182 



Essay towards a Grammar of the Berber Language. 



Perhaps Ejcdom was found too similar In sound to Yaxdom, when this last 

 had been vulgarized into Ixdum : and the imperative is also nearly the same. 

 Hence the Maltese say Naxdom, for the first pers. sing, and in the plural, 

 Isdomu, Taxdomli, Najrdomh, Equally by the form Qa.rdamar, the Ber- 

 bers avoid the awkward ambiguity of Arabic, in which "she does" and 

 "thou dost" are the very same word. 



But besides these similarities of verbal inflection, we have the pronouns : 

 I, &c. We, &c. Thou, &c. Ye. 



Berber 

 Shemitic 



Some might think it paradoxical to maintain, that the two families of 

 languages drew from a common source ; and that the Berbers have retained 

 the older forms. Yet h priori this is not at all incredible ; and in some of 

 the forms there is the appearance of it. Nakhi, pi. Nakni ; compared 

 with Natta (for primitive Na^Qa), pi. NoQni, appears regular ; but Anoki, 

 pi. Enaxnu, is less so, and has no Hebrew parallel. Ani (Hebrew), and 

 Ana (Arabic), appear corruptions of Anoki. Nay may seem older than 

 Na. On comparing the Coptic pronouns Anoki (I), Anon (We), with 

 suffixes i and n ,- and in the second person, suffixes and verbal inflexions, 

 K (in sing.). Ten (in plural) ; our prevailing conviction is, that various 

 African languages must originally have had some of these points in common 

 with the Shemitic. 



Moreover, in Berber and in the Shemitic tongues, an and in (mm) are 

 plural terminations ; which seem common to nouns and verbs ; just as in 

 Turkish they form Adam (a man), Adam/rtr (men) ; and Baqdi (he looked), 

 Baqdi/«r (they looked). In some dialects of Celtic, in forms a plural j 

 nay, and many German and Anglo-Saxon plurals are thus formed. These 

 coincidences weaken our argument, by inspiring a suspicion that n is so 

 easy of utterance, that it is not too much to suppose the coincidences acci- 

 dental. The same suspicion recurs, when we find adjectives formed by 

 adding an in Berber and in Arabic, and in Latin (-anus), &c. 



It remains to add ; that verbals are formed by prefixing m, alike in Cop- 

 tic, Arabic, and Berber; while in the same three languages, ^ or 9 indicates 

 a feminine, sometimes prefixed, sometimes suffixed, in nouns and in verbs. 



In Coptic it is notorious, that initial 9 has this force, being also the fe- 

 minine of the article. In Hebrew, however, it seems to have been less 

 acknowledged than it deserves, that or < (for we need not distinguish the 

 letters in Hebrew, though we must in Arabic) is the real and sole mark of 

 the feminine. Indeed the termination a often appears to have this force ; 

 but this a is nothing but a corruption of at. This is proved ( 1 ), by the 



