The foundation of the reconstruction of Proto-Karen was laid by André-Georges Haudricourt in 1946, with revisions in 1953 and 1975.
Haudricourt applied the comparative method to forms from two Karen languages, Pwo and Sgaw, from A Comparative Dictionary of the Pwo-Karen Dialect (1922) by W. C. B. Purser and Saya Tun Aung.
Each of these languages has six tones, four in open syllables and two in checked syllables (those ending in a glottal stop).
By comparing the lexical incidence of these tones, Haudricourt established eight correspondence sets, later labelled I to VIII by Gordon Luce, six in open syllables and two in checked syllables.
The two languages had similar inventories of initial consonants, distinguishing aspirated, unaspirated and implosive stops and having only voiced sonorants.
Implosives and sonorants were aligned between the two languages, but aspirated and unaspirated stops yielded three correspondence sets.
Moreover, the initial correspondence sets occurred only with certain tone correspondence sets, as follows:[1][2][3]
Tone and initial correspondences between Pwo and Sgaw Karen
Initials (illustrated with bilabials)
Tones
*A (modal)
*B < **-s
*B' < **-ʔ
*D < **-p/t/k
pʰ:pʰ < *voiceless aspirated stop
1h:1h (III)
2h:2h (VI)
2h:1h (V)
3h:3h (VIII)
m:m < *voiceless sonorant
p:p < *voiceless unaspirated stop
1b:1h (II)
ɓ:ɓ < *implosive stop
pʰ:p < *voiced stop
1b:1b (I)
2b:2b (IV)
3b:3b (VII)
m:m < *voiced sonorant
This fits a common pattern in languages of the Mainland Southeast Asia linguistic area, including Tai languages, Hmong-Mien languages, Vietnamese and varieties of Chinese, in which a four-tone system, reflecting earlier final segments, develops a register distinction conditioned by the manner of the initial, leading to a tone split.
The varying treatment of the first tone is also found in Tai and Chinese languages.[4][5]
Haudricourt's reconstruction was further supported by subsequent reporting that voiced stops and voiceless nasals are retained by other Karen languages, such as Geba.[6]
Manson gave a sample of diagnostic words for use during field elicitation to classify Karenic languages:[7]
Diagnostic words for proto-Karen tone and initial categories (Manson 2009)[7]
Haudricourt originally viewed the correspondence set V as irregular, and so reconstructed included only the three proto-tones *A, *B and *D.
He added the proto-tone *B' in his 1975 revision.[8][9][10]
Haudricourt's *B' class has been accepted by most modern workers on Karen, but is not included by Luangthongkum.[11]
This class is not distinguished after originally voiced initials, but a similar merger is common in Chinese varieties.[12]
Correspondence class V is not reflected as a distinct class in any modern Karen language, being merged with *A in Sgaw and Pa'O, with *B in Pwo, with *D in Kayan, Kayaw and Kaya, and with both B* and D* in Bwe-Gaba.[13]
Luangthongkum has suggested that the words in class V might reflect an earlier final segment, a view that Haudricourt had also expressed.[11]
Only Pa'O has a full set of nasal and stop codas, though many occurrences of -p, -t or -k are found in loans from Shan or Pali.
Other Karen languages may have nasalized vowels instead of nasal codas, and only glottal stop codas.
Some have only open rhymes.[17]
Most linguists accept Haudricourt's revised reconstruction with three proto-tones *A (modal), *B and *B' in open syllables, with checked syllables forming a separate category *D.[11]
However, Luangthongkum accepts only *A, *B and *D.[18]
Burling, Robbins (1969). Proto-Karen: a reanalysis. Occasional Papers of the Wolfenden Society on Tibeto-Burman Linguistics. pp. 1–116.
Jones, Robert B. (1970). "Proto-Karen: A Reanalysis. by Robbins Burling; Lahu and Proto-Lolo-Burmese. by James A. Matisoff and A. L. Becker". The Journal of Asian Studies. 30 (1): 230–231. doi:10.2307/2942792. JSTOR2942792.
Haudricourt, André-Georges (1946). "Restitution du karen commun". Bulletin de la Société de Linguistique de Paris. 42 (1): 103–111.
——— (1953). "A propos de la restitution du karen commun". Bulletin de la Société de Linguistique de Paris. 49 (1): 129–132.
——— (1975). "Le système des tons du karen commun". Bulletin de la Société de Linguistique de Paris. 70 (1): 339–343.
Jones, Robert B. Jr. (1961). Karen linguistic studies: Description, comparison, and texts. University of California Publications in Linguistics. Vol. 25. Berkeley and Los Angeles: University of California Press.
Cooke, Joseph R. "Karen Linguistic Studies: Description, Comparison, and Texts by Robert B. Jones". Journal of the American Oriental Society. 82 (2): 241–246. doi:10.2307/597941. JSTOR597941.
Henderson, E. J. A. (1964). "Robert B. Jones: Karen linguistic studies: description, comparison, and texts". Bulletin of the School of Oriental and African Studies. 27 (3): 662–665. doi:10.1017/S0041977X00118762. S2CID162423942.
Shintani Tadahiko. 2014. The Zayein language. Linguistic survey of Tay cultural area (LSTCA) no. 102. Tokyo: Research Institute for Languages and Cultures of Asia and Africa (ILCAA).
Shintani Tadahiko. 2015. The Kadaw language. Linguistic survey of Tay cultural area (LSTCA) no. 106. Tokyo: Research Institute for Languages and Cultures of Asia and Africa (ILCAA).
Shintani Tadahiko. 2016. The Nangki language. Linguistic survey of Tay cultural area (LSTCA) no. 109. Tokyo: Research Institute for Languages and Cultures of Asia and Africa (ILCAA).
