Hostname: page-component-cd9895bd7-gvvz8 Total loading time: 0 Render date: 2024-12-26T07:02:19.521Z Has data issue: false hasContentIssue false

Gitksan

Published online by Cambridge University Press:  02 March 2016

Jason Brown
Affiliation:
School of Cultures, Languages and Linguistics, University of Aucklandjason.brown@auckland.ac.nz
Henry Davis
Affiliation:
Department of Linguistics, University of British Columbiahenry.davis@ubc.ca
Michael Schwan
Affiliation:
Department of Linguistics, University of British Columbiamichael.schwan@alumni.ubc.ca
Barbara Sennott
Affiliation:
Urban Gitxsan Societybarbaraharris@shaw.ca
Rights & Permissions [Opens in a new window]

Extract

Gitksan (git) is an Interior Tsimshianic language spoken in northwestern British Columbia, Canada. It is closely related to Nisga'a, and more distantly related to Coast Tsimshian and Southern Tsimshian. The specific dialect of Gitksan presented here is what can be called Eastern Gitksan, spoken in the villages of Kispiox (Ansbayaxw), Glen Vowell (Sigit'ox), and Hazelton (Git-an'maaxs), which contrasts with the Western dialects, spoken in the villages of Kitwanga (Gitwingax), Gitanyow (Git-anyaaw), and Kitseguecla (Gijigyukwhla). The primary phonological differences between the dialects are a lexical shift in vowels and the presence of stop lenition in the Eastern dialects. While there exists a dialect continuum, the primary cultural and political distinction drawn is between Eastern and Western Gitksan. For reference, Gitksan is bordered on the west by Nisga'a, in the south by Coast Tsimshian and Witsuwit'en, in the east by Dakelh and Sekani, and in the north by Tahltan (the latter four of these being Athabaskan languages).

Type
Illustrations of the IPA
Copyright
Copyright © International Phonetic Association 2016 

Gitksan (git) is an Interior Tsimshianic language spoken in northwestern British Columbia, Canada. It is closely related to Nisga'a, and more distantly related to Coast Tsimshian and Southern Tsimshian. The specific dialect of Gitksan presented here is what can be called Eastern Gitksan, spoken in the villages of Kispiox (Ansbayaxw), Glen Vowell (Sigit'ox), and Hazelton (Git-an'maaxs), which contrasts with the Western dialects, spoken in the villages of Kitwanga (Gitwingax), Gitanyow (Git-anyaaw), and Kitseguecla (Gijigyukwhla). The primary phonological differences between the dialects are a lexical shift in vowels and the presence of stop lenition in the Eastern dialects. While there exists a dialect continuum, the primary cultural and political distinction drawn is between Eastern and Western Gitksan. For reference, Gitksan is bordered on the west by Nisga'a, in the south by Coast Tsimshian and Witsuwit'en, in the east by Dakelh and Sekani, and in the north by Tahltan (the latter four of these being Athabaskan languages).

The primary reference on the Gitksan language is Bruce Rigsby's Gitksan Grammar (Rigsby Reference Rigsby1986). Earlier work on the phonetics and phonology of the language includes Rigsby (Reference Rigsby1967, Reference Rigsby1986), Wickstrom (Reference Wickstrom1974), Hoard (Reference Hoard, Cook and Kaye1978), Ingram & Rigsby (Reference Ingram and Rigsby1987), Rigsby & Ingram (Reference Rigsby and Ingram1990), and more recently Brown (Reference Brown2008a, b, Reference Brown2010). The Gitksan orthography presented below was first developed by Bruce Rigsby and Lonnie Hindle in their Short Practical Dictionary of the Gitksan Language (Hindle & Rigsby Reference Hindle and Rigsby1973), with orthographic contributions from Powell & Stevens (Reference Powell and Stevens1977) in their language learning textbooks.

This Illustration provides an outline of the more prominent features of the phonetics and phonology of Gitksan. Further details of the language can be found in Rigsby (Reference Rigsby1986) and Brown (Reference Brown2008a). The data presented here are based on the speech forms of two female Gitksan speakers: Barbara Sennott and the late Doreen Jensen, sisters who grew up speaking Gitksan with their parents at home, regularly spoke Gitksan growing up in the village of Kispiox, and the surviving sister continues to do so with other speakers of the language today. The speech of both sisters is representative of the Eastern dialect. The examples presented throughout the text are spoken by Ms. Sennott, with two tokens for each form. The narrative at the end of the text is told by Mrs. Jensen.

Consonants

Gitksan has a rich set of consonants, including a set of glottalized plosives and affricates, and a set of glottalized sonorants. In the notation used here, glottalization is indicated following the segment for plosives and affricates, and preceding the segment for sonorants. This convention partially reflects timing properties, not necessarily a fundamental difference in segment type with respect to ejectives versus glottalized sonorants (see Carlson, Esling & Fraser Reference Carlson, Esling and Fraser2001 for the phonetics of Nuuchahnulth, and Howe & Pulleyblank Reference Howe and Pulleyblank2001 generally), and is consistent with the representation of glottalization in the Gitksan practical orthography.

While there do not exist many true minimal pairs, the vast majority of consonants can be shown to contrast in word-initial, prevocalic position. Exceptions are [x] and [xʷ], which are partially neutralized with [j] and [w] respectively in word-initial position: the former set only occur before obstruents and the latter set only before vowels. Also, the voiceless plosives and affricates are allophonically voiced prevocalically, thus leaving only pre-consonantal or word-final forms as examples below (voiced allophones will be transcribed throughout). There are no instances of [ˀl] in word-initial position (see Krauss & Leer Reference Krauss and Leer1981), which may be an accidental gap.

The plain ‘velar’ plosives and fricatives are usually phonetically prevelar in their place of articulation, though the labialized velar versions are produced slightly more posterior at the velum. Velar versions of the plain stops occur only before [s], [ɬ], which has led Rigsby (Reference Rigsby1986: 157–159) to suggest that the underlying forms are the prevelars, with a process of velarization that takes place in the environments before an [s] or [ɬ] (see Tarpent Reference Tarpent1987 for a similar treatment of Nisga'a). The prevelar variants will be represented with the superscript [ʲ].

The affricate [ ’] is rare in the language, in contrast to other language families in the Pacific Northwest, where the segment type is fairly common. The glottalized plosives and affricates are characterized by glottal closure preceding the oral closure, with ejective allophones in word-initial position. Also in contrast to other language families of the area, including most languages of the Na-Dene stock (Athabaskan and Tlingit) to the north and east, as well as the Wakashan and Salish languages to the south and west, is the lenis nature of these ejective allophones, which can cause them to be perceived as voiced stops and affricates by researchers (see discussion in Ingram & Rigsby Reference Ingram and Rigsby1987, Rigsby & Ingram Reference Rigsby and Ingram1990). These lenis ejectives are also incidentally found in the adjacent Athabaskan language Witsuwit'en (Wright, Hargus & Davis Reference Wright, Hargus and Davis2002, Hargus Reference Hargus2007). There is an obvious connection between the tendency towards decomposition of glottalized stops into glottal-stop–plosive sequences and their lenis quality. The sonorants are uniformly preglottalized, even in word-initial position. While the raised glottal stop [ˀ] has been used to indicate glottalization for sonorants, the implementation can range from a full glottal stop to creaky voicing during the sonorant (Um Reference Um1998: Chapter 5).

The list below includes examples of each consonant in the environment of a low vowel. Some consonants, namely the voiceless plosives and affricates as mentioned above, do not appear in prevocalic position, and so words were selected in which they appear in postvocalic position. In the list, voiced allophones of stop and affricate phonemes are also presented and exemplified (e.g. [p] followed by [b]). In the orthography, underlining indicates a uvular consonant.Footnote 1

Vowels

The phonological inventory of vowels for Eastern Gitksan is given above. The inventory for Western Gitksan is slightly different, as it can be argued that an additional long/short contrast for the mid vowels is historically emerging (Rigsby Reference Rigsby1986: 202).

There is a great deal of variation in the production of vowels; for instance, the mid front vowel space often overlaps with that of the high front vowels, and the high back vowels can be found in a more high-central position. Variation in low vowel production ranges from the back of the vowel space to the front. The vowels which are phonologically contrastive in stressed position (all except /ə/) are listed here:

The reduced vowel /ə/ appears in affixes and some function words, and is subject to a great deal of colouring by the neighboring consonantal environment. Adjacent to laryngeal or uvular consonants, /ə/ surfaces as a low vowel [a]:

In most other contexts, it tends to surface as [ɪ]:

Since schwa exhibits such variation (including further variation, to be discussed in the next section), it has been placed in an idealized central location in the vowel chart above.

The qualities for each of the above vowel examples (including the allophones of schwa), as well as the examples illustrating the vowel length contrast below, are illustrated in Figure 1, with ellipses defining one standard deviation.

Figure 1 F1 × F2 plots for Gitksan vowels. Ellipses define one standard deviation. The grid lines mark intervals of 750 Hz for F2 and 200 Hz for F1.

Vowel length is contrastive, and there are minimal pairs to illustrate this:

There are short mid vowels that occur as allophones of the long vowels in positions preceding a sonorant (Rigsby Reference Rigsby1986: 199–203). This results in allophonic short mid vowels in Eastern Gitksan (Rigsby notes a few exceptions with either short mid vowels not preceding a sonorant, and long mid vowels preceding a sonorant). The following alternations illustrate this shortening:

Unlike nearby Tlingit, vowel length does not interact with vowel quality, e.g. giving rise to a subsidiary tense–lax distinction (e.g. as reported by Boas Reference Boas1917; Maddieson, Smith & Bessell Reference Maddieson, Caroline and Bessell2001 characterize the short vowels as being more centralized). The lack of a vowel quality enhancement based on this contrast makes the distinction more difficult to perceive for researchers working on the language.

Conventions

There are numerous instances of consonant–vowel interactions in the language. As mentioned above, there is a pervasive rule of voicing, whereby plosives and affricates that precede a vowel become voiced. This is illustrated with the following sets of alternations:

While the voicing process affects both plosives and affricates, it does not affect fricatives. Hence, the process provides evidence for the class of plosives and affricates in the language (i.e. as a phonological class of ‘stops’). While the fricatives are immune to the prevocalic voicing that affects plosives and affricates, there is a lenition process that affects the voiced uvular plosive [ɢ], optionally rendering it as a voiced uvular fricative [ʁ]; for example, bogabaga ‘kiss’ is optionally realized as [ˌb oɢaˈb aɢa] ~ [ˌb oʁaˈb aʁa]. This optional lenition only affects the uvular plosives (see Rigsby Reference Rigsby1967, Reference Rigsby1986: 154).

There is also a process that shifts the timing of lip rounding from labialized velar plosives to a following vowel. As Rigsby (Reference Rigsby1986: 162–164) has pointed out, sequences of an underlying labialized plosive followed by a high front unrounded vowel surface as a plain plosive followed by a high back rounded vowel. The same process applies to schwa, as illustrated below with the alternation found in gipaykw ‘fly’ and gipaygum jixts'ik ‘airplane’. In this example, the epenthetic schwa that occurs between the root and the suffix is rounded to [u].

There is thus neutralization between high front vowels, high back vowels, and schwa following a labialized plosive.

There is pervasive spirantization of velars present in the language, which is also the most prominent phonological feature of the Eastern dialects versus Nisga'a and the Western dialects. As pointed out by Rigsby (Reference Rigsby1967, Reference Rigsby1986), this spirantization is especially evident in postconsonantal plosives, which become fricatives: /q a l-k sə/ [ɢa l x sə] ‘through a corridor or passageway’.

Finally, there is a process of vowel lowering in the language. Within roots, only low or mid vowels are allowed adjacent to uvular and laryngeal consonants (with some exceptions; see Brown Reference Brown2008a). In morphological contexts, such as affixes and reduplicants, this consonantal effect on the vowel is grammaticized, resulting in an active lowering process whereby a vowel of any height will become a low vowel (see Tarpent Reference Tarpent1983 and Shaw Reference Shaw, Bosch, Need and Schiller1987 for similar observations on Nisga'a). This is illustrated with the prefix /sə-/ ‘pick, gather’:

The same phenomenon can be shown with reduplication, where reduplicant vowels (reduplicants are prefixes) surface as [u] adjacent to (underlying) labials, as [a] adjacent to uvulars or laryngeals, and as [ɪ] elsewhere.

Stress

Rigsby (Reference Rigsby1986: 213–217) is responsible for the primary observations on Gitksan stress. He notes that in lexical words (verbs, nouns, and adjectives), stress falls on the rightmost vowel of the root. Examples include [ɢaˈns] ‘dog salmon’ /qən iːs/ and [ˈl aːɢa l d iˀj] ‘I examined it’ /l aːql-tə-ˀj/ (examine-?-t-1sg.ii) (Rigsby Reference Rigsby1986: 213). In compounds stress is found on the rightmost root-vowel of the rightmost member of the compound. In contrast, in preverbs, which have some lexical content, stress is on the leftmost vowel. Phrasal stress in Gitksan falls on the rightmost root-vowel of the head word. Suffixes are invisible to stress assignment (Rigsby Reference Rigsby1986: 216 notes only two exceptions).

Syllable structure

The word-level prosodic structure of Gitksan has been previously treated in Wickstrom (Reference Wickstrom1974) and most completely in Rigsby (Reference Rigsby1986). The syllable template of Gitksan allows for clusters of consonants in onset or coda position, though it does not allow for adjacent heterosyllabic vowels, or major processes of diphthongization (with vowel hiatus resolved through glide formation). Some examples of words with initial consonant clusters include:

Three-member clusters and larger are also found. The following clusters are in word-final position, which, because of consonantal suffixes, is a richer context for clustering:

An example of a larger cluster, involving five consonants in word-final position and derived from affixation is [ ɪl k s tɬ d a w] / i l k s-tt a w/ (melt-pass=cn ice) ‘the ice is melted’. More detailed discussion of consonant clusters and syllable structure constraints can be found in (Rigsby Reference Rigsby1986, Brown Reference Brown, Christodoulou and Lyon2008b).

Plosive–sonorant clusters (in that sequence), whether heterosyllabic or tautosyllabic, are nonexistent in the lexicon of the language. The mirror image sonorant–plosive sequence is, however, acceptable, whether across a syllable boundary or tautosyllabically in syllable coda position:

This results in a curious gap in consonant cluster sequences in the language. The ban on stop–sonorant clusters does not follow from the sonority hierarchy in onset position (as stop–fricative and fricative–sonorant clusters are allowed), and thus differs from neighboring Salish languages (which strictly obey the hierarchy), as well as from Wakashan and neighbouring Athabaskan languages, which generally allow no complex onsets at all (aside from Witsuwit'en, which does allow limited complex onsets; Hargus Reference Hargus2007: Chapter 19).

Transcription

The narrative text that follows is an adaptation of ‘The North Wind and the Sun’, which could be appropriately titled ‘The Wind and the Sun’, spoken by the late Doreen Jensen. The broad transcription and the orthographic version (with morpheme breakdowns) of the text follow.

Broad transcription

Orthographic transcription with interlinear English gloss

The top line of each entry is an orthographic representation. The second line indicates where the affix and clitic boundaries are within words.

Acknowledgements

Many thanks go to our Gitksan consultants and teachers, especially the late Doreen Jensen. Thanks to Molly Babel, Clarissa Forbes, Karsten Koch, Zoe Lippsett, Douglas Pulleyblank, Alyssa Satterwhite, and especially to Bruce Rigsby for commenting on an earlier draft. Thanks also to two anonymous reviewers and to Adrian Simpson for extensive comments and guidance. Funding for this project was made possible by a Phillips Fund for Native American Research grant and a Jacobs Fund grant awarded to the first author. Any errors are the responsibility of the authors.

Footnotes

1 Morpheme glosses are based on the Gitxsan Online Dictionary conventions, and are as follows: i = Series I person marker, ii = Series II person marker, iii = Series III person marker, 1 = first person, 2 = second person, 3 = third person, antip = antipassivizer, assoc = associative, attr = attributive, ax = A = (transitive subject) extraction marker, caus = causative, cl.cnj = clausal conjunction, cn = common noun connective, cntrst = contrastive, comp = complementizer, dem = demonstrative, detr = detransitivizer, dist = distal, distr = distributive, emph = emphatic, foc = focus, incep = inceptive, indp = independent, ipfv = imperfective, neg = negation, nmlz = nominalizer, pass = passivizer, ph.cnj = phrasal conjunction, pl = plural, pn = proper noun connective, prep = preposition, prosp = prospective aspect, restr = restrictive, sg = singular, sx = S = (intransitive subject) extraction marker, t = ‘T’ suffix, tr = transitive, vi = intransitive verb, vt = transitive verb. A hyphen (-) marks an affix boundary and an equals sign (=) a clitic boundary.

References

Boas, Franz. 1917. Grammatical notes on the language of the Tlingit Indians (University of Pennsylvania Museum Anthropological Publications 8(1)). Philadelphia, PA: University of Pennsylvania Museum.Google Scholar
Brown, Jason. 2008a. Theoretical aspects of Gitksan phonology. Ph.D. thesis, University of British Columbia.Google Scholar
Brown, Jason. 2008b. An unexpected gap in Gitksan consonant cluster phonotactics. In Christodoulou, Christiana & Lyon, John M. (eds.), Papers for the 43rd International Conference on Salish and Neighbouring Languages (University of British Columbia Working Papers in Linguistics), 1421. Vancouver: University of British Columbia.Google Scholar
Brown, Jason. 2010. Gitksan phonotactics. Munich: Lincom Europa.Google Scholar
Carlson, Barry F., Esling, John H. & Fraser, Katie. 2001. Nuuchahnulth. Journal of the International Phonetic Association 31 (2), 275280.Google Scholar
Hargus, Sharon. 2007. Witsuwit'en grammar: Phonetics, phonology and morphology. Vancouver: University of British Columbia Press.Google Scholar
Hindle, Lonnie & Rigsby, Bruce. 1973. A short practical dictionary of the Gitksan language. Northwest Anthropological Research Notes 7, 160.Google Scholar
Hoard, James E. 1978. Obstruent voicing in Gitksan: Some implications for distinctive feature theory. In Cook, Eung-Do & Kaye, Jonathan (eds.), Linguistic studies of Native Canada, 111119. Vancouver: University of British Columbia Press.Google Scholar
Howe, Darin & Pulleyblank, Douglas. 2001. Patterns and timing of glottalisation. Phonology 18 (1), 4580.Google Scholar
Ingram, John & Rigsby, Bruce. 1987. Glottalic stops in Gitksan: An acoustic analysis. 11th International Congress of Phonetic Sciences (ICPhS XI), Talinn, Estonia, 134–137.Google Scholar
Krauss, Michael E. & Leer, Jeff. 1981. Athabaskan, Eyak, and Tlingit sonorants (Alaska Native Language Center Research Papers 5). Fairbanks: Alaska Native Language Center.Google Scholar
Maddieson, Ian, Caroline, L. Smith & Bessell, Nicola. 2001. Aspects of the phonetics of Tlingit. Anthropological Linguistics 43 (2), 135176.Google Scholar
Powell, J. V. & Stevens, Russell. 1977. Gitxsanimx: Gitksan language, Books 1 & 2. Kispiox: Kispiox Band.Google Scholar
Rigsby, Bruce. 1967. Tsimshian comparative vocabularies with notes on Nass-Gitksan systematic phonology. Presented at the International Conference on Salish Languages.Google Scholar
Rigsby, Bruce. 1986. Gitksan grammar. Ms., University of Queensland.Google Scholar
Rigsby, Bruce & Ingram, John. 1990. Obstruent voicing and glottalic obstruents in Gitksan. International Journal of American Linguistics 56 (2), 251263.CrossRefGoogle Scholar
Shaw, Patricia. 1987. Nonconservation of melodic structure in reduplication. In Bosch, Anna, Need, Barbara & Schiller, Eric (eds.), Papers from the Twenty-third Regional Meeting, Chicago Linguistic Society (CLS 23), 291306. Chicago, IL: Chicago Linguistic Society.Google Scholar
Tarpent, Marie-Lucie. 1983. Morphophonemics of Nisgha plural formation: A step towards Proto-Tsimshian reconstruction. Kansas Working Papers in Linguistics 8 (Studies in Native American Languages II), 123214.Google Scholar
Tarpent, Marie-Lucie. 1987. A grammar of the Nisgha language. Ph.D. dissertation, The University of Victoria.Google Scholar
Um, Hye-Young. 1998. Laryngeals and laryngeal features. Ph.D. dissertation, The University of Texas at Austin.Google Scholar
Wickstrom, Ronald William. 1974. A phonology of Gitksan, with emphasis on glottalization. MA thesis, University of Victoria.Google Scholar
Wright, Richard, Hargus, Sharon & Davis, Katherine. 2002. On the categorization of ejectives: Data from Witsuwit'en. Journal of the International Phonetic Association 32 (1), 4377.Google Scholar
Figure 0

Figure 1 F1 × F2 plots for Gitksan vowels. Ellipses define one standard deviation. The grid lines mark intervals of 750 Hz for F2 and 200 Hz for F1.

Supplementary material: File

Brown sound files

Sound files zip. These audio files are licensed to the IPA by their authors and accompany the phonetic descriptions published in the Journal of the International Phonetic Association. The audio files may be downloaded for personal use but may not be incorporated in another product without the permission of Cambridge University Press

Download Brown sound files(File)
File 29.4 MB