Sunday, February 17, 2008

The Bob-fest


LI has been so pressed this week that our ambition – to advance a little along the line of the theme of the social animal – has been totally fucked. Fortunately, we’ve spent the last two days attending lectures at the Bob-fest – the conference commemorating Robert Solomon. The academic custom of reading papers at conferences, as anybody knows who has actually gone to a conference and had papers read to him or her, is not exactly the most exciting activity in the world. It ranks somewhere around TV coverage of the Tour de France – long stretches of time go by without anything seeming to happen, and then everybody gets briefly excited, and then ennui stretches out again. This conference was a bit more personal, a bit more eccentric, and instead of the standard way of asking questions, those with questions and comments would go down, at the end of the panel, to the panel table and sit around and speak. I did not hear all sessions, but of the four I heard, the highlight was surely the paper read by my friend, Janet McCracken, entitled " Grief and the Mnemonics of Places: A Thank You Note", which ranged over the funeral games in the Iliad to the place of the dog at Zoroastrian funeral rituals to connect a number of seemingly disparate topics – the dead as companions, the need to mark the earth for the dead, the relation of the human and the animal (and, especially, that subclass of domesticated animal, pets) from the perspective of death. I’m merely flying over the paper giving its grosser features, but McCracken’s paper was just the kind of Shandian essay I wrote about last week, re Ian Hacking’s paper on cyborgs.

Afterwards, I was in a group around McCracken, talking about animals, and of course animal intelligence, and we all sort of marveled that animal intelligence is always measured by seeing how close animals can come to human intelligence. That is surely wrong. An ant or dog or whale born with a human brain would quickly die. Intelligence, if it has any meaning at all, is connected to animal existence, and so one would want to produce a number of intelligence models fitting different animals. As a counterpart on tests of, say, the human intelligence achievable by parrots, it would be interesting to see how close humans can approach to parrot intelligence. Of course, it is pretty difficult to probe into parrot intelligence, but surely we can devise models for parrot seeing, hearing and flight such that they could be fed into a virtual environment, in which we could insert a human subject.

So, LI was not wholly negligent of the social animal theme we so abruptly aborted.

Later, thinking of animals, humans and intelligence, it struck me that one of the reasons I have problems with the machine model of humans and animals is that, in my experience, those who are best with machines – engineers, for instance – are really piss poor at human relationships. My old man, for instance, became pretty well known in the HVAC field. He had a certain diagnostic genius for what was wrong with, say, a 13,500 BTU roof unit that had been underperforming. He could track a ductwork problem down to its malign root. And yet, he couldn’t read the face of a man who was obviously trying to con him out of ten thousand dollars – a situation that my old man faced, alas, all too frequently, as he seemed to be on some list of gulls passed around in the confidence game. As for women, well, forget it. This is, of course, an absurdly limited sample – but I have met a lot of engineers in my life, and a lot of people who make their money repairing machines. It seems to me that generally an inverse relationship held between psychological insightfulness and mechanical ability. Although I wouldn’t say this is a hard and fast rule – my brothers, for instance, are both engineer/repair men, and both of them are fairly interested in what makes human beings tick (which already distinguishes them from most IBM engineers I’ve met). One has adopted an attitude of therapeutic nihilism – since it is impossible to rely on rule based behavior, here, it is foolish to judge – while the other is more inclined to a sympathy theory, and is very good at empathizing and advising – as I well know.

Obviously, this is touch and go evidence, a set of ordinary superstitions. But it does perhaps point to a nuance in the relationship between biology and positivism in the 19th century – given a certain notion of social engineering that went with utilitarianism and was carried into the scientific ideology of the late nineteenth century, biology was often the science from whence sprang the dissenters. Scientific psychology, in order to gain credibility, often veered closer to engineering than to biology. And this made a lot of difference.

11 comments:

Dominic said...

Biology - "that wild empiricism masquerading as a science" (Badiou).

It's certainly true that machines designed by people are very unlike the person-machines that designed them. Ray Kurzweil's notion that once we are able to design machines with greater than human intelligence, the machines will just go on designing smarter and smarter machines, doesn't to my mind take account of the reason for this.

An intelligence designed by an intelligence will tend to be a projection and amplification of some aspect of the designing intelligence; this isn't ultimately a fact about intelligence, but a fact about design. Animal intelligence (including human animals) isn't designed, and isn't a projection (or purposeful rationalisation) of this kind. Hence the current notion that maybe we can grow some kind of AI in some kind of virtual vat...

Roger Gathmann said...

Dominic, IT's post from a while ago about Ranciere dawdled a bit around the notion that we all have an equal intelligence - which is a phrase that no teacher can quite swallow.

But to my mind, intelligence is not a quality that is reducible to roles - whether the roles are those of teacher or student, engineer or humanist - but is instead about existence. Intelligence fills out the question: what is it like to be a ... - bat, human being, android, cyborg. In a way, an intelligent vacuum cleaner is one that works, an unintelligent one is a vacuum cleaner that is broken. The vacuum cleaner is already smarter than I am at vacuuming up dirt - I'll never compete, there. I'm smarter than a vacuum cleaner because I find dirt offensive - I live in a richer world of symbols. Or so think I.

However, the more I pursue this 'smartness' idea, the more it looks like the traditional chain of being which was decisively snipped by Darwin in the 19th century, and has never recovered its scientific status.

I do think it is sort of funny that those who design and engineer these smart machines are just those who - as anybody who has been around engineers knows - are rather proud of their clumsiness at social human interaction. How many times have I heard it said, by engineers, that engineers generally grade out as introverts on the personality tests they take (and, weirdly, believe in!). It is part of the folklore of engineering.

Dominic said...

I wonder how Ranciere would take to the suggestion that human intelligence fills out the question of what it is like to be a human being. The question would be equally askable of one person as of another: what is it like to be you? (With the automaton, golem or slave-born Muscovite presumably lacking even the rudiments of an answer. What is it like to be Commander Data? But Data qualifies as "sentient" precisely because this is in fact a question for him; because in addition to processing information about his environment and reacting accordingly he also seeks, in experience, the means to qualify his own being).

However, I'm not convinced that it is like anything to be a bat. To be a bat is to be a bat; there isn't anything that isn't being a bat that it's like, as such, except insofar as bats themselves are like other things. I suspect that what the questioner really means is: if I were to have echolocation added to my sensory awareness of the world, what would that be like for me? What in terms of my other human experiences would that experience resemble? Which is a decidedly human, and un-bat-like, thing to be asking.

Engineering and computery types spend a lot of time in the company of entities that are very unlike human beings - one wouldn't generally even think to ask the question "what is it like to be a webserver?" - and have a strong tendency to be the sorts of people who find that sort of company agreeable.

This has less to do with a reduced capacity for emotion and empathy than people think; engineers have plenty of empathy; just not with people. They have traditionally been hard to recruit to leftist causes, because the sentiment of social solidarity does not come easily to them. If one could portray the dictatorship of the proletariat as some sort of inexorable Rise of the Machines, a lot of them would be all for it.

Roger Gathmann said...

I take "like" there to function as the filling out of the being of the bat - since the bat never is all it is in any one instant.

Actually, IT's project, which is to revive the collective subject, contours a problem, here, that resembles Sartre's problem of joining a certain, seemingly radical individualism inscribed in Being and Nothingness with the collective subjects described in the Critique of Dialectical Reason. This is the continuum that shuttles us from the animal to the student - for indeed, if there is any sense in the notion of a "role", it has to be derived from a thing that at least blends many roles. In mechanical terms, it contains many affordances. And it is just here that the engineering model starts spluttering, for there is one thing above all others that engineers design for - control. And the emergence of affordances in use are exactly those things that aren't designed for. I think of these two ways of thinking as non-evolutionary and evolutionary. It is no surprise that when you look at, say, I.D. institutes who claim that "plenty of scientists" support Intelligent design, the striking thing is how many of them are engineers.

But anyway, getting back to the what is it like to be question, I don't think the referent is outside the species essence of bat, if it is properly posed. But the level at which it is a question about a particular x bat, human intelligence comes abruptly to a halt. The limits of the world are the limits of my Gattungswesen, as Doc Holiday would always say, before plugging another desperado.

Dominic said...

And it is just here that the engineering model starts spluttering, for there is one thing above all others that engineers design for - control. And the emergence of affordances in use are exactly those things that aren't designed for.

Dennett calls these emergent-affordances-in-use exaptations; hackers call them hacks.

I like the idea that the likeness of a bat is a moving image of bat-being. The moving image of a machine is reducible to a static image, a phase space or state diagram, since the machine's movements are ideally regular. Except that machines in their actual worldly embodiment are not quite so regular; hackers hack, users spill coffee cups, stray neutrinos flip bits when no-one's looking...

Roger Gathmann said...

Of course, one of the ironies is that the property of silicon that made it ideal for chips was just such an accidental discovery.

"In early 1955 Bell Labs researchers encountered a major problem with pitting on the surface of silicon wafers during high-temperature diffusion. This problem was overcome by chemist Carl Frosch during a serendipitous accident in which the hydrogen gas carrying impurities through the diffusion furnace briefly caught fire, introducing water vapor into the chamber. The resulting “wet-ambient” diffusion method had covered the silicon surface with a layer of glassy silicon-dioxide (SiO2)."

Roger Gathmann said...

ps - thinking about the Badiou quote, I am once again puzzled. Taken at face value, I think that quote is nonsense. It seems to derive from a point of view in which all the sciences will eventually reduce to physics - a program that has never, ever worked. It hasn't even worked for chemistry. You can find the old DINO philosophers still trying to do it, like Kim, but every time they set up their logical structures, they seem comically encoded to come up with a set of answers that simply ignores the wrecks on the reductionist path while reproducing the arguments that led to them.

I might be wildly wrong about Badiou - I'm no expert - but surely the one thing biology is not is wild empiricism.

I get rather irritated about that. Philsophers of consciousness often set up thought experiments in which they summon up, ex nihilo, various conscious creatures, zombie and the like, thereby ignoring the whole lesson of evolution tout court. Aristotelian biology ain't ever comin' back, but it certainly lives in philosophy departments.

Dominic said...

I think the complaint isn't so much that biology's destiny is to be subsumed into physics, but that biology lacks axiomatic anchor-points of its own (and hence is always at risk of hostile takeover bids from physics or some other more axiomatised discipline). The modern evolutionary theory discussed by Dennett and Dawkins certainly partakes copiously of the mathème, however - but doesn't all that computer modelling make real, experiment-driven biologists a bit uncomfortable?

Badiou's anti-vitalism also comes into play here, perhaps: what is this bios anyway? For Badiou, the one thing it isn't is a concept, and every attempt to install it as a concept, to yoke bios to logos, leads to confusion, muddle, disaster...

Anonymous said...

"Just because I don't care doesn't mean I don't understand" - Homer Simpson. You can't be sure all these people lack empathy just because they lack sympathy.

Roger Gathmann said...

Mr. Lawrence, this was, approximately, what my Mom used to say about the Old Man. I might have sounded more critical than I meant to be - simply observing a social custom.

Dominic - I believe that the entire field of biology, however uncomfortable they might feel with some parts of the neo-Darwinian synthesis, have pretty much gotten with the mathematics since the days of Haldane and Sewell Wright. Once you take it that living things must have dna, and that they reproduce, you are on your way. No need for vitalism. On the other hand, the idea that vitalism is the inevitable and only opposite to the reductionist program has made many a bio magus, like Ernst Mayr, hit his head against a wall out of sheer frustration.

But I'm more interested in the being-like relationship, the system of referents of an organism within which it lives, than the rather sterile fights about vitalism.

Simbaud said...

"Kathryn Stewart, director of the Orion Academy, a high school for high-functioning kids in Moraga, California, calls Asperger's syndrome 'the engineers' disorder.'"

From the Holodomor to Gaza: NYT softfocuses on famine - the spirit of Walter Duranty lives!

  When Gareth Jones, a former secretary of David Lloyd George, made a walking tour in Ukrainian agricultural districts in 1933, he wrote a s...