¹ Cree is a member of the Algonquian language family.

The narratives in the text were recorded by Mr. Joel Linklater of Fort: Albany, Ontario, on the west coast of James Bay, in July, 1958. Thanks. are also due to Mr. Alan Wheatley, Principal, Moose Fort Indian Residential School, for his help, and to Mr. Andrew Faries of Moose Factory,. Ontario, for assistance in the transcription and translation of the texts.

The conventions of Cree syllabic spelling have been followed in writing personal prefixes and aspect markers separately: e.g., mo·na ma·ka n’ki·ohci nipaha·na·n. Since they really form part of the larger word unit it would be better to write, mo·na ma·ka n’ki·-ohci - nipaha·na·n. - and so consistently throughout.

² Part I, Preliminary Edition, Summer Institute of Linguistics, Glendale, California, 1954.

The author wishes to express sincerest thanks to Dr. K. L. Pike for many helpful suggestions in the preparation of this paper.

³ Pike, K. L. Cf., “On Tagmemes, née Gramemes”, IJAL 24. 4 1958: 273–78.Google Scholar

⁴ Bloomfield, L., Language, N. Y., 1933; p. 166–7 Google ScholarPubMed. For Bloomfield the tagmeme had “structural meaning”, i. e., was a “feature of meaningful, arrangement”. Cf. C. F. Hockett, Lg. 23 1947:321 fn. 4.

⁵ Cf. “Language, etc.”, as per fn. 2 above. Of interest also in showing the beginnings of divergence from Bloomfield is Pike, K. L., “Taxemes and Immediate Constituents”, Lg. 19 1943:65–82.Google Scholar

⁶ It is hoped to do this at the earliest opportunity. This paper is, in effect, a sort of first run over the ground, although, in the light of the restricted material, all formulae must be treated as strictly tentative.

⁷ Cf. Sacred Stories of the Sweet Grass Cree, Nat. Mus. of Canada, Bull. No. 60, Ottawa, 1930; p. 6.

In the present text unmarked vowels are short. Long vowels are marked with a raised dot, with the exception of /e/, which is always long.

⁸ A permissible, tentative hypothesis on the analogy of 2.1.1.d’, q.v.

⁹ A safe assumption on the basis of other TA — TI parallels, although only an inanimate object is attested.

¹⁰ Cf. L. Bloomfield, “Algonquian”, LSNA, p. 112, ff.

¹¹ Description in terms of zero (#), is preferred for three reasons:

• 1)

  1) Zero may be regarded as a convenient allomorph for a 3rd person, pronominal indicator, {o w-} (Cf. o- mo·soma — his moose; wiyaw — his body.), inasmuch as the 1st and 2nd person possessor prefixes are the same in shape as the subject-marking prefixes.

• 2)

  2) In the Plains Dialect, {o} is known to occur also as a subject prefix with certain preterite forms.

• 3)

  3) The use of the zero allomorph for a 3rd person subject indicator makes for a more symmetrical description.

¹² The use of +, #, is properly restricted to an “occurrence” diagramme which, as the name suggests, is concerned primarily with the occurrence or non-occurrence empirically of items in any given tagmemic formula. The ‘order” diagram makes no commitment as to whether any given collation of items may or may not occur in an actualized formula, taut is concerned rather with showing the sequence required, if they should occur. Hence commas are more appropriate between tagmemes than the +, #, in a strictly “order” diagram. Throughout this paper (except where otherwise specified), +, #, should be read only in the light of restrictions imposed by actual, textual manfiestations of any formulae. Cf. NOTE to Clause-type 1.

¹³ This is, in fact, one of the commonest expressions in colloquial Cree: — ekwa·ni, “that’s all”, or “that’s it”.