4 INTR0J)UCTrON TO TMK DIOTIONAKY. 



This method is applicabk; to tho preparation of a dictionary ; but in 

 writing texts of the language, everj^ word and sound nuist be hiid down as 

 it flows from the lips of the native informant, pure and unaltered, in these 

 laniiiiaiies, (;very sound of the current speech is, or may be, significant ; 

 iiiHliaiii;(i(l ))y imaginary phonetic rules derived from the study of literary 

 languages, every word and syllable should be, as it were, photographed 

 \\ith its peculiar short, long and duplex, clear and obscure vowels, drawl- 

 ings and sto])s of the voice, noises and clangs. The law of accentuation, as 

 observed in the language of the MAklaks, sustains this principle in a singular 

 manner ; for in this western tongue accentuation is much more a syntactic 

 than a morphologic feature, the position of the accent being very generally 

 determined by the run of the sentence. There are but a few polysyllabic 

 words that never shift their accent. 



One of the manifold consequences of following fanciful phonetic rules, 

 often engendered by the desire of using as few types as possible, is the 

 arbitrary suppression in literary publications of sounds existing in a lan- 

 guage. Thus the Mohawk dialect of Iroquois is repi'esented to have twelve 

 oi' thirteen sounds only, while in reality it has no less than twenty-six. 

 Iiulian texts can convey their full meaning only by accurate phonetic 

 transcription ; and when they pass down to posterity in this shape, as a 

 true and faithful monument of the tribe who produced them, others may 

 discover phonetic or other laws of the language which our studies have failed 

 to reveal to our own understanding. 



After the al)solute form of the word, the Dictionary gives in the majority 

 of cases its distributive form, derived from it by what is called distributive 

 reduplication. When the various phonetic modes of forming the distribu- 

 tive from the absolute form have been studied attentively in the Grammar, 

 the absolute form will readily suggest itself when looked for in the Diction- 

 ary, though in some of the more difficult cases indications are given to help 

 the reader in his search. Many no?nina adoris and other terms occur in the 

 distributive form alone, or are more frequent in this than in the absolute 

 form. 



The definitions are presented in their etymologic order, which is the 

 order of their historic evolution. It is true that in many instances the form 



