The aims and scope of I, Me, Mine: Back to Kant, and Back Again are ambitious, and few scholars are as well situated as Béatrice Longuenesse to attempt the task of bringing figures such as Kant, Sartre and Freud into dialogue with recent work in analytic philosophy of mind and language. Both historians and non-historians will appreciate the book’s clarity, subtle analyses and provocative suggestions.

Longuenesse’s primary thesis is that a Kantian view of self-consciousness – namely, that our fundamental form of self-consciousness is an awareness of oneself as a thinking subject rather than as an object (whether as a Cartesian soul or as a living body) – is defensible on multiple fronts. She argues that such a view is successful in handling recent concerns within analytic philosophy, such as accounting for the use of the first-person pronoun as immune to error through misidentification, and that it possesses advantages over rival views, like those of Wittgenstein, Evans, Anscombe and Sartre. Moreover, she argues that this Kantian view can be naturalized along Freudian lines, that is, can be offered without positing an unknowable transcendental subject (p. 227). In presenting these views, Longuenesse offers engaging and illuminating discussions of a variety of other topics, including personal identity, self-deception and our awareness of ourselves as moral beings.

Part I situates Kant’s view of self-consciousness within twentieth-century theories of self-consciousness. Chapter 2 considers the analytic tradition’s driving question of what accounts for the special epistemic and linguistic status of our use of the first-person pronoun ‘I’, namely that it is (using Shoemaker’s phrase) ‘immune to error through misidentification relative to the first-person pronoun’ (IEM). In opposition to accounts (such as Evans’) according to which the use of the first-person pronoun possesses IEM in virtue of one’s possession of information about one’s own body, Longuenesse advances the Kantian view that this special status ultimately derives from a self-consciousness of one’s own activity of thinking. This special form of self-awareness ‘is a consciousness of being engaged in an activity of binding one’s representations in such a way as to come up with concepts, combined in judgments, connected in inferential patterns’ (p. 4).

Throughout the book, Longuenesse characterizes this form of self-consciousness in normative terms. Consciousness of one’s own body is to be distinguished from ‘consciousness of the reason-giving unity of one’s mental activity’ (p. 4, my emphasis; cf. pp. 29, 37, 188). The ‘I’ in ‘I think’ expresses a ‘taking responsibility’ for one’s acts of unifying representations. (p. 200, n. 51). And:

Although I too am inclined to interpret transcendental apperception as an awareness of oneself as subject to norms and reason-giving, Longuenesse’s support for such an interpretation is spread thin. For example, her initial presentation of this view in section 2.2 offers sparse textual support: a reference to the entire first half of the B-Deduction (B132–44), a couple of passages concerning the interdependence of one’s capacity to use ‘I’ and synthesis (B135 and A117n.), and Longuenesse’s gloss of A446/B404 as asserting ‘that in the proposition “I think,” “I” represents nothing further than “the I or he or it (the thing) that thinks”’ (p. 27).Footnote ¹ None of this textual support offers clear evidence that Kant conceived either the use of ‘I’ or synthesis in normative terms.

The footnotes in section 2.2 suggest that such an interpretation will be supported in chapter 4. One finds there an interpretation of the famous opening statement of §16 of the B-Deduction: ‘The I think must be able to accompany all my representations’. Longuenesse suggests that we understand the accompaniment of a thought by ‘I think’ to indicate ‘that I am in a position to provide justification for my judgment’ (p. 77). But no further textual support or analysis is given for the contention that transcendental apperception involves an actual or possible awareness of one’s justification. (Likewise, the otherwise nuanced analysis of this passage in section 7.2.1 makes no mention of normative considerations such as justification.) Such a normative reading of transcendental apperception is very promising for making sense of key argumentative moves within the transcendental deduction, but Longuenesse’s extraordinarily brief outline of the transcendental deduction of both editions (pp. 78–81) likewise makes no appeal to any such normative notions.

Chapter 3 focuses primarily on Sartre’s account of self-consciousness. On this account, every act or state of consciousness possesses two aspects: each consciousness is ‘thetic’ or ‘positional’ insofar as it has an intentional object, and each consciousness is at the same time ‘non-thetically’ or ‘non-positionally’ self-conscious insofar as it is aware of itself but not as an object. Longuenesse suggests that non-thetic self-consciousness provides Sartre with an explanation for the use of ‘I’ as subject, insofar as non-thetic self-consciousness grounds what Sartre calls ‘reflection’, i.e. a consciousness whose intentional object is consciousness itself. She develops further her interpretation of Sartre by examining it in contrast to Evans’ reading of Sartre and by considering Anscombe’s influential view that the first-person pronoun is non-referential when used as a logical subject. For Longuenesse, all three philosophers make the same mistake: their focus on how self-consciousness is related to awareness of one’s own body overlooks Kant’s insight that the fundamental form of self-consciousness is a consciousness of oneself as a thinking subject.

For example, Longuenesse charges Sartre with giving ‘insufficient attention’ to non-thetic self-consciousness of mental activity due to his focus on non-thetic consciousness of one’s own body (p. 66). But this construal of Sartre overlooks several passages in Sartre’s works that focus primarily on non-thetic self-consciousness of mental activity. First, his argument for non-thetic self-consciousness in the Preface to Being and Nothingness is that non-thetic self-consciousness is the ontological foundation for knowledge, and hence mental activity (Sartre Reference Sartre1984: 9–17). Second, Sartre characterizes the chapter ‘Immediate Structures of the For-Itself’ as a ‘description of non-thetic self-consciousness’ (1984: 120). The chapter goes on to describe such consciousness as engaged in mental activities, e.g. the pre-reflective cogito and consciousness (of) belief. Third, Sartre’s (Reference Sartre2003) ‘Conscience de soi et connaissance de soi’ discusses – independently of considerations of the awareness of one’s own body – non-thetic self-consciousness as what accounts for the Cartesian cogito. Even if one interprets Sartre as holding that such consciousness is not possible without the body, Sartre does take care to describe non-thetic self-consciousness of mental activity.

Longuenesse’s charge against Sartre – that he failed to pay sufficient attention to self-consciousness of mental activity – is encouraged by the neglect of key aspects of Sartre’s account of being-for-others. She claims that, according to Sartre, ‘I exist for myself as known by others … as a consequence of applying to my own body the knowledge I have of the body of others, and thus as a consequence of the fact that I thereby come to experience my own body differently’ (p. 57). But Sartre in fact holds that the primary way in which I exist for myself as apprehended by others is through ‘the look’, and this amounts to an awareness of others not in the form of knowledge of others’ bodies but rather in others’ being ‘given to me directly as a subject[s]’ (1984: 341). To be sure, Sartre makes explicit that this awareness of oneself as apprehended by others does not constitute an awareness of others as ‘extra-mundane’ (1984: 345). Nevertheless, Sartre rejects identifying knowledge of others’ bodies as the source either of one’s original awareness of others or of one’s awareness of being apprehended by others.

The above misconstrual of Sartre is a symptom of a much more general issue with how Longuenesse engages with past philosophers, namely, that insufficient attention is given to their motivations, goals and methods. While she rightfully points out that Sartre’s ontological claims are more palatable when understood ‘in epistemological and phenomenological terms’ (p. 56), she overlooks that Sartre himself makes such a point about his approach in discussing awareness of one’s body: ‘it is most important to choose the order of our bits of knowledge’ (1984: 402).Footnote ² This claim might seem inconsistent with Sartre’s rejection of the ‘primacy of knowledge’, i.e. the treatment of epistemology as first philosophy. But in context, Sartre arguably means that we must properly order our philosophical knowledge, i.e. our descriptions of phenomena insofar as those descriptions carry with them ontological commitments. More generally, Sartre systematically presents his views in Being and Nothingness by following such a method: the description of non-thetic self-consciousness of mental activity (part two, chapter one) has priority over the description of one’s awareness of others (part three, chapter two), which in turn has priority over the description of one’s awareness of one’s own body (part three, chapter two), which in turn has priority over the description of one’s awareness of others’ bodies (part three, chapter three).

Returning to Kant, Part II aims to explain how Kant conceives of consciousness of oneself as a thinking subject, while also maintaining that such self-consciousness does not provide one with any knowledge of one’s metaphysical nature. Using Descartes as a foil, Chapter 4 presents how the Transcendental Deduction offers a positive account of self-consciousness. Kant’s account has two advantages over Descartes’. First, whereas Descartes’ cogito merely makes a connection between self-consciousness and thinking, Kant’s account explains that connection. Second, Descartes illicitly makes claims about the nature of the thinking subject, whereas Kant holds that all that can be affirmed on the basis of self-consciousness alone is that one exists and not the manner in which one exists. Longuenesse’s contrast of Kant with Descartes is illuminating, but recent scholarship has established that Kant’s conception of self-consciousness can be understood adequately only in context of the philosophical reception of Descartes’ cogito by his German predecessors (cf. Wunderlich Reference Wunderlich2005; Thiel Reference Thiel2011: 349–54, 372–80; and Dyck Reference Dyck2014: 173–98).

In chapters 5 and 6, Longuenesse offers a nuanced interpretation of Kant’s diagnosis of the illusions underlying the first three paralogisms. In all three, the rational psychologist draws an invalid inference from true premises containing an equivocal term. For example, the first paralogism fails to differentiate two notions of ‘subject’: a metaphysical subject in which qualities inhere, and a logical subject of which concepts are predicated. Longuenesse treats these two senses of ‘subject’ as exhausting Kant’s uses of the term (pp. 108–9). But Kant seems to have an alternative use of ‘subject’, namely one referring to the ‘thinking subject’ (cf. Bxxiii, B278) that carries out acts of thinking. Longuenesse herself seems to recognize this use of ‘subject’ in Kant (p. 112), and she appeals to such a sense in characterizing Kant’s ‘consciousness of oneself as subject’ (my emphasis) as ‘a consciousness of being engaged in a specific kind of mental activity’ (p. 19; cf. section 2.2). By overlooking this sense of ‘subject’ in her discussion of the paralogisms, Longuenesse is led to an awkward interpretation of the first paralogism’s minor premise: ‘I, as a thinking being, am the absolute subject of all my possible judgments, and this representation of myself cannot be used as a predicate of any other things’ (A348). She writes: ‘Kant clearly means: “In the judgment ‘I think’, ‘I’ cannot be present otherwise than as subject, and cannot be predicated of something else.”’ (p. 115). Yet this reading does not cohere with how Kant himself describes the rational psychologists’ restatement of the minor premise: ‘Now in all our thinking the I is the subject, in which thoughts inhere only as determinations, and this I cannot be used as the determination of another thing’ (A349, my emphasis). The phrase ‘all our thinking’ makes evident that the minor premise’s scope is not limited – as it is on Longuenesse’s interpretation – to thinking the judgement ‘I think’. With the notion of ‘thinking subject’, a more straightforward reading of the minor premise is available: ‘I, as a thinking being, am what carries out acts of judgement, and this representation of myself is of what makes predications in judgements rather than what is predicated in judgements’. On this alternative reading, Kant’s description of the ‘logical subject of thinking’ can be understood, following Kant’s suggestion at A350, to refer to the transcendental subject.Footnote ³

The aim of chapters 7 and 8 is to show that Kant has the resources to naturalize his account of the transcendental unity of apperception, insofar as it is awareness of oneself as subject to both theoretical and practical norms. Longuenesse’s argument is that Freud offers such a naturalized account because his views are similar in relevant ways to Kant’s. This comparison is sometimes very illuminating. For example, she offers a careful and persuasive account of Kant’s views concerning our ‘motivated blindness’ regarding our motivations (section 8.2.3) in comparison with Freud’s notion of repression (8.4.3). Less persuasive, however, is her case that Kant’s transcendental unity of apperception ‘has its counterpart in the organization of mental processes Freud calls “ego,” governed by the reality principle’ (p. 189). For Kant’s transcendental unity of apperception unifies our experience by means of the categories, whereas Freud’s reality principle does so only ‘according to elementary logical rules’ (p. 189).

Despite the critical comments raised above, I, Me, Mine demonstrates the continuing importance of Kant’s thought for philosophical debates concerning self-consciousness. Anyone interested in such issues would benefit from engaging with Longuenesse’s provocative comparisons of Kant with figures in both the analytic and continental traditions.