Imagini ale paginilor


unacquainted, from having been so too. The relative pronoun in Chinese can be proved to have once been a substantive meaning “place,” and it would seem that the Hebrew relative 'asher had the same origin, 'asru in Assyrian, 'athar in Aramaic signifying “a place.” The Assyrian pronoun mala," as many as,” is merely a fossilized substantive meaning “fulness," and the Ethiopic lăli and cīyā, which, when combined with suffixes, express the nominative or accusative of the personal pronoun, really signified originally "separation ” and “entrails."1 The Malay ulun, I,” is still “a man" in Lampong, and the Kawi ngwang, “I," cannot be separated from nwang, “a man." In Japanese the same word may stand for all three persons; but this is because it was primitively a substantive, such as “servant,” “worshipper,” and the like. Even now the Chinese scholar will say, ts'ie (“the thief "') instead of “I," while tsián (“ bad") and ling ("noble ") are used for “mine” and “thine.”.

“ The inhabitants of Ceylon,” says Adelung, : " have seven or eight words to denote the second personal pronoun,” and Pott remarks that even German is still so much influenced by the habits of an earlier barbarism as scrupulously to avoid the employment of the second personal pronoun, recourse being had, where Er and Sie fail, to the uncivilized method of denoting the personal pronoun by means of a substantive. In Greek we find ode o avp used as the equivalent of "I," and a somewhat unsatis




[ocr errors]

1 Prätorius : “Z. d. D. M. G.” xxvii. 4 (1873).

Endlicher : Chines. Grammatik," pp. 258-89.
3 “ Mithridates," i. 233.
* “ Die Ungleichheit menschlicher Rassen,” pp. 5, 6.


factory attempt has been made to derive this pronoun itself, the Latin ego, the Sanskrit aham, from the root agh, “speaking,” which we have in the Latin ad-agi-um, "a proverb,” the Greek -, and the Gothic af-aik-an, “to deny.” However this may be, we must always bear in mind the possibility of tracing symbolic words to conceptual ones, and of discovering that what we have imagined to be the pronominal root is really a reduced and mutilated form. Above all, we must not fall into the mistake of confounding these pronominal roots with the classificatory suffixes, a mistake which has been perpetrated in the classification of roots as material and formal. It is perfectly true that some of the suffixes, such as -tar, or our own -ward, or the person-endings of the Aryan verb, can be referred to old nouns and pronouns; but what is true of some of them is not true of all, while even these suffixes are not identical with pronominal roots but belong to groups of words containing both pronominal and predicative roots.

are brought back to our startingpoint. Roots are the phonetic and significant types which underlie a group of words in a particular family of speech. Each family of speech has its own stock of roots, its own common heritage of words, which serve, like its grammar and its structure, to mark it off from every other family. We have seen how the various races of man have started with different grammatical conceptions and modes of constructing the sentence; they have equally started with different lexical types. Roots are for the dictionary what the mental ways of viewing the relations of the sentence are for grammar. Allied lan

And So


guages must agree in their roots as well as in their grammar.

But it is not necessary that the roots possessed by each member of a family of speech should all be the same. We find cases and case-endings in Latin which do not exist in Greek, while the Greek terminations in - Je and - Lev are equally unknown to Latin. Similarly in the vocabulary, one dialect may retain words which have been lost by another, or drop words which are in use in the remaining cognate tongues.

This is one of the causes of the difficulty experienced by etymologists in finding a derivation for every word in the lexicon, that is to say in settling the root to which it must be referred. Unless we have allied words in cognate dialects with which to compare our recalcitrant word, no etymological tact or scientific attainments will enable us to determine its roots and connections. The logicians tell us that we can draw no inference from a single instance; it is just as impossible to discover an etymology for an isolated word. But there may be other reasons for this impossibility besides the simple one that a word may be the last waif and stray of an otherwise extinct group. Languages borrow words from their neighbours, and it may very well happen that the word whose derivation we are seeking may be a foreign importation which has slightly changed its appearance in being naturalized. We know from Livy (vii. 2) and Festus that the Latin histrio (hister), “ a play-actor,” and nepos, “a spendthrift,” were borrowed from Etruscan, and the inscriptions have further informed us that the Latin Aulus was originally the

Ed. Müller, p. 165.


Etruscan Avile, “the long-lived one,” but there is little doubt that many words exist in Latin which were also introduced from Etruria, but of whose parentage our ignorance of the old Etruscan language forbids us to give any account. Maize and hammock seem genuine English words enough, but they have come to us through the medium of Spanish from the dialect of the natives of Hayti. To search for their etymology in the Aryan family of speech would be parallel to M. Halévy's endeavour to explain agglutinative Accadian from the Semitic lexicon. But there is yet a third reason for the existence of roots peculiar to only one out of a group of allied languages. Even in its most advanced and cultured state, language never wholly resigns its power of creating new words, and with them new roots. It is true that the inventions of the nursery are nipped in the bud or confined within the nursery walls ; it is also true that words like the Kafir angoca, mentioned before, could never be introduced into literary idioms like English and French; but it is also true that the native instinct of language breaks out wherever it has the chance, and coins words which can be traced back to no ancestors. The slang of the schoolboy, the argot of the large towns, Americanisms, and thieves' cant, all contain evidences that the creative powers of language are even now not extinct. The murderer Pierre Rivière invented the word ennepharer for the torture to which, as a boy, he subjected frogs, and the word calibène for the instrument with which he killed birds. Prince “Plon-plon" can be assigned no

Humboldt : “Travels” (Engl. transl.), i. p. 329. ? Charma : “ Essai sur le Langage” (1846), p. 66.


[ocr errors]



parentage, any more than the game of squails with its śwoggle and absquatulate. Du Mérit refers to the purely musical names given by children to those they are fond of, and Nodier tells a curious story to account for the origin of a lady's falbala. A witty prince of the last

? century, Marshal de Langlée, entered a shop with the intention of testing the assurance of the milliner in it. He therefore coined the word falbala on the spot, and immediately asked for one. The milliner at once brought him the dress called volant, which with its light floating points reminded her of the root involved in the newlyinvented word, and perhaps called up the sound and signification of folâtre or flotter. Even natural science has added to the stock of Aryan roots. To pass over Van Helmont's gas, Neckar invented sepal to denote each division of the calyx, Reichenbach the expression “Od force," and Guyton de Morveau the chemical terms sulfite and sulfate. Here, however, we have a reference to sulphur, just as M. Braconnot's ellagic acid, the substance left in the process of making pyrogallic acid, is merely galle read backwards. To find the process of word-making in full vigour, we must look elsewhere than to the scientific age. We have something better to do than to spend our time in inventing new words; that employment must be left to the disciples of Irving and other theological enthusiasts. The heritage we have received is large enough for our wants; our part is to

1 “Notions de Linguistique," p. 211.

? Falbala has been borrowed by most of the European languages under various forms, appearing in English as furbelow. It is first found in De Caillières (1690). 3 Whewell : "History of the Inductive Sciences,” ii. p. 535.

Whewell : op. cit. ii. p. 547.

« ÎnapoiContinuați »