Charles on… anything that comes along

Back in 1972, when the world was convulsed (or so it felt) with the doings of two men over a small board with 32 black and 32 white squares in Reykjavik, I was getting into chess about as much as I ever would. Played against my father, played at a chess club locally. Wasn’t that good, but good enough to enter a national tournament (we were living in Malta at the time, so “national” would roughly equate to sub-county in the UK) and get 4.5 out of 9.

And Bobby Fischer was playing Boris Spassky, and I devoured the papers to learn the latest twist - how he hadn’t turned up for the second game, the dissection of the overhead lamp (revealing two dead flies - the perfect grainy detail that most people writing a fictional account would overlook), the intervention of James Slater (who I’d not heard of) to up the prize fund.

It was amazing, the biggest show in any town; as convulsive as, just under a decade later, the matches between Borg and McEnroe would be in tennis. Like McEnroe, Fischer had that self-destructive streak, doing stupid things that got him defaulted, arguing with organisers, yet also able to pull remarkable things out of nowhere. I followed the games, but to be honest they were far too deep for me. The test of how well you understand a game is whether, given a position, you can figure out what the next move should be. I hardly ever could. But it was an interesting experience. My father (my most regular chess opponent; when we first began he’d start without a queen; then it was a rook; then a bishop; then level) and I bought all the instant books that came out from the match, and read them and played some of the games through, trying to puzzle them out.

To be honest, for me at least it was like being a dog trying to understand quantum physics. Chess at that level isn’t any less comprehensible than club chess; all the pieces are visible, all the possible moves are known. But the top players can see both the broader sea of possible moves and simultaneously focus on the few that will yield dramatic results. In a way, they’re like mathematicians, who also tend to peak in their 20s or 30s.

As the Guardian noted about Fischer, chess was the game that kept him sane; the rest of the time he was just mad. In the years since his virtual disappearance, there have been repeated rumours that he would pop up on the internet chess servers, play games showing flashes of his old brilliance, and then disappear again.

But he did have one really, really good idea which grew out of his frustration with the fact that increasing computer power meant that whoever had studied openings to death - whoever, basically, had the best computer analysis - came into a match better prepared. Because every match starts with the pieces on the same squares, you can analyse for any move. Use a computer and you can analyse the game, and your opponent, into dusty draws.

Well, blah. Fischer suggested a tack that would confound the computers: FischerRandom. Mix up the pieces on the back row at the start of the game. (The king has to be between the rooks, so you get 960 possible variations, apparently.) Rules are here - and keep some ideas from standard chess, such as opposite coloured bishops and so on.

Voila! At a stroke, the wonk who spends all the time just analysing openings has been neutralised. Once again you have to use your mental energy working from first principles, trying to see moves and possibilities. I think it’s the best possible idea that could be applied to human chess, since even with computer help nobody could prepare for what their next match would throw at them.

The irony, of course, is that computers would excel at FischerRandom - aka Chess960 - since they’d be able to figure out the decision trees far beyond people’s abilities (though their general strategic algorithms might be troubled by the lack of the standard starting positions, and those which rely on big opening books would have to toss them out). But it would make the games between humans more interesting, because more random. (Intriguing to see that there’s a project which is trying to analyse it to death, if I’m reading it correctly. Oh well. Some people can’t leave a good thing alone. It’s in our natures.)
And it’s that inability to rely on a specific set of opening moves that I find so attractive about go - the ancient Oriental game. You have a 19×19 board (so 381 places, more than four chessboards’ worth), and it starts empty - so that within 20 or so moves you’re certain to be playing an entirely new game, if not to the universe then almost certainly to you. It’s a slow, subtle game, and computers can’t get within spittting distance even of good club players. Opening theory is general, and computers get so badly lost in the opening phases that the game’s pretty much lost for them by the time the real fighting starts.

I wonder if Bobby Fischer ever tried go. Though I doubt it would have kept him sane. Nothing could.

These posts might be related (the database thinks..):
• My latest at the Guardian: why do we want people to be computers, and three podcasts (10 February 2008; score: 27.77%)
• Farewell to Bill Gates: I doubt we'll see his equal again (12 January 2008; score: 27.57%)
• "A full house of dysfunctionality": why we all hate automated answering systems (7 September 2004; score: 24%)