“Roots are the phonetic types produced by a force inherent in the human mind; they were created by nature,” etc., etc. Thus speaks Max Müller. Whitney, who is rarely of the same mind, says, notwithstanding, that all the Indo-European languages are descended from one primitive, monosyllabic language, “that our ancestors talked with one another in simple syllables indicative of ideas of prime importance, but wanting all designation of their relations.” In the other camp it is sustained that roots are the result of learned analysis, but that there is nothing to prove that they really existed (Sayce); that they are reconstructed by comparison and generalisation; that, e. g., in the Aryan languages, roots bear much the same relation to Sanskrit, Greek, and Latin words as Platonic ideas to the objects of the real world (Bréal). It has been calculated that the number of articulate sounds which the human voice is capable of producing amounts to three hundred and eighty-five. These sounds, for physiological reasons, constitute a fundamental theme in the various words created by man. Later on, linguists in comparing the vocables used in different languages, established the frequent recurrence of certain sounds common to several words. These have been isolated, but we must not see in them aught besides _extracts_. Moreover, “the first stammerings of man have nothing in common with phonetic types so arrested in form and abstract in signification, as _dhâ_, to place, _vid_, to see, _man_, to think, and other analogous words.” To sum up. In the first thesis roots come into existence, _ab initio_; words are derived from them by reduplication, flexions, affixes, suffixes, etc.; they are the trunk upon which a whole swarm of languages has proliferated. In the second thesis, words come first; then the common element disengaged by analysis, but which never existed as such in the pure and primitive condition.