Proto-Athabaskan is the reconstructed ancestor of the Athabaskan languages.
History The Athabaskan languages have been recognized as a genetic grouping for more than a century. The relationship between Northern Athabaskan languages was first acknowledged in 1836 by Albert Gallatin, who named them after Lake Athabasca. They were later connected to the Oregon Athabaskan languages and the Apachean languages. Edward Sapir was the first to speculate about the nature of the proto-language in 1914 and published a reconstruction of Proto-Athabaskan phonology in 1915, although further reconstruction would not be published until 1956 by Harry Hoijer. In 1964, Michael E. Krauss published a new phonological reconstruction that constituted a significant reworking of Sapir's reconstruction. Eung-Do Cook and Keren Rice have also worked on the phonology as well as syntax, while Gillian Story and Jeff Leer have reconstructed grammatical items. Krauss and Leer together published a book-long paper reviewing the sonorants in 1981. Leer reconstructed the classifier in 1990, and he collected a comparative lexicon including reconstructed terms in 1996. The 2005 book Athabaskan prosody offered a survey of the issue of tone and prosody in Proto-Athabaskan and its daughter languages, and has reconstructed lexical items.
Tonology The Athabaskan languages are well-known among North American Indigenous languages for having many tonal members; however, several languages, including all of the Pacific Coast Athabaskan languages, lack tone. The correspondences between Athabaskan tone systems and their implications for Proto-Athabaskan have been the subject of linguistic inquiry for over a century.
The early work on Athabaskan languages ignored phonemic tone in grammatical and lexical works, although it was mentioned by linguists such as Émile Petitot and Adrien Gabriel Morice. Edward Sapir's first fieldwork on Athabaskan languages was with Chasta Costa and Kato, both Pacific Coast Athabaskan languages that lack tone. He documented tone in Tlingit, a Na-Dene language, in 1914 when working with Louis Shotridge. He then encountered tone in Tsuutʼina (Sarcee) and Gwichʼin, both low-marked languages, and thus concluded that Proto-Athabaskan must be reconstructed as a tonal language with two tones: marked low tone and unmarked (high) tone. However, Li Fang-Kuei came across apparently contradictory findings in Chipewyan. He discovered that Chipewyan's high tone corresponded to Sapir's reconstructed Proto-Athabaskan low tone. This pattern, called "tonal flip-flop," in which different languages had seemingly opposite tones for lexical items that were clearly cognate, is documented across the Athabaskan family.
Li attributed the pattern in Chipewyan to a "glottal stop which has been lost". In 1964, Michael E. Krauss published a paper advancing this hypothesis in which he argued that Proto-Athabaskan, instead of tone, had glottalization contrasts which developed independently into tones in the daughter languages or in some cases were lost. He additionally pointed to the Eyak language, which is related to Athabaskan and has glottalization contrasts but no tone, as support for his theory. This theory has been elaborated further by Krauss, Jeff Leer, and John Kingston, and has served as a model for further synchronic studies of tone in Athabaskan languages by authors such as Keren Rice, Gary Holton, and Siri Tuttle. Furthermore, Krauss indicated in 1979 that Jeff Leer's discovery of the toneless Tongass dialect of Tlingit, a Na-Dene language, "strongly suggests that earlier Tlingit was also toneless" like Eyak and Proto-Athabaskan.
Origins There have been varying proposals as to when Proto-Athabaskan was spoken, when it diverged from Eyak, and when it diverged into its daughter languages. Most proposed datings are low compared to other proto-languages due to the relatively little internal diversity within Athabaskan. Morris Swadesh, Alfred Kroeber, Harry Hoijer, and Dell Hymes used glottochronology to estimate dates of divergence within Athabaskan, presenting dates ranging from between 1300 and 1800 years before the present. Krauss has proposed that Proto-Athabaskan was spoken between 2000 and 2500 years ago, while he posits that Proto-Athabaskan and Proto-Eyak diverged around 3400 years ago. However, James Kari has countered these estimations, saying that the homogeneity within Athabaskan can be explained by typological linguistic factors as well as common territorial practices. He has thus argued that Eyak separated from Athabaskan no earlier than 8000 years ago, while Proto-Athabaskan continued to be spoken until about 3000 years ago. Archaeologically, Proto-Athabaskan–Eyak speakers have been linked with the Northwest Microblade tradition of 4000 B.C. that occupied large areas of Alaska and northwestern Canada, although Krauss and Victor Golla have called this association "highly speculative".
Proto-Athabaskan was likely spoken in the interior of what is now eastern Alaska, the Yukon, and parts of British Columbia. This is indicated by the greater degree of linguistic differentiation among the Athabaskan languages now spoken in these same areas. There is little apparent Eskaleut influence on Athabaskan languages except for in Deg Xinag and Dena'ina, which are adjacent to Yup'ik territory, suggesting a non-western origin for the language family. Furthermore, of the Northern Athabaskan languages, only Dena'ina included coastal territory, which also indicates an interior origin. However, this proposal, which is often attributed to Michael Krauss and Victor Golla, has seldom been critically examined. Joseph A. P. Wilson has argued that the Urheimat of Proto-Athabaskan was instead in southwestern Alaska because some reconstructed Proto-Athabaskan lexemes refer to technology developed west of the traditionally proposed region. A southwestern homeland is also more compatible with the Dene-Yeniseian hypothesis. The relative lack of Eskaleut influence on Athabaskan may be due to a rapid expansion of the latter family.
Proto-Athabaskan likely first expanded westward in Alaska and southward into the Interior Mountains of British Columbia. The Pacific Coast Athabaskan languages were likely a result of an early migration southward from British Columbia, while eastern Northern Athabaskan languages arrived at a later date to the Mackenzie River basin. Finally, the date at which the Southern Athabaskan languages arrived in the American Southwest has been debated upon. The earliest known date of Athabaskan habitation in the region is 1300 A.D., although earlier theories suggested an arrival just before the Coronado Expedition in the 16th century. Most hypotheses have argued for a route taken through the Great Plains, but more recent studies have suggested a route that went at least partially through the Rocky Mountains. By the time of European contact, the descendants of Proto-Athabaskan occupied a total area of approximately , the most of any Indigenous language family in North America.
Phonology Notation
Athabaskanists tend to use an Americanist phonetic notation system rather than IPA. Although some Athabaskanists prefer IPA symbols today, the Americanist symbols are still in common use for descriptions of Proto-Athabaskan and in comparisons between members of the family. In the tables in this section, the proto-phonemes are given in their conventional Athabaskanist forms with IPA equivalents following in square brackets.
The symbols conventionally used to represent voiced stops and affricates (e.g. ) are actually used in the Athabaskan literature to represent unaspirated stops and affricates (i.e., ). This convention is also found in Athabaskan orthographies since true voiced stops and affricates are rare in the family, and unknown in the proto-language.
Since Americanist phonetic notation is not formally standardized, there are sometimes both historic and modern symbols for the same sounds. In the following tables, the older symbols are given first with newer symbols following. Republication of older materials may preserve older symbols for accuracy, e.g. Krauss 2005, which was previously an unpublished manuscript dating from 1979.
Consonants
The traditional reconstruction of the Proto-Athabaskan sound system consists of 45 consonants, as detailed in the following table.
Obstruents Bilabial Alveolar Postalveolar Velar Uvular Glottal median lateral plain labial plain labial Stop unaspirated aspirated glottalized ~ ~ Affricate unaspirated ~ ~ ~ ~ aspirated ~ ~ ~ ~ ~ glottalized ~ ~ ~ ~ ~ Fricative voiceless ~ ~ ~ voiced ~ ~ ~ ~ Sonorants Nasal ~ ~ ~ Approximant ~ ~ ~
First person singular fricative
The fricative sound in the first person singular pronoun in Proto-Athabaskan has uncertain phonetic reconstruction, and is represented variously as or . In Athabaskan languages, it usually has a reflex of , but in Eyak it appears as and in Tlingit as . However, in Kwalhioqua-Tlatskanai, it seems to have been in at least some forms of the first person subject verb prefix. It does not correspond well with other fricatives, a situation that led Krauss to consider it as unique. This proto-phoneme is not given in the table above, but is assumed to be a part of the Proto-Athabaskan inventory.
New consonant reconstruction
A newer reconstruction by Leer constitutes a reorganization of the system. Velars are reinterpreted as palatals, labialized postalveolar affricates are reinterpreted as retroflex consonants, and other labialized consonants are removed. Leer also adopted the argument by Keren Rice that there was no need to distinguish between and . The resulting system is somewhat simpler than the traditional one, with 8 fewer phonemes.
Obstruents Bilabial Apical Lateral Laminal Postalveolar Retroflex Palatal Uvular Glottal Stop/Affricate unaspirated ~ ~ aspirated ~ ~ glottalized ~ ~ ~ Fricative voiceless ~ voiced ( ) ~ Sonorants Nasal ~ Approximant
Leer has argued for an asymmetric lack of retroflex fricatives in the Proto-Athabaskan, saying that "PA lacked distinctively reflexed and as opposed to plain and ". Although Leer did not include and in his list of reconstructed consonants, those two proto-phonemes nevertheless appear in a variety of reconstructions in the same article.
Vowels
Leer also offered a vowel system consisting of four long or full vowels and three short or reduced vowels which are more centralized.
Front Back Full Reduced Reduced Full High Mid ~ Low
The following table is adapted from Leer 2005 and shows the vowel correspondences between Proto-Athabaskan and the better documented Athabaskan languages.
Language Full vowels Reduced vowels Proto-Athabaskan prefix stem Denaʼina i a u i ə ~ ∅ ə ə ə Deg Hitʼan e a o e ə ə ə ʊ Koyukon i a o u ə ~ [∅] ə α ~ ʊ ʊ ~ α Upper Kuskokwim i a o u ə ~ [∅] ə ʊ ʊ Lower Tanana i a o u ə ~ [∅] ə ʊ ʊ Ahtna i(ˑ) e(ˑ) a(ˑ) u(ˑ) e ~ ∅ e a o Tanacross i(ˑ) e(ˑ) a(ˑ) u(ˑ) e ~ ∅ e a o Upper Tanana i(ˑ) e(ˑ) a(ˑ) u(ˑ) i ~ ∅ ɵ ~ a a o Hän i e æ u ə ~ ∅ ɵ ~ ə a o Gwichʼin i[pal] i[pal] e ~ i i(o)[pal] ə a a o Northern Tutchone i i e u e ʌ ʌ o ~ ʌ Southern Tutchone i e a u e ʌ ʌ o ~ ʌ Tagish-Tahltan i(ˑ) e(ˑ) a(ˑ) u(ˑ) e e ~ i a o Tsekʼehne/Sekani i e a u ə ~ ɪ ə ~ i a o ~ ʊ Witsuwitʼen i ~ e i ~ e ~ ɛ a ~ e u ~ o ə ~ ∅ ə ə o ~ ə[rnd] Dakelh/Carrier i e ~ i a u ~ o ə (~ ∅) ə ə ə[rnd] Slave i e a u ɛ ɛ a o Dëne Sųłiné/Chipewyan (Li) i e ~ ə ~ ɛ a u ɛ ~ ə ɛ ~ ə a o Tsuutʼina i a o u i i o u Navajo i(ˑ) e(ˑ) a(ˑ) o(ˑ) i ~ a i ~ a a o Apache (Hoijer) i(ˑ) e(ˑ) a(ˑ) o(ˑ) i i ~ a a o Hupa (morph.) e e a o ə ə α ʊ Hupa (phonemic) e(ˑ) e(ˑ) a(ˑ) o(ˑ) i i a o Mattole (Li) i(ˑ) e(ˑ) a(ˑ) o(ˑ) i i a ~ i o Galice (Hoijer) i(ˑ) e(ˑ) a(ˑ) o(ˑ) a a a a[rnd] Tututni (Golla) i e a u ə ə ə ə[rnd]
Tone