ro 



and use of them.'''"- That they had letters before they became 

 general in Britain may be evidently inferred from the following* 

 facts that the Cornish had 21 letters exclusive of double ones, 

 and the Armoricans all the letters of the French or English al- 

 phabets, except W and Y. 



Prior to the intercourse of the Romans wiih the Celts, the 

 extent of their understanding seemed limited to a degree scarcely 

 credible. The mental powers of the wild Indians were perhaps 

 more comprehensive. The most uncivilized of these had names 

 for numbers and understood decimal arithmetic. The former, 

 on the otiier hand, seemed incapable of enumerating their herds, 

 flocks or armies ; and it would also appear, that the lapse of 

 time into minutes and hours, and that of days into weeks and 

 months, passed almost unnoticed. The same source, whence the 

 Roman letters were received, seems also to have supplied all the 

 Celtic tribes with the novel ideas of numbers, their use, and also 

 of the division of time. These inferences are deducible from the 

 Celtic names, which are derivatives from the Roman ones ; not the 

 latter from the former, as some etymologists would persuade us; 

 for, as the earliest liistory of the Celtaj is the history of savages, 

 instruction could not be communicated by ignorance. The opin- 

 ion originated chiefly from the apparent absurdity of supposing 

 those Celtic tribes so uninformed as facts evince. With the excep- 

 tion of those words derived from the Latin, and a few others, which 

 happen to have a resemblance in sound and sense, of which the 

 greater part are found in one dialect only, and some in two, no lan- 

 guages can be more distinct. If they approached as nearly as the 

 Greek in resemblance to the Russian, which nevertheless is not a 



191. The antiq of Irel. p. 107. 



