LANGUAGES OF THE INDO-EUROPEAN FAMILY. 299 



ation even where the radical m or w is missing, as in the Anglo-Saxon, 

 ^, the Frisian fyf, the Dutch vijf and our English ^ve, which follo-w 

 the analogy of the Gaelic and Irish coig and cuig. The Coptic five, 

 Tou, cannot help us here. Such cases, however, are no more to be 

 accepted as ofi'ering opposing testimony to those which vouch for the 

 truth of the general principle here illustrated than were the Irish 

 criminal's ten witnesses, who sought to negative the evidence of ten " 

 men that had seen him commit the crime for which he was being 

 tried by stating that they had not. , 



Without referring to Semitic roots I may instance some additional 

 examples among Indo-Euroj>ean words of the presence of the Coptic 

 article. The Sanskrit udan, the Greek hudor, the Gaelic and Irish 

 ad, signifying water, have thrown off what the old Phrygian retained 

 in bedu and the Slavonic in voda. Another Sanskrit word pavaka, fire, 

 on the other hand retained the article, while the Latin focus and the 

 Gothic hac rejected it; but the Sanskrit urana, goat, becomes the 

 Lithuanian haronas, as the Greek rhigos and orego are transformed 

 into the Jjxiii\ frig us and porrigo. Bopp is quite right when ho says 

 "the Latin Rog (rogo, interrogo) appears to be abbreviated from 

 Frog."^* This is seen in the Sanskrit ^racA and the German fragen 

 both meaning to ask. Another instance in which the Sanskrit shows 

 an afiinity with the Aeolic and Sabine dialects of Greek and Latin is 

 afforded by the word pum, a man, the Latin homo. The Welsh oer 

 and the Gaelic and Irish, fitar, cold, the Greek phren and the Latin 

 renes, the English rap and the French frapper, the Greek husteros 

 and the Latin 2^osterus, the Welsh oes and the Greek bios, the 

 English order and the German fordern, completely set at notight 

 every law of phonetic change forming part of the physical science of 

 language in the attempt made by such means to account for their 

 differences. The science of language has a place among the historical 

 as well as among the physical sciences ; and its historical element is 

 as distinct from the physical as ax*e the objects of Paleontological 

 from those of Mineralogical study, the fossils from the mere strata 

 in which they are imbedded. Following out the analogy, we may 

 compare the subjects of our present philological researches to the 

 Crinoids of many formations, some of which are still attached, or 

 may we not say articulated, to the old Coptic foundation, while 

 others, that once occupied the same position, have floated free, and 



** Bopp's Comparative Grammar, i., 116. 



