The origin of the graph b in the Thai Michel Ferlus

To cite this version:

Michel Ferlus. The origin of the graph b in the . Abramson, Arthur. Southeast Asian Linguistic Studies in Honor of Vichin Panupong, Chulalongkorn University Press, pp.79-82, 1997. ￿halshs-00922729￿

HAL Id: halshs-00922729 https://halshs.archives-ouvertes.fr/halshs-00922729 Submitted on 8 Jan 2014

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée dépôt et à diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. The Origin ofthe Graph < 6> in the Thai Script

Michel Ferlus

Centlo National de [a Recherche Scienlifique, Paris

It is now a well acceptcd fact that the Thai script originated Aom a type of ancignt . This borrowing took plaÇe by the XIIIth century, at tho latest, because the scrip used in the most ancient text known in a Thai languagg the famous Ramgamhæng stele etgraved around 1300 AD, is considered as ao innovatior or the basis of a previous script. The Khmer writiog was adapted to Thai in a very simple way. Wheû the phonetic unit was the same in the both languages, the Khmer graph was tâkgn without any modification to tanscribe the same sound in Thai. For example, the graph tanslitterated by t and pronounced k in Khmer was taken to transcribe Thai L (today spelled a). In t1le same way, the Khmer $aph translittelated ,kü was take[ to transcribe Thai kh (oday spelled tt). The Khmer graph translitterated g ând prcnounçed g (today k) was taken to transcribe Thai g (today pronounced kÀ and spelled n). But when a phonetic unit in Thai was not represented in Khmer, the $aph of a similar sound was taken and modified. The Khmer gaph t was modiûed to]Kl, to transcribe the aûcient

Thai non-voiced dorsal 2tr (today pronounced kù and spelled 1l in the chat). In the same way the Khmer graph g was modified to G to transcribe the arcient Thai yoiced dorsal fricative y (today plonounced kt and spelled n in the elphabet chart)... The gaphs Kâ (1) and G (n) arc now obsolete and have been replaced by ftà (1|) and g (çl) iû lext orthography. Thai (regular graphs) (modified graphs) kk ft n k ti kh k l kn Kh "t x>kn g g>k g n g>kn G n y>kt In spite of somc modifications in the forms of lette$ over the centuri€s, the relatioû betwegn the regutar graphs and thsir derived forms has been preserved until today and is qüte clear in the modem Thai script. On the basis of zuch examples, it has beer somevr'hat hastily concluded that every gaph of the Thai script derived from a homologous one in the Khmer script. That is true for ahrost all of the alphabet, but for the græh à, the so-called plcr ptram (r.{), the relation with its supposed model in ancient Khmer script seems to us spufious. For the purpose of our demorstration we shall compare the ancient Thai gaphs for pà (tl) and â (t{) in Sukhothai style (Songvitaya l98l) aod Faklûam style (Vimonkasem 1983)

Southeast Asiai Linguistic Studies in Honow ofvichin Panupong, edit€d by Arthur S. Ahamson, Chulaloûgkom University Press. 1997. 80 FERLUS

with those in and in ancient Klmer scripts, Preangkorian (Grawengkij l98l) and Angkorian (Phirunsam l98l).

ph b

Pallava Khmer heangkoriaü fo Angkorian .r l, clô{2m Thai Sukhothai rJt9 rru znj Fakkham il s.,J w moderrl ,4 ÿi

A simple examination of the both gaphs p, and ô shows us üat for pà the Thai graph is clearly derived from the Khmer protoqpe, while for ô the forms of the Khmer and Thai graphs arc loo disimilar to be considercd as related. In ancient Khmer, the graph b was rarely used (Jacob 1960) because, in general, the graph ÿ was used to transcdbe the arcieot phoneme b as well as v (Ferlus 1993). This is due to the fact that the Indians who introduced cultule to anciant pronounced Sanskrit ÿ as b but kept the two distinct in Sanskrit texts. WTen the Pallava script was adapted to âncient Klmel the gaph ÿ was taken to aanscribe the two Khmer phonemes v and b, which let to ambiguity, as the graph à was rarely used. Until the xlth century the graph ô, mostly used in sanslsit, is mquestionably of Pallava t)?, but from this period a new form of ô, presumably borowed fiom Mot script, was intoduced.

before xlth c. C' CJ Cf from XIth c. ôn

Where does the Thai gaph ô come fiom ? Before any answer one must remember a peculiar consonantal shift in ancient Khmq. At an unknown date, the labial and æical non voiced initial were glottalized, that is p > 6 and t > 4 while the ORIGIN OF < b > IN TIIAI SCRIPT 8I

gaphs remained unchanged. As a result, in the Thai script the graphs p ard t repres€nt the anciert preglottalized initials 6 and d (today lJ and n), while the initials p and t are ûallscribed by modified forms of the same gaphs (today ! and n). Wlen the ancient Thai people borro,,ÿed the Khmer script ard, with a remarkable precisioû, adapted it ûo their own language, they used the gaph ÿ for the sourd w, approximately restoring ils origiûal value in Sanskrit. They then needed ûo create a new graph to tmnscribe the sound b, for \rhich no letter was available in the ancient Kbmer system. Our hypothesis is that this à was made by doubling the graph p, the sound b being phonetically sufûciently close to 6 (w tten p) to suggest such a derivatiou.

p\J 6 à U +U >Irub Later, this new gaph â was modified to B to transcribe v (today pronounced f and spelled r,l) while pà was modified to Pi to traNcribe f (today sp€tled Âl). In the fottowing chart wç show and .ke explicit the evolution of the graphs p]1 p ô vand their derived forms since Pallava script to modem Thai.

Pâllâvâ âncient Klmer ancient Thai modem Thai

ph tJ t0 po ph pr Bl po Ph f P.l f p îJ U p>6 p U 6 U6 P ,CU p tJp o Ofi u ; b upL B ry$ ÿ'1 f ô OO v/b v b lca

One could note that it would be more accute ûo tanlitterate â as Æ but to avoid useless complications the traditional traNlitteration is preserved. Finally, we believe that this study is another illusaation of the importaûce of using both epiglaphic and tinguistic eüderces, in paxticular historical phonetics, in tlle study ofüe evolution ofwriting systems. 82 FERLUS

REFERENCES

Fe us, Michel (1993). Essai de phonétique historique du khmer. Mon-Khmer Sndies XXI: 57-81. c,rawellgù, Sai,,ud ( I 981). 'Ihe Prc-angkofiü Scnpls. M. A. thesis, Silpakom University, Bangkok [In Thai]. Jacob, Judith M. (1960). The Structure ofthe Word in . Bulletit ofthe School of Oriental and Aûicaû Studies 23i 351-68.

Phirunsam, Suntarce (1981)- The Angkoriaû Khmer Scripts. M. A. thesis, Silpakom University, Bangkok [n Thai]. Songvitaya" Anaû (1981). the Charucteristics of the Sqipts aûd theb Uhogaphy in Sukhothai Incriptions. M. A. thesis, Silpakom Univeffity, Bangkok. Un Thail. Vimonkasem, Kannika (1983)- The Script "faLkhaû" iû Ltsciptioùs ofNorthem . M- A. tbesis, Silpakom UdveBity, Bangkok. Un Thail.

Mise àjour, oct. 2013