the vocalism of sino-tibetan

The Vocalism of Sino-TibetanAuthor(s): Robert ShaferSource: Journal of the American Oriental Society, Vol. 61, No. 1 (Mar., 1941), pp. 18-31Published by: American Oriental SocietyStable URL: http://www.jstor.org/stable/594341 .

Accessed: 12/06/2014 19:48

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

American Oriental Society is collaborating with JSTOR to digitize, preserve and extend access to Journal ofthe American Oriental Society.

http://www.jstor.org

This content downloaded from 185.44.78.113 on Thu, 12 Jun 2014 19:48:31 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=aos

http://www.jstor.org/stable/594341?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


THE VOCALISM OF SINO-TIBETAN *

ROBERT SHAFER BERKELEY, CALIFORNIA

PART 2. CONSONANTAL FINALS

Table 11. Medial *-e-

meaning 0. B. Bu. L. Ch. S.

1 to not exist med-pa miet (622) hmedt (Lao) 2 to wink ...... hmit miet (622) 3 to slip adred-pa d'iet (880) ...... 4 to separate; to open p'yed (pf.) ...... piet (720) pie1 5 butterfly p'ye-ma-leb lip-pra -p'e-hlep 6 to roll up ...... lip lep ...... 7 to fold Iteb-pa ...... t'lep' ......

Ideb-pa .. d'iep (992) [dap] 8 that which is above, top ste. ...... tieng (999)

9 to beat, strike lbrdegs (pf.);te - den' fT 'tieng (999) ...... ~ (Mantgati group) ..... 10 know mk'yen-pa ...... kien' (368) 11 to be sleepy, to sleep mya'nt ...... cmnien (629) 12 to be flat; board pyan . 'Pien (733) 13 to taste tem pj'jtiem (997) ...... 14 to lick lem- (W.) ..... **'liem (997) *---- 15 to be full, overflowing Item-pa ...... ...... tiem' (1222) tM 16 memorial tems-yig tim-hmat ...... 17 forehead .. .... ...... ...... ctien (995) t (Lao) 18 sky ...... ....t.. ien (996) te4n (Lao) 19 shoulder ...... ...... ckien (371) k'Edn (Ls) 20 palace, temple ...... ...... ...... d'ien' (474) den 21 firm, solid ...... ckien (369) ken

22 reeds ...... ... ... ...... ckiem (374) k'8m

23 unite in one, put together.. ckiem (374) Wkm

24 add, increase . ..... ... t... ctiem (997) Vtfm (Ls) 25 black point, spot, dot;

punctuate, mark off .... ... .... 'tiem (1162) Y..m (Ls) 26 leg ...... ken. ken 27 to pull .... ... ..... ckien (169) nkt 28 jaw, cheek ...... kiep (345) kem (Ls)

Notes to Table 11

1. Lao "finished, consumed." 4. 0. B. pres. dbyed-pa. 6. L. "to turn up." 7. 0. B. (1) "to double down, turn in (as the hem),"

(2) " bend round or back, turn round, double down." Lao dsp.

8. Ch. "top of head; summit." 9. 0. B. pres. rdegs-pa.

10. Ch. "see," probably the original ST meaning. Not in Lusei, but Khimi (S. Kukish) has k'ei "see " (any final nasal becomes -A in Khimi).

11. Ch. "to close the eyes, to sleep." The seeming phonetic irregularity of this and the following word in Burmese is only apparent, as TB *-in became Bu. -an; see medial -i- table. Ka. myen "to fall to sleep," but Bahing myel- "to be sleepy." ST *myeln?

12. Ka. byen " to be flat and wide," p'un-pyen " board, plank" (pun "tree"), Nung siA-byen "board, plank" (sinC "tree "). Ch. " flat; signboard, tablet ": cognate * Continued from JAOS, 60 (1940). 337.

18



SHAFER: The Vocalism of Sino-Tibetan 19

to J4 p'ien' "board, tablet; slip, slice, leaf, sheet, card" (735), S. pC7n, ben "surface, plaque, blades, numeral classifier for planks and other flat things," pen "even, flat," Lao p'en " flat things," but L. per " flat and thin, flat," Dim. -p'er (g6-) " flat," < *pyern?

14. The root in W. Him. is, however, 1eb-; but Bahing lydm " tongue," EE.Him. 16m. The Ch. is based on Cant. 'Urm.

15. Ch. "to fill up, wedge up." S. "full," Lao t~dm. Cf. also 0. B. bitams (pf.), Itams-pa (pres.) " to be full," gtams-pa "full."

16. 0. B. yig "inscription." Bu. " designate, set up (as a mark or memorial) ," hmat " make a mark; to mark, note down."

17. Cf. Ka. [tan, ka-tan, 1a-t'an] " forehead," Moshang [tVal] " face."

Due to Wulff's excellent Daic comparisons, there can be little doubt that the equation for the above table is 0. B. -e- = Bu. -i- -L. -e-, - C- Oh. -ie-

=S., Lao -e- (--). Just as Old Bodish is not archaic in its final

vowels and diphthongs, neither is it archaic in its medial vowels and diphthongs. Thus -y- cannot occur after an initial affricate, dental, sibilant or liquid. That this was not always the case in Bodish is shown by the westernmost dialects, Sbalti and Burig, in a number of respects morei archaic than the written language. One of these is the preservation of original *-ya- after the above mentioned initials. Thus, 0. B. Ite-ba " navel-string,"

19. S. " arm," Lao "forearm, wrist." 20. S. "seat of the mighty; altar," Lao den " throne;

altar." 21. S. "firm, solid; heart of tree," ken "hard part

of fruits," Lao ken "hard, solid," ken "heart of tree." 23. S. " mix, intermix; mingled," Lao kdm " together,

with." 25. S. "to mark, note, paint," Lao "to note, write,

paint, sketch." 26. Not L., but Thado, Mikir, hence length of vowel

uncertain. Lao gen. 27. Lao kW'n. 28. S., Lao "cheek." Ka. [99-kap] " jaw" (Hz.). Cf. also N diep "butterfly" (225), Ka. 1Il-tp (N.).

Sbalti ttia " navel "; 0. B. legs-mo " good," Sbalti lyax-mo, Burig Iyay-mo (the guttural spirants are from *-gs) ; 0. B. tVeg-pa "to lift; to endure," Sbalti tcyak'-pa " patience," Burig tVyak- " to lift." That this represents the original state of the Bod- ish language and is not a regular phonetic change is shown by the fact that Sbalti and Burig generally have -e- corresponding to 0. B. -e-, and -ya- only occasionally corresponding to 0. B. -e- and then after the above named initials. Unfortunately the Sbalti and Burig materials are too limited to be able to cite the WW. B. form with -ya- corre- sponding to 0. B. -e- in the following table.

Table 12. Medial *-ya-


1 to think snyam-pa ...... ...... Ad 'Nahum < n- (935) 2 to lift, raise btegs (pf.) . . ...... Df4k < t- (759) ...... 3 to sew btsems (pf.) . . ...... Xf cj9m < - (876) ...... 4 mind sems .... .stem (801) ...... 5 fly yan (W.) yaA W cjang (632) ...... 6 soft nyaih7 (nem) 4 'niom<n- (935)

Notes to Table 12

2. 0. B. pres. ddegs-pa. For cognate form with -ya- in WW. B., see preceding paragraph. Ch. "to mount," i. e. intr. to the 0. B. verb.

In the above table it was found that *-yra is represented in Chinese by -ip-. The instances in which Ch. -is- occurs in the comparative material are not many but indicate that the. a of the diph-

3. 0. B. pres. dts'em-pa. Ch. " needle." 4. Ch. " thought, intelligence." 5. Probably originally used for any flying insect, as

the W. Him. compounds indicate. W. Bodish bu-yaA " bumble-bee."

thong represents an original *a which has been palatalized by a preceding y or palatalizing consonant.



20 SHAFER: The Vocalism of Sino-Tibetan

Table 13. Ch. -is-


1 weep k'rab- ...... tap' j. 9 k'ap (524) 2 water, fluid t9'ab ...... ...... b ap < t' (876) 3 deep zab-pa ..... . . cm (870) 4 jungle ...... ram cim (555) ...... 5 breath ...... 'd-sak ...... sik (780) ......

Notes to Table 13

2. Ch. "fluid, liquor." 4. Ch. "forest." 5. Ch. "to breathe."

Table 14. Medial -i-


1 joint tstigs -t.'atg ...... ...... ...... 2 leopard gzig sats ...... ...... ...... 3 dirt, filth nyag-nyig nyat . ...... 4 one *ktyig gtsig tats ...... i tslk <t' (1265-A) ......

tnyik-tsin ( Tsangla) 5 two nik-s! ( Bahing ) hnats ...... .......

6 bamboo smyig-ma hmyat . ...... ...... 7 sound of sobbing hig hats ...... ...... 8 eye mig (myak) mit ...... 9 pheasant (sreg-pa) rat9 -hrit ...... ......

10 to bind bkyigs (pf.) k'yaA k'it 4+ (kiet) (325) 11 to bite tuig- (Kanauri) . .i..k(866) tilk 12 name minA maA hminh cmmdng (633) 13 neck IaA ...... 'lhang (588) 14 to bind btsins (pf.) tan ...... 15 long rnh-ba 15 long ri..b, hraA ...... .......... ......

extend srii-ba h 16 bedding tVih- (Sbalti) -t'an ...... ..h.... ...... 17 sound ....r.. ar& rik' ...... 18 alive *srin (W.) [hrai&] hrin ...... ...... 19 heart snyin hnats nin ...... ...... 20 tree, wood sii sats t'ih csien (803) 21 dangling, waving, float.

ing (in wind) lin-ne -hats ...... ...... sin-mo )

22 new tsi-mo (Tsangla) sat ...... csin (803) ka-uik (Rgyarong)

23 year -nih hnats ...... (cnien) (669) ...... 24 shadow, shade grib -rip hlim [ram] 24a top, summit *t'yip? ...... tip tilip ...... ...... 25 sleep ib (Bunan) ip ...... ...... 25a soft hnip . i'zdm<hn- (928) MnM 26 suck bzibs (pf.) -(fp) tifp 27 to fold ...... . fidp<t' (781) tsip 28 fish scale -hlip lidp (550) grip 29 to whisper ..

... .nip (667) yip

30 to pinch ...... (hnyap) . njap (66) hn.p (Ls) 31 fine, thin, slender zim-bu ......

did csdrm (1057) 32 to gather (of clouds) gtibs-pa 1 tim tim ......

sdim (Rgya.)5




Table 14. Medial -i- (continued)

meaning 0. B. Bu. L. Ch S. 33 house k'yim 'im in Jd cdm < g- (247) ...... 34 mud ddzim-pa t~im' ...... tsjidm ( 1165) ...... 35 tongs ...... ...... ...... cg'dtm (298) gim (Ls) 36 satiated ...... ...... ...... Jo tim' (246) In 37 to pierce ...... ...... CtsPim (1074) tsim 38 to close pid- (W.) pit ...... pit 39 extinguish ...... -mit m~it (621) ...... 40 one *id (W.) ...... . jt (175) 'e t 41 happy, felicitous skyid-po ...... ...... kjft (325) 42 anvil tsid ...... ...st < t' (1227) ...... 43 seven ...... ...... ...... ts'tit (1055) tet 44 heavy Idziid-pa ...... rit 45 calf of leg byin-pa ...... 'b'jen (736) ...... 46 liver mtg'in-pa -s&An tin

sin (Kanauri ) 47 finished, exhausted zin-pa ...... 'dz'4epn (1080) ...... 48 ripe smin-pa hma4' hmin ...... 49 nails (sen-mo) ...... ......

tsin-an (Tsangla) -S&A tin thin (Kanauri) ...... ...... ......

50 weigh ...... k'yin k'in ......

Notes to Table 14

1. Not in Lusei, but Mikir sek < *sik. 2. Bu. " small animal of tiger species." 3. I have not seen it stated before, but in 0. B. re-

duplication, the first part has the vowel a, the second the original root vowel, except the much rarer exact reduplication of the root.

4. Ch. " single, alone; single piece." 5. T~aurasya nkc'si, Khambu tfts-(-tsa). 7. Bu. "to scream." 9. 0. B. "partridge." 10. 0. B. pres. aktyig-pa. The form in the L. column

is Old Kukish. 11. Thebor tsi- (-pdA) " to bite, to sting," Garo talk-,

thlk- " to bite." Ch. " sting of poisonous animal." S. " to give a blow with the beak, to pick."

14. 0. B. pres. dtg'ift-pa. 16. Bu. "cloth." 17. Pankhu mrit, where final approached a palatal.

L. riig " loud." 19. Ngente dialect of Lusei and in Pankhu (also Cen-

tral Kukish). 20. Ch. "firewood, fuel." 21. Bu. "fluttering." 24. Lao [ram]. 26. O. B. pres. addzibs-pa. S. "to sip." 27. S. "to plait, to roll up," Lao thfp "to fold." 28. Ch. "fishes with bristly fins." S. "fin (of fish)."

For the loss of S. g- before -r- in Ch., cf. the word for indigo." 30. Bu. "to pinch, squeeze; blacksmith's tongs." Ch.

" pincers, tweezers; to pinch." S. " to seize with pincers; to pinch," Lao "to pinch, to press." There may have been two roots, *nip " to pinch " and *nim " low," which due to vowel gradation, vowel leveling, and change to finals from stop to nasal consonant, have become probably inextricably interwoven in meaning and phonetics. Thus the idea of pinching and squeezing are naturally

associated, and from the latter, to squeeze down, to press down, etc. A few of these TB words which seem most closely to resemble the forms cited here are Rawang mag-nm "to nip, to pinch, tweezers," Bu. hnip "(press on or down, to crush," nim' "low," hnim " bring down; oppress," Bahing nip- (tr.) nim- (obl.) "to compress or express," L. (nem) " to press," (hniam) " low."

31. Ch. "thin silk thread, fine, delicate, small." 32. Rgyarong, Bu. "cloud." L. "dark." 33. The reason for L. -n is not known. It is however,

an intra-Kukish problem, as final -m is preserved in Southern and Naga Kukish. The connecting semantic link with Chinese is afforded by N. Naga Kukish yfm " village." Ch. " hamlet."

34. Bu. "to ooze through; swamp, swampy ground," Ch. " to soak, tinge, go through."

35. See also K. 380. 36. Lao 'jm. 37. S. "drive in, strike in (enfoncer, faire penetrer),"

Lao tUm "put (a wedge in a piece of wood) ." 38. Kanauri c. p. pi-pi. Cf. Ch. N pjwi' <-dc "to

shut" (717), 8f piei' "to shut" (609). S. "to shut, stop up, to cover, hide."

39. Ukhrul -gi-mit, Rawang h4-mit, Abor mit, Garo ki-mit-a.

40. Lao 'at. 41. Ch. " auspicious, lucky, good," also "fortunate,

prosperous." 42. Ch. " iron block." 43. Lao MM. 45. Ch. "knee-cap, knee, leg." 47. 0. B. probably the perfect of a present ddzin-pa,

as suggested by JAschke. Ch. " exhaust, use up, finish." 49. Tsangla gad-ia "hand, arm." Besides the above, cf. ' cts8'ang "green" (1085),

Ka. kg-siA "green (vegetation) " (Hz.), ka-tsni "green, raw, unripe " (Hn.).



22 SHAFER: The -Vocalism of Sino-Tibetan

With certain exceptions to be noted below, the equation for the above table is *-i- = 0. B. -i- Bu. -i-=L. -i-= Ch. -id-= S., Lao -i-, -i-.

The above equation for Burmese would have applied to proto-Burmese. At the time of the intro- duction of writing into Burma, this medial -i- had affected the final and then disappeared. The actual equation, therefor, for Burmese is *-ik = Bu. -ats, *-in Bu. -an, *-in = Bu. -ax. The process was

probably *-ik > -dik > -ats, *-in> -dih > -an, etc. The above equation is somewhat obscured in the table by the interchange of -IS and -i which occurred to an unusual degree in proto-Burmese in the words in the first part of this table, nos. 10, 19-23.

We do not know what caused this last mentioned change. One might consider that it was largely a peculiarity of Burmish, as LIusei generally agrees much better as to final consonant with 0. B. than Burmese does, except that in four of the six roots where the interchange occurs in Burmese, a final dental consonant is found in Chinese (nos. 10, 20, 22, 23) where a final -k or -ng would be expected. In Lusei a final glottal stop after -h frequently changes it to -n. Whether that was the cause of the change in Bu; and Ch. must be determined by further investigation. See below for Ch. -ie-.

In Kukish the process of palatalization of the final by -i- has not been carried so far as in Bur- mese. Both -it and -ik occur in Luseii and other

languages of Kukish proper, but the conditions under which one and the other occur have not been determined.

The Chinese words in the last part of this table, no. 40 and following, come in a different class from the preceding. Karlgren's -i- only occurs before dental finals, and the writer at first considered that the original vowel was *-i- and that the -e- was an epenthetic vowel occurring in Chinese and peculiar to the dental final (cf. proto-Bu. *-aik < *4ic mentioned above). But the Daic comparisons indicate that this conclusion is incorrect. For the two Daic comparisons show S. -e-, Lao -ed-.

But contrary to the Chinese -ii- which occurs only before dental final consonants, the Daic equation S. -e- Lao -ed- occurs before other finals:

n 'zzip < A- "to enter, go in, penetrate; to cause to enter, bring in, put in " (943), S. hnip " to drive in, to insert, to put in," Lao hnedp " to insert, to prick "; Thebor, Kanauri kep "needle," Mantiati tsab, S. 1cem, Lao ce'am, but 0. B. [1atb]. Not enough comparisons involving this Daic equation have been found to be able to do more than pose the problem.

In table 14, it was observed that Ch. -ta- is equivalent to S., Lao -i-, -4-. But Wulff has pre- sented enough comparisons to show that in some words the Chinese phoneme is equivalent to S. -eye-, Lao -yd- (-ie-):

Table 15

meaning Ch. S. Lao

1 beautiful M ram' (35) iyfm 'yam (iem)

2 a hoe; sharp-pointed, 2 csidm (862) steem 8yam (siem)

sharp 58tYEm 8yam (siem)

3 corner , cledm (552) hliyfm hlyam (liem)

4 first month E ctsidng (1198) ts'yoi& (tsiei)

5 voice, sound v ciadng (397) siyan syai& (sieA)

6 to change k pwdtin' (590) play ......

7 continuous ; cltin (551) lyan (lien)

Notes to Table 15

2. S. (1) "small pickaxe," (2) "to sharpen to a

point." Lao (1) "straight spade," (2) "to sharpen to

a point, pointed."

5. S., Lao " word, language, sound." Rawang k'a 814-

ni " voice " (ka " speech ").




While two Daic phonemes may have fallen together in the Ch. -id-, it seems rather that this merely represents the samprasdrana gradation in Daic (this appears most clearly in Lao) of an original -i-; that 1 spidng "cake, pastry" (741), S. pen "meal, flour," Lao pen " cake, bread" represent the low and guna grades respectively,

and perhaps similarly with A# sJaim " thread" (1125), S. sen, Lao sehn. L. ptiin, Lao byah (ptieh) " flat," may represent the gunia and samprasarana grades.

Medial back vowels must be considered in connection with the final consonant.

Table 16. Final *-uJk


1 poison dug tok d'uok (645) 2 to kindle dugs-pa ...... ...... ......

to burn tug tok ...... ...... ...... dug5 Brgfo 3 to sit ddug-pa -t'ok ...... 4 six drug ktrok ruk * liuk (563) 5 to bore p'ug (pf.) p'ok ...... 6 cavity, excavation sbugs puk b'iuk ...... 7 thick stugs-po t'uik tt'k X tuok (etudes, 894, no. 25) 8 to call k'ugs-pa ...... kuk R k'uk (498) 9 to grunt nug-pa ...... Alk ......

10 to erect btsugs (pf.) tsuik fuk (intr.). ...... 11 belly, stomach ze-bug watm-puik *puk ,J piuk (53) ...... 12 dark color smug-po ...... muk ...... ...... 13 dark, foggy rmugs-pa muik ...... ...... 14 owl ...... th'im-buk *biuk ...... 15 radish la-p'ug ...... ...... & *b'iuk ...... 16 bent kug ...... *k'uk kiuk (487) 17 to bend bgug (pf.) ...... guk 18 to cuckoo kug-rtse ...... ...... *guk... 19 to heap, accumulate, spunts (pf.)

pile up punt- (Kanauri) punA' W b'iuk (52) ...... to fill buil-5

20 to guard, keep, watch bsruns (pf.) hruA' ...... 21 to imprison ...... k'un ...... guik

22 to roast ...... cxung (474)Ls

23 stone lun (Tsangla) kyok lun. 24 finger ...... -k'y6 ZU ......u. 25 river klun& k'y6.. 26 lofty ...... csfung (308) AsAn (Ls) 27 tree, trunk ...... 'd-k'uini ku.i.

'd -k in' 28 to feel pain, to love gduns-pa ... cd'uong (1145) ...... 29 cooked, ripe ...... . uk<d'- (912) sik

Notes to Table 16

3. Bu. in dA-tVok "to kneel." 5. 0. B. pres. dbugs-pa. 6. L. "cave." Ch. "to make a cave to live in"; cf.

0. B. pres. in note 5. 7. 0. B. also dtWug-po, mt'ug-pa, all employed much

as in the English sense of " thick," i. e. " thick (cloth,

liquids, woods) ." Bu. only in reduplicated form "thickly." L. "deep" (consequently doubtful if it be- longs here). Ch. "thick, substantial, firm."

8. L. "to shriek." Ch. "to cry, weep, lament." 10. O. B. pres. ddzugs-pa. 11. O. B. "'maw or fourth stomach of ruminating

animals." Bu. " the outside of the belly." Not in L. or



24 SHAFER: The Vocatisrm of Sino-Tibetan

Kukish proper except S. Kukish, but in most of the other Kukish groups.

12. Perhaps cognate to no. 13.

13. S. (hma'sk) "fog, obscure," Lao (same form) " fog."

14. Ch. based on Korea pok, Hakka fuk, Pek. ful.

15. 0. B. also gui-dmar-la-p'ug "carrot." Ch. based on Sh. vo " turnip," C. fuk " scabbard," Hak. fuk, Kor. pok " kind of turnip," Sw. hok, Pek. fu " kind of turnip." There is another pronunciation of this character *bak ( ?'): Sh. bo " turnip," C. pdk " turnip, radish," Hak. p'et, Sw. pOak.

16. Ch. also "to bend." Kukish *k'uk "knee," see note to No. 17.

The equation for Table 16 is *-u Bu. -o-, -ui- = L. -a-, -u- = Ch. -u-, -uo- = S., Lao -it-, -ii-.

The writer originally considered the possibility that Bu. -o- and -ui- are the Bu. correspondents of *-My and -'f- respectively, that L. -u- might correspond to Daic -u- and L. -it- to Daic -ti-, and that Bu. -ui- and L. and Daic -t- came from the loss of a final *-s. The table will show that there

17. 0. B. pres. dgugs-pa. S. " to fold, to bend (as the knee)."

18. Kor. kuk, Hak. k'iuk, Sw. kiok, Pek. chii. Sonant initial based on aspiration in Hakka.

19. 0. B. pres spuA-ba. L. " to increase," Ch. " to fill; full."

20. 0. B. pres. sru-A-ba. Ngente dialect of L. 21. S. "prison," Lao gak. 22. S. "to cook," Lao " to cook in steam." 23. Dial. of Tavoy kid'. 25. S. (gla'an). 27. Bu. (1) "stalk or branch (of fruit or flowers),"

(2) "large branch, bough of tree." Cf. also la cpungg<-m "wind" (18), Ka. buW "to

blow," n-bun "air, wind, atmosphere."

is no good evidence for any of these possibilities, although this may be due to lack of evidence. But at present it seems rather that vowel length and Bu. -o-, Ch. -uo- have arisen independently in each linguistic group, the last two due to vowel gradation. The Ch. -uo- may be the samprasarana grade, *ua-, which has become -uo- by assimilation.

Table 17. Final *-ok, -on


1 below hhog 'ok ...... ...... ......

2 helmet rmog mokl. (hmw.9k) (Ls) (krok (intr.)

3 to be afraid skrog-pa k'rok (tr.) jDw'kjong (470)

4 servant g-yog yok ......

5 to hit, strike, touch p'og (pf.) ... p'uk (757) ...... 6 to heap together, to jumble, ddzog-pa (pres.) dz'uk (1106) ......

to throw disorderly together

7 to assemble, to gather, tots'ogs (pf.) ...... tg'uk (1106) ......

meet; an assemblage (of men)

8 rice ...... kok kuk (506) ...... 9 to wear, to dress plog (Kanauri) ...... b'jsk (43)

9a pulley ....... ...... lk (576) r'ak

10 thousand Stof' ...... ...... ......

11 onion btson ts8'oi (WW. B.) . .ts'ung (1120) ......

12 tube; quiver donA-po tonA' ...... cd'ung (1150)

13 tend (cattle) skyon-ba (pres.) kyA .......

14 small coin donA-tse . .cdng (1150) d9'ai (t'Vo) (Ls)

15 trade tWMon t'oi .......

16 to burn .on& ...... clung (585)

17 boat 1. lon ......

18 cubit ...... -to-& toA.

19 to unite bdofis (pf.) ...... ...... cd'tung (1150)



SHAFER: The Vocalsm of Sino-Tibetan 25

Table 17. Final *-ok-,oh (continued)

meaning 0. B. Bu. L h. S. 20 wild yak dbron pron ...... 21 to grasp, seize blons (pf.) ...... . . l 'lung (585) 22 buttocks, sitting part dp'ons pon ...... ...... ...... 23 the above gonA k'o.. 24 obscured rmons (pf.) hmo ...... m cng (611) ...... 25 want, need, indigence gyo ...... ...... gn c'fg (472) ...... 26 face gdo ...... d'ung (1149) . ...... 27 summit Itons to ...... 'tiwong < t- (1251) ......

28 size, volume bon

...p'iwong (31) tcamel rfa-bo.(

29 insects bon-bu ...... ...... & 'pong (31) 30 to sow tA'ok' (Vayu) )

to cultivate (earth) ti'oA- (Bahing) 3 .j ......t ong<t' (1270) 31 crooked k'yog-po . . k'iwok (492)

bent dk'yog-po 31a curved, crooked gyog-pa . g'iwok (485) ...... 32 sufficient th'og-pa ...... .. tsPiok (1104) 33 shell, rind kog-pa, skog-pal k'ok ...... k'ak (506)

bark p'yi-kog 5 34 to bake, roast pog- (W.) pn ...... pdk (698) 35 what is uppermost tVog .. t'dk < t- (1234) ...... 36 to flay, peel . ...... gJ pak (574) pa'?k (pok) (La) 37 flower me-tog ...... t.'ak mMi (Ls) 38 two (zun) ...... . . dng<s- (1243) 89'.tA (son) (Ls)

Notes to Table 17

2. Bu. "hat," S., Lao "hat, cap." 3. Ch. " to fear." 4. Bu. "man." 5. 0. B. pres. dp'og-pa; the verb does not necessarily

mean to strike in order to hurt. Ch. " to beat, strike against."

6. 0. B. pf. btsogs. Ch. "to collect together; clan, family, tribe; class, kind."

7. 0. B. pres. dts'ogs-pa. Ch. " crowded," also " to collect; a crowd."

8. Ch. "grain, cereals." 9. Ch. " to put on; clothes." 13. 0. B. pf. bskyafis. 14. Ch. "copper, bronze." S. "gold," Lao "copper." 16. Ch. "fire; to light, to warm." 19. 0. B. pres. sdofs-pa; cf. daA i' with." Ch. " fit

together, join; together with." 20. Bu. "bison." 21. 0. B. pres. IoA-ba, len-pa.

The equation for Table 17: ST. *-o- = 0. B. -o- Bu. -o- =- L. -o- = Ch. -u-, _iw-o, 4- c S., Lao -4'9-, before guttural consonant finals.

This hardly presents an accurate picture for Lusei. Briefly there are two types of o, the ordi- nary short -o- as the British pronounce hot, and -d-, the latter probably corresponding to a long

22. Bu. "thigh." 23. Bu. "top, summit." 24. 0. B. pres. rmoit-ba. Bu. " very dark." Oh. " dim,

dark." 25. Ch. " empty, exhausted, helpless, poor; limit, end,"

etc. But cf. the 0. B. word also with 31a, note. 26. Ch. "pupil of eye." 0. B. and Ch. words differ

considerably semantically, but both are perhaps cognate to 0. B. mt'oA-ba " see."

27. Bu. "mountain," Ch. " mound, peak." 28. Ch. "peak of a hill," also "hump of a camel." 29. Ch. "bee, wasp"; semantically closer to 0. B.

(buW-ba) "bee." On the other hand the Ch. word may be compared with 0. B. sbrafz-bu " bee " < *sbroA-bu1 by vowel levelling.

30. Ch. "sow, plant." 31a. Ch. " bent, cramped; in straits, poor." 33. Bu. "bark." Ch. " shell, husk." 35. Ch. "high, elevated." 38. 0. B. " a pair, couple," Ch. " a pair, couple; both."

vowel in the other vowel series of LuSei. Besides this there is the diphthong -ua-. All three phonemes correspond to an -o- in other TB languages. One would expect that L. -ua- would correspond to Bu. -wae. Sometimes it does and sometimes with Bu. -o-, leading to the conclusion




that each group developed its samprasdrana grade independently.

One might draw a similar conclusion regarding the development of the guna grade from the pre- ponderance of -u- over -iwo- in Ch. in the above table, except that Daic has -a'a- almost regularly for TB -o- in the few Daic comparisons found. However, there are not enough of these to establish definitely the equation for Daic, and consequently

the question of whether ST had an -o- must be left open for the present. In that connection, one must consider the fact that Burmese has no -o- before dental or labial finals. This, together with the fact that many of the -d and -n finals in 0. B. are not real finals but suffixes, makes it impossible now to present any conclusive data on medial -o- before dentals and labials.

Table 18. Medial *-u- (before dentals and labials)

meaning O. B. Bu. L. Ch. S.

1 to blow -mut (Rgya.) hmut ...... ...... ......

2 to rub bsud sut ...... MIJ (fwat) <s- (920) tsuit

3 to suck ...... ...... dut f iwt t<t- (1235) tut (dut)

4 to boil (tr.) prut ...... 'a p'iuat (47) ...... 5 to blow dbud-pa (pres.) t...... ...... ...... v

6 dust cloud bud p'ut ...... ...... f fun

7 dirt, dung brun . ..... .piun' (199) ffin (Lao)

8 noble, honorable btsun-pa ...... ...... tsuan (1112)

9 warm water ...... ...... .cuan (1316) 'fn

10 part ...... ...... ...... b'uan' (29) buzn (Lao)

11 all kun -kun ...... kiuen (292)

12 to wrap dt?'un-pa ...... fuin ......

13 obscure {dmun-pa hmun

14 box ...... Pup ...... ...... 'up (Lao)

15 to suck ...... tsup (Meithei) tsup 16 hole rlubs lup 17 rough rtsub-po sup ......

18 to gush, to spout forth, overflow dbrub-pa mrup ...... ......

19 posteriors rkub lan-kup ...... ......

20 pendant ears ...... tdp (11) I.-tp (Ls)

21 round zlum-po liarh hlum ......

22 three gsum suth tVum - csam (766) sdm (Ls)

23 belly ...... ...... pum ...... pdm (Lao)

24 100,000 dbum ...... pum ...... vum

25 warm ...... lumh lum ......

26 block, bench, stool, table, desk ...... k'uiii k'um ......

27 paddy mortar ...... tg'urh sum ...... ......

28 boiled grain for brewing glum gyumh ...... ......

29 contracted (as limbs) dk'um-pa kum' kum gum (k'um)

crooked k'ums-pa k.Im ...... ......

to contract skum-pa ki'tih ...... ......

30 to stoop .num' ...... ..m 31 to cover klubs (pf.) k'yu ...... ..... .... gl

k'rumh 32 bulge 'umh (n.) um (v.) ...... 'm

33 bud mum (arch.) -mum ......

34 to hold in the mouth ...... ...... *um Pt c-dm (238) am




Notes to Table 18

2. Bu. " to wipe." Ch. " wipe clean, cleanse, to brush." S. " rub, scratch."

3. Lao tMU (dut). 4. Ka. prut "to boil, bubble (as boiling water) "

(Rn.). Ch. " to bubble up, boil." 5. 0. B. pf. bus. S. " to blow (as for lighting tinder)"

(see final -u table). 6. Bu. "dust lying on the ground." S. "very fine

dust." 9. S. " lukewarm," Lao 'un. 13. Bu. "to be dim (as the eyes), dusky (as sky)." 15. L. (fdp, fdp'). Lao tkap. 17. 0. B. " uneven, rough (as skin, cloth, country),

bristly (hair)." Bu. only in redupl. "long and shaggy (as hair and grass) ."

18. Ka. prup " to gush forth." Bu. "to be sub- merged, overwhelmed, swallowed up."

19. Bu. "nape of neck" (i. e. "neck-posterior" ?). 21. Bu. " to make round." L. "ball." 24. L. " all." S. " abundant." 26. L. " bedstead." 28. Bu. "wheat."

The equation for the above table is that, before dental and labial finals, -u-= 0. B. -u- = Bu. =u - L. -u-, -it- = S., Lao -ii-, -Ii-. There is no indication that the length of vowels agrees in L. and Daic.

As for Chinese, the table must be divided in two parts, those words with dental finals and those with labial finals. For the first of these, the equation is *-- Ch. -ua- and for the second *-- Ch. -a- in the majority of words compared.

There is no medial -u- in KarIgren's transcrip- tion for dental finals. Of all the dental finals found in Karlgren's Etudes, -uan, -uat not only look like the most probable correspondents of ST -un and -Ut (and were so considered by Simon and WulfE working independently, as well as by myself), but the comparative material, meager as it is, also indicates this. The relationship of the Ch. words entered under nos. 3 and 11 to the ST words entered there, is therefor doubtful. But regarding no. 11, it may be pointed out that in Table 14, it was found that Ch. -ij- corresponds to TB -i-, i. e. the -e- of the Ch. diphthong is not represented in TB. Therefore one may infer that the -e- of Ch. -ug- is likewise not represented in TB.

Regarding the conclusion that ST -u- is represented in Ch. by -a- before labial consonants, it should be noted that Karlgren has reconstructed no back vowels, and no diphthongs of which back vowels are an element, before labial final con-

29. Bu. (1) "in a contracted, shrunken condition," (2) "arched, convex." L. "concave." S. "arched, concave."

30. S. "crooked." 31. 0. B. "to cover (as the body with ornaments),"

pres. klub-pa. Bu. " to cover, overspread (as with cloth, blanket or clothes)," S. " to cover," Lao gum. The only way of determining whether a word heard in modern Burmese had -y- or -r- is to find the same word in the old literature or Arakanese. Where Judson was unable to do this, he gave both forms. The comparisons would indicate -y- as correct.

32. L. "to develop or enlarge (as the breasts)." S. "swelled, inflated."

34. The form in Lugei column is a reconstruction for Southern. TB, rather than for Kukish. It is based upon Mikir om (*u- before a final consonant usually but not always becomes o- in Mikir; the reason for the occasional retention of u- is not known), Rawang im (Rawang i- < *i- or u-), Ka. md-urm (Rn.). These are all consistent with' a TB *um. However Tiungli (Ao Naga) has am where we should expect em from *um. However TAungli am does not argue for *am which should likewise become em in Tiungli. Lao 'am.

sonants. Even the phoneme, if such it was, which he writes -w- is only found after labial initials in the labial final series, leading to the conclusion that it arose from the labial initial and not from the final or as the representative of a labial vowel. One may also point to FW tap, transcribing the Prakrit form of Skr. stf2pa (K. 955).

Unfortunately for such a conclusion, two of the Ch. words in which *-X- appears to have become a are found in S. with -a- and -a- respectively (nos. 22, 34), although the table shows that *-u- is generally retained in Daic. Whether the change in Daic is due to initials, tone, or some other factor is a problem which may prove difficult to solve because of the paucity of Ch. labial finals.

In the previous tables, no ground was found for believing that long vowels correspond to long and short to short in the only languages where length has been distinguished,-Lusei, Siamese and Lao. However, the materials were scanty.

On the medial *-a- of ST we have much more abundant materials, yet many of the comparisons are between Lusei, which distinguishes length of vowels, and Bu. or 0. B. which do not; or between Siamese and Lao on the one hand and 0. B. or Burmese on the other. Yet because of the greater amount of comparative material and because no other serious problem is involved in the consideration of ST *-a-, the factor of vowel length will be the main consideration in the following table.




Table 19. Medial *-a-


1 phlegm ...... k'dk ...... kl'k (Ls) 2 to cackle ...... ...... kdk OR kdk (41 1) 3 branch ...... -kak kdk *kdk (or) *k'dk . 4 granary ...... ... ... ts'ang (1036) tA'4i 5 earthen cup or bowl ...... . . a d'anng' (210) 'an

6 steel ..... ... .... ckang (304) k'4t (Lao) 7 to rise, raise .

... .. cngdng (5) nan

8 betel-nut .... ... ... . cdng (514) la I (Dioi) 9 saddle ...... ...... ...... c an (4) an

10 curtain ...... mudn' (595) man 11 pipe ...... ...... ...... 'kudn (447) kan (Lao) 12 indigo rams ...... Am (376) grdm 13 deep purple, violet ..

... .km' (298) karh (Ls)

14 hairpin ...... J ctsdm (1028) tgd*h 15 bold, daring ....*. ...... ...m... J 'kdm (300) kd*-hdn 16 humble ...... ...... ..... l'k'dm (147) kldm 17 to cut, engrave ...... ...... ...... dz'dm' (1165) tsarn 18 to put or throw into gains (pf.) ....... cy.am <g- (62) gdam (Ls)

the mouth agam-pa (pres.) .cYam < g'- (149) 19 noble, brave dam-pa ...... .tam (1166) tam (Lao) 20 abundant ...... ram ...... 1dm' (376) ...... 21 to carry on shoulder ...... tm ...... ctdm (1166) 22 shore skiam kta'm (arch.) kam k'am' (303) ......

k'dm 23 road lam Ilm lam ...... ...... 24 to smell (tr.) bsnams (pf.) na.m ...... ......

(intr.) innam-pa nam h nam 25 hair sam ...... sam

ts'am (W.) ta ......am<s- (850)

26 tongs skam-pa ...... the-gap (Meithei) ;W kap (345) rkam-pa

27 pole Itgam hloar ...... ...... 28 mouth k'aih- kam ...... 29 bear (n.) warn (Bunan) warh ...... ...... 30 side (of body) snam- namh ...... ...... ...... 31 to weave btags (pf.) ...... ta'. tdk

dt'ag-pa pres.) *tak' (or) *t'ak ...... 32 to die ...... ...ina... m cmmWang (1299) 33 kill bsad (pf.) sat t'at' ? sat<s- (1154) ......

gsod-pa (pres.) 34 eight brgyad hrats riat A pwat (682) 35 goose Aai t. an ...... nan' 244) han (Ls) 36 much, many inan-po ...... mai c mng (598) 37 to eat gzan-pa ...... cts'dn (1030) th'dn

food bzan tstdn' (1030) zan

38 kind of turnip ...... ....... ...... cm4n (595) mdn (Ls) 39 shield, buckler ...... ...... ckdn (296) kdn (Ls) 40 to divide in half, half a.

... .*ptn $ pudn' (689) pdn

!IJ p'udn' (689) 41 box sgam ...... ...... f ck'dn (71) kdp (Lao) 42 elephant ...... .zi 'ang <dz- (787) dzan (Ls) 43 high ...... ........ n' f.. iiang'< d'- (856) . 44 artisan, workman ...... ...... ...... dz'iang' (385) dzian 45 young lady, woman ...... ...... ...... * ifnjiang (541) ndan (Ls ) 46 style, kind, sort .... j. ........ . ang' <z- (212) 'yani




Notes to Table 19

1. S. " to spit from the bottom of one's throat." Lao "to spit."

3. L. "fork (of tree, etc.)." Ch. is based on Korea kak "a branch" (Gale 904). Korea kak can only be from kdk, kjiak, kak, and the corresponding aspirate initial. Since the phonetic 1f nearly always occurs in words pronounced kak, lk, the reconstruction may be considered almost certain. The character is given by Karlgren, no. 411, with a different pronunciation and meaning. If S. kak1 " crossroad " is the same root, this entry should have been placed with those where the vowel length does not agree.

4. Lao sdn "storeroom, treasury." 5. Ch. "bowl, basin," also "jug." S. "water jug." 6. Ch. also "worked iron." Lao "melted iron." 7. Rawang &An "to climb, to be steep, to ascend."

S. " to lift up." 9. Lao 'In. 10. Lao hmdn. 12. Chinese has the phonetic ckam'. 13. S. "blood red," Lao " dark, black." The meaning

"purple" occurs with this root in Can. 15. The simple root occurs in Nung and Dioi. 16. S. "to respect, to fear." 17. S. " cut, carve." 18. Ch. " hold in the mouth." S., Lao "mouthful." 19. Ch. "gall, courage." Lao "to dare." 20. Ch. " to overflow; excessively," also " lavish." 21. Ch. "to carry on a pole," also "carry on the

shoulders." 22. Bu. " cliff," Mod. Bu. kam. L. (1) "bank, shore,"

(2) "precipice." Ch. "bank, cliff."

The writer is well aware that Karlgren did not employ the circumflex mark over a to denote length of the vowel, but only to denote the quality of the vowel. But just as in the Aryan languages of India the quality of the vowel depends in general on the original length of that vowel, so seems to be the case in the Sino-Tibetan languages as nearly as may be determined from the descriptions of the phonemes. The question here is whether Karl- gren's -a- corresponds to -a- in ST languages other than Ch., and his -a- to ST -a-.

The above table falls into several parts. Nos. 1 to 19 show L. -a- = Ch. -a- = S., Lao -a-, while nos. 25, 26, 31-34 show L. -a- Ch. -a- = S., Lao -a-. Nos. 35-46 show evidence in conflict with the equations just cited. But of these apparently conflicting examples nos. 42-46 fall into a special classification as they all have initials followed by -i-. Karlgren's ?Rtudes show no -8- vowel after -i-. One may infer from this either that the vowel was originally short as in Chinese and that the L. and S. initial absorbed the -i- with compensatory

23. While L. shows short a in this word, Dimasa has

24. 0. B. (1) pres. snom-pa. 25. 0. B. sam only in comp. for " moustaches." 26. Ch. " pincers." i kap " a smith's pincers " (345).

Ka. lo-kap " tweezers " (Hz.), Abor, Miri me-gap " tongs" (me *" fire").

27. Bu. "spear." 28. Bu. in comp. only. 31. L. is quite irregular, if not incorrect, as the other

Kukish dialects indicate *tak', tak'. S. "to plait," Lao tdk.

32. L. "to die, die out, become extinct," Thado "to lose," Meithei "lost," Kanauri "to conceal," Ch. "to die, pass away, lost, disappear."

34. S. pet, Lao pet, but Nora writes pyat. 35. 0. B. Aan < ian by assimilation of final to initial.

Ch. " wild goose." 36. Ch. " vast," also " plentiful; great." 37. S. "to eat (of bronzes)," Lao s4n. 38. S., Lao " yam, potato." 39. S. "defend against, shelter, protect," Lao "pro-

tect." 40. Ch. (2) " cleave, separate, distribute; half." S.

"to carve, to cut," pan "piece, slice," Lao pan " to split," pan "piece, part," pan "to divide." The reconstruction in the L. column is based on W. Kukish: Kwoireng ka-p'dn, Khoirao a-p'an "half."

43. Cf. also ? iiang' < d'-" up, above, upon; top, summit " (856).

44. Lao dzan. 46. S. " fashion, model, custom, species," Lao ydn

" manner, fashion."

lengthening of the following vowel, or that the vowel was originally long in ST but that it was affected in Ch. by the preceding -i-. These last five examples are therefor probably not really in conflict with the above equation.

Excluding them from consideration, the examples pointing to the validity of the above equations are 25, those against 7. Quite a number of variations in the length of the vowel will be found in these tables and notes between languages as closely related as Siamese and Lao. Therefore considerably more fluctuation is to be expected for languages as diverse as Lusei, Chinese, and Daic. Therefore the evidence points rather strongly to the conclusion that there was both a long and short a in ST. However the evidence is still very weak regarding Lusei.

It will be noticed that Burmese has two phonemes corresponding to -m in other TB languages, one written -m and always associated with the falling tone and other written -in and always without the falling tone, when following -a- (nos. 20-30). In




the first group, with -m, the three words with Ch. correspondences have -'- in Ch., while in the second group, with -im, the one word with a correspondence with Ch. has -a- in the latter language. There is not sufficient evidence as yet to show whether this is significant or accidental. It was seen above that a long -J- in Lusei, Siamese, or Lao or an -a- probably indicates the root had a long vowel in ST. Consequently we may infer that the vowel in the following words had -d- in ST: 47) 0. B. mag- pa "son-in-law," Bu. -mak, L. mdk-pa; 48) Bu. hak "to hawk" (intr.), "to gag" (tr.), L. hdle " choke "; 49) 0. B. ihags-pa " cold wind," J# *laxl " cold, icy, frozen " (reconstruction based on Kor. hak, Hak. lok, hok, Sw. ole, hak, Pek. lo; if the x preceded 1, we should probably have had *xiale which is not the case as shown by the lack of yodicization in this word in Hakka and Swatow); 50) 0. B. bag " narrow space; slight, insignificant," cognate to sbeg-pa " lean, lank, thin," * balk " thin, slender, narrow, small " (764); 51) 0. B. bleag (pf.), dgegs-pa (pres.) "to hinder," 3 *kalk "to stop, hinder " (411-A) (Karlgren calls this a " re- cent character," but the word exists in all the dialects of China from Korea to Annam and in Japan, and the 0. Ch. pronunciation is certain), cf. S. klc "to detain, shut up"; 52) 0. B. mtstah, dtscah "error, X tstale (777); 53) 0. B. sgah "hill," Bu. ktah- (259),14 cleAh "mountain ridge " (304); 54) 0. B. pah, p'ah " bend or hollow formed by belly and thighs in sitting, lap; bend or hollow formed by arm and chest in carrying something," Mikir p"ah "shoulder," fM *'pang "shoulder," 'ptang " thigh bone " (694); 55) gah-bu, Igah-bu " husk, pod, shell," Q ckldng "husk of grain, chaff " (306); 56) 0. B. rdzai " storage chest," A dz'ang' "storeroom, treasury" (1034); 57) 0. B. mdzahs-pa "wise, gentle, noble," bzai-ba "good # ctsing "commendable, good, right, happy " also "virtu- ous" (1034); 58) 0.B. btsad (pf.), gtsod-pa (pres.) " to cut," L. s&t'; 59) Bu. lean' " sulphur," L. kdt; 60) 0. B. brtan-po " firm," W "tan " solid " (966); 61) Bu. leyan "to be well," tldn "to be in the habit of eating or drinking anything "; 62) Bu. han' "to stop" (tr.), L. han "to cease"; 63) Bu. Ican' "to mark across, intersect," L. lcan "to cross"; 64) 0. B. han-pa " bad," h9 ngdt (959); 65) 0. B. Ian btab (pf.), pres. Ian addebs-pa "to answer," X top (71), cf. S., Lao t9'9p; 66) 0. B. yab, g-yab " fan," Bu. yap, L. zap' (v.); 67) O. B. gtam " talk " (n.), gdams (pf.), pres.

ddoms-pa " advise; advice," X cdtam " talk, chat " (969); 68) 0. B. latm " appetite," lcam-1atm " de- sire, longing for," Lao Palm "avide de, aimer"; 69) 0. B. lPam-bu "apricot, peach," leam-g-yag "cherry," ffj cca'm "mandarin orange," iJ ckam "sweet, nice" (298), Lao la'm "tamarind," but Nungish gam " to be tasteful," larm zi si " orange, lemon"; 69a) 0. B. yah-po "light (in weight)," L. zdan.

By a like criterion, the following roots probably had a short -a- in ST: 70) Burmish-Loloish *rak "fowl" (Maru ra < *rak, Northern Loloish *rak, Bu. le-rak), L. -rale "duck"; 71) Bu. lyalc "'to lick," L. iakl', cf. S. Riyea (hla), Lao hyea (is the loss of -Ic due to the final glottal stop found in L.?), O.B. bldags (pf.), Idag-pa (pres.) look as if they also belong here, but the 0. B. Id- would probably be reflected in L. as it or ti- and the most plausible explanation is that once there were 0. B. *Ideg-pa (pres.), *blags (pf.) as the dropping of the -d- of Id- frequently occurs in the 0. B. pf. but an understanding of the 0. B. Id- will require an extensive study of Old Bodish archaic documents and its transformations in the modern dialects; 72) Bu., L. nah "thou"; 73) Bu. ranh "to deny," L. tah; 74) 0. B. rmah-lam " dream," Bu. maak (735), L. mah; 75) 0. B. nad "ill," L. nat'; 76) 0. B. btsad (pf.), gtsod-pa (pres.) "to break," L. tsat (see no. 58); 77) 0. B. snabs "snot," Bu., L. hnap; 78) 0. B. tVab " fireplace," L. tap; 79) 0. B. hab (given as meaning " mouthful," but used in expressions for devouring greedily), Bu. hap " to bite at; to snatch in eating," L. hap "to bite, to snap "; 80) 0. B. sram "otter," L. -hracm; 80a) 0. B. dah " and," L. dah "another," S. ddh " also, and," Lao ddh; 80b) Bu. kyak "to be cooked," 0. Bu. klyak (Tin, JBRS, 1933, p. 32), Bu. kPyak "to cook," L. tlak " to boil or cook without salt."

In the following comparisons there is no indication of the length of the vowel: 81) 0. B. hag "speech," Bu. cak (arch.); 82) 0. B. nag-po " black," Bu. nak; 83) 0. B. lag " hand," Bu. haic; 84) 0. B. 1cag-po "difficult," Bu. Pcak; 85) 0. B. gzags (pf.), ddzag-pa (pres.) " to drop, drip," Bu. tsakc; 86) gtsa'-pa " clean," Bu. tsah; 87) 0. B. tVah-po, Bu. tani (irregular regarding aspiration); 88) 0. B. tah- "fir, pine," Bu. tVda "firewood," t~'s-ru "' pine"; 89) 0. B. brnahs (pf.), rnant-ba (pres.) "to be choked," Bu. nah; 90) gnahs (pf.), gnah-ba "give," Bu. hnah; 91) 0 B. blean-pa "to press (the foot against)," Bu. kan "kick."



OBERMANN: Votive Inscriptions from Ras Shamra, 31

Chinese, Siamese and Lao -a- is represented in 0. B., Bu., and L. by what is written here as -a-: 92) 0. B. dzag "robbery," M dz'ak "robber" (949); 93) 0. B. dah "meadow," Lao dah "plain, the country"; 94) 0. B. dhahs (pf.), pres. dhah-ba "to be out of breath; to fear, to be terrified," S. iah " to be benumbed, stupefied, astonished," iAM "immove- able, numbed; 95) 0 B. btab (pf.), idebs-pa (pres.) "to strike," S. top, Lao top; 96) 0. B. tam-pa "complete, full," L. tam " much, many," S. N'm

"several, abundant"; 97) V ltudn "short, de- ficient (1015), S. tw9n " cut off, amputated, truncated," Lao tw9n "cut off, amputated, shortened"; 98) I ckcimc "metal, gold'" (386), S., Lao gaih (S. kim borrowed from Ch.); 99) M tang' "stool, bench" (982), S. tah, Lao tah "chair, seat"; 100) C CxiUqTn "smoke" (174),1 S., Lao gwdn (cf. Ka. and Luhupa 1ut); cf. also 101) Rawang a-1cah "permission," Ka. a-1cah, '

"kPng "to be willing, consent, permit" (314). What few correspondences there are between Ch.

and S. and Lao do not show any correlations between -a- in Ch. and the same vowel in S. and Lao but indicate rather that the shift from -a- to -a- or -a- occurred independently in each linguistic group. The only direct correspondence of Ch. -a-

with Daic -a- so far found is one of Wulff's comparisons: 102) a cpuan " strenuous, energetic " (709), S. pwan "without rest, with persistence."

In the above tables and comment not all the medial vowels and diphthongs to be found in Lusei, Chinese and Siamese and Lao have been considered due to lack of comparative material in which such vowels and diphthongs were found. The same ap- plies to *-r, *-I, and *-s, which will be considered at a later date.

But enough material has been given to show that phonetic equations for Sino-Tibetan languages are possible, contrary to the opinion of Wulff and others. For the ability to make these equations I have often been much indebted to the work of Wulff and others.

Much, it must be admitted, yet remains to be done on the vocalism of Sino-Tibetan-the deter- mination of the total number of vowel gradations, questions of whether a certain vowel was original Sino-Tibetan or due to vowel gradation in in- dividual groups, the cause for vowel gradation, the question of whether long vowels were original in Sino-Tibetan, etc. Many of these questions will probably have to be determined statistically, due to conflicting evidence, after most of the possible comparisons have been made. 1 Mikir i-hon is given only by Stewart.

VOTIVE INSCRIPTIONS FROM RAS SIIAMRA

JULIAN OBERMANN YALE UNIVERSITY

AMONG the mass of epigraphic material from Ras Shamra, only two inscriptions have so far been found to be of votive, dedicatory character. They were discovered during the campaign of 1934, and the first report of their find, together with photographs of the objects on which they had been inscribed, was published by Claude Schaeffer in the second issue of Syria 1935 (see the Plate at the end of this article). In the same issue, a treatment of the find is offered by Rene Dussaud, who with customary resourcefulness concerns him- self only with the reading and interpretation of the two texts as such. He makes no attempt at appraising their specific significance, though he is undoubtedly aware that the two inscriptions are

of a kind to claim critical scrutiny quite aside from the fact that they are written in the alphabet and dialect of Ugarit. It is this specific import of the find that I venture to submit in the present paper. It may be best demonstrated by consideration of three points: (1) the arrangement of the two inscriptions, (2) their paleography, (3) their opening formula. In consequence of the last mentioned point, moreover, we shall be led to an understanding of the two documents vastly at variance with that suggested by the French scholar.

I. Arrangement

It has been the tacit assumption of scholars that Ugaritic script is to be read as horizontal rows



the vocalism of sino-tibetan

Documents