Game theories

SEVEN GAMES: A Human History

Author: Oliver Roeder

Publisher: W.W. Norton & Company

Price: $26.95

Pages: 320

As an adolescent, I fancied myself a devotee of the ancient Chinese board game Go. I had never actually played it, but in one of my favourite novels Shibumi, by Trevanian — the super-assassin hero praised Go as a fitting pastime for the kind of warrior/poet/lover he was and that I hoped to become. “Go is to Western chess what philosophy is to double-entry accounting,” he says, probably before killing someone with a toothpick. Sadly, when I actually tried to play Go, it turned out be … a board game, and a difficult one at that. I gave it up when I realised that in return for the labour of truly learning Go, I would become not an enlightened international assassin, but just a guy who could play a game.

Oliver Roeder, a student of games and game theory, is deeply aware of the tension between what games are and what people project onto them. His “group biography” of seven classic games — checkers, chess, Go, backgammon, poker, Scrabble and bridge — is in many ways an interrogation of these questions: Are games more than their rules and playing pieces? Are they metaphors for deeper truths of the human experience? Is chess “life in miniature,” as the former world champion Garry Kasparov once said? Or is it just a board game — risk with more rules and a boring map?

Mr Roeder is himself a player, and he structures much of his narrative around his entry in major competitions. He begins with the long history of games, going back 5,000 years, and asks: Why does almost every society engage in games and why have certain games survived for centuries? His suggested answers range from the simple — because we’re bored — to the complex, as presented by the aptly named philosopher of games Bernard Suits, who coined the phrase “lusory mind,” that is, a mind freed from practical or real concerns and thus able to ponder immaterial problems.

But this book is not truly a philosophical inquiry. Nor is it a deep consideration of any of the games or their star players, though such people provide much of the joy of the book. There is Marion Tinsley, an abstemious Midwesterner of steady habits and deep religious faith, and the greatest checkers player who’s ever lived; and Nigel Richards, an aloof ascetic who lives off the grid in Malaysia but emerges to dominate every Scrabble competition he enters. And while backgammon’s roots are buried in Egyptian tombs, it was a Russian expatriate named Prince Alexis Obolensky who popularised the game as a jetsetter’s pastime in the 1960s. But none of these people are the book’s central characters. Instead, this is the story of people like Jonathan Schaeffer (checkers), Gerald Tesauro (backgammon) and Jason Katz-Brown (Scrabble). They are very good players. But they are here for their skill at programming computers.

Each of the primary sections of this book reads like a repeating myth of hubris told with different characters but the same ending. Every game has its history, its champions, its quirks and its community, and then comes the programmer who believes he can teach a computer to play it. Each time, devotees of the game stake their claim that their pastime is a pure expression of ineluctable human creativity, and then, as the programs improve, the players are stripped of their illusions. The human champion ends up in a hotel ballroom across a game board from some young programmer who sits next to a box that tells him what to do, and each time the box eventually wins. The game they thought was an art is just another mechanism, no more inaccessible to the brute strength of microprocessors than, say, assembling an automobile.

By the time computers defeated human champions at chess, then poker, then backgammon and Go, the human players had submitted to their betters. High-level poker, Roeder tells us, no longer involves playing “the man, not the cards,” but rather memorising ideal strategies spat out by programs running millions of variations on any given hand. Instead of gamblers noting one another’s tells, the game is dominated by young men wearing sunglasses and headphones so they won’t be distracted. After wins or beats, players check their play against the programs they’ve purchased, to make sure they chose the software’s recommended course of action. You begin to realise that people aren’t playing one another with the help of computers; the computers are playing against other computers, using humans.

Oddly, the only game yet to be conquered by computers is contract bridge, and my sense of relief at this lone holdout from the onslaught of AI was extinguished when Roeder let us know that the game, a relatively recent addition to the pantheon of pastimes, is now dying out, its players trending older and smaller in number. That might not be a coincidence: Perhaps by now we’re so used to playing against, or for, our digital masters that we’ve lost interest in something so messy and human that its strategies can’t be simulated by an online bot.

To play a game is, Professor Suits said, to try to solve a voluntary problem. But it is also a voluntary engagement with another human, temporary but total, to create a shared narrative of attack, defence, stalemate, victory and loss. The one thing a computer can never do is enjoy another computer’s company. Let us hope that no new genius arises to teach them how to do that.

©2022TheNewYorkTimesNewsService