perors, or the four largest fruits. But this is not all. The number four is still easy to transform into images, but that is no longer the case when we come to higher numbers, such as seven, and, with still more reason, 20, 100, 1,000, etc. Yet the large numbers are not more difficult for us to conceive than the small ones. This is because we represent them by conventional signs, or the figures.
We must not, however, forget that some savage peoples can not count beyond four or five. Sir John Lubbock tells in his paper an anecdote of Mr. Galton, who, on one occasion, made a comparison of the arithmetical comprehension of a Damara savage of South Africa and a little dog. According to Mr. Galton, the comparison was not to the advantage of the man.
Let us now examine Sir John Lubbock's experiments. He wrote on his cards such words as go, bone, water, food, etc., in phonetic orthography, so as not to trouble his dog's head with the difficulties of English spelling; also words without significance to the animal, such as simple, nothing, ball, etc.; and he had cards with nothing written on them.
Van the dog soon learned to distinguish the blank cards from the written ones; then he learned to attach an idea to some of the latter; and finally was able to fetch to his master the card that corresponded with his wish. To get a single meal he had to fetch some eighteen or twenty of these cards, and he made no mistakes.
Sir John Lubbock concluded from this success that Van had learned to read. In one sense, this conclusion is absolutely false, but that is not the sense in which Sir John regarded the matter. In another sense it was true, and this is the point on which we need light.
There was never a dog whose master has not said and thought a thousand times that he only lacked speech. In fact, the dog seems to comprehend speech, and speaks in his expression. His eyes behind which, according to Madame de Staël's fine expression, he seems to conceal a human soul—interrogate, supplicate, and answer; his ears are erected, or lop over; his tail wags, and his whole body assumes marked attitudes, not to be misinterpreted, of desire, joy, attention, anger, repentance, fear, shame, and submission. Could he better express all of his feelings if he spoke? Should we understand him any better if he should say to us when he had been guilty of some misdeed, "I deserve to be punished, but don't, I pray you, be too hard on me," or if, after he had been corrected, he should politely thank us for our moderation? We perceive at once the distance between his language and ours. One is natural, the other conventional.
Does he understand our conventional language? He does, and he does not, but in the more exact sense he certainly does not. He understands us when we give him our usual orders: "Down!" "Come here!". "Go back!" "Give me your paw!" "Now, the other one!" "Seek it!" "Bring it here!" "Get out!" But we forget that
