1898-99. | THE USE AND ABUSE OF PHILOLOGY,. gi 
istic is inevitably retained.* Secondly, this attempt at linguistic iden- 
tification must also be qualified a failure because ¢saa, even if supposed 
to signify “headgear” as in the case in question, cannot be compared to 
tsau, which is the Egyptian for “crocodile.” 
From this last remark we may deduce this corollary: in all philo- 
logical comparisons, both words, while homonymous, should also be 
synonymous. This is so evident that we need not insist. There is no 
lack of homonymous terms in all languages, and if the philologist’s busi- 
ness was merely to discover consonances, his task would certainly not be 
a very arduous one. It must be admitted, however, that there are some 
cases when this synonymy of homonymous words needs be but relative. 
As illustrative of the appropriateness of this qualification, I may point to 
the etymology of the English word “loafer,” which is said to come from 
the German /awfer, a runner, which is itself derived from /aufen, to run. 
Passing from the letters to the words themselves, we cannot help 
noticing that some of the latter are more ancient, more immutable, and 
simpler than others ; they reappear under a similar—though not neces- 
sarily identical—form in divers cognate dialects; in a word, they are the 
roots of the language. These are the essence of a dialect and, as far as 
practical, with them only should comparisons be attempted. But in this 
case care should be taken to choose only equally radical words for the 
purpose of identification. A living language is subject to inexorable 
laws of growth and mutations, and any resemblance between a modern 
accidental term and an old root of a different tongue must be the result 
of purely fortuitous coincidence. 
A rule of analogous import demands that test words be compared, as 
far as possible, only with synonyms from one of the oldest forms of the 
language, not from one of its modern derivative idioms. To render this 
principle clearer by contrast, I shall give an instance of an evidént 
violation of the same. Rev. C. Petitot, in an essay on the Déné 
languages,t gives the consonance between the Déné word aa, “he has 
said,” and the French a dz, as in some way confirmatory of the unity of 
race between the American and the European nations from whose vo- 
cabulary the two words are extracted, Now, it seems to me that the 
* In another paper, ‘‘Déné Roots,” published in the Transactions of the Canadian Institute (Vol. IT.), 
I have called attention to the absence of diacritical marks denotive of this explosion in the texts of the ‘‘Moun- 
tain Chant” by Dr. W. Matthews, hinting at the same time that, as the words which lack it are otherwise 
quite identical with their northern Déné equivalents, this most important peculiarity had possibly escaped the 
transcriber, and giving my reasons for this surmise. A copy of the paper sent to Dr. Matthews and accom- 
panied by a note pointing to that passage failed to elicit a declaration that his rendering of the Navajo texts 
was faultless. Shall we apply in this case the maxim : Quz tacet consentire videtur ? 
t Paris, 1876, p. xvi. 
