This slim volume contains eight papers originally presented in the AHRC-funded project Transcendental Philosophy and Naturalism (2005–8), directed by the late Professor Mark Sacks. The broad variety of approaches and topics reflects admirably Dr Sacks's own scholarship on the nature of transcendental philosophy. The editors should be congratulated on successfully carrying the project through to completion.

The subject matter ranges over the intersection of naturalism and transcendental argument in epistemology, philosophy of mind, logic, moral philosophy and radical scepticism. As one might expect, Kant's philosophy provides the natural centre of gravity for the text.

Although the contributions in this volume are unlikely to displace defining statements on transcendental argument found in the work of Walker, Stroud or Stern (see comments below on Stern's included contribution), Transcendental Philosophy and Naturalism offers an engaging encounter with the omnipresent force of contemporary naturalism as it bears on the standing of transcendental argument. Whereas in the mid-twentieth century the debate tended to focus on the character and limits of transcendental argument, this volume is a timely expression of the extended reach of the naturalist worldview. Assimilation, rather than limitation, may more accurately capture the spirit of the contemporary naturalist encounter with transcendental philosophy.

Hilary Kornblith's contribution, ‘Reasons, Naturalism, and Transcendental Philosophy’, offers a bald expression of the hard edge of this naturalist view. Kornblith's primary line of attack falls on the approach developed by McDowell – with its inspiration in Sellars and Kant – that cognition answers to the world from within the space of reasons. Kornblith begins with the governing thought that an understanding of belief demands treating knowledge as a ‘natural phenomenon in the very same way that we examine other natural phenomena; that is, empirically’ (96). Although well executed, Kornblith's early and undefended appeal to ‘information processing’ (96–7) as the central problematic fails to engage its target. Where one views propositional content as simply given in causal descriptions of physical states, the naturalist triumph is an unearned prize. The contested domain concerns the source and standing of the inferential aspects of belief, and how these relations are, arguably, constitutive of the content of perceptual states. Neither Kant nor McDowell denies that it is the causal contact of our sense organs with the world that prompts belief. Neither queries the fact that our ability to adjust and refine belief develops, gradually, as we move from infancy to greater levels of self-conscious reflection. The distinctly transcendental concerns pertain to how the deliverances of receptivity are drawn into beliefs that answer rationally to the world. The relation of causal interaction and the informing nature of belief is left relatively untouched by Kornblith's challenge. This is a double pity insofar as McDowell himself lays claim to a form of naturalism that operates within the space of reasons. Exploration of this larger clash of competing conceptions of naturalism would have proved particularly interesting.

Donald Davidson's work, closely allied as it is with themes exploited and furthered by McDowell, forms the subject matter of Adrian Haddock's ‘Davidson and Idealism’. This article is one of many in the collection that looks to the influence of Kant's transcendental idealist doctrine as a way of capturing the idea that the demands of ‘mindedness’ entail definite boundary conditions for belief's bearing on the world. Haddock's chief concern centres on the complex crosscurrents present in the work of Davidson, Kant and Wittgenstein on the topic of the transcendental constraints judgement enforces. Haddock develops the thought that reflective practice that discloses that we ‘enjoy a priori knowledge grounded in the way we are minded’ (38) is a theme deeply implicated in Davidson's celebrated ‘On the Very Idea of a Conceptual Scheme’. Although the argument connecting Davidson to Kant through Jonathan Lear's reading of Wittgenstein is somewhat circuitous, as the author himself admits, the thread developed is well worth the effort. Given Davidson's denial of the intelligibility of scheme/content dualism, and that Kant is read habitually as the father of that dualism – ‘thoughts without content are empty, intuitions without concepts are blind’ – the assertion that the linkage between the two is sufficient to nest Davidson within the extended family of transcendental idealist positions is a provocative claim. This contribution will prove valuable for contemporary Kant scholars, adding to a growing awareness that Davidson's work sounds a contemporary echo of Kant's empirical realist claims about why empirical belief is answerable only to a single unified spatial/temporal world of objects determined uniquely by causal engagement. One might fault the paper only for its omission of discussion of the enabling role truth-conditions play in Davidson's account of interpretation and how this anchors Davidson's challenge to empiricist appeals to sense-data as information bearers.

Peter Sullivan's ‘Is Logic Transcendental?’ explores sceptical challenges trained against the authority of logic. Sullivan's aim is to test the ‘easy’ answer of Thomas Nagel (among others) to the sceptic's attempt to throw into doubt the universal and necessary nature of judgement. Nagel's response is modelled on the thought that the mounting of a sceptical argument against the authority of logic is ultimately self-defeating – in order to be successful, the sceptic requires the very structures of thought that are being drawn into question. Sullivan suggests that this easy answer, while it may offer an exemption from the sceptic's challenge, does not adequately carry the day, as ‘being exempted from an exam seems less creditable than passing it’ (160). The worry, emanating from what Sullivan describes as a ‘robustly realist setting’, concerns whether ‘our way of thinking, and the laws whose acknowledgement essentially structure it, [is] the only way there could be, or the only system of laws that could play that distinctive structuring role’ (162). Sullivan finds in Sacks's work a neutralizing strategy that closes the alleged realist gap between the demands of ‘any experience we can conceive of’ (180) and the unrestrictive generality of ‘any experience’ in toto. This strategy stands or falls largely on the claim – clearly Kantian in flavour – that the sceptic needs to discharge the burden of the extension of universality beyond the range of ‘possible experience’ if the alleged gap is to be invoked. The transcendental strategy effectively reduces this ‘gap’ to zero. Borrowing from Sacks, Sullivan then reclaims Nagel's ‘easy’ answer although now with a fresh framework that moves from ‘exemption’ to a new ‘constituting’ metaphor for modelling the expressive force of judgement.

Sullivan's defence of logic serves as a useful reminder of the authority Kant assigns to his table of judgements. Kant, it will be recalled, offers little explanation for his table of judgements apart from expressing why the four modes of judging (Quantity, Quality, Relation and Modality) share a threefold mode of expression (B107–10). Although recent Kantian scholarship has helped clarify the important role of the table of judgements, Sullivan's contribution here should add an additional arrow to that transcendental quiver.

The contributions from Robert Stern and Patricia Kitcher are, to some measure, embryonic statements of work more fully realized in material subsequent to Sacks's 2005–8 project (see Stern Reference Stern2010 and Kitcher Reference Kitcher2011). Stern's perceptive reconstruction of Korsgaard's argument for the ‘value of humanity’ in ‘Reflections on Korsgaard's Transcendental Argument’ offers a productive take on her celebrated appeal to practical identity as a relevant ground for valuing oneself as a rational agent. Stern acknowledges that his transcendental reconstruction is at odds with Korsgaard's occasional appeal to the psychological category of ‘need’ as the basis for the adoption of identity. Given the challenges Stern has offered elsewhere to Korsgaard's work, this sympathetic reconstruction of this portion of her approach should serve as a valuable resource for those interested in housing Korsgaard's interpretation within the transcendental stable.

Kitcher's ‘The Unity of Kant's Active Thinker’ attempts to reclaim, by means of transcendental concerns Kant develops in the Transcendental Deduction, a substantive thesis concerning the relation of mental content and the requirement of reflective consciousness. For Kant scholars this contribution may be the most satisfying insofar as it tightly tracks detailed Kantian themes. At the heart of Kitcher's approach – here and elsewhere – is the idea that consciousness is fundamental to ascriptions of rational judgement and, more generally, to the ground for ascribing mindfulness to others. The cutting edge of Kitcher's treatment is the thought that non-conscious systems of symbolic combination (various AI approaches, for example) are inherently blind – and therefore inadequate – to the requirements of mental unity and understanding that stand as necessary conditions for the possibility of belief that is rationally answerable to the world. Even in the best case example where self-consciousness is modelled by ‘reflective’ programming that monitors the symbolic states of a system, ‘[s]ince [mere] reflecting or monitoring consciousness [can] not account for the relations of epistemic dependence across the states of a rational cognizer, it [can] not explain rational cognition either’ (72). Here again, Sellars's argument invoking the demands of the space of reasons intrudes on the naturalist programme of a purely causal system of engagement with the environment.

The challenge posed by scepticism to the naturalism/transcendental divide is taken up most directly in Penelope Maddy's ‘Naturalism, Transcendentalism, and Therapy’ and in A. W. Moore's ‘Vats, Sets, and Tits’. Maddy develops an argument in favour of the priority of what she calls ‘Second Philosophy’. She identifies this with a form of methodological naturalism that endorses a scientific vision of knowledge where the ‘second philosopher’ begins ‘with perception and common sense, gradually refines her observations, devises experiments, formulates and tests theories, always striving to improve her beliefs and her methods as she goes along’ (121). The general tone of Second Philosophy is consonant with what one finds in Arthur Fine's work in the philosophy of science. Like Fine, Maddy attacks First Philosophy – represented by positions such as Bas van Fraassen's constructive empiricism and Kant's transcendental idealism – as unhealthily directed to second-order concerns about epistemic conditions of enquiry that are unlikely ‘to convince [one] that a legitimate enquiry is involved’ (126). Looking approvingly to the later Wittgenstein, and to Austin, she argues that enquiry operates naturally and best when its investigations are empirically driven. Higher-order reflection that motivates First Philosophy's concern with justifying empirical claims from a transcendental standpoint – to secure an authority for empirical belief that is inoculated against worries of scepticism etc. – is identified, on this reading, with a set of extraneous commitments. We are reminded of Wittgenstein's claimed disdain for transcendental enquiry: ‘We have got onto slippery ice where there is no friction and so in a certain sense the conditions are ideal, but also, just because of that, we are unable to walk. We want to walk: so we need friction. Back to the rough ground!’ (quoted (142) from Philosophical Investigations, I, §107). In a similar vein, Maddy develops a response from Austin which, directed against First Philosophy, aims ‘to get us over a certain mental tick’ (152).

Naturalism, thus understood, offers a ‘therapeutic’ response designed to tranquilize the offending over-reach of reflective judgement. Maddy's form of naturalism essentially pathologizes the impulse towards transcendental reflection – regarding it as an unnecessary, and suspect, conceptual overlay. As with Fine, one may find this interpretative strategy ultimately unsatisfying: a kind of charmed sleepwalk through the thicket of reflective practice. While one may grant that the character of transcendental reflection can fail to meet the Kantian enthusiasts’ desire for an Archimedean fixed point, it is less than a mock victory to cast off transcendental reflective practice merely on the deflationary grounds that immediate, natural enquiry (whatever that is) will look ‘curiously’ on ‘Kant's transcendental purposes … and how they are to be achieved’ (134).

Moore discusses Putnam's now well-trodden ‘brain-in-a-vat’ engagement with scepticism. He develops three distinct versions of the radical sceptical worry: the Vat Paradigm, the Set Paradigm and Moore's rather unique Tit Paradigm. This last version points to what Moore identifies as the most worrying form of radical scepticism, a form bound up with transcendental idealism in which the subject is unable to assert knowledge claims beyond the domain of the empirical (‘phenomenal bubble’) and yet is able ‘to achieve an insight, beyond his powers of expression, into the possibility that he is subject to precisely such limitations’ (47). Moore ultimately rejects the plausibility of this position, suggesting that, while we can understand the temptation to think that we could be in such an epistemic position, this temptation does not itself vindicate the view because ‘[t]he fact is, we cannot treat that to which our thinking is answerable as some kind of phenomenal bubble, for a reason famously articulated by Wittgenstein: “in order to be able to draw a limit to thought, we would have to find both sides of the limit thinkable (i.e. we should have to be able to think what cannot be thought)” ’ (51).

There is a faint echo of Maddy's approach here in so far as Moore is suggesting that the best we can do is to offer a diagnosis of the temptation to model the idea that thought can meaningfully extend beyond the limit of the sphere of the knowable. Of course Moore, unlike Maddy, makes no claims against transcendental analysis per se – he is in fact deploying it in the paper. Moore's point is rather that we can neither endorse nor even concede the coherence of the idea that thinking ‘is answerable only to our phenomenal bubble’ (52; emphasis added). With the relevant Wittgensteinian intuition in place, we come to see that the idea of the experiential as bounded by the sphere of the merely ‘thinkable’ is a thought too far. As such, the possibility of a form of radical scepticism along the lines made possible by the governing Kantian thought that our empirical beliefs pertain merely to appearance and not things-in-themselves can be set aside.

Where this leaves the Kantian idea of experiential belief as a conditioned achievement – with the logical space of the unconditioned flowing immediately from this pairing – is left as an uncomfortable dangler. Moore's brief concluding suggestion that we might view things-in-themselves as ‘syncategorematic’, as ‘totally free of perspective, whether cultural, biological, or of any other kind’ (52) would need a good deal of further elaboration if it is to have significant purchase.

The volume is rounded off with a contribution from Joel Smith titled ‘Strawson on Other Minds’. This essay serves as a useful reminder of the seminal work of Strawson, and his generation in the mid/late twentieth century, as they tackled the transcendental/empirical divide. Smith defends Strawson's use of transcendental arguments for the existence of other minds against a range of challengers. Ultimately, though, he finds Strawson's argument wanting insofar as it assumes a privileged role for ‘first-person first’ (203ff.) ascriptions of the relation of mindfulness and behaviour that Smith finds unmotivated. Smith then directs attention to an empirical reconstruction of Strawson's approach. He suggests that an empirical canvassing of all non-first-person accounts might reveal it as a null set and ‘that any empirically plausible version … was committed to a first-person first account’ (204). Smith ultimately rejects this naturalized version of Strawson's argument on empirical grounds.

Smith's conclusion is a fitting final statement of the enduring methodological gulf between transcendental argument and naturalism's empiricist canon.