“I’s” Identity Crisis

In this post I will explain Strawson’s argument for how we cannot understand how we could come to ascribe psychological states (pstates) to dualist entities. Subsequently, I will consider an objection to Strawson’s view and defend him by showing just how primitive the concept of a person is.

A dualist entity is a physical body and non-physical (no-)subject or mind composite. If you are a dualist entity, then the “you” who are interpreting these letters is (ontologically) distinct from the “you” whose hand is on the coffee thermos. Physical predicates are ascribed to the body; p-states are ascribed to the non-physical subject relevantly “attached” to that body. Consequently, there are two uses of the word “I” (98): there is the “I” that is in pain or is elated; then there is the “I” that is 5’11’’ and would benefit from going to the gym. Strawson addresses two dualist views: the No-Ownership view and the Cartesian view, in order to show that we couldn’t make sense of the ascription of p-states to these non-physical subjects.

The No-Ownership view holds that it is an illusion that we ascribe p-states to others, for there is no “owner” of subject experience at all (95). There is some unique causal relation between a body and experience, but that experience may be causally dependent on some other body. For we can imagine a case where my visual experience is contingent on the openness of body A’s eyelids and the orientation of body B’s eyeballs (88). The theorist thinks there are two senses in which one’s experience might be said to belong to a thing, (1) there is the loose sense in which some experiences may belong to a body, then (2) there is the sense in which some experiences belong to a Cartesian ego (96). There is a temptation to say that it is a contingent fact that all my experiences are had by (or are causally dependent on) a particular body and that it appears a necessary truth that all my experiences belong to an associated Cartesian ego (96), meaning that the ego must be the owner of experience. But the sense of “belonging” in (2) is an illusion, for the only sense of “belonging” that the No-Ownership view permits is the “logically transferrable” kind, where things like experience are only contingently owned. So experiences owned in the sense of (1) are logically transferrable, for the relation between experience and that body is contingent. But the Cartesian ego solely functions so as to own experiences in a necessarily logically non-transferrable way (96), for recall it seemed a necessary truth that the experiences belonged to the Cartesian ego. But no experience can be owned in this logically non-transferrable sense, so it cannot be the case that “you” (as in the “you” who feels a pain) own your experiences; your experiences are only “owned” qua they are causally connected to a body. So on this view, we cannot ascribe p-states to anyone (subject) at all.

Strawson claims that this view is incoherent because it necessarily makes use of the notion of possession that the view wishes to deny (96). When the No-Ownership theorist states the contingent fact “all my experiences are had by a particular body”, he makes essential use of the word “my” – a possessive word that establishes a necessary connection between experience and the subject denoted by “my”, and this is the notion of possession that the theorist argues against. He could not revise it as, “all experiences are causally dependent on a particular body”, as this would simply be false, for there are clearly other subjects of experience. So it seems the theorist means to say, “all experiences had by a certain person are contingently, causally dependent on a particular body” – but this rephrase is not equivalent to, “all my experiences are had by a particular body” (97). To say, “all experiences had by a certain person” are dependent on a body, is to talk about those experiences necessarily of a certain entity (in a logically non-transferable sense) – and we saw that the No-Ownership theorist cannot accept this (97). So the view must be incoherent; we are not forced to conclude that we do not ascribe conscious states at all.  Moreover, Strawson can make sense of the logical non-transferability of ownership (i.e. the necessity of “all experiences had by a certain person”). Consider the fact that we can only identify and ascribe p-states insofar as they are manifested in the behaviors of an identifiable person (97). As there are no identifiable Cartesian egos floating around to which I ascribe p-states, it takes the identification of a person’s behavioral expression to ascribe a p-state to anything. Because p-states can only be identified in this way, the identity of p-states depends on an identified person whose states they are (97). As this is the way we must ascribe p-states, if p-states are ascribable at all, then they they must be logically non-transferrable (98). This is because I must identify a person if I am to identify a p-state at all; this identity requires that the ownership must be logically non-transferrable (98). So if we can refer to particular p-states at all, then the No-Ownership theorist cannot maintain his position, for his logical transferability is violated. Consequently, the theorist must say that we cannot refer to particular p-states at all, and this seems absurd (98).

The Cartesian view holds that it is a linguistic illusion that both physical and psychological predicates apply to one and the same entity. On this view, reference to a person is reference to one or both of two distinct substances (mind and body), which share no properties or states (94). So the method of verification of my own mental state will be different from the method of verification I use to determine your mental state; the former is verified through introspection and the latter is verified via behavioral signs. We identify a subject of experience by referring to “the subject which stands to that body in the same special relation as I stand to this one” (101).

Here’s Strawson’s worry for the Cartesian view. The Cartesian identifies a mind as “the mind that stands to that body in the same special relation as I stand to this one” (101). However, this way of identification makes essential use of the word “I”. It is a necessary condition of consciousness that if we can ascribe p-states to ourselves, then we have the capacity to ascribe them to others (101). That is, I can only think of myself as in some p-state if I can think of some other as being in a p-state, for to think that “I am in pain” requires me to be able to think that “X is in pain”. “I” cannot be used because I have not established if there are my experiences at all (101), as I have no way of seeing that there are other minds. Because I cannot see other minds, I cannot think of other subjects; if I cannot think of other, I can have no notion of self (if I cannot think X is in pain, I cannot think I am in pain). Consequently, the Cartesian can only maintain that “all experience stands in relation to a particular body” – but not “my experience” (101). And as this is all the Cartesian can maintain, he has no resources to ascribe p-states to others (and so to himself), for there is no identifiable self or other.
Because we cannot make sense of how we ascribe p-states to dualist beings, Strawson thinks we must take the concept of a person as logically primitive (meaning that minds are to be analyzed in terms of persons, not vice versa): the kind of entity such that both physical states and p-states are ascribable to it.

But there is reason to be skeptical of Strawson’s claim that in order to develop the concept of self one must be exposed to other subjects of consciousness. Consider the following. In 2050, humans create the first androids – they are functionally and aesthetically equivalent to humans but lack conscious experience and thus genuine mental states. By 2100, the androids have eradicated all humans on earth. By 2150, they’ve outgrown their aggressive tendencies and now are pursuing advanced scientific projects, including the cloning of a human from genetic material recently recovered at an archaeological site. The clone (which because it is biologically human has mental states) develops until early infancy and then is given to a couple of androids to raise. As the clone is raised, her interactions with the androids are sufficiently similar to those she’d have with humans, such that she comes to believe that they have p-states. She sees androids wince when they stub their toes, say (on the basis of functional behavior), “that must have hurt”. Growing up in this environment, the clone would come to ascribe psychological predicates to the androids [1] (even though they have no p-states) and so develop a self/other concept, and as such can ascribe p-states to herself. So it would seem that you can develop the idea of “my” without ever being exposed to other subjects of consciousness. This means that the Cartesian is entitled to use “I” when he refers to “the subject which stands to that body in the same special relation as I stand to this one”, and so can ascribe p-states to both himself and other subjects.

Here is how Strawson might respond to this worry. Suppose he grants the intuition that the clone acquires the self/other concept. The only way that she can manage this is if she never recognizes the androids as androids at all – that is, if she never recognizes androids as entities that do not have p-states. So in order to develop her self/other concept, she must mistake the androids for persons – that is, she must mistake them to be subjects of experience. For consider if she didn’t. Then she could not come to ascribe mental states at all, for she has recognized no other and consequently no self, and so has no tools to think of anyone as a subject of experience – this was the original problem with the Cartesian view. She can only acquire the self/other concept if she mistakes androids as persons – as the kind of thing to which both physical and psychological predicates apply. Even though the clone acquires the self/other concept in the absence of other subjects of experience, this still requires her to have the concept of a person, so the concept of a person is still going to be logically primitive. That she mistakenly acquires the concept of a person doesn’t matter, for a real person who unwittingly finds himself in a virtual world with virtual trees will still acquire a non-virtual concept of tree from the experience. So even if you can acquire the self/other concept in the absence of other subjects, this still requires you to have the prior concept of a person, and so the Cartesian cannot avoid Strawson’s conclusion in this way.


[1] Perhaps if there were still humans, they would be able to correct the clone and allow her to see the androids as androids.  However, because there are not, she has no sense of reference by which she can recognize that the androids do not have subjective experience.  So as far as she knows, she can ascribe to them psychological predicates.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s