Specks of dust, brooms and comets

Previously in this thread: I argued that the *-it-s vs. *-ut-s alternation affecting 寐, 穟, 季 and 悸 in Shi Jing rhyming is due to a dissimilatory sound change in early eastern China turning *Cʷut-s to *Cʷit-s. To that end I began to assemble a word-family based on a root *Pˤut (where ‘P’ stands for either of *b, *m-p- or *m-pʰ), with dust-related meanings including ‘powdery’, ‘duster=broom’ and ‘comet’, out of ‘broom’. So far, the family includes 孛 MC bwot, bwojH ‘comet’, 勃 MC bwot ‘powdery’ (of soil), and 彗篲 MC zwijH ‘broom, comet’. I have described how MC zwijH was derived out of root *Pˤut through lenition of the original labial initial, voicing of the prefix, and dissimilatory shift of *Cʷut-s to *Cʷit-s.

I now examine the initial in root *Pˤut. Was it *b-, *p- or *pʰ- ? Consider 嘒 ‘to chirp, of insects’, OC (BS) *qʷʰˤi[t]-s > MC xwejH. This is an old word: it rhymes as *-it-s in Odes 197 and 222, and also occurs with the meaning ‘small-looking’ (as glossed in the 毛傳:“微貌”) in Ode 21  小星, at the beginning of the first line of stanzas 1 and 2:

嘒彼小星 ‘small are those starlets’ (tr. Legge)

The maning of 嘒 in Ode 21 cannot simply be ‘small’, as 小 occurs two words later. In that ode, a prince’s concubines compare their fate  to that of a prince’s principal wife, describing themselves as faint stars in the sky. The idea is that of  being faint, almost invisible, insignificant. Supposing the root in 嘒 ‘small’ is the same as in 勃 MC bwot ‘powdery’ (of soil), we may infer the more specific meaning ‘tiny as a particle of dust’. In order to evolve to MC xw-, an OC labial stop would have to be aspirated, and to be lenited intervocalically: *Cə-pʰ-. The evolution would be parallel to that of 彗, via *Cə-ɸ- (lenition) >  *ɸ- (prefix loss) and finally xw-.

To my knowledge, this evolution path has not been proposed before. Are there any parallels ? the word 華 ‘flower’, MC xwae, is given the OC reconstruction *qʷʰˤra in Baxter and Sagart (2014) but this ignores 郭璞 Guo Pu’s comment to Fangyan 1. 7. 23 , that “江東呼華為荂, 音敷” (BS *pʰra); and the Shuo Wen entry for 葩 (reconstructable as *pʰˤra) : “葩, 華也”. It makes more sense to propose OC 【*Cə-pʰˤra 】 ‘flower’, with the same evolution as for 嘒:  through *Cə-ɸ-  >  *ɸ- > xw-  (I am using thick brackets【】for reconstructions that modify those in Baxter and Sagart 2014).

This parallel supports treating the root in 孛勃彗篲 as *pʰˤut, with the basic meaning ‘dust’ (v. and n.). 孛 ‘comet’, MC bwot and bwojH, then goes back to 【*m-pʰˤut 】 and 【*m-pʰˤut-s 】, where the prefix is the  instrumental nominalizer called *m1c– by Baxter and Sagart (2014:55). A broom (out of which we may suppose the meaning ‘comet’ arose) is an instrument of dusting. 彗 and 篲, both ‘comet’ and ‘broom’, involve the long, iambic form of *s-, more specifically *s2-, a prefix deriving circumstantial nouns including names of instruments (Baxter and Sagart 2014:56), plus *m1b-, which turns a noun into a verb of volitional/controlled action: thus 彗 and 篲【*sə-m-pʰˤut-s】’instrument for dusting=broom=comet’. 勃 MC bwot ‘powdery’ (of soil) would have had *N-, the stative-intransitive voicing prefix (Baxter and Sagart 2014:54): 【*N-pʰˤut】. Finally 嘒  ‘small as a speck of dust’ had an iambic prefix of whose function little can be said: 【*Cə-pʰˤut】 . To recapitulate:

Root pʰˤut ‘dust (n.), to dust’:

孛 ‘comet’【*m-pʰˤut 】 and 【*m-pʰˤut-s 】> bwot and bwojH

勃 ‘powdery’ (as soil) 【*N-pʰˤut 】> bwot

彗篲 ‘broom, comet’ 【*sə-m-pʰˤut-s】> zwijH

嘒  ‘small as a speck of dust’ 【*Cə-pʰˤut】> xwejH

And, from another root:

華 ‘flower’ 【*Cə-pʰˤra 】 > xwae

Cite this post as: Laurent Sagart, "Specks of dust, brooms and comets," in Sino-Tibetan-Austronesian, 14/02/2021, https://stan.hypotheses.org/1272.

Alternating *-it-s and *-ut-s rhymes in the Shi Jing: a dissimilatory change after all

In a recent post I listed six cases of unexpected rhyming between *-it-s and *-ut-s in the Shi Jing, noting that this always occurred after a labial initial. I treated these examples as resulting from an assimilatory process spreading the labial feature onto the vowel. From the sections of the Shi Jing where these rhyme contacts occur, I deducted that the change had taken place in (north)western China.

Guillaume Jacques immediately commented that the reverse interpretation might be worth considering:  a dissimilatory process changing *-ut-s to *-it-s after a labial. In that case, the four labial-initialed *-it-s words (as  reconstructed in Baxter and Sagart 2014) 寐 *mi[t]-s, 穟 *s.[ɢ]ʷi[t]-s, 季 *kʷi[t]-s, 悸 *[g]ʷi[t]-s must have their OC rhyme  emended to *-ut-s : they evolved to *-it-s through labial dissimilation only in eastern China, while continuing to rhyme as *-ut-s in western and northern regions (Daya, 王風 Wang Feng and 衛風 Wei Feng sections).

In a first response,  dated 28-02-2020, I resisted this alternative treatment, missing the point that the change could be dissimilatory and the vowel in the OC reconstructions 寐 *mi[t]-s, 穟 *s.[ɢ]ʷi[t]-s, 季 *kʷi[t]-s, 悸 *[g]ʷi[t]-s, inferred as it was from MC,  could be innovative. In a second comment to the same post four months later I finally saw the light, citing Lushai mut ‘to sleep’ as a potential cognate of 寐 ‘sleep’ in support of reconstructing 寐 as *mu[t]-s in OC. In this post I present more evidence for the dissimilatory view. I am indebted to Bill Baxter for bringing some of the relevant facts to my attention. In what follows I will take as established an east-west dialect distinction in Shi Jing times, described in a forthcoming paper by Ma Kun and myself.

If one of the four words: 寐穟季悸 rhyming as *-it-s or *-ut-s depending on the region had a second reading in MC *-t, we would directly know the direction of change: as suffixed *-s is a necessary condition for the change to occur, the s-less form would show the original vowel. That is, unfortunately, not the case. We can still look for a word-family relating *-it-s and *-ut-s words, and also including s-less members.

Let us consider the words for ‘broom’ and ‘comet’. A comet—really— looks like a certain type of broom, so it is not surprising that a Chinese word combines the two meanings: 彗 *s-[ɢ]ʷi[t]-s > zwijH > huì ‘broom; comet’. A synonym is 孛 *[b]ˤ[u]t-s > bwojH > bèi ‘comet’. These two words have the rhymes involved in the *-ut-s / *-it-s alternation, and both have labial or labialized initials. 孛 has a second, s-less reading: *[b]ˤ[u]t > bwot > bó ‘comet’. If 彗 and 孛 can be shown to be etymologically related, then the direction of change cannot be from *-it-s to *-ut-s because the s-less reading of 孛 is compatible with *u and *ə, but not with *i.

But are they related ? as reconstructed in Baxter and Sagart (2014), the two words involve very different roots: [ɢ]ʷi[t and [b]ˤ[u]t. Initial [ɢ]ʷ- in the former was reconstructed because the phonetic series of 彗 includes words with MC x- and hj-, both compatible with an uvular, and a voiced obstruent was needed to voice the prefix. But that reconstruction is not inscribed in marble. I am going to argue that one can derive both 彗 and 孛 out of the labial-initial root in 孛.

The two readings of 孛 ‘comet’ are based on a root Pˤ[u]t (where ‘P’ stands for a labial stop, and [u] stands for either *ə or *u). There is an external cognate: Tangkhul phut ‘broom’ (Bhat 1969). VanBik (2009) reconstructs Proto-Kuki-Chin *phut ‘dust, powder’ which is probably related—a broom is an instrument for dusting. Matisoff’s STEDT web site reconstructs PTB *pʷut ⪤ *hwut ASHES / DUST, which includes Written Burmese phut ‘dust’.  Schuessler (2007:171) offers 勃 ‘powdery’ (sc soil) as a Chinese cognate (reconstructible as *[b]ˤ[u]t > MC bwot in the Baxter-Sagart system). This word is most likely related to 孛 ‘comet’ via ‘broom’. Baxter and Sagart (2014) reconstruct OC independently from TB, but since OC *ə usually corresponds to a in other ST languages, the vowel outside of Chinese argues, though not conclusively, for *Pˤut over *Pˤət.

Let us now suppose that 彗 MC zwijH ‘broom; comet’ contains root *Pˤut. We can explain the  z- initial as a voiced outcome of the OC *s(ə)- prefix,  and the -w- medial as the lenited version of the labial stop in the root. The prefix would have undergone voicing assimilation on the stop—we would need to reconstruct either of OC *bˤ-, *m-pˤ-, or *m-pʰˤ-, which all converge into [b]- before MC, according to the BS system (I am excluding *N-pˤ- and *N-pʰˤ which would only make sense in an intransitive verb). The *[b]- in the OC reconstruction of 孛 in Baxter and Sagart (2014) represents either one of those three  onsets. Lenition of the stop would be more natural in an intervocalic environment, suggesting *sə-[b]-. Evolution to zw- would be via *sə-b- > *sə-β- (lenition) >  *s-β-  (loss of prefix vowel) > *zβ- ( s- > z- voicing assimilation) and finally zw-(defricativization). The appearance of -w- before *u would be what triggered the change  from *-ut-s to *-it-s in the east: *zwut-s > *zwit-s: in other words, a dissimilatory change.

There is a problem with pharyngealization. I have assumed that the root common to 孛 and 彗 had a pharyngealized initial *Pˤ. Now OC pharyngealized words normally do not evolve to MC division 3, yet 彗, MC zwijH, is a third-division word. Why is that ? while in general loosely attached preinitials/prefixes (i.e. having the form *Cə) can occur in all MC divisions, *sə- as reconstructed in the Baxter-Sagart system only ever occurs in MC third-division words. There is something in that prefix which is incompatible with root initial pharyngealization, causing it to be lost. I suspect the prefix was [si] rather than [sə] as Baxter and Sagart reconstructed it, and the front vowel prevented coarticulatory anticipation of the pharyngealization gesture in the following consonant.

Let us go back to the proposed role of medial -w- in triggering the change from *-ut-s to *-it-s: this fails to explain the participation of 寐 in the change, as the initial is a plain *m. The price to pay for our analysis is that we need to stipulate a minor change turning OC [mu] into [mʷu] in the eastern pronunciation, before the *-ut-s > *-it-s change took place. But the idea that medial -w- triggered the change does fit in rather well with the absence of MC labial stop initials among the four words involved. Under a dissimilatory account, the -it-s/-ut-s alternation is conditioned by a labialized initial *Cʷ- preceding *-ut-s, not by a labial consonant preceding *-it-s.

I have shown that deriving 彗 and 孛 from a single root *Pˤut is feasible, even if it requires us to add a hitherto unknown lenition process to our OC reconstruction toolkit. This supports the dissimilatory analysis. In a forthcoming post, I will add new members to the proposed word-family and investigate the phonological and morphological variation it contains.


Baxter, William H. and Laurent Sagart. 2014. Old Chinese: a new reconstruction. New York: Oxford University Press.

Bhat, D. N. Shankara. 1969. Tankhur Naga vocabulary. (Deccan College Building Centenary and Silver Jubilee Series 67). Poona: Deccan College Postgraduate and Research Institute.

Schuessler, A. 2007. ABC etymological dictionary of Chinese. Honolulu: University of Hawaii Press.

VanBik, Kenneth. 2009. A reconstructed ancestor of the Kuki-Chin languages. STEDT monograph 8. Berkeley, University of California.

STEDT: https://stedt.berkeley.edu/~stedt-cgi/rootcanal.pl

Cite this post as: Laurent Sagart, "Alternating *-it-s and *-ut-s rhymes in the Shi Jing: a dissimilatory change after all," in Sino-Tibetan-Austronesian, 14/02/2021, https://stan.hypotheses.org/1233.

The PAN coda *-X in Favorlang

In Sagart (2019 here) I reconstructed a new PAN consonantal ending : *X [χ]. This is one source of the Kra-Dai tone B. In the same paper I showed that at the end of Austronesian words, two previously reconstructed h-type sounds: *H1 and *H2, evolved to different tones in Kra-Dai: *H1 to tone C and *H2 to tone A. In Table 7 of that paper I further assigned my *X to two words having no Kra-Dai cognates: *qeCaX ‘chaff’ and *duRiX ‘thorn’.  The reason for claiming these words ended in *X was that, while there was evidence in each case for a final voiceless fricative (Amis ʔtah ‘chaff’, Aklanon dugih ‘thorn’), that fricative could neither be *H1 nor *H2, as shown by the combination of  cited Saisiyat, Amis and Bunun forms: only *X was consistent with all the evidence.

Favorlang, an extinct Austronesian language of the Formosan west coast, recorded by Dutch missionaries in the mid-17th century brings additional evidence. I rely on Happart’s Favorlang vocabulary of 1650, as published in Campbell (1896).

As represented in those sources, Favorlang regularly loses the PAN endings *H1 and *H2:

*-H1: bao ‘new’ < baqeRuH1; bagg’ (o chau) ‘a live coal’ < baRaH1; ema < *qumaH1 ‘a field’, atshia ‘a salting’ < qasiRaH1.

*-H2: torro ‘three’ (in manna-torro-us ‘three times’) < tuluH2; manna-pida ‘to be often repeated’ < *pijaH2; charina ‘ear’ < CangilaH2; abo < qabuH2 ‘ash’; kairi ‘left’ < ka-wiRiH2; bato ‘stone’ < batuH2.

Yet *qeCaX ‘chaff’ is reflected in Favorlang with a final back-of-the mouth fricative, as chach or chagh ‘the powder of ground seed; the refuse of bruised corn’ (understand ‘the by-product of pounding grain: chaff’). PAN *q is regularly lost in Favorlang, and PAN *C evolves regularly to ch. Dutch orthographic ch writes [x] and gh writes [x] or [ɣ] : this may have been close to the value of the fricatives in chach/chagh: [xax], [xaɣ] or similar. The first ch- reflects PAN *C; the last consonant: –ch or –gh, my *X. It cannot reflect *H1 or *H2, since as shown above both go to zero in Favorlang.

The behaviour of final *-X in Favorlang is parallel to that of final -h in the Amis cognate ‘tah ‘chaff’: this -h cannot reflect *H2, as final *H2 goes to zero in Amis. It could reflect *H1, since Amis reflects final *H1 as –h, but *H1 is contra-indicated by Saisiyat kæ-ʔsæʔ ‘chaff’—expect ʔsæh if PAN was *qeCaH1. Thus geographically distant Amis and Favorlang agree in having in ‘chaff’ a final back-of-the-mouth fricative that can neither reflect *H1 nor *H2.


Happart, Gilbertus (1650 [1896]) Woordboek der Favorlangsche Taal. Published in English translation as Happart’s Favorlang vocabulary in W. Campbell (ed.) The articles of Christian Instruction in Favorlang-Formosan Dutch and English. London, Kegan Paul, Trench and Trübner, 1896, pp. 122-199.

Sagart, Laurent. 2019. A model of the origin of Kra-Dai tones. Cahiers de Linguistique Asie Orientale, 2019, 48 (1), pp.1-29. ⟨10.1163/19606028-04801004⟩

Cite this post as: Laurent Sagart, "The PAN coda *-X in Favorlang," in Sino-Tibetan-Austronesian, 24/06/2020, https://stan.hypotheses.org/963.

Labial fortitions in Kra-Dai

In his study of Proto-Kam-Sui (PKS) initials, Ferlus (1996) tentatively reconstructed initial *ˀl- based on a set of four comparisons with high-register tones, including ‘wasp’ and ‘boar’:

Dong Shui Maonan Mulao
wasp la:u A1 lu A1 du A1 -lu A1
boar lai B1 -la:i B1 -da:i B1 -la:i B1

The probable Austronesian prototypes are *waNu ‘honeybee’ and *waNiS ‘boar’s tusk, boar’ in Blust’s PAN reconstructions. Recall that *N was a lateral, probably simply [l], in PAN, until it shifted to *n after Proto-Southern-Austronesian, in PMP. As a daughter of PSA (here), PKD retained *N as [l]. Evolution of *waNiS to tone B in PKS is unexplained, but has a parallel in the reflex of *gumiS ‘moustache, beard’ in Kra: Buyang mui 11 (B) ‘body hair, feather’.

In PSA, initial *w- became *ʔw- with phonetic glottal stop: ʔwalu, ʔwaliS. This ʔw- underwent fortition to *p- in Proto-Hlai (Ostapirat *pəlu A, Norquest *p-lu: A). In PKS, *ʔwal- simplified to ʔwl- and ultimately to Ferlus’s *ʔl-.

The Kam-Sui word for ‘blood’, Ferlus *pʰr-, Thurgood *phla:t 7, can now more confidently be related to PAN *huRaC ‘artery, blood vessel, blood vein’ (Blust) with fortition of initial *XuR- to *pʰr- via [ɸr-] vel sim., parallel and similar to the fortition of *ʔul- to *pl- in Hlai.


Ferlus, Michel. 1996. Remarques sur le consonantisme du proto kam-sui. Cahiers de Linguistique Asie Orientale 25: 235-278.

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

Ostapirat, W. 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.

Cite this post as: Laurent Sagart, "Labial fortitions in Kra-Dai," in Sino-Tibetan-Austronesian, 15/05/2020, https://stan.hypotheses.org/894.

Southern Austronesian *sa-puluq-at in numbers 10 to 19

(view clickable tree)

Jump to Southern Austronesian master post

This post returns to a subject discussed in a 2010 conference paper (here). New facts are presented and the interpretation is slightly modified.

Several Philippine languages belonging to the South-Central Cordilleran subgroup in northern Luzon: Kankanay, Kalinga, Bontok, Ifugao, Balangaw, Inibaloi, have a conjunction reflecting Proto-South-Central Cordilleran *et [ət] ‘and, and then’. In the Central Luzon group of Philippine languages, Kapampangan, not in direct contact with the Cordilleran languages, has at ‘and’, reflecting either *et [ət] or *at. Tagalog, a Central Philippines language, has at ‘and’, which can only reflect *at, unless it is a loan from neighboring Kapampangan at ‘id.’ (personal communications from L. Reid and R.D. Zorc, April 2007). Kapampangan is known to be a source of loanwords to Tagalog. Among the Central Cordilleran languages reflecting *et ‘and’, Isinai, Kalinga and a few others use a phonologically reduced form of that morpheme between pulu ‘ten’ and a following unit numeral. Thus in Kalinga, ‘ten’ is simpulu < *sa-n(g)a-puluq, but ‘eleven’ is simpu:lut osa (Tryon et al. 1995, Vol. 4, 43-45). That last form is analyzable as /sin-pulu=t osa/, where /=t/ ‘and’, /osa/ < *əsa ‘one’. Similarly, ‘twelve’ is simpu:lut duwa /sin-pulu=t duwa/, where /duwa/ < *duSa ‘two’ and /=t/ ‘and’. Kapampangan, the only Central Luzon language with a reflex of *et, does not use it in compound numerals. Although Tagalog may have borrowed the conjunction from Kapampangan, it does use a reduced form of at in compound numerals above ‘20’, in a very similar way to Kalinga and Isinay:


twenty’ < *daduha ‘two’ + nasal linker + *puluq ‘ten’

dalawampút isá

twenty-one’ < *daduhatwo’ + nasal linker + *puluq ‘ten’ + =(a)t ‘and’ + *esa ‘one’

dalawampút dalawá

twenty-two’ < *daduhatwo’ + nasal linker + *puluq ‘ten’ + =(a)t ‘and’ + *daduha ‘two’

The puluq-t-numeral form cannot easily have been borrowed from Cordilleran into Tagalog, as the languages are not in contact: if Tagalog borrowed the conjunction from Kapampangan, Tagalog shows a development parallel to Cordilleran in its use of the conjunction.

There are two possible accounts of the development of the conjunction itself in the Philippine languages. In the first, the conjunction’s oldest form in the Philippines was *et [ət], regularly evolving to at in Kapampangan, and loaned to Tagalog as at. Under that account, the conjunction, inherited only in the South-Central Cordilleran and Central Luzon groups, reconstructs to the northern clade of Philippine languages in the phylogeny of Gray, Drummond and Greenhill (2009), which this post follows for Malayo-Polynesian. Use of the conjunction between *puluq and a following numeral in Cordilleran and in Tagalog must be the result of parallel innovations.

In a second account, the conjunction’s oldest form in the Philippines was *at. Kapampangan and Tagalog inherited it in that shape, while in the Cordilleran languages the presumably unstressed vowel underwent centralization, to [ət]. Under that account, Tagalog being part of the southern clade, both *at ‘and then’ and the *puluq=at+numeral construction reconstruct higher, to the ancestor of the southern and northern Philippine clades.2

Whether as *et or *at, the Philippine conjunction is reminiscent of the form ato ‘and, with’ in Amis (central Amis: Fey 1986, Namoh 2013) and of the conjunction katu in Paiwan. In Nataoran Amis (Nataoran Amis: Bril fieldwork), atu may occur in numeral expressions (Isabelle Bril, p.c., 11 February 2014):

a cacay-en atu la-tusa ku ni-pabeli itakuan-an tu pawli

HORT one-UV and half NOM PERF.NMZ-give me-OBL ACC banana

Give me one and a half banana (lit. ‘let it be one and a half the giving a banana to me’)

Both Namoh (2013:136) and Bril treat ato~atu as originally composed of a ‘and’ plus =to ‘oblique common noun marker’, drawing a parallel to aci ‘and + personal noun’ (<a ‘and’ + =ci ‘personal noun marker’). The oblique common noun marker to itself is analyzable into t- oblique and u or o common noun marker.

In a northern Amis numeral system recorded by Tsuchida (here), tu, a shortened form of atu, serves in numerals between 11 and 19 as a linker between the locally innovated word for ‘ten’: sabaw , and the following unit number, e.g. sabaw tu tulu ‘13’, sabaw tu lima ‘15’.

Similarly, in a Paiwan dialect recorded by Chang Hsiou-chuan (here), a conjunction katu occurs between *sa-puluq and a following unit number, e.g. tapuluʔ katu ʔunəm ‘16’. katu is analyzeable as katu ‘and-oblique noun marker’, or possibly kaatu ‘andoblique noun marker’.

While *sa-puluq, the Puluqish word for ‘10’ ends in *-q, which correspoponds to -k in Kra-Dai, the Kra-Dai word for ‘10’ ends in *-t, as if it came from *pulut. The Philippine developments suggest an explanation. The following scenario is based on the hypothesis that the oldest Philippine form of the conjuction was *at. It is also possible to explain the facts under the view that it was *et, but the explanation is slightly more complex.

Comparison of Amis and Paiwan leads one to posit a conjunction *atu ‘and’ between *sa-puluq ‘10’ and a following numeral. In PSA the presumably enclitic conjunction was reduced to *at through syncope. Proto-Malayo-Polynesian and Proto-Kra-Dai, the direct daughters of PSA, inherited both *at and the compound numeral construction with it. Proto-Malayo-Polynesian in turn transmitted these to its immediate daughter Proto-Philippines, but the conjunction and construction were lost in Malayo-Polynesian outside of the Philippines. For this to occur, a single loss is required as such a node is supported with 98% posterior probability in Gray, Drummond and Greenhill (2009). Soon after Kra-Dai branched off, and independently from developments in the Philippines, *at ‘and’ was further reduced to /t/, and the resulting *sa-puluqt was reduced to *sa-pulut in PKD. This then evolved to Proto-Hlai *apu:c, Proto-Kra *pwlot. The outcome in Proto-Kam-Tai is unknown because these KD languages have abandoned Austronesian numerals for Chinese ones.

Cite this post as: Laurent Sagart, "Southern Austronesian *sa-puluq-at in numbers 10 to 19," in Sino-Tibetan-Austronesian, 07/05/2020, https://stan.hypotheses.org/881.

iwa vs. siwa ‘nine’ in Puyuma

(jump up to Puluqish)

The Walu-Siwaish word for ‘nine’ is reconstructed as *Siwa with capital *S. This is expected to give iwa in Puyuma, where *S evolves to zero. As recorded by Li and Tsuchida (here), Nanwang Puyuma indeed has iwa, but Thomson 1875 gives siva for the ‘tribe at Pilam’, which is 卑南, modern Beinan, Nanwang Puyuma.  Cauquelin’s dictionary of the same dialect gives both iwa and siwa, the latter  used in serial counting. Puyuma initial s- reflects PAN (and PWS) *s.

In an insightful paper, Blust (2001) stated that Formosan languages reflect *Siwa while Malayo-Polynesian languages reflect *siwa. He argued that the irregular shift of *S to *s in ‘nine’ in PMP was motivated by the proximity of *sa-puluq ‘ten’, which begins with *s-.  The existence of siwa in Puyuma shows that the shift from *S to *s in ‘nine’ is older than PMP. If Blust’s analogical explanation is correct, then Puyuma, or an ancestor of Puyuma, must once have had *sa-puluq  for ‘ten’. This is precisely what the etymology I have recently presented for *puluq ‘ten’ implies.


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

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

Thomson, J. 1875. The straits of Malacca, Indo-China, China or ten years’ travel, adventures and residence abroad. London: Sampson Low, Marston, Low and Searle.

Cite this post as: Laurent Sagart, "iwa vs. siwa ‘nine’ in Puyuma," in Sino-Tibetan-Austronesian, 11/04/2020, https://stan.hypotheses.org/664.

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) examplified 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, only *S2 is possible. More examples can be added. By my count at least 15 reconstructable Austronesian words include S2.

See this table.

15 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). Th 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 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).


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.

Cite this post as: Laurent Sagart, "The merger of PAN *S1 and *S2," in Sino-Tibetan-Austronesian, 05/04/2020, https://stan.hypotheses.org/618.

OC ‘wing’ and ‘next day’

Baxter and Sagart (2014) (hereafter ‘B&S’) reconstruct ‘wing’ as *ɢʷrəp > MC yik. Although the MC form ends in -k, and ‘wing’ constantly rhymes as *-ək in the Shijing odes—is quite a prolific rhyme word in the Shijing—OC *-p was proposed on the basis of the phonetic element 立, OC *k.rəp, which occurs in some tokens of the Shang graph, alongside a naturalistic drawing of a wing. Moreover, with or without , the graph for ‘wing’ serves as a jiajie for ‘next day’, presumably because the two were close or identical in pronunciation—it does not seem likely that the two are etymologically related. In Middle Chinese and Modern Standard Chinese, or yik > yì ‘next day’ is a homophone of ‘wing’ yik > yì ‘wing’. Note that, as with ‘wing’, the (modern) characters for ‘next day’ include phonetic , implying OC *-p here too. The replacement of final *-p by *-k in OC is best treated as the result of a sound change: any other kind of explanation would hardly explain the two words simultaneously. B&S therefore assign ‘next day’ the same OC form as ‘wing’, i.e. *ɢʷrəp. They assume that dissimilation between the labial elements in the onset and final caused the *-p coda to shift to *-k, in both words. Interestingly final nasals seem unaffected by the change, for instance

*prəm > pjuwng > fēng ‘wind (n.)’

consistently rhymes with other *-əm words, such as *səm > sim > xīn ‘heart’ in the odes. The labial initial in ‘wind’ has not triggered a change from *-m to *-ŋ. Such a change will eventually take place, but later, after the Shijing period. Change of *-p to *-k in words with labial initials (even over medial -r-) is very early, predating Shijing rhyming, perhaps in terminal Shang or initial Zhou times.

As justification for the *ɢʷr- onset in ‘wing’, Baxter & Sagart (2014:386 fn 30) note that often writes *[ɢ]ʷrəp-s > hwijH > wèi ‘standing, position’. This word they treat as a contraction of a honorific *ɢʷəʔ (as in 有商 ‘the Shang’) and root *rəp ‘to stand’, with a nominalizing *-s suffix. They suppose that in ‘wing’ and ‘next day’, phonetic stands for *[ɢ]ʷrəp-s, whence the onset proposed for the two words.

The phonetic is not associated with the graph for ‘wing’ before Zhou bronze script: there is no need to assume final -p and dissimilation from a labial onset: B&S reconstruct *ɢ(r)ək-s > yiH > yì ‘different’.

These reconstructions resolve an interesting range of issues, but there is a difficulty: Qiu Xigui (2012) observed that occurs in the Shijing and Shujing in some of the same contexts where was used in the oracular inscriptions. This is difficult to explain with our reconstructions *ɢ(r)ək-s and *l̥ək: in the current framework there is no way that an OC *ɢ- or *ɢ(r)- can evolve to a late OC lateral; but *m-r- does evolve regularly to late OC *l- (Baxter and Sagart 2014:133). Consequently it is preferable to reconstruct as *m-rək. Further, would be a perfect phonetic for ‘wing’ in Zhou times is ‘wing’ itself were *m-rək. Replacing initial *ɢʷr by *m-r- would not prevent labial-to-labial dissimilation, since m is labial too: a terminal Shang-time or initial Zhou-time *m-rəp would dissimilate to early Zhou *m-rək just as well. In addition, the new proposal would avoid the complication of supposing that in ‘wing’ and ‘next day’, phonetic stands for . If *k is a prefix in *k.rəp > lip > lì ‘stand (v.)’ (our reconstruction leaves the possibility open) then the root *rəp is eligible for prefixation by *m-, which in verbs marks actions controlled by the subject.

An OC *m-r- would give *z- in Proto-Min (p. 134, table 4.39). Indeed, ‘wing’ has *z- in Proto-Min (Norman 1974:34).

Externally, the new reconstruction *m-rəp ‘wing’ would make an attractive match for Proto-Tangkhulic *raap ‘rib’ (Mortensen) and the etymon #558 PTB *s/b-ram RIB in the STEDT. The folded wings of a bird are located along its sides, like the ribs.


OC (proposed)

Proto-Min (Norman 1974)


*m-rəp > yik > yì ‘wing’


*s/b-ram RIB

*m-rəŋ > ying > yíng ‘fly’


*s-b-(r/y)aŋ FLY (n.) / BEE

where OC *m-r- appears to correspond to the /b-r-/ onset strings in the two STEDT reconstructions.


Baxter, William H. and Laurent Sagart. 2014. Old Chinese: a new reconstruction. New York: Oxford University Press.

Norman, Jerry. 1974. The initials of Proto-Min. Journal of Chinese Linguistics 2.27-36.

Qiú Xīguī 裘錫圭. 2012. Bǔcí “[yì]” zì hé Shī, Shū lǐ de “[shì]” zì 卜辭 “異” 字和詩、書裏的 “式” 字. Qiú Xīguī xuéshù wénjí 裘錫圭學術文集, vol. 1, 212–229. ( 6 v.). Shànghǎi 上海: Fùdàn dàxué chūbǎnshè 復旦大學出版社.

Cite this post as: Laurent Sagart, "OC ‘wing’ and ‘next day’," in Sino-Tibetan-Austronesian, 21/02/2020, https://stan.hypotheses.org/526.

A dialectal sound change in the Shi Jing: -it-s > -ut-s after labial consonant

[This post was revised March 12 and 15, 2020]

There are at least six examples in the Odes of OC *-it-s rhyming with *-ut-s (whether original OC *-ut-s, or *-ut-s from OC *-up-s). This happens only when *-it-s is preceded by a labial consonant. Most likely, a dialectal sound change has taken place, changing *-it-s to *-ut-s after a [+labial] consonant, and allowing words pronounced as *-it-s elsewhere to rhyme with *-ut-s:

*-it-s > *-ut-s /[+labial] __

The locations in the Odes where this rhyming can be seen are:

  • 58.5 *mi[t]-s with *sə-lu[t]-s,
  • 60.1 悸 *[g]ʷi[t]-s with *sə-lu[t]-s,
  • 65.2 穟 *s.[ɢ]ʷi[t]-s with 醉 *Cə.tsu[t]-s [added March 15]
  • 241.3 季 *kʷi[t]-s with 對 *[t]ˤ[u]p-s,
  • 241.4 季 *kʷi[t]-s with 類 *[r]u[t]-s,
  • 256.4 寐 *mi[t]-s with 內 *nˤ[u]p-s.

Presumably each of the twelve words above ended in [uts] in the pronunciation these odes were composed in, whence the particular rhyming pattern. Note that rhymes twice with *-ut-s in different stanzas of ode 241, each time with different words: this shows the phenomenon is not a fluke, but a dialectal characteristic.

This rhyming tends to make the OC vowel in *[t]ˤ[u]p-s and *nˤ[u]p-s more secure, as the alternative to *u in both cases is *ə, not *i. The brackets around [u] are no more needed.

Odes 241 and 256 are part of the Da Ya, a section believed to be associated with western Zhou and (north-)western China at the time. Odes 58, 60 and 65 are from the Guo Feng section (58 and 60 from 衛風, Hebei-Henan boundary;  65  from  王風 , We may be dealing with a northern/northwestern isogloss.

For the numbering of the Odes readers should refer to Appendix B in Baxter (1992), and to Mattis List’s Shijing Rhyme Browser (March 12, 2020: thanks, Mattis, for modifying  your browser to allow queries by OC rhyme! reader, to get the browser to list all Shi Jing occurrences of a certain OC rhyme, type the rhyme in B&S system, within double angle brackets, like this: «ək», in the query window).

A counterexample ?

There is in the entire Da Ya and Wei () Feng sections of the Shi Jing a single example of *-it-s after a labial NOT rhyming as *-ut-s: in 245.4 (Daya), ‘ear of grain’, OC *s.[ɢ]ʷi[t]-s (same word as , the older graph) rhymes with , MC bajH. This MC form could go back to either of *-at-s or (following a labial consonant) *-ots , but the fact that the phonetic series’s head-word is 巿 *put ‘knee covers’ suggests that the MC reading with -ajH is anomalous, even if from *-ot-s, and that Ode 245 read as *-ut-s.

Consequently there is no strong counterexample to the change *-it-s > *-ut-s /[+labial] __ in the Da Ya, Wei () Feng and Wang (王) Feng sections of the Odes, and we may regard this as a constant dialect feature of these sections of the Shi Jing.

Outside of the Daya, Wei () and Wang (王) Feng, the change does not occur at all. In the Guo Feng we have:

  • 30.3 *mi[t]-s with 嚏 *[t]ˤi[t]-s,
  • 53.1 *pi[t]-s with 四 *s.li[j]-s,
  • 110.2 *mi[t]-s and *kʷi[t]-s with *[kʰ]i[t]-s.

Odes 30, 53 and 110 belong, respectively, to Bei Feng 邶風 (Henan, between Zhengzhou and Anyang), Yong Feng 鄘風 (Henan, near Bei) and Wei Feng 魏風 (south Shanxi): a region on the middle course of the Yellow River valley.

In the Xiaoya there are no examples of the change, at all. Where *-it-s occurs after a labial consonant, it rhymes with *i words:

197.4 嘒 *qʷʰˤi[t] -s and 淠 *pʰˤi[t]-s and *mi[t]-s with 屆 *kˤr[i][t]-s,

222.2 qʷʰˤi[t] -s and 淠 *pʰˤi[t]-s with *s.li[j]-s and 屆 *kˤr[i][t]-s.

In these two instances, the presence of 屆 kˤr[i][t]-s and 駟 *s.li[j]-s in the rhyme sequences guarantees that 嘒* qʷʰˤi[t] -s, 淠 *pʰˤi[t]-s and 寐*mi[t]-s were pronounced with *i vowel.

[added Feb. 14, 2021: find a sequel to this post here]

Cite this post as: Laurent Sagart, "A dialectal sound change in the Shi Jing: -it-s > -ut-s after labial consonant," in Sino-Tibetan-Austronesian, 06/02/2020, https://stan.hypotheses.org/450.

Chinese-Tibetan vowel correspondences I

Hill (2019: §33, §224) reconstructs a vowel *e in the Sino-Tibetan parent language (Proto-Trans-Himalayan in his terminology; I will be using the term PCT ‘Proto-Chino-Tibetan’, meaning the language one would reconstruct if only Old Chinese an Tibetan were available). He argues that this vowel is preserved unchanged in OC, and that its reflexes in WT are conditioned by the ending: one has WT e before labials, i before velars, and a before dentals (including –r and –l). Below are examples of the correspondence OC *e to Written Tibetan a in words without a coda (1), with labial codas (2-3), velar codas (5-6), PCT *-q (7) and PCT *-ʔ (8-9)

OC *e : WT a

1. 溪 *kʰˤe > khej > xī ‘valley with stream in it’ : WT rka ‘small furrow conveying water from a conduit to trees and plants’
2. 占 *tem > tsyem > zhān ‘prognosticate’ : WT gtam ‘talk, discourse, speech’
3. 攝 *kə.n̥ep > syep > shè ‘catch, gather up’ : WT rnyab ‘seize or snatch together’
4. 貞 *treŋ > trjeng > zhēn ‘straight; correct’ : WT draŋ-po ‘straight’
5. 脛 *m-kʰˤeŋ-s > hengH > jìng ‘leg, shank’, 牼 *m-kʰˤeŋ > heang > kēng ‘shank bone’ : WT rkang ‘marrow, leg bones, stalk’
6. 清 *tsʰeŋ > tshjeng > qīng ‘clear (adj.)’, 淨 *m-tseŋ-s > dzjengH > jìng ‘cleanse (v.t.)’ : WT g-tsang-ba ‘to be clean, pure’
7. 舐 *Cə.leʔ > zyeX > shì ‘lick’ : WT ldag pa ‘to lick’
8. 庳 *N-peʔ > bjieX > bì ‘low, short’, 鼙 *[b]ˤe > bej > pí ‘small hand-drum’ : WT p’ra-ba ‘thin, fine, minute; little, small’
9. 諀 *pʰeʔ > phjieX > pǐ ‘slander’ : WT p’ra-ma ‘slander’

Similarly, the correspondence OC *e : Tibetan e is not limited to words with labial endings. Below are examples in words without a coda (1), ending in some kind of back-of-the-mouth fricative (2-3) and, in apparent violation of Dempsey’s law, in a velar (4-5):

OC *e : WT e

1. 佳 *[k]ˤre > kea > jia ‘good’ : WT dge ‘happiness, welfare, happy, propitious’
2. 秝 *[r]ˤek > lek > lì ‘successively, sequence’ : WT re ‘one at a time’, etc.
3. 譬 *pʰek-s > phjieH > pì ‘example’ : WT dpe ‘pattern, model; example, illustrative parable’
4. 積 *[ts]ek-s > tsjeH > jī ‘to hoard’ : WT rtseg-pa, pf. (b)rtsegs ‘to lay one thing on or over another, to pile up, stack up, build up’
5. 易 *lek-s > yeH > yì ‘easy’ : WT legs-pa ‘good, happy; neat, elegant, beautiful …’

These two correspondences of OC *e are not in complementary distribution.

Hill, Nathan W. 2019. The Historical Phonology of Tibetan, Burmese and Chinese. Cambridge: Cambridge University Press.

Cite this post as: Laurent Sagart, "Chinese-Tibetan vowel correspondences I," in Sino-Tibetan-Austronesian, 08/01/2020, https://stan.hypotheses.org/413.