[…] human activity in general, and linguistic activity in particular, may in many instances be an end in itself, a play: talking, more often than not, contributes very little to information; many people talk because they like it, not because they have anything to communicate. But this does not imply that linguistic evolution is not determined by the principle of the least effort. Talking is often as a game, but a game is only worth while if he who plays the game sticks to the rules. Cheating makes sense only if the game ceases to be an end in itself, and there is, for chatterboxes, no reason why they should cheat in the game of talking. The rules of the game of speech are laid down by its communicative uses.
- Martinet (1962), Pag. 139