Bermudez has written a thought-provoking book on self-consciousness that will stimulate a much more open-minded attitude to the subject than we have enjoyed to date. His combination of sophisticated philosophical analysis and imaginative use of empirical data will be an interdisciplinary model for future work in the area. Bermudez's argument that language-use is not a necessary condition for self-consciousness is appealing, but he does not go far enough when he confines the concept of self-consciousness to knowledge one has of oneself as an object. His book leaves openings for a view of oneself as a quintessential subject.
2. The orthodox philosophical presupposition about self-consciousness is that it is an attribute not available to non-language-using creatures, whether these be human infants or other species. Bermudez calls this the "Thought-language Principle" (p. 12-13). His book is an elaborate attempt to overthrow this principle. He begins by seeking to refute it with a philosophical argument -- the appeal to a vicious circle.
3. As ingenious as this approach is, I want to suggest that the challenge to the thought-language principle does not come primarily from philosophical problems, or an awareness among philosophers of problems in the orthodox reading of self-consciousness, but from the mounting body of scientific evidence that is anomalous from the perspective of the thought-language principle. I believe that in reality it is the scientific dog here that is wagging the philosophical tail, whereas Bermudez's is the opposite point of view. It is notable that most of the scientific evidence was produced in the 1970s and 1980s, and the philosophical response is occurring in the 1990s. I am suggesting in effect that philosophers were forced to respond to this challenge, or become irrelevant. It is admirable that Bermudez has made a bold and far-reaching attempt to meet the challenge head-on.
4. For Bermudez and other philosophers, the scientific evidence begins with the work on perception in the 1970s of J. J. Gibson (1979). Gibson's theory of affordances indicated that an animal's awareness of its environment is a matter of "taking in" the environment for the opportunity it affords for satisfying its needs, whether these be a good place for a lair, a place that may hide a predator, or a spot that would bear its weight. Relying on inference to the best explanation, Bermudez sees in this scenario the requirement to recognise in the animal a primitive form of self-consciousness.
5. In the case of infants lacking language some of the scientific data Bermudez relies upon are the following (p. 248ff).
a. The reaching-behaviour of 15-week-old infants is adjusted to the physical distance separating them from the object (Field (1976).
b. This adjustment may be in place as early as two weeks after birth (Bower 1972.
c. Infants look to where they perceive another individual's gaze to be directed (Scaife and Bruner 1975).
d. The crucial embedding of object-focused attention in social contexts begins at about the age of 6 months when infants begin to switch their gaze back and forth between object and caregiver (Newson and Newson 1975).
e. Neonates are capable of discriminating their own cries from the cries of other infants (Sagi and Hoffman 1976).
f. Infants between 12 and 21 days old can successfully imitate three distinct facial acts -- lip protrusion, mouth opening, and tongue protrusion (Meltzoff and Moore 1977).
g. Nine-month-old infants are capable of recognizing maternal pointing as a cue for directing their attention (Murphy and and Messner 1977).
h. Infants as young as 3-days-old seem to prefer their own mother's voice over the voice of another infant's mother (DeCasper and Fifer 1980).
i. Babies are capable of adapting the speed of their reach to the speed of the moving object (Von Hofsten 1982).
j. Evidence that infants with a mean age of 45 hours revealed a clear preference for the face of their mother rather than the face of a stranger (Field, Woodson, Greenburg, and Cohen 1982).
k. Infants display perspectival sensitivity to the spatial implications of their changing position (Acredolo, Adams, and Goodwyn 1984).
l. Two-week-old infants were more likely to stop crying when they heard the sound of their mother's voice than when they heard the sound of a female stranger (Bremner 1988).
m. "The most important feature of the new behaviour at 9 months is its systematically combining interests of the infant in the physical, privately known reality near him, and his acts of communication addressed to persons. A deliberately sought sharing of experiences about events and things is achieved for the first time" (Trevarthen & Hubley 1978).
6. In the light of this scientific evidence Bermudez concludes:
"Distinguishing self-awareness involves a recognition of oneself as a perceiver, an agent, and a bearer of reactive attitudes against a contrast space of other perceivers, agents, and bearers of reactive attitudes. It can only make sense to speak of the infant's experience of being a performer in the eyes of the other if the infant is aware of himself as an agent and of his mother as a perceiver" (pp. 252-3).
7. In sum, the scientific evidence forces us to push the phenomenon of self-consciousness back into species that lack language abilities and back into pre-linguistic stages of human development. We can only do this by abandoning the Thought-Language Principle.
8. Bermudez takes the approach of developmental psychology to explain the evolution of fully developed self-consciousness from primitive forms. He appeals to what he calls "The Acquisition Constraint" which he describes as follows:
"If a given cognitive capacity is psychologically real, then there must be an explanation of how it is possible for an individual in the normal course of human development to acquire that cognitive capacity" (p. 19).
9. In other words, self-consciousness could not have issued fully formed from the head of Minerva. The evolutionary path Bermudez maps out contains three stages. The lowest stage consists of basic sentience, or "the capacity to feel sensations". The stage above that is "stimulus-response behaviour". The final stage is "intentional behaviour". This stage "emerges when the number of stimulus-response correlations potentially relevant to any given action becomes so great that the need arises to choose between a range of possible responses to a given situation" (p. 192).
10. Now it is one thing to abandon the idea that self-consciousness is a matter of having "I" thoughts (thoughts immune to error through misidentification), but quite another to give positive content to the idea of a self-consciousness that is pre-linguistic. To do this Bermudez draws on the work of the philosopher Christopher Peacocke (1992) and his concept of a mental state with a nonconceptual content. According to Bermudez's understanding:
"The general thought is that it is theoretically legitimate to refer to mental states that represent the world but do not require the bearer of those mental states to possess the concepts required to specify how they present the world as being. These are states with nonconceptual content" (p. 50).
11. Nonconceptual content is representational if it can meet correctness conditions. Thus if an infant represents its mother as looking at the object that the infant wanted her to look at, then the nonconceptual content is representational if indeed the mother is looking at that object for that reason.
"On this model, correctness of a content is then a matter of instantiation: the instantiation by the real world around the perceiver of the spatial type that gives the representational content in question" (Peacocke 1992, p. 62).
Equipped with the notion of a nonconceptual content, Bermudez draws on developmental psychology to arrive at abilities that reveal more and more of the properties of fully developed self-consciousness such as is found in the "I" thoughts of language users.
12. His first step is to broaden the concept of a nonconceptual content into that of a nonconceptual point of view. A creature, we learn, does not have a nonconceptual point of view until it has the ability to distinguish its experience from permanent features of the environment which instantiate a given experience. This ability depends on conscious memory.
"When a creature finds its way back to a particular place that it then consciously remembers . . . it is emerging from a continuous present and moving toward possession of a temporally extended point of view" (Bermudez 1998, p. 177).
But it also depends on "basic inductive generalizations at the nonconceptual level" (p. 184)
13. To get a better grasp of the nonconceptual point of view, Bermudez provides the example of the acquisition of the representational content of an individual's body as a body among other bodies (albeit a special one by virtue of the individual's discovery that he can exercise intentional control over bodily members, and cannot do this over other things in the environment). This discovery must be understood in terms of a representational content of one's own body being a special case. We can follow his line of thinking from what he calls "the simple argument" (p. 135).
a. The self is embodied.
b. Somatic proprioception provides perceptions of bodily properties.
c. Somatic proprioception is a form of self-perception.
d. Therefore, somatic proprioception is a form of self-consciousness.
14. Quite apart from the merits of "the simple argument", I would like to draw attention to an inconsistency in Bermudez's standpoint. The reader will recall that consciousness is approached from the outside -- from the perspective of a scientific observer. And yet the foundational building block of his theory of self-consciousness is the concept of a mental state with a nonconceptual content. But how could Bermudez arrive at the notion of a mental state with a nonconceptual content from a third person point of view? Surely he could only do so from a neo-behavioural standpoint such as that of Dennett (1969, p. 118ff).
15. The mental state of neonate or animal is a hidden variable from the point of view of the scientific observer. But how could such an observer ascribe the attribute of consciousness to the neonate or the animal if he did not draw on his own experience of nonconceptual contents to give meaning to the attribution? It seems to me that Bermudez's foundational concept crosses the line from the third person viewpoint to a second person viewpoint (Symmetry Thesis: and if second person, necessarily first person), and this vitiates his attempt to stick resolutely with the third person standpoint. Bermudez says that his explanation of self-consciousness in neonates and animals must be considered to be an inference to the best explanation. But this "best explanation" is only appealing to a person who understands what self-consciousness is from a second person point of view (us -- mother and me).
16. Fudjack (2000) identifies an ambiguity in the concept of self-consciousness. We can take 'self-consciousness' to mean 'consciousness with a sense of self' as opposed to 'consciousness of self'. It is this distinction that is missing from Bermudez's reflections. Indeed, if we refer to an earlier book edited by Bermudez with Eilan and Marcel (1995, p. 3) we find that the emphasis is on the concept of the self as an object. "Self-consciousness is a matter of representing oneself as an object." This emphasis is continued in Bermudez's new study. The difficulty is that if no recognition is made of a consciousness with a sense of self, to what is the nonconceptual representation of the self being attributed? Clearly there is only one answer available to Burmudez: the nonconceptual content is attributable to a body or an organism. Now what happens, in this case, when the body is not at a given moment experiencing a nonconceptual content of the appropriate kind, such as somatic proprioception? In that case we are left with consciousness sans self. Being a primitive self means enjoying experiences of certain nonconceptual contents. Being a developed self means enjoying "I" thoughts as expressed in canonical linguistic form. In both cases we are left with nothing better than a sophisticated serial theory of the self.
17. If, however, Bermudez had been able to do justice to consciousness with a sense of self, then he could argue that all cases of nonconceptual perceptual content would be able to modify that sense of self and become genuinely revealing of self-consciousness. And the means for coming up with this idea all lie embedded in Bermudez's book as follows:
a. Bermudez adopts Peacocke's idea that there are nonconceptual contents that are not representational in nature (pp. 214-5) [NOTE 1].
b. Bermudez offers the analysis that attention splits the content of consciousness into background and foreground (pp. 138-9) [NOTE 2].
c. Bermudez mentions with acceptance the view that the content of consciousness is an integrated sensory field (p. 141) [NOTE 3].
18. If we combine these propositions we come up with a consciousness that includes a sense of self by containing a nonconceptual nonrepresentational content, which accompanies every "content" singled out by attention as a nonrepresentational background. In this scenario we have a sense of self that is never absent from a consciousness taken up with a content, and when that content happens to give information about the self as object, that information is taken up into the sense of self which adjusts to it. In that sense information about the self as object can at last become genuine self-consciousness. We thus arrive at a form of self-consciousness in which the self of which we gain consciousness is a subject and not an object (see Evans and Fudjack 1976).
19. One interesting consequence of this position is that it puts the self within consciousness in the position of a nonconceptual non-representational content. This would be a content minus correctness conditions. It would therefore be a self that is immune to error through misidentification, which is just what we would predict of a subject of consciousness. In this connection Peacocke's nonconceptual, non-representational content, translates into properties that a self has in virtue of what it is like to be that self (Peacocke 1983, p. 5).
20. Once we have made a place for a sense of self we are in a better position to handle what Bermudez calls the spatial awareness component of a nonconceptual point of view. Bermudez places his main emphasis for reaching the conclusion that infants and animals possess self-consciousness on evidence drawn from perception studies. This is evident from the following, which reveals the fine-grain nature of much of his analysis. The spatial awareness component of primitive self-consciousness requires (p. 220):
a. awareness that one is navigating through the environment, which requires
b. a degree of understanding of the nature of space, which requires
c. a grasp of the distinction between the spatial relations that hold between places and the spatial relations that hold between things, which is manifested in navigational behaviour that
d. satisfies the following two minimal conditions: (i) not being reducible to particular sequences of bodily movements, (ii) not being driven by sensitivity to features of the environment that merely covary with spatial features of the environment;
e. and implicates the following three cognitive capacities: (i) the capacity to think about different routes to the same place, (ii) the capacity to keep track of changes in spatial relations between things caused by one's own movements relative to those things, (iii) the capacity to think about places independently of the objects or features located at those places.
21. With the idea of the sense of self as a nonconceptual nonrepresentational background content to the foreground content of consciousness, we can make of 20a-20e implicit features of the self, and not explicit representational contents as we must in Bermudez's model. We can say that 20a- 20e are components of the implicit context for actions and events that are focally singled out in spatial orientation. 20a- 20e are encapsulated in the self, and the self does not rely upon explicitly summoning up these contents in order for it to successfully "navigate through its world". In part at least, the function of the background self is that of providing the spatial context for such navigation.
 Bermudez acknowledges "what Peacocke calls sensational properties, that is, the nonrepresentational properties that an experience has in virtue of what it is like to have that experience."
 The following is a direct quote from Martin (1995) by Bermudez (1998 p. 138):
"An attention shift is all that is required for the properties of the limb doing the exploring to become the focus of attention, and this does not involve a complete change in the structure and content of awareness. Relevant here is some form of the distinction between focal awareness and peripheral awareness, in particular, the possibility of awareness of something even though one is not attending to it. This idea is most familiar from vision, where it seems clear that items that the boundaries of the visual field are consciously registered, even though they are not being attended to. Peripheral awareness seems to require at the least the following. First, items in peripheral awareness can generally be brought into focal awareness, either by moving the body into an appropriate perceptual relation to an object (by turning one's head, for example) or by focusing on one element in a complicated perceptual experience (as when one tries just to listen to the violins in a symphony), Second, instances of peripheral awareness can have implications for action and reaction without bringing focal awareness to bear, as when one flinches from something seen out of the corner of the eye... Even when the attention is fixed firmly on the proprioceptive dimension of tactile awareness, the exteroceptive dimension remains phenomenologically salient in background awareness."
 Bermudez (p. 141) quotes Ayers (1991, p. 187):
"Our perceptions of the world," says Bermudez, "form what Michael Ayers has described as an integrated sensory field. He captures the phenomenology of the situation as follows." "A judgement which links the objects of different senses may itself be, and very often is, an immediate perceptual judgement, directly grounded on the deliverance of sense. This it is not normally a result of inference, habitual association or the like (although in a few peculiar cases of disorientation some form of inference may be to the point) that I judge the object I feel with my hand to be the object I see. Quite simply I perceive it as the same: the identity enters into the intentional content of sensation, and of my total sensory 'field',"
Acredolo, L. P., Adams, A., and Goodwyn, S. W. (1984) The Role of Self-Produced Movement and Visual tracking in Infant Spatial Orientation. Journal of Experimental Child Psychology 38: 312-327.
Ayers, M. (1991) Locke: Epistemology and Ontology. London: Routledge, 2:187.
Bermudez, J. L. (1999) Precis of "The Paradox of Self-Consciousness". PSYCOLOQUY 10(35) ftp://ftp.princeton.edu/pub/harnad/Psycoloquy/1999.volume.10/ psyc.99.10.035.self-consciousness.1.bermudez http://www.cogsci.soton.ac.uk/cgi/psyc/newpsy?10.035
Bermudez, J. L. (1998) The Paradox of Self-Consciousness. Cambridge MA. MIT Press.
Bermudez, J. L., Marcel, A., Eilan, N. (1995) The Body and the Self. Cambridge MA. MIT Press.
Bower, T.G.R. (1972) Object Perception in Infants. Perception 1: 15-30.
Bremner, J. G.1988. Infancy. Oxford: Basil Blackwell.
DeCasper, A. J., and Fifer, W. (1980) Of Human Bonding: Newborns Prefer Their Mother's Voices. Science 208: 1174-1176
Dennett, D. C. (1969) Content and Consciousness. New York.
Evans, C. O. (1970) The Subject of Consciousness. Humanities Press. http://www.Mental-states.net/SOC.html
Evans, C. O. and Fudjack, J. (1976) Consciousness: An Interdisciplinary Study. http://www.Mental-states.net/consciousness.html
Evans, G. (1982) The Varieties of Reference. Oxford. Clarendon Press.
Field, J. (1976) Relation of Young Infants' Reaching Behaviour to Stimulus Distance and Solidity. Developmental Psychology 12: 444-448
Field, T. M., Woodson, R., Greenburg, R., and Cohen, D. 1982 "Discrimination and Imitation of Facial Expression by Neonates." Science 218: 179-181.
Fudjack, J. (2000) The Subject of Consciousness Revisited. http://www.Mental-states.net
Gibson, J. J. (1979) The Ecological Approach to Visual Perception. Boston.
Martin, M. G. F. (1995) Bodily Awareness: A Sense of Ownership. In Bermudez, Marcel, and Eilan (1995).
Meltzoff, A. N. and Moore, M. K. (1977) Imitation of Facial and Manual Gestures by Human Neonates. Science 198: 75-78.
Murphy, C. M., and Messner, D. J. 1977. "Mothers, Infants, and Pointing: A study of a Gesture." In Shaffer, H. R. (ed). 1977. Studies in Mother-Infant Interaction, Proceedings of the Loch Lomond Symposium, Ross Priory, University of Strathclyde, September 1975, New York: Academic Press.
Newson, J., and Newson, E. (1975) Intersubjectivity and the Transmission of Culture. Bulletin of the British Psychological Society 28: 437-445.
Peacocke, C. (1983) Sense and Content. Oxford, OUP.
Peacocke, C. (1992) A Study of Concepts. Cambridge MA. MIT Press.
Sagi, A., and Hoffman, M. (1976) Empathic Distress in the Newborn." Developmental Psychology 12: 175-176
Scaife, M., and Bruner, J.S. 1975. "The Capacity for Joint Visual Attention in the Infant." Nature 253: 265-266
Trevarthen, C. & Hubley, P. (1978) Secondary Intersubjectivity: Confidence, Confiding and Acts of Meaning in the First Year. In Lock, A. (Ed.) Action, Gesture, and Symbol: The Emergence of Language. Academic Press.
Von Hofsten, C. (1982) Foundations for Perceptual Development," Advances in Infancy Research 2: 241-261.