I was just reading Roger Caillois’ book about games when he brought up the argument that competitions (agôn) are not unlike theatre plays: according to him, they have a dramatic arch, and as such they need an audience. Obviously, that second argument has been refuted by theatre theorists quite some time ago. The competition can go on even without an audience, while a play without an audience will just be a rehearsal.

Interestingly enough, that brought me back to the question of narratology in games, where the same argument has been brought up: all games have a dramatic arch, even if unintended, and therefore a “narrative”. I personally find the term applied in such a manner way too broad. The thought that Tetris is telling me about life and the human condition in an uncaring universe is simply making me cringe.

It made me wonder though: What if you could construct a Turing test for video games? If you can write a simple program to make a computer play a game on another computer up to the end, does that tell us something? I think so: it would be a proof that the game is based solely on rules and contains no narrative.

If that computer fails at some point or we would feel the need to rewrite the program to account for some special cases, we may have a case of a narrative game. Narration does not follow strict rules. It can go this way or another. Behind a layer of symbols, it contains meaning that can’t be understood by a simple program.

Narration always needs a human agent to be understood and, therefore, to work. This is why theatre needs an audience and competition does not, and some games have a narrative and some have not.