Flying Man?

Back in 2019, I included the following in a tangent near the end of a longer post:

“The thrust of the famous cogito ergo sum was already anticipated in Augustine’s Confessions. A more detailed version was developed by Avicenna, in an argument known… as the “flying man”. He proposed a thought experiment, considering someone in counterfactual absolute sensory deprivation from birth, with the intent of asking whether awareness could be completely independent of sensibility. He argued that the person in absolute sensory deprivation would still be aware of her own existence, due to a pure immediate reflexive awareness intrinsic to the soul and independent of the body. This kind of claim would have been accepted by Plotinus, but rejected by Aristotle or Hegel. Medieval Augustinians, however, enthusiastically adopted many of Avicenna’s ideas.” Peter Adamson, a scholar of Islamic philosophy, has a nice short article on this.

What Descartes Proved

The Latin subtitle of Descartes’ Meditations advertises that they not merely assert but demonstrate the existence of God and an immortal soul. However, as noted in the Cambridge Descartes Lexicon, neither the word “immortality” nor “immortal” even appears in the body of the text. Descartes does convincingly argue that even in the very act of doubting everything, it is apparent that something is going on — that there is awareness or “thinking”, even on the extreme assumption that everything it thinks is mistaken.

“What then did I formerly believe myself to be? Undoubtedly I thought myself to be a man. But what is a man? Shall I say a rational animal? No, for then I should have to inquire what is ‘animal’, what ‘rational’; and thus from one question I should be drawn on into several others yet more difficult. I have not, at present, the leisure for any such subtle inquiries. Instead I prefer to meditate on the thoughts which of themselves spring up in my mind” (Descartes’ Philosophical Writings, trans. Norman Kemp Smith, pp. 203-204).

Descartes argues that bodily phenomena are not clearly “me”, but that I cannot separate myself from my awareness. “Thinking? Here I find what does belong to me: it alone cannot be separated from me. I am, I exist. This is certain. How often? As often as I think” (p. 205).

He makes a subtle transition from arguments about the indubitable existence of some awareness to arguments about something that has the awareness.

“What is a thinking thing? It is a thing that doubts, understands, affirms, denies, wills, abstains from willing, that also can be aware of images and sensations” (p. 206). Elsewhere he also includes love and hate. Using contemporary vocabulary, Gueroult calls this a thinking subject (Gueroult, Descartes, p. 40), a word Descartes himself does not use in this context. Descartes identifies it with his own view of the soul. Though Descartes thinks it is separable from the body and the world, he treats it as a concrete “me”.

This concrete me-thing is supposed to have some very strong properties. Descartes claims it is a substance — apparently in the sense of something underlying — and that we can know something is a substance without knowing it as a substance. Secondly, “Interior to the Cogito, consciousness [French conscience, which predated Cudworth and Locke’s English word] and consciousness of consciousness are identical” (Gueroult, p. 99, my translation, brackets and emphasis added). Thirdly, the Cartesian soul is supposed to have “infinite” free will.

On my reading, the earlier transition from the indubitability of the existence of some concrete awareness — which I take to be genuinely irrefutable — to the assumption of a “thing” that has the awareness, is only valid if the “thing” is strictly identified with the awareness itself. But if the “thing” is identical with the awareness we started with, then the transition to calling it a substance is not justified. Conversely, if “thing” does have strong enough meaning to justify calling it a substance, the transition from thinking to a thinking thing was unjustified.

The identity of consciousness and consciousness of consciousness seems to be a completely unproven and unprovable assumption — one that I think Aristotle, Kant, Hegel, and Ricoeur each convincingly suggest we should oppose. Gueroult quotes Descartes in a letter making the weaker claim that (contrary to what Locke would assert a bit later) the soul is capable of thinking of many things at once. If we accept this, that would make a simultaneity of consciousness and consciousness of consciousness possible in particular cases, but it still falls far short of establishing their identity.

Like many voluntarists, Descartes makes a leap from the much weaker claim that we make real choices, to the ultra-strong claim that our power of choice is completely unconditioned, and that divine omnipotence exempts it from the natural order altogether.

He argues at length that an omnipotent God is the necessary foundation of the razor-thin subjective certainty he discovered by introspection. The arguments about God are introduced as follows: “Those [ideas] which represent substances are without doubt something more, and contain in themselves, so to speak, more objective reality (that is to say participate by representation in a higher degree of being or perfection) than those which represent only modes or accidents; and again, the idea by which I apprehend a supreme God, eternal, infinite, immutable, omniscient, omnipotent, and the creator of all things which are on addition to Himself, has certainly in it more objective reality than those ideas by which finite substances are represented.”

“Now it is manifest by the natural light that there must be at least as much reality in the efficient and total cause. For whence can the effect draw its reality if not from its cause? How could this cause communicate to it this reality if it did not itself have it? And hence it follows… that what is more perfect, i.e. contains more reality, cannot proceed from what is less perfect” (pp. 219-220).

“By the name God I mean a substance that is infinite, immutable, independent, all-knowing, all-powerful, and by which I myself and everything else, if any such things there be, have been created. All those attributes are so great and so eminent, that the more attentively I consider them the less does it seem possible that they can have proceeded from myself alone; and thus, in the light of all that has been said, we have no option save to conclude that God exists” (p. 224).

Out of nowhere, we suddenly have an unexplained appeal to a fully formed late-scholastic distinction between substance, mode, and accident, and other scholastic reasoning about the superiority of causes to their effects. The implicit claim seems to be that these very particular historical formulations are universally given in intellectual intuition.

Descartes’ subsequent argument is literally that God is the efficient and total cause of everything, therefore He exists. Need I point out that only an existing being could be an efficient and total cause in the first place (assuming there is such a thing as a total cause)? And that the argument is therefore patently circular? Though still debatable, Anselm’s ontological proof and the various arguments of Aquinas were much more persuasive than this.

Plotinus, Augustine, and Avicenna all anticipated Descartes’ strong sense of interiority. What was relatively new in the metaphysics of Descartes was his narrow point about the otherwise empty certainty that “awareness exists”, but even that was partially anticipated by Avicenna’s “flying man” argument.

Subjectivity in Plotinus

Plotin: Traité 53 (2004) is Gwenaëlle Aubry’s contribution to the French series Les écrits de Plotin. It is a translation of and commentary on the treatise Porphyry placed at the very beginning of the Enneads when he edited his teacher’s works. In this essay Plotinus broaches the question who “We” are.

Aubry translates the title of Plotinus’ essay as “What Is the Animal? What Is the Man?”. (The classic English translation by Stephen MacKenna called it “The Animate and the Man”.) But she says “The real question of treatise 53 is not that of man, but rather of the subject…. [T]here appears something like a subject in the modern sense of the term, that is to say a reflexive consciousness, capable of asking itself about its operations and its identity” (p. 17, my translation throughout).

Plotinus develops his unique view of soul (psyche) as in effect a sort of unmoved mover. “We”, it will turn out, are for him neither the pure soul nor a union of soul and body. Aubry says here that for Plotinus the subject is not a substance and has no identity. It is rather a “pure power of identification” (p. 18).

It was precisely the classical identification of subject with substance that led Augustine — who was deeply impressed by Plotinus — to insist that the mind is not a “subject”.

“Finally, the theme of the immaculacy of the separated soul is another way of underlining the responsibility of the “We”: it is on us, ultimately, that the ethical decision depends, understood as a choice of identification with that which exceeds us or with that which hinders us, with what founds us or with what weakens us — with the divine or with the animal in us” (p. 18).

“The itinerary of treatise 53 is nothing other, finally, than the passage from the me to the self. What the treatise proposes to its readers, what it proposes to ‘us’, is not to discover our identity — for identity we do not have. It is not to define our essence — for the essence is not ‘us’, but the self — it is to identify ourselves with another object than that with which we spontaneously confuse ourselves” (ibid).

For Plotinus, a mere conversion to interiority is not sufficient to disclose the self, Aubry says. That just gives us a sensible, empirical “me”, a subject of passions. “The whole effort of the treatise… is to redirect consciousness away from that immediate and fascinating object, to orient it toward the impassible and separated soul in which essential identity resides. Thus, the work of definition imposed by the Delphic precept [know thyself] is inseparable from an ethical work: the self cannot be determined except at the price of a renunciation of its first object of identification” (p. 20, brackets added).

She quotes Plotinus asking, “that which investigates, which examines and poses these questions, what could it be?” (p. 23).

“At this point, the treatise takes up a new orientation, engages itself on a radically unexplored path: for the subject in the classical sense of the term — substance, the subject of attribution — is substituted in effect, by the detour of a phrase, a modern subject, possessed of consciousness and reflexivity. The Plotinian project here distinguishes itself radically as much from that of the De Anima [of Aristotle] as from that of the First Alcibiades [of Plato]. It is no longer only to examine the various faculties to distinguish among them which are common to the soul and the body, and which are proper to the soul alone. It is no longer only to decide whether man is the soul, the body, or the mixture of soul and body. It’s about, writes Plotinus, asking oneself about the very thing that does the investigation: the philosophizing subject takes itself as the object of investigation. The conversion is no longer only to interiority, but to consciousness: and that consciousness takes the form of an immediate reflexivity” (ibid, brackets added).

Plotinus seems to have been the first to claim this sort of immediate reflexivity of consciousness. As Aubry goes on to note, for both Plato and Aristotle, reflexivity only comes through the mediation of another. “The other is not only another self, but it is through the other, insofar as the other is at the same time like me and other than me, that I have access to myself…. The access to interiority requires a detour, either by way of exteriority, or by otherness” (p. 25).

Hegel and Paul Ricoeur, I would note, have each in their own way again taken up Plato and Aristotle’s emphasis on mediation and the need for a detour. But even though Plotinus claims an immediate reflexivity, he does not claim that it is or has an identity.

For Plotinus, according to Aubry, “the we is neither the incarnate, empirical me, nor the separated essential self; it is the passage from the one to the other” (p. 27). In the related essay “A Me Without Identity? The Plotinian We” in Le moi et l’interiorite (2008), she writes, “Returning into itself, thus, [the ‘We’] does not know itself as a unity, nor in its identity with itself, but as a mixture of two terms, where neither is, properly speaking, ‘itself'” (p. 108).

Narrative Identity

Chapters 5 and 6 of Ricoeur’s Oneself as Another are concerned with the concept of narrative identity, developed in his earlier three-volume work Time and Narrative . We get at least the beginning of an answer to the doubts that occurred to me about a concept of identity based on reflexivity rather than sameness, when he applied it to Aristotle. I have not yet consulted the earlier work, but the treatment of narrative identity here itself seems a bit literary and elliptical. On the other hand, reading this I had the thought that literary theory was finally giving something back to philosophy, after so much borrowing in the other direction.

Narrative identity is intended to be a sort of Aristotelian mean between an identity of character — which according to Ricoeur follows the pattern of sameness — and an opposite pole he introduces, associated with what he calls self-constancy, which is to follow the pattern of reflexivity. Self-constancy is associated especially with keeping promises, and more generally with being reliable. This provides a more concrete model of what identity based on reflexivity is supposed to look like. Self-constancy involves success in a constant, self-directed effort. The self-directedness of the effort makes it clearly reflexive, in a temporally extended sense that seems much less problematic than an instantaneous reflexivity.

This notion of self-constancy reminds me of what we would get if we separated out Brandom’s ethical use of a responsibility or imperative to aim for consistency in one’s commitments, and directly gave it a more explicit temporal dimension that for Brandom arises mainly in a larger context. Narrative identity itself seems like the kind of thing that for Brandom is constructed writ large by Hegelian genealogy. (See also Time and Narrative; Narrated Time; Narrative Identity, Substance.)

Ascription of Actions

After the disappointing result from traditional analytic semantic approaches to action, Ricoeur turns to the pragmatics of action, and to applying Strawson’s notion of ascription to persons.

He discusses Aristotle’s distinctions of willing and unwilling actions and choice at some length. Unlike Donald Davidson, who only had a modern notion of (physical) cause to work with, Aristotle had a neutral concept of arche or “principle” that applies equally well to ethical and physical instances, like his broad notion of “cause” as a reason why. According to Ricoeur, Aristotle ascribes actions to a principle that is a “self”. Ricoeur also notes that Aristotle speaks of us as synaition (co-responsible for, or co-causing in Aristotle’s broader sense) our dispositions and character.

Aristotle himself did not actually use a word like “self” in this context, but attributed choice to “either intellect fused with desire, or desire fused with thinking, and such a source is a human being” (Nicomachean Ethics, Sachs translation, p. 104). Even the term “fused with” turns out to be an interpolation by the translator here — the Greek just has “intellect and desire”, and says nothing about how they are related. I agree there is a kind of reflexivity within the thought and desire involved here, but I’ve been taking it to be of the adverbial sort. I have so far used the term “self” either adverbially, or for a matter-of-fact emotional constitution inter-articulated with an intimate but anonymous transcendental but historical ethos. (Later note — in an earlier work, Ricoeur had proposed a notion of ethical Self as an aim, which I am now adding into my own view. Such an interpolation seems at least compatible with the broad spirit of Aristotle, despite its anachronistic character at a literal level.)

I’m awaiting further clarification of how Ricoeur’s ipse identity is supposed to work in a positive sense (through a sort of continuity of development?); how that would apply to the combination that is mentioned but not elaborated on by Aristotle; and whether the application of ipse identity — which I suspect would be warmly welcomed in a Thomistic context — is intended to be understood as historically Aristotelian, or as a post-Aristotelian original thought. The novel semantic category of ipse identity seems well suited to capture intuitions uniting self with responsibility, and potentially to solve some difficulties with which I have struggled. But so far, its application here is not fully explained. (For the beginning of a resolution, see Narrative Identity. For an actual resolution, see Self, Infinity.)

Turning to Strawson, Ricoeur argues that ascription of actions to persons is different from logical attribution of properties to objects, and that it implicitly involves the kind of reflexivity found in self-designating utterance. (I can grant the difference between ascription and attribution, but it is as yet unclear to me in what way he wants us to see that ascription necessarily involves reflexivity, since ascription does not involve self-designation.) He says we first ascribe actions to persons, and only then do we ask about their intentions. Motives, he says, are mainly relevant in hindsight when we ask about an action that has occurred. Also, the “who” behind an action is expected to have a definite answer, whereas motives depend on other motives, and so on indefinitely. The notion of an agent as the “who”, Ricoeur says, is this time successfully reached. Its actual meaning depends on the whole related network of the “what”, “why”, and “how” of the action.

Ricoeur nonetheless finds a difficulty in Strawson’s approach as well. The “who” again turns out to be subordinated to an ontology that reduces away its specificity — this time, an ontology of generalized “somethings”. Ricoeur had argued previously that the reflexivity of a self makes it not properly analyzable as a thing at all, because “things” are understood as having the simple idem kind of identity, but selves have the reflexive, ipse kind of identity. He makes the further point that ascription of an action to a self differs from ordinary description, in that it implies an attribution of responsibility.

He notes that for Aristotle, ascriptions of actions have ethical or juridical significance from the start. He also notes that ascription of an action implicitly involves a judgment that the action is within the agent’s power. Then, there are questions of how we assess responsibility for the whole chain of effects of an action, and how we apportion shared responsibility among multiple agents. He concludes that we still have work to do to understand the thinking initiation of actions, and that the framework of simple ascription of actions to selves is still too abstract to do the job.

Pragmatics of Utterance

The second chapter of Ricoeur’s Oneself as Another concerns speech acts in context. Here we begin to consider reflexive acts of self-designation. Like Brandom, Ricoeur emphasizes that saying is a form of doing. “I” is not an identifying reference, but more resembles terms like “here” and “now”. Linguistically, it is a performative associated with affirmation, promising, and similar actions.

Reflexivity applies to the utterance, rather than to the subject of the utterance. It “is not intrinsically bound up with a self in the strong sense of self-consciousness” (p. 47). Meanwhile, it is not utterances but speakers that refer to things, making reference an action rather than a property.

At a later stage, the performative “I” does after all get assimilated to an identifying reference tied to a body, but Ricoeur thinks it will be necessary to step outside the analysis of language and consider our status as incarnated beings to understand this.

Inferentialism vs Mentalism

Brandom’s “inferentialism” or emphasis on material inference effectively makes what I call ethical reason the most important thing in the constitution of subjectivity — not psychology, and not some putative immediate mental presence, or universal transparent representational medium, or supposedly perfect reflexivity.

This is not to deny that there is such a thing as immediacy; it is rather to specify that immediacy is not foundational, and has nothing to do with certainty. Immediacy has a very different role to play, in showing us the world’s “stubborn recalcitrance to mastery and agency” and providing occasions for learning. (See also Mind Without Mentalism; Psyche, Subjectivity.)