Last night I saw Werner Herzog’s Cave of Forgotten Dreams. 3-D makes everyone look small and doll-like, like the Indians/Native Americans they used to have in panoramas in glass display cases in museums. It’s just not worth it.

One of the archeologists in the movie talks about how stone age men 30,000 years ago invented two notions of the world around them: fluidity — the capacity for people and objects to suddenly change their nature (think metamorphosis in Ovid)– and permeability — the linkage between the natural and the supernatural world.

Then, he continued, man shouldn’t be called homo sapiens, but rather homo spiritualis, which is what distinguishes us from the Neanderthals, etc; it is our figurative art represents out perceived linkage to the spiritual world that, (I say) shrink though it may, never quite vanishes into oblivion. To seek a morality in the rationality of science is itself a kind of moral attitude.

Meanwhile, through the movie, which was a bit long, I found myself distractedly thinking about ways of knowing the world. In May 1996 I wrote what may have been the first paper on (financial) Model Risk in Risk Magazine (the Goldman version of the paper is on my website at www.ederman.com…). There I wrote

This reliance on models to handle risk carries its own risks. In this report we analyze the assumptions made in using models to value securities, and list the consequent risks.

At that time I classified models into three types: fundamental, phenomenological, and statistical. Since then I’ve modified this categorization somewhat: in my forthcoming book I distinguish ways of understanding the world as follows: theories, models and intuition. (Not the absence of a serial comma, British style; my copy editor at Simon and Schuster insists on serial commas, and I give up. Americans need everything spelled out for them, and want a comma before the conjunction.)

I’ve discussed some of this stuff lately with M. Taleb (the M. is the French honorific, not an initial) and I comprehend where we differ. He thinks (as I see it) that models are inferior to heuristics, because he comes from a trading environment where heuristics and common sense matter, and indeed they do. Foolish to rely on models of human affairs rather than common sense.

Heuristics are indeed a way of dealing with the world, a sort of intuitive generalization from experience. But heuristics can take you only so far. That’s where scientists come in. I come from a scientific background, and my beef with models is that they are not theories, and one had better remember that.

Theories, I argue, are a way of penetrating to the heart of things, perhaps even to the spiritual heart — and that requires intuition and a connection to a world of strange notions that emerge, fluidly and permeably, from somewhere unknown and sometimes turn out to be true.