without further explanation, so that we may say that the deaf and dumb make use of the very same gesticulations which were customary in the pantomimic intercourse of the Indians from Hudson's Bay to the Gulf of Mexico. Thus by means of his usual signs a deaf and dumb Englishman was able to make himself understood by some Laplanders at a show. Finally, it is said (though the statement is subject to grave doubts) that the unfortunate Laura Bridgeman, a blind deaf-mute, cut off from all external instruction, used the ordinary pantomimic movements. 16

Thus at the period of the first development of speech, there were a number of expedients for the communication of thought, while at the same time, as man is of all creatures the most sociably disposed, necessity urged him in some way or other to make himself intelligible to his neighbour. Yet it is still difficult to explain the first attempt at speech. A purpose of communicating an idea to another person by means of the vocal organs must not be assumed, for that would imply a consciousness on the part of the speaker that a sound would serve to communicate an idea. Even if the first speaker had connected a particular sound with a particular idea, yet as any sound may be connected with any idea, he had no prospect of being understood." Any elucidation of this obscure process would be inconceivable, had not each of us been at one time obliged to work himself up from a speechless condition. In speech, each child is obliged to repeat the experiments of mankind, only that in his course of development a great number of intermediate stages are passed over by the aid of instructors. The awakening of the power to comprehend speech, and the creation of speech, may therefore be observed anew in every child. L. Geiger rashly asserts that no new words can be invented. Young America ought to have taught him the contrary. The party name Locofoco, the name of a secret society Kluklux, the sectarian name Mormon, are arbitrary inventions. Schurlemurle, as a beverage of mixed wines is named in Würzburg, and picnic, can scarcely be derived from older expressions. Any one who has watched children must be astonished

16 Tylor, Early History of Mankind, pp. 21, 44, 69, 86.

17 Steinthal, Psychologie und Sprachwissenschaft, pp. 84, 370. Berlin, 1871.

"Mama and Papa."

III

at the doubt whether articulate sounds can be combined into new groups. 18 In South Africa the inhabitants of barren districts quit their settlements for a time, leaving the children under the care of a few aged people. The young ones forthwith begin to make a language of their own; the more lively are followed by the less developed, so that in the course of a single generation the nature of a language may be altered in this manner. Two words which are echoed in every language were created by children, and are created anew by every child, namely, the sounds papa and mamma. The elementary sound ma or pa is by no means an attempt to speak, but merely an exercise of the vocal organs proceeding from an inward physical impulse without purpose or consciousness, in no way better or higher than the twit-twit of our chaffinches. But as long as man has wandered on this earth, parental love in blissful delusion has misunderstood the child as if a call had been intended, as if the child were yearning for its father or its mother. That these first exercises of the vocal organs determined the sound of the future word, whereas the interpretation of the parents determined the meaning, is shown by the fact that, in a certain number of languages, the sound ba stands for father, and ma for mother, while in an equal number the converse is the case. 19 Other childish words for mother are aithei (Gothic) and atta (Sanscrit), the latter applying also to the elder sister. Atta exists also in Latin and in Greek, and also in Gothic, as an endearment for father, whence comes also the term aette for grandfather in German dialects. The lisping child has to pass 20 through various stages in the comprehension of language; for it must first learn by experience that when it cries ba or ma the parents approach or that pleasure is given to those present; the sound is then for the first time purposely uttered by the child, and it is not till much later, and not without the aid of the parents, that one sound is used as a call for the father and the other for the mother. Months and even years pass by before it dawns upon the perception that mama and papa are not proper names, but

18 Max Müller, Science of Language, vol. ii. p. 54.

19 A list of names for father and mother in every part of the world is to be found in d'Orbigny, l'Homme américain, p. 79.

20 Bacmeister in the Allgem. Zeitung. 1871.

that with all children they are the designations of their nurses and guardians. Only at a later age does the child further discover that these names are given to its progenitors, and their full and true purport is understood even by adults only when they have experienced the joys and anxieties of parenthood. The course of development in tender years is thus approximately, if not completely, similar to the first attempts to speak made by our race.

The richness of a language is always determined by the need for communication; and we must suppose this to have been very small in the earliest evolutionary stages of our race. The English boast of a vocabulary of 100,000 words, but their fieldlabourers are said to be satisfied with 400. A clergyman in a Friesland island states that he could reckon no more in the case of a workman of his parish. As Kleinpaul21 informs us, a man of average education has from 3000 to 4000 different words at his disposal, a great orator 10,000, while in the institutions of the deaf and dumb, at Berlin, no less than 5000 signs are employed. That the number of expressions increases with the need for expression is shown by the numerical terms, which among barbarous nations seldom extend beyond 20. Alexander von Humboldt was the first to trace the origin of numerical groups of 5, 10, and 20 units to the number of the parts on the hands and feet, so that with six-fingered hands we should have arrived at the duodecimal system. 22 Exceptions exist, however, especially in the Australian family, which make use of only two numerical terms; thus for I is said netat; for 2 naes; for 3 naes-netat; for 4 naes-naes; for 5 naes-naes-netat; for 6 naes-naes-naes. 23 Other Australian dialects have an independent expression for three, and in one linguistic region of those parts, the numerical terms reach as far as 15 or 20.24 Orton maintains that the Zaparos on the Napo river in Ecuador can count only up to three, but express higher numbers by raising the fingers; 25 and the Prince of Wied 26

21 Zeitschrift für Völkerspsychologie und Sprachwissenschaft, vol. vi. p. 354Berlin, 1869.

22 Life of A. von Humboldt, edited by Carl Bruhns, vol. iii. p. 9.

23 Latham, Opuscula, p. 228. 24 Tylor, Primitive Culture, vol. i. p. 220.

25 James Orton, The Andes and the Amazon, p. 170. 1870.

26 Reise nach Brasilien, vol. ii. p. 41. Frankfort, 1825.

Vocabularies.

declares the same of the Botocudos.

113

Closer research might,

however, reveal more favourable facts respecting most of the nations mentioned, for it also has been disputed that numerical terms above three exist among the Abipones. In reality, however, they say, instead of four, "ostriches' toes;" for five they use two expressions, for ten they say "fingers of two hands," for twenty, "fingers and toes on hands and feet." 27 We ourselves have no expression for ten thousand such as exists in Greek, nor for a hundred thousand (lak), or for ten millions (kror), such as exist in Hindostanee, the richest language of the world in expressions for high numbers, reaching as far as 51 figures, owing to the fact that these terms were employed in many ways by the Sankhjâ philosophers and the Buddhists in their numerical juggleries. The word million was unknown to the nations of the classical age, while the term milliard has come into circulation only in this century.

A comparison of languages of scantily developed races shows that the perception of specific differences arose much earlier than the recognition of generic characters. Savage hunting tribes have names for the beaver, wolf, and bear, but none for animal. 28 The Tasmanian languages are wanting in expressions for tree, fish, and bird; but there is no lack of appellations for the individual species.29 The same may be said of the North American Indians, for in the Chocta language, while there are names for the white, red, and black oak, there is none for the genus oak. When we take nourishment, whether it be soup, bread, meat, vegetables, or porridge, we always use the word eat, but the Hurons vary the expression according to the nature of the food. 30 The Eskimo again, have particular expressions for fishing, depending on the implements employed.31 The

27 Dobrizhoffer, Geschichte der Abiponer, vol. ii. p. 202.

28 Greek has also no word for animal in so far as (wov includes man, for which reason the song, "Mensch und Thiere schliefen feste " (men and animals were sleeping fast), cannot be translated into Greek. Steinthal, Zeitschrift für

Völkerpsychologie, vol. vi. p. 480. 1869.

29 Lubbock, Prehistoric Times, p. 466.

30 Charlevoix, Nouvelle France, vol. iii. p. 197. Paris, 1744.

31 Latham, Varieties of Man, p. 376.

Malays distinguish between red, blue, green, and white, but they have no word for colour. The Tasmanians have no adjectives, so that they say "stone-like" instead of hard; "moon-like" instead of round; "long-legged" instead of high.

Languages are variously provided with sounds. The Arabs are destitute of the clicking sounds of the Hottentots, and we are deficient in many Arabic consonants, but the greatest paucity is found in the South Seas. The Polynesians have only ten consonants at their disposal, f, k, l, m, n, ng, p, s, t, v, and even these exist in full purity and completeness only at Fakaafo and Vaitupu,32 while the inhabitants of the Tupuai group to the south of Tahiti have preserved only eight, m, n, ng, p, r, t, v, and one with a marked guttural sound.33 A like paucity of sounds has arisen in the Sandwich Islands by deterioration, and is not primitive and simple, for other Polynesian languages, which have remained richer in consonants, have preserved the more archaic forms. If with this be connected the fact, that the enunciation of the Bushman language, especially owing to its clicking sounds, imposes the greatest exertions on the vocal organs, we might be induced to conclude that in the primordial attempts at speech a greater stock of sounds was brought into use.34 Still there are scholars who maintain the opposite,35 so that a universally valid rule must not as yet be laid down.

II. THE STRUCTURE OF HUMAN LANGUAGE.

THE foreign languages, whether ancient or modern, which we Europeans study during our school years, all possess a greater or less number of grammatical forms, by the help of which a definite function in the sentence is allotted to the radical sounds. This

32 See Gabelentz, Die melanesischen Sprachen in the transactions of the phil. Gesellschaft der Wissenschaft, p. 253. Leipzic, 1861.

33 Hale, Ethnography, p. 142.

34 W. H. J. Bleek, Ueber den Ursprung der Sprache, p. 53. Weimar, 1868.

35 Whitney, Language and the Study of Language, p. 467: "The tendency of phonetic change is always towards the increase of the alphabet.”