Analogies all the way down?

There's a fuzzy story, nicknamed "Turtles all the way down". It's said that someone buttonholed Thomas Huxley (or Bertrand Russell, or William James, or someone else) after a lecture, and put forth the proposition that our Earth, really flat, is resting on the back of a giant turtle; the question (by Huxley, Russell, or whoever) of what that turtle is resting on is answered dismissively, "It's turtles all the way down, young man!".

Douglas Hofstadter has a notion that for human cognition, "It's analogies all the way down." His Stanford lecture "Analogy as the Core of Cognition" is the place I first ran across it. He calls this notion

a refrain that I’ve chanted quite oft in the past, to wit:

One should not think of analogy-making as a special variety of reasoning (as in the dull and uninspiring phrase “analogical reasoning and problem-solving,” a long-standing cliché in the cognitive-science world), for that is to do analogy a terrible disservice. After all, reasoning and problem-solving have (at least I dearly hope!) been at long last recognized as lying far indeed from the core of human thought. If analogy were merely a special variety of something that in itself lies way out on the peripheries, then it would be but an itty-bitty blip in the broad blue sky of cognition. To me, however, analogy is anything but a bitty blip — rather, it’s the very blue that fills the whole sky of cognition — analogy is everything, or very nearly so, in my view.

End of oft-chanted refrain. If you don’t like it, you won’t like what follows.

Fairly far into the lecture (or perhaps it's a chautauqua), he says stuff that reminds me of the fun I have when I and a good friend are thought-riffing; and of how adrift I feel when that sort of thing becomes too infrequent:

Note that what I have just described is not problem-solving, which has traditionally played such a large role in modeling of thought and been tightly linked with “analogical reasoning”; no, everyday thought is not problem-solving or anything that resembles it at all; rather, it is a nonrandom stroll through long-term memory, mediated by high-level perception (which is simply, to echo myself, another name for analogy-making).

To be sure, thought does not generally take place in a sealed-off vat or an isolation chamber; most of the time, external events are constantly impinging on us. Therefore the purely self-driven flow that the “central loop” would suggest is just half of the story — it is the contribution from within one’s private cognitive system. The other half — the contribution from outside — comes from inanimate objects impinging on one’s senses (skyscrapers and sunsets and splashes, for instance), from animate agents seen mostly as objects (mosquitos that one swats at, people that one tries not to bang into as one hastens down a crowded sidewalk), or from other cognitive agents (conversations with friends, articles read in the paper, email messages, scenes in movies, and so on).

I find the themes Hofstadter explores there analogous to some of the work of George Lakoff, who, before he became a popular lecturer with a political agenda, seemed entirely capable of doing actual science. His Women, Fire and Dangerous Things (hereafter, WFaDT) was a breakthrough work in cognitive linguistics; the later Metaphors We Live By, written with Mark Johnson, is a bit more accessible and builds on his earlier work.

I'll save my carping about his slide into partisanship for later. Maybe for never. :)

Work like Hofstadter's and Lakoff's, along with the recently-popular "mirror neuron" stuff, seems to shed light on a lot of things, from the "homunculus" model -- the little guy sitting in a control seat behind the eyeballs, running things -- to why thinking is so hard.

My current take is that even if mirror neurons are not the big deal that Ramachandran made of them in this Edge piece, there's something to the idea that metaphors and analogies are deeply intertwingled with human thought, and indeed with the brain hardware we think with.

Why does this matter? Lakoff makes the case in WFaDT that for much of history, philosophy had a tacit premise that the trick of perfecting philosophy was to get "the" categories right. The notional grail: there's an inherent taxonomy of categories, and once that's debugged, human perception will perforce be subject to ultimate reason.

This gets called into question if all our symbolic thought is running as quasi-opportune patterns overloaded on nonsymbolic hardware.

I suspect abstract categories just don't map the way the color red does. I suspect that an awful lot of "abstract" human thought isn't as highflown as many of us wish. Perhaps I state the obvious.

I have a lot I'd like to say about this stuff, but it will have to wait for another time. For now I'll close with another excerpt from the Hofstadter lecture.

This viewpoint may be overly ambitious, and may even — horrors! -- be somewhat wrong, but I have observed that many good ideas start out by claiming too much territory for themselves, and eventually, when they have received their fair share of attention and respect, the air clears and it emerges that, though still grand, they are not quite so grand and all-encompassing as their proponents first thought. But that’s all right. As for me, I just hope that my view finds a few sympathetic readers. That would be a fine start.

Read the whole thing. He goes interesting places.