In "Introduction To Objectivist Epistemology", Ayn Rand gives an explanation of an experiment in which a man walks into a forest and some nearby crows fly away into the tree tops. They do not return to the forest clearing until the man leaves. Two men walk in, and then leave with the same results. Three men walk in and leave, same results. But, when five men walk in and only four walk out, the crows return.
In ITOE, Rand also explains that:
My question is: How did the crows understand "2", "3", and possibly "4" (though they obviously didn't grasp "5")? Why didn't they get confused after "1"?
asked Jan 01 '11 at 16:29
David Lewis ♦
The crows didn't actually use numbers. But beyond that, I don't believe the research exists to prove exactly how crows realize only three hunters have left. I can provide a speculative answer that is consistent with general principles of perception and memory at the animal level of cognition.
Pairs of things are symmetrical in the simplest sense. The power of simple symmetry to higher cognition cannot be over-stated. What is entirely random becomes, when mirrored, as in a Rorschach ink-blot, meaningful. Pairs don't need to be counted, 1,2, in order to be recognized as being a pair, they are their own kind of thing. "Two-ness" is a perceptually-given meaning.
Consider triplets next. A triplet is hard to confuse with a pair, yet it is also distinctively symmetrical, as its extensive appearance and use in culture proves. The three Virtues, the Holy Trinity, etc. Endless figures of speech testify to our ability to make one "chunk" of three things. Three dots are a triplet, a single, distinct thing, or a triangle, the simplest polygon. We tend to automatically perceive any three of the same thing as a three-part unit.
We suppose these intelligent crows are capable of this level of cognition also. It is a level of abstracting from perceptions involving three of something, the triplet structure itself. It remains a perceptual abstraction, not mediated by the sort of serial matching operations that constitute primitive counting.
Humans can recognize more complex configurations, but crows, perhaps, can only go that far or a little farther. The answer, then, is that the crows recognize pairs of men (or other like things) as a specific type of thing, and even triplets of men as another specific type of thing. But larger configurations are all non-specific pluralities of men, and equivalent to one another. As such, they don't prevent the bird from holding that the men are gone when only four (one of his non-specific pluralities) have left.
Man's perceptual abstractions are also limited, and that is why concepts and that serial matching oepration we call counting become so important. It is a whole different ball game when concepts and language become involved, despite the fact that we have the same set of senses, memory, and abstractive ability as most animals do. And this, of course, is why the crow epistemology comes up in Objectivist epistemology in the first place.