I Am A Strange Loop [Englisch] [Taschenbuch]

Douglas Hofstadter is an exceptionally bright and witty man, with a gift for analogy. This no doubt makes him entertaining company and a pleasure to have as a teacher, but at the same time it sometimes gets in the way of the message he's trying to convey- the allegories and metaphors become the dominant message, and the core gets lost in translation.

This is of course exactly what happened with Hofstadter's 1979 tour-de-force "Godel, Escher and Bach"; it was roundly praised to the heavens by scores of reviewers, none of whom seemed to notice that it was in fact a very clever way of presenting a theory of conciousness and intelligence. This bothered Hofstadter as well, as he tells us in the introduction to "I Am a Strange Loop", and so he set out to tell the story again, this time in a more straightforward manner. I'm not so sure he succeeded.

The bulk of "I Am a Strange Loop" is devoted to explaining Godel's Incompleteness Theorem, with a minimum of math and a lot of allegory and allusion. Much of it seems repetitious, and all of it is, I think, wasted, as the end product of all this attmepted explanation seems to be simply one more metaphor- that what's going on in the brain/mind is something very much like what's going on in Godel's theory: That a theory, or a formula, or a sentance, or a "thing," can contain within it a complete representation of itself. Hofstadter calls this a "strange loop", and believes that, combined with input from outside that adds to this (and other) loops is the wellspring from which consciousness springs.

I first heard this notion expressed in the following manner (although I don't recall who wrote it): Every living thing has in it some representition of the outside world. A plant has in some sense a representation of the sun, that allows it to bend towards it. A bacterium moving along a gradient of nutrients contains within it a representation of this source of nutrition. A bee has representations of hive, flower, sun, and other concepts that guide it goal-seeking behavcior. And so on, up the evolutionary line. When that representation become complete and complex enough to include itself, that is the birth of consciousness. This is not a particularly original notion, although when Hofstadter wrote GEB back in the 70s it wasn't a particularly widely held idea in psychology. At the same time, it was't a completely alien idea, either.

In the last few chapters Hofstadter toys with some more or less current ideas in the philosophy of mind, like Chalmer's "zombie", and presents us with a few more allegories and clever tales, none of which, I think, end up clarifying this position terribly well. One is left with the feeling that Hofstadter has a very strong intuitive sense of how conciousness evolves from these strange loops of self-representation, and what's he's struggling to do is to let us share his intuitions. I find that I share some of these intuitions with him, particualrly with his notions of where the self is represented, and representations of others alongside the self, and I think there's a germ of some powerful explanation hiding in there, but I can't seem to provide any more illumination than can Hofstadter.

And that in turn reminds me of something I was told at the beginning of my teaching career: If you can't explain something clearly and simply to another person, then you don't fully undertsand it. I think that's where Hofstadter is with respect to consciousness: He has a lot of intutions and parallels he can pull out, but in the end, he doesn't really have a theory of conciousness.

You have certainly enjoyed the sensation of looking into a mirror that itself reflected a mirror, making a tunnel of reflections that went as deep as you could see. The same sort of thing happens when you take a television camera and turn it onto a monitor that is showing what the television camera is taking a picture of. But there is something spooky about such a loop. In fact, when young Doug Hofstadter's family was looking to purchase its first video camera, Hofstadter (showing in youth the sort of interest in self-reference that he would turn into a writing career) wondered what would happen if he showed the camera a monitor that itself showed the camera's own output. He remembers with some shame that he was hesitant to close the loop, as if he were crossing into forbidden territory. So he asked the salesman for permission to do so. "No, no, _no_!" came the reply from the salesman, who obviously shared the same fears, "Don't do _that_ - you'll break the camera." And young Hofstadter, unsure of himself, refrained from the experiment. Afterwards he thought about it on the drive home, and could see no danger to the system, and of course he tried it when they got home. And he tried it again many times; video feedback is one of the themes in Hofstadter's monumental and delightful _Gödel, Escher, Bach_ (known by millions as GEB) from 1979, and it comes back for further discussion (with more advanced hardware) in Hofstadter's new _I Am a Strange Loop_ (Basic Books). As in his other books, Hofstadter has written a deeply personal work, even though he is taking on the eternal philosophical bogey of consciousness, and has written once again with a smoothness and a sense of fun that will entrance even casual readers with no particular interest in philosophy or consciousness or mathematics into deep and rewarding thought.

Hofstadter's theme here is consciousness, or "I" or (and he shuns religious connections to the word) the soul. Humans have consciousness. Dogs seem to have some ability to understand what other dogs (and humans) are feeling, some way of representing themselves and others within their own brains. Goldfish, well that's pretty iffy. Mosquitoes have no capacity for self-knowledge. And go further down that scale. How about the neuron itself? Is there any consciousness there? After all, mosquito neurons aren't really much different from human ones, they are just more numerous and tangled in humans. Further down: DNA molecules - conscious or not? Further: carbon atoms - wait a minute, there's not even the possibility that an inanimate atom could have consciousness. Thus the great paradox, looked at repeatedly from different viewpoints here: inanimate matter, properly organized, yields consciousness. We take it all for granted, but it is all profoundly puzzling. Every human brain working at the symbol level (but very much dependent on neural and chemical foundations) "perceives its very own 'I' as a pusher and a mover, never entertaining for a moment the idea that its star player might merely be a useful shorthand standing for a myriad infinitesimal entities and the invisible chemical transactions taking place among them." The "I" is an illusion, an effective one that has great survival value for its possessors. This could be dense stuff, but Hofstadter's analogies are brilliant, as are many of his puns; he reminds us, "Just as we need our eyes in order to _see_, we need our "I"'s in order to _be_!" Hofstadter is fun to read.

Hofstadter's last book, _Le Ton beau de Marot_, was a long meditation on language and translation, and contained many reflections about his wife Carol, who sadly and suddenly died of a brain tumor in 1993 before she was 43. Carol reappears many times in the current work; it is clear that she and Hofstadter had an unusually deep and affectionate marriage, "one individual with two bodies". He is able to write movingly of what he has learned from the loss, how Carol's mind, her "Carolness" or "Carol-consciousness" has been incorporated into his own "I". He isn't Carol, and carries only an imperfect copy of Carol's soul within his own soul, but he shows how her strange loop has been incorporated into his, and just how strong and loving such an incorporation must be. It is a deeply humanistic vision of empathy, the sort of generous personal insight that shows that though souls might be merely the product of atoms and neurons interacting, might be merely illusory, they can still be grand and fully empathetic. Hofstadter has written another book to increase our wonder over the workings of our wonderworking brains.

In the "strange loop," Douglas Hofstadter has come up with a pretty fertile metaphor. The problem is that the book doesn't do a whole lot to explain it. If you can "dig" or "grok" or "intuit" that consciousness is a strange loop, then you won't need the long portions of this book that attempt to promote this thesis. If you cannot so grok, then reading those same portions will be confusing and unhelpful.

This is not Hofstadter's fault. Trying to understand consciousness in this way is like "the art of seeing one's own eye" - it pushes up at the limits of language and reason. Good writing can only get you so far.

There are other portions that are quite enjoyable and these are the ones that are less thesis-driven and more literary. Hofstadter's youthful attempt at his own Socratic dialogue is fun and -although he apologizes at length for its immaturity- actually pretty good. I could have read a book-length chat between his "Plato" and "Socrates" (who seem -anachronistically- to be aware of computers and fruit-canning machines).

But even these bits could have done with a bit more editorial direction. The main problem with this book is Hofstadter's isolation within the closed-universe of the academic philosophy of mind. He clearly attaches an undue importance to this vanishingly small world. Hofstadter's snipes at John Searle are embarrassingly frank in their personal bitterness. I have never thought Searle was worth taking very seriously, but Hofstadter has little sense of humor about him or his work.

The same problem colors Hofstadter's frequent digressions into ethics, since his ethical positions seem to stem more directly from the cultural values of the academy than from his own ideas. He makes clear that he is pro-animal rights and pro-choice, since animals have consciousness and fetuses do not. He proposes that there exists a hierarchy of consciousness, with "small-souled" beings (e.g. fetuses, vegetables, retarded human beings) at or near the bottom and "large-souled" beings (e.g. adult humans) at the top. I happen to agree with him here, but Hofstadter's ethical discussion would be greatly enlivened by a familiarity with mysticism and religion, especially Buddhism and Aristotle. Instead, his horizons seem limited to journal-page arguments with Dennett, Churchland and Searle (ethical geniuses none).

Specifically, Hofstadter conflates "degree of consciousness" with "relative right to exist." If he fails to recognize that this is a nonsequitur, he does at least acknowledge that it commits him to a troublesome implication: that 2-year old humans have less right to exist than adult humans. He deals with this as follows:

"Even though I sincerely believe there is much more of a soul in a twenty-year-old than in a two-year-old (a view that will no doubt dismay many readers), I nonetheless have enormous respect for the potential of the two-year-old to develop a much larger soul over the course of a dozen or so years."

It is entirely obvious that fetuses (not to mention spermatazoa) have this same potential, so to the extent that "potential" is the reason for Hofstadter's pro-choice views, his argument is unsatisfactory.

He gives another: cuteness. Perhaps, he suggests, the sensation of cuteness reflects a protective instinct in humans. But this is clearly a positive fact and not a normative one - Badtz Maru is cute, but this alone does not give him rights. Indeed, it seems to me that the OPPOSITE is often true. Babies are precious by virtue of their limited awareness, not in spite of it. (Which is a greater tragedy: the death of child or a highly realized being?) Viewed with a cold eye, the "strange loop" theory of consciousness simply has no necessary ethical implications. Like Dennett, Hofstadter is a terrific thinker but a hamfisted ethicist - an unreflective mouthpiece for the ideology of the academy.

By far, the most useful contribution here is Hofstadter's specific discussion of video feedback as a metaphor for consciousness. Again, you either "grok" this or you don't - there is just no explaining something this weird. In the most novel and interesting portion of the book, he uses Marvin Minksy's term "telepresence" to explore the notion that consciousness is not singular, discrete or correlated with a spatial location or any single "body." He suggests, I think rightly, that mind exists wherever there is sufficient feedback of information, and that it spills over from one feedback loop into another, without respect for bodies, matter, or location. However, in my view, the same prejudices that prevent Hofstadter from confronting the ethical implications of his views also commit him to a reductive, ateleological worldview (again, he is here in lockstep with Dennett here), and this forces him over and over again to explain mind as some sort of emergent (and therefore anomalous) property of information itself. It also forces him to spill gallons of ink unnecessarily in an (unsuccessful) attempt to salvage free will. Finally, it keeps him from exploring the teleological (and much more parsimonious) alternative: that information exists in order to facilitate the emergence of mind.

At the end of the day, this is a self-indulgent book-length footnote to Hofstadter's masterpiece, GEB. Rather than pick at these scraps, the reader should take the opportunity to read or re-read that work.