March 1, 1888.] 



KNO^A^LEDGE 



115 



EVOLUTION OF LANGUAGE. 



By Ada S. Ballix. 



IX.— DEMOXSTRAXIVE AND PRONOMINAL ROOTS. 



IROFESSOR MAX MULLER has distinguished 

 two kinds of roots — the predicative, denoting 

 either existences or ;ictions, and the demon- 

 sfrafive, marking positions or localities of 

 existences or actions. Thus, taking as a con- 

 crete example the root eat, which simply 

 expresses an action, and the roots /lere, 

 there, i/onder, expressive of positions in which an ac- 

 tion may take place, we can, using them in a Chinese 

 fashion, get the sentences : Eating here = I eat ; eating 

 there = thou eatest ; eating yonder := he eats, which 

 would be expressed in gesture language by imitating the 

 movements of mastication and pointing to oneself for " I 

 eat," to one's interlocutor for " thou eatest," and to the person 

 referred to for " he eats," or, in the case of the absence of that 

 person, by making the sign used to represent him. 



I have discussed as exhaustively as space permitted the 

 first class of these roots described in a former article * — 

 namely, those signifying action or quality, whence by far the 

 greater number of words is derived — and I have shown the 

 probable origin of these " units of speech. ' It is now time 

 to consider the second and less important class, composed of 

 roots which signify position. There are not more than a 

 dozen of these roots, and some of them were probably 

 derived from one original. From this class come the 

 interrogatives, roots generally containing k, the demonstra- 

 tives, containing f or n. the personal pronouns, possessives 

 and relatives, with the adverbs of position and direction. 

 It is very noticeable that suchlike words are not confined to 

 any one family of languages. 



It has been suggested, as Whitney says, not without some 

 degree of plausibility, that the pronominal roots in Indo- 

 European and other languages have a sort of physical 

 raison d'etre. Ma, our me, is produced by closed lips, as if 

 shutting out the rest of the world, while the demonstrative 

 ta, our that, is produced by the thrusting forward of the 

 tongue in the mouth, as if to point at the object signified ; 

 and the same suggestion may be made with regard to Semitic 

 pronouns, as ani, Hebrew and Arabic, " I "; atlu ov ant, 

 '■ thou," which, as a final enclitic, becomes ta ; asher or 

 Arabic hatha, " that " ; also Arabic thalik. 



On some such theory may perhaps be explained the 

 resemblance of Aryan and Semitic pronouns, as, for 

 example, Hebrew hoo, Arabic hooa, " he " ; Hebrew ham^ 

 (pronounced like " lame "). Arabic humma, " them." 



Our primitive fathers, like children and savages, did not 

 distinguish between object and subject. The accusative of 

 the first person historic;illy preceded the nominative ; hence 

 ma, mi are attached to the first person of the verb, not ego, 

 aham. The Sanskrit aham, according to Benfey, grew 

 out of an older form, maham. The nominative plural in 

 Lith. is iiie~s, in 0. Prus. mes, in O. Slav, mu, in Ossetic 

 mach, in Lapp mi, in Pali mayam, and in Greek and 

 Latin there are the verbal endings men and mus, as in 

 tuptomen, scribimus. The change of in into vj is by no 

 means rare. A gentleman from Cierman Switzerland told 

 the late Professor Key that where he lived all the people 

 said mir sagen in preference to loir sagen ; and in the 

 Lithuanian the nominative and accusative ve-du and mu- 

 du coexist. The Sanskrit plural nominative is vayam, 

 formerly doubtless mayam, as in Pali. The Zend is vaein, 

 and the Gothic veis. The interchange of the liquids m and 



» Vol. ix. p. 286. 



71 is very common. In Sanscrit in the oblique cases of the 

 dual and plural, we find 7idn and mis, in the Zend dual no 

 and ne, Greek dual ndi, Latin plural nominative nos. 

 Although the Lapp language is not supposed to belong to 

 the Indo-European family, its pronouns closely resemble 

 them. Thus, according to Fiellstrum t : — 



The Nom. is mon, I todu, thou soden, he 



„ Gen. „ mo, of me to, of thee so, of him 



From parne, son, comes parnam, my son ; from nipe, knife, 



niput, thy knife ; from aija, grandfather, aijahs, his 



grandfether. In a related language, Tcheremissian, we 

 find i :— 



ata-m, my fiither ischlene-m, 



ata-t, thy „ ischtene-t 



ata-sha, his „ ischtene-she, 



ata-na, our ,, ischtene-ne, 



ata-da, your „ ischlene-da, 



ata-shl, their „ | ischtene-sht, 



Eask § gives the Lapp pronouns as follows ;- 



Nom. mon, I don, thou 



PI. mi, we di, ye 



Gen. mo, of me di'i, of thee 



Yhael i| gives the Finn pronouns as : — 



Nom. mina, 1 sina or tdm.a, thou 



PI. me, we te, j'e or you 



Gen. minun, mine tiimdn, thine 



I do 



thou doest 

 he does 

 we do 

 you do 

 thev do 



son, he 

 SI, they 

 si'i, of him 



hiin or se, he 

 lie, they 

 hanen, his 



Out of a list of ninety negro languages, in more than 

 seventy the pronouns of the first person consist of w«, man, 

 me, mi, inu, na, ne, 7ii, in, or ;« and n as a prefix, and m, 

 n, or nj is used in a similar way in Asia, Siberia, and 

 America. 



The plural of the first personal pronoun is natui-ally an 

 extension of the singular, and whether we find it with an 

 a, a ic, or a v, we can easily trace its original form to the 

 radical consonant m, with its idea of inwardness or sub- 

 jectivity. Similarly with the second person in forms with 

 (/, th, dh, or .~, we trace the original t — the thrusting forward 

 of the tongue as in pointing to a person or object ; and 

 hence we find a close relationship existing between the 

 second personal pronoun and the demonstrative. The 

 Sanscrit neuter nominative tat = this. T/iis, that in 

 Arabic is dha and dhii ; Hebrew, zeh. The Slavonic 

 languages have demonstrative pronouns beginning \vith t, 

 like the Greek to, tov ; in Mongol ere this cori'esponds to 

 tere, th.at, like our here and there, and oiiha and touba = this 

 and that. The Lapp demonstrative is daat or dat. The Sanscrit 

 relative and interrogative is ka-s, with the dhdtii, kinn, Ionic 

 Kov, Kij, KOTcpos; guichat and quirhere =^ our ivhat and tohere. 

 Our ichicli is from a former irhiik. In Portuguese we find 

 5Mem srt6e .?" who knows ? " Spanish qiden sabe ? Sardinian 

 chini, interrogative and relative: Turkish kirn or kim, with 

 a variety kih : while the Finnish has cu-ca (Lat. quisque) 

 and ken-kfi as interrogatives, ciiin as a relative, and cu-mpsi 

 as the equivalent of the Latin nter (formerly cuter). Old 

 Slav, has the forms chlso or cheso for interrogation, and 

 Mongolian ke7i ? who, which. One form of the interrogative 

 in Lapp is gi, " who ? " gen. geen, " whose t " g superseding 

 the Finnish k. In Chinese inten-ogative adverbs the k 

 sound predominates — ki-to, " how much ? " kl-td, " how 

 great?" khshl, "at what time?" "when?" sien-M-nien, 

 " how many years ago ? " k'l at the beginning of a sentence 

 is interrogative (quomodo), and in Chinese literature hi, 

 ho, and yen are used as interrogatives. The striking 

 resemblances of distinct families of speech with regard to 



t Gram., p. 32. 

 § Eask's Gr., p. 79. 



t Ibid., pp. 20, 21. 

 II P. 41. 



