One might think that Matthew Boyle’s Transparency and Reflection is a bit of a disappointment to those Kant scholars who, like me, have long been fascinated (or frustrated, or both) by Kant’s theory of transcendental self-consciousness. For, as Boyle tells us, he thought about writing a book about Kant’s cryptic statement that “it must be possible for the I think to accompany all my representations” (B131) but decided against it.Footnote 1 While he thought that this Kantian thesis encapsulates “a crucial idea about the importance of self-consciousness to the rational mind” (Boyle, Reference Boyle2024, ix), the account appeared to suffer from insuperable difficulties and had to be abandoned. When he then discovered “a more Sartrean way” of thinking about self-consciousness, he found in it the key to avoiding the difficulties that plagued his earlier, more Kantian orientation. And yet, I would like to suggest that Boyle’s book is a thoroughly stimulating and thought-provoking contribution not only to contemporary scholarship on self-knowledge but to Kant scholarship in particular. That this is so is a testament both to the deep historicity of Boyle’s thought and his ability, demonstrated so vividly in this work, to draw out the importance of apparently idiosyncratic historical developments to contemporary and, indeed, ultimately to universal philosophical concerns. My aim in this commentary is two-fold. First, I would like to make explicit the ways in which I think Boyle’s Sartrean account of the structure of consciousness can help illuminate Kant’s account of transcendental self-consciousness and its role in cognition. Second, I will suggest that Kant’s account of transcendental self-consciousness can help us address an apparent gap in Boyle’s response to what he calls “the Anti-Egoist challenge.”
Boyle insists, against contemporary orthodoxy, that the philosophical interest of the topic of self-knowledge lies in the light that it can shed on the nature of our minds. Confronting us in the familiar topic of the “transparency” of self-ascription is not merely a puzzle about what is distinctive of this knowledge qua knowledge. Instead, Boyle proposes that an analysis of self-knowledge (and its possibility) reveals something important about the nature of our minds that justifies the connection that philosophers have historically seen between self-consciousness and rationality. The proper analysis of self-knowledge shows us to be rational animals:animals with the ability “to think and choose in a distinctively self-determined manner” (Boyle, Reference Boyle2024, 10), for whom the question of “how we ought to live” is inescapable.
Now, Kant is a paradigmatic example of the kind of philosopher who takes seriously the connection between rationality and self-consciousness. He says, for example, that “The I is the foundation of the capacity for understanding and reason, and the entire power of cognition” (25:10) and is “the reason why I call myself an intelligence.” (B158n). But as Boyle points out, it is very difficult to see what this connection is really meant to be, and most contemporary scholarship in a broadly “Kantian” vein appears to assume rather than explain it.
Yet Boyle remains sympathetic to the Kantian project, and suggests that we can vindicate its core insight: that there is a fundamental connection between rationality, self-determination, and self-awareness. We can illuminate this relationship, Boyle contends, only if we draw a distinction between two forms of self-awareness: reflective and pre-reflective self-awareness; and he insists that we “must give central place to pre-reflective self-awareness in explaining how we know our minds and how this knowledge is linked to rationality” (Boyle, Reference Boyle2024, 19). Boyle does not, however, locate the resources for such a distinction in Kant’s work, and instead turns to the work of Jean-Paul Sartre. My main aim here is to defend the idea, to which I suspect Boyle would be sympathetic, that there are in fact resources in Kant’s philosophy, which the Kantians themselves may have overlooked, for thinking that a kind of pre-reflective self-awareness is exactly what is at issue in explaining how it is that the “I think” is able to accompany all my representations.
I proceed as follows: First, I would like to introduce the concept of nonpositional consciousness that Boyle draws from Sartre’s work, and I will suggest that this kind of consciousness has two features that do important work for Boyle in explaining transparent self-knowledge. Departing from Boyle’s own terminology, I will call these features its “performative” character and its “organizational function”. Second, I want to suggest that drawing on these features of Boyle’s analysis will enable us to understand Kant’s theory of apperception in new and interesting ways. Transcendental self-consciousness, we will see, can helpfully be thought of as a kind of nonpositional consciousness, and viewing it in this light will help us make progress in understanding Kant’s claim that such consciousness constitutes the form of thinking.
1. Two Features of Nonpositional Consciousness
The capacity for explicit self-knowledge, on Boyle’s view, relies on a kind of “pre-reflective” self-awareness that is, in a sense, hidden in plain sight. This kind of consciousness, Boyle argues, is a pervasive yet implicit feature of our mental lives, and it enables us to explain certain puzzling facts about self-knowledge—key among those for Boyle’s purposes, the problem of transparency: the fact that we can self-ascribe states not by turning inward and considering how things stand with us psychologically, but by looking outward and considering questions about how things are or appear to be. My aim in this section is to consider the nature of this kind of consciousness and the role it plays, not only in making self-knowledge possible, but in the structure of consciousness more broadly.
Boyle finds the resources for developing an account of pre-reflective consciousness in Sartre’s works, where it contrasts with a more familiar kind of consciousness that he calls “positional”. To the extent that our consciousness is of or about something, that is, insofar as it constitutes an intentional relation to an object, our consciousness is positional: it “posits” something outside of itself to which it is related. Since nonpositional consciousness, by contrast, does not involve the positing of some object that it is “of” or “about”, it cannot be characterized as an intentional relation to an object (Boyle, Reference Boyle2024, 67). Boyle thinks that appreciating this distinction not only equips us to solve the puzzle of transparency (Chapter 4) but also helps us understand the semantic and justificatory basis of explicitly first-personal representation (Chapter 5). An account of nonpositional consciousness can play this dual theoretical role, on Boyle’s account, because such consciousness has two important features, which I will refer to as its “performative character” and its “organizational function”. My aim in what follows is to unpack these two features and explain the philosophical significance Boyle sees in them.
Let us begin with the performative character of nonpositional consciousness. If, as Sartre maintains, nonpositional consciousness is not to be understood in intentional terms, what sense can we make of the idea that it is a kind of awareness at all? To answer this question, Boyle distinguishes the content of a state of consciousness from the mode in which that content is presented, and situates nonpositional consciousness at the level of mode rather than content. Nonpositional consciousness expresses awareness of a given mode of presentation of a content, where this awareness is itself partially constitutive of the very mode in question. We might say that it is the grasp of the act of representing that we have in the act of representing. It is, accordingly, an awareness that we have in performing the act in question, and thus, I suggest that it is a kind of performative awareness that is a constitutive part of the representing. A particular representational content, say, of a cat, can be represented in a number of different ways: A cat can be perceived, a cat can be imagined, a cat can be desired, for example, and on Boyle’s view, these distinct modes are grasped by the subject in the very act of representing. This “grasping”, on Boyle’s model, is not “an explicit representation of oneself as in [the relevant state]”; instead, it shows up in “a certain intelligent responsiveness to what is in fact one’s being in” the relevant state (Boyle, Reference Boyle2024, 98). For example, a subject who perceives the cat will be disposed to think of the cat using the demonstrative “this”; she will be disposed to gain information about the cat in certain distinctive perception-dependent manners; and so on (ibid.). There is a sense, then, in which the subject is aware of her seeing just insofar as she sees, but this awareness is not “of” the act of perceiving – she does not stand apart from her state and observe herself in it. It is, instead, an awareness she has “in” the act of perceiving and is, in that sense, performative.
It is because this performative awareness of our modes of representation is built into those very modes that Boyle can claim to have unpacked Sartre’s mysterious claim that “any positional consciousness of an object is at the same time a nonpositional consciousness of itself” (Sartre quoted in Boyle, Reference Boyle2024, 66). And for just this reason, Boyle can use his Sartrean analysis to solve the puzzle of transparency: the essentially self-conscious character of our mental states means that for any subject who possesses the relevant concepts, just being in the relevant state gives them all the justification they need for self-ascribing it. As Boyle points out, however, we have not yet uncovered the basis within this nonpositional consciousness for ascribing those states to a unified subject that I designate by the use of the first-person pronoun. My nonpositional consciousness apparently makes me aware that there is, for example, perceiving, imagining, thinking, or believing going on, but what justifies my ascribing those states as modifications to a single subject? We have not yet uncovered the basis within nonpositional consciousness for ascribing those states to a unified subject that I designate by the use of the first-person pronoun—and what Boyle calls the “Anti-Egoist challenge” presses that no such basis can be found (see esp. Ch 5.2).Footnote 2
Boyle’s response to the Anti-Egoist challenge centers on the case of egocentric spatial representation, which he thinks holds clues for a more general response that can cover several representational domains. His point of departure is John Campbell’s distinction (Reference Campbell2002) between monadic and relational egocentric thought. Whereas monadic egocentric thinking simply ascribes a spatial property to an object (being here, being to the left, etc.), relational egocentric thinking explicitly represents the object’s spatial relation to the subject (being where I am, being to the left of me, etc.). Boyle (Reference Boyle2024) maintains that a genuine cognitive advance is required to transition from the former kind of representation to the latter (110–11). To represent an object’s egocentric spatial location in explicitly relational egocentric terms, a subject must grasp two generalities: First, that any two spatially located individuals can occupy that same relation, and second, that she is one of the relata, which in turn requires an appreciation that she could also feature as such in any number of other possible (spatial) relations. In grasping this second generality, the subject is able to recognize herself as “a single topic on which all monadic egocentric representations bear, albeit in diverse ways” (Boyle, Reference Boyle2024, 120).
But what could possibly enable her to grasp the latter generality? Boyle answers this question by appealing to the fact that the mode of presentation itself contains, albeit implicitly, the ground(s) of our reflective awareness of ourselves as constituting such a topic. It does so because it presupposes an organizational structure that would be impossible without a constant reference point. Egocentric representation is a systematically organized whole in virtue of which the subject is able to orient herself and engage practically with her surroundings. But such an organized whole requires a constant reference, a central point of connection and orientation. It requires, moreover, that the subject herself occupies that central location, such that it constitutes what we might call a “point of view”. Such a “point of view” functions as an organizational principle that systematically relates all of these spatial representations to one another, integrating them into a unified field of possible action and perception. The first-person concept, Boyle maintains, is the device by which we represent the systematic integration of our representational states in relation to a single organizational principle: “its fundamental role is not to designate something of which we are aware, but to posit a certain unity among our states of awareness themselves” (Boyle, Reference Boyle2024, 124).
Pace the Anti-Egoist, then, at least in the domain of egocentric spatial representation, our consciousness does afford a semantic and epistemic basis for introducing the first-person concept. And Boyle believes that other representational domains exhibit (and presuppose) a similar systematic interconnectedness, centered on a single point of view, which likewise justifies the introduction of the first-personal concept (Boyle, Reference Boyle2024, 128). Boyle illustrates this general strategy in relation to a particular mode of representation—belief. For believing, Boyle argues, presupposes what he refers to as “a single epistemic view on the world”: an
epistemic standpoint [that] has a certain interconnectedness… propositions are held true as parts of a single point of view are held true together, even before the holder of this point of view thinks any explicitly conjunctive thought about them: they are held true as parts of a single outlook on what is the case, and stand in rational relations to one another in virtue of this. (Reference Boyle2024, 129)
Crucially, on Boyle’s view, it is not merely an external fact that these relations obtain; it is also the case that we are nonpositionally aware of this unity, that is, that we ourselves occupy this standpoint, for this awareness is a condition on the possibility of belief as a mode of presentation.
I hope that the foregoing has done something to bring out the depth and ambition of Boyle’s account. If Boyle’s interpretation of Sartre is along the right lines, he has unearthed in Sartre’s account of nonpositional consciousness the materials for a solution to the problem of transparency and a robust response to the Anti-Egoist challenge. But as I said in opening, I continue to believe that Kant would have been just as good a philosophical companion as Sartre, perhaps even a better one. And to set the stage for what I think could be a fruitful conversation, I want to close this section by raising a philosophical question for the account that has just emerged.
While he offers a brief sketch of how a full account might run, Boyle (Reference Boyle2024, 130) does not attempt to extend his analysis to cover the full range of representational domains in which we characteristically employ the first-person concept. We are left to wonder whether the notion of a point of view can be generalized to justify the presence of the first-person concept in our self-ascription of desires, imaginative states, fears, and so on. But even if we can show that each representational domain exhibits the features that justify the introduction of the first-person concept, a more fundamental question concerns the identity of the subject across different representational domains. Even if we discern a single point of view within each mode of presentation, the worry remains that Boyle’s account leaves us with as many subjects as we have modes of presentation. To make sense of the idea that it is the same subject who believes, perceives, desires, and so on, we would need to make sense of the idea that each of the points of view internal to these various modes are themselves integrated into an overarching point of view. This would seem to require yet another kind of nonpositional consciousness, one that is deeper still than the awareness of a unified epistemic view, which could incorporate the systematic relations not only within but also between the various modes of presentation.
2. Transcendental Apperception as Nonpositional Consciousness
My aim in this final part of my commentary is to put Kant and Boyle (back) into what I think could be a fruitful conversation with one another. On the one hand, I want to suggest that Boyle’s account of nonpositional consciousness, and especially his emphasis on what I have called its performative nature and organizational function, can help us make progress in understanding Kant’s theory of transcendental self-consciousness. On the other hand, I want to suggest that this theory, so illuminated, can help us avoid the problem of disunified domains of systematically integrated representation, thus vouchsafing the identity of the subject across different domains or representational “modes”.
On the model of transcendental self-consciousness I will propose, transcendental self-consciousness exhibits all the central features of nonpositional awareness. This consciousness is not intentionally (positionally) directed at anything internal or external to the subject but is a precondition of all such intentional (or positional) consciousness.Footnote 3 It is, I would like to suggest, a kind of performative consciousness in the following sense: it constitutes a subject’s grasp of a certain mode of representation, which grasp is expressed in the activity of representing itself. The mode of presentation that we grasp in transcendental self-consciousness is what Kant simply calls thought—the distinctively intelligible way of representing objects that distinguishes it, as a mode of representation, from the sensible way in which objects are given to the mind in intuition. And, as we will see, a certain unity is presupposed by this particular mode of presentation, and this unity plays an organizational role, one that is analogous to the role of a “point of view” that Boyle identifies within particular modes of presentation like belief. This unity provides the justification required for us to ascribe the representations in question (thoughts) to a single unified subject (I). It is on the basis of this unity, presupposed in all thinking, that it must be possible for me to accompany all the representations that are “something for me” with the “I think”.
Whereas typical readings regard transcendental self-consciousness as a species of object-directed representation or positional consciousness (where the object “posited” might be myself, my mental states, my acts of synthesis, the laws of the mind, etc.), my reading makes transcendental apperception a condition on the possibility of all conscious object-directed representation (or thinking). And because the unity that this consciousness presupposes is a precondition of all intentional consciousness, not simply one specific mode of presentation, the account of transcendental self-consciousness that I want to develop can avoid the problem of disunified domains of systematically integrated representation. It would therefore justify the introduction of the subject not just as the center of some specific mode of presentation, but as at the center of all conscious object-directed representation as such.
The key to appreciating all of these points is to understand the sense in which transcendental apperception constitutes the form of thinking. For I will claim that on Kant’s view self-consciousness is the form of all consciousness of objects, a form that the subject is not in turn conscious of, but that is simply presupposed in all of her conscious representations of objects. Transcendental apperception is the form of intellectual consciousness, inherent in the faculty of the mind that Kant calls “the understanding”. This form is automatically expressed in thought, without the requirement that the subject who is combining the representational elements of the thought be positionally conscious of her own activity. The only consciousness that she has of her activity is in performing it, and performing it is simply to represent the given content intelligibly, that is, in a logically organized or structured way. My first claim, then, is that apperception is a subject’s grasp of a certain mode of presentation, namely, thought, and that this grasp is expressed in the activity of thinking itself. As we will see, thoughts have a distinctive logical organization, and apperception is the representational activity that imparts that structure to the content given in sensibility.
We can make progress toward seeing why I would attribute such a view to Kant by considering the question of how the manifold that is given in intuition comes to be thought. Kant’s answer to this question is relatively straightforward: while the manifold “can only be given in the intuition … [it is] thought through combination [Verbindung] in a consciousness” (B135). Thus, if we can understand what “combination” is and how the manifold is combined in a consciousness, we will have a decent understanding of how thinking is possible.
Fortunately, Kant opens the B-Deduction with an account of combination. There, he claims that “[a]mong all representations combination is the only one that is not given through objects but can be executed only by the subject itself, since it is an act of its self-activity [Selbsttätigkeit]” (B129). The activity in question only contributes one distinctive feature of the resulting representation, namely, its unity. Prior to combination, a representation already contains a manifold, and that manifold is already subject to synthesis, but the synthetic manifold is not yet represented as a “synthetic unity” (B130–31). We therefore see that whatever else thought is, it is a kind of unity that is the result of an act of the subject that Kant goes on to identify as pure or original apperception (B132).
If thought is, as I have claimed, a certain combination of (given) representations, then we can start to see the basis for the claim that the unity of such a representation consists in the logical relations between the elements of the representation. Once a thought has a given logical form, the order and relations of its parts become necessary for the identity of the thought. Consequently, each thought is a unity, a whole that is “held” together by the logical relations between its constituents. When I say that transcendental apperception is the organizing principle or the formal condition of thought, what I mean is that it is a grasp of the logical structure that is a necessary condition of this mode of presentation (thinking) and that is expressed in it.
To see that this is Kant’s view, we can start by noting that he says “[transcendental apperception] is only the formal condition, namely the logical unity of every thought” (A398, my italics). This transcendental apperception is designated by the expression “I”, which Kant claims “contains a priori a certain form of thinking” (A685/B713). Apperception thus plays its role in making thought possible by supplying the distinctive logical form that characterizes this mode of presentation, rather than representing any distinctive kind of object. In the Paralogisms, Kant makes this point explicitly, saying that “[transcendental apperception] in itself is not even a representation distinguishing a particular object, but rather a form of representation in general, … for of it alone can I say that through it I think anything” (A346/B404, cf, also B411). These passages make it clear that Kant conceives of transcendental apperception as the form of thinking and that, as a condition of thinking, it itself does not represent anything but only makes a certain kind of representation possible. For a representation to be “accompanied by” the I think, then, is not for that representation to be represented in any given way (e.g., as mine), but rather for it to be combined in such a way that it inheres, as a thought, in a subject.
What we have seen so far is that apperception is expressed in thinking itself, and is the basis of the fact that all thought has a synthetic unity that consists in the logical connections between the elements of the thought. What I would like to highlight now is that apperception plays a crucial organizational role; indeed, it functions as an organizing principle of thinking. For we will see that the synthetic unity characteristic of this mode of presentation presupposes a prior unity that pertains to apperception as such, that is, to the act itself, and this unity Kant calls “qualitative”, or “formal” unity. Kant describes this unity as a thematic unity, which can best be understood by analogy with the unity that determines the organization of a work of art (like a play or a story), and which functions to establish connections within the sensible given manifold that render it intelligible.
In explaining the synthetic unity of thinking, Kant tells us that this unity is the expression of a more fundamental unity, one that is prior to thought and that must be “added” to the manifold in combination. Now, Kant tells us that the unity that is added to the manifold cannot be understood in terms of the category of unity, since the category already requires combination. He claims, therefore, that
we must seek this unity (as qualitative, §12) someplace higher, namely in that which itself contains the ground of the unity of different concepts in judgments, and of the possibility of the understanding, even in its logical use. (B131)
It is this qualitative unity, then, that makes possible the synthetic unity of thought, and that Kant refers to as the unity of apperception.
If we now look at the passage in §12 of the Analytic, to which Kant refers back at B131, we see Kant explaining that, in its formal use, the concept of unity, which he calls “qualitative unity,” cannot be ascribed to things, but must be understood as pertaining to the cognition of things, and has a formal rather than a material use. He describes the unity as follows:
In every cognition of an object there is, namely, unity of the concept, which one can call qualitative unity insofar as by that only the unity of the comprehension [Zusammenfassen] of the manifold of cognition is thought, as, say the unity of the theme in a play, a speech, or a fable. (B114)
From this passage, it follows that the unity in question is expressed by any concept through which an object is cognized, so this unity is expressed in thinking. And the unity must be understood as pertaining to the form of the cognition insofar as the manifold in it is taken together or comprehended. As examples of this kind of “taking together”, Kant points to “the theme of a play, a speech, or a fable”.
These examples suggest two things: First, that the taking together is made possible through a kind of intelligible connection of the elements of the manifold; second, that the intelligible connection is established in accordance with a principle (an organizational form). I take the first suggestion to be the result of the fact that the theme of a play or a story constitutes a kind of intelligibility since it is in virtue of the theme that the “manifold” of the action of the play, or the words and sentences of the speech or story, makes sense. Thus, qualitative unity (the taking together of the manifold) should be understood as the basis for establishing intelligible connections between the elements of the manifold such that the manifold constitutes the content of a thought (a concept). It is important to notice that the theme of a play or a story is the principle of connection between the elements of the “manifold” that it contains. It is the theme that explains why the sentences or words or ideas in a story are presented and connected in one way rather than another. In this way, the theme is a principle of organization, since without it there is no connection between the elements, and thus, no play. In the same way, qualitative unity is a principle of connection between the elements of the manifold of sensibility; without such a principle, there is no thought or concept.
As to the second point, then, reflecting on the examples Kant provides suggests that he is committed to the idea that this intelligible connection must be understood holistically; that is, the connections must be understood as presupposing an idea of a whole. The idea that I am referring to here is an idea of the form of the whole of cognition. It is this idea that informs the structure of all thinking, and which is therefore the form of thinking itself. We might think of this idea as “the rational” as such. We can see how this is meant to work by looking again at how Kant conceives of the kind of thematic unity displayed in works of art.
Works of art are produced in accordance with a concept, viz., the theme that is the central organizing principle of the materials, which concept is also the principle of the unity of the manifold, the unity of the work itself. Kant says that such works are “comprehended [zusammengefasst] under a concept or an idea that must determine a priori everything that is contained in it” (5:373). But note that he says the same thing about the cognitions of the understanding, too (A67/B92). In a play or a story, the theme is a principle that functions to connect various elements (sentences, words, scenes, etc.) in a meaningful way, or in a way that makes sense, and also in such a way that the resulting product is unified and connected in accordance with an aim (whatever is to be expressed or communicated by the work).Footnote 4 Correspondingly, the unity of apperception is a principle that functions to connect the elements of a representation in such a way that the resulting representation is intelligible; that it can be thought. Transcendental apperception connects the elements of our intuition in such a way as to express its unity, by giving the manifold of intuition the form of thinking.
Note, too, that the theme, as a principle of the connection of the manifold, cannot really be abstracted from the play, story, song, and so forth. In one sense, of course, it precedes the work, in the sense that it is the “purpose” and the reason for it, and in that sense the work is created so as to fulfil that purpose. But in another sense, the theme only really exists within the work as a whole; it exists through the connections that it makes possible. In that sense, too, we might say that the theme is formal: it shapes the content of a work for a purpose or end and does not incorporate any content “of its own”, independently of the content of the art work. On my formalist reading, transcendental apperception, as the thematic unity of thought, is formal in the same way that the theme of a work of art is formal. It does not in itself represent anything, but it contributes something essential to the possibility of a certain kind of representation, namely the necessary connections between the representational elements in a thought that make it a synthetic unity.
3. Conclusion
If this account is along the right lines, then transcendental apperception, like nonpositional consciousness, is “the kind of awareness of being a thinker one has in thinking, as opposed to a merely abstract awareness that I am a being who thinks” (Boyle, 123). Moreover, a basis for ascribing our various representations to a unified subject can be found in an account of the presuppositions of thinking, understood as the most fundamental mode of conscious intentionality. Boyle’s account of self-knowledge thus gives us a model for a fruitful understanding of Kant’s theory of transcendental self-consciousness, which in turn allows us to solve the problem of the unity of the subject. It is one of the many virtues of Boyle’s richly rewarding work that it can inspire in us a fresh feeling for the depth and vitality of Kant’s philosophy, and a renewed reason for us to sustain our interest in it.
Acknowledgments
I am grateful to Drew Christie, Bill de Vries, Casey Doyle, Max Edwards, Mark Okrent, and Paul Schofield for reading and discussing Transparency and Reflection with me in a reading group during Fall 2024. I also wish to thank the audience at our Author-Meets-Critic session in Toronto (January 2025) for their constructive feedback and illuminating discussion.
Claudi Brink is an Assistant Professor of Philosophy at the University of New Hampshire. Her research focuses on Kant’s theoretical philosophy, with a particular emphasis on his account of cognitive spontaneity and transcendental self-consciousness. She is currently writing a Cambridge Element on Kant’s account of cognitive spontaneity.
Open access
