When The Voice Goes Out…Kun Chang 張琨 (次瑤) 1917-2017

"Collect, observe, and study carefully. Get your data right and let the language speak for itself."

These were the words he offered to students who came to enroll in his graduate seminar on Chinese linguistics. These were also the words he held dear to his heart, always with microscopic examination, be it Chinese or Sanskrit, Tibetan or Miao-Yao. This conviction was, in fact, what kept him going for more than seven decades in research, a conviction that accounted for his accomplishments both as a linguist and a teacher. Professor Kun Chang, of the University of California at Berkeley, passed away on April 4, 2017, at the age of 99.

Chang received his BA in Chinese from the Tsing Hua University, Beiping, in 1938. Shortly thereafter, he was recruited by Professor Fang Kuei Li, of Academia Sinica, as an assistant to work on languages other than Chinese in southwestern China, including Miao-Yao, and also Tibetan. He followed Li to Yale in 1947, where he eventually received his PhD in Linguistics, with a dissertation on the Kathinavastu, a comparative study of a Sanskrit Buddhist text. He began teaching at the University of Washington, Seattle, in 1951, and in 1963, when Professor Yuan Ren Chao retired from the University of California at Berkeley, he came to join the faculty in the Department of Oriental Languages. He retired in 1989. Chang was elected a lifetime Academician of Academia Sinica in 1972.

Chang started out his linguistic pursuit under the supervision of Professor Li in the 1930s and conducted laborious field investigations of aboriginal languages, with a particular emphasis on Miao-Yao. In 1947 he published a paper in the Bulletin of the Institute of History and Philology, reporting his findings of a well-defined tonal system of Miao, a system that had long baffled scholars with its seemingly irregular and often confusing tonal variations among dialects. The system consisted of four basic tones, according to his analysis, roughly corresponding to the traditional 4-tone system in Chinese, each of which had undergone further splits and coalescences under different phonological conditions. A revised version of the paper in English appeared in Language in 1953. The findings, based mostly on Miao dialects, provided for the first time a methodological and comparative framework that made subsequent studies, both descriptive and diachronic, possible within the field. Chang himself continued to work on the reconstruction by studying tonal issues in Yao (1966) and eventually incorporated his findings of both Miao and Yao into a seminal publication entitled "The Reconstruction of Proto-Miao-Yao Tones" (1972). With an amassed data of 69 Miao-Yao dialects, based on his own field work notes and also reports by others, spoken in China, Vietnam, Thailand and Laos, he reported the actual values of all tones in all 69 dialects, a corpus that enabled him to further confirm his reconstruction of a proto-system with four tones, the last category also being segmentally marked by a consonantal coda -p, -t, or –k. More importantly, in critical details, he provided an account of the ways in which further tonal splits had occurred under the influence of the initial consonant in a syllable, conditions including voicing, aspiration and pre-nasalization. As a result, some contemporary dialects could have up to eight or even more tones, but some with a fewer number because of coalescence in pitch between derived tones. The proto 4-tone system and its subsequent development as conditioned by the initials readily recalls what we are all too familiar with in the Chinese language. The resemblance, however, was not necessarily a justifiable reason to classify Miao-Yao as members of the Sino-Tibetan family. On the other hand, he turned to cognates shared by Chinese and Miao-Yao as evidence of this historical association, a claim that some scholars discounted, assuming such cognates were merely cases of language borrowings. In an interview he gave in 1983, Chang suggested that, language typology being such a complex and formidable topic for any one scholar to tackle with confidence, it would be advisable for all interested parties...