Where Did the Rejangs Come From?
Rejang is an isolated Austronesian language with roughly 200,000 speakers in five major dialect areas occupying the Barisan highlands in the Indonesian provinces of Bengkulu and South Sumatra. Rejang country is surrounded on three sides by various Malay dialects, and its western edge meets the Indonesian (Indian) Ocean. It is not to be confused with another language group with the same name occupying a territory near the mouth of the Rejang River in Sarawak, Malaysia. At one and the same time, however, there may well be a connection. McGinn (1999) raised this question and gave typological evidence. Section 3 of this paper continues along the lines of that quest. At the same time, Section 2 raises a new set of possibilities in search of a subgrouping hypothesis for the Sumatran Rejangs.
Rejang is a fairly well-studied language, at least from the point of view of its historical phonology, thanks in the main to a paper by Robert Blust (1984), which demonstrated that in this language there have been more changes in the vowels than in any other known Austronesian language. McGinn (1997, 1999) added further information, including the claim that pre-Rejang once had same stress pattern as contemporary Malay. In this pattern, the stress falls on the ultimate when the penult is schwa; otherwise on the penult. (In contemporary Rejang stress falls on the ultimate.) The advantage of reconstructing the Malay-type pattern for pre-Rejang is that a number of vocalic changes that had been described as irregular were shown to be regular.
The basis of the present paper is a set of 100+ ordered changes in phonology and morphology linking Rejang and Proto-Austronesian. (See Appendix 1 for a sample of the first 50+ changes.) The central rules in Rejang historical phonology affected unstressed reflexes of PMP *a. The rules are shown as (31a-c) in Appendix 1, and summarized below.
(31a-c) *a > « /-V:C__(C[-velar])
This is a composite of three changes affecting unstressed *a word-finally in polysyllables.
31a. The first change affected pre-Rejang diphthongs
*aw and *ay (from PMP *aw, *ay, *ey), raising the
nuclei to *«w and *«y (reflected as «w, «y in Lebong and Pasisir dialects; in Musi ie, uo; in Kebanagung «e, «o; and in Rawas uy, iw), e.g. *pisaw > Lebong pis«w `knife’ and *matay > Lebong mat«y `die’.
31b. The second change affected etyma with the shape
CV:CaC except when the final -C was a velar; thus
*ta:Nan > ta:N«n `hand’, *surat > su?«t `letter’ and
*zalan > dal«n `road’ but *a:nak > a:na? `child’ and
*da:qan > da:n `branch’.
31c. The third change affected etyma with the shape CV:Ca; thus *bu:Na > *bu:N« (> *buNi > buN«y)
`flower’ but *t«ka: > *t«ka: (> t«ko) `come’.
Rule 31 is central because it is dependent both for its regularity and its phonetic plausibility upon its interaction with two global patterns reconstructed for early pre-Rejang: the Malay-type stress pattern, and a set of syllable reductions (disyllabic and monosyllabic etyma reflecting PMP trisyllables and disyllables). See McGinn (1999) for discussion.
Of the three changes expressed in (31), change (31b) is the most promising for subgrouping purposes because it is both regular and typologically unusual. (Standard Malay shows virtually a mirror image change, reflecting *-eC as -aC in word-ending syllables: *taneq > Malay tanah; *qutek > Malay otak.) In this paper I shall explore the possibility that the three factors just mentioned (change 31b, Malay-type stress pattern, set of syllable-reductions) might be useful in determining the position of Rejang in relation to other Austronesian languages. The quest is for a subgrouping hypothesis, hence an `external’ interpretation of the historical phonology, which would add to the contribution Rejang has already made to the study of sound change.
The basis of my subgrouping quest is the set of the first 50 changes shown in Appendix 1. These are assumed to represent the earliest changes in Rejang based on their relative ordering. The first six changes (not spelled out) merely identify Rejang as a member of the PMP subgroup. The next changes after PMP are the important ones for my purpose, especially (31b), as already mentioned. After (31a-c), the stress pattern shifted to the contemporary pattern (word-final) and the language began to diverge into the contemporary five major dialects.
2. Bedayuh (Land Dayak)
I assume that Rejang has no close relatives in Sumatra. In McGinn (1999) and in my SEALS X paper, I presented some data linking Rejang typologically with a set of Bornean languages, especially the Melanau dialects in Sarawak, one of which is called `Rejang’ (see below). During the discussion that followed my paper,
pointed out that the “Bedayuh” languages spoken in the Serian District of Sarawak regularly show -«C reflecting PMP last-syllable *-aC except before velars, therefore satisfying the general description of Rejang change (31b). I have since read a brief account of Bedayuh phonology in Court (1967a), which does indeed suggest a resemblance to the Rejang rule. More on this below. In addition, there is a small bit of onomastic evidence pointing in the same direction. Three Sarawak place names in the Bedayuh district correspond suggestively with the names of villages in the Lebong dialect area of Rejang (Sumatra): Sarawak’s “Serian” (pronunciation unknown) corresponds with Lebong’s “Sien” [si«n]; Batu Lintang (pronunciation unknown) corresponds with Lebong’s Butaw Litang [butaw litaN]; and Tapuh (pronunciation unknown) corresponds with Lebong’s Topos [topos] (spelled Tapus on standard maps). On the strength of these threads of evidence, I now intend to explore the Bedayuh languages around Serian (as well as the Melanau group around the Rejang River) during my field trip to Sarawak in December 2000. Unfortunately, what linguistic information I have found through library research about the Bedayuh languages in general (see bibliography) does not suggest any close connection with Sumatran Rejang. The exception to this statement is the data provided by Court (1967a) mentioned above; moreover, Robert Blust (personal communication) has send me a wordlist for the Tapuh dialect provided to him by Donald Topping which corroborates Court’s comment with respect to the existence of at least one Bedayuh dialect showing second-syllable schwas reflecting *a except before velars, e.g. beside Tapuh t«N«tn `hand’, su?«t `letter’, ber«s `husked rice’ and j«r«tn `road’ (cf. PMP *taNan `hand’, *surat `letter’, *beRas `husked rice’ and *zalan `road’) one finds «maN `father’, l«mak, `fat’, turakN `bone’, anak `child’, and deya? `blood’ (cf. PMP *tuqelaN `bone’, *anak `child’, and *dalaq (*-q = velar) `blood’). Finally, an interesting phonetic similarity is the feature of pre-stopped final nasals reported for both Rejang and Tapuh-Bedayuh; thus, beside Tapuh bul«tn `moon’ one finds the `pausal forms’ of Rejang (e.g. bul«n [bul«dn] `moon’ reported in the literature (Voorhoeve 1955, McGinn 1982).
These data sets provide prima facie justification for investigating a possible genetic link between the Sumatran Rejangs and the speakers of Bedayuh-Tapuh. Thus a major goal of my upcoming field trip is to collect data from Tapuh and other dialects of the Serian district, and to reconstruct pre- (or Proto-) Bedayuh for purposes of comparison with pre-Rejang.
3. Mukah Melanau.
Pre-Rejang and contemporary Malay share the same stress pattern (McGinn 1997); and both share with Melanau a number of other typological similarities which might turn out to be actually shared innovations. Since the immediate ancestor of Malay is known to derive from western Borneo (Adelaar 1992), it is thus reasonable to look in that direction in search of a homeland for Rejang. I have speculated in print (McGinn 1999) that all three languages might be part of a subgroup. However, as I also pointed out in that paper, there are alternative explanations for the similarities one finds. In fact, most of them can be accounted for either as shared retentions from PMP (changes 1-11 in Appendix 1) or as independent inventions involving typologically similar languages (changes 12-30 in Appendix 1). Nonetheless, the changes (besides the reconstructed Malay-type stress) that seem most promising as a possible link between Rejang and the Melanau group is the set that I have labeled `Blust’s Law’ in McGinn (1999). Although these changes have occurred independently in many languages in Borneo, Sumatra and Java (Adelaar, personal communication), it is remarkable that they occurred in the same order in Malay and Mukah Melanau (Blust 1997). If it is accepted (as I have claimed) that they also applied in the same order in Rejang, that would be a beginning toward establishing a subgroup including pre-Rejang, pre-Melanau and Proto-Malay in a subgroup. To establish such a claim it will be necessary to reconstruct pre- Melanau alongside pre-Rejang and Adelaar’s (1992) Proto-Malay. This becomes the second goal of field work in Sarawak.
4. Concluding Remarks
Highly innovative phonologies often correlate positively with long-distance migrations (Blust 1991; Ross 1991), whereas dialect uniformity within an area suggests a relatively short period of occupation (Sapir 1916, 1949). Rejang’s high number of phonological innovations (especially in the vowels and diphthongs) is not matched by a corresponding high degree of dialect diversity. On a modified Swadesh 200 wordlist of basic vocabulary, the Lebong, Pasisir, and Musi dialects shared around 95% cognates among themselves, compared with around 87% shared cognates with the Kebanagung dialect, and around 79% with the Rawas dialect. These measures suggest that a migration resulting in loss of contact with the homeland may have occurred in fairly recent times, perhaps as little as 1000-1,500 years ago. If so, it is reasonable to hope that an extra-Sumatran point of origin (homeland) will eventually be found using linguistic reconstruction techniques, and that the results will add to our growing body of knowledge about the movement of Austronesian settlers through the large islands of Southeast Asia (Bellwood 1995).
Appendix 1: EARLIEST 50+ CHANGES
CHANGE pre-REJANG MUKAH MALAY
1-6 PAn > PMP (6 changes) YES YES YES
7. PENULT STRESS EXCEPT *« YES MAYBE YES
(All affixes unstressed) (reconstructed)
Morphological Changes in Western Indonesia
8. Lost of INSTR prefix *hi- YES YES YES
9. Loss of GOAL suffix* -en YES YES YES
10. Reanalysis of *-in- as Passive YES YES YES
11. Retention of infixes *-um- , *-in- YES YES NO
`BLUST’S LAW’ SET (BL)
12. Prepenultimate *a > « YES YES YES
13. Schwa Syncope (SS) YES YES YES
14. Consonant Reduction (CR) YES YES YES
15. Prepenultimate *i > « YES YES YES
16. Prepenultimate *u > « YES YES YES
`BLUST’S LAW’ AND MORPHOLOGY
17. *maN- > m«N- ; *maR- b«(r)- YES YES YES
18. Ablaut: -u-, -i- become infixes NO YES NO
19. *-um- > -«m- and *-in- > -«n- YES YES NO
20. *b-, *p- > zero in Trans. verbs YES YES NO
BACK TO PHONOLOGY
21. Final *-q > ? YES YES NO (-h )
22. Initial *q- > zero YES YES NO ( h-)
23. *-q- > zero between like vowels YES NO ( ?) NO ( -h-)
24a. *e- > zero in trisyllables YES YES NO
24b. *e- > zero ATB (*epat > pat) YES YES NO
25. *-eR- > zero in trisyllables YES YES(dian) NO
26. *-«V- > V (*buh«k > buk) YES YES YES
27. Derived diphthong: *-aqi > *ay YES YES n.d.
28. *Z- > d- (*dal«n) YES NOBlust NO
29. *z- > j- (*zaRi > ji?«y) YES n.d. YES
CHANGES pre-REJANG MUKAH MALAY
30. *-Z- , -z- merge as -j- YES YES YES
31a. *a Raising: *-ay, *aw > «y, «w YES NO NO
b. *a>«/V:C__C[-velarl]# YES NO NO
c. *a>«/V:C__# YES NO NO
32. *u-Lowering (*niuR > *nioR) YES NO(ñuh) NO
33. *-R- > -l- except/C[-cor]V VC YES NO NO
34. *R, *r merge as > h ATB YES NO NO
35. *w- > b- (*bal«t) YES NO NO
36. V-Coal. *ai > e (pet, pat) YES NO NO
37. *-j > -g (*qul«j > *qul«g) YES NO NO NO(Rawas)
38. *-j- > -g- (*p«gu: > p«gew) YES NO NO
39. *-j- > zero (*pa:ay > pay) YES NO NO
40. STRESS SHIFT TO ULTIMATE YES YES NO
NO (Blust 1988)
41-49. Vowel changes conditioned by new word-final stress pattern:
Gloss PMP late pre-Rej Rejang (Lebong)
broom *sapu *supu: supa:w
rope *tali *tili: tila:y
brain *qut«k *uto:k oto:?
tooth *ip«n *ipe:n epe:n
chicken *manuk *monu:k mono:?
sky *laNit *leNi:t leNe:t
50. (affected forms that systematically escaped change 31c):
*-a: > -o: /-«C #
PMP Rejang Gloss
*t«ka t«ko come
*lema lemo five
*depa depo fathom
51. (secondary changes after 31c applied):
*«: > i: > «y / __#
eye *mata > *mat« > *mati: > matay (Lebong)
flower *buNa > *buN« > buNi (Kebanagung)
52. *«: > o: / «C__C[+dorsal]
hear *d«N«:R > *d«No:R > t«Noa
hold *p«g«N > *p«goN > goN
DIAGNOSTIC ITEMS IN SEARCH OF A SUBGROUPING HYPOTHESIS
A. Diagnostic items linking Rejang and Malay (taken from Blust 1981, 1982, 1992).
1. Metathesized reflex of PMP *quDip `alive’. Rejang: idup `alive’
2. Innovated numerals `SEVEN’, `EIGHT’ and `NINE’. Rejang: tojoa?, d«lap«n, s«mbil«n. Cf. also Malayic (Adelaar 1992) and Proto-Chamic (Thurgood 1999:37)
B. Diagnostic items representing changes affecting Rejang but not Malay.
Group I – Metathesis affecting two words.
PMP Pre-Rejang Rejang Malay
wood *kahiw > *kihaw ki«w kayu
dove *punay > *panuy ponoy punai
(cf. fire *qapuy > *apuy opoy api )
Group II – Regular change, e.g. *-aC > -«C in unstressed syllables except before velars (*quDaN > udaN). Note reconstructed penultimate stress (McGinn 1997).
Gloss PMP Pre-Rejang Rejang Malay
round *bulat *bu:lat>*bu:l«t bul«:t bu:lat
smoke *hasap *a:sap>*a:s«p as«:p a:sap
rain *quZan *u:jan>*u:j«n uj«:n hu:jan
hand *taNan *ta:Nan>*ta:N«n taN«:n ta:Nan
road *zalan *da:lan>*da:l«n dal«:n da:lan
Group III – Irregular Rejang reflexes of PMP segments.
Gloss PMP Pre-Rejang Rejang irregular Malay
hear *dengeR *t«NoR t«Noa *d- > t- d«ngar
egg *qateluR *t«noR t«noa *-l- > -n- t«lur
break *pataq *patiq patia? *a > i patah
bone *tuqelaN *t«laN t«lan *u > « tulang
five *lima *l«ma l«mo *i > « lima
claw *silun *s«lon s«lon *i > « (cakar)
Group IV – Analogical change: Loss of *b- and *p- in transitive verbs
Gloss PMP Rejang-Lebong Malay
kill *bunuq unua?, m-unua?, n-unua? bunuh
hold *pegeng gong, m«-gong, n«-gong p«gaN
give *beRey l«y, m«-l«y, n«-l«y b«ri
choose *piliq elea?, m-elea?, n-elea? pilih
borrow *pinzem iny«m, m-iny«m, n-iny«m pinjam
C. LEXICAL INNOVATIONS
Gloss Rejang Pre-Rej. Malay
heavy b«n«k *b«n«g b«rat
lightning s«mitoa (Keb.) *s«mituR kilat
to stab tuj«ah (Keb.) *tujaq tikam
ear ti?u? *tiruk t«liNa
wild pig jaoa? *jauq babi hutan
ashamed sele? *selek malu
sit t«mot *t«mot duduk
door baN *baN pintu
cf. Proto-Chamic *ºaN `door;hole’ poss. borr. fr. MK (Thurgood 1999:312)
D. FUNCTION WORDS
Gloss Rejang Musi Malay
not coa tidak
not a iso bukan
not yet ati b«lum
Don’t! daN jaNan
jib«a? (Lebong dialect)
to may ke
at na? di
there di sana
here pio sini
Adelaar, K. Alexander. 1992. Proto-Malayic: A
reconstruction of its phonology and part of its
morphology and lexicon. Pacific Linguistics Series C
No. 119. Canberra: Research School of Pacific
Studies, Australian National University.
___________. 1995. Borneo as a crossroads for
comparative Austronesian linguistics. In Bellwood,
Fox and Tryon, pp. 87-95.
Bellwood, Peter, James J. Fox, and Darrell Tryon, eds.
1995. The Austronesians: Historical and Comparative
Perspectives. Canberra: Department of Anthropology,
The Research School of Pacific and Asian Studies.
Bibi Aminah Abdul Ghani. 1992. The Bintulu dialect: an
introduction. Sarawak Museum Journal 43:321-41.
Blust, Robert A. 1981. The reconstruction of Proto-
Malayo-Javanic: an appreciation. Bijdragen tot de
Taal-, Land- en Volkenkunde 137(4):456-469.
__________. 1982. An overlooked feature of Malay
historical phonology. Bulletin of the School of
Oriental and African Studies, Vol. XLV.2:284-299.
__________. 1984. On the history of the Rejang vowels
and diphthongs. BTLV 140:422-450.
__________. 1988. Papers in Borneo and Western
Austronesian Linguistics No. 3. Sketches of the
morphology and phonology of Bornean Languages 2:
Mukah (Melanau). Pacific Linguistics A-78:151-216.
__________. 1991a (ed.). Currents in Pacific
Linguistics: Papers in honor of George W. Grace.
Pacific Linguistics Series C, No. 17. Canberra.
__________. 1991b. Sound change and migration
distance. In Blust (ed.), pp. 27-42.
__________. 1992. The Austronesian settlement of
mainland Southeast Asia. In Adams, Karen L. and
Thomas John Hudak, eds., Papers from the Second
Annual Meeting of the Southeast Asian Linguistics
Society. Tempe, AZ: Arizona State University Program
for Southeast Asian Studies.
__________. 1997. Ablaut in Northwest Borneo.
Clayre, Iain. 1973a. A preliminary note on focus and
emphasis in Melanau – a language of Coastal
Sarawak. Lingua 31:237-69.
__________. 1973b. Notes on spatial deixis in
Melanau. Anthropological Linguistics 15(2):71-86.
Chou, Shu Hsiu. 1999. Melanau and Indonesian: a
comparative grammatical study. Unpublished thesis,
Tamkang University, Taiwan.
Collins, James T. (ed.). 1990. Language and oral
traditions in Borneo. Williamsburg, VA: The Borneo
Research Council Inc., Proceedings Series Vol. Two
Court, Christopher. 1967a. Some Areal features of
Mentu Land Dayak. Oceanic Linguistics VI: 46-50.
__________. 1967b. A Distinctive Feature Analysis of
the Phonemes of Mentu Land Dayak. Phonetica
__________. 1967c. Pointing and Asking: A Note
on Deixis in Mentu Land Dayak. Bijdragen tot de
Taal-, Land- en Volkenkunde 123.
Jussem Dundon, Stephen. 1989. Bidayuh Language
and Dialects. Sarawak Museum Journal. 61
(New Series.): 407-413.
Kroeger, Paul R. 1998. Language classification in
Sarawak: a status report. Sarawak Museum Journal
LIII.78 (New Series):137-174.
Martin, Peter W., ed. 1992. Shifting patterns of language
use in Borneo. Willimsburg, VA: Borneo Research
Council Proceedings Series, Vol. 3.
McGinn, Richard. 1982. Outline of Rejang syntax.
__________. 1997. Some irregular reflexes of
Proto-Malayo-Polynesian vowels in the Rejang
language of Sumatra. Diachronica XIV.66-107.
__________. 1999. The position of the Rejang
language of Sumatra in relation to Malay and the
`ablaut’ languages of northwest Borneo. In
Zeitoun, Elizabeth and Paul Jen-kuei Li, eds.,
Selected Papers from the Eighth International
Conference on Austronesian Linguistics. Taipei,
Taiwan: Academia Sinica, pp. 205-226.
Nais, Datuk William. 1988. Bidayuh – English Dictionary.
Kuching: Persatuan Kesusasteraan Sarawak.
Ray, S.H. 1913. The Languages of Borneo.
Sarawak Museum Journal. 1(4):1-196.
Ross, Malcolm D. 1991. How conservative are
sedentary languages? Evidence from western
Melanesia. In Blust (ed.), pp. 433-451.
Sapir, Edward. 1916, 1949. Time perspective in
aboriginal American culture. In David G. Mendelbaum
(ed.), Selected writings of Edward Sapir. Berkeley
and Los Angeles, pp. 389-462.
Thurgood, Graham. 1999. From ancient Cham to
modern dialects: two thousand years of language
contact and change. Oceanic Linguistics Special
Publications 28. Honolulu: University of Hawai’i
Topping, Donald M. 1990. A dialect survey of the Land
Dayaks of Sarawak. In Collins, ed.
Voorhoeve, P. 1955. Critical survey of studies on the
languages of Sumatra. ‘S-Gravenhage: Martinus