1 A comment on Gedney's proposal for another series of voiced initials in Proto-Tai revisited.1 Graham Thurgood California State University, Chico 1. INTRODUCTION Gedney is well-known for a number of his contributions to Tai linguistics, with one of the best-known contributions being his checklist for determining tones in Tai dialects (reprinted as Gedney 1989c:202 (but originally appearing in 1972)). The first column of the checklist contains the different groups of initial sounds in Proto-Tai. The first three rows contain various voiceless sounds; the fourth row contains voiced initial sounds. In short, the first column contains four distinct groups of proto-Tai initials. Proto-Tai tones A B C D-short D-long Voiceless friction sounds, *s, *hm, *ph-, etc. 1 5 9 13 17 Voiceless unaspirated stops, *p, etc. 2 6 10 14 18 Glottal, *÷, ÷b, etc. 3 7 11 15 19 Voiced, *b, *m, *l, *z 4 8 12 16 20 Smooth syllables Checked syllables Chart 1: Gedney's checklist While these four patterns in the Gedney checklist account for most of the patterns found in the Tai languages, they do not account for all the patterns. In a paper first printed in 1979 (reprinted in 1989a) entitled “Evidence for another series of voiced initials in proto-Tai”, Gedney quite tentatively posited one more pattern, a second series of voiced initials for Proto-Tai (PT). Gedney's paper was written to account for roughly fifty forms with a pattern of initial and tonal correspondences that do not fit into the chart given above, that is, these forms follow a pattern different from the patterns of all the other Tai words. Although the anomalous nature of these forms must also have been apparent earlier to Li, in Li's Handbook (1977) the majority of these forms are simply reconstructed with voiceless aspirated stops for Southwestern (SWT) and Central Tai (CT)), with Li frequently appending a note that, in Northern Tai (NT), the forms are reconstructed, not with a voiceless onset, but with a voiced onset. 1 This paper is a revised version of "A comment on Gedney's proposal for another series of voiced initials in proto-Tai", which appeared in 2002 in Studies in Southeast Asian Languages, edited by Robert Bauer. Pacific Linguistics, pages 169-183.2 Table 1a Gedney forms: The tones and the initials (the patterns) Proto-Tai tone Proto-Tai Initial Siamese (SWT) LM (CT) Saek (Northern Tai) Yay (Northern Tai) *ph- ph- 5 ph- 1 ph- 2 p- 1 *A *B- ph- 5 ph- 1 ph- 4 p- 4 *b- ph- 1 p- 4 ph- 4 p- 4 *ph- ph- 2 ph- 2 ph- 6 p- 2 *B *B- ph- 2 ph- 2 ph- 5 p- 5 *b- ph- 3 p- 5 ph- 5 p- 5 *ph- ph- 3 ph- 3 ph- 3 p- 3 *C *B- ph- 3 ph- 3 ph- 6 p- 6 *b- ph- 4 p- 6 ph- 6 p- 6 *ph- ph- 2 ph- 3 ph- 4 p- 2 *DS *B- ph- 2 ph- 3 ph- 6 p- 1 *b- ph- 4 p- 4 ph- 6 p- 1 *ph- ph- 2 ph- 2 ph- 6 p- 2 *DL *B- —— —— —— —— *b- ph- 3 p- 5 ph- 5 p- 5 Table 1b Gedney forms: The tones and the initials (examples) Proto-Tai tone Proto-Tai Initial Siamese (SWT) LM (CT) Saek (Northern Tai) Yay (Northern Tai) *ph- phom∞ phyom¡ phram™ piam™ head hair *A *B- fii∞ phey¡ phay¢ pay¢ boil; ulcer *b- ph‰‰≥¡ --- ph‰‰≥¢ pe≥¢ expensive *ph- phaa™ phaa™ phaa§ pa™ split; hew *B *B- thii™ thay™ thii∞ ti∞ thick; dense3 *b- phii£ --- phii∞ pi§ older sibling *ph- ph- 3 ph- 3 ph- 3 p- 3 *C *B- phuu£ phow£ phuu§ pu§ person; male *b- sa™phay¢ pay§ ph©©§ pay§ daughter-in-law *ph- phak™ phyak£ phrak¢ piak™ vegetable *DS *B- khop™ khop£ ©ap§ hap¡ bite *b- khrok• lok¢ --- cok¡ mortar *ph- phaak™ phyaak™ phraak§ pyaak™ forehead *DL *B- --- --- --- --- *b- phaat£ --- --- paat∞ lay cloth Table 1a illustrates the patterns that were noticed both by Li and by Gedney (The Table 1b forms are simply some examples). The first column lists the five tone classes reconstructed for Proto-Tai (PT), which are indicated by A, B, C, DS, and DL. The syllable final in tone classes A, B, and C ends either in a vowel or a sonorant, a group constituting what Tai scholars term ‘live’ syllables. The syllable finals in tone classes DS and DL end in a stop, constituting what Tai scholars term ‘dead’ syllables, with DS occurring with short vowels and DL with long vowels. The second column represents the classes of Proto-Tai initials, using the bilabials as the example: *ph- represents the voiceless aspirated stops, *B represents the pattern documented by Gedney and Li, and *b- represents the voiced stops. The primary focus of interest in Table 1 is pattern of tonal variation for *B forms. In the Southwestern and Central Tai dialects, the tones of these forms pattern as if they had had Proto-Tai voiceless obstruent initials; in the Northern Tai dialects, these forms pattern as if they had had Proto-Tai voiced obstruent initials. Thus, in Siamese, a dialect of Southwestern Tai, and in Lungming, a dialect of Central Tai, tones of the *B forms pattern tonally with the voiceless aspirated *ph- series, while in Saek and Yay, both Northern Tai dialects, the *B forms pattern tonally with the *b- series. This tonal alternation pattern was obvious to Li. Although he reconstructed the majority of these forms with voiceless aspirated stops in his Handbook (1977)—the way that the forms reconstructed in Southwestern and Central Tai, but he added notes for most of these forms noting that they patterned as if they reconstructed with voiced obstruents in Northern Tai languages. It was Gedney (1989a), however, who first fully recognized that these forms were a major problem for the reconstruction of proto-Tai. In the careful way that Gedney's work is always done, he meticulously arranged the evidence, showing that just as Li had also noticed that the tones of the Southwestern and Central Tai reflexes would require4 them to descend from earlier proto-Tai voiceless obstruents, while the initial and tonal patterns of the Northern Tai reflexes would require that the forms descend from earlier voiced obstruents. Then, Gedney discussed the problem, which is what kind of initials could this pattern of tones and initials have come from? The problem is that Proto-Tai already has a well-established series of