I didn't have to read much game theory. I read the first chapter, I think, of a book by two guys called Luce and Raiffa or something, you didn't have to read much, just to get the notation, that's all you need. And the notion essentially was - what would you expect - well no, let me step back a bit - Classical Game Theory, which Luce and Raiffa, and it goes back to von Neumann and Morgenstern, what they're doing is saying, what would a reasonable man do in a contest situation if he can assume that his opponent is reasonable as well. So you and I are playing a game, I say, 'Dawkins is not only reasonable, he's pretty smart, so I can work out what he'll do, what had I better do?' And I must assume that he's being just like me, and can we find some kind of rational solution to the game, assuming - each of us assuming the other guy's pretty clever. In animals, you don't say, 'I think this animal's pretty clever,' it's probably as thick as two planks, there's no need to think it's clever. What you have to do is to look for what I call an Evolutionary Stable Strategy. An evolutionary stable strategy is simply a behaviour or a strategy, call it what you will, which has the property that if every member of the population does it, or almost every member of the population does it, then any mutant doing anything else possible doesn't do as well against the members of the population as the members of the population are doing against one another. And that can be formulated mathematically, very simply, it's not difficult, you don't have to be a mathematician. And you can then ask: will this, that or the other strategy satisfy this mathematical condition? And it turns out, actually, or it turned out later, to be really rather a powerful technique, probably because it just is so simple. And while I was in Chicago I worked out the formal conditions for something to be what we call an ESS nowadays, an evolutionary stable strategy. And I tried it out on George Price's problem, as I think of it, the problem of retaliation, and found that, in fact, George had been quite right, that you can make symbolic, symbolising, ritualised behaviour stable if you introduce retaliation as a possible alternative. And I invented a game which is now known as the Hawk-Dove Game, remember this was during the Vietnam War, in Chicago, Hawks and Doves, that's sort of popular. And at that stage, I thought, well, I suppose I'd better quote Price, because that's where the whole idea had started from. And I searched through the pages of Nature and the indexes and it didn't seem to have been published, and I couldn't find any record of it being published anywhere. And indeed, it hadn't been. And I had this problem of how do I get in touch with George Price, and found out what happened to it. And in the end I wrote to Nature and asked for his address. And they had an address in London, in Charlotte Street. And then I tried to contact him. And I can't remember, I think in the end I actually had to go and bang on the door, but that's from memory.