THE following texts in the Tsimshian dialect of the Tsimshian language were written down by Mr. Henry W. Tate, a full-blood Indian of Port Simpson, British Columbia, in Tsimshian, with interlinear translation, according to the alphabet adopted by Bishop Ridley in his Tsimshian translations of the Gospel (published by The Society for Promoting Christian Knowledge). This material was revised by me, with the assistance of Mr. Archie Dundas, a full-blood Tsimshian from New Metlakatla, Alaska. Apparently some slight differences in dialect have developed between the Tsimshian of the older people who staid in British Columbia, and the younger generation who migrated to Alaska. The phonetics, as given here, are those of Archie Dundas.
The following alphabet has been used to represent the sounds of the Tsimshian.
a short a with a strong leaning towards ê, which depends largely upon the following consonant. Before m, n, w, the a is fairly pure. Before l, l, k*, it is almost ê.
â long sound, always pronounced with retracted lips, and therefore more like ti.
b, d distinctly sonant, but more strongly articulated than in English.
E obscure weak e, as in flower.
ë with glide towards i.
g* distinctly sonant, anterior palatal, with affricative glide towards y, more strongly articulated than English g.
g distinctly sonant, middle palatal, like English g in good, but more strongly articulated.
g the same, velar.
h as in English.
i, î continental i.
î, open i, is in hill.
k*, k*! surd and fortis of g*.
k, k! surd and fortis of g.
q, q! surd and fortis of g.
l sonant 1, with full glottal articulation and long continued.
l! the same, with great stress of articulation.
m as in English.
m! the same, with great stress of articulation.
n with fuller glottal articulation than in English.
n! the same, with great stress of articulation.
o, ô as in short and long
ô like o in German voll.
â like aw in law.
p, p! surd and fortis of b.
r a very weak, strongly sonant middle palatal trill.
s the tip of the tongue is turned up and touches the palate just behind the alveoli. The teeth are closed, and the air escapes laterally. The sound effect is intermediate between s and sh.
t, t! surd and fortis of d.
u, û like oo in root.
w as in English, but more strongly sonant.
w! the same, with greater stress of articulation.
X velar aspirate, like ch in German Bach.
y as in year, but more strongly sonant, with full breath.
y! the same, with greater stress of articulation.
dz, ts, ts! affricative sonant, surd, and fortis, with purer s sound than the s described before.
0
indicates parasitic vowels which accompany some short and all long vowels. These are glottal stops with the weakened timbre of the preceding vowels. â0, for instance, sounds almost like äA (where A indicates a very weak a), â0 like âA, i0 like îi, ê0 like êI. After short vowels, the sound resembles the simple glottal stop.- connects proclitics and stems.
= indicates typographic division.
In the following texts the grammatical forms given by Archie Dundas have been given preference over those of Mr. Tate, who tends to substitute the forms peculiar to direct discourse for those characteristic of indirect discourse. Thus, he writes,--
lukdî'dEt dîl lgû'0lgEdit (instead of lukdî'dagA dîl lgû'0lgEtga0) she and her daughter were hungry (p. 72, lines I and 2).
nâ'0kE hanâ'0x galgâ'lxdEt dEda nE-wî-la'kdEt (instead of nâ'0kE hanâ'0xga0 a galgâ'lx gEsgE nE-wî-la'ktga0) the woman lay with her back to the great fire (p. 72, line 18).
ada lat sagait-dâ'dEt da dEm dô'xdEt (instead of ada lat sagait-dâ'ga0 asgE dEmt dô'xtga0) then, when she gathered them in order to take them (p. 74, lines 4 and 5).
It is in accord with this tendency that he omits very often the terminal -ga0 indicating absence, which Dundas uses regularly. The same tendency may be observed in the texts dictated to me by Matthias, a native of Old Metlakatla in 1886, and printed in A. C. von der Schulenburg's Grammar; 1 while a text dictated to me by Mrs. Morison, a native of Port Simpson, which may be found at the same place, 2 does not show the same tendency. Examples taken from Mr. 'Fate's manuscript are:--
Ada la dEm ksE'rEt (instead of ksE'rEtga0) when she went out (p. 72, line 27). da g*îk yâ'0 hanâ'0x (instead of hanâ'xga0) then the woman went again (p. 74, line 16).
In other cases Mr. Tate uses the indicative where Dundas prefers the subjunctive.
gaksta! wâ'0gA wî-mExmê'0 (instead of gakstat wâ'0sgA wî-mExmê'0) behold! he found a large grouse (p. 74, line 13).
wula wâ'lga wula wula-dza'bEdEt (instead of wula wâ'lsga) thus did the hunters.
He also often omits the possessive -t of the third person, and the demonstrative terminal -t.
In the written texts, inaccuracies of construction creep in easily. I should have preferred a revision of the texts with Mr. Tate, but this was not feasible.
I published a text with interlinear translation and grammatical notes in the "Zeitschrift für Ethnologie," 1908, pp. 776-797. A grammatical sketch of the language is contained in the "Handbook of American Languages." 3
FRANZ BOAS.
69:1 Die Sprache der Zimshîan Indianer, pp. 188 et seq.
69:2 Ibid., pp. 183 et seq.
69:3 Bulletin 40 of the Bureau of American Ethnology.