Tuesday, March 29, 2011

What happened next XVI

[Today's post was written before yesterday's comments came in. As you may already have seen, Davee got the right next move while T.C. and David Sedgwick identified What Happened Next.

As chance would have it, I bumped into David (S) at Golden Lane last night. We got talking about von Neumann and David mentioned the "smokescreen theory", something that I intended to come back to in a later post. For the meantime, just remember that some folk believe that there could be more to von Neumann's shenanigans than meets the eye.]

White to play
John von Neumann v A.N. Other, World Open 1993

Yesterday we left John von Neumann considering the ninth move of a game he was playing at the 1993 World Open. Given that he had just three legal moves to chose from, and only one that was even remotely plausible, you might think that what he did was was play 9. bxc3 pretty sharpish. In fact that was the move that White chose, but it wasn't What Happened Next. When faced with the position at the head of today's blog what von Neuman did first was nothing whatsoever for forty minutes.

How did he come to take so long over an absolutely forced move? Did he get drunk and fall asleep at the board a la Tkachiev? Did he nip to the toilet and get himself locked in a cubicle? According to 'Shoes on Fire', Hans Ree's excellent article on the case (republished in The Human Comedy of Chess, Russell Enterprises Inc. 1999), von Neumann did neither of these things. During that extended pause he didn't leave the board at all. He just sat there staring at the ceiling.

Jigger-all might have been what the man behind the White pieces was doing, but there was some activity at his board nonetheless. At some point a spectator approached the table, took a note of the position and then scurried away. It was only some time after this had happened that von Neumann roused himself and managed to make move.

So, what actually was going on during von Neumann's long think? With the benefit of hindsight it seems he was probably starting the chess world off on a ride that would lead directly to the French team's shenanigans at the Olympiad last year. I say "probably" because precisely what it was that von Neumann was up to in Philadelphia has never been proved for sure. Most likely, though, he was using a computer to help him win chess games. If he was, I believe that would make his case the earliest recorded instance of such behaviour at a chess tournament.

Should we accept von Neumann's - presumably unintended - claim to a place in chess history? Let's review the rest of the evidence.

Being dead is usually a pretty good alibi

von Neumann's catatonic episode alone would be enough to get noses twitching. As it happens, though, it was not the first time during the tournament that his behaviour had aroused the curiosity of his fellow competitors. In the fourth round, for example, he lost on time after 9 moves. In the second, though unrated and unknown, he managed to draw with a Grandmaster - Iceland's Helgi Olafsson.

According to Ree (who was quoting John Watson's report in Inside Chess) Olafsson said of that game,

"I was sure I was playing a complete patzer, he had no idea what he was doing and I even thought he was on drugs. He took way too much time to reply to obvious moves and he was very strange. But I made a bad mistake in the opening and I was punished with a draw."

Still, the odd spanner in the works not withstanding, von Neumann's World Open was successful enough that he secured $800 by sharing first place in the unrated section. The Tournament Director had become suspicious, however, and suggested to our man that he both prove his identity and take part in an independent test to demonstrate his chessing skills. At this point von Neumann took to his heels never to return, the prize money that he'd 'won' left uncollected.

If we deduce from this strange behaviour that von Neumann must have been cheating in some way, what is there to suggest that the employment of silicon-based assistance was his method? Well there's the taking of several minutes per move over even forced recaptures (Ree speculates that the mega-long think in the eighth round was the result of a transmission error and von Neumann playing 8. Bxc3 instead of the machine's recommendation of 8. bxc3) for a start, but most telling of all is his name.

John von Neumann was a mathematician who'd made important contributions in the field of computer science amongst other areas. Not the guy at the '93 World Open, another J vN. The chess tournament von Neumann was a "black American with a Rastafarian hairdo" (Ree); the computer geek was born in Hungary. As biographical details go, being black and being Hungarian are not necessarily mutually exclusive, I suppose, but being dead and playing in chess tournaments definitely don't go together [although ...].

Since the Hungarian computer bod John von Neumann died in 1957, I think we can assume that the fellow at the World Open was not him, but an impostor. Unlikely don't you think, that somebody clearly up to no good would choose that particular nomme de guerre were he not benefitting from the help of a Bloody Iron Monster ((c) Michael Stean)? So, yes, I think we can conclude that on the balance of probabilities the '93 World Open von Neumann was the first chesser to be seduced by a computer's charms.

Eighteen years on Feller, Marzolo and Hauchard are surely not the last to follow in von Neumann's footsteps, just the latest*. Two decades ago, a man who will almost certainly never be identified brought a genie out of a bottle and we're never going to get it back in. The question is, how much damage will von Neumann's legacy do?

[EDIT 29/3/11 3:30pm] I should have made it more clear that the account of the von Neumann case outlined above is very much as given in Hans Ree's article - JMGB

What Happened Next? Index

* as the Bard of Barking would say.


Campion said...

Somewhat ironically, von Neumann wasn't just a famous/renowned mathematician: he was known as something of a calculating prodigy, and his speed of calculation has almost become the stuff of legend among the tribe. Which is to say, I'd be less surprised by the real JvN rising from the grave to push wood in Philadelphia, than by him taking 40 minutes over a move. (I don't recall if he was actually good at the game in any serious way, though.)

Morgan Daniels said...

Von Neumann to Jacob Bronowski in the back of a taxi:

'Chess is not a game. Chess is a well-defined form of computation. You may not be able to work out the answers, but in theory there must be a solution, a right procedure in any position. Now real games... are not like that at all. Real life is not like that. Real life consists of bluffing, of little tactics of deception, of asking yourself what is the other man going to think I mean to do. And that is what games are about in my theory.'

Jonathan B said...

Two wonderful snippets.

Talking of games, Morgan, last night I observed your board after about six moves and played the "work out how they could possibly have reached this position" game. As usual with you, I failed.