Why are signals between animals believed? I mean, if I say (roars) like this, meaning, 'If you don't go away I'll hit you,' why do you believe me? I mean, maybe I'm bluffing, you know. There is a real problem about honesty in any symbolic communication, there are problems about honesty when we're talking to one another, why aren't I lying all the time? And the answer there is, you'll remember what I said, and if I lied last time, you won't believe me next time, and so on. But why do animals, on the whole, believe, if I can put it that way, in the truth of the signals they're receiving? And this is the problem that Zahavi addressed himself to. And he came up with an answer which still attracts a lot of interest, though I've always been rather cynical about it, less so now than I was, which is that signals - the honesty of signals are guaranteed if the signal is expensive. If I can do that and it's cost-free, then I'll lie. But if it's genuinely costly for me to do it, for some reason or another, I use a lot of energy doing it, or I have to spend hours beforehand developing my muscles in order to be able to do it, or something, then it, in some sense, guarantees the honesty of the signal. Now, this is an interesting sort of chapter in the history of science, in a way. Zahavi is not merely - well I was going to say he's - he's an in principle non-modeller. He really does not believe that formal mathematical models illuminate anything. And he believes that only verbal models, and very intuitive verbal models, are any use. My picture of the history, and it may be quite different from his, probably is, that he talked about the handicap principle at a number of meetings, and so on, and I felt rather strongly that it wasn't very satisfactory that one just have an idea which was going the round which had never been written down and published. So I persuaded him to write an account of it for The Journal of Theoretical Biology, because I felt it needed to be something we could get our teeth into. And he did that. And then...

[Q] Could you just explain the relation between cost as you were talking about it just now, and handicap as he understood it?

Well, I would explain it if I understood it, but, you see, it's a purely verbal model and I don't understand the parallel.