Dennett's Second Argument

Dennett can be read as offering a second, more intricate argument for MDT over TTC on the basis of color phi. At the heart of this argument is Dennett's ingenious distinction between ``Stalinesque" and ``Orwellian" explanatory hypotheses. In order to introduce the latter type of hypothesis, Dennett gives us an interesting parable:

Suppose I tamper with your brain, inserting in your memory a bogus woman wearing a hat where none was (e.g., at the party on Sunday). If on Monday, when you recall the party, you remember her and can find no internal resources for so much as doubting the veracity of your memory, we would still say that you never did experience her; that is, not at the party on Sunday. Of course your subsequent experience of (bogus) recollection can be as vivid as may be, and on Tuesday we can certainly agree that you have had a vivid conscious experiences of there being a woman in a hat at the party, but the first such experience, we would insist, was on Monday, not Sunday (though it doesn't seem this way to you). ([5], p.115-116, emphasis his)

 
Figure 3: Schacter's Model in ``AIish" Terms

Such post-experiential revisions of memory Dennett aptly calls `Orwellian." Another method for fooling posterity is to ``stage show trials, carefully scripted presentations of false testimony and bogus confessions, complete with simulated evidence" ([5], p.117). Dennett calls such a ploy `Stalinesque,' and gives us another story to show it at work: As you stand on the corner a long-haired woman dashes by. Before you are conscious of that woman, a subterranean memory of an earlier woman with eyeglasses contaminates your experience, and so you hallucinate glasses on the woman dashing by.

 
Figure 4: Overview of

 
Figure 5: Flow-Chart View of

How do the two types of hypotheses relate to ? Our TM-based explanation of is Stalinesque, and is one Dennett himself seems to consider in broad strokes (120); there is no harm in identifying Dennett's Stalinesque explanation of with our E. The Orwellian explanation of runs as follows.

Shortly after the consciousness of the first spot and the second spot (with no illusion of apparent motion at all), a revisionist historian of sorts, in the brain's memory-library receiving station, notices that the unvarnished history in this instance doesn't make enough sense, so he interprets the brute events, red-followed-by-green, by making up a narrative about the intervening passage, complete with mid-course color change, and installs this history in the memory library for all future reference.

Now here is Dennett's second argument:

Arg is formally valid: it conforms perfectly to modus tollens. But premise (2) certainly seems peculiar -- because there are surely differences between Stalinesque and Orwellian explanations. After all, we just got through visiting some of these differences! Dennett's position, on further reading, isn't that there is literally no difference, but rather that (in his words) there is ``a difference that makes no difference." His justification for this attitude, by his own admission (p. 126, [5]), rests on this principle:

(P)

If all the evidence there is or could be fails to distinguish between two competing theories T and T' of phenomenon p, then there is no difference between T and T'.

But of course (P) is verificationism (= operationalism) incarnate -- as Dennett admits (p. 125, [5]. This means that Arg is really an enthymeme to be unpacked as this argument:

It seems to us that, at least in principle, there is evidence that could support one theory and not the other. (If nothing else, could there not be a fact of the matter as to which theory is true, so that if an omniscient creature is in principle possible, it's in turn in principle possible for such a being to enlighten us about the facts?) If so, then (6) is false. However, for the sake of argument (and space) we gladly concede (6). There remains another vulnerable spot in Arg , of course: verificationism. Though Dennett happens to be one, there aren't many verificationists around these days. At the very least, everyone must concede that verificationism is itself in need of independent defense if anything like Arg is to be compelling. Now though Dennett doesn't provide a defense of verificationism simpliciter, he does try to provide a defense of what he calls (p. 132, [5]) first-person operationalism, the view that if the subject herself cannot make judgments concerning competing theories (or explanations) T and T' for p on the basis of what seems to her to be the case, then there is no difference between these two theories. This new principle (call it `(P')') is of course a variant on (P), and can be used in the obvious way to build a variant of Arg (which we can call Arg ). How does this argument fare? Not well, we submit; here's why. Arg has the following premise.

(6')] The subject herself cannot make judgments concerning the competing Stalinesque and Orwellian theories of phenomenon .

And here is Dennett's case for 6':

Ask a subject in the color phi experiment: Do you judge that the red spot moved right and changed color because it seemed to you to do so, or does it seem to you to have moved because that is your judgment? Suppose the subject gives a ``sophisticated" answer:

I know there wasn't actually a moving spot in the world -- it's just apparent motion, after all -- but I also know the spot seemed to move, so in addition to my judgment that the spot seemed to move, there is the event which my judgment is about: the seeming-to-move of the spot. There wasn't any real moving, so there has to have been a real seeming-to-move for my judgment to be about.

But [this] argument is fallacious. Postulating a ``real seeming" in addition to the judging or ``taking" expressed in the subject's report is multiplying entities beyond necessity. Worse, it is multiplying entities beyond possibility: the sort of inner presentation in which real seemings happen is a hopeless metaphysical dodge since those who are inclined to talk this way are eager to insist that this inner presentation does not occur in some mysterious, dualistic sort of space perfused with Cartesian ghost-ether. When you discard Cartesian dualism, you really must discard the show that would have gone on in the Cartesian Theater, and the audience as well, for neither the show nor the audience is to be found in the brain, and the brain is the only real place there is to look for them.

This is an extraordinarily weak bit of ``reasoning." First, the ``sophisticated answer" is precisely the sort of answer we have received from some of our subjects (recall Figure 1 and our ``Web-based" recasting of phi). Second, to multiply entities beyond necessity is not to commit a logical fallacy. (Logic has nothing at all to say about what is or is not necessary in the explanatory sense. Logic alone is unable to provide help to the empirical scientist trying to rationally select between competing theories differing in regard to the size of their ontologies.) Third, to say that the sophisticated answer is to multiply entities beyond possibility is just plain wrong, and clearly so: our earlier marriage of to Schacter's model constitutes, as we noted, a perfectly coherent (though confessedly not a wonderfully detailed) picture of a scheme that can undergird the so-called sophisticated answer. Fourth, and finally, it is downright bizarre to say that this picture (or some related picture) isn't ``found in the brain." What would it be like to find such a thing in the brain? Would we actually find the flow graph of Figure 5 in the shape of neurons and dendrites? The proponent of our computationalist brand of TTC would presumably be in the same boat as (say) the proponent of universal grammar: no proponent of such a grammar expects to read off any of the principles of this grammar directly from a PET scan, or from an inspection of real ``wet" brain matter. Likewise, TTC, if accurate, need not allow flow graphs and modules like `Phenomenal Consciousness' (see Figures 2 & 3) to be found in this manner.

We conclude that explaining phi doesn't require Dennett's exotica. Good ol' fashioned computation, in the context of traditional theories of cognition, does just fine.