Precis of I Me mine

1 Book Symposium. Philosophy and Phenomenological Research First published: 30 May 2019 DOI: https://doi.org/10.1111/phpr.12599 Précis of I, Me, Mine Béatrice Longuenesse, New York University* Central to I, Me, Mine is the thesis that self-consciousness has two fundamentally different but intimately related aspects. One aspect is the consciousness of one’s own body and especially one’s own body in action. The other aspect is the consciousness of one’s mental unity or more precisely, the consciousness of being, oneself, engaged in a mental activity by which one brings rational unity into the contents of one’s mental states. A second, related thesis is that the latter (consciousness of mental unity) is more fundamental to self-consciousness than the former (consciousness of one’s own body) even though the two are in fact, for us human beings, intimately connected. By ‘self-consciousness,’ I mean the capacity to think of a state as one’s own state, that is to say, the capacity to ascribe that state to oneself in thought and language, using the first-person pronoun ‘I’ or (in language) any equivalent grammatical form such as the first-person inflection of the verb. In thus relating self-consciousness to the use of ‘I,’ I do not mean that selfconsciousness just is the (propositional) self-ascription of states or properties. What I mean is that self-consciousness is the necessary condition for (propositional) self-ascription. Selfconsciousness is more than mere self-feeling or subjective state—a state (e.g., pain or pleasure or the sensation of redness) such that being in that state just is being aware of being in that state. It also more than mere de se state: a state in the content of which the subject of the state is * This is the original version of the Précis published in Philosophy and Phenomenological Research which I agreed to shorten by request of the editor. 2 involved, such as: seeing a flying object moving toward me. To be self-conscious is to have the capacity to integrate one’s conscious states—subjective, de se, or directed at objects—into propositional contents in which ‘I’ occupies the subject place (in Aristotelian form: <S is P>) or the argument place (in post-Fregean predicative form: <Fa>, where a=I). This explains the second thesis outlined above: the consciousness of being, oneself, engaged in a mental activity by which one brings rational unity into the contents of one’s mental states is more fundamental to self-consciousness than consciousness of one’s own body. The former is what Immanuel Kant called “transcendental unity of consciousness” or “transcendental unity of self-consciousness” or “transcendental unity of apperception.” That unity is not a given. Rather, it is a unifying act, whose role is not only or even primarily to integrate subjective or de se states into propositional form such that one can ascribe those states to oneself (‘I’m in pain’, ‘I have a sensation of red,’ and so on). Rather, the role of that act is primarily to unify the contents of object-directed perceptual states into propositional form. Such unity of consciousness, Kant claims, is what eventually gets expressed in the thought ‘I think,’ whatever the content of that thought might be (e.g.: “I think this proof is valid” or “I think this is a tree”—two examples I analyze in the course of the book). Several questions are looming here. For now, I will mention only two, others will come up below in the course of the discussion with my critics. Here’s the first: does that unifying act necessarily involve the agent’s being conscious of being engaged in it? In other words: is that act properly speaking an instance of selfconsciousness? Is it not the case, rather, that the rational unity of the contents of one’s consciously accessible mental states is the outcome of an activity that occurs largely at the subpersonal level and of which one is, therefore, not conscious? This point is discussed in 3 connection with Kant’s view in chapters 2, 4, 5 and 6, and in connection with Jean-Paul Sartre’s view, comparing the latter with the views of Gareth Evans on the one hand and Elizabeth Anscombe on the other, in chapter 3. The second question takes me back to the first aspect of self-consciousness I mentioned at the beginning. In cases in which one is conscious of being engaged in bringing rational unity into the contents of one’s mental states (for instance when one consciously and painstakingly progresses through the steps of a proof or one counts the pears in a basket), to what extent does that self-consciousness (consciousness of being engaged in thinking or counting) involve consciousness of one’s own body, or of oneself as an embodied entity? And a related (third!) question: is ‘I,’ in ‘I think,’ a referring term? If so, is its referring role premised on the conception of oneself as an embodied entity? To the former (the second) question, my own answer is: yes, ‘I’ in ‘I think’ is a referring term, as it is in all other uses. In chapter 2, I voice my agreement with Gareth Evans and Sydney Shoemaker on this point, pace Elizabeth Anscombe’s view, discussed in chapter 3. In my view, in all instances of its use ‘I’ refers to the ‘I’-user, namely to the entity that thinks the proposition in which ‘I’ is in use. Pace Gareth Evans, however, I argue that no further conception by the ‘I’-user of what her fundamental nature might be is needed for her to meaningfully refer to herself as the ‘I’-user. Nevertheless, in many instances, propositions in which ‘I’ is in use have in the predicate place concepts of bodily properties. In those cases, even though ‘I,’ per se, is still meaningfully used in virtue of the sole “thinker-rule,” the proposition does express a self-consciousness that includes a conception of oneself as embodied. The same is true in cases in which one ascribes to oneself object–directed perceptual states whose contents could not have been verified otherwise than by taking into account the connection of the relevant object to our own body and its position in the 4 world. This means that an integrated conception of the referent of ‘I’ is a conception of it as an embodied entity. This is what I mean when I say that the two fundamental aspects of selfconsciousness listed at the beginning, while being distinct, are nevertheless intimately connected. The theses just outlined are developed from different perspectives throughout the book, appealing to a variety of philosophical traditions and methods. Part One discusses analyses of self-consciousness and the first person in recent and contemporary philosophy of language and mind, introducing along the way comparisons with the views of authors such as Immanuel Kant, Sigmund Freud or Jean-Paul Sartre. Part Two is entirely devoted to Kant’s analysis of ‘I think,’ its role in cognition and the metaphysical illusions generated, according to Kant, by misunderstandings concerning that role. Part Three comes back again to present times, making a detour through Freud’s meta-psychological analysis of the structures of mental life, which, I argue, offers resources for a naturalization of Kant’s view of the mental structures governing epistemic and moral normativity and therefore our uses of ‘I’ in the epistemic ‘I think’ and in the moral ‘I ought to.’ Part One has three chapters. Chapter One introduces the book’s overall line of argument. Chapter Two discusses Wittgenstein’s Blue Book discussion of the distinction between uses of ‘I’ as subject and uses of ‘I’ as object and recent analyses of that distinction. I try to highlight the difference between that distinction and Kant’s distinction between consciousness of oneself as subject and consciousness of oneself as an object. I focus especially on the similarities and 5 contrasts between Gareth Evans’s characterization of self-consciousness as self-reference1 and Kant’s characterization of self-consciousness as the unity of consciousness that is the necessary condition of object-representation. This unity of consciousness, according to Kant, is expressed in the thought ‘T think.’ Now according to Evans’s reading of Kant’s view, absent a conception of oneself (or what he calls an “Idea of oneself”)2 as an embodied entity, ‘I think’ is merely “formal” or expresses the formal unity of our representations. In other words, absent a conception of oneself as an embodied entity, ‘I’ in ‘I think’ has no referential role at all. This is, of course, Evans’s own position, and he takes himself to agree with Kant on this point. In contrast, I argue that pace Evans, having a conception of oneself as an embodied entity is not, for Kant, a necessary condition for ‘I’ to have a referential role in the phrase ‘I think.’ In ‘I think,’ says Kant, ‘I’ refers to (or, in Kant’s vocabulary, “represents”) an “I, or he, or it, (the thing) that thinks” of which all we know is that it (I) thinks (think). Nevertheless, not only the conception of myself as an embodied entity, but the actual consciousness of myself as an embodied entity is often a necessary condition for asserting with confidence the validity of a particular content of my thought—e.g., when I assert ‘I think this is a tree’ with a confidence that partly rests on the fact that I take into account, when verifying my judgment, my own position with respect to the object I see, the quality of the light, the angle at which I see the object and so on. Nevertheless, even though as a matter of fact Kant’s self-consciousness as unity of consciousness is intimately 1 Cf. Evans (1982), 213: “The essence of self-consciousness is self-reference, that is to say, thinking, by a subject of judgments, about himself and hence, necessarily, about a subject of judgments.” 2 Cf. Evans (1982), 210; Longuenesse (2017), 24, 39fn.12. 6 connected with consciousness of one’s own body, the latter cannot be described as a necessary condition of the former. In fact, the contrary is true, I argue: I argue that the kind of self-consciousness expressed in the thought ‘I think’ (but not the thought ‘I think’ itself), namely Kant’s “unity of consciousness” or more completely “transcendental unity of consciousness” is a necessary condition for all uses of ‘I,’ whether (in the vocabulary of post-Wittgensteinian discussions) uses of ‘I’ “as subject” or “as object,” and whether those uses depend on the consciousness of one’s own body or just on the awareness of subjective states. I illustrate the connection and possible disjunction between the two types of self-consciousness (consciousness of mental unity, consciousness of one’s own body) by appealing to a pathological case (Oliver Sacks’s “disembodied lady”). In the course of his description of the case, Sacks appeals to Freud’s concept of “ego.” Freud’s “ego” is an “organization of mental events” which is, says Freud, essentially a “bodily ego”3 in that its contents are provided by the connections between our body and the world of objects around us. The point will be taken up again in chapter 7 of the book. Chapter 3 addresses again the question of the respective priority of self-consciousness as unity of consciousness and self-consciousness as consciousness of one’s own body. Here I draw on Jean-Paul Sartre’s analysis of what he calls “non-thetic” or “non-positional” (self-) consciousness. Sartre’s conception of consciousness can be summarized in two notorious statements. On one hand, all consciousness is consciousness of something, or: consciousness is intentionality. This is what Sartre also calls “positional” or “thetic” consciousness. On the other, positional consciousness, namely, consciousness of an object (consciousness as intentionality), is at the same time non-positional (non-thetic) (self-)consciousness: a 3 Freud (1923), SE19:25; Longuenesse (2017), 36. 7 consciousness of the intentional state which is not consciousness of that state as an object (although the state can eventually be made, itself, an object of consciousness, by directing attention to it). In non-thetic (self-)consciousness says Sartre, “there is no room for ‘I’.” And yet, non-thetic (self-) consciousness can be expressed in a first-person statement which is just the expression of that non-thetic (self)-consciousness, not the expression of a consciousness of oneself as an object one directs one’s attention to: the attention remains directed at the object the state is intentionally directed at. Sartre identifies two such types of (non-thetic) selfconsciousness that can come to be expressed in first-person statements: consciousness of one’s own bodily agency, expressed in sentences such as ‘I’m hanging a picture.’ And consciousness of one’s mental agency, expressed in sentences such as ‘I’m counting cigarettes.’ I argue that surprisingly enough, Sartre gives short-shrift to the latter (consciousness of mental agency) and says more about the former (consciousness of bodily agency and more generally, consciousness of oneself as an embodied entity). I argue that this limitation in Sartre’s analysis explains some of its inconsistencies. I argue that identifying those inconsistencies also sheds light on similar strengths and weaknesses in views otherwise as opposed as those of Evans on the one hand and Anscombe (in her celebrated paper “The First Person”) on the other. All three authors (Sartre, Anscombe and Evans) share an avowedly anti-Cartesian position that privileges the consciousness of one’s own body over the consciousness of one’s mental agency. I emphasize again, by contrast, the importance of Kant’s insight into the latter aspect of self-consciousness. This provides the transition to part 2 of the book. Part 2 is devoted to a detailed examination of Kant’s view of the “transcendental unity of apperception” (TUA) and its expression in the proposition ‘I think.’ In chapter 4, I examine TUA and ‘I think’ as they first appear in the central argument of the Critique of Pure Reason, the 8 Transcendental Deduction of the Categories. I develop the point I introduced in Chapter 2: according to Kant, using ‘I’ in the proposition ‘I think’ expresses the consciousness of being, oneself, engaged in a mental activity whose contents are unified according to logical rules. That self-consciousness does not depend on consciousness of one’s own body. Rather using ‘I’ in ‘I think’ is premised on nothing more than the consciousness of a mental activity one takes to be one’s own in virtue of the fact that one takes oneself to be accountable for the correctness of its contents and their connections. I argue that even though he criticizes Descartes’s cogito argument — which he misunderstands — Kant nevertheless maintains, on his own grounds, that ‘I think’ entails ‘I exist.’ Even more radically: ‘I think,’ Kant maintains, just means ‘I exist thinking.’ The thinker’s existence is warranted, for that thinker, by the very consciousness of being engaged in thinking. Nevertheless, against Descartes or the standard rationalist metaphysician he takes to be inspired by Descartes, Kant maintains that the mere thought ‘I think’ gives the thinker no justification for the claim that she exists as a thinking substance distinct from the body. In chapters 5 and 6, I examine Kant’s refutation of three of the arguments he attributes to rationalist metaphysicians, deriving from the mere assertion that I think the claims (1) that I exist as a substance, (2) that I exist as a simple substance, (3) that I exist as a substance conscious of its own identity at different times. I will not review Kant’s refutations here. Let me just highlight two especially significant points. The first is the fact that Kant’s argument trades on the contrast between the first-person and the-third person standpoint on the self (the referent of ‘I’). Kant’s general claim against the rationalist metaphysicians is that they confuse the standpoint which necessarily, each thinker has on herself in virtue of being engaged in thinking, and the third-person standpoint of science or of 9 metaphysics. It is a necessary feature of the first-person standpoint on one’s own thinking that one take oneself to be the metaphysical subject of the thought; it is a necessary feature of that standpoint that one take oneself to be engaged in one process of thinking in which the very meaning of each step is constrained by the meaning of all others so that a thought is one indivisible whole; finally, it is a necessary feature of that standpoint that one take oneself to be one and the same subject conducting one’s reasoning from one step to the next and so, remaining one and the same at different times. This is the first-person standpoint on thinking, justified by the mere fact that it is constitutive of the very act of thinking. But, Kant argues, none of those features give any inkling of the nature of the mind-independent process underlying what, to the thinker’s mind, appears (intellectually) in the ways just outlined. The importance of this point cannot be overstated. It is at the core of Kant’s against the rationalist metaphysicians’ illusions. But it remains open to us to extend it into an acknowledgement of the fundamental difference and equal relevance of the first-person standpoint on one’s thinking and the scientist’s third person standpoint on the brain processes at work when we rationally think. The second point I want to emphasize is that the difference of standpoints I just outlined is especially relevant to Kant’s third argument listed above: his argument concerning the identity at different times of the thinker using ‘I’ in the proposition ‘I think.’ It is a necessary condition of thinking that the thinker take herself to be one and the same author of the thought throughout the process of thinking it. But this, says Kant is not sufficient to justify the metaphysical claim that the entity bearing the act of thinking remains one and the same throughout what the thinker takes to be one and the same act of thinking. To have access to the numerical identity at different times of oneself as the entity one, in fact, is, one needs to complement the first-person standpoint 10 with the third-person standpoint that allows us to track our own identity in space at different times as a physical entity. In fact, it is this aspect of Kant’s view that led Evans to think that he was in agreement with Kant in claiming that only a conception of oneself as a physical entity among other physical entities makes ‘I think’ a referring expression rather than the expression of the mere formal unity of one’s thoughts. Kant, in contrast, is satisfied to just say that even though, in each instance of thinking ‘I think,’ ‘I’ refers to the thinker of that thought, none of the assumptions the thinker thereby makes suffices to warrant any claim concerning the nature or even the numerical identity at different times of that thinker. One might question whether this leaves any room for claiming that ‘I’ (used in ‘I think’ as Kant understands it: as the expression of the transcendental unity of consciousness) is a referential term at all. I claim that it does, but the nature of the referent is not thereby determined beyond the mere statement of the ‘thinkerrule.’ For more determination, one needs more information: whether first-person information (from what Sartre called “non-thetic” self-consciousness, whether bodily or mental) or thirdperson information (as we can get about any other entity in the world). I suggest, moreover, that it is open to us to reject Kant’s claim that from the standpoint of morality, a conception of the referent of ‘I’ in ‘I think’ as a purely thinking being, grounding both our bodily and our mental existence as appearances, gains new traction. I suggest that it is open to us, pace Kant, to salvage Kant’s characterization of epistemic and moral normativity while considering the referent of ‘I’ in ‘I think’ and in the moral ‘I ought to’ as an empirically determinate person, capable of tracking its own identity at different times and of laying out the normative laws of its own action. This leads to the third part of the book. Part 3 builds on a suggestion made in chapter 2: Freud’s structural view of the mind offers a naturalized version of Kant’s transcendental account of mental life. Freud calls ‘ego’ that aspect 11 of our mental life that is structured according to what he calls the “reality principle,” in which mental contents are connected according to elementary logical rules so that a consistent view of the world could be presented. Freud’s “ego,” I argue, has important similarities to Kant’s “transcendental unity of apperception.” I acknowledge the obvious fact that Freud’s view of the mind has different methods and goals from Kant’s. Freud’s approach is clinical: identifying the source of specific types of mental dysfunction. But it is also, in his own words: “metapsychological,” attempting to answer the question: what kinds of structure of mindedness have to be at work so that those dysfunctions and their eventual therapeutic processes can be accounted for? Freud’s ‘ego,’ I argue, is a naturalistic counterpart to Kant’s “transcendental unity of apperception” that finds expression in ‘I think.’ In the context of Freud’s investigation, it is quite clear that ‘I’ refers to a natural entity: a minded body. That entity need not be what we are conscious of in every instance of thinking ‘I think.’ And yet, the ego’s contents are originally provided by the body, which is the repository of all information from the outside world. This point was made at the end of chapter 2 and is taken up again in chapter 7. A similar connection is outlined in the case of morality. This is the object of chapter 8. Here the structures of mental life presiding over the use of ‘I’ in Kant’s moral ‘I ought to’ are compared to Freud’s account of the emergence of morality in the mental structure he calls ‘super-ego.’ I argue again that Freud’s metapsychology offers resources for an account in naturalistic terms of the emergence of those very normative capacities that Kant thought called for a non-naturalistic account. I argue that such an account does not necessarily amount to a debunking argument against the (formal) content of Kant’s categorical imperative, but certainly calls for a careful evaluation of its role in our ethical life. 12 The final chapter (chapter 9) takes stock of the overall argument of the book and outlines directions for further research. I am extremely grateful to Philosophy and Phenomenological Research for organizing a panel to discuss this book and to the critics who were willing to take part in this panel. Every single one of their suggestions and objections gave me a lot to think about. More often than not, trying to respond to their searching comments made me wish I could write the book all over again. I hope at least my next productions will be better for having had to think in unison with these outstanding critics. *** References (for the Précis and Replies) Anscombe, G.E.M. 1975. “The First Person.” In Mind and Language: Wolfson College Lectures, 1974, edited by Samuel Guttenplan, 140–49. Oxford: Oxford University Press. Evans, Gareth. 1982. The Varieties of Reference. Oxford: Clarendon Press. Freud, Sigmund. SE. The Standard edition of the complete psychological works of Sigmund Freud. Edited by J.Strachey et al. London: Hogarth. _____________. 1911. Formulations on the Two Principles of Mental Functioning. In SE 12: 213–26. _____________. 1914 Remembering, Repeating, and Working Through. In SE 12: 145–56. _____________. 1923. The Ego and the Id. In SE 19:1–66. Kant, Immanuel. [A/B] Critique of Pure Reason. Translated and edited by Paul Guyer and Allen Wood. Cambridge University Press. 13 Lichtenberg, Georg Friedrich. 2012. Philosophical Writings, selected from the Wastebooks. Translated by Stephen Tester. SUNY Press. Longuenesse, Béatrice. Forthcoming. The First Person in Cognition and Morality. Oxford University Press. Peacocke, Christopher. 2014. The Mirror of the World. Subjects, Consciousness, and SelfConsciousness. Oxford University Press. Sartre, Jean-Paul. 2018. Being and Nothingness. Translated by Sarah Richmond, with a Foreword by Richard Moran. Oxford: Routledge. Strawson, Peter. 1966. The Bounds of Sense. London: Methuen. Vaihinger, Ernst. 1922. Die Philosophie des ‘als ob’, Leipzig: Felix Meiner.