David Bellos: Well, are humans the only species that have language?  It’s a question that's been asked many times, but to think about it I’d like you to imagine a dog.... and he’s a very thoughtful dog and he can hear humans barking.  He can hear that amongst the funny noises humans make there are a number of signals with fixed meanings like "walk," "sit," "heel," and he ponders as to whether the rest of the noise they make is just barking or whether it constitutes a language, and I'd like to fix on that idea.  I mean, obviously it’s a silly fable, but it’s not completely silly in that, precisely because we do not understand the noises that dogs make except for a few individual signals, such as "Let me out of here" or "Take me for a walk" or "There's an intruder," we therefore say it’s not a language because it consists only of a discretely or a fixed number of specific signals that don’t change, and the rest is just dogs barking.

Dogs could, in this fable, have exactly that same understanding of human language. The problem is that we can’t translate between dog and human.  If we could, then we might know whether dogs have a language or not.  The condition for the existence of a language, or what we think of as a language, is its translatability.  So the boundary between our species and others is indeed an unbridgeable gulf until we learn to translate them.  If we could translate any of them, we would then have a much more intelligent and interesting view as to what makes human language different from.  But there's no earthly reason, I mean, given that language, as I've said at the beginning, is a form of human behavior, why a dog or a cow should have a human form of behavior.  They’re not humans, so it’s a species difference.  What would a dog want to say anyway that would be of interest to us?  Only a limited number of things where we interact on specific points. 

And so the argument that only human language is language and that animal communication systems, however sophisticated they are, and some of them are quite sophisticated, are not languages because they consist of discrete signals is a kind of—it’s a circular argument; it’s a self-fulfilling thing.  And I think we should be a little bit more interested in the complexity and the variability of animal communication systems and less rigid about this distinction between what is a language and what is not a language.