© C.O. Evans
 In The Bounds of Sense Strawson makes some remarks which seem to invite the theory I am advocating, and, in conjunction with these remarks, he makes others which seem to preclude the possibility of any theory such as mine succeeding. For this reason it is worth seeing how my theory stands in the light of what he says. Strawson maintains in that work that we can ascribe certain experiences to ourselves (presumably ones which we ascribe to ourselves not on the basis of observation) without invoking any criteria of personal identity. As he says,
When a man (a subject of experience) ascribes a current or directly remembered state of consciousness to himself, no use whatever of any criteria of personal identity is required to justify his use of the pronoun "I" to refer to the subject of that experience. When "I" is thus used, without any need or any possibility of its use being justified by empirical criteria of subject-identity, it does not, however, lose its role of referring to a subject. 152
Now in terms of the distinction between the persons-approach and the self-approach, Strawson's reference to 'the subject of experience'
is equivocal. On the one hand the subject of experience can be understood to be referring to what we take the self to be - i.e. a person for some theorists, and a body for others - or it can be understood to be referring to what we experience ourselves as being. Since Strawson's is exclusively the persons-approach we are, of course, not in any doubt about what he means by 'the subject of experience'. But the point is that this closes his mind to one of the possibilities suggested by his description. And that is that the 'I' can be used for completely non-ascriptive purposes - criterionless or otherwise. It is only when the speaker intends to communicate to another person that it makes sense to say that he is ascribing an experience to himself (criterionless self-ascription), and it is only when the intention to communicate to another is the reason for an avowal that the persons-approach interpretation of the speaker's words is the right one. But we also engage in self-address with the intent of bringing home to ourselves a certain experience we are having. We do this often because attempting to express ourselves by describing the experience is a way of focusing attention on the experience the better to appreciate it. We are then confined to the self-approach, and in that case, I maintain, mention of the subject of the experience has nothing to do with ascribing an experience to an owner, and hence nothing to do with criterionless self-ascription. In such cases the 'I' is used symbolically and not referentially. It is used as an 'I'.
The consequences of a failure to demarcate these two logically different types of situation manifests itself in what Strawson says in the following passage:
It is easy to become intensely aware of the immediate character, of the purely inner basis, of such self-ascription while both retaining the sense of ascription to a subject and forgetting that immediate reports of experience have this character of ascriptions to a subject only because of the links I have mentioned with ordinary criteria of personal identity. Thus there arises a certain illusion: the illusion of a purely inner and yet subject-referring use for "I". If we try to abstract this use, to shake off the connection with ordinary criteria of personal identity, to arrive at a kind of subject-reference which is wholly and adequately based on nothing but inner experience, what we really do is simply to deprive our use of "I" of any referential force whatever. It will simply express, as Kant would say, "consciousness in general". If we nevertheless
It is noteworthy that Strawson acknowledges a 'purely inner' use of 'I', but because of his unquestioned assumption that 'I' can only be used in a referring sense, he concludes that any such use of 'I' must be illusory. We have seen that he is right to deny any such inner directed use of 'I' as a referring expression but this does not rule out the possibility of the use by philosophers of the word 'I' to symbolize the subject standing over against an experience to which he is attending. Strawson's position rests on his rejection without argument of the Wittgensteinian claim that the word 'I' has two entirely different senses - one of which is a referring sense and the other of which is not. It is this premiss of his that leads him to conclude that any attempt to identify the subject of inner experience must lead to 'a pure, individual, immaterial substance'; i.e. a Pure Ego. In terms of my theory we can say that Strawson was right to reject the idea of a Pure Ego, but wrong in thinking that no alternative could be arrived at on the basis of inner experience. The identification of the self with unprojected consciousness is a clear alternative, and it gives us a self which is experiential and the experiential content of which is rich in diversity and variety. It is the antithesis of Strawson's 'bare form of reference'.
We may speculate that Strawson's conviction that awareness of the purely inner basis of self-ascription leads inevitably to the Pure Ego can be traced back to the assumption that a study of consciousness in order to determine its subject can have as its outcome no other result than adoption of a Pure Ego Theory (thesis) or a Serial Theory (antithesis). Until now a third alternative has not been sufficiently explicitly stated to demand consideration. For those who enjoy dialectical language, I am offering a reconciliation between thesis and antithesis in a higher synthesis. We may view Strawson's argument as a challenge to produce a theory of the self based on inner experience which is not a Pure Ego Theory. The theory I offer is my answer to that challenge.
In my introduction I remarked that we have native knowledge of the self and that, therefore, the identity of ourselves as subjects of
states of consciousness could only be determined by our looking to our own cases. This is the course I have tried to follow, and the theory I have been putting forward in this chapter is the outcome. I mentioned MacNabb's remark that we have an experience of the self 'more internal than the most personal emotion we feel'. I now suggest that this is the sort of remark we would make were the self indeed identical with unprojected consciousness. Unprojected consciousness has the experiential character and the interiority which MacNabb's description portrays the self as having. The logical features of unprojected consciousness also make it understandable that the Augustinian formula should seem so applicable to the experience we each have of ourselves: if no one asks me what I am, I know; if I am asked, I know not.