Sunday, March 18, 2012

Language, Flat Earth, and Goldilocks: Three riffs on evolution

A blonde girl with big lips surprised me after my talk in Toronto yesterday. Double surprised me, in fact. I thought she’d been asleep but I guess she’d been listening with her eyes shut. I had just argued that language, analysed à la Chomsky, reveals the stages by which our minds evolved. Minds don’t leave fossils. So, the common wisdom says, we can only get at cognitive evolution indirectly, by examining artefacts, like cave paintings and tools. Finding that the capacity for language is itself a fossil bed of mental evolution turns the conventional wisdom on its head. But Eyes Shut had been reading articles by a journalist—not religious or christian, she hastened to add—who had found lots of problems with evolution and she found his case convincing. Her question: so what did I think about that?

The classic Kruger-Dunning conundrum! To know whether you can trust the journalist and his assessment of evolution, you need to judge his level of expertise. But to judge expertise, you need expertise. Which makes the journalist useless: if you’re expert, you don’t need to rely him; and if you aren’t, you can’t. But people engage in because charades. They don’t believe things because they’re true, but because they sound like what they want to hear. So, they assume that unreliable sources are reliable and, worse still, assume that they have gained expertise by reading them.

Eyes Shut seemed more open than this, but I know from experience how difficult Kruger-Dunning delusions are to dislodge. My mother is always Kruger-Dunning me. She insists on having unearthed all manner of esoteric truths in health and healing by reading about herbs, and electro-medicine, and magnesium, and potato skin soup fasts, and coconuts, and …. But without expertise, you have no way of knowing whether what you’re reading—electro-medicine, evolution, etc.—gives correct answers, or competent answers, or complete answers, let alone whether it asks the right questions to begin with. Anyone can make a case sound convincing by ignoring everything that’s inconvenient and, if you’re not expert, you don’t know what they’re hiding from you, or from themselves.

So far as evolution and language go, though, I can answer with some expertise. And the power of Darwin’s idea seems to me little short of miraculous.

For instance, I recently finished a paper that argues against the “geometric hypothesis”. In brief, think of the brain like a computer. Language is one of the tasks it can perform (like playing dvds, or word processing). My job is to discover the program that the brain runs which makes it fit for language. Hence, what are building blocks of the program and how are they put together? The geometric hypothesis says that some combinations of building blocks are ruled out, even though the computational system would have no trouble with them. The opposite view, which I believe in, is that there are no such restrictions: the brain welcomes all inputs. It’s like kosher versus treif: some combinations of food are fine for the digestive system (meat and milk, meat and fish, …), but you’ll never find them on a Lubavitcher buffet.

There are two ways to disprove the geometric hypothesis.

In one, you go off and describe hundreds of languages (in this case, we’re interested in pronouns, verbs, and words like this/that, here/there, hither/thither). You then pool all the data to see which types of languages the brain is capable of producing. Then you simmer away for years to boil the data down to a set building blocks all combinations of which are used by some language or other—hoping that the whole thing doesn’t blow up in your face like an ill-set pressure cooker.

Given the hundreds of languages you need to document, the hundreds of hours that go into describing each one, the hundreds of hours that go into conducting initial, then larger, then yet larger cullings of the data, and the hundreds of hours that go into devising and evaluating successive proposals, a reasonable estimate is that it has taken this approach some 100,000 “thought hours” to show that we’re not at a Lubavitcher buffet: the geometric hypothesis is wrong.

Here’s the Darwinian alternative. You consider whether geometries are evolutionarily necessary (they aren’t), whether they’re evolutionarily stable (they aren’t), and whether they offer informational, hence adaptive, advantages (they don’t). In fact, if we ever had geometries, evolution would expect us lose them. So, if it’s a sunny day and you have some good coffee, you can probably get all this thinking done under 10 hours.

Darwin didn’t have that much to say about language. Indeed, we can apply his ideas to the geometric hypothesis only because several scientific and mathematic revolutions separate us from him. For his ideas, in such radically foreign intellectual terrain, to deliver in 10 hours what nose and grindstone only churn out after 100,000 strikes me as close to miraculous.

* * *

“Thought hours” is a useful way of guaging the robustness of an idea and the concept came up again in my conversation with Eyes Shut, in a moment of superb irony. As said, it’s up to experts to assess critiques of evolution. However, if the problems are so obvious that an enthusiastic amateur can unearth them, then you’d think that theory would have been debunked long ago. There is, after all, no “scientific establishment” that protects bad ideas. There is only a scientific disestablishment, that finds the false and roots out the wrong. The bigger the ideas are, the more credit you get for making them fall.

However, Eyes Shut observed, it had taken centuries for people to stop believing that the earth is flat. “Evolution, your days are numbered” was the subtext, I guess. The comparison, though inapt, is fascinating.

Evolution and flat-earthism may be of great age, but, in its time, evolution has withstood hundreds of thousands of hours of criticism, modification and reform. Flat-earthism dies once you watch a ship vanish over the horizon: hull first, mast last. Age in thought hours, not years, is what matters.

What keeps me smiling as I type (and reread) this is the latest twist in the evolution of anti-evolutionism. Flat-earthers are the archetypal rejectors of progress, the possessors of undislodgeable delusion. Evolution, by contrast, is an idea so revolutionary that, long after it became the mainstay of the natural, social, and cognitive sciences, the religious still struggle to accept it. Yet at least one anti-evolutionist wants to equate evolution with the flat-earthism, the symbol of (religious) recalcitrance.

* * *

Before Toronto, I was at a mine in northern Quebec. The highlight was watching professional geologists in action. Scrutinizing tray after tray of dreary rocks, consulting the chemical analysis, plotting and comparing the drill sites, then having an animated discussion about how long this now upturned, underground volcano had been active and underwater, how long it had remained hot after it ceased erupting, why millions of years later it had withstood the pressures that had distorted the neighbouring geology.

Here’s how the discussion did not run:

“Let’s pretend that god created this bit of the earth so that it looks like there was a volcano here.”
“Yes, and let’s pretend that he put a basalt cap at this end so that we would think that the convection currents would have remained active longer.”
“Good idea, and let’s also pretend that this a magma chamber and that this is swarm of barren dykes.”
“Great because if we pretend that god did all of that, then we’ll know where to look for which metals.”
“Yes, but let’s just hope though that god made this pretend volcano like all those other ones.”
“True, you can’t be too careful. You know how god loves dykes. He might have just put them there to test us.”

Geology makes a nonsense of the bible. But so does physics. So do linguistics, genetics, and cosmology. And yet only evolution seems to hold a special place in the hostilities of the godly, as a shibboleth of the faithful.

I think it’s because evolution is the goldilocks science. Linguists are too abstract for anyone to care to question them. And geologists and physicists are too concrete for anyone to dare to question them (the bible belt likes its oil and mining companies and the GPS that guide them). But evolution is about us. It hits home. Plus you can deny it without personal cost. Deny physics, and your GPS navigator becomes black magic (“Begone Satan, I shall perform no U-turn in 100 yards”). Evolution is—as the faith-fooled bleat—“just a theory”.

And yet, as genetics advances, it builds a bridge between evolution and technology (think of the bible belt’s love of GM soya). The debunkability of evolution is declining market.

Besides, it’s only a matter of time before the religious discover a special fondness for evolution. As cognitive science finds the fossils that reveal why the mind creates gods, and why groups of minds create religions, the godly will suddenly proclaim that evolution provides the ultimate proof of god’s existence: he has programmed us to believe in him! Well, we’ll deal with irony when we come to it. First, let’s deal with his charade of making gold mines look like superannuated subaquatic volcanoes…

No comments:

Post a Comment