152 



nants involved ;i kind of palatalization of the preceding 

 vowel. This is of great significance for the vocalic colouring 

 of the Eskimo language. Tlie short sounds fa] and \u\ 

 become [«] and fv], the long sounds [a-J and fy] are 

 diphthongized, the former more strongly, the latter slightly; 

 [«■'] or [«'J does not sound exactly like a diphthong because 

 a retains its deep pitch until the last moment before the on- 

 glide of the following consonant ; the modiâcation which a 

 thereby undergoes lies on the way toward an /, but it often 

 gets no farther than to e or ä, thus either \üc] or \de, ä-ä\, 

 seldom {ù'j\. As a common symbol for all these glides I gener- 

 ally use [«•'], as in: [tama'H]^ [аЧ-а'Ч]^ [гкгЧге]'^ [tama''sa\* 

 [sa'Hut]^. The finer shades of this diphthongizing of a will 

 be given in g 26. \>j'] is lightly palatalized and diphthongized 

 in \t'jOs-'jt\ ^. 



Besides there may everywhere be found examples of a and 

 и which are lightly palatalized |e, 0] by a following t, n. к or s. 

 as far as a is concerned, also by A:, 7] (and m. j>?); even if I 

 have written [a] or [u] in these cases and left the palatalization 

 unindicated, still it must as a rule be understood. 



Palatalized consonants (like those in Kussiau), I have not 

 met with in Greenlandic. 



1 have yet to mention the uvular glide-sound which 

 is caused by the uvularization of vowels, namely the one which 

 after a short vowel introduces the stopped consonant q. I 

 consider it to be identical with the voiced fricative y, as in : 

 [e''5'e] ' \t'^(£t^q'dq\^. which I generally designate in a different 

 manner, thus: [eç'é] [er^er^o^]. This glide-sound gets an 

 independent value in the consonant-group [r^|, which is no 

 doubt merely a reduced [^'j (cf. § 14|. 



' whole, entirely (3 pers. sing.) ^ not until now ' once (formerly) * all 

 (3 pers. plur.) * place-nanae * a large needle used in covering kajaks (with 

 skin). ' corner, corner of the month * the little finger 



