*q   

*qa particle expressing confirmation: indeed

*qa₁ locative particle

*qabaŋ₁ boat, canoe

Note:   Also Tsou apaŋə ‘boat, canoe, Favorlang/Babuza abak ‘small boat or sampan’. Given the independent evidence for PAn *layaR ‘sail’ we can be sure that PAn speakers had boats with sails. However, we cannot be certain that they possessed the outrigger, since *saRman ‘outrigger float’ and other terms connected with the outrigger canoe complex are reconstructible only to PMP. As noted in Blust (1999) the Austronesian settlement of Taiwan may have been accomplished with bamboo sailing rafts, leaving open the possibility that the *qabaŋ was a simple dugout canoe used on interior rivers.

*qabaŋ₂ float

Note:   Also Toba Batak habaŋ-habaŋ ‘fly back and forth, of a flag’.

*qabaRa₁ shoulder

Note:   Also Gaddang afa ‘shoulder’, Tboli abal ‘shoulder’, Sa'a hala ‘shoulder’, Sa'a apala ‘shoulder’. In addition to meaning ‘shoulder’, unaffixed reflexes of PMP *qabaRa mean ‘to carry on the shoulder’ in various CEMP languages, and in Chamorro. Although one would normally expect the body part term to be semantically primary and the verb derived, PAn *qabaRa-an (‘place of shoulder-carrying’) suggests that *qabaRa originally meant ‘carry on the shoulder’. Some WMP and OC languages irregularly reflect this form without the initial syllable (Maranao waga, Tiruray wara, Mono-Alu fala, Nggela vala, etc.). This may indicate that *qabaRa contains a prefix *qa- seen also in *qa-lima ‘hand’ and perhaps *qa-qaqi ‘foot’ (although tentatively I have reconstructed the latter as *qaqay). PMP *baRa ‘hand, arm’ seems clearly to be distinct, as it is reflected with the meaning ‘hand’ in the great majority of witnesses.

*qabaRa₂ topmost hand of banana stalk

Note:   Probably a metaphorical extension of *qabaRa₁ (the ‘shoulders’ of the banana stalk). Although Mansaka abága ‘top hand on a stalk of bananas’ and abagá ‘shoulder’ carry different accents, this divergence may have arisen from historical changes which were intended to more clearly distinguish between forms no longer regarded as related. For Molima Chowning (n.d. b.) gives a single lexical entry ʔavala-na, avala-na ‘shoulder; topmost hands of banana’.

*qabaS notch cut in a tree

Note:   Possibly a product of convergent innovations.

*qabated sago grub

Note:   Also Tiruray lebuted ‘the larvae of the coconut black beetle, Oryctes rhincerus Linn.’, Samoan ʔafato ‘grub, larva (some are edible)’, Rennellese ahato ‘Longicorn beetle, Olethrius tyrranus Thompson, Wolff, highly desired as food’. Entirely apart from the evidence for disjuncts *qabated and *qabateR, there are several noteworthy irregularities in this comparison. First, although Tongan appears to confirm the initial *q-, it has first-syllable /o/ rather than the anticipated **a/. Second, Samoan has an initial glottal stop for expected zero, finally and Rennellese has initial zero for expected glottal stop.

*qabateR sago grub

Note:   Also Ilokano bátar ‘a kind of white larva’, b<in>átar ‘infested by the bátar larva’, Bontok bátel ‘the grub of a kind of wood-boring beetle’, Ifugaw bat-batól ‘great green caterpillar covered with hair’, Ifugaw (Batad) bātol ‘a large larva, of a long-horned beetle … or a honey bee’, Simalur batəl ‘white woodworm’, Mentawai batara ‘beetle larva’, Sasak batah ‘worm (in maize, trees, rice)’, batah pandan ‘larva of the pandanus beetle’, Mongondow obatok ‘large larva in e.g. a sago palm or a coconut palm’, Rembong wate ‘wood grub’, and Nggela vato ‘a species of grub, eaten as food on Mala’.

For Paulohi of the central Moluccas Stresemann (1918) gives hatele ina-i (‘mother of the sago grub’) as the name of the adult growth stage. In a similar vein, Elbert (1975) notes that among the Polynesian-speaking Rennellese of the Southeastern Solomons "Informants are uncertain as to the "mother" of the ahato, but possibly the manumogi, which deposits eggs in certain trees." This suggests that at least by PCEMP times the mature growth stage was represented lexically by a genitive construction *ina ni qabateR.

*qabatiR sago grub

Note:   Also Ilokano bátir ‘a kind of larva, white worm’, batir-en ‘plant disease caused by the bátir’, Nauna kahek ‘sago grub’.

*qa(m)bej wind around

Note:   With root *-bej ‘wind around repeatedly’.

*qabi take hold of, grasp

*qabin hold or carry under the arm

Note:   Also Maranao aben ‘hold in the lap’, Kenyah (Long Wat) bin ‘carry on the back’, Melanau (Mukah) bin ‘carry on the back’. Toba Batak abiŋ ‘carry in the lap; carry off a woman by force’, abiŋ-an ‘lap, place where one carries a child’, Nggela avih-i ‘carry a child under the arm’. Although the gloss given to this form is justified by very similar meanings in both WMP and OC languages, an essential part of the meaning in PWMP appears to have been a condition that the contact between thing carried and person carrying was on or near the lower abdomen.

*qabiq finished, used up; all, every one

Note:   Also Javanese kabéh ‘(in) all, every one’.

*qabis finished, used up; all, every one

Note:   Also Sangir kebiʔ ‘all, every’.

*qabqab to eat or drink voraciously (derogatory)

Note:   Also Aklanon hábhab ‘to eat sloppily and hurriedly’, Cebuano habháb ‘for pigs to eat; for people to eat (derogatory), Manobo (Western Bukidnon) habhab ‘of a person, to eat as an animal eats without the use of hands, used only in a derogatory or joking sense; of a pig, to eat in its characteristic manner’, Tausug habhab ‘to eat or drink voraciously’.

*qabu ash, cinder, powder

Note:   Also Nias awu ‘ash; hearth, kitchen’, Balinese awu ‘ashes, dust’, Balinese awon ‘ashes, ash’, Sasak abu ‘ash; ash-gray (specifically of carrier-pigeons)’, Sasak awu ‘ash from firewood in the kitchen’, Leti apu ‘ash’, Manam apoaŋ ‘fireplace, ashes’. *qabu is among the most widespread and stable morphemes in the Austronesian family. It referred to ashes, and prototypically to the ashes in the fireplace (hence its recurrent replacement of PMP *dapuR in the meaning ‘hearth’). Dyen (1953:29) erred doubly in positing *abuh, the erroneous initial following from his acceptance of Dempwolff's use of Tongan efu ‘ash’ (actually a reflex of *dapuR), and the erroneous final from his overconfidence in the evidentiary value of thematic -/h/ in the Tagalic languages.

Apart from the meanings given here it is likely that some form of *qabu was used to designate a type of plumage for domestic fowls in PPh (cf. Ilokano ábo ‘cock with gray plumage and reddish tail’, Cebuano manuk ŋa abúhun ‘gray-colored cock’), and perhaps in PWMP (cf. Sasak abu ‘ash-gray, specifically of carrier pigeons’). Finally, both the reduplication of *qabu in various daughter languages, and forms assigned to *paR-qabu-an here may be products of convergence. The same explanation probably accounts for the agreement of Kenyah (Long Wat) umaʔ avew ‘kitchen’ with Kambera uma aü ‘cooking place, place of the hearth fire’.

*qabu qabu fish sp.

*qabud strew, scatter, sprinkle

Note:   Sundanese awur ‘strewing; to strew, scatter’ is assumed to be a borrowing from Javanese.

*qabug dust

*qabuk dust, sawdust

Note:   Also Bintulu abuk ‘sawdust’, Tongan afu-afu ‘fine drizzling rain’. With root *-buk₁ ‘decay, crumble; powder’.

Although reflexes of *qabu ‘ash’ and *qabuk ‘dust’ clearly are distinct in languages which preserve final consonants, the inclusion of Oceanic reflexes under *qabuk rather than *qabu is based entirely on semantic fit. Whereas *qabu evidently referred to ashes, and prototypically to the ashen residue in the hearth, *qabuk evidently designated a range of fine particulate matter produced without combustion (despite occasional semantic cross-over, as in the Sasak and Banggai reflexes). Although only the meanings ‘dust’ and ‘sawdust’ can be justified by the glosses of cognates in primary branches of MP, the meaning ‘powder’ can be added for PWMP. Consideration of the root *-buk₁ suggests further that PMP *qabuk probably referred additionally to the powder produced by weevil-infested, or decaying wood. Finally, Richardson (1885) cites Malagasy maŋ-àvoka under hávoka, but with a question mark indicating uncertainty regarding the shape of the unaffixed stem.

*qábut to reach, overtake, catch up with

Note:   Also Hanunóo áput ‘reach, reaching, as with the hand; arriving’, Hawu abo, abu ‘meet, come across, encounter’. Central Sama um-abut ‘extend, stretch, be enough’, ta-abut ‘be reached, overtaken’ are assumed to be borrowings from a Central Philippine language.

*qaCay liver

Note:   Also Isneg agtáy ‘liver’, Casiguran Dumagat agtë ‘liver’, Ifugaw alte ‘liver’, Bontok agtéy ‘liver’, Pangasinan altéy ‘liver’, Sambal (Botolan) ágtay ‘liver’, Bikol katóy ‘liver’, Iban atau ‘liver, heart’, Makassarese ati ‘inner self, courage, heart’; ati-ati ‘heart-shaped; do one's best’, Numfor kè-n ‘liver’, Banoni date ‘liver’, Gilbertese ato ‘liver; clot of blood, clotted blood’. For reasons that remain unclear, most members of the Cordilleran and Central Luzon subgroups of Philippine languages (Blust 1991) reflect *qaRtay; although *-R- in these reflexes appears to be infixed, it has no identifiable meaning. Similar phonological problems are found with some other forms (e.g. PPh *qaRta; reflexes of *qata ‘outsiders, alien people’), but in most cases a doublet can be reconstructed.

A number of other interesting problems are connected with this form, including the following:

1)Dempwolff reconstructed *qatay ‘liver’, but it is clear that this term had a cultural importance greatly transcending simple reference to the physical organ. Many attested glosses refer to the liver as the inner self, the seat of the emotions, will, and desire and mind, in short as the cultural equivalent of the metaphorical ‘heart’ in Indo-European languages. The absence of competing reconstructions for such essential verbal notions as ‘to think’, ‘to feel’, and ‘to desire’ is also consistent with a hypothesis that *qaCay whether alone, affixed, or in combination with other free morphemes, represented these meanings;

2) nominal compounds or descriptive phrases which make use of the morpheme for ‘liver’ as the metaphorical ‘heart’ sometimes show a similar semantic structure cross-linguistically (‘big liver’, ‘burning liver’, ‘white liver’, etc.); examples of such usages are cited here if a fairly clear-cut meaning can be reconstructed, even when the complete phonological form is unattainable directly. Undoubtedly, many more expressions relating to the emotions will be reconstructed with *qaCay when more complete descriptive sources become available. No such morphologically complex expressions have been recorded for any Formosan language, but this may simply reflect the rudimentary stage of lexicography in the Formosan field. On the other hand, although the Philippines is lexicographically well-represented, very few compounds with a reflex of *qaCay have been recorded, and, surprisingly, none of these relate to the emotions;

3) PMP *qatay-an is only weakly attested in the data, but is proposed to unite several disparate observations under a single explanation. Next to the meaning ‘liver’ many glosses include the meanings ‘inside, inner part’ and this semantic component of PMP *qatay no doubt lies behind the diachronic change of *dalem ‘in, inside’ to e.g. Ilokano dálem Elat raro Watubela lamno, Alune lale ‘liver’ and Buruese lale-n ‘liver; inside, heart, mind’; lale-t ‘heart, attitude, will’. Similarly, expressions such as Malay dalam hati ‘in one's heart; to oneself, secretly or silently’ (borrowed as Tagalog dalam-hatíʔ ‘extreme sorrow’) appear to be restructurings of *qatay-an with replacement of the suffixed ‘locative focus’ marker by a locative preposition. The expected Mussau form is **ateana, but this form apparently was reanalyzed as /atea/ plus the 3sg. possessive pronoun -/na/, giving rise to a new free form: /atea-gi/ ‘my liver’, /atea-m/ ‘your (sg.) liver’, /atea-na/ ‘his/her/its liver’;

4) even in its purely physical sense there is more to be said about *qaCay than simple reference to the liver. The glosses cited here for Malay, Rembong, Rotinese, Wetan and Kwaio all suggest that the liver was conceived as part of a larger anatomical unit. In all glosses except that of Rembong this larger anatomical unit includes the heart, while in all glosses except that of Malay it includes the lungs. In field elicitation ‘liver’ and ‘heart’ are often confused by Austronesian-speaking informants, probably because in common sacrificial animals such as chickens they are in closer physical proximity to one another than is the case in the human body.

The association of liver and lungs is more subtle. Fox, J. (1974) notes that in Rotinese formal dyadic language (also known as ritual parallelism) many lexical items are paired with at least one ‘linking term’ to form a convention association for purposes of achieving rhetorical effect. Among the entries in his dictionary for which a linking term is given Fox lists ate ‘liver’. The link is ba ‘lungs’, and for the latter also a single linking term is given: ate ‘lungs’. Strikingly, a kind of converse association appears in Simalur of the Barrier Islands west of Sumatra, where PAn *baRaq ‘lungs’ is reflected as bala ‘lung, also: liver’ (Kähler 1961). Whether these agreements indicate that a dyadic link connected *qatay and *baRaq in some type of formal ritual language in PMP, or simply that the liver and lungs were conceived as part of a single larger anatomical unit is unclear, although the apparent absence of comparative evidence for other dyadic sets favors the latter interpretation. Perhaps to be connected with this interpretation are the occasional reflexes of *qaCay meaning ‘intestines’: Sangir tatuno ate ‘intestines, used in maledictory expressions’, Proto-Ambon *(h)ata(y), Lakalai la hate viriviri ‘intestines’.

*qaCeb deadfall trap to catch small mammals

Note:   Also Itbayaten kateb ‘coconut-crab trap’, Sangir atebeʔ ‘bamboo trap for crabs’.

*qaCi ebb, of water in streams

Note:   Also Kapampangan kati(h) ‘low water-level’, Mansaka atiʔ ‘dry up (of water); evaporate’; Lou ramet ‘reef; dry reef’, Mota meat ‘ebb, low tide’.

*qaCi- prefix variant of the *qali/kali- set

Note:   For an explanation of the analysis and function of this affix, which is almost invariably fossilized in the words in which it appears, cf. Blust (2001).

*qaCipa soft-shelled turtle