Date of Award


Document Type


Degree Name

Master of Arts (MA)



First Advisor

Richard S. Pittman


From the introduction: "Based on [the] recorded data [of a natural conversation], rather than on suggested or controlled informant responses to out-of-context situations, a procedure will be described by which the larger grammatical sequences (i.e., sentences and clauses) of Dakota may be delimited from the larger speech continua in which they are imbedded.

"1.3 There are today two main views regarding the fundamental procedure to be followed in the linguistic analysis of a corpus of data, each with its group of adherents. One group considers that the analyst will obtain best results, save time, and produce a more accurate final description if he begins first with the identification and description of the smaller units, and then proceeds through word, to phrase, then to clause, and finally to discourse.

"The other offers identical rewards to those who apply this procedure in reverse, i.e., who begin by cutting the discourse first into paragraphs, then sentences, phrases, until the ultimate building blocks of grammar--the morphemes--are isolated.

"1.4 The following sections of this paper will describe the method of elicitation used in obtaining the language data utilized herein, the phonology, and the procedure by which the sentence and clause boundaries within this Dakota discourse can be determined.

"The final section will present a phonemic transcription of the conversation text, with a literal translation; a free translation of the same text; and an inventory of particles and inflectional morphemes occurring throughout the data."

Included in

Linguistics Commons