Counting Consciousnesses

for Brain and Behavioral Sciences, CONTINUING COMMENTARY on "Time and the Observer: the 'Where' and 'When' of Consciousness in the Brain", vol. 17, no. 1, March 1994, pp. 178-80

Counting Consciousnesses: None, One, Two, or None of the Above?

Daniel C. Dennett and Marcel Kinsbourne
Center for Cognitive Studies
Tufts University

In a minute there is time

For decisions and revisions which a minute will reverse.
--T. S. Eliot, "The Love Song of J. Alfred Prufrock"

In a second there is also time enough, we might add. In his dichotomizing fervor, Bogen fails to realize that our argument is neutral with respect to the number of consciousnesses that inhabit the normal or the split-brain skull. Should there be two, for instance, we would point out that within the neural network that subserves each, no privileged locus should be postulated. (Midline location is not the issue--it was only a minor issue for Descartes, in fact.)

As one of us (Kinsbourne, 1982) has pointed out, it follows from the nonexistence of a privileged locus that the limit on the number of consciousnesses that could theoretically be housed in the brain (given suitable disconnections) is the minimal complexity of the neuronal substrate that suffices for this kind of functioning. There

could be many, and certainly more than two. Not all these consciousnesses may be to Bogen's liking. Given a lateral (coronal) transection, the posterior sector may be precluded from controlling behavior, while the anterior one be sorely lacking in information to guide spared action. The separated left or right hemisphere, in contrast, is more fully equipped with input and output possibilities, depleted though they are.

Bogen claims that our argument "puts excessive emphasis on introspection." To the contrary, it goes further to discredit introspection than Bogen himself can countenance. What on earth is the Multiple Drafts Model but a denial of the singularity of consciousness? In our view, Bogen's duality of consciousness is just as simplistic as Descartes's singularity. Our references to (apparently) unified normal awareness, to which Bogen takes such exception, are in the service of the very position that pleases him: the apparent unity is not a necessary reflection of any unity in the neuronal substrate. So Bogen's arrow misses its mark; or perhaps he was just using the BBS format to propagandize his own, unrelated hobbyhorse.

Johnsen's point concerns a confusion about the stream (or streams) of consciousness. We speak of a "parallel stream of conflicting and revised contents" and he asks: parallel to what? Here is what we meant: The apparently single and unified "stream" is in fact composed of many different, largely independent, contemporaneously evolving content elements. These occasionally conflict with each other, occasionally mutually support each other. Coalitions of such elements take turns dominating their alternatives until they all fade away. That temporary ascendancy is our substitute for the more traditional idea of an entrance into a privileged sphere or theater as the "mark" of consciousness or awareness.

Some commentators have wanted to read our substitute as a variation on, or version of, the traditional idea, and since the crucial difference is hard to keep in focus, we turn to an analogy which may help anchor the discussion--but only if the points of comparison are carefully marked.

You go to the racetrack and watch three horses, Able, Baker and Charlie, gallop around the track. At pole 97 Able leads by a neck; at pole 98 Baker, at pole 99 Charlie, but then Able takes the lead again, and then Baker and Charlie run ahead neck and neck for awhile, and then, eventually all the horses slow down to a walk and are led off to the stable. You recount all this to a friend, who asks "Who won the race?" and you say, "Well, since there was no finish line, there's no telling. It wasn't a real race, you see, with a finish line. First one horse led and then another, and eventually they all stopped running." The event you witnessed was not a real race, but it was a real event--not some mere illusion or figment of your imagination. Just what kind of an event to call it is perhaps not clear, but whatever it was, it was as real as real can be.

Notice, first, that verificationism has nothing to do with this case. You have simply pointed out to your friend that since there was no finish line, there is no fact of the matter about who "won the race" because there was no race. Your friend has simply attempted to apply an inappropriate concept to the phenomenon in question. That's just a straightforward logical point. You certainly don't have to be a verificationist to agree with it.

Notice that each horse's career can be precisely tracked, including the spatio-temporal intervals during which it led (if it ever did). The same must be true, surely, for events occurring in the brain. At different times and places different contentful processes may dominate ("be in the lead"), but no such time or place is privileged (the "finish line"). What counts in the analogy as being conscious? Simply running well--contributing to the "dominant focus of neuronal activity" (Kinsbourne, 1988) for some (unspecified) period of time. No doubt the property of being in the lead is a property which has precise temporal boundaries in the case of the horses, and its analogues in the brain may be presumed to be just as determinable (e.g., some property of relative neuronal dominance) but such domination does not confer some extra property of awareness (so that moving into the lead is becoming conscious, and ceding the lead is lapsing into memory or unconsciousness). The succession of dominance is what gives the stream of consciousness its seriality (such as it is), but it is a feature within the stream of consciousness, a sufficient but not necessary condition of being a conscious content.

Johnsen claims that a sentence of ours makes sense "only if read as saying both that there is an 'order in which we experience events to occur,' i.e., a single subjective sequence, and that it differs from the order in which we experience the events." This was not our intended reading (if it even makes sense). We said what we meant: the standard presumption breaks down--for quite mundane reasons. Of course we can specify times before consciousness of an item has begun and after which consciousness of that item has ceased, but it is in the nature of the phenomena that this timing principle does not apply at all scales. Similarly, the standard presumption that political events can be put into a unique time sequence breaks down when we choose our events carefully. Which came first: Clinton's victory or the closing of the polls? It is only those who have a "finish line" model of consciousness who cannot tolerate leaving such questions unasked and unanswered.

Revonsuo summarizes our replies to earlier commentators and says they are not very clarifying. He correctly analyzes Dennett's position in earlier work, and sees that our joint view is consonant with it. As he says, sub-personal cognitive psychology "evades" consciousness, but this only means: don't look for a consciousness module, any more than you should look for an honesty module or a shame module.

He then asks some questions, to which we here supply the answers:

"Is consciousness, like belief, an observer-relative 'calculation-bound entity' or 'logical construct'?" No, but heterophenomenological objects are. (Feenomanism [Dennett, 1978, 1991] is a perfectly real phenomenon--a religion--but Feenoman is not real.)

"Why do microtakings have to have their effects on guiding action before they can reach the status of consciousness?" It is not that microtakings must first have their effects on guiding actions and then acquire some additional property of consciousness; their having these effects is constitutive of their being conscious takings.

"Why is there 'no crisp way of telling exactly which parts of the multiple parallel streams are conscious'? Not, as he surmises, because of observer-relativity, except in the minimal sense that it is the observers' concept(s) of consciousness that break down (as noted above) at this point.

Revonsuo perpetuates one large (but common, and forgivable) misreading of Dennett's position on observer-relativity and reality. Beliefs, according to Dennett, are quite real even if no one ever attributes them to their subject, and they are as discoverable-in-principle as genes, to use Revonsuo's example. Notice, by the way, that this comparison is particularly apt. According to current thinking, there don't turn out to be any Mendelian genes--Mendel didn't quite carve nature at the joints. So we face a terminological choicepoint: do we say there never were any genes or that genes turn out to be rather different from what their "discoverer" claimed they were? In fact there has been some heated disagreement among biologists, but the general trend certainly seems to be to keep the term "gene" and abandon Mendel's definition. But in a free country (and science is a free country) this lexical decision could go either way.

To those critics who claim that what we have not so much provided a model of consciousness as a denial of its very existence, we can reply, in a similar spirit, that we take consciousness to be rather different from what they think it is, but those who hate to see consciousness robbed of some of its "defining" properties can keep their "essences" if they insist--we'll simply have to declare then that consciousness, so defined, does not exist. Something that is rather like that consciousness--enough like it to be called consciousness by the lexically lax!--does exist. That's a realistic variety of realism.


Dennett, D. C., 1978, Brainstorms: Philosophical Essays on Mind and Psychology, Cambridge, MA: MIT Press.

Dennett, D. C., 1991, Consciousness Explained, Boston: Little, Brown.

Kinsbourne, M., 1982, "Hemispheric specialization and the growth of human understanding," American Psychologist, 37, 411-20.

Kinsbourne, M. 1988, "Integrated field theory of consciousness," in A. J. Marcel and E. Bisiach, eds., Consciousness in Contemporary Science, Oxford: Oxford Univ. Press., pp.239-56.