Loading...
Please wait, while we are loading the content...
Similar Documents
The quantum and the homunculus (session QA1 addition)
| Content Provider | ACM Digital Library |
|---|---|
| Author | Etter, Tom |
| Abstract | Is art, and poetry in particular, like searching for a needle in a haystack, or is it like making hay? To the needle searcher, the haymaker is no friend. If an idea in your head is trying to get out, and if you are carefully feeling your way toward just those words that will express it, you don't want to be hit with a heap of interesting but irrelevant possibilities. Computers have a tendency to do just that, which makes me a bit skeptical about their use in poetry.Actually, my concern here is less with computers as aids or hindrances to the human poet than with computers as poets in their own right. We might forgive them for distracting us from our ideas if they could find beautiful enough words for their own. Presumably they can find beautiful words in the dictionary. But what about their ideas: do these even exist? Does a computer have an inner life? Can it say “I” and mean it? This is the big and scary question lurking in the background whenever we talk about computer art.For some people the immediate answer is “Obviously not.” For others it's “Why not?” I find that most of the philosophical statements on this subject tend to be one or the other of these answers in fancy dress. Neither is very satisfying. The first bears an uncomfortable resemblance to racism and sexism: “A mere machine could never do that!” The second, which is commonly heard in the computer community, is certainly more friendly and polite. It extends the benefit of the doubt to the computer on the basis of appearances, just as we do to a friend who writes us a check. But friendliness and politeness are not always well placed; I remember saying “Excuse me” after stubbing my toe on a rock.What it comes down to is that we don't seem to have any way to gain access to the inner life of a computer, apart from appearances. It's not just that we lack knowledge, it's that we can't imagine how to extend our present knowledge in any way that would help. The word “I” isn't even in the vocabulary of science. Will bigger telescopes or linear accelerators get it there?Crazy at it sounds, they actually may. To put it less cryptically, quantum mechanics, the science of very small things, has forced us to adopt certain ideas that utterly demolish the foundations of the old way of thinking that excludes “I”. More important, it offers subtle but real hints of how to bring “I” (and “you” too!) back into the world. To someone who has not lived with quantum ideas, it's hard to convey just how revolutionary they are. Partly this is because they have been so terribly successful in the old-fashioned way, giving us not only the bomb but the transistor and hence the micro-computer.Physics books tell us that it makes no sense to ask what happens to an electron between the time it is emitted and the time it is detected. This is sometimes made to sound like traditional positivism, which no one can take very seriously. But it's not that at all; it really is serious. The history of the electron not only can't be known, it can't even be imagined without contradictions. And this is not just a philosophical tenet, but a hard logical consequence of well established empirical laws. Between observations the electron drops out of history, out of imaginability, out of furthest ken of a sane mind. Since the world is made of things like electrons, it is made of little bits of insanity.How can many little bits of insanity add up to a world which for the most part seems not only sane but quite orderly? There was a similar question asked, and answered, in the last century, which is how can the random “off the wall” behavior of many atoms add up to predictability? The answer then was called statistical mechanics. Much of the mathematics of that answer applies to the present case, though the underlying fundamental problem is profoundly different, and its solution, which might be called statistical logic, is far from complete. Still, such as it is, statistical logic provides an immense amount of useful knowledge about ordinary material objects. It's real science, the best we have.I will here go out on a limb and saw some things about “quantum madness” that go beyond the usual negatives. I believe that in the quantum domain there is a breakdown of two very basic distinctions, distinctions which seem quite rigid and absolute in the ordinary world. The first is between subject and object, or more precisely, between the perspective of the observer and the state of the thing observed. The second is the ancient and familiar distinction between universal and particular. The quantum domain contains subjects, objects and universals only in potential, and the potentials there are diffuse and interchangeable. How these potentials become realized in the macro-world is a problem in statistical logic, and, I believe, a soluble one.By macro-world here is meant a classical world, a world of pure external facts. This world is of course a figment of our imagination, an ideal, like the geometer's perfect circle. Facts in reality are observed by real observers, whose observations are matters of fact in their own right. Now, however, we are pretending that the observer is a homunculus, a tiny creature of theory who tiptoes onto the world stage so discretely as to give not the slightest sign of his presence, nor leave the slightest trace.This homunculus is what is left in the limit when the observer gets infinitely small. What does that make him? Well, for one thing, you can't describe him by giving information, which, according to physics, always entails energy, of which he has none. Thus you can't say where he is. All you can enumerate is the bare plurality of his possible perspective; to choose among them would be to give information. Also you can't say what he is observing. All you can do is lay out a plurality of ideas or universals, describing how for him these could be associated with matters of fact, but not saying how they actually are, for again, this would entail giving information.Shrinking the observer down to the homunculus extracts from the concrete reality of self the abstract ideal of perspective, and from the concrete reality of experience the abstract idea of the universal. We must do this to obtain the very notion of an objective world, but by doing it we lose the notion of a subjective world. There is close parallel here to the Heisenberg uncertainty principle, which says that if the position of an object is a matter of definite fact, its momentum cannot be a fact at all (though it still makes sense to speak of the range of its possible momentum, which is analogous to the plurality of perspectives or universals).I suggest that this parallel is no accident, and that the uncertainty principle is a projection into the “nerwhol” of the classical complementarity of inner and outer. Quantum mechanics is about the potentialities for a classical world in the nerwhol. Though the potentialities there for external facts may get mixed up with potentialities for thoughts and selves, even the deepest chaos of the nerwhol does not lose the mutual relationships of complementarity among these three dimensions of being! From this seemingly faint hint of order in the wilderness, I believe that much of the actual mathematics of quantum theory can be deduced.The most fundamental problem in quantum mechanics is to understand the genesis of the classical world. But of course real macroscopic things aren't perfectly classical, especially when they are alive. We might say that life occurs when quantum madness is not quite cancelled out by large numbers. This means that perspectives and meanings get mixed up with matters of brute fact even on the scale of ordinary objects. But in a living being these mix-ups are not just chaos, as they are in the nerwhol, since they are tied to the orderly almost classical histories of non-living things. A perspective which is a matter of fact is a self, while a meaning which is a matter of fact is an experience. We get the first hint of how “I” and “my ideas” might enter the language of science, of how bodies might have minds.What does all this have to do with computer poetry? A poem is the expression of our inner life. But what, apart from its expression, is our inner life? H.G. Wells wrote “I suspect that all things unspoken in our souls partake somewhat of the laxity of delirium and dementia.” What I have been suggesting is that this is the same delirium and dementia that characterizes the state of an unmeasured electron. The human head is like a giant electron. Its delirium is of course vastly more complicated than that of the electron, and it extends a huge network of logical tendrils into the sunlight of classical sanity. These tendrils are its ongoing process pf quantum measurements. Toward their classical ends they become computer-like; we can meaningfully describe them as acts of information storage, information gating etc. Within the pattern of these acts one finds the expression of thoughts, and on special occasions, the composition of poems. One can think of a poem as a big complicated measurement of the nerwhol.There's no way to dissect the quantum delirium in the human head. The scalpel and the microtome find only the classical ends of its tendrils. If one didn't know better one might conclude that the brain is a computer. Unfortunately, we may have to wait quite a while for the science that will tell us in detail what the brain really is. But in the meantime we can be pretty sure, based on what we learn from physics, that a computer can't write a poem. |
| Starting Page | 4 |
| Ending Page | 7 |
| Page Count | 4 |
| File Format | |
| ISBN | 0897911709 |
| DOI | 10.1145/324409.324411 |
| Language | English |
| Publisher | Association for Computing Machinery (ACM) |
| Publisher Date | 1985-10-01 |
| Publisher Place | New York |
| Access Restriction | Subscribed |
| Content Type | Text |
| Resource Type | Article |
