Wiktionary:About Proto-Algonquian

Proto-Algonquian (PA) is the ancestor of all the Algonquian languages. Their relatedness and descent from a common source was recognised early on; a decade before William Jones' famous "philologer" speech on Proto-Indo-European, native Mohegan speaker Jonathan Edwards reported to the Connecticut Society of Arts and Sciences on the Algonquian "language [family that] is spoken by all the Indians throughout New England. Every tribe, as that of Stockbridge, that of Farmington, that of New London, &c. has a different dialect [i.e. language], but the language [family] is radically the same."[1] Proto-Algonquian is now one of the best-studied, most thoroughly-reconstructed proto-languages.[2][3]

Proto-Algonquian was spoken around 2 500 to 3 000 years ago[4] somewhere around the Great Lakes. It descends, like the languages Wiyot and Yurok, from Proto-Algic.


Proto-Algonquian had four basic vowels, which could be either long or short:

long: *i·, *e·, *a·, *o·
short: *i, *e, *a, *o

The same inventory of eight vowels was found in Proto-Algic, but Proto-Algonquian did not inherit its inventory directly from Proto-Algic. Rather, several sound changes left pre-Proto-Algonquian without short *i and *o, and they only later redeveloped,[5] apparently as the result of morphophonological shortening. They are thought to be of no great antiquity as independent phonemes within Proto-Algonquian; nonetheless, they are used in reconstructions.[2][6]


Proto-Algonquian had a slightly smaller number of consonants than Proto-Algic.

Proto-Algonquian consonant phonemes
labial alveolar postalveolar/
velar glottal
stop / plosive p t /t/ č /tʃ/ 1 k ʔ 2
fricative central s š /ʃ/ h
possibly lateral ɬ /θ/ or /ɬ/ 3
sonorant nasal m /m/ n /n/
lateral or rhotic l /l/ or r /r/ 4
semivowel w y /j/
1 Whether č was an independent phoneme or merely an allophone of t before i is disputed.[7] The evidence for its phonemicity is very weak; t is only found before i in some reconstructions of the single questionable, possibly onomatopoeic word *ti·nti·wa, and č is only found elsewhere in the (also possibly onomatopoetic) word *čapo·nk- and in clusters.
2 The nature of this phoneme is not certain, but it probably was indeed /ʔ/.[2][8]
3 The nature of this phoneme is uncertain; it was most likely /θ/ or /ɬ/. Following Bloomfield, many authors have written θ, but Picard argues that it was more likely /ɬ/, and Proulx reconstructs both it and its Proto-Algic antecedent as ɬ. In daughter languages, it became θ, t, l or n (in isolation; in clusters, it became several other things, such as s).[9]
4 Bloomfield reconstructed this phoneme as l, but Goddard argues it is more likely r.

Reconstructing Proto-Algonquian's consonant clusters is difficult because their evolution in different child languages has been complex. In addition to the consonants above, two consonants of uncertain identity can be found in clusters before p and k. Bloomfield and some scholars after him have arbitrarily used the symbols x and ç to represent them (the consonants are not /x/ and /ç/). Goddard believes that Bloomfield's x can be reconstructed as s and that Bloomfield's ç is the non-nasal alveolar sonorant (which is either l or r, as described above).

See alsoEdit


  1. ^ Lyle Campbell, American Indian Languages
  2. 2.0 2.1 2.2 Ives Goddard, Comparative Algonquian (1979)
  3. ^ Lyle Campbell Historical Linguistics: An Introduction (2004)
  4. ^ Ives Goddard, Central Algonquian Languages, in Trigger (1978), volume 15 of the Handbook of North American Indians
  5. ^ Paul Proulx, Proto-Algic VI: Conditioned Yurok reflexes of Proto-Algic vowels, Kansas Working Papers in Linguistics 27:124–138 (2004)
  6. ^ Howard Berman, Two Phonological Innovations in Ritwan (1982)
  7. ^ Ives Goddard, A New Look for Algonquian (1994)
  8. ^ Lucy Thomason, Proto-Algonkian Phonology and Morpho-Syntax (2006), in the Encyclopedia of Language and Linguistics 10, second edition
  9. ^ Marc Picard, in Principles and Methods in Historical Phonology: From Proto-Algonkian to Arapaho (1994), argues that an original /ɬ/ shifting to /θ/ in some daughter languages (and /θ/ shifting to /t/) and to /l/ in other daughter languages (and /l/ shifting to /n/) is more plausible than an original /θ/ shifting straight to /t/ in some daughter languages but via an unpreserved /ɬ/ to /l/ and then to /n/ in other languages.