There are two elements of Kant’s thought that perhaps sit uneasily together. The first is his apparent commitment to a philosophy that is adequate to what he sometimes calls ‘common’ aspects of cognition or understanding (cf. B3; A184/B227; Groundwork, 4: 391, 402, 412; for some recent discussion see Grenberg Reference Grenberg2013; Sticker Reference Sticker2014; Callanan Reference Callanan2019). The second is his commitment to the value of a self-consciously exercised capacity for rational reflection. Interpreters traditionally focus upon the latter element, with its emphasis on self-directed questions (‘How ought I to act?’ etc.), demands for intellectual autonomy (‘Dare to know! Think for oneself!’ etc.) and arguments for internally determined tribunals of judgement as a necessary condition for rational responsibility. What this focus neglects, as generations of critics have noted, is the apparent value (and perhaps rationality) of ordinary non-reflective responses to various situations. Merritt’s very impressive book aims to defend Kant against hyper-intellectualist mischaracterizations and objections by focusing on his account of reflection as it emerges in a range of his texts, from metaphysics to logic to anthropology.

Merritt’s exegetical aim is to show that there is enough evidence to show that Kant has an account of reflection that points away from the necessity of continuous acts of self-conscious introspection into the grounds of one’s judgements. Merritt’s Kant holds that one might simply judge well of the world in a non-introspective manner just because one possesses a good reflective character, and this rational character ‘infuses the whole package’ of ground-level responses (p. 5). A reflective status for judgements can be achieved without those individual judgements themselves being accompanied with distinct mental acts of reflection. Merritt thereby links reflection to virtue and to a status that is attributed to a person considered as a whole rather than focusing on the reflective character of this or that faculty of the mind. Merritt’s scrupulous and scholarly work is, I think, valuable in drawing our attention to the importance of these kind of considerations for Kant’s overall picture of the human being. In order to do justice to that overall picture one must also account for the elements of Kant’s thought that led readers to form the introspective and intellectualist interpretation in the first place. I will first outline her strategy regarding the first point and then briefly raise some questions regarding the latter.

A rough outline of the book’s aims might be as follows. Kant occasionally appears to make a generalization in the form of what we might call the reflection claim:

One might hesitate from endorsing (R) on the grounds that it is regularly empirically refuted, whether (R) is interpreted constitutively as a property that all judgements attain, or normatively as a status all rational and responsibly made judgements ought to attain. Someone who wished to defend the reflection claim must find grounds for softening its initial implausibility.

The matter is complicated by the possibility that there isn’t a univocal sense of ‘reflection’ at work in Kant’s texts. Merritt claims that Kant in fact holds to two different notions of reflection (which she labels ‘reflection-c’ and ‘reflection-n’ respectively (pp. 18–19): one which concerns a frequently tacit state associated with the task of thinking itself and another which concerns the more self-conscious judging to which we might attribute a normative status. Merritt claims that Kant intends for (R) to hold with regard to reflection-c. It can sometimes look like explicit acts of reflection-n is what Kant is after, but since this would surely be ‘a highly sophisticated cognitive skill of some kind, and hardly the sort of thing that could be required of anyone with each and every judgment’ (p. 33), this reading looks problematic.

Merritt claims that (R) is more plausibly interpreted as holding with a ‘merged’ account of reflection and claims that reflection-c entails some level of enaction of reflection-n. Reflection has a normative dimension, since it involves an activity that need not accompany every judgement but rather must accompany every sound judgement; on the other hand, however, Merritt conceives of this notion of reflection as ‘internal to sound judgment: it is nothing separate from considering the objective cognitive question in the right spirit, or with the right frame of mind’ (p. 8, emphasis in original). It is a normative status one achieves then, but not necessarily by adopting a distinct cognitive attitude towards one’s act of judgment. This conception of reflection involves ‘a subjective orientation to … thinking that consists in taking a certain interest in oneself as the one who settles the question’ (p. 8). If we take reflection in this way, we can begin to see how Kant might have held (R) since it is more plausible that each and every sound judgement is one made in the right frame of mind, so understood.

The structure of the book is as follows. Chapter 1 examines Kant’s distinctions between types of logic in order to set out the claim that (R) refers to reflection-n. The remainder of the work sets out to argue ‘for a conception of reflection-n as embedded in appropriately developed cognitive capacities’ (p. 52). The claim is not that explicit reflection might not be necessary for a ‘life lived well’ but merely that non-explicit reflection is sufficient in explaining the truth of (R). Chapter 2 sets out Kant’s ‘maxims’ of ‘healthy human understanding’ (such as ‘to think for oneself’) in the third Critique and elsewhere. Chapter 3 works on material both from the first Critique and the Anthropology to argue that the constitutive and normative conceptions are both intertwined in Kant’s conception of judgement. Chapters 4 and 5 present the case for what Merritt calls the ‘specification thesis’, which argues that the rules of the healthy understanding are general rules for a healthy theoretical and practical reason, and that the practical employment of these rules is a specification of the general understanding. Chapters 6 and 7 argue for the connection between reflective character and moral virtue considered as a skill.

As should be obvious, this is a very stimulating and erudite book, one that will be rightly regarded as central to this topic for years to come, and I would recommend it to Kant scholars working on any issue in Kant’s Critical thought. Merritt has a rare facility with primary texts ranging from the first Critique to the logical works to the Anthropology to the Metaphysics of Morals. As I cannot do justice to the detail contained in its pages here, I will instead only gesture towards some initial questions of a general nature. For instance, I would have liked to have seen more to motivate the problem to which the book devotes itself. As Merritt notes, Kant certainly asserts something like (R) in the Amphiboly, and it appears elsewhere in the Jäsche Logic and other logic notes. Merritt observes throughout that the sources like the Jäsche Logic should be handled with care (Merritt is a paragon of scrupulousness when it comes to handling sources), yet even considered together this is not a lot of textual evidence and, moreover, this evidence is open to different interpretations.

To focus upon the primary example: it is not obvious to me that (R) as it is stated in the Amphiboly is a demand that every token judgement requires reflection. The text there seems consistent to me with it being a demand for a single reflection into the cognitive sources related to a judgement-type (specifically, a priori judgements). It is also important to note that the claim is made within the pages of a metaphysics text – the question of who is being addressed is surely relevant here. When Kant says that ‘[t]his transcendental reflection is a duty from which no one can escape if he would judge anything about things a priori’ (A263/B319) he is making the claim that the task of determining whether concepts belong to understanding or sensibility is a necessary condition of validating different classes of a priori judgements. This duty of the metaphysician might be that all metaphysical judgement-types require a single prior reflection on their general validity, yet it does not follow from this textual evidence that Kant held that the same duty holds for each and every sound judgement made by each and every subject in each and every context.

Merritt motivates her concern with (R) by consideration of an anxiety with Kant’s ethics, which is that the categorical imperative test might be put forward as a token act of reflection that is to accompany each token sound moral judgement. If we accept this, then Kant’s ethics seem absurdly intellectualized; if we do not then it is unclear when rational reflection is relevant to ethical life and when it is not. Merritt is committed to giving a reading that preserves the Kantian thought that ‘a life that is lived well can only be reflective through and through’ (p. 7), yet without this being achieved through an intellectualized reading of (R). It would have helped, however, to have had more general discussion at the beginning concerning why (say) a type-reading of (R) would not satisfy the demand for a life that is ‘reflective through and through’.

In the Amphiboly passage mentioned above, the context seems to suggest that the type of reflection universally demanded concerns whether a concept belongs to this or that faculty. This looks like a job for introspective reflection – how would one achieve this task via the ‘typically tacit handle that one has on oneself as the source of a point of view on how things are or what is worth doing’ (p. 28)? Similarly, the texts regarding the maxims of a healthy understanding seem to invite a plain interpretation whereby they are something for the subject to deploy consciously in their deliberative lives and are less obviously read as the hope that these maxims would infuse one’s ground-level thinking and become part of a kind of tacit ‘knowing how to make good use of one’s cognitive capacities’ (p. 48, emphasis in original). If that were Kant’s meaning here, one would think that he would clarify it, since the plainer reading of ‘thinking for oneself’ is just the demand to keep asking oneself outright ‘is this right?’, ‘Should I accept it?’, ‘What if everyone did that?’ and so on.

When considering how this account relates to the practical philosophy there are similar concerns. There is a well-known passage in the Groundwork (one I discuss in greater detail in Callanan Reference Callananforthcoming) where Kant outlines what he calls the ‘natural dialectic’ (4: 405). This is, roughly speaking, a phenomenon whereby subjects come to talk themselves out of their grasped moral responsibilities by making exceptions for themselves, taking moral short-cuts, etc. Moral philosophy itself is needed, Kant claims, because it is the only thing that can provide an antidote to this naturally occurring tendency towards self-corruption. It seems clear to me that someone who did not regularly reflect on just what might be going on within their own minds in cases like these would simply fail to have a reflective character, yet such reflections must surely be explicit acts of introspection if they are to bring a subject to awareness regarding their own corruption. By resisting the focus on faculty-talk, Merritt risks neglecting the evidence that pushes one towards considering the importance of introspection for Kant: it is because we are dimly aware of conflicting internal voices (subsequently identified as the faculties of desire and rationality respectively) that we are required to think more explicitly about how to resolve their conflict.

My questions here concern just the risk of over-egging the less introspective elements in Kant’s thought at the cost of now under-selling the traditional emphasis upon explicit introspection within the overall intellectual worldview. As Merritt knows, when Kant set out the first rule of the healthy understanding he surely did so in awareness that he was repeating something he had previously stated in an answer to the question of the nature of enlightenment. In that latter context he surely did mean that one must reflect as to what seems to be the rational thing to do by one’s own self-conscious lights. It was necessary to do so, he thought, so as not to give in to the temptation to outsource the responsibility for thinking to whatever cultural and emotive elements happened to be infusing one’s capacity to judge. It seems plausible that he wanted us to do this quite frequently. It might be that a demanding account is just what is demanded if resisting the sirens of unthinking is a worthwhile endeavour.