In his recent book, Sweet Dreams: Philosophical Obstacles to a Science of Consciousness (MIT press, first paperback edition, 2006), philosopher and cognitive scientist Daniel Dennett intends to undertake “some revision and renewal” of the views on consciousness he famously put forward in his 1991 magnum opus on the mind, Consciousness Explained. “I didn't get it all right the first time, but I didn't get it all wrong either”, he now puts it. As the title indicates, Dennett is now confronting what he regards as philosophical barriers to a science of consciousness, and it is certainly true that a number of the ideas he criticizes really do deserve to come under attack.
One idea is that of zombism, and it is taken on by Dennett in chapter 1. This remarkably popular ignis fatuus is the idea that there resides in us, somehow, a metaphysically supernumerary, causally inefficacious, mystical ingredient that is forever undetectable by science. Is this a ghost or some kind of ectoplasm? No, say the zombists, it is the genuinely undeniable fact that we have a qualitative awareness of reality—that we are able to appreciate the blueness of the blue sky or the distinctive dark, roasted sensation we savor each morning, as we down our Starbucks cup. In contrast to human beings, the zombies are like the unfortunate creatures we find in fictional works like the Night of the Living Dead horror film, going through the motions of human behavior like perennial somnambulists bereft of all qualitative awareness of the world. Unlike the fictional zombies, however, the philosophical zombies are supposed to be utterly indistinguishable from us. In other words, no rotting flesh, glazed eyes or staccato ambulation; they walk, talk, speak, act, and look all the way down to the smallest physical constituent exactly like we do. Yet, they are still supposed to have no qualitative awareness of any kind – and thus, the zombist argument goes, it is clear that science can never give us the full story about human consciousness. This type of argument is a good example of a philosophical ‘sweet dream’ (or, rather, bad nightmare) blocking progress for the study of consciousness. Dennett rightly objects to this idea, and points out, among other things, that we should take pains to be fully clear about the zombie scenario in the philosophical literature and not, in the face of the sheer implausibility of the suggestion, replace it with some generously whitewashed notion. We should also, he argues, realize that the notion of some mystical, scientifically undetectable difference between us and the zombies is simply nonsense.
Another popular philosophical confusion taken on by Dennett is Frank Jackson's famous philosophical vignette about Mary the imaginary color scientist, who has all the physical information there is to obtain about what happens when we see ripe tomatoes, blue skies, red roses and so on – while having lived her entire life in an exclusively black-and-white environment. When the doors to her dichromatic laboratory are finally opened and she beholds the optic splendor of the world, she surely learns something new, Jackson reasoned—and therefore qualitative or phenomenal consciousness eludes the grasp of science, which is again taken to be limited to ‘the physical’. Although Jackson recently has retreated from this position, interest in the argument and its implications is still strong, as witnessed by the 2004 MIT Press publication of the anthology There's Something About Mary: Essays on Phenomenal Consciousness and Frank Jackson's Knowledge Argument, featuring contributions from many leading philosophers of mind including Dennett himself. Dennett's case against Jackson's Mary scenario is developed in chapter 5 of Sweet Dreams, and it features his own philosophical vignette about the curious character RoboMary—an imagined robot whose hardware is equipped for color vision, but who is brought online with only black-and-white video cameras. RoboMary, however, promptly corrects this oversight by writing some code that enables her to colorize the input from her cameras, bringing herself into the same mental state she would have been in, had she experienced colored objects directly. This shows, according to Dennett, that there is nothing special about the direct qualitative experience of color.
One might object to Dennett's argument that he is switching the topic entirely by bringing in a robot, but he replies to this that RoboMary is not fundamentally different from us, since we similarly “are robots of sorts” (p. 125). This response, however, spotlights a profound problem in Sweet Dreams and in the rest of the Dennettian oeuvre—namely Dennett's well-known commitment to a strictly reductionist, materialistic, third-person approach to the mental, and his consequent rejection of all kinds of self-awareness or introspection as viable sources of knowledge in their own right. In a profound sense Dennett himself has thus been lulled into a sweet dream that simply represents the opposite side of the counterfeit zombist coin, where he ends up denying that we have an genuine qualitative awareness of reality – while famously maintaining that we really amount to mere robots, zombies or machines designed by Mother Nature to facilitate the storage and passing on of cultural information (known as ‘memes’). On this view, the people who are doing real work on consciousness are the ones who are “leaving it behind” (p. 144). So, even if we grant that there is a lot of nonsense in contemporary philosophy of mind, Sweet Dreams serves as good evidence of the fact that a proposed pharmacon can be as bad as the illness itself. This becomes fully clear as we turn from the heavily polemical chapters 1, 4 and 5 of the book, to the remaining four more heavily constructive ones.
In chapter 2, for example, Dennett argues in favor of heterophenomenology—a theory that, as we know from his other works, offers an exclusively third-person approach to the study of consciousness as a basic methodology for psychology, cognitive science and so on. As heterophenomenologists we begin our research by adopting a radically agnostic stance towards the actual presence of consciousness in human beings, and we proceed by extracting and purifying texts from our (apparently) speaking subject's to “generate a theorists fiction, the subject's heterophenomenological world”. That theorist's fiction—the fictitious construct that comprises the subject's reported conscious beliefs, experiences, feelings and so on—is then delivered to behavioral- and neuroscientists who will decide whether those alleged phenomena have any real existence, or are mere illusions or fairy tales. If, for example, data from our brain scanner doesn't line up with the reported subjective experience, it should simply be assumed that the subject is wrong about his own experience, and the problem will then merely be to explain how he came to have such a deluded belief.
In Sweet Dreams, Dennett continues to expound this almost neo-Skinnerian methodology, here developed at length from the perspective of Martians arriving on earth to study human consciousness. ‘Agnosticism’, he tells us, ‘is built into all good psychological research with human subjects’, and the lack of radical skepticism about the conscious lives of human beings produces only bad research. This line of reasoning, however, seriously misrepresents our options when it comes to the choice of psychological research methods. As Dennett would have it, one either gullibly believes everything the subject says, or one approaches him as a heterophenomenological item whose conscious life one radically mistrusts—to the point of being noncommittal about whether he might be a zombie, a robot, or even a dressed-up parrot. Clearly, however, there is a panoply of epistemic options here, and the falsehood of unrestrained gullibility does not imply the truth of radical agnosticism. As an analogy, think of forensic practice. Here, it would make no sense to treat testimony by witnesses as comparable either to fairy tales or, alternatively, to infallible declarations from the Almighty. Surely, the picture is much more subtle, and in forensic epistemology it is a well-know fact that certain kinds of witness testimony are highly reliable, others are moderately reliable, and others still amount to little more than outright fabrications.
No less troubling than Dennett's continued defense of heterophenomenology is his apparent continuing allegiance to classical functionalism. This theory of the mental was famously introduced in the mid-20th century by Turing, Putnam and others, who were all taken in by the seeming boundless prospects of the brilliant technological innovation of the modern computer. Their essential ideas were (1) that the mental is simply the execution of complex computational algorithms by an appropriately powerful hardware device and (2) that machines will be able to think, just like us, if only we could get a fast enough processor, a large enough hard disc, and a good enough piece of software. The latter idea quickly morphed into the resoundingly failed strong AI project, that marketed itself by making strikingly irresponsible claims—such as MIT AI-lab guru Marvin Minsky's famous boast, quoted by Searle in 1992, that we will be lucky if the next generation of computers will “keep us around the house as household pets”—but that still, despite 50-odd years of work and nearly endless amounts of funding, has yet to present us with an independently intelligent being equal to a retarded chimpanzee. Dennett, however, doesn't see it that way. To him functionalism was a “tempting proposal” that made a “minimalist empirical wager” according to which “neuroanatomy doesn't matter; chemistry doesn't matter.” These didn't matter to the classical functionalists, of course, because all that matters is the computational algorithm – whether that one is executed by a brain, a computer, or a Rube Goldberg device. Refusing to denounce classical functionalism and strong AI for ignoring the brain (and, indeed, also the mind) for several decades, Dennett merely observes in a laconic tone that ‘[m]any of the fond hopes of [the] opportunistic minimalists have been dashed … and they have learned that no, if you leave out x, or y, or z, you can't explain how the mind works.’ (P. 19). This, one would think, represents a serious scientific blunder on the part of the functionalists. But no, according to Dennett, it was in fact “a good idea in many ways, but … it has not been a politically astute ideology.” (P. 19). Since the functionalists were, in fact, fundamentally misguided in both their methodology and conclusions, this statement is as puzzling as a claim to the effect that pursuing astronomy without ever looking at the sky in many ways was a good idea, even if the project did turn out to be lacking in political astuteness.
The troubles for Dennett continue as we turn to his new Fantasy Echo theory of consciousness, outlined in chapters 7 and 8. Developed from his previous Multiple Drafts theory of consciousness, Fantasy Echo holds that consciousness is “a matter of content-bearing events in the brain achieving something a bit like fame in competition with other fame-seeking … events.” (p. 160). The problem that must be tackled by a theory of consciousness, then, is to explain the functional power of attention-grabbing–that is to say, how “relatively few contents” become elevated to such “political power” without a single attention-giving source. (p 161-2). The solution, in turn, must be sought in the human propensity to “relive or rekindle contentful events” – i.e. in the human ‘habits of self-stimulation’ that lead, via echo amplification, to the position of ‘fame’ for some events in the brain.
There are many objections to this theory, but an obvious one is that it relies heavily on the distinctively Dennettian approach, where mental content and intentionality is assumed to be primary and consciousness is constructed out of those notions as a subsequent and dependent phenomenon. But don't we, rather, start with a basic notion of consciousness, and then get to the idea of mental content—by realizing (1) that many different conscious mental processes (e.g. seeing the apple, thinking of the apple, desiring to eat the apple) can have the same content and (2) that the same conscious mental process can have varying contents (thinking of the apple or the sun or blueberry muffins)? Only by differentiating such varying aspects of consciousness, then, can we get to the idea of mental content—in contrast to, say, mental processes. But to then turn around and declare that consciousness itself is simply reducible to this secondary aspect seems to be a serious mistake.
Analogously, imagine someone who argues that biological organisms are unreal; what is truly real, this person maintains, is disease. Allow me the notion of disease as a primitive, he would go on, and I can provide you with a wonderfully reductive theory of biological organisms. Organisms, after all, are nothing more than a location for disease and potential disease. Of course, the answer would be to deny this person the axiomatic use of ‘disease’ and to insist that one first must have the notion of an organism and then, as a metaphysical and epistemological dependent, the notion of a pathological and abnormal condition of that organism. Related to this problem, it is discouraging to learn that Dennett still is using the Intentional Stance technique, where essential distinctions are systematically blurred by the ‘instrumental’ bunching together of everything from macromolecules, thermostats, lightening, computers, robots and, yes, human beings, into the category ‘rational agents’, unsurprisingly leading to the conclusion that we are really no different from the robots we construct.
Although Sweet Dreams does critically confront some dubious philosophical notions such as zombies, ghostly qualia and omniscient color scientists locked up in black-and-white rooms, the work does so on the basis of Dennett's basic commitment to reductive materialism along with his frequent appeals to the heterophenomenological methodology, the inversion of epistemic hierarchy, and a systematic blurring of essential distinctions—and thus must itself be deemed more of a sweet dream than a long-overdue wake-up call to the fledgling science of consciousness.
