[53] How were primitive terms (roots or words) formed? A much-debated and still unsolved question. Man had at his disposal one primary element, the interjection. By all accounts this remained sterile, unfertile; it did not give birth to words; it remained in articulate language as a mark of its emotional origin. A second proceeding was that of imitation with the aid of sound, onomatopœia. From antiquity to the present time, it has been regarded as the parent, _par excellence_. This was accepted by Renan, Whitney, Tylor, H. Paul, etc.; rejected by M. Müller, Bréal, P. Regnaud, etc. No one disputes the formation of many words by onomatopœia, but those who question its value as a universal process say that “if in certain sounds of our idioms we seem to hear an imitation of the sounds of nature, we must recollect that the same noises are represented by quite different sounds in other languages, which are also held by those who utter them to be onomatopœia. Thus it would be more just to say that we hear the sounds of nature through the words to which our ear has been accustomed from infancy” (Bréal). I have observed that those who study the spontaneous formation of language in children, claim for them but little onomatopœism. On the other hand, a word created by undoubted onomatopœia is sometimes by means of association, or of strange analogies, transferred successively to so many objects that all trace of the transformations of meaning may be lost, and the imitative origin actually denied. Such was Darwin’s case, cited above, where the onomatopœia of the duck finally served to designate all liquids, all that flies, all pieces of money. If the successive extensions of the term had not been observed, who could have recovered its origin? [54] Sayce, _loc. cit._, IV., §§ 3-5. [55] We cannot doubt, however, that there is in the child (and so too for primitive man) a period of pure and simple denomination, when, in the face of perceived objects, he utters a word, as a spontaneous action, a reflex, with no understood affirmation. But this act is rather the prelude, and attempt at speech, an advance towards language proper. [56] There is in Iroquois a word that signifies, “I demand money from those who have come to buy garments from me.” Esquimaux is equally rich in terms of this sort. Yet we must recognise that these immense composite words, themselves formed from abbreviated and fused words, virtually imply the beginning of decomposition. [57] _Lectures on the Origin and Growth of Religion_, ed. 1891, p. 196. [58] Whitney, _The Life and Growth of Language_, Chap. X. Sayce, _op. cit._, VI., 28, rejects them absolutely. [59] Renan, _Histoire générale des langues sémitiques_, pp. 128 and 363.