SWANTON] TUNICA, CHITIMACHA, AND ATAKAPA LANGUAGES 11 
PHONETICS 
In this discussion the phonetics must be regarded merely as a 
means to an end. Of course nothing was to have been expected in 
the line of phonetic exactness from any writer earlier than Gatschet, 
and Gatschet’s system was by no means adequate to express the 
sounds which American languages contain. There is reason to believe, 
however, that even in his time the phonetics of the three varieties 
of speech under consideration had broken down very considerably, 
and a further lapse of 30 years has not improved them. Atakapa 
and Tunica are particularly disorganized; Chitimacha is preserved 
somewhat better, but it is not possible to furnish a satisfactory 
detailed table of phonetics. We have here only certain approxima- 
tions. 
The system adopted is based, so far as possible, on the simpler 
system for the “phonetic transcription of Indian languages’ con- 
tained in the report of the committee of the American Anthropological 
Association—which had the duty of attempting some unification in the 
work of American philologists—which was published as volume 66, 
No. 6, of the Smithsonian Miscellaneous Collections. The second 
method of recording vowels has been adopted: @ as in English 
father, @ as in fare, @ as in final, @ as in hat, @ as a in fate, e as in met, 
7 as In pique, 7 as in pin, O as in note, o almost as in not, @ as inrule, wu 
asin put. In the few cases where nasalization of vowels occurs it is 
indicated by a hook placed beneath, .. The consonants are so poorly 
distinguished that most of thesigns willbesufliciently described by say- 
ing that they are to be pronounced nearly like the hard English sounds 
represented by the same letters. In Chitimacha, however, we have 
a series of intermediate stops, for which I have employed the common 
surd signs p, t, and k, and a series of surd aspirates which I have 
written p‘, ¢‘, and k‘. There appear to have been two corresponding 
affricatives, tc and tc‘, but they are now difficult to distinguish. In 
addition to these we have s, the dental sibilant, c, the prepalatal sibi- 
lant, and z, the palatal spirant. #—like ng in smg—is not found in 
Tunica but occurs in both Atakapa and Chitimacha. It seems to vary 
between the nasal sound accompanying a vowel and the sound of m. 
Tunica alone contains a velar r, and Atakapa alone a bilabial f, of 
rare occurrence, which Gatschet sometimes writes v. In Atakapa we 
also find an initial ¢/, which probably stands for an original surd J, 
the one generally written ¢. In the same language ts takes the place 
of tc, but in Tunica we find exactly the reverse condition, while 
Chitimacha contains both. Chitimacha and Tunica employ both 
s and c, which appear at present to be somewhat confounded. Ata- 
kapa employs only c. Besides its use in Chitimacha in the cases 
already mentioned we sometimes have occasion to use ‘ to indicate 
