#63 On magic tricks

Dear Mat,

First off, we’re accusing one another of “essentialism” and “foundationalism”. I think we’re both against those things and think that science should be done accordingly. Truce? More importantly, I feel we’ve lost sight of what these skyhooks are, the ones for which I have a “hatred”. I don’t think we’re talking about discredited skyhooks from the past like creationism or vitalism. Personally, I’m thinking of spooky theories of consciousness — which Popper was into and Deutsch refuses to dismiss — and a kind of magical version of knowledge creation “ex nihilo” as in Deutsch’s idea of creativity. These remaining skyhooks are, tellingly, all about the knower, not the known.

With that in mind, here’s something interesting about magic tricks: every bit of stage magic has three audiences.

The first audience are fans of Uri Geller and think that real (supernatural) magic is actually happening on stage.

Most people nowadays are in the second audience and don’t believe that anything supernatural has happened. Instead they think, “There’s some trick to it, something special that only magicians can do, on another level of expertise, beyond my ken.”

That’s consistent with the data, a good placeholder theory until a better one comes along. You certainly can’t accuse it of resorting to supernatural skyhooks. But it’s a skyhook nonetheless. Indeed the skyhooks that I criticise are of this allegedly non-supernatural kind, put forward by people who call themselves naturalists or scientists or rationalists or sceptics. They can’t imagine how consciousness or creativity could be anything other than a real trick, rather than craned-up from cruder phenomena.

This second audience at the magic show think to themselves:

How is the magician doing this? It can’t be real magic. But let’s see, the only way I could do it would be if I set all this up before hand, had a whole collection of stacked decks, rehearsed with hired plants in the audience, had some ludicrous contraption on the inside of my suit, spent literally years pracitising one trick that only takes up two minutes of the show…  a whole bunch of things that are so ludicrous, inelegant, elaborate and distributed in time and space from the event on stage, that there must be some as yet unknown but real trick to it. Not real magic, but a real non-supernatural trick that is much simpler than all the laborious, implausible, hard to specify steps to do it the hard way that doesn’t involve any special abilities.

But here’s a useful definition of a magician: someone willing to spend all their time offstage, doing — by any means physically possible — whatever it takes to make something onstage look physically impossible or unprecedented.

That’s it. That’s the trick. It happened before the show started. And the third audience are weirdos (apparently) like me who try and figure out the elaborate craned-up explanations of how a magician tricked a room full of people into thinking there’s a real trick. I concede that most people hate, fear, or are simply uninspired by this. But I rudely point out that the successful explanations of tricks, so far, have been of this offstage kind.

In fact, for me, the juiciest episodes in stage magic occur when someone is actually given the longwinded offstage explanation of how a trick was performed, but still insists that it is wrong or some kind of decoy told to outsiders to protect the secret, real trick of the magician. Explanations, even when they work, are frequently ignored if they dispel an illusion.

But if you don’t yet have an offstage explanation for how the magician went to great but mundane lengths to fool you, should you default to the non-supernatural skyhook and say it’s a real trick? The only answer has to be: sure!

…but I wouldn’t bet on it. That’s all it comes down to: confidence and the updating web of explanation. There’s no transcendent, essential, foundational reason and no one’s saying the skyhook is definitely wrong, merely that it’s a bad bet.*

Most importantly, I think modern cognitive science (Popper was too old for it and it’s Deutsch’s blindspot) tells us something about why we default to skyhooks and assume the magician is performing a real trick. It’s because of the natural assumption that causes and effects are proximate in time and space. We evolved to think this way: the explanation for the trick must involve something happening onstage at that moment. But as we’ve begrudgingly learned in physics and biology, complex phenomena have explanations that are distal, not proximal. E.g. scientists and philosophers accept that the wondrous array of species on earth is explained not by spontaneous creation but by an unimaginably complicated interconnected history of minute changes by degree and a lot of time.

But most philosophers and scientists — professional creative and conscious thinkers — thinking about creativity and consciousness  (tricks they’ve seen performed in their own lives) are still betting that the magician does something onstage. To think otherwise is to undermine their self, their occupation and their view of epistemology, which always privileges the knower over the known.

Thinking in terms of offstage explanations, the third audience (such as hardcore naturalist philosophers and cognitive scientists) would say that natural science should feed back into epistemology. For instance, it suggests that there’s no downward causation, explanatory arrows go one way, causation itself is emergent and that all of these asymmetries (which undermine but don’t erase skyhooks) are byproducts of the low entropy early universe and the resultant asymmetry of time. They have confidence in the sciences of thermodynamics and cosmology that got them that insight and now use it to reflect back on epistemology and scientific explanation itself.

But the loopiest thing about contemporary thought is that we now know something about the offstage nature of thought itself. One of the natural sciences — cognitive science — has caught up with the sacred domain of philosophers: their own thoughts, qualia, creativity. We have more confidence in various scientific theories** than we do in the coherence of the philosopher’s knower, or in “creativity”  or “consciousness” or “Popper” or “Deutsch” or “Mat” or “Jamie”. These things are wonderful tricks and I bet that their explanation will be amazingly elaborate, more than worth the price of entry and that most people won’t believe it even when it’s given to them, especially because it will be an explanation of why they won’t believe it.


*I also think most skyhooks violate the hard-to-vary criterion advocated by Popper and Deutsch, because they’re usually ill-defined. They remain on the books only because craned-up explanations of consciousness and creativity are of the type that people simply can’t or won’t imagine.

**Incidentally the next big implication from cognitive science is that the human brain apparently gains knowledge in a way that (I think?) Popper would consider impossible, i.e. Bayesian credences and inference to best explanation, or maybe even something like abduction.

Also published on Medium.