Tim Kenyon
Department of Philosophy
University of Waterloo
Waterloo Ontario
Canada N2K 3M6

tkenyon@watarts.uwaterloo.ca

Copyright (c) Tim Kenyon 2000

PSYCHE 6(13), October 2000
http://psyche.cs.monash.edu.au/v6/psyche-6-13-kenyon.html

KEYWORDS: self-consciousness, nonconceptual content, self-awareness, concepts.

REVIEW OF: Jose Luis Bermudez (1998). The Paradox of Self-Consciousness. Cambridge, Mass: MIT Press. ISBN 0-262-02441-1. xv + 338pp. US#18.00 pbk.

The paradox that Jose Luis Bermudez is concerned to defuse comprises the following propositions, each of which is either widely held or defended by prominent writers, and which are inconsistent as a set.

1. The only way to analyze self-consciousness is by analyzing the capacity to think 'I'-thoughts.
2. The only way to analyze the capacity to think a particular range of thoughts is by analyzing the capacity for the canonical linguistic expression of those thoughts (the Thought-Language principle).
3. 'I'-thoughts are canonically expressed by means of the first-person pronoun.
4. Linguistic mastery of the first-person pronoun requires the capacity to think 'I'-thoughts.
5. A non-circular analysis of self-consciousness is possible.
6. The capacity to think 'I'-thoughts meets the Acquisition Constraint (i.e., there must be an explanation of how humans normally acquire this capacity).

Bermudez's book is a lucid and relatively accessible discussion of these propositions (especially 1, 2, and 4), their implications, and their interrelations.

The problem may not be immediately apparent upon examining this set of statements. But its essence is this: jointly, these propositions set such demanding criteria for the correct ascription of the capacity to think 'I'-thoughts that there is no way to explain how an agent not already possessing this capacity could come to develop it. This could have been set up as the difficulty, rather than including Proposition 5 and making the issue one of overall inconsistency. And indeed, not all of the other five propositions contributing to the problem are analyzed in depth. Propositions 3 and 6 are almost accepted at face value, presumably on the two reasonable assumptions that there are only so many hours in the day, and that the audience at whom the book is targetted accept these statements as plausible.

Accordingly, Bermudez focuses mainly upon the notion of nonconceptual content. If nonconceptual content is both real and significant in human cognition, then Proposition 4 is open to doubt, if not falsified outright, and so we have independent grounds to reject it - resolving the inconsistency and the paradox. Much of the book is devoted to discharging the antecedent of this claim; Bermudez marshalls empirical evidence from a range of sources in order to provide independent grounds for identifying Proposition 4 as the problem. He does not reject Proposition 4 tout court, but portrays it as depending upon a "classical" conception of content with two components: a "Conceptual- Requirement Principle" and a "Priority Principle". The former holds that ascriptions of content to an agent are constrained by the agent's conceptual repertoire, while the latter holds that an agent must possess language in order to have genuine concepts. Bermudez does not dispute the Conceptual-Requirement Principle, but characterizes the Priority Principle as untenable in the face of evidence supporting the notion of nonconceptual content.

That evidence is culled primarily from three sources: (i) dishabituation experiments on pre-linguistic infants, which identify nascent abilities to distinguish self from (m)other or pick up on basic purposiveness; (ii) primitive self-conceptions arising from proprioception and visual bodily monitoring; and (iii) proto-concepts of selfhood underlying spatial awareness and navigation. Indeed, throughout the book Bermudez synthesizes results from various psychologists and philosophers, explaining and extending the ideas with confidence. For example, the work of John Campbell on spatial awareness, and that of Christopher Peacocke on concepts and content, is deeply influential on Bermudez's view, and his outline of the relevant aspects of their views is very clear and direct. (Especially in comparison with some of the original presentations of these ideas.) Similarly, in Chapter Five Bermudez discusses J.J. Gibson's theory of ecological optics, canvassing it for elements supporting a primitive non-linguistic concept of self that will enable him to break out of the problematic circularity identified above. This discussion could easily serve as part of a general introduction to Gibson's theory and its continued relevance to psychology and philosophy.

The book's main argument is strong, though there are doubts to be raised. The argument hinges on a collection of empirical results in psychology, so the strength of those empirical results is a matter of importance. Some of this evidence, as mentioned, relies upon release-from-habituation results used in testing infants, a method employed with some caution due to its congeniality to over- interpretation. In fact, those finding the Priority Thesis plausible will incline toward this general challenge to Bermudez's empirical evidence, namely, that the linguistic richness of our descriptions of "self-conceptual" behaviour in non-linguistic animals and prelinguistic humans inevitably overdescribes the internal capacities implicated in the behaviour. This is a concern to which Bermudez is sensitive, warning that '[t]he fact that the infant regulates his behavior in conformity with his mother's responses does not license the conclusion that he is thinking about either himself or his mother in any particular way' (p. 253). The empirical arguments for nonconceptual content are all balance-of-evidence in character, leaving room to disagree reasonably by offering a psychologically deflationary account of the relevant phenomena.

Bermudez also considers studies in the phylogeny of language in making the case for nonconceptual content (pp. 76-9). But the scant evidence that this area yields is shakier still, and does not find the author on his most familiar ground. That Bermudez motivates the move from developmental to phylogenetic considerations by citing the chestnut "Ontogeny recapitulates phylogeny" is one indication of this short section's rather marginal status.

This much is just to say that the main argument is likely to engender debate. Bermudez's discussion is not, however, completely free of more substantial difficulties: one notices in particular its relatively narrow focus. For a book on self-consciousness, this one has very little to say about our sense of ourselves as selves. Another concern arises from the missing figures who have played large roles in the recent history of this and related issues. Simon Baron-Cohen's work on autism is naturally suited for discussion in Chapter Nine (on the relation between awareness of self and awareness of others) and was notable by its absence. Donald Davidson's influential paper 'Thought and Talk' seems relevant, but goes unmentioned, as do the views of Peter Carruthers, whose work was recently the focus of much commentary in this journal. William Lycan, Daniel Dennett, Owen Flanagan, David Rosenthal -- the problem seems either more serious or more geographical as one considers a list of prominent recent theorists on consciousness whose work presumably has some bearing on the question of self-consciousness, and who do not get even as far as Bermudez's bibliography.

Why expect that these theorists would be relevant? A very natural approach, if one wants to explain self-consciousness, is to ask what consciousness itself is, and then see if an answer to the more specific question falls out of one's account. Or, if we want to know how a subject could come to have an 'I'-thought, we might intuitively begin by noting some features of what it is to have a thought simpliciter, and then ask what more specific orientation of that capacity would generate an 'I'-thought. Neither approach is prima facie committed to a notion of non-conceptual content. At a finer grain, Bermudez seems to endorse this line of reasoning: he argues that an explanation of deliberate self-reference ought to be based on Grice's intention-theoretic account of communication, 'on the entirely reasonable assumption that an account of the intention to refer linguistically to oneself can be derived only from an account of communicative intent in general' (p. 277). That no similar approach to self-consciousness in general is mentioned places the work somewhere on the borderline between excessively narrow in scope and forgivably programmatic. The book could hardly have been so tightly argued had its scope been much broader.

Jose Bermudez has assembled some powerful and fascinating evidence against the thesis that language is prior to, or constitutive of, thought. He has furthermore presented the case in favour of nonconceptual content in what must be its canonical form to date. This is an important book for anyone interested in the notion of nonconceptual content or the developmental psychology of the self. I expect that future discussions of these issues will have to place themselves relative to this one.