- A language spoken by about 75,000 Indians in the southern part of the state of Oaxaca. Data were gathered during field trips to the town of Juchitán in the years 1944–50, under the auspices of the Summer Institute of Linguistics. The author wishes to express appreciation to Kenneth L. Pike for his assistance in the preparation of this paper at the University of Oklahoma during the summer of 1950, and to Charles F. Hockett for valuable constructive criticism given at Cornell University in the current academic year.
- This paper was read, in abridged form, at the meeting of the Linguistic Society in Ann Arbor, July 29, 1950.
- The consonant phonemes of Isthmus Zapotee are: voiced fortis L, N, W; voiceless fortis p, t, k, s, š; voiced lenis n, l, w, y, r (apical flap); voiceless lenis h (actualized as a homorganic voiceless nasal or lateral preceding voiced nasal or lateral), f (occurring only in borrowed words),? glottal stop, occurring as part of the syllable nucleus; voiced lenis varying freely to voiceless b, d, g, z, ž; voiced lenis m, varying to fortis in word-medial position; rare phonemes r (apical voiced trill) found in only three native-origin words but now being introduced in borrowed words, and b, a bilabial voiced trill found in only one word: berenb', 'ant lion'. The vowel phonemes are i, e, u, o, a. The tone phonemes are high and low, high symbolized with an acute accent, low left unmarked. For further discussion of tone, see below.
- In the above example and all nonbracketed examples, phonemic notation is used except for those specific features which are the subject of the discussion. Thus the nonphonemic stress is regularly written, and in certain cases vowel letters are phonetic rather than phonemic. In certain transcriptions of tone, it will be obvious from later statements that the transcription is not even phonetic, but reflects what we thought we heard in early stages of the language.
- We have discovered one—and only one—instance of a sequence of three like vowels. In this instance, as with the sequence of V1V2V2 there is a perceptible syllable break between the first and second moras, therefore the second and third moras are treated as a complex syllable nucleus and as such receive the stress according to the regular rule. Examples: na'il? 'sour', bi'ii? 'became sour'.
- We are considering a sequence of two vowels separated by a glottal stop to be the nucleus of a single syllable. The reference is not to a phonetic syllable, however, but to that which has been termed a "phonemic syllable" (e.g. see Kenneth L. Pike, Phonemics, pp. 65,90). These nuclei comprise units of timing [compare Einar Haugen, "Phoneme or Prosodeme?" Language 25.378 (1949)], taking approximately the same timing as a sequence of two oral vowels, and act the same as to tone alternation and stress placement.
- In addition, we are considering the sequence of vowel plus glottal stop as a syllable nucleus since such an interpretation simplifies the statements regarding stress placement and consonant cluster distribution. We have then six types of syllable nuclei, the first simple, the remainder complex: V, VV, V?, V?V, VV?, V?V?.
- The distribution of stress is based on grammatical words (i.e. isolatable groups of morphemes which may occur as whole utterances). These grammatical words may also coincide with rhythmic breaks within larger utterances, but this phenomenon has not been studied. In the remainder of this paper we are relating the analysis of "phonemic" and "nonphonemic" stress to the above definition of "word", regarding such word boundaries as a basis for phonemic analysis. This is, of course, in accordance with only one trend of phonemic theory. At the close of the paper the conclusions will be restated in terms of a different phonemic theory.
- Clusters of two unlike vowels are not as common but pattern the same as like vowels, i.e. as complex nuclei.
- We describe a sequence of tones on a complex nucleus by hyphenating the descriptive words, and a sequence of two single nucleus syllables with space between the words.
- This type of bi-phonemic interpretation of a single segment is not entirely new to the field. It has been done in such instances as the abstraction of the phonetic syllabicity of /n/ in English "button" as the vowel phoneme /ə/. A similar type of phenomenon may possibly be found in the occasional pronunciation of English /t/ and /d/ in utterancefinal position. In this position the occlusion of the two may occasionally appear homophonous and voiceless, but the length of the vowel preceding the phoneme /d/ must be abstracted as part of the consonant phoneme, leaving it in contrast with the phoneme /t/, which does not have as part of its actualization any preceding vocalic length.
- For a dialect of Zapotee in which normal tone contrast is taken over by length at one minor point in the pattern, see Eunice Pike, "Problems in Zapotee Tone Analysis", I.J.A.L. 14.161 (1948). A similar type of analysis has been done by William E. Weimers in Senadi, a language of northern Ivory Coast [Language 26.1 (1950)]. Welmers finds it convenient to interpret a combination of voice consonant with low tone as a consonant followed by a vowel with low tone, the vowel being a "zero alternant" of the final /ì/.
- Recent acoustic data [See Martin Joos, Acoustic Phonetics, p. 105 (supplement to Language 24.2, 1948)] shows that no segment is completely isolated but is always affected by the preceding and following segments. Phonemic analysis of that segment must abstract these preceding and following slurs as part of that phoneme.
- The final statement covers all of the loan words found in our present data. We assume, however, that with further acculturation additional Spanish loan words may ultimately bring in a conflict with the native pattern and introduce stress contrasts which at present do not exist.
- Choose new content alerts to be informed about new research of interest to you
- Easy remote access to your institution's subscriptions on any device, from any location
- Save your searches and schedule alerts to send you new results
- Export your search results into a .csv file to support your research