Puluqish *paka-/*maka- abilitative

(Jump to Puluqish master post)

Prefixed *paka-, originally a causative with *pa- of stative verbs prefixed with *ka- (Zeitoun 2000), retains its causative function in a  broad range of Formosan languages: Saisiyat (pak-), Pazeh, Mayrinax Atayal and Mantauran Rukai (examples in Zeitoun’s paper); Siraya (Adelaar 2012:117), Kavalan (paq -̴ paqa-, Li & Tsuchida 2006:17), Central and Northern Amis (Bril, in press), and Puyuma (as in pa-ka-ulaŋ ‘make crazy’, Cauquelin 2015). paka-causatives are also found in the Philippines (Liao 2011).

In the Puluqish languages of Taiwan (Amis, Puyuma, Paiwan), in addition to paka-causatives, one also finds abilitative verbs with paka-: in the words of Liao Hsiu-chuan (2011:858), paka- expresses an actor’s ability to perform an action: ‘can V, able to V’. The present discussion is largely based on Liao’s  findings on the distribution and functions of paka- and maka- abilitatives, and on her examples for Taiwan and the Philippines.

Here is an example of paka- abilitative in Paiwan, cited by Liao from  Wolff (1995:567):

su=paka-qati-n ‘you (SG) can do it.’ (qati ‘succeed, achieve’).

In addition to paka-, Paiwan and Amis have an Actor-Focus prefix maka- which is also abilitative. A Paiwan example:

maka-qati ‘can do something’ (Wolff, ibid.)

Bril finds maka- abilitatives in Northern Amis too (p.c.):

maka-tengil tiya suwal
(he) could hear these words

Wolff (ibid.) derives Paiwan maka- from an earlier *<um>paka-. This seems likely. I assume the same explanation works for Amis.

Philippine languages use maka- not paka- in abilitatives (Liao 2011:859). maka-abilitatives are also found elsewhere in MP, as in Malagasy, under the form maha- (maha-fotsy ‘qui peut blanchir’).

paka- abilitatives maka-abilitatives
Northern Amis + +
Puyuma +
Paiwan + +
MP-Philippines +
MP-Malagasy +
Kra-Dai ? ?

In view of the fact that Siraya has been claimed to be phylogenetically close to PMP, it is interesting that Siraya uses paka-Vstat to form causatives of ma-statives:

ma-kuptix ‘clean’, paka-kuptix ‘to purify’

In addition paka-Vstat preceded by the auxiliary verb ’lpux ‘can’ (Adelaar 2011:133) results in a abilitative construction ‘lpux pa-ka-V ‘can V, able to V’:

paka-‘lpux=kaw pa-ka-kuptix ĭau-an=da

AS-can=2s.NOM. CAUS-v1-clean 1s-OBLl=ADV

‘you are able to purify me’

In this example, paka– preceding the auxiliary is not a prefix but a copy of the main verb’s initial syllables, which here coincide with the main verb’s prefix. See Adelaar (2011:135) on ‘anticipating sentences’ (AS).

This construction provides a missing link between PAN *pa-ka- ‘causative of nonfinite stative verbs’ and abilitative. It is likely that *paka- abilitatives originate in a similar construction with paka-Vstat preceded by an auxiliary verb meaning ‘can -, able to’: elision of the auxiliary verb resulted in paka- acquiring its abilitative meaning.

These facts are susceptible of phylogenetic interpretation.  *paka- and/or *maka- abilitatives are only found in the southern languages Paiwan, Amis, Puyuma, and in PMP:  the innovations creating these two forms should therefore be assigned to proto-Puluqish.

Since proto-Kra-Dai is a Puluqish language, one would like to know whether these two innovations were present there too. Unfortunately the Kra-Dai languages do not preserve the left edge of words well enough for us to know the answer.

References

Adelaar, Alexander. 2011. Siraya, Retrieving the Phonology, Grammar and Lexicon of a Dormant Formosan Language. Berlin: De Gruyter Mouton.

Bril, I. in press. Northern Amis. Handbook on Formosan languages, ed. by P. Jen-kuei Li, E. Zeitoun, R. De Busser.  Leiden: Brill

Cauquelin, J. 2015. Nanwang Puyuma-English dictionary. Institute of Linguistics, Academia Sinica, Taipei, Taiwan.

Liao Hsiu-chuan. 2011.  Some morphosyntactic differences between Formosan and Philippine languages. Languages & Linguistics 12, 4:845-876.

Wolff, John U. 1995. The position of the Austronesian languages of Taiwan within the Austronesian group. Austronesian Studies Relating to Taiwan, ed. by Paul Jen-kuei Li, Cheng-hwa Tsang, Ying-kuei Huang, Dah-an Ho & Chiu-yu Tseng, 521-583.
Taipei: Institute of History and Philology, Academia Sinica.

Zeitoun, E. 2000. Concerning ka-, an overlooked marker of verbal derivation in Formosan languages. Oceanic Linguistics, 39.2: 391-414.

The merger of PAN *S1 and *S2

The merger of PAN *S1 and *S2 is an innovation of eastern Walu-Siwaish

Working in the tradition of Dyen (1971), Tsuchida (1976:250, 308) distinguished two major PAN sibilant phonemes, *S1 and *S2. Examples of the former are many; Tsuchida (1976:250 and index) exemplified the latter with seven lexical items:

  • *S2uni ‘chirp’,

  • *S2uReɬa ‘snow’,

  • *kaS2iw ‘tree, wood’,

  • *CuS2uR ‘to thread’,

  • *q4uS2uŋ ‘mushroom’,

  • *Cuma[S2H1] ‘body louse’,

  • *guS2am ‘skin disease’.

The ending in *Cuma[S2H1] is ambiguous with *H1 if the evidence is limited to Saisiyat and Atayalic (as in Tsuchida 1976:202), but when Kavalan tumes and Amis tomes are included, *H1 is excluded. More examples can be added. By my count at least 14 reconstructable Austronesian words include *S2.  Click  PAN *S2_table of 14 exx_modified April 2024.

14 words is not a negligible number. *S2 occurs in all positions: initial, medial, final.

The distinction between *S1 and *S2 is consistently observed in Saisiyat (S ≠ h), Pazeh (s ≠ h), Atayal (s ≠ h), Seediq (s ≠ h), Thao (ʃ ≠ 0), Siraya (x ≠ 0), Bunun (s ≠ 0), Tsou (s ≠ 0), Kanakanabu (s ≠ 0) and Rukai (Maga dialect: s ≠ 0; Mantauran dialect ʔ ≠ 0). It is lost in Saaroa (0 = 0), Kavalan (s = s), Amis (s = s), Paiwan (s = s), Puyuma (0 = 0), PMP (0 = 0), Proto-Kra (word-medially: s = s), Proto-Hlai (word-finally: -t = -t), Proto-Tai (word-medially: s = s). The evidence for the merger in Kra-Dai will be detailed below.

Treatment of *S1 and *S2 in extinct Taokas, Favorlang/Babuza, Hoanya, Papora, Basay and Trobiawan cannot be fully determined due to the fragmentary character of the evidence.

Geographically the languages that merge *S1 and *S2 are spoken along the eastern and southern coasts of Taiwan. Those that keep them distinct are found on the west coast and in the central regions. Saaroa, a central Taiwan language that merges *S1 and *S2 is special: its distance from the eastern coast, and its affiliation with Tsouic argue strongly that the same merger took place independently on the east coast and in Saaroa.

I now present the evidence for the merger of *S1 and *S2 in the Kra-Dai branches. The evidence is as limited as the examples of *S2 are. To my knowledge, only *CumeS2 ‘louse’, *kaS2iw ‘tree, wood’, and *qaluS2ip-an ‘centipede’ are reflected in Kra-Dai.

In the Kra branch, Proto-Kra *sui A ‘firewood’ (Ostapirat 2000) is the most probable reflex of *kaS2iw ‘tree, wood’—several Formosan languages reflect *kaS2uy instead of *kaS2iw. This shows that the PK reflex of *S2 in word-medial position is *s-. *s- is also the reflex of *S1 in the same position: PAN *duS1a ‘two’ = PK *sa A.

In the Hlai branch (Norquest 2015) we have *CumeS2 ‘louse’ reflected by Proto-Hlai *hməːt ‘flea’ and ‘to winnow’ *tapeS1 by *wet, showing the merged reflex of *S1 and *S2 is -t in word-final position.

In the Tai branch, Proto-Tai *q.sip D (Pittayaporn 2009) ‘centipede’ corresponds to PAN *qaluS2ip-an ‘centipede’ (I take final *-an to be suffixal, compare *waNiS-an ‘boar’, *RiNaS-an ‘pheasant). PAN medial *S1 also goes to s- in Proto-Tai: PAN *qaS1aN ‘grain before husking’, PT *sal A ‘husked rice’ (Pittayaporn 2009).

References

Dyen, I. (1971) The Austronesian languages and Proto-Austronesian. In T. Sebeok (ed.) Current Trends in Linguistics, Vol. 8:5-54. The Hague and Paris: Mouton.

Norquest, Peter. 2015. A Phonological Reconstruction of Proto-Hlai. Brill.

Ostapirat, Weera. 2000. Proto-Kra. Linguistics of the Tibeto-Burman Area 23.1:1-251.

Pittayaporn, Pittayawat (2009) The phonology of proto-Tai. Unpublished Cornell U. dissertation.

Tsuchida, S. (1976) Reconstruction of Proto-Tsouic phonology. Tokyo: Study of languages and cultures of Asia and Africa monograph series N° 5.

Yet another word for ‘ten’ in Formosan languages

up one level to Eastern Walu-Siwaish

A hitherto undescribed etymon for ‘ten’ occurs in languages of the north and east coasts of Taiwan: Sakizaya, Puyuma, and north Formosan (Kavalan, Basay, Ketagalan).

In Sakizaya, Amis’s close relative, bataʔan serves for ‘ten’ in and multiples of ten (data collected by McNaught, 2017; see the Sakizaya page in Eugene Chan’s ‘numeral systems of the world’s languages’ (here). There no trace of a reflex of *puluq, unlike in Amis and in Puyuma.

Cauquelin’s Puyuma dictionary (Cauquelin 2015) only has puɭuʔ and məəp for ‘ten’ but ‘twenty’ is maka-bəʈaʔan. The prefix maka- serves in ‘ten’ and multiples of 10, eg maka-telun ‘30’, maka-pətəl ‘30’ etc. It is possible that maka-bəʈaʔan was simplified from maka-ɖua-bəʈaʔan: removing ɖua would cause no ambiguity.

Kavalan has Rabtin ‘10’, zusabtin ‘20’ where –btin is the numeral ‘10’ (Li & Tsuchida 2006). One source of Kavalan /i/ is *qa. Vowel syncope is common in Kavalan although the conditions governing this process have not been elucidated. So -btin < *b()t()qan. Kavalan merges *t and *C into *t, so *b()t()qan may originate in *baCaqan. Basay is similar to Kavalan: labatan ‘10’, lusa batan ‘20’, as is Ketagalan ɭabat-an ‘10’, ɭusa batan ‘20’ (Ogawa, cited in Ferrell 1969). All these forms are regular outcomes of *baCaq-an. Puyuma provides the evidence for reconstructing -C- as against -t-.The formative Kavalan Ra-, Basay la-, Ketagalan ɭa– is a distinct morpheme.

A tentative etymology can be offered. Tsuchida (1988) glossed the Bunun word bataqan (< baCaq-an, *bataq-an) in Qato and Idhokan dialects as ‘racks (L-shaped – to carry woods)’. Nihira’s Bunun vocabulary gives ‘a carrying board on the back’. The name of a carrying device for multiple objects is a potential source of ‘ten’. Bunun is a Walu-Siwaish language, like the languages where *baCaq-an occurs in ‘10’ or ‘20’: we may assign *baCaq-an ‘carrying device’ to Proto-Walu-Siwaish and the derivation out of it of a new word for ‘ten’ to Proto-Eastern-Walu-Siwaish.

References

Cauquelin, J. (2015). Nanwang Puyuma-English dictionary. Institute of Linguistics, Academia Sinica, Taipei, Taiwan.

Ferrell, R. (1969) Taiwan Aboriginal groups: problems in cultural and linguistic classification. Monograph No. 17, Institute of Ethnology, Academia Sinica. Nankang: Academia Sinica.

Li, P. J-K, and S. Tsuchida (2006) Kavalan dictionary. Language and Linguistics monograph series A19. Nankang: Institute of Linguistics (preparatory office), Academia Sinica.

Nihira, Y. (1983) A Bunun Vocabulary (2nd edition). Privately published.

Tsuchida, S. 1988. Comparative word lists of Bunun dialects. Report of the research carried out in 1983. Unpublished manuscript.

The etymology of *puluq ’10’, at last

(Jump to Puluqish master post)

In Sagart (2008), a modification of the phylogeny in Sagart (2004), I set up a Puluqish group of Austronesian languages including all of PMP, all of Kra-Dai, plus three southern Formosan languages: Amis, Puyuma and Paiwan. This group was defined by the innovation *puluq for ‘ten’ (another innovation has since been identified, see the master post for Puluqish).  My subgrouping is in contrast to the standard view (e.g., in Blust’s phylogeny) that *puluq existed in PAN. I was not able, at the time, to explain how *puluq arose. My inability to do so at the time was interpreted as an indication that *puluq was etymologically opaque, as befits a PAN form, just like the numerals 1-4. If,  on the contrary,  *puluq was an innovative form, as I claimed, it must originate in another word or expression shifting to ’10’ just before proto-Puluqish, and traces of that original expression should be found, hopefully in modern Puluqish languages, or not too far from them on my tree.

The etymology of *puluq ‘ten’ can now be given. Amis has a root /poloʔ/, orthographic polo’ < *puluq, which occurs in all Amis dictionaries. Below I discuss three forms in Namoh’s large (Central) Amis-Chinese dictionary (Namoh 2013), where the most detailed information can be found:

  • si-polo’ 因分離而獨居,或分家 ‘living alone or separated due to separation’
  • ma-sipolo’. 分離,分闊,分居,單身,未婚 ‘separate, split, separation, single, unmarried’. Example sentence: Milaliw ko fafahi ni Kuraw, saka masipolo’ ciira i matini. 古饒的妻子昨天離家出走,所以他現在是單身 ‘Kuraw’s wife left home yesterday, so he is now single’
  • ma-si-polo’-ay (deverbal noun out of ma-sipolo’). 寡婦,搞婦,穌夫。單身 ‘widow, widower; single person’.

The Amis affixes ma- (stative) and -ay (nominalizing) are well-known. masipolo’ is a stative verb based on sipolo’; masipolo’ay is a deverbal noun based on masipolo’. It is not entirely clear what the function of si- in sipolo’ is. There is a prefix si- in Amis with existential or possessive meaning ‘have, be’ (Bril, p.c., circa 2020) .  In the English index at the end of Namoh’s dictionary, masipolo’  is given the gloss  ‘separated from and left alone’.  One could gloss si-polo’ as ‘for a state of separation to exist’; ma-si-polo’ as ‘be in a state of separation’, and ma-si-polo’-ay as ‘a N in a state of separation’. A gloss like ‘to separate from, and leave alone’ can be supposed for the stem polo’.

It is easy to see how polo’ could be put to use in counting numbers ten and above: the meaning ‘separate from, and leave alone’ well describes the mental operation of isolating n sets of ten objects and putting them aside before expressing the remainder.  For instance, ’35’ can be expressed as ‘three sets of ten, put them aside, and five’.

The process by which *N-puluq became the main form for ’10’ in the Puluqish languages can be reconstructed thus. The word for ‘10’ in Proto-Eastern-Walu-Siwaish (the immediate ancestor pf Proto-Puluqish) was *baCaqan (here). ‘35’ in that language might have been expressed simply as telubaCaqan-lima ‘3 times ten, five’, or more explicitly as telubaCaqanpuluq-lima ‘3 times 10, leave those alone, five’. *baCaqan being redundant, *N-baCaqanpuluq-N was simplified to N-puluq-N in Proto-Puluqish, with *N-puluq acquiring the meaning ‘N times 10’. But *baCaqan was predictably retained in northerrn Puluqish languages where there was no remainder to ‘leave alone’, i.e. in ’10’ and ’20’:

  • Sakizaya t͡sət͡saj a bataʔan ‘10’ (one-baCaqan), tusa bataʔan ‘20’ (two-baCaqan),
  • Puyuma maka-bʈaʔan ‘20’

We have a small but independent piece of evidence confirming  that *puluq and its derivatives were part of the vocabulary of counting: Pokpok Amis nom-a-sipolo’ ‘7’ (Li and Toyoshima 2006, #83l.1). This is a subtractive numeral: ‘three left aside’, where nom reflects *Nem ‘3’ (details here).  Here sipolo‘ does not directly refer to ‘ten’ but to the act of ‘leaving (three units, out of ten) aside’.

A consequence of the above is that in proto-Puluqish, ‘ten’ must have been not *puluq, but *sa-puluq with *sa-, the short form of the PAN numeral *isa/*esa ‘one’, preceding *puluq: ‘one (set of ten), put aside’.  Paiwan (Sagaran dialect, recorded by Hsiou-Chuan Chang) tapuluʔ  ‘ten’ directly reflects proto-Puluqish *sa-puluq. *sa-puluq is also the form that can be reconstructed for PMP (Blust’s Austronesian comparative dictionary).

In Kra-Dai, Ostapirat (2000) reconstructs Proto-Kra *pwlot, and Norquest (2015) gives Proto-Hlai *fu:t. Ostapirat (2004) had Proto-Hlai *apu:c, with final *c  (an error in my opinion; see here).  The vowel *a before the string *pu:c appears to reflect the *a in proto-Puluqish *sa-. As reconstructed by Ostapirat, proto-Hlai occasionally retains the first vowel of Austronesian words, but never the consonant before that vowel: proto-Hlai *ata A < *maCa ‘eye’, aka:i C < *Caqi[] ‘excrement’, *ura:ŋ A < *qudaŋ ‘shrimp’,  utu A ‘head louse’ < *quCuH2, *ipan < *nipen or *lipen ‘tooth’ etc.

Thus proto-Puluqish *sa-puluq is an adequate source of the Paiwan, proto-Malayo-Polynesian and proto-Hlai words for ‘ten’. Lack of *sa- in Amis and Puyuma puɭuʔ and polo’ is due to a simplifying innovation: there was no need for *sa- after the meaning ‘ten’ became entrenched with *puluq and the original semantics were lost. This innovation should be added to the six shared innovations of Amis and Puyuma (a.k.a ‘northern Puluqish’) described here.

Although Puyuma shows no trace of the *sa- in *sa-puluq, indirect evidence that it once possessed the long form exists: the serial counting word for ‘nine’ in Nanwang Puyuma, siwa, can only be explained as the result of analogical alignment on the *s- at the beginning of *sa-puluq (more on this here).

I am not aware of cognates of polo’ with semantics related to ‘separated, put aside’ outside of Amis. It is perhaps significant that the lexical source of the Puluqish word for ‘ten’ comes from Amis, a Puluqish language.  If *puluq were a PAN word, that would be a coincidence.

References

Blust, R. 2001. Malayo-Polynesian: new stones in the wall. Oceanic Linguistics 40, 1: 151-155.

Li, Jen-kuei and Masayuki Toyoshima (eds). 2006. comparative vocabulary of Formosan languages and dialects, by Naoyoshi Ogawa. Asian and African lexicon series 49. Institute for Languages and cultures of Asia and Africa, Tokyo University of Foreign Studies.

Namoh, Rata. 2013. O Pidafo’an to Sowal Misanopangcah [dictionary of the Amis language]. Taipei: Nan t’ien.

Norquest, Peter. 2015. A Phonological Reconstruction of Proto-Hlai. Brill.

Ostapirat, Weera. 2000. Proto-Kra. Linguistics of the Tibeto-Burman Area 23.1:1-251.

Ostapirat, Weera. 2004.  Proto-Hlai sound system and lexicons. in: Ying-ching Lin, Fang-min Hsu, Chun-chih Lee, Jackson T.S. Sun, Hsiu-fang Yang and Dah-an Ho (eds), Studies in Sino-Tibetan languages, papers in honor of Professor Hwang-cherng Gong on his seventieth birthday, 121-175. Nankang: Institute of Linguistics, Academia Sinica.

Sagart, L. (2004) The higher phylogeny of Austronesian and the position of Tai-Kadai. Oceanic Linguistics 43,2: 411-444.

Sagart, L. (2008) The expansion of setaria farmers in East Asia: a linguistic and archaeological model. In Sanchez-Mazas A, Blench R, Ross M, Peiros I, Lin M, eds. Past human migrations in East Asia: matching archaeology, linguistics and genetics, 133-157. Routledge Studies in the Early History of Asia, London: Routledge.

Northern Puluqish (master post)

(view clickable tree) (to list of innovations, sorted by node)

(Jump up one level to Puluqish master post)

(Jump down one level to Southern Austronesian master post)

There is evidence that Amis and Puyuma subgroup together within Puluqish, forming a northern group opposite Paiwan, MP and KD. Here are innovations uniquely shared, as far as I know, by the two Northern Puluqish languages.

  1. Initial of *DuSa ‘2’ aligned on initial of *telux ‘3’ in the serial-counting series. Puyuma retains a contrast between a serial-counting series of numerals and another series for counting objects and people. Amis has lost distinctions among numeral series. Puyuma z reflects the PAN voiced initial *D [ḍ] in the series for objects and people, but has replaced it with the same initial as ‘3’ PAN *telux in the serial counting series. The single series of Amis continues the Proto-northern Puluqish series for objects and people, at least for ‘2’. Table 1:
Rikavong Puyuma Tamalakaw Puyuma Amis
2’ (objects and people) zowa zuwa tosa
‘2’ (serial) towa ʈuwa
‘3’ tiɭi ʈeɽi tolo

Table 1: Analogical alignment of initial of ‘2’ on ‘3’ in northern Puluqish. Sources for Puyuma: Rikavong: Suenari (1969:152); Tamalakaw: Tsuchida (1980:287). Thao has tusha ‘2’ but contra Sagart (2004), PAN *D [ḍ] evolves regularly to t- (Ross 2015), therefore analogy plays no role there.

2. ‘one’ *sasay: Amis cacay ‘one’, Puyuma sasaya (< *sasa-y-a) ‘one’ (counting things). These forms reflect PAN *sa, short form of PAN *isa/esa ‘one’,  with reduplication, plus intrusive -y- inserted beween final -a in the word-base and the linker -a- following it. Thus Puyuma has ɖaɖua-ya ‘two’,  iwaiwa-ya ‘nine’, but ta-taɭu-a ‘three’, pitu-pitu-a ‘seven’, waɭu-waɭu-a ‘eight’ (all counting things).

3. ten’  *puluq. By the etymology proposed here, a word *puluq  meaning ‘to put aside’ was recruited in counting numbers above 10 in Proto-Puluqish; *puluq ‘(n) [sets of 10] put aside’ could only arise preceded and followed by a numeral.  In numbers 11 to 19,  the numeral for ‘one’ probably had the shape *sa-, as still in the other Puluqish languages: Paiwan tapuluʔ < *sa-puluq and PMP *sa-puluq. Once *puluq ‘put aside’ was lost as an independent morpheme, *puluq could be reanalyzed as a morpheme meaning ‘ten’ and *sa- could be dispensed with. That reduction had taken place in the ancestor of Amis and Puyuma, but not elsewhere in Puluqish.

4. ‘ten’ *mukeCep. In both languages, inherited *puluq is attested only as the serial-counting word for ‘ten’ and multiples of ten. When counting human and non-human referents, it has given way to an innovated form *mukeCep: Amis moketep, Puyuma mukʈəp. The expression of numerals 11 to 19 is complex and will not be discussed here.

5. ‘bone’ *ukak:  Amis ‘okak ‘bone’, Puyuma ukak ‘bone’;

6. ‘cloud’ *kuCem: Amis kotem ‘black cloud’, Puyuma kuʈəm ‘(big black) cloud’ (mi-a-kuʈəm ‘big black clouds promising rain’);

7. ‘activity, skill’ *Nemak: Amis dmak ‘an activity, a happening; the things one does; to move’ (Fey); ‘activity, affair, business; to move; situation’ (Namoh); Puyuma lemak ‘know-how; skill’.

8. ‘hat’ *kabuŋ: Amis kafoŋ ‘hat; anything worn on the head’ (Fey); Puyuma kabuŋ ‘hat’, mi-kabuŋ ‘wear a hat’ (Cauquelin). This is a cultural item. Its phylogenetic value is low.

Northern Puluqish clashes with Blust’s East Formosan subgroup. For a critique of East Formosan, see  Sagart (2015).

References

Blust, R. (1999) Subgrouping, circularity and extinction: some issues in Austronesian comparative linguistics. In: Elizabeth Zeitoun and Paul Jen-kuei Li (eds.) Selected Papers from the Eighth International Conference on Austronesian linguistics, 31-94. Taipei: Institute of Linguistics (preparatory office).

Ross, Malcolm (2015) Some Proto Austronesian coronals reexamined, in Elizabeth Zeitoun, Stacy F. Teng & Joy J. Wu (ed.), New Advances in Formosan Linguistics, Asia-Pacific Linguistics, Canberra, Australia, pp. 1-38.

Sagart, L. (2004) The higher phylogeny of Austronesian and the position of Tai-Kadai. Oceanic Linguistics 43,2: 411-444.

Sagart, L. (2008) The expansion of setaria farmers in East Asia: a linguistic and archaeological model. In Sanchez-Mazas A, Blench R, Ross M, Peiros I, Lin M, eds. Past human migrations in East Asia: matching archaeology, linguistics and genetics, 133-157. Routledge Studies in the Early History of Asia, London: Routledge.

Sagart, L. (2015) ‘East Formosan’ and the PAN palatals. Paper presented at the 13th International Conference on Austronesian Linguistics, Taipei, July 18-22, 2015.

Suenari, Michiko (1969) A preliminary report on Puyuma language (Rikavong dialect). Bulletin of the Institute of Ethnology, Academia Sinica, 27:141-163.

Tsuchida, Shigeru (1980) Puyuma (Tamalakaw dialect) vocabulary with grammatical notes and texts. in Kuroshio no Minzoku, Bunka, Gengo pp.183-307. Tokyo: Institute for the study of languages and cultures of Asia and Africa, Tokyo University of foreign studies.

Michel Ferlus et l’interface austronésien/kra-dai

Michel Ferlus avait fait à la conférence austronésienne 11-ICAL d’Aussois en 2009 une présentation où il résumait une nouvelle théorie des relations entre austronésien, kra-dai (il utilise le terme de Benedict: tai-kadai) et malayo-polynésien. Elle peut être consulté ici, en versions française et anglaise. Elle répondait à un de mes articles (2004) où je proposais que le kra-dai est une branche de l’austronésien, notamment sur la base d’innovations partagées dans les système des numéraux.

Ferlus déplore aujourd’hui le peu d’attention accordé à son texte, allant jusqu’à parler de “refus du débat scientifique”. Je vais donc donner ci-dessous une liste de problèmes contenus dans sa présentation qui sont, me semble-t-il, la cause du manque d’intérêt de ses collègues. Précisons que personne n’a abandonné l’idée du “out of Taiwan” suite à la présentation de Ferlus: j’ai moi-même employé ce terme dans un article paru l’an dernier dans la revue Rice (ici).

Tout d’abord, résumons rapidement l’idée de Ferlus.

Ferlus pense que l’austronésien, originellement parlé dans le bas-yangtzé, s’est d’abord répandu vers le sud le long de la côte chinoise. De là, des groupes austronésiens auraient peuplé Taiwan les uns après les autres (ainsi, il n’y a pas de proto-formosan dans la théorie de Ferlus). Les austronésiens restant sur le continent auraient ensuite continué vers le sud jusqu’au Guangdong, où ils seraient devenus malayo-polynésiens. Là, ils auraient été recouverts par la famille kra-dai, laquelle aurait emprunté beaucoup de vocabulaire de base au malayo-polynésien, donnant l’impression d’une parenté génétique entre le kra-dai et l’austronésien. Par la suite, les malayo-polynésiens seraient passés du Guangdong aux Philippines, d’où ils auraient peuplé tout le Pacifique. Ils n’auraient pas laissé de traces sur le continent. Parallèlement à leur expansion vers le sud, ils auraient, au nord, influencé les langues de Taiwan, leur transmettant les nombres de 4 à 10, mais apparemment rien d’autre sur le plan linguistique. La transmission des nombres malayo-polynésiens aux langues de Taiwan se serait faite de façon graduée, les langues du sud recevant plus de nombres MP que celles du nord. Outre les nombres, des éléments culturels auraient été transmis par les malayo-polynésiens des Philippines aux formosans du sud, en particulier un certain type de poterie (“red-slipped pottery”, terme que Ferlus traduit par “lissé-rouge”), et le riz indica.

Voyons les problèmes inhérents à cette théorie.

1. Une famille kra-dai distincte ?

Laissant de côté les emprunts au chinois, le vocabulaire de base kra-dai contient une couche principale austronésienne et une couche secondaire, plus petite, austroasiatique. Si le kra-dai était une famille distincte, il devrait exister une couche lexicale indigène très fondamentale qui ne serait ni AN, ni AA, ni chinoise. Ferlus ne fait aucun effort pour caractériser une telle couche. Notons qu’elle ne comprendrait ni les pronoms personnels, ni les nombres, et bien peu de noms de parties du corps. En l’absence de caractérisation lexicale, l’idée d’une famille kra-dai distincte n’est pas recevable.

2. Les kra-dai auraient acquis l’agriculture des austroasiatiques.

Selon Ferlus, les premiers TK auraient été des végéculteurs (du taro, en particulier), et le vocabulaire kra-dai de la riziculture serait entièrement d’origine austroasiatique. Pourtant :

  • “riz décortiqué”, Proto-Tai *sa:l A, proto-Kra *sal A. Ce mot est d’origine austronésienne: PAn *qaSaN “grain non décortiqué”;
  • “riz cuit”, Proto-Kra *m-laɯ C. Ce mot est d’origine austronésienne: PAn *beRas “grain décortiqué”
  • “rizière inondée”, Proto-Kra *na A, proto-Hlai *ana B, également d’origine austronésienne : PAn *bena “champ de plaine”.

Il n’y a pas lieu de supposer que les kra-dai aient jamais eu un mode de vie non-agricole.

3. L’interaction malayo-polynésien/kra-dai au Guangdong : un scénario peu clair.

Selon Ferlus (p. 5) vers le 2e siècle avant notre ère, le malayo-polynésien du Guangdong “aurait été recouvert par une poussée du TK originel venant de l’intérieur.” Il ajoute : “Ce scénario explique pourquoi le vocabulaire commun au TK et à l’AN appartient au lexique fondamental.” Or, le gaulois a été recouvert par le latin, sans que le latin tradif parlé en Gaule reçoive du vocabulaire fondamental gaulois. Les explications sont insuffisantes.

4. La date du départ des malayo-polynésiens du Guangdong vers les Philippines.

Ferlus place ce mouvement vers 3000 avant notre ère. Or il n’y a pas trace d’un mode de vie néolithique aux Philippines jusqu’à 2000 avant notre ère.

5. “La parenté du vocabulaire partagé entre le TK et l’AN est à placer au niveau du PMP”

C’est une erreur. Les mots austronésiens en kra-dai n’ont subi aucune des innovations phonologiques du malayo-polynésien: *C et *t sont toujours distincts, *N et *n également, *S est toujours une sifflante, les métathèses de -h final n’ont pas eu lieu. En particulier, le mot “oeil” cité par Ferlus, PAN *maCa, devenant *mata en PMP, est *m-ʈa A en Proto-Kra (Ostapirat 2000, sur la foi du Gelao de Qiaoshang), et non *m-ta A qui serait la forme correspondant au PMP *mata.

Sur le plan lexical, les innovations malayo-polynésiennes dans le domaine des nombres sont présentes en kra-dai, mais on trouve aussi en kra-dai des mots non-malayo-polynésiens, par exemple PAN *puja ‘nombril’, remplacé par *pusej en PMP, et pourtant reflété par Proto-Kra *m-ɖaɯ A, Proto-Tai *ɗwɯ A, Proto-Hlai *urɨ A. La seule explication est que le kra-dai est issu d’une langue du sud de Taiwan ayant les innovations dans les nombres, mais pas toutes les innovations lexicales, et aucune des innovations phonologiques du malayo-polynésien.

6. Transmission des nombres malayo-polynésiens aux langues de Taiwan : vraiment, seulement les nombres ?

Les nombres de 5 à 10 appartiennent au lexique modérément fondamental. Ils sont plus facilement empruntables que les pronoms, parties du corps, noms des éléments de base de la nature ; que les verbes aller/venir, mourir, manger, dormir ; mais sont plus difficilement empruntables que le vocabulaire culturel (techniques, commerce, calendrier etc.). Il n’est pas concevable que des nombres soient transmis par contact sans que du vocabulaire culturel le soit aussi. Si des nombres malayo-polynésiens ont été transmis aux langues du sud de Taiwan, on devrait aussi y trouver toute une couche de mots culturels malayo-polynésiens. L’existence d’une telle couche ne devrait pas être trop difficile à établir, étant donné les innovations phonologiques très visibles du malayo-polynésien. Mais personne ne l’a jamais vue. Les emprunts malayo-polynésiens dans les langues de Taiwan… sont très peu nombreux, et généralement plutôt récents (tabac, écriture etc).

7. Transmission des nombres malayo-polynésiens aux langues de Taiwan : le pattern d’implications est ignoré.

J’ai montré dans mon article de 2004 que les nombres malayo-polynésiens apparaissent dans les langues de Taiwan selon un pattern constant: la présence de *puluq ’10’ implique celle de *Siwa ‘9’ et *walu ‘8’, qui impliquent *enem ‘6’, qui implique *lima ‘5’, qui implique pitu ‘7’; mais l’inverse n’est pas vrai. Dans la géographie, plus on descend vers le sud, plus le paradigme malayo-polynésien est complet. De ce pattern, Ferlus ne retient que la ressemblance croissante (en termes du nombre de formes communes) avec le malayo-polynésien du nord au sud de Taiwan. Le pattern d’implications est ignoré.

8. Transmission des nombres malayo-polynésiens aux langues de Taiwan : les nombres bas vont plus loin que les hauts.

On sait depuis Greenberg que les nombres sont d’autant plus faciles à emprunter qu’ils sont élevés : ainsi ’10’ s’emprunte plus facilement que ‘9’, qui s’emprunte plus facilement que ‘8’, qui s’emprunte plus facilement que ‘7’, etc. Dans le modèle de Ferlus, les nombres malayo-polynésiens qui sont empruntés par le plus grand nombre de langues (et donc remontent le plus au nord) sont ‘5’, ‘6’ et ‘7’. Ceux qui restent confinés au sud de Taiwan, et donc qui sont empruntés par le plus petit nombre de langues, sont ‘8’, ‘9’ et ’10’. C’est l’inverse du pattern habituel.

9. Transmission d’éléments culturels malayo-polynésiens au sud de Taiwan : le cas de la poterie lissé-rouge.

La poterie lissé-rouge est plus ancienne à Taiwan qu’aux Philippines : elle est dominante vers -2100 avant notre ère dans le sud-est de Taiwan, mais ne commence à apparaître qu’après -2000 aux Philippines (Hung 2008: 23, 107). En général, les artefacts austronésiens communs à Taiwan et aux Philippines sont plus anciens à Taiwan (ibid.). Leur transfert a évidemment eu lieu du nord vers le sud.

10. Transmission d’éléments culturels malayo-polynésiens au sud de Taiwan : le cas du riz indica.

Ferlus propose (p.6) que le riz indica a été introduit à Taiwan depuis les Philippines par les malayo-polynésiens qui l’auraient, suppose-t-on, amené avec eux, depuis le Guangdong, en 3000 avant notre ère. C’est impossible. Le riz indica apparaît en Inde du nord par hybridation vers 2000 avant notre ère, et ne devient une composante stable de l’agriculture dans cette région que dans la période 1600-1000 avant notre ère (Silva et al 2018). Par la suite, il se répand dans l’Asie du sud-est, continentale et insulaire, à la faveur de l’expansion indienne du premier millénaire de notre ère. Depuis l’état hindouisé de Srivijaya à Sumatra, par le commerce maritime, il remonte vers le nord jusqu’à Taiwan. Les malayo-polynésiens des Philippines jusqu’à Madagascar cultivent traditionnellement du riz “japonica tropical”, aussi appelé “javanica”. Ce riz a des précurseurs biologiques dans les riz japonica des aborigènes de Taiwan (travail en préparation).

11. Pourquoi le malayo-polynésien n’est-il pas représenté à Taiwan?

Ferlus demande pourquoi le malayo-polynésien “qui serait né sur l’île de Taiwan n’y est plus représenté aujourd’hui”. C’est parce que les innovations caractéristiques du malayo-polynésien se sont produites aux Philippines, après la migration. La langue du sud de Taiwan dont le malayo-polynésien s’est séparé il y a 4000 ans a évolué vers… le paiwan moderne. Le paiwan est la langue de Taiwan la plus proche du malayo-polynésien. Ils ont les mêmes nombres et des innovations lexicales communes : les mots *alap ‘prendre’ et *Cazem ‘tranchant’.

Retournons la question à Ferlus : pourquoi le malayo-polynésien qui serait né au Guangdong n’y est-il plus représenté aujourd’hui ?

Conclusion.

Bien que le chemin suivi par les Austronésiens depuis le continent ne soit pas le même chez Ferlus que dans la théorie standard, l’arbre phylogénétique impliqué par son modèle n’est pas différent du mien : les branches formosanes se séparent les unes après les autres du tronc commun, puis c’est au tour du malayo-polynésien. Un tel arbre a le potentiel d’exprimer l’idée que (à la différence du modèle de Blust) les langues formosanes ont des degrés de proximité différents avec le malayo-polynésien. Il contient potentiellement les noeuds auxquels accrocher les innovations successives *pitu 7, *lima 5, *enem 6, *walu 8, *Siwa 9, *puluq 10, expliquant ainsi simplement la hiérarchie d’implications entre ces nombres dans les langues de Taiwan. Dès lors, pourquoi rejeter mon modèle au profit d’un autre qui n’explique pas la hiérarchie d’implications ?

références

Hung, Hsiao-chun (2008) Migration and Cultural Interaction in Southern Coastal China, Taiwan and the Northern Philippines, 3000BC to AD 100: The Early History of the Austronesian speaking Populations. Canberra: Unpublished PhD dissertation, Australian National University.

Sagart, L. (2004) The higher phylogeny of Austronesian and the position of Tai-Kadai. Oceanic Linguistics 43,2: 411-444.

Search OpenEdition Search

You will be redirected to OpenEdition Search