Because most philosophies that frown on reproduction don't survive.

Monday, July 18, 2011

A Foundation of Determinism

Paul Krugman recently did a Five Books interview with The Browser, talking about his five favorite books. The books are: Asimov's Foundation series, Hume's An Enquiry Concerning Human Understanding, two books by Lord Keynes, and a book of essays by economist James Tobin, one of Krugman's old teachers. Of Foundation he says:
This is a very unusual set of novels from Isaac Asimov, but a classic. It’s not about gadgets. Although it’s supposed to be about a galactic civilisation, the technology is virtually invisible and it’s not about space battles or anything like that. The story is about these people, psychohistorians, who are mathematical social scientists and have a theory about how society works. The theory tells them that the galactic empire is failing, and they then use that knowledge to save civilisation. It’s a great image. I was probably 16 when I read it and I thought, “I want to be one of those guys!” Unfortunately we don’t have anything like that and economics is the closest I could get.
This sounded vaguely familiar, and looking around I realized I'd heard about this affection of Krugman's for Foundation before, but again it strikes me as underlining a lot of what I find basically unappealing about the approach to economics and humanity which Krugman has.

To me, the interesting thing about the Foundation books is their historical sweep. Asimov said he got the idea from Gibbon's Decline and Fall of the Roman Empire, and there's a wonderful sense of the sweep of history, of loss and of scale in the books. (Of which, to my mind, it's only worth reading the golden age novels, not the later ones tacked on.) But what's incredibly frustrating is the gnostic determinism of the vision -- that somehow, a really brilliant group of guys can working through a whole bunch of equations and calculate what is going to happen for centuries into the future, as if the universe were some gigantic difference engine with the people in it as gears, moved through predictable motions by the forces exerted on them.

This view, it seems to me, is not just incorrect, but also fairly dangerous. Wrong because it fails to take into account human free will, which is, it seems to me, one of the most defining elements of humanity. Dangerous because it offers the illusion of control, that one can take some great action and reshape society if one can only get people to stop acting as persons and play their appropriate parts in the grand machine.

To see economists as like Asimov's psychohistorians seems to me to glorify economics far above than its proper level. What economics is fairly good at is explaining how certain mechanisms work "all other things being equal" -- yet what makes it frustratingly approximate is the degree to which things are so seldom equal. As soon as people starting thinking of the economy as some great machine with levers just waiting to be pulled (whether it's liberals convinced that if only we could put through a couple more trillion dollars worth of stimulus everything would be fine or conservatives convinced that we can always raise tax revenues by lowing tax rates) they set themselves up to cause more harm than good.

9 comments:

Lucius Quinctius Cincinnatus said...

If men were the automatons that psychohistorian men purport them to be, then being automatons themselves, the pyschohistorian men could have never invented the amazing nonsense that they call psychohistory where men are but automatons.

That's a paraphrase from one of Robert Heinlein's Lazarus Long novels.

Joseph M said...

The question is: Is economics any less a fantasy than psychohistory? Just once, I'd like, before the fact, to have an economist lay out his arguments, make concrete, non-trivial predictions - and get held to them.

Loved Foundation as a kid, even liked the psychohistory idea of men as atoms: individuals are unpredictable (that's what that whole 'Mule' bit was about) but that, in aggregate, we tend to behave in predictable ways. As SciFi - sure! Let's run with it. But in the real world? Ummmm - maybe not.

That Krugman would think economics is a sort of psychohistory-lite - yep, giving a wee bit too much credit to economics, and revealing that economics aspires to a fantasy (and economists aspire to being wizards!).

Darwin said...

The question is: Is economics any less a fantasy than psychohistory? Just once, I'd like, before the fact, to have an economist lay out his arguments, make concrete, non-trivial predictions - and get held to them.

Well, there are some fairly basic predictions that would be easy. For instance, ask an economist what would happen if we instituted price controls to keep the price of gas down to $2/gal, run the experiment, and his prediction will probably turn out to be fairly accurate.

The thing is, many things people want economists to provide insight into (How can we get unemployment down? Are housing prices too high?) are questions that are far more complex and far harder to test.

I'd tend to say that economics is useful -- but that people need to be studiously modest about it's abilities.

Banshee said...

Asimov wrote Foundation when he was a kid who liked history and was flirting with communism's inevitable history. But mostly, psychohistory wasn't supposed to be taken seriously; it was a plot device. Similarly, all his repeated "But actually, that was all part of the plan!" problem stories.

Asimov being Asimov, he had to go ahead and break the system during the stories. (Seldon's recording droning on about some future economic problem, while all his followers were waiting for a solution to their invasion problem, was a hilarious poke at fans taking psychohistory too seriously.) There are also a few moments poking fun at Gibbon, IIRC, because he's not quite as omniscient as all that, either.

So yeah, I'm disturbed that anybody over the age of 16 would take psychohistory seriously as anything other than a plot device, much less take it as an influence in the way Krugman does. Enjoying it as a retelling of the Roman Empire or a fun Golden Age sf story, that's different.

John Beegle said...

Dear Darwin,

I loved the early Foundation books as a kid too. "Blind Oracles" is a good cautionary book for intellectuals who are tempted to confuse models with reality, but since you're discussing economics ... I'd be curious to learn whether you lean more towards Keynes or Hayek?

Darwin said...

John Beegle,

I'm not a full blown Austrian by any stretch (being a pricer by trade, and self taught at that, macro is a bit alien to me) but generally speaking I've found more insight in what I've read of Hayek than what I've read of Keynes.

Joseph M said...

Darwin - I'd happily take studious modesty from economists. Your example - that price controls will create predictable results - is about the level I'd be comfortable with. If only they'd keep it that simple, we'd all be better off. Well, except for maybe Goldman Sachs...

Tom Simon said...

For what it’s worth, Asimov himself considered the idea of psychohistory rather dodgy. I believe it was John W. Campbell who insisted on the guff about the Seldon Plan when they first discussed the idea.

In fact, Asimov made the Galactic Empire so big to try to fudge one of his own principal doubts about the concept. He often said that he could not imagine something like the Seldon Plan working with a population smaller than the quadrillions of the Empire, and he had a sinking feeling even that wasn’t big enough.

Then, just to underline the point, he wrote the two stories about the Mule, which knocked the whole scheme into a cocked hat. It turned out that the only reason the Seldon Plan ever worked at all was that the Second Foundation had their thumbs firmly on the scales at all times.

Asimov was naive about economics and a sheer dummy about politics, but to do him credit, he was not dumb enough to believe that psychohistory was realizable or desirable. It takes a Krugman to be that foolish.

kkollwitz said...

"we can always raise tax revenues by lowering tax rates"

I don't think any conservative belives this is "always" so; but rather that the modern welfare state taxes at such a rate where it is incrementally so.