148 



EARLY HISTORY OF MANKIND. 



have appeared chimerical. Even our own language may 

 sometimes receive light from the study of distant mem- 

 bers of our family. Where, for instance, are we to seek 

 for the root of our comparative better 1 Certainly not in 

 its positive, good, nor in the Teutonic dialects in which 

 the same anomaly exists. But in the Persian we have 

 precisely the same comparative, behter, with exactly the 

 same signification, regularly from its positive beh, good."* 



The second great family is the Syro-Phcenician, com- 

 prising the Hebrew, Syro-Chaldaic, Arabic, and Gheez 

 or Abyssinian, being localized principally in the coun- 

 tries to the west and south of the Mediterranean. Be- 

 yond them, again, is the African family, which, as fa^as 

 research has gone, seems to be in like manner marked by 

 common features, both verbal and grammatical. The 

 fourth is the Polynesian family, extending from Mada- 

 gascar on the west through all the Indian Archipelago, 

 besides taking in the Malayan dialect from the continent 

 of India, and comprehending Australia and the islands of 

 the western portion of the Pacific. This family, how- 

 ever, bears such an affinity to that next to be described, 

 that Dr. Leyden and some others do not give it a distinct 

 place as a family of languages. 



The fifth family is the Chinese, embracing a large part 

 of China, and most of the regions of Central and North- 



* Wiseman's Lectures on the Connexion between Science and 

 Revealed Religion, i., 44. The Celtic has been established as a 

 member or group of the Indo-European family, by the work of Dr. 

 Prichard, on the Eastern Origin of the Celtic Nations. " First," 

 says Dr. Wisemen, " he has examined the lexical resemblances, 

 and shown that the primary and most simple w r ords are the same in 

 both, as well as the numerals and elementary verbal roots. Then 

 follows a minute analysis of the verb, directed to show its analo- 

 gies with other languages, and they are such as manifest no casual 

 coincidence, but an internal structure radically the same. The 

 verb substantive, which is minutely analyzed, presents more strik- 

 ing analogies to the Persian verb than perhaps any other language 

 of the family. But Celtic is not thus become a mere member of 

 this confederacy, but has brought to it most important aid ; for 

 from it alone can be satisfactorily explained some of the conjuga- 

 tional endings in the other languages. For instance, the third per- 

 son plural of the Latin, Persian, Greek, and Sanscrit ends innt, nd, 

 vti, vto, nti, or nt. Now, supposing, with most grammarians, 

 that the inflections arose from the pronouns of the respective per- 

 sons, it is only in Celtic that we find a pronoun that can explain 

 this termination; for there, too, the same person ends in nt, and 

 thus corresponds exactly, as do the others, with its pronoun 

 hwynt, or ynt." 



