Problemos ISSN 1392-1126 eISSN 2424-6158

2025, vol. 108, pp. 95–109 DOI: https://doi.org/10.15388/Problemos.2025.108.7

‘Subjectless’ Subjectivity: Anscombe and Sartre on Self-Consciousness

Zhihan Li
Tsinghua University, Beijing, China
lzhphilo@gmail.com
ORCID https://orcid.org/0009-0005-9784-0405
https://ror.org/03cve4549

Abstract. Anscombe famously argues that ‘I’ is not a referring expression, otherwise, a ‘Cartesian Ego’ would inevitably appear. She thinks the form of self-consciousness expressed in ‘I’ is ‘subjectless’: it is consciousness that does not involve a self-object (or Ego). Sartre’s theory of consciousness offers a strikingly parallel and mutually illuminating framework. He similarly denies the presence of the Ego within consciousness, arguing instead that the Ego is a transcendent object constructed through reflection. For both of them, the fundamental form of self-consciousness is pre-reflective. I argue that Anscombe’s position encounters two difficulties: (1) how the non-referential ‘I’ can express subjectivity, and (2) how self-consciousness is about an embodied subject. I propose that Sartre’s accounts of (1) the ‘fundamental ipseity’ and (2) the ‘lived body’ offer a phenomenological framework that helps to respond to Anscombe’s difficulties.
Keywords: self-consciousness, self-reference, bodily awareness, Anscombe, Sartre.

„Besubjektis“ subjektiškumas: Anscombe ir Sartre’as apie savimonę

Santrauka. Gerai žinoma, kad Anscombe teigia, jog „aš“ į nieką nenurodo, kadangi kitu atveju neišvengiamai pasirodytų „dekartiškas ego“. Anscombe mano, kad savimonės forma, išreikšta kaip „aš“, yra „besubjektė“: tai yra sąmonė, kuri neįtraukia savęs kaip objekto (arba Ego). Sartre’o sąmonės teorija mums siūlo neįtikėtinai panašią schemą; Anscombe ir Sartre’o teorijos patikslina viena kitą. Sartreʼas panašiai neigia Ego sąmonėje, vietoj to teigdamas, kad Ego yra transcendentinis objektas, konstruojamas pasitelkiant refleksiją. Tiek Anscombe, tiek ir Sartre’as laikosi nuostatos, kad pamatinė savivokos forma yra ikirefleksyvi. Aš teigiu, kad, laikantis Anscombe pozicijos, iškyla dvi problemos: 1) kaip nereferencinis „aš“ gali išreikšti subjektiškumą; bei 2) kaip savimonė gali būti susijusi su kūnišku subjektu. Teigiu, kad Sartre’o 1) „pamatinė savastis“ bei 2) „gyvenamasis kūnas“ atveria fenomenologinę prieigą, galinčią padėti išspręsti problemas, su kuriomis susiduria Anscombe teorija.
Pagrindiniai žodžiai: savimonė, autoreferencija, savo kūno suvokimas, Anscombe, Sartre’as

_______

Received: 09/10/2025. Accepted: 29/11/2025
Copyright © Zhihan Li, 2025. Published by Vilnius University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (CC BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Introduction

In her famous paper “The First Person” (1981), Elizabeth Anscombe advances a highly unusual view: the first-person pronoun ‘I’ does not refer. If it does refer, she claims, the ‘Cartesian Ego’ would inevitably appear, which must be avoided. When I say, e.g., ‘I am writing’, I express self-consciousness in writing, but I do not refer to the person myself who is writing. The self-consciousness expressed by the non-referential ‘I’ is, as she puts it, ‘subjectless’ (Anscombe 1981: 36), in the sense that it is consciousness that does not involve a self-object (or Ego). As I will argue, Anscombe’s view faces two main problems: (1) if ‘I’ does not refer, how can it nonetheless be used to talk about oneself?; and (2) How can a ‘subjectless’ form of self-consciousness be about an embodied subject? Although Anscombe’s work is typically regarded as a logical-grammatical approach to the first person, the discussion below will show that her position is inseparable from phenomenology. Her phenomenological observations – particularly regarding self-consciousness and bodily awareness – play a foundational role in shaping her views. Although Anscombe and Sartre are often seen as belonging to very different philosophical traditions, my analysis will reveal a deep resonance between their accounts. Similar to Anscombe, Sartre denies the presence of any form of Ego within consciousness and holds that the fundamental form of self-consciousness is pre-reflective. I will argue that Sartre’s accounts of (1) the ‘fundamental ipseity’ of pre-reflective consciousness, and (2) the consciousness-body relation in the ‘lived body’, respectively, suggest phenomenological avenues for addressing Anscombe’s problems, while also illuminating both the core insight and the limitations of Anscombe’s account.

1. Anscombe: ‘I’ does not Refer

Although she does not mention it, the immediate background of Anscombe’s paper is clearly Wittgenstein’s (1969) famous distinction between the use of ‘I’ as subject and its use as object. Wittgenstein discovers a special feature of the use as subject: it is impossible to confuse myself with someone else. By contrast, ‘the use as object’ does allow for misidentification. Shoemaker (1968) calls this feature of the first-person pronoun “immunity to error through misidentification” (hereafter, IEM), and, since then, it has become a classic topic in analytic philosophy of the first person. Anscombe’s paper is an intervention in this debate. She rejects one influential interpretation of IEM – namely, that ‘I’ necessarily refers to the correct subject – because she believes that such a view entails the existence of a ‘Cartesian Ego’. She maintains that the IEM feature is grounded in the fact that: “‘I’ is neither a name nor another kind of expression whose logical role is to make a reference, at all”¹ (1981: 32). For her, the key to properly understanding the use of ‘I’ (as subject) lies in grasping the form of self-consciousness it expresses, and this opens a path for comparison with phenomenology. Now, let me begin to present her main arguments.

In order to argue that ‘I’ is not a name, Anscombe proposes her famous ‘imagined society’ thought experiment (1981: 24). Let us imagine a special society in which each person has two names. Each person has a unique name (“B”–“Z”), marked on their backs and the top of their chests, visible only to others. The other name is the same for everyone: the letter “A”, marked on the wrist and visible to oneself. When seeing others acting, members of this society attribute the action to the person identified by the mark on their back or chest. When attributing an action to themselves, they likewise must look at their own wrist to know that “A is ……”. This thought experiment is meant to show how different our ordinary “I” is from “A”. At first glance, “A” seems to share the essential feature of “I”: each user of “A” uses it to refer to themselves. But, Anscombe points out, the users of “A” differ from the users of “I” in lacking the self-consciousness possessed by the latter (1981: 24). For Anscombe, the basic condition for the use of a name is that it involves criteria of identification, by which we can determine whether the reference is correct. This is exactly the case with “A”. By contrast, the use of “I” (as subject) does not involve any such criterion of identification, for, in using “I”, we do not identify any object as ourselves. Anscombe thinks the exclusion of self-identification is precisely the essential mark of self-consciousness. Since “I” must express self-consciousness, it cannot involve self-identification as “A” does, and therefore it cannot be a name. For Anscombe, the users of “A” manifest not ‘self-consciousness’ but rather ‘consciousness of a self’ (1981: 25), which she sees as dangerously prone to the postulation of a ‘Cartesian Ego’. This wariness toward the Cartesian Ego runs throughout her thought, as it will become more apparent later.

If ‘I’ is not a name, might it instead be some other sort of referring expression – say, a demonstrative (‘this’, ‘that’, etc.)? Anscombe notes that demonstratives are susceptible to reference failure. For example, someone might point to a box and say, “This is all that is left of poor Jones”, when, in fact, the box is empty (1981: 28). Here the demonstrative ‘this’ fails to refer. But, as Wittgenstein and Shoemaker observed, ‘I’ is immune to such failures. However, Anscombe thinks the idea of ‘guaranteed reference’ is mysterious (1981: 30–31). To resist the idea that ‘I’ refers, she introduces another well-known thought experiment. Imagine that I am in a state of complete sensory deprivation: “Sight is cut off, and I am locally anaesthetized everywhere, perhaps floated in a tank of tepid water; I am unable to speak, or to touch any part of my body with any other” (1981: 31). Yet I can still think, for instance, “I won’t let this happen again!”. Even in such an extreme situation, ‘I’ remains available for use, and my thought still manifests self-consciousness. If, in this situation, ‘I’ is a referring expression, what does it refer to? It cannot refer to a body, since, in the absence of all sensory access, I have no way of reaching my body at all. So, it can only refer to a Cartesian Ego – and, for Anscombe, that is unacceptable, as we can give no criterion of identification for such a pure mental entity (1981: 31). Given that the Cartesian Ego must be avoided, she concludes by reductio that ‘I’ does not refer. Once again, she insists that self-consciousness cannot be understood as consciousness of a self-object.

As we can see, although Anscombe’s work is often regarded as a logical-grammatical analysis of ‘I’, her central concern is self-consciousness. This is because she adopts a Fregean account of reference, according to which, a referring expression is characterized by two aspects: reference (Bedeutung) and sense (Sinn). The ‘sense’ is the ‘mode of presentation’ (Art des Gegebenseins) of the object, and Anscombe uses the notion ‘conception’ to capture this aspect. What, then, is the conception of ‘I’? Anscombe’s answer is that it can only be ‘self-consciousness’ (1981: 26), since it is self-consciousness that expresses the way in which the subject is presented. She further holds that the conceptions associated with referring expressions must be ‘mediated conceptions’ – that is, the way they reach their object is always mediated by mediating representations such as observation, inference, testimony, etc. The “A” used by the imagined society involves mediated conceptions: they always reach themselves through some mediating representations, and, for that reason, “A” is a referring expression. By contrast, the peculiarity of ‘I’, in her view, is that it expresses ‘unmediated conceptions’ (1981: 34), i.e., a form of self-consciousness that reaches the subject without mediation. This is why a person in a state of complete sensory deprivation can still have self-consciousness: he can be conscious of himself without any mediating representations. Anscombe boldly proposes that unmediated conceptions are ‘subjectless’: “These conceptions are subjectless. That is, they do not involve the connection of what is understood by a predicate with a distinctly conceived subject” (1981: 36). To illustrate, she offers another vivid example borrowed from William James: a man named ‘Baldy’ was sitting in a moving wagonette, and he accidentally fell out on the road. Strangely, Baldy asked, “Did anyone fall out?” When told that Baldy fell out, he said “Did Baldy fall out? Poor Baldy!” (ibid.). Baldy knows that ‘someone fell out’, knows that the poor guy is called ‘Baldy’, and even sees a body lying on the ground; yet, however much information he possesses, he cannot realize ‘I fell out’. For Anscombe, Baldy’s behavior shows the ‘lapse of self-consciousness’ (ibid.), the lapse lies in that he ‘looked for a subject’ (ibid.) – that is, he can only reach himself via mediated conceptions. The ‘subjectless’ nature of self-consciousness lies in the fact that it is not consciousness of a self-object – an object of mediated conception – but is unmediated conception of oneself.

Although the above-presented thought experiments seem highly intuitive, Anscombe reveals their deeply counter-intuitive conclusion: ‘I’ does not refer. Yet, our ordinary intuition about ‘I’ is that it refers to the very speaker here and now. Is there a possibility of reconciling this tension? In my view, Anscombe’s view faces two main problems:

Problem (1): If ‘I’ does not refer, how can we account for the fact that we consistently use it to talk about ourselves, for instance, in saying “I see an apple”? Such a sentence plainly expresses the subject’s perceptual state. But if the ‘I’ here does not refer, in what sense is the subject speaking about his own perceptual state? Anscombe shifts the discussion from the semantics of ‘I’ to the phenomenon of self-consciousness via the notion of ‘conception’; thereby the problem can be restated as follows: if self-consciousness, as unmediated conception, is ‘subjectless’, how is it nevertheless related to a subject? ‘Unmediated’ is, so far, only a negative characterization: it tells us that self-consciousness does not reach the subject through any mediating representations, but it does not positively explain how it reaches the subject. Without such an account, Anscombe risks the charge of advancing a kind of no-self view – a charge that, in fact, many people levelled against her.

Problem (2): As a determined opponent of Cartesianism, Anscombe insists that self-knowledge must be knowledge of an embodied human subject (1981: 34). Her ‘unmediated conception’ is concerned with bodily states, such as ‘I am sitting’, ‘I am writing’, and ‘I twitched’ (1981: 33–34). It is not concerned with mental states, such as ‘I hope, fear, love, envy, desire’ (1981: 35), which she labels “Cartesianly-preferred thoughts” (ibid.). This is because she thinks “the Cartesianly-preferred thoughts are not the ones to investigate if one wants to understand ‘I’ philosophically” (ibid.). Why does she hold such a radical view toward mental states? The reason is that one’s bodily condition can be directly seen by oneself or others, which makes bodily self-knowledge “directly verifiable or falsifiable” (ibid.). By contrast, mental self-knowledge lacks this feature. Thus, she thinks that bodily self-knowledge should be the starting point for philosophically understanding ‘I’, as judgments about the body have direct means of verification.² Even if we grant her starting point, however, it is unclear how she could move from bodily self-knowledge to mental self-knowledge. Why is the unmediated conception of our ‘actions, postures, and movements’ about a particular body? For Anscombe, the answer seems almost trivially obvious: when I am unmediatedly aware that ‘I am raising my hand’, I – and anyone else – can see that very hand being raised, thereby directly verifying it. But what, then, is our unmediated conception of mental states about? A Cartesian Ego? Anscombe would certainly deny that. If it is about a body, it lacks the means of direct verification. It remains obscure how she could explain that unmediated conception of mental states is conception of a body. As a result, ‘Cartesianly-preferred thoughts’ or mental states become a blind spot in her account.

The central ideas of Anscombe, as we shall see, invite a productive dialogue with phenomenology. Her key point is a strict distinction between unmediated conceptions and mediated conceptions as two radically different modes of presentation – with the former pertaining to subjects, the latter to objects. In phenomenological terms, they are two different kinds of manifestation: self-manifestation and hetero-manifestation.³ According to typical phenomenological views, they are essentially different. Hetero-manifestation is the domain of intentionality: consciousness is always consciousness of something, and that something is apprehended as other. Self-manifestation, by contrast, is the domain of self-consciousness, which is wholly different from intentionality. If self-consciousness takes the self as other, it would objectify the subject. From the phenomenological standpoint, conceiving self-consciousness as a form of intentionality is characteristic of the higher-order (or reflective) theories of self-consciousness, which face many well-known difficulties.⁴ Anscombe’s rejection of the equation “self-consciousness = consciousness of a self” reflects a similar concern. Thus, many phenomenologists insist that self-consciousness must be pre-reflective. As we will see later, Sartre – one of the most thoroughgoing defenders of pre-reflective consciousness – develops an account that resonates deeply with that of Anscombe’s, and offers a phenomenological way for addressing both of the above-outlined problems.

2. Sartre: The Transcendence of the Ego

Like Anscombe, Sartre doubts the existence of any ‘self’ or ‘ego’ within consciousness. In The Transcendence of the Ego (2004), he famously argues that the Ego is not an inhabitant of consciousness, but rather an external and transcendent object of consciousness: “the Ego is neither formally nor materially in consciousness: it is outside, in the world” (Sartre 2004: 1). Sartre’s central points are summarized nicely by Williford:

The phenomenological subject, in this sense, is just the stream of consciousness itself, and the ultimate origin of ‘personality’ is not to be found in a personal ego-entity, but in the self-manifesting stream of consciousness. The self-manifestation of the unified stream is sufficient for basic selfhood. (2011: 208)

To defend his view, Sartre advances three lines of argument: the Ego’s presence in consciousness is neither necessary, nor possible, nor factual.

(1) Against the necessity of the Ego

The traditional view holds that the Ego is a necessary condition for consciousness to have unity and individuality. First, the Ego was said to provide synthetic unity of consciousness; without it, consciousness would be chaotic and unstructured. Second, the Ego was thought to guarantee individuality, distinguishing one person’s consciousness from another’s. Yet, Sartre argues that, for consciousness to have unity and individuality, the Ego is unnecessary (2004: 3–4). Drawing on Husserl’s phenomenology of internal time-consciousness, Sartre thinks that the unity of consciousness arises from its own temporal structure. Through the interlocking triad of primal impression, retention, and protention, each moment of consciousness retains the just-past and anticipates the about-to-come, binding discrete experiences into a unified, continuous flow – without the need for an Ego to provide unity. As for individuality, each stream of consciousness is intrinsically distinct from other streams of consciousness. No Ego is needed to guarantee individuality.

(2) Against the possibility of the Ego

Sartre contends not only that the Ego is unnecessary, but that its presence in consciousness would be a hindrance. Consciousness, for Sartre, is entirely transparent and empty – it is conscious of something and is exhausted by this intentionality. If there is an Ego in consciousness, “it would divide it, slicing through each consciousness like an opaque blade” (2004: 4). That is because the Ego is essentially heterogeneous with consciousness: it is opaque and transcendent. The existence of the Ego would introduce an opaque center in consciousness, which would destroy the transparency of consciousness. The transcendence of the Ego lies in the fact that it shares with perceptual objects the structure of adumbration: it always appears in partial profiles and can never be grasped in its entirety. For example, a character trait like ‘brave’ belongs to Sartre’s ‘Ego’. As a disposition, it is manifested only piecemeal in concrete acts, and never wholly present. Such a transcendent nature precludes the Ego from being an inhabitant of consciousness; it can only be an object of consciousness.

(3) Against the factuality of the Ego

Finally, Sartre points to our actual lived experience. In engaged, unreflective activities, we are in a state of ‘self-forgetfulness’: there is no felt presence of an ‘I’. When absorbed in reading a book, what we are conscious of is the book and the characters of the book (2004: 7), and not of ‘I am reading the book’. When spontaneously chasing after a departing tram, what we are conscious of is the ‘tram-needing-to-be-caught’ (2004: 8), and not of ‘I need to catch the tram’. In fact, “there is no I on the unreflected level” (2004: 7). Only when we focus on ourselves by reflection does the ‘I’ appear.

From these considerations, Sartre concludes: there is no Ego in the original structure of consciousness. The Ego is a transcendent object, it is an ‘other’ that can never be manifested in its entirety. He thinks the source of the Ego is reflection. Reflection disrupts the world-directed living flow of consciousness, objectifying it and thereby generating the Ego as an object. As Anscombe would put it, consciousness of an Ego is not ‘self-consciousness’, but rather ‘consciousness of a self’. Sartre even suggests that the sure and certain content of consciousness is not “I am conscious of this chair”, but that “there is consciousness of this chair” (2004: 9). This, of course, invites the same question posed to Anscombe: Does such an account amount to a no-self view? How does it differ from Lichtenberg’s (2000) notorious suggestion that we should say ‘it thinks’ rather than ‘I think’?

Let us recall the distinction mentioned in the previous section between self-manifestation and hetero-manifestation. For Sartre, the Ego can only be manifested by hetero-manifestation. Self-consciousness, as self-manifestation, is essentially different from hetero-manifestation. It is in the analysis of self-manifestation that Sartre avoids the charge of the no-self view. In the next section, I will outline Sartre’s account of self-manifestation and use it to address Anscombe’s Problem (1).

3. Sartre on Self-Manifestation – Addressing Problem (1)

Sartre’s account of self-manifestation arises from his investigation into the nature of consciousness. Following Husserl, he thinks that the essence of consciousness is intentionality – i.e., that “all consciousness is consciousness of something” (Sartre 2021: 9); or he would say that consciousness always ‘posits’ an object. So, intentionality is ‘positional consciousness’, which belongs to hetero-manifestation. Crucially, Sartre thinks that, for a positional consciousness to be given to the subject, it does not require a second-order positional consciousness – that is, reflective consciousness. Instead, Sartre maintains, that every positional consciousness is given to the subject pre-reflectively by ‘non-positional consciousness’ – “Any positional consciousness of an object is at the same time a non-positional consciousness of itself” (2021: 11). Non-positional consciousness is non-intentional and non-objectifying; it is not a higher-order consciousness but an integral part of every consciousness, making every consciousness directly presented to the subject. Thus, non-positional consciousness is the self-manifestation of consciousness, it is the fundamental form of self-consciousness. For Sartre, non-positional consciousness is the precondition of reflective self-consciousness: “nonreflective consciousness is what makes reflection possible: there is a prereflective cogito, which is the condition of the Cartesian cogito” (2021: 12).

Sartre’s ‘non-positional consciousness’ corresponds precisely to Anscombe’s ‘unmediated conception’: both designate the pre-reflective, non-objectifying form of self-consciousness. As mentioned in Problem (1), Anscombe’s difficulty is: if self-consciousness, as unmediated conception, is ‘subjectless’, how is it related to a subject? Sartre’s phenomenology of consciousness suggests a way for addressing it.

Sartre’s rejection of the Ego does not entail a rejection of subjectivity. On the contrary, he holds that consciousness is inherently subjective, and this subjectivity is the ‘fundamental ipseity’ (2021: 160) that is expressed by non-positional consciousness. It is the pre-reflective ‘fundamental ipseity’ that makes the reflective ‘Ego’ possible, just as pre-reflective consciousness makes reflective consciousness possible (ibid.). How, then, does non-positional consciousness express fundamental ipseity? For example, when I see an apple before me, I have a positional consciousness, ‘There is an apple’. This is accompanied by a non-positional consciousness that, if expressed in language, would be, ‘I see there is an apple’. The crucial point is that this ‘I’ does not express an Ego (in Anscombe’s words, it does not refer). On Sartre’s account, the ‘I’ merely expresses the ‘self-presence’ (ibid.) of consciousness. The ‘I’ is not an entity, but it designates the very manner in which consciousness is presented – it is always presented to ourselves. This self-presenting manner is precisely the most basic subjectivity, or ‘fundamental ipseity’. This point is developed by Zahavi, whose formulation can help us understand Sartre’s view more clearly.

Zahavi (2014) maintains that phenomenal consciousness has not only a ‘what-it-is-like’ dimension but also a ‘for-me’ dimension. Every ‘what-it-is-like’ is, more precisely, ‘what-it-is-like-for-me’. The taste we experience when eating a lemon is, more exactly, ‘lemon-taste-for-me-ness’. To avoid the Lichtenbergian ‘no owner’ model of experience, experience must have a minimal dimension of selfhood, which Zahavi calls the ‘for-me-ness’ of experience. Importantly, Zahavi stresses that ‘for-me-ness’ is not an extra quale to the ‘what-it-is-like’, as if in eating a lemon one experiences both the sourness and a feel of ‘I’. Rather, “the ‘me’ of for-me-ness is not in the first instance an aspect of what is experienced but of how it is experienced; not an object of experience, but a constitutive manner of experiencing” (Zahavi et al. 2016: 38). In other words, the ‘me’ in ‘for-me-ness’ is not an object ‘Ego’, but expresses the first-personal manner in which the lemon is experienced. The subjectivity of experience does not require an Ego, because the first-personal manner of experiencing is itself a ‘minimal self’. This first-personal manner of experiencing is exactly Sartre’s ‘non-positional consciousness’, for it captures the self-presence of experience. And Zahavi’s ‘minimal self’ captures Sartre’s ‘fundamental ipseity’.⁵

With this in view, we can see how Problem (1) can be addressed. The way in which her ‘unmediated conception’ is related to a subject is precisely through the fundamental ipseity that non-positional consciousness expresses. When one says ‘I see there is an apple’, the ‘I’, as Anscombe insists, does not refer. Yet, it is an expression of subjectivity, for it expresses that the experience of the apple is ‘for-me’.

On this basis, we are in a position to evaluate Anscombe’s unusual claim that ‘I’ does not refer. Her reasoning is that reference, as she understands it, must correspond to a mode of presentation that is mediated conception (hetero-manifestation). What ‘I’ expresses, by contrast, is unmediated conception (self-manifestation). Because the two modes of presentation cannot be confused, ‘I’ cannot refer. Thus, the question of whether ‘I’ refers turns, in the end, on how one understands the relation between semantics and phenomenology. Must ‘reference’ (a semantic matter) correspond only to hetero-manifestation (a phenomenological matter)? Or can self-manifestation also be regarded as a kind of reference? I think this is basically a matter of stipulation. If ‘reference’ is defined so that it must correspond to hetero-manifestation, then Anscombe is right to say that ‘I’ does not refer. But if we broaden the concept of ‘reference’ to include self-manifestation, then it is right to say that ‘I’ refers. This is not unacceptable for Anscombeans, for her core idea is the strict distinction between mediated and unmediated conceptions. If Anscombeans accept that reference can include unmediated conceptions, they would remind us how reference to others and ‘reference’ to oneself are entirely different. My aim here is not to take a stand on how ‘reference’ should be stipulated, but rather to suggest that, by drawing on phenomenology, we can better see where the truth of Anscombe’s striking view lies. ‘I’ does not refer, because subjectivity cannot be manifested by hetero-manifestation. The way in which the non-referential ‘I’ expresses subjectivity is by expressing the for-me-ness of experience. Anscombe’s key insight is that self-consciousness cannot be identified with ‘consciousness of a self’, and, in phenomenology, we find both a resonance with, and a richer articulation of, that insight.⁶

4. Sartre on Embodiment – Addressing Problem (2)

Anscombe maintains that ‘unmediated conceptions’ concern bodily actions, postures, and movements, which is directly verifiable. But how can unmediated conceptions of one’s mental states be about a body? As noted earlier, this is a blind spot in Anscombe’s account. For Sartre, this problem cannot be solved if we cling to the traditional picture of the consciousness-body relation. On the traditional view, body is inert and spatial, consciousness is active and non-spatial; and consciousness inhabits in body, issuing commands to it. And yet Sartre insists that the true consciousness-body relation is that consciousness ‘exists its body’ (2021: 441) – to borrow his deliberately unorthodox phrasing. To see what this means, let us begin with Sartre’s well-known passage:

The problem of the body and its relations to consciousness is often obscured by the fact that we posit the body right from the start as a certain thing, possessed of its own laws and capable of being defined from the outside, while we reach consciousness through the type of inner intuition that is distinctive to it. If…… I try to join it to a certain living object, constituted by a nervous system, a brain, some glands, some digestive, respiratory and circulatory organs, ……I will encounter insurmountable difficulties: but these difficulties stem from the fact that I am not trying to join my consciousness to my body, but to the body of others. In effect, the body whose description I have just sketched is not my body, as it is for me. (2021: 409)

To grasp the consciousness-body relation, Sartre argues, we must start from the first-person perspective and investigate how the body as subject, the body ‘for me’, is given. In the first-personal perspective, the body is not a physiologically structured object defined from the outside, and our awareness of the body is not an inner observation of it. As subject, the body is lived body – “it is lived, and not known” (2021: 435).

For Sartre, in normal circumstances, the lived body is transparent and is wholly directed toward the external world, and it is given to us in pre-reflective consciousness. Sartre points out that the body is the center of perception: it is the zero point from which we represent the spatial relations between external things and ourselves, so that their orientation, distance, and size are presented perspectivally. Whenever we represent an object as being near or far, left or right, large or small, we implicitly apprehend our body. Yet the body, as the center of perception, is not itself an object of perception: “Only we do not see this center as a structure of the perceptual field in question: we are it” (2021: 427). The body is also the initiator of action, our most closely accessible instrument. When we are aware that “this painting is to be hung” or “this hammer fits well in my hand”, we also implicitly apprehend our body. Sartre thinks the instruments we use in action form an instrumental series, e.g., “hand–hammer–nail–board …”, with the body as its starting point. But “we do not employ this instrument [i.e., body]; we are it” (2021: 434). Sartre argues that, if the body had to be perceived and employed as an object, this would presuppose that there is a body* as a subject perceiving and employing it. And if that body* could only be perceived and employed as an object, this would require a body** – which generates an infinite regress. Thus, Sartre concludes, the body as subject is invisible and unusable: “It is the instrument that I am unable to use by means of another instrument, the point of view on which I can no longer take a point of view” (2021: 441). Since the body we live as subject is pre-reflective and non-objectified, our awareness of it must be non-positional consciousness. When perceiving and acting with our body, we are conscious of external objects positionally, and conscious of our body non-positionally. This also accords with common sense: in much of everyday action, our attention is on the external world, while “the body is what is neglected, ‘passed over in silence’” (2021: 442). Sometimes ignoring one’s body is even a requirement: e.g., when playing tennis, one should keep one’s attention on the coming ball, not on one’s body. And sometimes the emergence of the body, such as a sudden pain, would disrupt actions. To use the body fluently, it should be “passed over in silence”.

From this, we can see that for an unmediated conception to be about a body, what is needed is not, as Anscombe suggests, the means of direct verification.⁷ Third-person verification treats the body as an object, making it the target of positional rather than non-positional consciousness. Consciousness of the body as subject is non-positional consciousness; it is about a body by ‘for-me-ness’, or more exactly, ‘for-my-body-ness’. As Sartre points out, when we perceive an object, we are always (pre-reflectively) conscious of its orientation, distance, and size relative to our body. When we use an instrument, we are always (pre-reflectively) conscious of its relation to the posture and movement of our body. The lived body is at once ‘passed over in silence’ and omnipresent. Even without positional consciousness of the body, we already have pervasive awareness of our body in our absorbed engagement with the world.⁸

A remaining question is: perception and intention in action are only a subset of mental states. How can consciousness of other kinds of mental states also be about a body? For Sartre, the answer lies in the nature of mentality itself. The essence of the mind is intentionality – i.e., the relation between the subject and the world. Following Heidegger, Sartre conceives this relation as ‘being-in-the-world’. The being-in-the-world is a being who deals with objects in the world, practically organizing them according to its projects. But in order to deal with worldly objects – using them, approaching them, avoiding them, etc. – the being-in-the-world must also be one object among others in the world. He writes “to say that I have entered the world, ‘come to the world’, or that there is a world in which I have a body comes to one and the same thing” (2021: 427). That is because only an embodied subject can truly have a ‘world’: the shirt is large or small, the axe is heavy or light, the mountain peak is near or far – all these make sense only for an embodied subject, and not for a Cartesian subject. As long as the subject exists in-the-world, he must have some understanding of his objectivity, even though, most of the time, this understanding is implicit, non-positional consciousness.

From this perspective, the so-called ‘Cartesianly-preferred thoughts’ that Anscombe sets aside – “I hope, fear, love, envy, desire” (1981: 35) – are in fact not ‘Cartesian’ at all. They all have to relate to the subject’s embodiment if they make sense. Hope and desire are directly linked to the subject’s projects toward the future. For Sartre, a project is not a private inner wish, but a practical projection of oneself into the world. In this sense, a project is world-directed; it always involves the grasp of pragmatic information of the world relative to one’s body. Thus, a project makes sense only for an embodied being-in-the-world. Fear, love, and envy are emotions. Sartre also rejects the idea that emotions are merely private inner feelings; he thinks that “emotion is a specific manner of apprehending the world” (2014: 35). Fear always apprehends the fearful aspect of the world; love and envy always apprehend the lovable and enviable qualities of the world. These, too, make sense only for an embodied being-in-the-world. Just as perception always involves the grasp of spatial information of the world relative to one’s body, emotion always involves the grasp of affective quality of the world relative to one’s body. This suggests that the implicit connection with the body is embedded in our egocentric modes of thinking about the world; and these modes can be spatial, pragmatic, affective, etc. This point does not depend on Sartre’s more grandiose philosophical vocabulary; it stands independently and has, in fact, found support within the ‘analytic’ tradition as well.⁹ In short, a wide range of mental states involve the subject’s apprehension of the world, and thus involve an at least implicit understanding of its embodiment. Without it, the subject would not be ‘in-the-world’ at all.¹⁰

Thus, unmediated conceptions of both bodily and mental states are always about an embodied subject. Third-personal means of verification neither apply to mental states nor adequately explain how unmediated bodily awareness is about a body. For Sartre, mentality is world-directed intentionality, and it is always egocentric. The egocentric modes of thought – including spatial, pragmatic, affective, etc. – all presuppose the subject’s understanding of its embodiment. This understanding of the body is non-positional bodily consciousness; it is precisely in this understanding that the subject ‘exists its body’.¹¹

5. Conclusion

Anscombe and Sartre are often thought to belong to very different philosophical traditions, yet their views on self-consciousness reveal a remarkable degree of mutual illumination. While this point has been noted in previous scholarship, my contribution lies in providing a phenomenological grounding for some of Anscombe’s intuitive observations and the problem she confronts. Although this paper focuses on how Anscombe benefits from phenomenology, conversely, Sartre also benefits from analytic philosophy. Through engagement with analytic philosophy, some of the seemingly grandiose elements of Sartre’s philosophy can be articulated more clearly, as reflected in contemporary research on pre-reflective consciousness and egocentric modes of thought mentioned earlier. Furthermore, Sartre can benefit from Anscombe in another significant respect. Anscombe’s linguistic approach to ‘I’ is part of her Wittgensteinian project of grammatical investigation. The ‘I’ is not only used by individuals in isolation; it is part of a public language, reflecting the language games and form of life of a community. Only by becoming a member of a linguistic community can we possess a full-fledged ‘I’. The topic of the subject acquiring a full self-conception through encounters with others is also discussed in Sartre’s account on the ‘look’ and the third ontological dimension of the body (see footnote 7). Anscombe’s lesson for Sartre is that the acquisition of a full self-conception occurs not merely under the threatening ‘look’ of others, but rather within harmonious exchanges of words and concordant rule-following practices. While this opens an interesting avenue of inquiry regarding intersubjectivity, its full elucidation remains a task for future research.

References

Anscombe, G. E. M., 1957. Intention. Oxford: Basil Blackwell.

Anscombe, G. E. M., 1981. The First Person. In: Metaphysics and the Philosophy of Mind. Oxford: Basil Blackwell, 21–36.

Aucouturier, V., 2023. Self-Consciousness and Uses of “I”: Sartre and Anscombe. In: Sartre and Analytic Philosophy, ed. T. Morag. London: Routledge, 65–88. https://doi.org/10.4324/9780429455919-4

Boyle, M., 2024. Transparency and Reflection: A Study of Self-Knowledge and the Nature of Mind. New York: Oxford University Press.

Evans, G., 1982. The Varieties of Reference. Oxford: Oxford University Press.

Howell, R. J., 2023. Self-Awareness and The Elusive Subject. Oxford: Oxford University Press.

Kriegel, U., 2009. Subjective Consciousness: A Self-Representational Theory. Oxford: Oxford University Press.

Lichtenberg, G. C., 2000. The Waste Books. Translated by R. J. Hollingdale. New York: New York Review of Books Classics.

Longuenesse, B., 2024. “I” and Self-Consciousness. In: Analytic Existentialism, eds. B. Marušić, M. Schroeder. Oxford: Oxford University Press, 54–81. https://doi.org/10.1093/oso/9780192864215.003.0003

Narboux, J.-P., 2018. Is Self-Consciousness Consciousness of One’s Self? In: Wittgenstein and Phenomenology, eds. O. Kuusela, M. Ometita, T. Uçan. New York: Routledge, 197–247. https://doi.org/10.4324/9781315626307

Sartre, J.-P., 2004. The Transcendence of the Ego. Translated by A. Brown. London: Routledge.

Sartre, J.-P., 2014. Sketch for a Theory of the Emotions. Translated by P. Mairet. London: Routledge.

Sartre, J.-P., 2021. Being and Nothingness. Translated by S. Richmond. New York: Washington Square Press.

Shoemaker, S., 1968. Self-Reference and Self-Awareness. Journal of Philosophy 65(19): 555–567. https://doi.org/10.2307/2024121

Williford, K., 2011. Pre-reflective self-consciousness and the autobiographical ego. In: Reading Sartre: On phenomenology and existentialism, ed. J. Webber. Abingdon: Routledge, 195–210. https://doi.org/10.4324/9780203844144

Wittgenstein, L., 1969. The Blue and Brown Books. Oxford: Basil Blackwell.

Zahavi, D., 2014. Self and Other: Exploring Subjectivity, Empathy, and Shame. Oxford: Oxford University Press.

Zahavi, D., 2020. Self-awareness and alterity: A phenomenological investigation (A New edition). Evanston, IL: Northwestern University Press.

Zahavi, D., Kriegel, U., 2016. For-Me-Ness: What It Is and What It Is Not. In: Philosophy of Mind and Phenomenology: Conceptual and Empirical Approaches, eds. D. O. Dahlstrom et al. London: Routledge, 36–53.