredients at least implicit, but often explicit, in definitions—then its scope of application is limited accordingly” (10).
In general, the distinctiveness of Waxman’s approach lies in his wholesale engagement with Kant’s more unwieldy terms and concepts, and readiness to endorse Kant’s claims as literally about psychological reality—rather than about concepts, the brain/organism, or the objective world. Despite his psychologistic approach having a foot in psychology and a foot in explication, Waxman is content neither with purely conceptual or exegetical arguments, nor ‘proto-cognitivist’ claims that Kant anticipated cognitive science on this or that front. But he ultimately agrees with many other interpreters that any staunchly ‘anti-psychologistic’ reading is a non-starter, as well as joining Kitcher (1990) and Brook (1994) in thinking that a naturalistic perspective on Kant may be able to contribute to contemporary (experimental) psychology and theory of mind.
I now turn to Waxman’s reading of Kant and his empiricist forerunners. Chapter 2 argues that Locke sets the stage for the psychologistic approach by conceiving consciousness on a scale starting with primitive animals: “Since in the entire absence of sensation no consciousness of any kind seems possible … its terrestrial advent would presumably have coincided with the appearance of the first sensation” (54). In Chapter 3, Berkeley is credited with “extending imagination into the cognitive sphere, thereby for the first time crossing the line separating reality from fiction (where Hume and Kant would follow). This is because the ability to represent space as transcending the divide between sight and touch is indispensable to all cognition of the physical” (85). For Kant, this consideration leads to the thesis that there is nothing in sensation (visual or tactual) that is intrinsically spatial, so the representation of space must be constructed in pure intuition (201). Chapter 4 links the foregoing to Hume’s well-known skeptical challenges to relational concepts:
By shifting the basis of belief in relations from objective experience to subjective feeling, Hume moved the topic from epistemology to psychology where, instead of needing to be justified by evidence and to follow as a conclusion from premises, belief is determined purely affectively, by association-constituting feeling, and nothing else (96).
For Waxman, Hume was the first to psychologistically attribute fully general concepts such as the uniformity of nature and the general causal maxim to humans and non-human animals alike (114). Waxman illustrates the theoretical development from associationism to Kantian a priori psychologism in terms of a speculative evolutionary development from a creature
capable of the kind of highly sophisticated, behaviourally efficacious conscious mentation that Berkeley and Hume devised their associationist psychology to explain. Thanks to some fortuitous mutation or other alteration in its genome, its progeny included creatures capable of the representation ‘I think.’ Having no evident selective advantage by itself, this neural capacity presumably could have established itself in the population only as a spandrel piggy-backing on some genetically connected trait that earned its evolutionary keep. (177)
Eventually enabling complex propositions and inferences, Waxman contends, the ‘I-think’ representation would have enabled behaviours with adaptive value. This evolutionary story complements Waxman’s interpretation of Kant’s logical forms of judgment as more basic than language and indeed making language possible:
The ability of linguistic propositions to blend with non-linguistic is, in its way, no less amazing than the mathematizability of nature. But isn’t this exactly what we should expect if language was originally crafted by creatures already fully conversant in the use of non-linguistic propositional representation, transcendental judgments included? (315).
The chapters of Part II centre around Kant’s psychologization of logic in relation to sensory and propositional representation. Waxman details how Kant adapts from Descartes the contentless representation ‘I think’, which “in and of itself, has no content [and hence] cannot be suspected of having borrowed any, whether from language or anywhere else” (158). Being non-linguistic, it is not attributable to convention and thus provides a purely psychological basis for logical generality that synthetically unifies all possible representable contents (174). According to this notion of pure apperception, Waxman argues that Kantian logic is based in a universal self-consciousness which includes “the totality of logical structures universals enable us to form—propositions, inferences, narratives, et al.” (149). Unfortunately, Waxman doesn’t elaborate on whether self-consciousness is for Kant always a form of ‘bright consciousness’, or could exist in ‘darker’ shades as he argues it does for Hume (120).
Kant brings what Waxman calls “intelligence”—which amounts to consciousness of universality and modality—to Humean representations of general relational concepts which lack any awareness of the “logical universe of possible representations” (171). Kantian concepts “enable us to consciously represent each and every associative combination as a grouping of denizens of the logical universe that are thereafter sortable not only by their sensible/imaginable properties but by their logical ones as well” (171). According to Waxman, Kant’s logical forms of judgment are what “make any ‘I think’-generated concept combinable with any other such concept in a single act of thought, or propositional representation” (167). Conceptual representation is thus explicated as the gradual elimination of degrees of “logical freedom” (150) by restricting the space of logical possibilities to what is representable given the content supplied by sensation and association.
According to Kant’s psychologization of logic, fixing the position of a term as subject or as predicate in a categorical judgment is a psychological act starting with the bifurcation of the logical universe (170). However, such a “logical form by itself cannot guarantee that a proposition will result. In particular, even if a would-be proposition is logically well formed, it would still fail to be a genuine proposition if, for other than strictly formal logical reasons, it cannot be thought without generating not just falsehood but one or another species of impossibility, e.g. … ‘water is not always H2O’” (175). For Waxman’s Kant, categorical form permits the unification of concepts in a single consciousness, but it does not allow a unified representation of the resulting propositions (175). In addition to the assertoric modality, then, we need Kant’s problematic modality, which asserts merely the relation between two propositions, suspending judgment on the propositions themselves. The logical form this assumes in us, hypothetical judgment, relates problematic propositions as ground to consequent in the assertoric modality. Together, hypothetical and categorical forms enable any combination of the totality of possible propositions to be unified in a single conscious act (175f).
Waxman mobilizes this account to show that Kant (1) does not restrict intelligence to language-using organisms, nor (2) does he impose our parochial linguistic structures on his model of the basic “building blocks of propositional thought” (22). Waxman takes his elucidation of Kant’s transcendental logic to counter both misconceptions:
Intelligence can be accorded to any creature, actual or possible, that is capable of pure apperception … even including beings so asocial as to be devoid of anything remotely analogous to language or socially grounded symbolic communication of any kind. This is not to deny that Kant regarded all non-human animals known to him as incapable of apperception and therefore unintelligent. (172)
But, Waxman adds, “that does not mean he would have persisted in that view had he known what we know today” (172). This is an interesting speculation, implying that animals have some form of awareness of universality and modality—not in Hume’s sense, but in Waxman’s more demanding Kantian sense of apperception—“an a priori logical universe that quite literally encompasses all possible conscious representations” (169). I fully agree that Kant’s logic is non-linguistic, but I want to know more exactly how Waxman understands our present knowledge of animal intelligence such that it could make Kant change his mind on this controversial topic.
Waxman regularly refers to his lengthier engagements with Kant’s first Critique in footnotes. These references will be necessary for scholars seeking to determine Waxman’s position on exegetical debates. Part II succeeds in presenting Kant’s account of logical forms and concepts as centrally relying on consciousness, and hence psychologistic, while emphasizing that propositional representation need not be construed as an evolutionary leap separating humans from other animals. The combination of these theses makes for an original, stimulating addition to works on Kant’s first Critique.
I now have some remarks on the book’s intriguing but somewhat ambivalent references to the sciences. Waxman’s general stance that the insights of Kantian psychologism extend beyond their own scientific and philosophical context is very ambitious, and I’m not sure his case is equally strong for each of the sciences he addresses. Waxman makes a convincing case that non-Euclidean geometries cannot falsify any of Kant’s statements about pure intuition, as the latter pertain to the necessary features of any representation of space, rather than any particular geometry:
the formal intuition of space is not only neither Euclidean nor non-Euclidean but completely indeterminate as regards number, limit, distance, metric, part-whole relationships, and everything else that makes space suitable for properly mathematical representation or objective representation of any kind (208-9).
When it comes to the life sciences, however, Waxman’s perspective is (perhaps inevitably) more in tension with the scientific context Kant was writing in.
Waxman is clearly not neutral on matters of philosophy of consciousness, stating at the outset that consciousness is for his purposes identical in existence to its neural correlates (13-14), but that there is nonetheless a distinct psychological reality (14, 57, 159, 162, 177f). As this commitment is not compared to any alternatives, I am curious as to why he has opted for this particular form of identity theory to defend the biological plausibility of Kant’s psychologism, rather than functionalism or some form of emergentism. Polák and Marvan (2018), for example, defend the view preferred by Waxman that neural correlates are not in a causal relation to conscious states but an identity relation—a position which philosophers of mind might like to see defended more explicitly in connection with Kant’s views. Also, they may ask whether there is a specifically Kantian motivation for understanding the “mystery of consciousness” as Waxman does, that is, “a purely physical existence that is at the same time irreducible to physical reality” (14). Waxman clearly wants a naturalistic position to complement his psychologistic interpretation, but he also does not want to impose too much recent theory on the historical theories. Given this concern, psychophysical parallelism—an early form of identity theory espoused in different (including Neo-Kantian) versions by German philosophers from Fechner to Feigl via Riehl and Schlick—could illuminate Kantian psychologism from this side of Darwin and experimental psychology (see Heidelberger 2004, Ch. 5). Of course, it would be possible for most Kant scholars to profess neutrality with respect to these debates, but Waxman has—to his credit—set different standards.
Waxman’s decision to adhere to Kant’s original terminology and to steer clear of issues of translation in one sense makes the book smoother. The book conveniently contains a glossary, and tailor-made terms such as ‘AUA [analytic unity of apperception] concepts’, and ‘dark consciousness’ helpfully remind us that a term is not being used in its familiar sense. Waxman’s capacious use of ‘consciousness’ clearly works in his favour insofar as Modern and Kantian philosophy become much more relatable simply by lowering the threshold (as we understand it) on what sensory and cognitive states count as conscious. Yet it would have helped for Waxman to illustrate the general features of (what Descartes, Locke and Kant viewed as) ‘dark consciousness’ in more descriptive neural or psychophysical terms—not simply because readers may have difficulty forgetting the current meanings of such terms and the controversies attached to them, but because it would improve the book’s case for making psychologism compatible with contemporary biology.
One of the book’s most interesting features is its rapprochement between Kant and Darwin. Frequent references to phylogeny and the evolutionary plausibility of psychologism (as contrasted with platonism and conventionalism) evidence Waxman’s eagerness to integrate Kant’s insights into a post-Darwinian landscape, which I take to be a very important, relatively neglected project:
[N]ot only is Kant’s psychologism consistent with evolution, it actually spotlights suitably primitive forms of empirical consciousness that would gain adaptational advantages from a priori consciousness. Most basically, formal intuitions can easily be conceived to be of use to minds grappling with the challenge of combining external sensations into a single, unified external sense capable of providing immediate access to sense-divide transcending objects. (306)
This illumination does not go both ways, however, since Waxman also says that transcendental consciousness is entirely outside the scope of scientific explanation (295)—“because the subjectivity constituted by apperception … becomes part of the explanation of the physical, it cannot itself be explained physically on pain of circularity” (23). This claim recalls circularity charges against naturalized epistemology and logic, which in my view can be convincingly refuted by pragmatic and holistic considerations. It also blatantly contradicts the book’s initial claim that “[t]he mental is causally and in every other way fully determined by its physical underpinnings, and so is in principle fully explicable by science” (21). This is surprising, after seeing Waxman go to great lengths to argue that although a priori, the most innovative aspects of Kantian philosophy are features of our natural constitution, different in complexity but not radically different in kind from the sensory sensitivity of an oyster. Placing a priori elements of Kantian psychologism beyond the reach of current and future biological explanation strikes me as a missed opportunity, and again makes me doubt what is meant by psychologism. Empirical research in animals can indicate how consciousness might have arisen, and in a footnote Waxman considers but then discards a few such accounts based on their comparatively restrictive definition of consciousness (63n). The reader may suspect that casting the net more widely would reveal theories more congenial to Waxman’s evolutionary take on subjectivity, such as Godfrey-Smith (2016).
The following passage is a good example of Waxman’s ambivalence towards his own naturalistic outlook:
The connection between the subjective psychological reality of consciousness and the objective physical reality of neurophysiology is a complete mystery in both directions, today and quite possibly for some time to come. Thus, Kantian logical forms of judgment pose no special mystery but instead are best regarded as simply an additional species of phylogeny-dependent neuro-psychological reality in addition to sensations, emotions, dreams, and the rest. (178)
Because Waxman has not restricted himself to purely exegetical argument, or the Kitcherian ‘proto-cognitivist’ angle described earlier, the reader may wonder how literally such claims are to be taken. One cannot but agree that (something similar to) Kantian a priori capacities such as the logical forms of judgment must have evolved somehow, just as there must be some neural correlates for the empiricists’ associative psychologies, as indicated in Part I. But the empirically-minded reader will want to know in more detail how forms of judgment (or their successors in contemporary science of mind) could be modelled and studied, if they are on a par with sensations and emotions.
A Guide to Kant’s Psychologism is bound to appeal to diverse philosophical audiences for its fresh take on Kant’s theoretical philosophy as a priori psychologism. It is also a lively, articulate instance of philosophical storytelling. Waxman avoids approaching Kant through the lens of contemporary philosophical problems where semantics, epistemology and metaphysics are concerned. When it comes to the sciences, however, the book makes us acutely aware of pieces of the puzzle of the mind in nature that Kant simply cannot have anticipated. Kant’s understanding of his current science needs to be confronted with today’s sciences in order to address all of the questions raised by an interpretation of Kantian philosophy as naturalistic a priori psychologism. Over the course of Waxman’s book, frequent references to neural correlates and phylogeny habituate the reader to seeing our biological reality in the same conceptual space as Kantian doctrines, which is surely a step in the right direction. As is perhaps Waxman’s intention, it is left to the reader to ponder the convergences and divergences between Kant’s account of the mind and current scientific knowledge, perhaps especially the life sciences. The book offers its non-Kantian readers a challenging, raw encounter with Kant’s theoretical philosophy, and will leave Kant scholars much to think about both on the old problem of psychologism and new ones arising from Waxman’s brand of naturalism.
Brook, Andrew. 1994. Kant and the Mind. Cambridge, New York and Melbourne: Cambridge University Press.
Cassam, Quassim. 1997. Self and World. Oxford: Clarendon Press.
Evans, Gareth. 1982. The Varieties of Reference. Edited by John McDowell. Oxford: Clarendon Press and New York: Oxford University Press.
Falkenstein, Lorne and Easton, Patricia, eds. 1997. Logic and the Workings of the Mind: The Logic of Ideas and Faculty Psychology in Early Modern Philosophy. Atascadero, CA: Ridgeview.
Godfrey-Smith, Peter. 2016. “Mind, Matter and Metabolism.” Journal of Philosophy 113 (10):481-506.
Heidelberger, Michael. 2004. Nature from Within: Gustav Theodor Fechner and his Psychophysical Worldview. Translated by Cynthia Klohr. Pittsburgh: University of Pittsburgh Press.
Kant, Immanuel. 1998 [1781/1787]. Critique of Pure Reason. Edited and translated by Paul Guyer and Allen W. Wood. Cambridge: Cambridge University Press.
Kitcher, Patricia. 1990. Kant’s Transcendental Psychology. New York and Oxford: Oxford University Press.
Longuenesse, Béatrice. 1998. Kant and the Capacity to Judge: Sensibility and Discursivity in the Transcendental Analytic of the Critique of Pure Reason. Translated by Charles T. Wolfe. Princeton: Princeton University Press.
McDowell, John. 1994. Mind and World. Cambridge, MA and London: Harvard University Press.
Polák, Michal and Marvan, Tomáž. 2018. “Neural Correlates of Consciousness Meet the Theory of Identity.” Frontiers in Psychology 24:1269. DOI https://doi.org/10.3389/fpsyg.2018.01269.
Sellars, Wilfrid. 1967. “Some Remarks on Kant’s Theory of Experience.” The Journal of Philosophy 64 (20), Sixty-Fourth Annual Meeting of the American Philosophical Association, Eastern Division (Oct. 26, 1967): 633-47.
Strawson, Peter F. 1966. The Bounds of Sense. London: Methuen.