I concur with Puccetti's criticism and argue that Dennett's multiple drafts theory cannot help us to understand any kinds of dissociations of consciousness. This is because the theory does not allow the distinction between real and apparent streams of consciousness. Such a distinction can be made on empirical grounds, however, with the help of empirically based accounts of consciousness.
2. For Dennett, to find out what the mental content of a person is is to make an interpretation of the person's external behavior from the intentional stance (Dennett 1987), that is, from the third-person point of view. Now, the basic common denominator in Dennett's theory of intentionality and his theory of consciousness is this: first, we can find neither consciousness nor beliefs at the level of internal brain processes; second, we cannot establish either the real content of a person's beliefs or the definitive content of the subject's actual stream of consciousness (because there are no such things); third, consciousness, like beliefs, cannot have an existence independent of observers and interpretations made from the outside (Revonsuo, in press b).
3. The dissociation between explicit and implicit access to information in brain-injured patients (for a review, see Schacter et al. 1988) provides us with empirical grounds for deciding which of the multiple drafts have to do with the subject's actual stream of consciousness, and which can be considered to be cases of mere "as-if consciousness." In a typical case of dissociation between explicit and implicit processing, the patient is selectively impaired in some cognitive domain, for example, recognizing faces (prosopagnosia). If requested to explicitly use this impaired function (e.g., a prosopagnosic patient is asked to recognize faces of relatives or celebrities), the patient utterly fails. Nevertheless, there is indubitable evidence from reaction time and interference tasks, skin conductance measurements, forced choice tests, and event-related brain potentials that, at some level, patients do have implicit access to the information they seem to be completely unaware of. Thus, although the patient's overt performance in recognizing familiar faces is at a chance level, we may have simultaneous evidence from behavioral and electrophysiological measures that the familiar faces do have effects different from the unfamiliar ones in the patient's brain. Consequently, they were, in a way, "recognized." Bruyer (1991) summarizes the findings concerning prosopagnosia and implicit knowledge:
"[W]e now have empirical data showing that some prosopagnosic subjects manifest simultaneously covert behavioral signs of recognition and overt behavioral signs of nonrecognition. What is (un)recognized can be the face familiarity, semantic properties of the person seen, or its name. These data suggest that the `conscious subject' does not recognize or identify familiar faces, while his/her information processing system does" (Bruyer 1991, 230).
4. According to Dennett, there is no deeper fact to be found to establish which one of these mutually contradictory drafts is the description of the real content of subjects' consciousness (or the real content of their beliefs). The patient just manifests one contentful trace with the content "p" ("this is a familiar face") and another with the content "not-p" ("this face is not familiar"). Neither of these traces is the "canonical version." If you look at such behavior from the third-person point of view, there simply is no determinate content of the patient's mind. There is no definitive way to translate the person's behavior to a single, coherent "heterophenomenological" account. For Dennett, the contents of mind are exactly as indeterminate as are the various external interpretations of observable behavior (cf. Dennett 1988, p. 66). Dennett's lesson is that in all cases of dissociations of consciousness, it is wrong to designate one of the contentful traces as speaking with the voice of the actual "conscious subject" (cf. Bruyer 1991, above). Thus, if we follow Dennett's theory, the split-brain patient and the prosopagnosic patient with implicit processing are just exceptionally clear cases of the multiple drafts of consciousness in action.
5. However, explicit information plays roles in the behavior of the organism entirely different from the roles of implicit information. The latter can only influence a very limited range of behavioral responses, it tends to be isolated from the global behavior of the organism, it amounts only to simple and fixed input-output relationships, and it has no survival value whatsoever for the organism. Lahav (1993), in comparing implicit and explicit information, concludes that explicit, conscious information constitutes a central, functional "junction of information," in which information from different sources and modalities is integrated to produce a unified and coherent body of behaviors. Now, different empirically based models and theories of consciousness (e.g., Baars 1988; Schacter 1990) contain converging theoretical conceptions of the neurocognitive architecture and function of consciousness. These conceptions accord well with the above characterization of explicit knowledge as global and flexible (Revonsuo, in press a). This means that we do have empirically based conceptions of consciousness with the help of which we can decide which of the traces or drafts to regard as revealing the actual content of consciousness and which to regard as only reflecting the functioning of nonconscious systems. The various behavioral and physiological outputs of the residual implicit systems can, of course, be interpreted as carrying mental content. Unfortunately, the "mental content" ascribed to these multiple drafts is only in the eye of the beholder.
6. In conclusion, the problem with Dennett's theory is the denial of the reality of subjective phenomenology (Revonsuo 1993). Consequently, the multiple drafts theory cannot help us to differentiate real cases of conscious information processing from apparent ones, because it does not admit such a distinction. Nevertheless, in cases of dissociations of consciousness we are strongly inclined to make such a distinction. And here empirically based conceptions of consciousness can be useful, even at this early stage of their development. According to some of them, explicit knowledge is not simply another stream of consciousness among other essentially similar implicit "drafts"; rather, it is qualitatively and functionally distinctive. Where we have global, flexible and integrated behavior, we have good grounds for believing that it is based on explicit knowledge and subjective awareness of the stimuli. If all we have are isolated, inflexible input-output relationships which do not contribute to the survival of the organism, we have every reason to believe that subjective awareness of the stimuli will not occur, no matter how the high-level semantic properties of the stimuli (e.g., face identity) are implicitly processed.
7. To return to the case of split-brain consciousness, there clearly is a difference between the "drafts" of implicit processing and those of disconnected right-hemisphere processing. The left hemisphere of split-brain patients is as oblivious to the processing in the right hemisphere as are patients with blindsight or prosopagnosia to the implicit processing of stimuli going on in their brains. In that sense, the right hemisphere processing is "implicit" from the perspective of the dominant hemisphere. Nonetheless, it is quite evident that both hemispheres can independently show flexible, nonautomatic cognitive abilities far beyond those of typical implicit processing. Although our current empirical theories of consciousness by no means allow us to define and count such things as centers of consciousness in split-brain patients, the hypothesis of double consciousness that Puccetti advocates is not inconsistent with the available empirical theories. In fact, it might turn out to be correct.
Baars, B.J. (1988) A Cognitive Theory of Consciousness. New York: Cambridge University Press.
Bruyer, R. (1991) Covert Face Recognition in Prosopagnosia: A Review. Brain & Cognition 15, 223- 235.
Dennett, D.C. (1987) The Intentional Stance. Cambridge MA: MIT Press.
Dennett, D.C. (1988) Quining Qualia. In A.J. Marcel & E. Bisiach. (eds.) Consciousness in Contemporary Science. New York: Oxford University Press, 42-77.
Dennett, D.C. (1991) Consciousness Explained. Boston: Little, Brown.
Lahav, R. (1993) What Neuropsychology Tells Us about Consciousness. Philosophy of Science 60, 67-85.
Puccetti, R. (1993) Dennett on the Split-Brain. PSYCOLOQUY 4(52) split-brain.1.
Revonsuo, A. (1993) Is There a Ghost in the Cognitive Machinery? Philosophical Psychology 6 (4): 387-405.
Revonsuo, A. (in press a) In Search of the Science of Consciousness. In A. Revonsuo & M. Kamppinen, (eds.) Consciousness in Philosophy and Cognitive Neuroscience. Hillsdale, NJ: Lawrence Erlbaum.
Revonsuo, A. (in press b) The Multiple Drafts Model and the Ontology of Consciousness. Behavioral and Brain Sciences.
Schacter, D.L. (1990) Toward a Cognitive Neuropsychology of Awareness: Implicit Knowledge and Anosognosia. Journal of Clinical and Experimental Neuropsychology 12:155-178.
Schacter, D.L., McAndrews, M.P. & Moscovitch, M. (1988) Access to Consciousness: Dissociations between Implicit and Explicit Knowledge in Neuropsychological Syndromes. In Weiskrantz L (ed.): Thought without Language, Oxford: Oxford University Press: 242-278.
This study was supported by the Academy of Finland.