Lastly a reference may be made to the fact that children have shown themselves capable of inventing the rudiments of a simple kind of language. Professor Horatio Hale of America has made a special study of these spontaneous child-languages. One case is that of twin American boys who when the talking age came employed not the English sounds that they heard others speak but a language of their own. Another, and in some ways more remarkable case, is that of a little girl who at the age of two was backward in speaking, only using the names ‘papa’ and ‘mamma,’ and who, nevertheless, at that age, and in the first instance without any stimulus or aid from a companion, proceeded to invent a vocabulary and even simple sentence-forms of her own, which she subsequently prevailed on an elder brother to use with her. The vocables struck out, though suggesting some slight aural acquaintance with French—which, however, was never spoken in her home—are apparently quite arbitrary and not susceptible of explanation by imitation.[73] ----- Footnote 73: For a summary of Professor Hale’s researches see Romanes, _Mental Evolution in Man_, p. 138 ff. ----- I think the facts here brought together testify to the originality of the child in the field of linguistics. It may be said that in none of these cases is the effect of education wholly absent. A child, as we all know, is taught the names of objects and actions long before he can articulate. Thus Darwin’s boy knew the name of his nurse five months before he invented the vocable ‘mum’. It is obvious indeed that wherever children are subjected to normal training their sign-making impulse is stimulated by the example of others. At the same time the facts here given show that the working of this impulse may, in a certain number of children at least, strike out original lines of its own independently of the direct action of example and education. What is wanted now is to experiment carefully with an intelligent child, encouraging him to make signs by patient attention and ready understanding, but at the same time carefully abstaining from giving the lead or even taking up and adopting the first utterances so as to bring in the influence of imitation. I think there is little doubt that a child so situated might develop the rudiments of a vocal language. The experiment would be difficult to carry out, as it would mean the depriving of the child for a time of the advantages of education.[74] ----- Footnote 74: Of course, as Max Müller says (_The Science of Language_, i., p. 481 f.), the facts ascertained do not prove that ‘infants _left to themselves_ would invent a language’. The influence of example, the appeal to the imitative impulse, has been at work before the inventions appear. Yet they do, I think, show that they have the sign-making instinct, and might develop this to some extent even were the educative influence of others’ language removed. -----