Person perception

Abstract

Peter Strawson holds that on a proper conception of personhood, the problem of Other Minds does not arise. I suggest that the viability of his proposal depends on a particular account of person perception. I argue that neither the theory theory nor the simulation theory of mindreading constitutes a suitable basis for this account. I then go on to defend Peter Hobson's notion of ‘feeling perception’ as an intersubjectivist alternative that, if properly developed, delivers a basis for a viable account of person perception. In developing that alternative, I draw on John Campbell's suggestion that self-understanding is constituted, in part, by an understanding of oneself as a cause of many events.

Keywords:

• Other Minds
• person perception
• intersubjectivity
• mindreading
• feelings

Acknowledgements

Thanks are due for helpful comments on various versions of this paper to John Campbell, Robert Frederick, Daniel Hutto, Adina Roskies and Marc Slors, as well as the participants in a conference on Other Minds and Moral Agency at the College of the Holy Cross in April 2007. I am grateful to Bentley College for a summer grant that enabled me to write this paper.

Notes

One reply (pointed out to me by Marc Slors) that may be available to the defender of what Goldman calls a ‘low-level’ version of ST is this. The ‘resonating’ of the emotion that gives rise to the capacity for the ascription of particular mental predicates to others is not conceptually mediated itself; conceptual resources only come into play once one actually ascribes predicates to oneself and others. So the Unmediated Resonance model may be able to meet the Symmetry Constraint after all. I am, in fact, sympathetic to this sort of view; but the question then arises what entitles us to think of it as an instance of ST. I don't want to argue about labels, however; if defenders of a low-level account of ST can, in fact, deflate the worry expressed in this section, such an account may remain a candidate for the approach to person perception I am sketching in this paper.

Again, if you think that some versions of low-level ST are compatible with Hobson's approach, I don't see a problem here. What I wish to argue against is the idea that in order to be able to understand another's psychological life, you have to operate with a grasp of the conceptual role of the first person that your ability to ascribe mental predicates to others depends on.

He stresses that he has not given a complete review of Meltzoff's account. It is worth noting that Meltzoff himself defends a version of TT rather than ST. See Stueber (2006, 118).

This is suggested, firstly, by the factual consideration that there are persons without proprioception who are nevertheless self-conscious. Secondly, such a move is supported by a conceptual reason: an account of common cause thinking that builds on the notion of a body image cannot respond appropriately to the suggestion that one might think of judgements about psychological states along the lines of ‘this body is thinking’.

By ‘simple’ I mean that an experience of pain may be had without the deployment of conceptual resources: I can be in pain without having a conceptual grasp of the source of the pain, while I may not be able to empathize with someone without possessing a range of concepts that allow me to entertain thoughts about her. But this is not to say that pain is monolithic: it may still comprise a sensory and an affective component (see Prinz 2004, 177).

Consider Decety and Chaminade's (2003) proposal to conceive of the neural correlate of feeling sympathy in terms of activity in a ‘shared representation network’. The thought is that sympathy is the result of a matching between the subject's own emotional reaction to the story with that of the story-teller: a story-teller who communicates a sad story with an expression of joy will be found less sympathetic than one who communicates the same story with an expression of the same feeling of sadness experienced by the subject herself. So the experience of sympathy is possible only if one operates with a quite sophisticated conception of self and other. This resonates with a commonsense understanding of sympathy: in order to feel sympathy with another person, I have to be in a position to experience feelings directed at that person. But if this is right, the kinds of feelings that constitute the intersubjective perspective cannot be described in terms of sympathy. A similar point holds with regard to the feeling of empathy. Peter Goldie (2000, 195f.) argues, convincingly I think, that the capacity for empathy also presupposes a sophisticated grasp of the person one is empathizing with, since ‘empathy is a process or procedure by which a person centrally imagines the narrative (the thoughts, feelings, and emotions) of another person’.

This use of the term ‘proprioception’ is unusual, given that it is commonly employed to describe the experience of the relative position of neighbouring parts of one's body. Obviously I am not concerned with spatial bodily positions here. But since I am seeking to describe simple feelings as embodied, the term seems better suited than the alternative notion of ‘introspection’.

Barbara Montero (2006) argues that we can be said to ‘proprioceive’ someone else's movements. This, of course, is a position that runs directly counter to the view I am advocating here. It seems obviously true to me that there is an experiential difference between movements performed by oneself and perceived movements of other persons. And by the same token there is an experiential difference between one's own and other's perceived feelings, even if these feelings can be directly accessed in the perceptual act.

In the rest of this section I am concerned with the question whether the reidentification of simple feelings involves conceptual (rather than representational) capacities. Since I take it, however, that it is a necessary condition of conceptual skills that representational capacities are in place – since I take it that one could not operate with concepts unless one was able to entertain representations of objects, or state of affairs – Hutto's proposal is of relevance for my enquiry: if his strong (anti-representational) claim holds, my parallel weaker (anti-conceptual) claim should hold too.

One worry this proposal might provoke is that the practical knowledge how to enact intentional attitudes is first-person knowledge: it is knowledge that is immediately accessible only to the agent. But this worry can be deflated, I think, by stressing the intersubjective outlook that primitive interactions between infant and caregiver give rise to: it is through such interactions that the practical knowledge at issue is acquired in the first place. And it is through such interactions that a grasp of the conceptual role of the first and third person is made possible. So the idea is that first-person knowledge is not available prior to the kinds of interactions at issue. I am grateful to John Campbell for this point.

I leave open the question whether there is a way of conceiving of representations that are not purely mental in this sense. Pacherie's and Dokic's notion of ‘motor representations’ (Pacherie and Dokic 2006, particularly 151f.) may offer such an alternative. The question arises, however, whether such ‘representations’ can still be said to be truth-apt in the sense discussed.

Additional information

Notes on contributors

Axel Seemann

Email: [email protected]