Chapter 4.7 Insects and other creepy-crawlies

Meredith Osmond

1. Introduction⇫

The Oceanic Lexicon Project is concerned both with reconstruction of POc terms and their meanings and with compiling a picture of the kind of culture that can be associated with these reconstructed concepts.¹ My intention here is not just to identify the insects that existed in the world of Proto Oceanic speakers but also to consider their role in it. Were they dangerous or merely pesky, harmless, insignificant or perhaps useful? Were they associated with the supernatural? The perceived role of an insect depends on the way in which it impinges on human lives, whether by biting or stinging, by being highly visible or noisy, by attacking woodwork or food plants, by being a valued food, and so on. Occasionally we find linguistic clues that throw additional light.

It should be noted that the term ‘insects’ is used here in its popular extended sense to include spiders, centipedes, and other creepy-crawlies such as grubs, worms and leeches.

The difficulty of the task is increased by the sheer size of the Oceanic world, involving as it does many different environments. Compared with other animals, insects are highly adaptable (capable of living in a wide range of environments), easily transportable (by man, animal or wind), and numerous in both kind and number. Consequently, it is difficult to plot the distribution of even a major order, such as scorpions or leeches. Ideally it would be useful to know whether a given creature exists in a region (in one or more readily identifiable species), and if so, whether it has been there at least as long as the first settlers. Even this information is of little use, however, unless we also have local names. Specialised entomological tomes, dealing with, say, the Hymenoptera of Vanuatu, rarely have this information.

For the time being, then, we make do with the most readily available information, culled from general wordlists fleshed out by ethnographic descriptions. Many apparent cognates show unexpected variation in gloss, so that the term for a mosquito in language A becomes that for sandfly in language B, wasp terms move to bee, centipede to millipede, grasshopper to cricket to cicada, and so on. I have to decide if this is due to change in distribution of the creature, or perhaps reflects the insect’s lack of importance to a community, or may be a deficiency in the wordlist.

Another factor to be considered is a seemingly high degree of phonological irregularity in some insect names. This manifests itself at times in collections of terms which share strong resemblance but for which no reconstruction is possible. At other times it results in doublets, where two or more very similar reconstructions can be made for the same insect. Terms for noisy insects are likely to be influenced by onomatopoeia, while terms for insects that jump or sting or tickle may be derived from relevant verbs. This may help explain why I can have several reconstructions for an insect that are based on quite limited cognate sets, but no one dominant form. In contrast, terms such as POc *kutu ‘louse’ and POc *laŋo ‘fly’ are particularly stable, their reflexes occurring in a very high proportion of languages throughout the Oceanic region.

One further concern of this chapter is the role of a particular prefix in insect names as a possible supernatural marker. In 1983 Robert Blust presented a paper at the Third Eastern Conference on Austronesian Linguistics entitled A linguistic key to the early Austronesian spirit world². It dealt with the frequent occurrence of an affix, often fossilised, in Austronesian languages, traceable back to PAn. Its typical form was qali- or kali-, although there were a number of variants. Blust listed 62 lexical sets containing over 570 examples from about 50 languages distributed geographically from Taiwan to Polynesia, with the vast majority occurring in Western Malayo-Polynesian languages. When these 62 lexical sets were sorted into semantic categories, 20 were for creepy-crawly life forms, the largest single category. Other categories were bats/birds (six sets including doves, owls), striking natural phenomena (six sets, including rainbows, echoes, whirlpools), body parts (five sets, including pupil of eye, hair whorl), and muddled psychological states (four sets, including being dizzy, talking/walking in one’s sleep). The puzzle for Blust was to assign a single conceptual principle that linked these apparently disparate meanings. If qali-/kali- was a morpheme, what did it mean?

He realised that in a substantial number of cases the referents had a connection — most commonly a dangerous connection — with the world of spirits. Insects in particular were associated with spirits of the dead. The qali-/kali- terms listed by Blust (updated in 2001b: 37) refer to the following creepy-crawlies: ant/termite, honey bee, bumble bee, beetle, butterfly, caterpillar, centipede, cockroach, crab, cricket, dragonfly, earthworm, firefly, flea, gecko, grasshopper, jungle leech, paddy leech, luminous millipede, scorpion, snake, spider, wasp. Other phenomena he lists (rainbows, whirlpools, sleep-walking etc.) are thought of as supernatural events/conditions. In other words, the qali-/kali- terms indicated a taboo, a warning that certain kinds of behaviour were to be avoided in the presence of these creatures or events. It would be an advantage for such taboos to be linguistically clearly marked, to be learned early by children.

Blust strengthens his argument by considering the kinds of insects which are not marked by qali-/kali-. He sorts them into creatures that (i) tend to invade human space (maggot, horsefly, housefly, both types of lice, mosquito, nit), or (ii) are economically important either because they are edible (sago grub), or because they cause damage to human crops or constructions (termite).³ In contrast, the marked creatures have little or no economic importance. In addition, some of these, particularly fireflies and luminous millipedes, have unusual properties which might be regarded as supernatural (pp 37-38).

Blust concludes his argument: ‘In short, then, the function of the qali-/kali- prefix evidently was to mark facets of experience that were regarded as spiritually dangerous, hence requiring special precautions of a sort likely to be violated by incompletely accultured children. It did this purely by lengthening the affixed word to an atypical quadrisyllable shape, hence marking the associated semantic categories as those requiring particular behavioural sensitivity’ (p.59).

One of my purposes in this study, then, is to see to what extent qali-denoted insects are regarded as creatures with supernatural associations in Oceanic languages. At first glance, it would seem that the Oceanic lexical forms no longer hold the same WARNING TAKE CARE message. Nonetheless, the affix, or traces of it, occurs in a large number of Oceanic languages in a range of insect terms, and I have been able to reconstruct it to POc level in a small number of items.

2. Lice (Order Anoplura: Family Pediculidae)⇫

One of the largest cognate sets in the entire lexicon of POc reconstructions is that reflecting *kutu ‘louse’. Its reflexes serve in places as a generic term for lice and fleas, often in compound form, when they refer to other creatures parasitic on plants or animals such as ‘pig louse’ or ‘dog flea’. The very size of the set (I have over 100 reflexes) and their consistency of referent, must reflect its relative salience for witnesses in contemporary languages. The set below is simply a representative sample. Also reconstructed is POc *tuma ‘body louse’. In Nakanai, Roviana and Gela, languages where reflexes of both *kutu and *tuma survive, reflexes of *tuma now refer to ‘flea, house louse, bed bug’, ‘house bug’ or ‘(mat-eating) moth’ respectively. As well, I have a widely reflected term for the louse egg or nit and a number of terms for the action of searching for headlice. This grooming activity was no doubt a regular feature of social life.

2.1. Head lice⇫

PAn		*kuCu	‘louse’ (Blust 2002)
PMP		*kutu	‘louse’
POc		*kutu	(1) ‘louse (generic)’; (2) ‘head louse’
Adm	Kaniet	uto	‘louse’
Adm	Lou	kut	‘louse’
NNG	Sio	kutu	‘louse’
NNG	Roinji	ɣutu	‘louse’
NNG	Wampur	gur	‘louse’
NNG	Gedaged	ut	‘louse, flea, plant louse, tick’
PT	Are	kutu	‘louse’
PT	Bwaidoga	utu	‘louse’
PT	Molima	ʔutu	‘louse’
PT	Motu	utu	‘louse’
MM	Vitu	ɣutu	‘louse’
MM	Nakanai	utu	‘louse, dog flea’
MM	Roviana	ɣutu	‘louse’ (ɣutu siki ‘dog flea’)
SES	Bugotu	ɣutu	‘louse’
SES	Gela	ɣutu	‘louse’
SES	Lau	ʔū	‘louse, flea’
SES	’Are’are	ū	‘louse, flea’
NCV	Mota	wutu	‘louse’
NCV	Raga	gutu	‘generic for biting lice and sucking lice’ (gutu-boe ‘pig louse’, gutu-n-manu ‘bird louse’)
NCV	Tamambo	hutu	‘louse’
NCV	Nguna	kūtu	‘louse’
SV	Lenakel	kur	‘louse’
SV	Kwamera	ur	‘louse’
NCal	Nemi	cīk	‘louse’
NCal	Nixumwak	ciɣic	‘louse’
NCal	Iaai	uto	‘louse’
Mic	Kiribati	uti	‘louse’
Mic	Kosraean	kut	‘louse’
Fij	Rotuman	ʔufu	‘louse’
Fij	Wayan	kutu	(1) ‘head louse’; (2) ‘generic for lice and fleas’
Fij	Bauan	kutu	‘crab louse, flea’
Pn	Tongan	kutu	‘louse’ (kutu-fisi ‘flea’, kutu-lotuma ‘bedbug’)
Pn	Rennellese	kutu	‘louse, bird louse, mite, leech, tiny insects of various kinds’
Pn	Samoan	ʔutu	‘louse’
Pn	East Futunan	kutu	‘head louse’
Pn	Tikopia	kutu	‘head louse’
Pn	Hawaiian	ʔuku	‘any small insects: louse, flea, mite’

2.2. Body lice⇫

Because clothes did not feature strongly in the lives of Proto Oceanic speakers, who did not make use of woven cloth, I have preferred to gloss the POc reflex of PAn *tumeS ‘clothes louse’ as ‘body louse’. These creatures would no doubt have continued to exist in Proto Oceanic household items such as the woven pandanus matting generally used as sleeping mats (vol.1,80). Although the term has survived in Proto Oceanic, its PAn/PMP meaning has been restored in Niuean and East Futunan reflexes, perhaps fortuituously, through the modern advent of clothing. It seems that in societies where woven clothes were not worn, the distinction between the different kinds of lice was not significant, and either term was used in places as a generic for functionally similar creatures.

PAn		*tumeS	‘clothes louse’ (Blust 2002)
PMP		*tumah	‘clothes louse’
POc		*tuma	‘body louse’
NNG	Kove	tuma	‘louse, flea’
NNG	Numbami	tuma	‘louse’
PT	Gapapaiwa	tuma	‘louse’
PT	Ouma	tuma	‘louse’
PT	Suau	tuma	‘louse’
MM	Nakanai	tuma	‘house louse, bed bug, flea’
MM	Bola	tuma	‘bug’
MM	Nehan	tuma(su)	‘bedbug’
MM	Roviana	tuma(o)	‘house bug’ (cf also tuma-rere ‘stinking black cockroach’, tuma-rititi ‘k.o. dragonfly’)
SES	Gela	tuma	‘(mat-eating) moth’
SES	Fagani	(a)umʷa	‘louse’
Pn	Niuean	tuma	‘clothes louse’
Pn	Tongan	tuma	‘k.o.louse’
Pn	East Futunan	tuma	‘body or clothes louse’

2.3. Nits⇫

In addition to POc *lisaq ‘nit’ I have reconstructed PWOc *lejaq ‘nit’. Both forms are securely based. No language has reflexes of both. Their formal similarity may be due to chance or may reflect an early borrowing.

PAn		*liseqeS	‘nit, louse egg’ (Blust 2002)
POc		*lisaq	‘nit’4
Adm	Mussau	lisa	‘louse’
Adm	Lou	lisa	‘nit’
NNG	Wogeo	lisa	‘nit’
NNG	Kove	lare	(vowel metathesis)
MM	Patpatar	lise	‘nit’
MM	Ramoaaina	lia	‘nit’
MM	Nehan	lih	‘nit’
MM	Halia	lisa	‘nit’
MM	Teop	niha	‘nit’
SES	Gela	liha	‘nit’
SES	Talise	lisa	‘nit’
SES	Kwaio	lita	‘nit’
NCV	Mota	lisa	‘nit’
NCV	Tamambo	lisa	‘nit’
NCV	Nguna	līsa	‘nit’
SV	Anejom̃	na-laθ	‘nit’
SV	Lenakel	(ki)lha	‘nit’
SV	Kwamera	(kʷa)resa	‘nit’
SV	Sye	ne-lis	‘nit’
NCal	Nemi	ñʰida	‘nit’
Mic	Kiribati	rina	(metathesis)
Mic	Ponapean	c̣īl	(metathesis)
Fij	Bauan	lise	‘nit’
Pn	Tongan	liha	‘nit’
Pn	Samoan	lia	‘nit’
Pn	Tikopia	rie	‘nits of head lice’
Pn	Māori	riha	‘nit’
Pn	Hawaiian	lia, liha	‘nit’

PWOc		*lejaŋ	‘nit’
NNG	Tuam	les	‘nit’
NNG	Mangap	leze	‘nit’
NNG	Yabem	lɛsɛŋ	‘nit’
PT	Dobu	neda	‘nit’
PT	Molima	neda	‘nit’
PT	Tawala	neda	‘nit’
PT	Kilivila	lesa	‘nit’
PT	Sudest	leleji	‘nit’
MM	Vitu	leda	‘nit’
MM	Bali	ledaŋa	‘nit’
MM	Meramera	lesa	‘nit; louse’

PWOc		*ka(R,r)oma	‘nit’
NNG	Labu	alɔma(la)	‘nit’
MM	Hoava	karoma	‘nit’

2.4. Search head for lice⇫

One POc term and one PROc term have been reconstructed with specific reference to searching for lice. Other terms with more general meaning, particularly POc *tirop, *tirop-i ‘look intently’, are sometimes used. In Kove (NNG) the expression is ravu tuma ‘feel for lice’ (Ann Chowning, pers. comm.).These terms indicate the likelihood of reciprocal grooming, an activity undertaken in many societies as part of social interaction.

PMP		*tin[d]ap	‘look intently’ (Dempwolff 1938)
POc		*tiro(p), *tirop-i-	‘look intently, look for (lice etc.)’
MM	Roviana	ti-tiro	‘search for’
SES	Lau	irofi	‘look at fixedly, look for’
SES	Kwaio	ilo(i) (falaina)	‘search hair (i.e. for lice)’ (falaina ‘hair’)
SES	’Are’are	iro	‘look for, collect’
SES	Sa’a	irohi	‘clear the head of lice’
SES	Arosi	(ha)irōhi	‘look for lice in the hair’ (ha- ‘verbal prefix’)
Pn	West Futunan	jiro(a)	‘look carefully, search for’
Pn	Māori	tiro	‘look into, examine’

POc		*tapu(s), *tapus-i-	‘seek lice’
PT	Motu	tahu(a)	‘seek, examine’
PT	Gumawana	tao	‘look for lice’
SES	Gela	tavuhi	‘seek, esp. lice in the hair’
SES	Ghari	tavu(a)	‘scratch, look for lice’
SES	Lau	afui	‘search a head for lice’
NCV	Avava	tap	‘pick fruit’ (John Lynch, pers. comm.)
NCV	Naman	tov	‘pick fruit’ (John Lynch, pers. comm.)

PROc		*pakit, *pakit-i-	‘search hair for lice’
NCV	Tamambo	vahi	‘groom hair, search head for lice’
NCV	Nguna	vāke	‘search hair for lice’
Mic	Woleaian	faxiti	‘search hair for lice’
Mic	Mokilese	pakit	‘delouse’

3. Butterflies (Order Lepidoptera = ‘scaly wings’)⇫

A number of formally similar reconstructions for butterfly have been made above the level of POc: PAn *(qali)-beŋbeŋ, PMP *kali-mbembeŋ and PSHWNG *kalə-mbombəŋ. From the first-mentioned one would expect the regularly derived POc *kali-boboŋ. It does appear, but its reflexes are with one exception limited to the North New Guinea subgroup. The most widespread Oceanic cognate set, yielding POc *(kau)bebek, cannot be derived directly from PMP.

PAn		*qali-beŋbeŋ	‘butterfly’ (ACD)
POc		*[kali]bobo(ŋ)	‘butterfly’
Adm	Loniu	popʷ(ilow)	‘butterfly’
NNG	Gedaged	kilibob	‘butterflies, a collective term’
NNG	Numbami	kaiᵐboᵐbo	‘butterfly’
NNG	Bariai	vovo	‘butterfly’
NNG	Kove	vovo	‘butterfly’
NNG	Maleu	na-lvovo	‘butterfly’
NNG	Tuam	bobo(im)	‘butterfly’
NNG	Gitua	bobo(koro)	‘butterfly’
NNG	Malalamai	vovo	‘butterfly’
NNG	Sio	bobo	‘butterfly’
NNG	Bing	kalbob	‘butterfly’
NNG	Bilibil	kilibob	‘butterfly’
NNG	Megiar	kabob	‘butterfly’
NNG	Manam	bo-bobe	‘butterfly’
NNG	Bam	ba-bob	‘butterfly’
NNG	Wogeo	bobo	‘butterfly’

POc *[kau]bebek has reflexes throughout almost the entire Oceanic region apart from NNG. The added -a in Molima, Dobu, Sewa Bay, Duau, Bwaidoga and ’Auhelawa reflects a regular process, but the labialisation of *k is unexpected.

POc		*[kau]bebek	‘butterfly, moth’
Adm	Lou	pepe	‘centipede’
Adm	Mussau	kau-bebe	‘butterfly’
PT	Molima	pepeʔʷa	‘butterfly’
PT	Dobu	pepekʷa	‘butterfly’
PT	Sewa Bay	pepekʷa	‘butterfly’
PT	Duau	pepekʷa	‘butterfly’
PT	Bwaidoga	bebewa	‘butterfly’
PT	Gapapaiwa	beba-beba	‘butterfly’
PT	Ubir	fefek	‘butterfly’
PT	Hula	pepe	‘butterfly’
PT	Sinaugoro	kau-bebe	‘butterfly’
PT	Mekeo	fefe	‘butterfly’
PT	Motu	kau-bebe	‘butterfly, moth’
MM	Bali	ka-bebeke	‘butterfly’
MM	Nakanai	bebe	‘butterfly’
MM	Tolai	bebe	‘butterfly’
MM	Mono-Alu	bebe	‘butterfly’
SES	Lau	bebe	‘butterfly, moth’
SES	Sa’a	pepe	‘butterfly, moth’
SES	Arosi	bebe	‘butterfly (generic)’ (used as first element in compound terms for partic. varieties)
NCV	Mota	pepe	‘a yellow butterfly’
NCV	Raga	bebe	‘generic for butterflies and moths’
NCV	Lewo	(le)pepe	‘butterfly’
NCV	South Efate	(li)pep	‘butterfly’
Mic	Kiribati	pʷepʷe	‘butterfly’
Mic	Marshallese	pʷapʷipʷ	‘butterfly, moth’
Fij	Rotuman	pepe	‘moth or butterfly of any kind’
Fij	Wayan	bēbē	‘butterflies and moths’
Fij	Bauan	bēbē	‘butterfly’
Pn	Tongan	pepe	‘butterfly’
Pn	Pukapukan	pepe	‘dragonfly’
Pn	Tikopia	pepe	‘butterfly, moth; gen. term for Lepidoptera’
Pn	Māori	pepe	‘flutter, moth’
Pn	Māori	pepepe	‘butterfly’

cf. also:

Mic	Puluwatese	(li)pʷekipʷek	‘butterfly’ (li- ‘nominal prefix’)
Mic	Woleaian	(ri)ɸexiɸex	‘butterfly, caterpillar’ (ri- ‘nominal prefix’)

One other reconstruction is restricted to Papuan Tip.

PPT		*qara-bembem	‘butterfly’
PT	Ubir	kara-bimbim	‘butterfly’
PT	Doga	ara-bembem	‘butterfly’
PT	Anuki	kara-bemem	‘butterfly’
PT	Are	ara-bembemta	‘butterfly’

In Western Oceanic, but with just two reflexes, I find —

PWOc		*bebelo	‘butterfly’
PT	Lala	ebebelo	‘butterfly’
MM	Torau	bebelo	‘butterfly’

— and in that part of MM which lies within the Northwest Solomons:

Proto Northwest Solomonic		*pepele	‘butterfly’
MM	Ririo	pepel	‘butterfly’
MM	Sisiqa	pe-pepele	‘butterfly’
MM	Babatana	pe-pepele	‘butterfly’
MM	Nduke	pepele	‘general term for butterflies and moths’
MM	Roviana	pepele	‘butterfly’
MM	Hoava	pepele	‘butterfly’
MM	Vangunu	pepele	‘butterfly’

cf. also:

TM

Buma

mebeli

‘butterfly’

What are we to make of these variations? Butterflies, like lice and flies, are commonplace, but whereas I have extremely stable cognate sets for the latter two, butterfly terms in Western Oceanic have many slightly different forms. It is as if there is wordplay, with small changes being deliberately made to a word. Was there something culturally significant about butterflies, perhaps some trace of the supernatural that the kali- prefix reflects, that gave rise to some taboo about the use of the normal term? Although I have examples of an association between butterflies and the supernatural in Sa’a (Ivens 1927:187), Maori (Andrew Crowe pers. comm.) and Easter Island (Steven Roger Fischer pers. comm.) I have no comparable examples from Western Oceanic. Malcolm Ross (pers. comm.) suggests an explanation involving a different kind of cultural taboo.

There is a widespread origin legend along the New Guinea north coast based on the activities of two brothers, Manub (blue dove) and Kilibob (butterfly), who, between them, ‘made their dwelling places, sun, moon and stars etc. and also the people, and gave them all their customs and usages’ (Pech 1991:81). It happens that such legends may be seen to belong to particular clans in their traditional form. Others may continue the legends but lack the right to use traditional names (Pech 1991:116). If this were the case, it would be the names, rather than the creature, which were subject to some kind of taboo. It may have been some such reason that is the explanation for the numerous variations in the WOc butterfly term. The EOc terms, however, are quite consistent, all derived from one of the WOc variations, *bebek, so evidently no longer subject to the same taboo pressures.

Although POc speakers evidently included moths and butterflies within one generic term, I have a lower-level reconstruction which is limited to moths, possibly a particular kind of moth. Samoan and Tikopia reflect *lele-fua while the Central Eastern Polynesian languages appear to have added a prefix of unclear function.

PNPn		*[pu]lele-fua	‘k.o. moth’ (pollex: *lele ‘fly swiftly’)
Pn	Samoan	lele-fua	‘moth’
Pn	Tikopia	rere-fue	‘large moth’
Pn	Rarotongan	purere-ʔua	‘large moth sp.’
Pn	Tahitian	pūre-hua	‘moth’
Pn	Māori	pūrere-hua	‘moth’
Pn	Hawaiian	pulele-hua	‘butterfly, moth’

4. Mosquitoes (Order Diptera = ‘two wings’: Family Culicidae)⇫

A POc reconstruction for ‘mosquito’, *ñamuk, is based on numerous cognates from all major subgroups, with little variation in meaning, together with a PMP antecedent. In some Papuan Tip languages reflexes form compounds to refer to a range of small flying biting insects. The PT reflexes have undergone a common sporadic vowel change, lowering of unstressed -u to -o.

A number of terms that refer to the mosquito have been reconstructed in addition to *ñamuk.

PMP		*ñamuk	‘mosquito’ (Blust 2002)
POc		*ñamuk	‘mosquito’
Adm	Aua	namu	‘mosquito’
Adm	Seimat	namu	‘mosquito’
Adm	Kaniet	ñamu	‘mosquito’
Adm	Loniu	ñamɔn	‘mosquito’
NNG	Lukep	nam	‘mosquito’
NNG	Kaiwa	namuk	‘mosquito’
NNG	Manam	nam	‘mosquito’
PT	Molima	namo-namo	‘fly’ (namo-kili ‘mosquito’, namo-kodu-kodu ‘k.o. sandfly’)
PT	Bwaidoga	nimoɣa	‘mosquito’ (namo-kili-kili ‘fruit-fly’)
PT	Maisin	namoɣi	‘mosquito’
PT	Minaveha	namo-namo	‘fly (generic)’ (namo-kina ‘mosquito’, vivia namo-namo ‘wasp, small red variety’)
PT	Gapapaiwa	namo-namo	‘fly’ (namo-kīri ’mosquito)
PT	Are	namo-namo	‘housefly’
PT	Motu	namo	‘mosquito’
PT	Sudest	ñamo-ñamo	‘fruitfly’
MM	Nalik	namu	‘mosquito’
MM	Nakanai	lamo	‘mosquito’
MM	Tolai	namu	‘sandfly’
SES	Bugotu	ñamu	‘mosquito’
SES	Gela	namu	‘mosquito’
SES	Kwaio	namu-namu	‘small flying insect (generic); sandfly, gnat’
SES	Arosi	[na]namu	‘mosquito’
TM	Buma	muko	‘mosquito’
NCV	Mota	namu	‘mosquito’
NCV	Raga	namu	‘generic for mosquitoes’
NCV	Paamese	a-namu	‘mosquito’
NCV	Nese	namɣo	(regular metathesis of final *-Vk)
SV	Sye	yomoɣ	‘mosquito’
SV	Anejom̃	n-yamʷ	‘mosquito’
NCal	Nixumwak	nabuc	‘mosquito’
NCal	Nemi	naguk	‘mosquito’
Mic	Marshallese	namʷ	‘mosquito’
Mic	Woleaian	ramʷu	‘mosquito’
Mic	Ulithian	lamʷo	‘mosquito’
Fij	Bauan	namu	‘mosquito’
Fij	Wayan	am	‘mosquito’
Pn	Tongan	namu	‘mosquito’
Pn	Tahitian	namu	‘mosquito’
Pn	Māori	namu	‘sandfly’

The following reconstruction shares the same second syllable as *ñamuk ‘mosquito’. Both *ñamuk and *simuk evidently referred to ‘mosquito’ in WOc, although *simuk may have had a broader semantic range than *ñamuk, its reflexes at times referring to other small biting flies. Southeast Solomonic terms reflect PSES *simi rather than †*simu. (Maringe si-simi is evidently borrowed from a neighbouring SES language). Reflexes of *simuk are not found in subgroups east of the Solomons.

POc		*simuk	‘mosquito, small biting fly’
NNG	Tuam	sum	‘mosquito’
NNG	Mato	simak	‘sandfly’
NNG	Labu	sumu(si)	‘mosquito’
PT	Wedau	imo(kini)	‘mosquito’ (kini ‘to sting’)
PT	Tawala	himo(kini)	‘mosquito’
PT	Tawala	himo-himo(kini)	‘sandfly’
PT	Dawawa	simo(kin)	‘mosquito’
MM	Mono-Alu	simuʔu	‘midge’
MM	Varisi	simu-simu	‘midge’
MM	Avasö	simuku	‘mosquito’
MM	Maringe	si-simi	‘housefly’
SES	Lau	simi	‘sandfly’
SES	Baegu	si-simi	‘midge’
SES	Longgu	simi	‘mosquito’
SES	Kwai	simi(sakʷalo)	‘mosquito’
SES	Kwaio	simi	‘fly, sandfly’
SES	Kwaio	simi(lakʷalo)	‘mosquito’
SES	Dori’o	simi(lakʷalo)	‘midge’
SES	Dori’o	simi(ni-ōne)	‘mosquito; sandfly’
SES	’Are’are	sime	‘mosquito’
SES	Sa’a	sime	‘mosquito’

Another term, POc *ma-kini(t), is also reconstructable, with reflexes thar refer to ‘mosquito’ in Western Oceanic, but in Remote Oceanic rather to the state of being stung. As a literal translation of a form of the verb *kini-t, ‘to pinch’, *ma-kini(t) means ‘to get pinched’, i.e. ‘get stung’. The languages in which reflexes of *ma-kini(t) are listed have all lost POc final consonants, and putative final *-t is reconstructed on the basis of etymology. In languages such as (PT) Gumawana, the meaning of the verb gini includes ‘to puncture, spear, inject, sting’. Reflexes of *kini-t appear frequently in PT languages as the second element in compound terms referring to biting insects. (See also POc *simuk above.)

POc		*ma-kini(t)	‘mosquito’
NNG	Bariai	makin-kin	‘mosquito’
NNG	Kove	makini-kini	‘mosquito’
MM	Nakanai	makili-kili	‘black gnats, sandflies’
MM	Bali	makini-kini	‘mosquito’
MM	Bulu	makini-kini	‘mosquito’
PROc		*makini	‘to be stung’
NCV	Nguna	makini-kini	‘itchy’
Pn	Samoan	maʔini	‘to sting, to smart’
Pn	Tongan	makini	‘to have a pricking or tingling sensation like pins and needles’ (also makini-kini implying duration)

cf. also:

Pn

Tikopia

kini-kini

‘insect, small brown, predatory on man, in houses (? flea)’

A tendency to identify small buzzing and biting insects with nasal-initial terms and frequent reduplication is noted with *ñamuk as well as POc *ŋiŋi(ŋ) and PWOc *ŋati-ŋati below, and is apparent also in the terms for sandfly/midge/gnat in the following section (POc *niku-niku, *nonok, *ŋi(s,j)i). Most are onomatopaeic, using n, ñ and ŋ almost interchangeably, making it difficult to trace cognate forms.

PMP		*ŋiŋ	‘buzz, hum’ (ACD)
POc		*ŋiŋi(ŋ)	(1) ‘buzz as a mosquito’; (2) ‘mosquito’ (onomatopoeic)
NNG	Bing	ŋiŋ	‘mosquito’
SES	Gela	ŋiŋi	‘buzz as a mosquito’
SES	Talise	ŋi	‘midge’
SES	Birao	ŋi	‘midge’
SES	Tolo	ŋi	‘mosquito’
Fij	Bauan	ŋīŋī	‘buzz as a mosquito’

PWOc		*ŋati-ŋati	‘mosquito’
NNG	Mapos Buang	ŋat-ŋat	‘mosquito’
PT	Gapapaiwa	nasi-nasi	‘small brown fly’
MM	Tabar	ŋati-ŋati	‘mosquito’
MM	Lihir	ŋet-ŋet	‘mosquito’
MM	Madak	ŋit	‘mosquito’
MM	Ramoaaina	ŋat-ŋat	‘sandfly’
MM	Patpatar	ŋati-ŋat	‘mosquito’
MM	Minigir	ŋati-ŋati	‘mosquito’
MM	Tolai	ŋati-ŋat	‘mosquito’
MM	Nehan	nat-nat	‘mosquito’

5. Sandflies, midges, gnats (Order Diptera: Families Chironomidae (non-biting), Ceratopogonidae, Psychodidae, Simuliidae (biting); Families Drosophilidae, Tephritidae (fruitflies))⇫

Cognates in the following sets refer to a range of very small flying biting insects. Vagueness of identity may be due to the fact that speakers have no real need to distinguish these insects more precisely.

POc *niku-niku from PMP *nik-nik rather than the expected †*ninik may simply be another example of the playful variation prevalent in some insect terms.

PMP		*nik-nik, *ñik-ñik	‘tiny biting insect: gnat, sandfly, fruitfly’ (ACD)
POc		*niku-niku	‘small biting fly’
PT	Bwaidoga	niku-niku	‘small fly that bites (like sandfly)’
MM	Vitu	niki	‘mosquito’
MM	Konomala	nuk-nuk	‘mosquito’
MM	Tolai	nuku-nuku	‘very small fly’
MM	Roviana	niku-niku	‘k.o. sandfly whose bite is painful’
MM	Hoava	niku-niku	‘midge’
MM	Vangunu	niku-niku	‘midge’

Polynesian reflexes of *nonok ‘sandfly, midge’ appear to refer to fruitflies and perhaps other small non-biting flies. The Cristobal-Malaitan terms reflect PCM *nono(i)tasi, literally ‘fly of salt water’.

PMP		*nek-nek	‘gnat, sandfly, fruit fly’ (ACD)
POc		*nonok	‘sandfly, midge’
Adm	Lou	(tip)non	‘sandfly,mosquito’
Adm	Baluan	(liplip)non	‘sandfly’
MM	Mono-Alu	nono	‘mosquito’
MM	Nduke	nonoɣo	‘mosquito’
SES	Gela	nonoke	‘ko. sandfly on the shore’
SES	Baegu	nono(āsi)	‘sandfly’
SES	’Are’are	nono(asi)	‘small stinging midge; gnat; sandfly’
SES	Sa’a	nono(asi)	‘midge, gnat’
SES	Arosi	nono	‘fly’
SES	Arosi	nono(iasi)	‘sandfly’
SES	Arosi	(ʔarai)nono	‘mosquito’ (ʔaraʔi ‘to bite, sting’)
SES	Fagani	nana(osi)	‘sandfly’ (vowel metathesis)
NCV	Mota	nono	‘a small beetle that comes on decaying fruit’
NCV	Lewo	ne-nono	‘sandfly, midge, fruitfly’
NCV	Uripiv	nunu	‘midge’
Pn	Niuean	nono	‘small beetle’
Pn	Tongan	nono	‘fruitfly’
Pn	Rennellese	nono	‘k.o. small fly found on rotten bananas’
Pn	Samoan	nono	‘white ant when winged’
Pn	Tikopia	nono	‘fruitfly’
Pn	Tuvalu	nono	‘small flying insect’
Pn	East Uvean	nono	‘small flying insect’

The next reconstruction is reflected mainly in compound terms which include a range of unidentified elements.

POc		*ŋi(s,j)i	‘sandfly’
Adm	Mussau	(kala)ŋisi	‘sandfly’
NNG	Malai	(maraŋ)ŋis-ŋis	‘sandfly’
NNG	Lukep	(bara)ŋis-ŋis	‘sandfly’
NNG	Singorakai	(mala)ŋis	‘sandfly’
NNG	Tami	siŋi-siŋ	‘firefly’ (metathesis)
NNG	Wab	ŋis	‘sandfly’
NNG	Bing	(ramaŋas)ŋis	‘sandfly’
NNG	Mindiri	(bɔrɔ)ŋis	‘sandfly’
NNG	Manam	(mara)ŋizi-ŋizi	‘k.o. gnat, small’
SES	West Guadalcanal	ŋiju	‘sandfly’

PWOc		*ki(r,R)i-ki(r,R)i	‘sandfly’
PT	Gapapaiwa	kiri-kiri	‘sandfly’
PT	Ubir	ire	‘sandfly’
PT	Dobu	(dagʷa)kili-kili	‘sandfly’
MM	Maringe	gri-gri	‘sandfly’

6. Flies (Order Diptera: Families Muscidae (houseflies, bluebottles), Tabanidae (horseflies), Bibionidae (Marchflies))⇫

6.1. Flies (generic); houseflies⇫

Blust (2002) has reconstructed both PMP *lalej ‘housefly’ and PMP *laŋaw ‘botfly, bluebottle’. The latter became in POc the generic term for flies, with reflexes often referring specifically to the common housefly.

Reflexes of POc *laŋo ‘fly’ are both numerous and widespread, occurring in all major subgroups. In a number of languages including Lou (Adm), Gela (SES) and Hawaiian (Pn), the term is used in compounds as first element, referring to a range of flies (housefly, blowfly, horsefly, March fly, bluebottle fly etc.).

PMP		*laŋaw	‘botfly, bluebottle’ (Blust 2002)
POc		*laŋo	‘fly’
Adm	Mussau	laŋo	‘housefly’
Adm	Lou	laŋ-laŋ(a)	‘fly, flying insect’
Adm	Lou	laŋ(et)	‘housefly’
Adm	Lou	laŋ-laŋa-n palawa	‘honey bee’
Adm	Lou	laŋ-laŋ ŋara	‘bluebottle, horsefly, March fly’
NNG	Tami	laŋo-laŋ	‘fly’
NNG	Kove	laŋo-laŋo	‘fly’
NNG	Kove	laŋo-vihi	‘blowfly, big, blue, noisy, bites and stinks’
NNG	Sengseng	laŋ	‘fly’
NNG	Sengseng	laij-i-yuyu	‘hornet with papery nest’
NNG	Sengseng	laŋ-oŋoŋ	‘insect that makes holes in wood’
PT	Motu	lao	‘fly’
PT	Lala	nalo	‘k.o. fly’ (metathesis)
MM	Tabar	raŋo	‘fly’
MM	Vitu	laŋo	‘fly’
MM	Nehan	laŋo	‘fly’
SES	Gela	laŋo	‘fly (generic)’ (first element in many binomial fly terms)
SES	Sa’a	laŋo	‘bluebottle fly’
SES	Arosi	raŋo	‘fly’
NCV	Mota	laŋo	‘bluebottle fly’
NCV	Raga	laŋo	‘generic for flies’
NCV	Paamese	a-laŋo	‘fly’
SV	Sye	(u)laŋ	‘a fly’
SV	Lenakel	(k)iaŋ	‘a fly’
SV	Southwest Tanna	(e)laŋ	‘a fly’
NCal	Nemi	nen	‘fly’
NCal	Jawe	nen	‘fly’
Mic	Kiribati	naŋo	‘fly, blowfly’
Mic	Woleaian	raŋo	‘fly’
Mic	Sonsorolese	rāŋo	‘fly’
Fij	Rotuman	laŋo	‘fly’
Fij	Bauan	laŋo	‘fly’
Pn	Samoan	laŋo	‘fly’
Pn	Tikopia	raŋo	‘fly’
Pn	Māori	raŋo	‘fly’ (eastern dialect)
Pn	Hawaiian	nalo	‘common house fly’ (metathesis; Used as first element in compounds for hornet, wasp, bee, bluebottle etc.)

6.2. Horseflies and March flies⇫

Members of the horsefly and March fly families suck the blood of mammals, inflicting a painful bite. Few terms have been collected, and no reconstructions have been possible.

7. Dragonflies and damselflies (Order Odonata = ‘flies with teeth’)⇫

Dragonflies and damselflies are arguably the most acomplished aerialists in the animal kingdom. Swift and agile, they can reverse direction in midair within one body length, hover with ease and fly backwards. They copulate in the air, typically over a body of fresh water, where the female lays her eggs. Although cognate sets have been elusive, a remarkable consistency of meaning has emerged in descriptive compounds meaning literally ‘copulate’ + ‘water’, from terms in SE New Britain to the Solomons to Fiji. Terms referring to copulation have evidently been replaced by euphemisms in many languages. Lack of cognacy in the terms for water is due to the tendency in some languages to generalise terms originally referring specifically to fresh water, rain, river, pond and so on. For instance, the second element in the Pendau (central Sulawesi) term for a dragonfly, tuntu/rano, translates literally as ‘lake’ (Phil Quick pers. comm.). POc *waiR ‘water’ is attested in the two following (non-cognate) sets by terms from SES, NCV and Fiji.

NNG	Mengen	roro-me	‘dragonfly’ (roro ‘copulate’, me ‘water’)
SES	Gela	hita-hita-beti	‘dragonfly’ (hito ‘to cohabit’, beti ‘water, stream’)
SES	Arosi	ʔani-wai	‘dragonfly’ (ʔani ‘fornicate’, wai ‘water’)
Mic	Chuukese	nifēfēcon	‘dragonfly’ (‘the one that copulates with water’; Davis 1999)
Fij	Wayan	dulu-dulu-wai	‘dragonfly’ (dulu ‘copulate’, wai ‘water’)
Fij	Bauan	ðai-ðai-wai	‘dragonfly’ (ðai ‘copulate’, wai ‘water’)

Other compounds retain the water connection, but vary the verbal concept.

PT	Sudest	wawa-eᵐba	‘dragonfly’ (wawa ?, ᵐbʷa ‘water’)
SES	Gela	gito-beti	‘dragonfly’ (‘steal water’)
SES	Tolo	ici-kolo	‘dragonfly’ (ici ?, kolo ‘water’)
SES	Lau	nalu-kafo	‘dragonfly’ (‘scoop up water’)
SES	Lau	tatara kafo	‘dragonfly’ (‘skim water’)
SES	Longgu	tatara-wai	‘large dragonfly with tail like a helicopter’ (lit ‘skim water’)
SES	Sa’a	tātara-wai	‘dragonfly’ (‘skim water’)
SES	Arosi	tātā-wai	‘dragonfly’ (‘skim water’)
SES	’Are’are	tatara-wa	‘dragonfly’ (‘skim water’)
NCV	Mota	roro-pei	‘dragonfly’ (roro ‘sink’, pei ‘water’)
NCV	Paamese	menmen-oai	‘dragonfly’ (men-men cf. munmun ‘drink’, oai ‘water’)
NCV	Southeast Ambrym	munmun-oi	‘dragonfly’ (‘drink water’)

Samoan retains the skimming element in seʔe-mū ‘dragonfly’ (seʔe ‘glide’, mū ‘flying insect’).

The Rennellese describe it as a sailing insect: manu-manu hogau (hogau ‘sailor, ocean voyager’).

There are isolated instances of dragonflies being used in sorcery, as a means of causing death in Kiribati (Rosemary Grimble 1972:26-27) and to facilitate theft in Tahiti (Henry 1971:391), but rather than reflecting something specific about the role of dragonflies in POc times, this is probably no more than an indication that certain insects were commonly regarded as agents of the gods and spirits.

8. Bees (Order Hymenoptera = ‘membrane wings’: Super Family Apoidea)⇫

The following reconstruction rests on a pair of cognates from different primary subgroups.

POc		*kororo	‘bee’
Adm	Lou	koror	‘bee (generic)’
MM	Teop	kororo	‘honey-bee/honey/(ear wax)’

cf. also:

PT

Sinaugoro

kororo

‘cricket’

Although the following reconstruction bears some resemblance to PMP *(n,ñ)ik-(n,ñ)ik, ‘tiny biting insect: gnat, sandfly, fruitfly’ (§5), it is lacking a final consonant and may simply have been generated independently on onomatopoeic grounds. The initial ŋ- of the Lau and Sa’a cognates is a common variation on n- and ñ- that occurs in terms for buzzing insects.

POc		*(n,ñ)i(n,ñ)i	‘bee, buzzing insect’
Adm	Seimat	nini	‘bee’
NNG	Kela	nini	‘mosquito’
NNG	Sio	ni	‘mosquito’
NNG	Labu	nene(ᵑgʷa)	‘fly (insect)’
NNG	Takia	nini	‘bee, wasp (generic)’
SES	’Are’are	nini(sua)	‘a bee which makes its nest in the ground and in trees’ (sua ‘to burrow’)
SES	Lau	ŋiŋi(dua)	‘bee’
SES	Sa’a	ŋiŋi(due)	‘native bee’

9. Wasps (Order Hymenoptera): Families Ichneumonidae, Vespidae (paper wasp), Sphecidae (mud-dauber wasp, mason wasp)⇫

I have been unable to identify a particular species that can be identified by POc *pupuk. It may be a name given to any wasp-like insect that infests wood.

PMP		*bukbuk	‘weevil that infests wood, bamboo, and rice; dust produced by the boring of this insect; tooth decay, dental caries’ (ACD) 5
POc		*pupuk	‘k.o. borer wasp or bee that infests wood and bamboo’
NNG	Gedaged	fuf	‘wood-borer’
PT	Sudest	vuvu	‘sugarcane borer’
PT	Gapapaiwa	vuvua	‘bee type; carpenter bee’
PT	Molima	wuwuwuva	‘borer wasp’
MM	Patpatar	huh	‘borer insect; wood that has been bored by wood-borers’
MM	Nakanai	vuvu	‘wasp and wasp-like insects’
MM	Tolai	pupuka	‘weevil; dust from weevil-eaten bamboos in house roof’
MM	Nduke	vuvu	‘black insect that eats wood’
MM	Maringe	fufu	‘bamboo borer’
SES	Gela	vuvu	‘a worm that bores into wood’
SES	Bugotu	vūvū	‘mason wasp’
SES	Tolo	vuvu	‘k.o. wasp which bores through wood’
SES	Longgu	vuvu(i)	‘a type of black wasp, digs in the ground, painful if it stings you’
SES	Lau	fufu	‘mason wasp’
Pn	Niuean	(laŋo) fufu	‘mason wasp’
Pn	Tahitian	huhu	‘xylocope or carpenter bee, large Hymenopterous insect, black, attacks wood’
Pn	Māori	huhu	‘larva of beetle sp., edible grub’
Pn	Hawaiian	huhu	‘wood-boring insect, worm-eaten’
Pn	Marquesan	huhu	‘white wood-eating worm’

cf. also:

PT	Minaveha	vuvu(ha)	‘bee, small black variety that bores holes in wood’
MM	Nakanai	vuvu(li)	‘k.o. stinging worm which is painful’ (reflects †*pupuni)
MM	Roviana	vuvu(nu)	‘wood-eating worm’ (reflects pupun_ rather than _pupuk)
NCV	Raga	huhu (gai)	‘generic for wood-boring beetle larvae’ (expect vuvu)
NCV	Raga	huhu (lua)	‘bee or wasp, hairy body, large head, painful sting’

The next reconstruction, POc *bubu ‘k.o. wasp’, may simply be a variant of POc *pupuk.

POc		*bubu	‘k.o. wasp’
NNG	Adzera	(wa)ᵐpup	‘hornet’
NNG	Kove	vuvuvu	‘wasp, kills spiders’
PT	Motu	bubu	‘an insect like a wild bee that destroys timber’
PT	Lala	bubu	‘hole-borer wasp’
MM	Roviana	bubu	‘Buzzer wasp (Odynerus superbum)’
MM	Nduke	bubu	‘black wasp, lives in the ground’
MM	Mono-Alu	(si)bubū	‘wasp’
SES	Sa’a	pū	‘mason bee, wasp’
SES	’Are’are	pū	‘wasp’

cf. also:

NNG	Gitua	bubu(la)	‘housefly, mosquito’
SES	Sa’a	pupu(oro)	‘an insect which digs up the ground and leaves a track’ (oro ‘stoop down’)
SES	Arosi	buburu(mʷago)	‘mason bee, v large variety’ (mʷago ‘earth, soil, brown?’. Also totora mʷago ‘mason bee’)
SES	Arosi	buburu(atoā)	‘k.o. cricket’ (atoā ‘noon’)
Pn	Tikopia	pupu(matau)	‘dragonfly’ (trad, embodiment of spirit; matau ‘fishhook’?)

I am unable to account for the existence of two very similar POc reconstructions for ‘wasp’, *manipo and *(n,ñ)i(n,ñ)ipo, although presumably they are contrasted with *pupuk/*bubu. Nor can they be divided according to their subgroups; reflexes of both appear in the Admiralties and in the Southeast Solomons. Wandamen (SHWNG) anibar ‘wasp/bee’ may be a non-Oceanic cognate. A third group reflects Proto Central Papuan *naniɣo. It seems that here we have yet another example of wordplay. It is difficult to know whether the resemblance between *(n,ñ)i(n,ñ)ipo ‘wasp’ below and *(n,ñ)i(n,ñ)i ‘bee, buzzing insect’ (§8) arose by chance or reflects common ancestry in some way.

POc		*mañipo	‘k.o. wasp’
Adm	Loniu	menih	‘large black bee; k.o. deep water seaweed which stings’ (exp meñipo)
PT	Iduna	manibo	‘blue insect, perhaps k.o. mud wasp’
SES	Bugotu	mañivo	‘wasp’
SES	Tolo	manivo	‘wasp’
SES	Lengo	manivo	‘wasp’
SES	Arosi	ma-maniho	‘hornet’

cf. also:

NNG

Yabem

baniʔ

‘generic for wasps, hornets’

POc		*(n,ñ)i(n,ñ)ipo	‘k.o. wasp’
Adm	Drehet	ninih	‘k.o. wasp’
SES	Longgu	ninivoi	‘k.o. wasp’
SES	Sa’a	niniho	‘hornet’

cf. also:

NNG	Manam	niniko	‘wasp’
	Proto Central Papuan	*naniɣo	‘wasp’
PT	Sinaugoro	naniɣo	‘small black and red wasp’
PT	Motu	naniɣo	‘k.o. manu manu. hornet’
PT	Lala	naniʔo	‘wasp’

In the next set we find the same lowering of an unstressed vowel from /-u/ to /-o/ in Papuan Tip languages that was noted in reflexes of POc *ñamuk (§4).

PWOc		*gumu	‘k.o. wasp’
PT	Dobu	gumo	‘paper wasp’
PT	Saliba	gumo	‘wasp, bee’
PT	Muyuw	gum	‘wasp, hornet’
PT	Molima	gumo	‘wasp’
PT	Tawala	gumo	‘wasp’
PT	Nimoa	gumo	‘black wasp’
MM	Maringe	gu-ɣumhu	‘mud wasp’

10. Ants (Order Hymenoptera: Family Formicidae)⇫

Although sources for a number of languages give a term for ants in general, I cannot reconstruct a POc generic. It is likely that the dominant ant species varies from place to place, and that the term for that species has become the generic term in that area. In Yabem (NNG), for instance, the generic term is lêsêʔ, in Sudest (PT) it is vwara-vwara, in Vitu (MM) duri-duri, and in Kwaio (SES), ta-galo. In Wayan Fijian kadi (from POc *kadik ‘stinging black ant’) has become the generic. In Niue and Tonga, lō (from POc *loRo ‘large red stinging ant’) is the generic form. However, a number of POc reconstructions can be made for ant taxa.

POc		*kadik	‘stinging black ant’
Adm	Nyindrou	karak	‘ant’ (Tryon 1995)
NNG	Kaulong	keh-keh	‘(?) (red) ant’
MM	Bulu	kadi	‘stinging black ant’
MM	Tolai	kədik	‘black ant’ (kərəkum ‘large red ant’)
MM	Nehan	kadik	‘tree ant’
MM	Simbo	kadiki	‘black ant’
SES	Gela	ɣadi	‘k.o. black ant, bulldog ant’
SES	Lengo	ɣadi	‘black ant’
SES	Longgu	hadi	‘black ant that bites’ (possibly a borrowing from Malango)
SES	Lau	kadi	‘bulldog ant’
SES	Kwaio	(kā)kadi	‘stinging black ant’
NCV	Mwotlap	ɣen	‘big stinging black ant’
NCV	Nguna	kādi	‘black biting ant’
NCV	Ambae	gadi	‘fire ant’
NCV	Nese	na-ɣajɣe	‘fire ant’ (regular final *-Vk metathesis)
SV	Ura	(w)asek	‘small stinging black ant’
SV	Lenakel	kasək	‘soldier ant’
SV	Anejom̃	n-ɣas	‘fire ant’
Mic	Woleaian	(u)xeṣi	‘k.o. black ant’
Mic	Carolinian	(wu)xeṣ	‘large black ant with a painful stinging bite, makes its nest near rotting trees’
Mic	Kosraean	kaṣ(kin)	‘k.o. ant’
Mic	Ponapean	kāc̣	‘ant’
Fij	Wayan	kadi	‘generic for ants; large ants’
Fij	Bauan	kadi	‘large black ant, stings’

Rufus Pech refers to the stinging red ant of the north New Guinea coast (lol in Tami) in the following way: ‘This is perhaps the most enterprising and obviously ubiquitous species of ant in the Melanesian lowlands. They build huge nests in trees by drawing and sticking together bunches of leaves, are carnivorous, and defend their territory ferociously.’ (Pech 1991:91). POc *loRo probably referred to this group of ants although witnesses in Eastern Oceanic have a wider range of reference.

PMP		*loRo	‘red tree ant’6
POc		*loRo	‘large stinging red ant’
Adm	Mussau	loa	‘red tree ant’
NNG	Tami	lol	‘stinging red ant’
MM	Vitu	loro	‘large red stinging ant’
MM	Nakanai	lolo	‘large red ant’
MM	Tabar	ror	‘red ants’
MM	Tangga	lo	‘large sugar ants’
SES	Gela	lolo	‘small black ant’
SES	Longgu	lolo	‘a small ant’
SES	Lau	lolo	‘k.o. ant’
SES	Kwaio	lolo	‘ant’
SES	Sa’a	lolo	‘red sugar ant’
Fij	Rotuman	roro	‘fruit fly’
Fij	Bauan	lō	‘small black ant’
Fij	Bauan	lolo	‘small ant’
Pn	Niuean	lō	‘ant (generic)’
Pn	Tongan	lō	‘ant (generic)’: kinds: lō hina (white ant), lō kula (red ant?), lō ʔuli (black), loʔata (see below)’
Pn	Tikopia	ro	‘ant, of various kinds’
Pn	Tahitian	rō	‘ant’
Pn	Māori	rō	‘ant, stick insect, mantis’
Pn	Hawaiian	lō	‘black insect, earwig’

cf. also:

Adm

Lou

lolɛ

‘small black sugar ant’

The next set is possibly a compound form of the above.

PPn		*lō-qata	‘large ant’ (PPn *qata ‘image, likeness’)
Pn	Tongan	lōʔata	‘k.o. black ant w painful bite’
Pn	Samoan	lōata	‘k.o. large black ant’
Pn	Nanumea	loata	‘large ant’

The three reconstructions which follow, POc *drui ‘ant’, POc *[driu]driu ‘small red or brown ants’ and PWOc *didi ‘small ant’ may all be variants of the same etymon.

POc		*drui	‘ant’
MM	Kandas	dui-dui	‘centipede’
MM	Siar	dui	‘centipede’
MM	Simbo	dui	‘ant’ (any ant?)
MM	Nduke	dui	‘small black ant, found in gardens’
MM	Roviana	dui	‘small ant’
SES	Tolo	(koko)dui	‘reddish-brown ant’
SES	Ghari	dui-dui	‘grey middle-sized ant’
SES	Lau	dudui	‘k.o. large yellow ant in trees’
SES	Kwaio	dui-dui	‘vinegar ant’
SES	To’aba’ita	dūdui	‘fairly big reddish ant, stings’
NCV	Nokuku	ʔuʔui	‘ant’
NCV	Nduindui	dui-dui	‘ant’

cf. also:

MM

Vitu

duri-duri

‘generic for ants’

POc		*[driu]driu	‘small red or brown ants’
MM	Sursurunga	diu	‘centipede’
SES	Arosi	diu-diu, di-diu	‘k.o. small brown ant’
NCV	Mota	ni-niu	‘small red ant’
NCV	Ambae	di-diu	‘ant’
NCV	Tangoa	ri-riu	‘red ant’
NCV	Tamambo	diu-diu	‘small sugar ants; general term for red and brown ants, just a nuisance with food’
Fij	Wayan	(kā)driu-driu	‘very small ants’

Given that all terms in the following set except Nduke are binomials, *didi may have been a generic for ‘ant’.

PWOc		*didi	‘small ant’
NNG	Labu	titi(nalo)	‘ant’ (nalo ‘small’)
PT	Dobu	didi(yauyau)	‘ant’ (yau ‘number, to increase in number’)
PT	Sinaugoro	didi(rima)	‘small black ants’
PT	Sinaugoro	riri(ka)	‘sugar ants’
MM	Tolai	(kara)didi	‘larvae of the red ant, used as food’
MM	Nduke	didi	‘wood-borer’

The next set consists of possibly related items, but there are too many irregularities to permit a POc reconstruction.

NNG	Vehes	kakak	‘ant’
NNG	Mapos Buang	kakok	‘ant’
NNG	Kumaru	kakok	‘ant’
NNG	Patep	kekeak	‘ant’
NNG	Zenag	kkewa	‘ant’
MM	Tolai	kakakau	‘small black ant’
SES	Arosi	ʔaʔaki	‘large black ant, bulldog ant, w painful bite’

Although the following terms for ‘ant’ appear to be related (reflecting POc *kal(a,o)) all except Marshallese carry additional unrelated morphemes which cannot be accounted for.

Adm	Loniu	kalɔ(n)	‘ant’
PT	Saliba	kalakala-pʷasipʷasi	‘black ants’
MM	Teop	ano(hi)	‘ant’
SES	’Are’are	(ta)karo	‘small black ant’
NCV	Nguna	(ma)kāla	‘red ant’
Mic	Marshallese	kal(lep)	‘big black ant’ (lep ‘big’)
NNG	Manam	kala(poaki)	‘k.o. red ant’
NCV	Mota	kalo	‘crawl, creep’
SV	Lenakel	(ma)kal	‘large brown spider’

11. Termites (Order Isoptera = ‘equal wings’)⇫

POc *ane is well-supported as a term for ‘termite/white ant’. There are occasional mentions of an association of white ants with the supernatural in SES languages. In Arosi ane-hau refers to ‘k.o. white ant that lives in the ground; these ants were thought to be incarnations of the dead’, while in Sa’a they are ‘the food of ghosts on Malapa’. However, they would have been regarded as pests, and it is not surprising that we have no record of their ever being referred to by a *kali- term.

PAn		*SayaN	‘white ant, termite’ (Blust 2002)
PMP		*anay	‘termite’ (metathesis) (Blust 2002)
POc		*ane	‘termite’
Adm	Mussau	āne	‘termite’
Adm	Pak	ān	‘termite’
Adm	Lou	(ŋ)an	‘termite’
Adm	Loniu	an	‘termite’
PT	Saliba	yane	‘termite’
SES	Bugotu	ane	‘termite’
SES	Gela	ane	‘termite’
SES	Lau	sane	‘termite’
SES	’Are’are	sane	‘termite’
SES	Sa’a	sane	‘termite’
SES	Arosi	ane	‘white ant which bores and destroys’
SES	Arosi	ane(ʔara)	‘k.o. white ant found in coconuts’
SES	Arosi	ane(hau)	‘k.o. white ant that lives in the ground; name for its nest’
SES	Arosi	ane(niragui)	‘white ant, common wood-boring’
SES	Arosi	ane(wado)	‘white ant, large’
Mic	Carolinian	anay	‘termite; to be infested with termites’
Fij	Bauan	yane	‘k.o. moth’
Pn	Tongan	ane	‘moth, or more strictly, its larva, that eats holes in clothes etc. Also applied to silverfish’
Pn	Samoan	ane	‘termite, k.o. white ant’
Pn	Tikopia	ane	‘termite’
Pn	Rarotongan	ane	‘termite’

12. Crickets, grasshoppers (Order Orthoptera - ‘straightwings’); Cicadas (Order Hemiptera= ‘half wings’)⇫

If a generic term existed in POc, either for grasshoppers, or for grasshoppers + crickets + cicadas (+ mantises), I have not been able to reconstruct it. Although I have a number of reconstructions, their reflexes show considerable variation in gloss. Some terms refer to ‘singing insect’, which may indicate either cricket or cicada. Lack of agreement in glosses may reflect indifference among speakers, although I would expect a generic term if particular varieties were not seen as important. Some languages have simply adopted verb forms for ‘grasshopper’ - etymons from Kove (piti-piti ‘jumping insect found in beach rubbish’) and Raga (visi[ribi] ‘taro plant hopper’) are from POc *pitik ‘spring up suddenly’/ POc *pitik-pitik jump repeatedly. In Dobu crickets are named kelei ʔana toeʔita, literally ‘wallaby’s teacher’ (Ralph Lawton. pers. comm.). A grasshopper named kokoru-bote in Arosi is remarkable for the use found for it. It is described as ‘a large grasshopper, put on one’s head to eat lice’ (from koru ‘eat’, bote ‘lice’). Together with mantises and stick insects, grasshoppers and crickets are valued as food. Kwaio has a term, faʔafulu glossed as ‘generic term for edible insects, incl. grasshoppers, stick insects etc.’ Their appeal would no doubt be as a kind of snack food, a chance variation on their regular diet.

POc		*sisi(r,R)	‘orthopterous insect’
NNG	Mengen	sisi	‘grasshopper’
NNG	Lukep	sis	‘grasshopper’
NNG	Kaulong	u-sis	‘grasshopper’
NNG	Sengseng	e-sis	‘grasshopper’
NNG	Gedaged	sis	‘grasshopper, locust’
PT	Iduna	sili(fa)	‘k.o. small grasshopper’
MM	Tangga	sisi	‘cicada’
MM	Nehan	sir-siri(alum)	‘cricket’
MM	Nduke	hiri-ri	‘k.o. cicada’
MM	Roviana	sisiri	‘k.o. cicada (Diceropyga obtecta)’
SES	Kwara’ae	sīsī	‘generic for beetles and insects’
SES	To’aba’ita	sīsī	‘generic for beetles and fireflies’
NCV	Nese	sis	‘chirp, of cicada’

cf. also:

MM	Tolai	titi(tiŋ)	‘k.o. small locust’
NCV	Mwotlap	titi(key)	‘grasshopper’

Any resemblance between POc *riŋa and PROc *liŋo-liŋo is probably fortuituous.

POc		*riŋa	‘orthopterous insect’
Adm	Drehet	riŋ	‘cricket’
PT	Molima	lia	‘cicada’
MM	Nehan	(kaka)riŋ	‘grasshopper’

PROc		*liŋo-liŋo	‘night insect, probably a cricket’
Mic	Kiribati	niŋo-niŋo	‘an insect which chirps at night’
Fij	Bauan	liŋo-liŋo	‘night insect, esp. k.o. moth’
Pn	Tongan	liŋo-liŋo	‘k.o. cricket’
Pn	Samoan	liŋo-liŋo	‘cicada, sings in daytime. ʔālisi sings at night’ (Stair 1983: 207)
Pn	Tikopia	riŋo-riŋo	‘cricket (Gryllidae). Rarely seen’

POc		*siko	‘grasshopper’
MM	Tabar	ciko	‘grasshopper’
SES	Lau	siko	‘grasshopper’
SES	Kwara’ae	siko	‘grasshopper’ (Whitmore 1966)
SES	To’aba’ita	siko	‘generic for a class of insects that includes locusts, grasshoppers, stick insects, mantises, cicadas’
NCV	Nese	(narra)sɣo	‘grasshopper’ (narra ‘?’)

POc		*sakʷa	‘grasshopper or stick insect’
NNG	Gedaged	sok(sok)	‘stick insect’
NNG	Takia	sok	‘large insect, prickly, with six legs, eaten’
PT	Molima	saga-saga	‘green mantis (eaten)’
MM	East Kara	saɣwa	‘grasshopper’ (reflects *saqa)
Mic	Carolinian	tāxa	‘grasshopper’

The next term, POc *lale ‘cicada’, is based on external evidence plus reflexes from two closely related languages. It remains an uncertain reconstruction.

PAn		*lalay	‘cicada’ (Blust 2002)
POc		*l(a,e)le	‘cicada’
PT	Molima	lele	‘a mountain insect with very attractive cry’
PT	Dobu	lele	‘cicada’

The next reconstruction is reliable in form only at PEOc level, and there is little consistency of gloss even within Polynesian languages.

PEOc		*(s,j)eqe	‘grasshopper ?’
SES	Bugotu	se-se	‘grasshopper’
Fij	Rotuman	jei	‘cricket’
PPn		*seqe	‘insect spp. including mantis, stick insect, locust’
Pn	Tongan	heʔe	‘locust, grasshopper’
Pn	Niuean	hē	‘grasshopper, locust’
Pn	Rennellese	seʔe	‘k.o. small cricket’
Pn	Samoan	sē	‘stick insect’
Pn	Tikopia	sē	‘stick insect; mantis’
Pn	Tokelauan	hē	‘locust, grasshopper’

John Lynch (pers. comm.) notes a number of NCV forms for ‘cicada’ (Mota puŋa, Naman bəgale, Neve’ei na/bugali Tape bəŋale, Paamese ha/vuŋalii). and Namakir biŋaleh ‘cricket’ which support PNCV *bu(g,ŋ)alis(V). This in turn suggests a possible relationship with the following PPn reconstruction, *kālisi.

PPn		*kālisi	‘cicada, cricket’
Pn	Tongan	kālihi	‘k.o. cricket’
Pn	Samoan	ʔālisi	‘cicada, cricket (night insects)’
Pn	Ifira-Mele	karisi	‘cicada’

PPn		*wāwā	‘orthopterous insect’
Pn	Niuean	vā	‘insect sp.’
Pn	Rarotongan	vāvā	‘cricket’
Pn	Tongarevan	vāvā	‘grasshopper, cricket’
Pn	Mangarevan	vāvā	‘insect sp.’
Pn	Tahitian	vāvā	‘praying mantis, stick insect’
Pn	Tuamotuan	vāvā	‘grasshopper, stick insect’

13. Mantises (Order Mantodea = ‘like a prophet’), stick insects (Order Phasmatodea = ‘like a ghost’)⇫

Although there is some confusion of gloss within cognate sets, it seems clear that POc speakers distinguished praying mantises (*papa) from stick insects (*mimis-mata). The ability to project a milky chemical painful to the skin which is reflected in POc *mimis-mata is a defensive mechanism of at least some stick insects (Serkan Alasya, CSIRO Entomology, pers. comm.).

PMP		*(kali)papa	‘praying mantis, grasshopper’7
POc		*papa	‘praying mantis’
MM	Tolai	pap	‘praying mantis’
MM	Roviana	papa(maho)	‘praying mantis’ (also ma/maho ‘green mantis’)
MM	Nehan	popo(hɔsoi)	‘walking stick insect’

POc *mimis-mata, literally ‘urinate’ + ‘eye’, reflects awareness that the stick insect can eject a fluid which causes pain if it touches the eye. One of the better-known Phasmatodea of the Oceanic region is Megacrania batesii, the pandanus stick insect.

POc		*mimis-mata	‘stick insect’
MM	Roviana	mimi-mata	‘walking-stick insect which ejects a fluid said to cause intense pain should it touch the eye’
NCV	Raga	me-merehi-mata	‘praying mantis’ (lit. ‘urinate on eye’)
Fij	Wayan	mīmī-mata	(1) ‘praying mantis’; (2) ‘stick insect (generic)’ (lit. ‘urinate in eye’)

cf. also:

PT

Molima

mata-seʔa-seʔa

‘large poisonous beetle’ (seʔai or seiʔa ’to squirt poison, as a millipede; lit. squirt in eye??)

Closely related to stick insects are leaf insects. The only terms I have located are descriptive compounds: Tawala (PT) hiaga luguna (hiaga ‘?’, lugu ‘leaf’), Tongan moko-moko tuʔaniu (moko-moko ‘crawling insect, tuʔaniu ‘midrib of coconut leaflet’), and Rennellese manu gaukei (‘insect’ + ‘leaf’).

Some varieties of mantis and stick insect are listed as edible in a number of languages.

14. Fireflies (Order Coleoptera = ‘sheath wings’: Family Lampyridae)⇫

Judging from ethnographic references and brief elaborations in wordlists, it is fireflies, among all the insects of the Oceanic world, which are most associated with supernatural qualities. For Takia (NNG) speakers, a firefly, lutot, is thought to be a messenger of someone’s death (Bugenhagen wordlist). The term for a firefly in Sissano (NNG), mas eyiaw, is a compound term, mas being defined as ‘devil; spirit of dead relative’ and eyiaw as ‘evil spirit; thief’ (Whitacre wordlist). Seligmann quotes a report from Bartle Bay (Wedau, PT) that ‘if a firefly enters a house at night, it is a sign that someone in the house will die.’ (1910:653). Molima (PT) speakers regard fireflies as emissaries of witches who enter the house at night in order to report back on who is there, and for that reason they are killed as soon as they are seen (Ann Chowning, pers. comm.). In Nakanai (MM), although its speakers consider that fireflies embody spirits of the dead, they are delighted if a firefly enters the house at night, because it is a sign that someone recently dead is paying a visit (Ann Chowning, pers. comm.). Ivens writes that fireflies are popularly classified as ghosts in Sa’a and Ulawa (1927:189), and in his Sa’a dictionary he defines fireflies (pulu-pulu) as ‘souls of dead persons’. The Tolo people of Guadalcanal regard cicadas and fireflies as the ghosts of unknown people, or ghosts of the forest (Ivens 1930:229). Tamambo (NCV) speakers are frightened of them, and try to avoid them as ‘little devils’ (Dorothy Jauncey pers. comm.). In Wayan Fijian, fireflies (tōtōvuata) are believed to be a sign of the gods inspecting fruit crops. The Rennellese see a firefly as the embodiment of a goddess.

The belief that fireflies are messengers of doom is an ancient one among Austronesian speakers. Blust (pers. comm.) provides additional evidence from Isneg (northern Philippines) that fireflies are precursors of death, and Karo Batak (northern Sumatra) where their presence in a house signals that thieves will come. Of all POc reconstructions, then, I should expect that for ‘firefly’ to be most likely to retain the kali-/qali- indicator of supernatural status. I have limited evidence supporting two reconstructions derived from PMP *qali-petpet, POc *qali-popot ‘firefly’ and POc *(k,q)ali-totop ‘firefly’, the latter with metathesis. POc *(k,q)ali-totop depends on one reflex, from Rennellese, for reconstruction of its first element and on another, Nduke, for reconstruction of its second element.

PMP		*qali-petpet	‘firefly’ (ACD: WMP)
POc		*qali-popot	‘firefly’
NCV	Lewo	le-popo	‘firefly’
NCV	Tamambo	vovo(mbo)	‘firefly; regarded as little devils and avoided’

PMP		*qali-petpet	‘firefly’ (ACD)
POc		*(k,q)ali-totop	‘firefly’ (metathesis)
NNG	Takia	lu-tot	‘firefly’
MM	Tangga	kel-tot	‘firefly’
MM	Nduke	alal-totovo	‘luminous flying insect, its light flashes on-off’
Pn	Rennellese	ʔagi-to	‘firefly, seen inland, a rare phenomenon; believed to be the embodiment of the goddess Sikingimoemoe, and might not be killed’

A putative reconstruction, POc *bulu-bulu ‘firefly’, must be regarded as dubious because witnesses are from adjoining subgroups where borrowing cannot be ruled out. Etymons may be derived from POc *bulut ‘gum’ whose reflexes can include ‘a torch (of gum)’, and thus extend to ‘light, firefly’.

MM	Nehan	bol-bol	‘lightning bug, near swamps’
SES	Gela	bulu (tora)	‘firefly; phosphorescent fungus’ (bulu ‘torch, lamp’)8
SES	Sa’a	pulu-pulu	‘firefly. Regarded as souls of dead persons, and killed when they enter a house’
SES	Kwaio	bulu-bulu	‘star; firefly. Believed (esp. when enters house) to be a messenger from an ancestral spirit.’
SES	Arosi	buru-buru	‘firefly; torch (of gum)’
SES	Arosi	buburu(atoā)	‘k.o. cricket’ (atod ‘noon’)
SES	’Are’are	puru-puru	‘star; firefly’
SES	To’aba’ita	būbulu	‘star; firefly; beetle sp.’

Other terms use reflexes of POc *mata ‘eye’, sometimes extended to ‘star’ and ‘firefly’.

NNG	Mengen	mata	‘eye; star; firefly’ (mata-balaŋ ‘bee/wasp’)
MM	Nakanai	mata(tabu)	‘star; firefly’
MM	Nakanai	mata(laso)	‘k.o. long glow worm’

Other languages may make similar associations between fireflies, stars and sometimes phosphorescence.

NNG	Sengseng	limlek	‘firefly; star’
MM	Roviana	pinoro	‘phosphorescence as seen in the sea or on fireflies etc.’ (pino-pino ‘a star’)

15. Beetles (Order Coleoptera)⇫

Although there are thousands of different beetle species, very few taxa are named in wordlists. Only two POc reconstructions have been made, and in the first listed below there are grounds for believing that the gloss given in many of the daughter languages is a recent adaptation. The large rhinoceros beetle (Oryctes rhinoceros), which bores into the crown of the coconut palm and is considered a serious pest in many parts of the Pacific, is believed to have been introduced from southeast Asia only in the 20th century (Gressitt & Homabrook 1985:34). A beetle similar in appearance and behaviour is Xylotrupes gideon, found in almost all parts of New Guinea. Presumably, speakers used an existing beetle term to refer to the introduced beetle. The second reconstruction, POc *kamak, probably refers to longicorn beetles, those with extremely long antenna (Family Cerambycidae), the larvae of which attack timber.

POc		*tabuRuRu	‘k.o. beetle’
NNG	Sengseng	taput	‘an enormous black beetle’
MM	Nakanai	tabuburu	‘coconut beetle’ (expect †tabururu)
MM	Tabar	taburu(ŋa)	‘beetle’
MM	Tolai	taburur	‘horned beetle’
MM	East Kara	tevu(ŋun)	‘rhinoceros beetle’
SES	Gela	tabelulu	‘rhinoceros beetle’
SES	Bugotu	tabilolo	‘rhinoceros beetle’
NCV	Uripiv	barur	‘yam weevil, Papuana beetle’ (loss of first syllable unexplained)
SV	Sye	n-tompi	‘k.o. beetle’

cf. also:

NNG	Mapos Buang	abu	‘rhinoceros beetle’
MM	Nduke	vure	‘rhinoceros beetle’

POc		*kamak	‘beetle, possibly longicom’
NNG	Kaulong	kamak	‘k.o. insect; large, edible, found inside the hollows of trees’
NNG	Yabem	kaŋ	‘beetle: capricorn beetle, stag beetle, homed beetle, weevil’
NNG	Sengseng	makah	‘longicorn beetle’ (metathesis)
PT	Iduna	kama	‘large insect with long fat body’
Pn	Tikopia	kama	‘insect, possibly of various types, attacks cultivated plants’

cf. also:

MM

Nakanai

komaga

‘long thin black beetle, edible, longicorn beetle’

A surprising omission from most wordlists are terms for the taro beetle (Papnana uninodis). Although these beetles may be as long-established in the region as the tubers which we know were cultivated in Proto Oceanic times, I have not been able to reconstruct a term for them other than PSES *ko[l,R]o-ko[l,R]o ‘beetle that attacks plants’. Damage to the tuber is caused by the adult beetles, and not by their larval stage which feeds on soil and roots (Robin Hide, pers. comm.). I can locate very few terms for the pest in Western Oceanic languages, none cognate. Although damaged tubers cause significant economic loss in today’s marketing terms, it may be that the damage was of little significance when tubers were primarily for home consumption.

PSES		*ko[l,R]o-ko[l,R]o	‘beetle that attacks plants’
SES	Bugotu	kololo	‘coconut beetle’
SES	Gela	kololo	‘taro-eating beetle’
SES	’Are’are	ʔoro-ʔoro	‘taro-boring beetle’
SES	To’aba’ita	ʔoro(kʷao)	‘k.o. white worm, attacks sugar-cane stems’ (kʷao ‘white’)

cf. also:

SES

Sa’a

ʔoroʔoro

‘beetle that bores into yams’ (for †ʔoloʔolo)

Wordlists contain a few references to edible beetles. Sudest (PT) has a term gilai ‘a large black sago beetle; people eat both the adults and the larvae’; Nakanai (MM) has komaga ‘an edible insect like a long thin black beetle’ and bureka ‘large (6 ins) edible insect found in trees’; Wayan (Fij) has gou ‘large beetle, Coleopterus sp., 2 inches long with hard shell. Eaten by some people’.

16. Cockroaches (Order Blattodea = ‘light avoiders’)⇫

Three POc reconstructions and one lower-level reconstruction can be made for ‘cockroach’.

PAn		*Sipes	‘cockroach’ (ACD)
POc		*ipos	‘cockroach’ (ACD)
Adm	Nali	yih	‘cockroach’
Adm	Nauna	ih	‘cockroach’
Adm	Drehet	ih	‘cockroach’
NNG	Mutu	up	‘cockroach’
NNG	Kaulong	e-yus	‘cockroach’
NNG	Sengseng	e-yus	‘cockroach’
MM	Tolai	ipi(na)	‘sp. beetle’ (ipina-koto ‘cockroach’)
NCal	Cèmuhî	iwet	‘cockroach’
NCal	Nemi	yuet	‘cockroach’

cf. also:

NCV

Raga

ihi

‘cockroach’ (for expected ivi)

POc		*musi	‘cockroach’
PT	Roro	muhi	‘cockroach’ (for expected musi)
MM	Maringe	mhi-mhi(gi)	‘stink beetle’
SES	Bugotu	muhu	‘cockroach, beetle’
SES	Gela	muhu	‘cockroach’
SES	Tolo	musu	‘cockroach’
SES	Ghari	mosu	‘small cockroach’

Final -u for †-i in the SES items appears to reflect vowel assimilation.

The next set shows a variety of reduplication patterns. Some Polynesian terms show possible contamination from English cockroach or Spanish cucaracha.

POc		*kʷaru-kʷaru	‘cockroach’
PT	Bwaidoga	kʷalu-kʷalu	‘cockroach’
MM	Nehan	kur-kuru	‘cockroach’
MM	Banoni	karo	‘small white cockroach’
Pn	Tongan	kakalu	‘large cricket/cicada sp.’
Pn	East Futunan	kakalu	‘that which smells bad’
Pn	Samoan	alalū	‘cockroach’
Pn	Tahitian	aʔararū	‘small beetle’
Pn	Māori	kekererū	‘stinkroach’
Pn	Hawaiian	ʔelelū	‘cockroaches’

PPn		*moŋa-moŋa	‘cockroach’ (pollex)
Pn	Tongan	moŋo-moŋa	‘any kind of beetle or cockroach’
Pn	Niuean	moŋa-moŋa	‘cockroach’
Pn	Samoan	moŋa-moŋa	‘cockroach’
Pn	Nanumea	moŋa-moŋa	‘beetle sp.’
Pn	Tokelauan	moŋa-moŋa	‘cockroach’

17. Grubs, caterpillars⇫

Grubs and caterpillars are the larval stages of beetles (Coleoptera) and butterflies (Lepidoptera). Particular kinds are more likely to be named if they are a useful food source, or if they attack crops.

One which is widely recognised is the edible sago grub (Order Coleoptera, family Curculionidae), a valuable source of protein. Sago grows in low-lying, swampy areas of Melanesia where other food plants do not thrive. Lack of the means of a varied diet may contribute to the specialised sago grub cultivation employed by the Labu (NNG) people.⁹ In Sissano the sago grubs, yiat, are used in food offerings to the spirits in some healing ceremonies (Whiteacre wordlist).

The Admiralties languages in the next set all have -e- for †-a-.

PMP		*qabated	‘sago grub’ (Blust 2002)
POc		*qapator	‘sago grub, edible’
Adm	Loniu	het	‘sago grub’
Adm	Ere	ehet	‘sago grub’
Adm	Titan	aet	‘sago grub’ (for †ahat)
Adm	Pak	keher	‘sago grub’
Adm	Penchal	kahɛt	‘sago grub’
Adm	Lenkau	kehetr	‘sago grub’
NNG	Sissano (Arop)	yiat	‘sago grub’
NNG	Kove	awatolu	‘sago grub’
NNG	Gedaged	gafat	‘big grub found in rotten sago logs. Eaten raw or roasted. A great delicacy’ (initial g- unexpected)
PT	Lala	avako	‘grub, found in sago palms’
SES	Gela	vato	‘k.o. grub, eaten as food on Mala’
SES	Ghari	vato	‘white worm in decayed tree’
SES	Arosi	hao	‘white grub in rotten wood, eaten’ (first syllable lost)
SES	Lau	safao	‘grub that eats sago palms’
SES	Kwaio	lafao	‘large edible grub, larva of stag beetle’
SES	Sa’a	sahao	‘the grub found in fallen logs of breadfruit tree, supposed to turn into the cockchafer beetle’
SES	’Are’are	rahao	‘big, white worm, lives in dead trees’
NCV	Mota	vato	‘white grub found in dead trees, eaten’
NCV	Lewo	varo	‘grub found in dead wood, edible’
NCV	Raga	avato	‘large white insect found in trees, edible, larva of longhorn beetle’
SV	Sye	n-avat	‘edible wood grub’
SV	Anejom̃	n-ahat	‘edible wood grub’
Fij	Wayan	avato	‘wood-boring grubs of various species’
Fij	Bauan	yavato	‘grub in a tree, develops into the coleopterous called qou’
Pn	Tongan	ʔofato	‘k.o. white grub found in timber, sometimes eaten’
Pn	Samoan	ʔafato	‘large edible grub, found in dry trees’ (ʔ- unexpected)
Pn	East Futunan	ʔafato	‘larvae which are eaten; large edible grub, found in dry trees’
Pn	Rennellese	ahato	‘larva of longicorn beetle Olethrus tyrranus’ (for †ʔahato)

cf. also:

PT	Sinaugoro	kavata	‘caterpillar’
MM	Nakanai	(k,g)avato	‘black lizard, found in rotten trees’ (kopa ‘larva of a beetle, found in trees and eaten’)

POc *muno may have been a generic term for caterpillars. Reflexes are not found in Central Pacific where a reflex of POc *qanupe is the most widespread term for ‘caterpillar’.

POc		*muno[-muno]	‘caterpillar, grub’
Adm	Lou	mon-muon	‘caterpillar’
MM	Simbo	mu-muno	‘worm’
MM	Nduke	mu-muno	‘looper caterpillar’
MM	Roviana	muno-muno	‘general name for caterpillars and grubs’
SES	Bugotu	muno	‘caterpillar’
SES	Lau	muno	‘caterpillar’
SES	Sa’a	muno	‘larva, chrysalis’
SES	’Are’are	muno	‘caterpillar’
SES	To’aba’ita	muna	‘k.o. caterpillar that turns into a butterfly’
Mic	Puluwatese	mūn	‘caterpillar’
Mic	Carolinian	mʷūl	‘generic term for maggots, larvae or small worms’
Mic	Woleaian	mʷura	‘caterpillar, silkworm’

cf. also:

MM

Nduke

meno

‘taro beetle’

WOc cognates of POc *qanupe refer to trepangs,¹⁰ also known as bêche-de-mer or sea slugs (ch.4, §5.2.1), while Eastern Oceanic cognates refer to caterpillars. Without external cognates, the meaning of POc *qanupe is indeterminate.

POc		*qanupe	‘caterpillar or k.o. sea cucumber (indeterminate)’
PWOc		*qanupe	‘sea cucumber, holothurian’
NNG	Kove	anuwe	‘trepang’
PT	Dobu	kanue	‘bêche de mer’
PT	Galea	anue	‘general term for all sea cucumbers’ (Lawrence Rutter pers.comm.)
MM	Nakanai	haluve	‘trepang’
PEOc		*qanupe	‘caterpillar’
SES	Ghari	nive	‘caterpillar’
Fij	Rotuman	aniha	‘caterpillar, maggot’
Fij	Bauan	(b)anuve	‘caterpillar’
Pn	Tongan	ʔunufe	‘caterpillar (generic)’
Pn	Samoan	ʔanufe	‘worm, caterpillar’
Pn	Tikopia	unūfe	‘caterpillar’
Pn	Nanumea	anufe	‘caterpillar, worm, slug, etc.’
Pn	Tuamotuan	anuhe	‘caterpillar’

18. Centipedes (Class Chilopoda): Millipedes (Class Diplopoda)⇫

Another *qali- term, POc *qalipan ‘centipede’, is reconstructable, probably embracing millipedes as well. Evidence of an association between centipedes and the supernatural comes from ethnographic descriptions as well as linguistic clues. For instance, Sengseng (NNG) speakers consider centipedes to be supernatural beings, and carefully remove them from their houses (Ann Chowning, pers. comm.). Elbert’s Rennellese dictionary defines ʔagi-paipai as ‘centipede, considered the embodiment of non-worshipped deities and a loathsome creature’. In his Tikopia dictionary Firth defines morokau as ‘centipede. Trad. held to be frequent embodiment of spirit’. There is also a Samoan term atua-loa ‘k.o. centipede with poisonous bite’, whose prefix atua- has supernatural significance (discussed in §27). In addition to the *qalipan cognate set I list non-cognate terms for centipede from a range of languages across several subgroups which all exhibit a reflex of the qali-/kali- prefix.

Reflexes of *qalipan are numerous and widespread. As well, I have a number of lower-level reconstructions, possibly for particular kinds of centipede. Two kinds of centipede/millipede are described in wordlists, categorised by behaviour or luminosity rather than number of legs. One, when threatened, squirts a kind of acid from between its scales which can cause blisters or even permanent eye damage, while a second is described as phosphorescent or luminous. A millipede in Arosi (SES), doŋa, is also singled out as having a vile smell. I have one lower-level reconstruction for the luminous variety, PPn *taqe-tuli ‘phosphorescent centipede or millipede’, whose first element is derived from POc *taqe ‘excrement’, implying strongly that the creature is considered unpleasant in Central Pacific languages. In Tahiti, however, where there are two kinds, not venomous, one short and thick and light brown, the other long, threadlike and phosphorescent, they were well regarded. Teuira Henry (1971:391) writes that ‘when a centipede fell upon a wound or bruise of any kind, it soon healed. If it crawled upon a sick person, a speedy recovery followed.’

PAn		*qalu-Sipan	‘centipede’ (ACD)
PMP		*qalu-hipan	‘centipede’ (ACD)
POc		*qalipan	‘centipede’
Adm	Mussau	aliena	‘centipede’
Adm	Wuvulu	ali-alifa	‘centipede’
NNG	Manam	alia	‘centipede’
NNG	Mangga	garivaŋ	‘centipede’
NNG	Yalu	kanif	‘centipede’
NNG	Wampur	ganef	‘centipede’
PT	Maopa	ɣaiva	‘centipede’
PT	Gumawana	ganiva	‘centipede’
PT	Saliba	kalihai	‘centipede’
PT	Duau	ganihana	‘centipede’
PT	Motu	aiha	‘centipede’
MM	Bola	ɣaliua	‘centipede’
MM	Halia	lihaŋ(a)	‘centipede’
MM	Nehan	hilaŋ	(metathesis)
MM	Nduke	livaŋ(a)	‘centipede’
SES	Gela	liva	‘centipede’
SES	Bugotu	liva	‘centipede’
SES	Lengo	aliva	‘centipede’
SES	Lau	safila	(metathesis)
SES	Kwaio	lalifa	‘centipede’
SES	Arosi	(karikari)ʔariha	‘centipede’
Fij	Rotuman	aniha	‘maggot; also applied to many kinds of caterpillars and millipedes’
Fij	Bauan	yaliva	‘a red millipede, possibly generic for millipedes’

cf. also:

Pn

Rennellese

ʔagipaipai

‘centipede’

Other terms for centipede which apparently exhibit a reflex of the qali-/kali- prefix include:

Adm	Drehet	kxane-pup	‘centipede’
PT	Molima	gani-geva	‘centipede’
MM	Ririo	kali-gava	‘centipede’
MM	Sisiqa	kala-gava	‘centipede’
MM	Babatana	kali-gava	‘centipede’

The compound terms below yield a reconstructed first element, PWOc *mʷali-.

PWOc		*mʷali-	‘millipede’
PT	Bwaidoga	mʷali(keke)	‘millipede (yellow, causing irritation or burn to skin when it squirts acid from between scales)’
PT	Kilivila	mʷani(ta)	‘centipede’
MM	Maringe	mali(so)	‘millipede’

cf. also:

PT

Molima

pʷali(keke)

‘millipede’

POc *weli (1) ‘fireworm, sea centipede’; (2) ? ‘k.o. millipede’ is included in chapter 4, §7.2.

PPn		*taqe-tuli	‘phosphorescent centipede/millipede’ (pollex)
Pn	Tongan	teʔe-tuli	‘phosphorescent centipede’
Pn	Rennellese	taʔ-tugi	‘long and slender centipede variety’
Pn	Samoan	tae-tuli	‘phosphorescent millipede’
Pn	Tikopia	tae-turi	‘millipede’

cf. also:

Fij

Rotuman

turi

‘phosphorescent centipede’ (Pn borrowing)

In Bugotu (SES) the term for a phosphorescent centipede, liva tora, is a literal description: (liva ‘centipede’, tora ‘phosphorescent’).

Reflexes of a reconstruction from a different semantic field, POc *taqe ‘excrement’ (see ch.2, p.60), have extended their meaning to refer to a centipede in Kahua (SES) and Bauan Fijian and a scorpion in Tamambo (NCV) and Anejom (SV), no doubt reflecting their common characterisation as painful stinger.

19. Scorpions (Class Arachnida: Order Scorpiones)⇫

A substantial number of terms for scorpions have been collected, but only a single reconstruction, Proto SE Solomonic *vari ‘scorpion’ (not listed here) has been possible. In the Solomons, many terms are compounds or phrases which include the word for centipede (Gela liva-ni mala, Birao kaukau-liva, Lengo aliva-ni-ɣoro). Although centipedes and scorpions belong to different zoological classes, the linking of the two creatures in this way may be due to the similar appearance of grasping pincers, and to the facts that both are capable of delivering a severe bite or sting, and both frequent the same habitat, under rocks or leaf litter. The scorpion is named as crab-like in Gapapaiwa (gʷaɣa-gʷaɣa ‘scorpion; idiom for one who moves slowly and fearfully on heights) (gʷaɣa ‘crab (generic)’) and possibly also in Molima (dowa ‘scorpion, land crab’), although Chowning adds a question mark to her gloss.

20. Spiders (Order Araneae)⇫

Although POc speakers would undoubtedly have named a range of spiders, I have not been able to link reconstructions with specific kinds of spider. I have only one lower-level example in which a name is assigned to a particular spider — PCP *tuku-tuku ‘k.o. spider that lets itself down by a single thread’. POc *lawaq, traceable back to PAn, with cognates right across the Oceanic region, may have referred primarily to ‘spider’ but by POc times had come to refer as well to ‘spider web’ and ‘k.o. fish net’ (see vol.1,212). I also find a number of reconstructions with formal similarities, including POc *koko, POc *gagao~*(g,k)a(g,k)a-, PWOc *kuku, and PEOc _*[ko]ko-miji. Some kind of word play has evidently taken place, although I have no ethnographic evidence that particular spiders are associated with any kind of supernatural belief or taboo. On the contrary, they are freely eaten in many communities. Wordlists of a number of languages including Molima and Sinaugoro (PT), Takia (NNG) and Hawaiian (Pn), record particular kinds of edible spider.

Reconstructions other than *lawaq have an unusually large number of reflexes which are compounds with a non-reflected element. These reconstructions may refer to particular kinds of spider. As Pawley (2000: 8) has noted, names for folk specifics in plants and animals, or, to be more exact, the secondary or modifying terms in binomials, are notoriously unstable and will, in most cases, not be reconstructable for a language as remote in time as POc.

PAn		*lawaq	‘spider’ (Blust 2002)
POc		*lawaq	‘spider, spider web’ (Also ‘fish net’)
Adm	Loniu	(wi)law	‘spider’
NNG	Tuam	lawag	‘spider’
NNG	Gitua	lawak	‘spider’
NNG	Malai	lawak	‘web’
NNG	Wogeo	lawa	‘spider’
PT	Bwaidoga	nawaya	‘spider’
PT	Gapapaiwa	nawa-nawa	‘spider web’
PT	Sudest	lawa	‘spider’
PT	Motu	vala-vala	‘cobweb’ (metathesis)
MM	Vitu	lava	‘spider’s web’
SES	Lau	lakʷa	‘k.o. large yellow spider and large web’
SES	Kwaio	lakʷa-lakʷa	‘spider web’
SES	’Are’are	rawa	‘spider, cobweb’
SES	Sa’a	lawa	‘spider’s web, spider’
SES	Arosi	rawa	‘spider, cobweb, small net’
NCV	Mota	(ta)lau	‘cobweb, either single line or web’
NCV	Nguna	(ka)lau	‘spider (web)’
SV	Anejom̃	ni-lva	‘spider web’
Fij	Bauan	lawa	‘fishing net’
Fij	Bauan	viritā lawa-lawa	‘cobweb’
Fij	Bauan	tina-ni-viritā lawa-lawa	‘spider’ (lit. ‘mother of cobweb’)

PPn *lewe is probably cognate, with reduplication and addition of the prefix *ka-, which may reflect *kʷakʷa, discussed below the next item.

PPn		*ka-lewe-lewe	‘cobweb, spider’ (pollex)
Pn	Niuean	ka-leve	‘cobweb’
Pn	Tongan	ka-leve-leve	‘k.o. spider’
Pn	East Futunan	ka-leve-leve	‘spider, spiderweb’
Pn	Samoan	ʔa(poŋā)-leve-leve	‘spider; spider’s web’
Pn	Tikopia	ka-reve-reve	‘spider’s web, all types’
Pn	Emae	ka-reve-reve	‘spider web’
Pn	Rarotongan	(pūŋā)vere-vere	‘spiderweb’ (pūŋā ‘fortress, retreat’; metathesis)
Pn	Hawaiian	(pūnā)wele-wele	‘spinning spider’ (metathesis)
Pn	Tuamotuan	(puŋā)vere-vere	‘a cobweb’ (metathesis)

In the cognate set below the Malai (NNG) term and those from the Malakula languages Nahai’i, Axamb and Nisvai (NCV) speak in favour of a freestanding POc *gagao, whilst the Molima (PT), Teop, Simbo, Torau, Kia, Kokota (all MM) and Gela (SES) terms point to *gaga- and *kaka- as forms occurring in compounds.

POc		*gagao~(g,k)a(g,k)a-	‘k.o. spider’
NNG	Malai	(a)gagau	‘spider’
PT	Molima	gaga(boʔa)	‘house spider’
MM	Teop	kaka(rebasu)	‘spider’
MM	Simbo	ɣaɣa(vere)	‘spider’
MM	Torau	kaka(piro)	‘spider’
MM	Roviana	kaga	‘spider with long legs, lives in the house’
MM	Kia	kaka(para)	‘spider’
MM	Kokota	kaka(fre)	‘spider’
SES	Gela	kaka(vere)	‘spider’ (probably borrowed from an Isabel language)
NCV	Raga	(ta)gaga	‘generic for spiders’
NCV	Naha’ai	ne-gegeu	‘spider’ (John Lynch, pers. comm.)
NCV	Axamb	na-gagao	‘spider’ (John Lynch, pers. comm.)
NCV	Nisvai	na-gao	‘spider’ (John Lynch, pers. comm.)

POc *koko and PWOc *kuku may be variants of the same term as *kaka. Two alternate hypotheses about their origins are (a) that *ko reflects the second syllable of *gagao, and (b) that the reflexes supporting *kaka-, *koko and *kuku reflect POc *kʷakʷa. Evidence for (b) is that (i) SE Solomonic languages retain k- and this is diagnostic of POc *kʷ-, and (ii) Western Oceanic (PT, MM) reflexes other than Simbo retain k- rather than leniting it to ɣ- or zero, the usual reflexes of *k-.

POc *koko becomes *[ko]ko-miji ‘k.o. spider’ in PSES with the addition of -miji an element which occurs also in other compounds meaning ‘spider’ (Dori’o [SES] kala-midi) although not found separately. It also may occur in Nakanai kamimisi ‘spider’, if intepreted as kami-misi with kami- < kali by assimilation. I am unable to suggest a meaning for POc *-miji, other than to note that I have found it only in spider terms.

POc		*koko	‘spider’
NNG	Patep	kɔɔ	‘spider’s web’
PT	Dobu	(wa)koko	‘brown house spider’
MM	Tolai	koko	‘spider: all kinds of house spider’
MM	Kandas	ko	‘spider’
MM	Halia	koko(ratski)	‘spider; starfish’
PSES		*[ko]ko-miji	‘k.o. spider’
SES	Lengo	ka-midi	‘spider’
SES	Longgu	ko-midī	‘spider’
SES	Tolo	koko-mici	‘large house spider’
SES	Talise	ko-mici	‘spider’
SES	Ghari	ko-mici	‘common spider’
SES	Malango	ko-miji	‘spider’

cf. also:

PT	Wedau	(wa)gogo	‘k.o. spider’
PT	Gapapaiwa	(wa)koko	‘spider type; A specific large species of spider which occupies our outhouse’
Pn	Rennellese	ko-miti	‘various spiders that bite’ (possible borrowing SES)

PWOc		*kuku	‘spider’
NNG	Bariai	kuku	‘spider’
NNG	Gedaged	kuk	‘spider’
NNG	Kilenge	na-kuku(n)	‘spider’s web’
NNG	Malalamai	kuku(lamlambe)	‘spider’
MM	Vitu	(pa)kuku	‘spider’
MM	Bali	(ma)kuku	‘spider’
MM	Nakanai	kuku(re)	‘spider’
MM	Marovo	kuku	‘huntsman spider’

PWOc		*ka(p,b)ilakʷa	‘k.o. spider’
NNG	Mapos Buang	kapik	‘large brown hairy spider’ (for ’)
PT	Molima	kapinoko	‘k.o. edible black bush spider’
MM	Nakanai	kalalaua	‘spider web’ (kala- for ^kabe- by assimilation)
MM	West Kara	kabelaua	‘spider’
MM	Tigak	kavelo	‘k.o. spider’
MM	Lamasong	kabiloŋ	‘k.o. spider’
MM	Minigir	(ka)kobiloko	‘k.o. spider’
MM	Tolai	kabiloko	‘spider; spiderweb’
MM	Ramoaaina	kabulak	‘spider, cobweb’
MM	Bilur	kabilak	‘spider’

The only reconstruction which can safely be attributed to a particular kind of spider is PCP *tuku-tuku ‘k.o. spider which lowers itself on a single thread’, from PCP *tuku ‘let down’.

PCP		*tuku-tuku	‘k.o. spider which lowers itself on a single thread’
Fij	Rotuman	fuʔ-fuʔu	‘k.o. spider that lets itself down on a thread’
Fij	Bauan	tuku-tuku	‘k.o. small spider’
Pn	Tikopia	tuku-tuku	‘spider (unid. but includes daddy long legs. Trad, a creature with silk thread, assoc, with spiritual beings Manu and Tanggaroa)’
Pn	Māori	tuku-tuku	‘spiderweb; ornamental latticework’
Pn	Tuamotuan	tuku-tuku	‘k.o. harmless spider’
Pn	Hawaiian	kuʔu-kuʔu	‘short-legged spider, lowers itself on a single thread’

21. Leeches (Class Hirudinea)⇫

There is a well-supported POc reconstruction for ‘leech’ based on reflexes in NNG, PT and NCV languages. I have not so far located any term for leech, cognate or otherwise, east of Fiji, and assume that the creatures are not found in Micronesia or Polynesia (with the exception of New Zealand, where the Maori term is ŋata i.e. ‘snake’) (Ross Clark, pers. comm.). Pawley has reconstructed a similar term, PROc *drumane ‘anemone’ (ch.4, §6.1) which is almost in complementary distribution, and the terms may be related.

POc		*droman	‘leech’
NNG	Mangga	domaŋ
NNG	Mapos Buang	domŋ	‘leech’
NNG	Adzera	tuaman	‘leech’
PT	Are	domeni	‘leech’
PT	Gapapaiwa	domani	‘leech’
PT	Tawala	domani	‘leech’
PT	Dobu	domana
PT	Molima	domana	‘mountain leech’
PT	Nimoa	dome
PT	Lala	toma	‘leech’
PT	Motu	doma
PT	Taboro	doma	‘leech’
NCV	Tamambo	ruma
NCV	Raga	rimʷa
NCV	Ambae	rimʷe

cf. also:

MM	Nakanai	soma	(for †doma)
MM	Tolai	domol

22. Worms⇫

22.1. Worm (generic)⇫

There is a widespread belief among Austronesian speakers that dental caries is caused by a small worm. Both the creature and the associated condition are named in different languages by reflexes of various reconstructed terms for ‘worm’. Blust comments on two WMP languages, Cebuano and Malagasy, whose reflexes of PAn *qulej ‘maggot’ refer also to ‘tooth decay’ or ‘toothache, supposed to be occasioned by a small worm in the tooth’ (ACD) (§23). Dobu (PT) has a term kimʷata ‘the supposed insect which causes caries in children’ (mʷata ‘snake’). The To’aba’ita term is wā-lifo ‘tooth decay, believed to be caused by worms’ (wā ‘worm, grub, maggot, caterpillar, larva’, lifo ‘tooth’). Other WMP languages refer to tooth decay by reflexes of PMP *bukbuk ‘weevil that infests wood, bamboo, and rice; dust produced by the boring of this insect; tooth decay’ (ACD) (§9). Codrington also reports the belief as existing in the Banks Islands (NCV) (1891:193).

It seems likely that POc *mʷata-mʷata, the reduplicated form of *mʷata ‘snake’, was a generic for worms and worm-like creatures, including caterpillars and millipedes. In places it has reverted to its unreduplicated form, either as a generic for snakes and other snake-like creatures, or, as in Micronesia and Seimat, to refer to worms alone. In Micronesia there are no snakes except on Guam, where they have been introduced in comparatively recent times (Jeff Marck, pers. comm.), and hence there is no need to maintain the distinction. I have not been able to ascertain if snakes are found in the Ninigo Islands, where Seimat is spoken.

POc		*mʷata-mʷata	‘generic for worms and worm-like creatures’
Adm	Seimat	wat	‘land worm’
NNG	Kove	mota	‘snake, eel, caterpillar’
NNG	Bing	mut-muat	‘caterpillar’
NNG	Yabem	moaʔ	‘snake, and as first element in compounds (worm caterpillar, slug, snail and millipede)’
PT	Gapapaiwa	sulata(ramoa)	‘caterpillar, worm (not earthworm)’
PT	Bwaidoga	mota-mota	‘grub, caterpillar which eats leaves of yams, taro etc.’
PT	Sudest	mʷata(wadi)	‘a small black millipede’
PT	Minaveha	mʷata-mʷata	‘caterpillar, worm, maggot or small snake’
PT	Dobu	mʷata-mʷata	‘earthworms, intestinal worms’
PT	Molima	mʷata-mʷata	‘intestinal worms’
PT	Kilivila	mʷateta	‘worm’
PT	Gumawana	moteta	‘snake, worm’
SES	Sa’a	mʷā-mʷā	‘maggot, worm’
PEOc		*mʷata	‘snake, worm’
SES	’Are’are	mā	‘generic term for snake, worm’
SES	Kwaio	wā	‘snake; various worms, various snake-like creatures’
SES	To’aba’ita	wā	‘worm, grub, maggot, caterpillar, larva’
SES	Lau	wā	‘(1) snake; (2) worm, maggot’
Mic	Kiribati	mʷata	‘grub, caterpillar, worm’
Mic	Mokilese	mʷac	‘worm’
Mic	Carolinian	mʷata	‘worm, earthworm’
Mic	Woleaian	mʷat	‘underground worm’
Mic	Puluwatese	mɔho	‘worm, general name (incl. intestinal worms)’
Pn	Māori	ŋata	‘snail, slug, leech’

22.2. Other worm taxa⇫

Blust has reconstructed PMP *bulati ‘roundworm, ascaris, intestinal worm; also earthworm?’ (ACD) without Oceanic reflexes. I have reconstructed POc *sulati ‘worm’ which evidently carries a common root, -lati. The Oceanic evidence does not permit further differentiation of meaning.

POc		*sulati	‘worm’
PT	Sudest	ula-ulari	‘worm’
NCV	Tamambo	sulati	‘earthworm’
NCV	Mota	sulate	‘worm’
NCV	Mwotlap	ni-slat	‘worm’
NCV	Raga	silosi	‘general term for worms incl. earthworms and intestinal worms’
NCV	Paamese	a-silati	‘worm’
NCV	Nguna	a-sulati	‘hookworm, tapeworm’

The next reconstruction, PNGOc *dimʷan ‘k.o. worm’, bears some similarity to POc *droman ‘leech’ but a contrast is preserved in some languages, e.g. Molima domana ‘mountain leech’, dimʷa-mʷana ‘worm’; Dobu domana ‘leech’, di-dimʷana ‘common earthworms, small lizards’; Gapapaiwa domani ‘leech’, dimo(moga) ‘earthworm’; Sinaugoro doma ‘leech’, dimo ‘worm’). The resemblance appears to be accidental.

PNGOc		*dimʷan	‘k.o. worm’
NNG	Tami	timʷa	‘worm, maggot’
NNG	Numbami	(mota)diŋana	‘worm’ (mota ‘snake’)
PT	Molima	dimʷa-dimʷa-na	‘earthworms’
PT	Dobu	di-dimʷa-na	‘common earthworms, small lizards’
PT	Sewa Bay	dimʷana-na	‘worm’
PT	Duau	di-dimʷana	‘worm’
PT	Gapapaiwa	dimo(moga)	‘earthworm’
PT	Gumawana	dimona-na	‘an earthworm’
PT	Muyuw	simʷa(mʷat)	‘maggots, worms’ (mʷat ‘snake’)
PT	Sudest	jimo-jimo	‘earthworm’
PT	Sinaugoro	dimo	‘worm’
PT	Hula	rimo	‘worm’
PT	Kuni	sima	‘worm’

PPn		*kele-mutu	‘earthworm, grub’ (pollex: PPn *kele ‘earth, dirt, soil’)
Pn	Niuean	kele-mutu	‘worm’
Pn	Tongan	kele-mutu	‘earthworm, grub’
Pn	Samoan	ʔele-mutu	‘grub in rotten wood’
Pn	Tuvalu	kele-mutu	‘earthworm, grub’
Pn	Tokelauan	kele-mutu	‘earthworm, grub’

cf. also:

Adm

Loniu

(te)keli-mʷet

‘earthworm, grub’

See also chapter 4, §7.1 and §7.2, for POc *ibo ‘sandworm’ and POc *weli ‘fireworm’ respectively.

23. Maggots⇫

POc *quloc continues the PAn term for ‘maggot’, *qulej. Some reflexes show /i/ instead of /u/ in the first vowel, a common sporadic change in many forms.

PAn		*qulej	‘maggot’ (ACD)
POc		*quloc	‘maggot’
Adm	Lou	kul	‘maggot’
Adm	Loniu	un	‘maggot’
Adm	Titan	ul	‘maggot’
PT	Motu	ulo-ulo	‘maggot’
PT	Lala	ulo(li)	‘worm’
PT	Suau	uro-uro	‘fly’
MM	Roviana	uloso	‘maggot’
MM	Roviana	ul-uloso	‘maggot (pl)’
MM	Maringe	n-ulho	‘maggot’
SES	Bugotu	ulo	‘maggot’
SES	Arosi	uro	‘worm, maggot’
NCV	Mota	ulo	‘maggot’
NCV	Raga	ulehi	‘small worm’
NCV	Tape	wiləs	‘maggot’
NCV	Namakir	ʔil	‘maggot’
SV	Sye	n-ilah	‘maggot’
SV	Anejom̃	n-ija	‘maggot’
NCal	Cèmuhî	únet	‘maggot’
NCal	Caaàc	kōlet	‘maggot’
Mic	Kiribati	ino	‘a worm, larva’
Mic	Puluwatese	wūl, yīl	‘maggot’
Mic	Kosraean	ulɛ	‘maggot’
Mic	Woleaian	iṛ	‘maggot, larva’
Fij	Wayan	ilo	‘maggot’
Fij	Bauan	ulo	‘maggot’
Pn	Samoan	ilo	‘maggot’
Pn	Tahitian	iro	‘generic for maggots, worms’
Pn	Marquesan	iʔo	‘worm, maggot’
Pn	Māori	iro	‘maggot, threadworm, vermin’
Pn	Hawaiian	ilo	‘maggot, grub; to creep, as worms’

The next reconstruction illustrates what was evidently a productive process in POc, the addition of -a(n) to a term X, where X could be water, sennit, people, pigs etc., to indicate a state of being ‘full of X’ or in the case of insects, infested with the insect. Besides POc *qulos-a(n) ‘be maggoty’ I can reconstruct PEOc *kutu-a(n) ‘(be) lousy’ (To’aba’ita ʔu-la, Samoan ʔutu-a, Pukapukan wutu-a). Further examples from SES languages show the process applied to other insect names: Sa’a pote ‘head louse’, pote-la ‘lousy’; To’aba’ita, Lau wā ‘maggot, worm’, wā-la ‘maggoty, wormy’, and I add a Polynesian example, Tongan moŋomoŋa-ʔia ‘infested with beetles or cockroaches’, from moŋomoŋa ‘any kind of beetle or cockroach’.

PMP		*qulej-an	‘have worms’ (ACD)
POc		*quloc-a(n)	‘be maggoty’
SES	Bugotu	ulo-a	‘maggoty’
NCV	Mota	ulos-a	‘maggoty, full of maggots’
Mic	Kiribati	(ka) ino-a	‘bring on putrefaction’
Fij	Bauan	ulo-ulo-a	‘wormy, maggoty’
Pn	Samoan	ilo-a	‘be maggoty’

Reduplication on verbs serves to indicate frequentative aspect, while its nominal counterpart means ‘full of, lots of N’.

PMP		*qulej-qulej	‘lots of worms; crawling with worms, extremely wormy’ (ACD)
POc		*qulo-quloc	‘full of maggots’
PT	Motu	ulo-ulo	‘maggot’
MM	Roviana	ul-uloso	‘maggots, when in numbers’
Mic	Kiribati	ino-ino	‘full of worms, worm-eaten’
Fij	Bauan	ulo-ulo	‘maggot (diminutive)’

24. Unspecified flying insect⇫

The following set consists of cognates with a range of glosses which can only be subsumed under one generalised reconstruction.

PPn		*mū	‘flying insect’ (pollex)
Pn	Tongan	mū	‘moth’
Pn	East Futunan	mū-mū	‘dragonfly’
Pn	Samoan	(seʔe)mū	‘dragonfly’ (seʔe ‘glide’)
Pn	Rarotongan	mū	‘big reddish moth that flies round lamps at night’
Pn	Māori	mū	‘insects’
Pn	Hawaiian	mū	‘general term for insects that eat cloth, wood, plants’

25. Semantic extensions and generalisations⇫

Chapter 8 contains a detailed examination of *manuk and *manu-manuk. POc *manuk evidently had two senses, one restricted, the other extended. The restricted sense was ‘bird’ or more precisely ‘bird + bat’, hence ‘flying vertebrate’. The extended sense was ‘creature’, probably excluding people and things that swam in the sea. In POc, diminutives were regularly formed by reduplication. So *manu-manuk meant either ‘small flying creature’, or simply ‘small creature’. Beyond that I have only fragments of information to go on as clues to how POc speakers classified their world of insects and other creepy-crawlies. It is likely that most languages had generic terms for butterflies, flies, spiders, hoppers, ants, grubs and so on, although I have been able to reconstruct reliable generic terms for only the first two.

There is some evidence that there were two broader categories in PPn, the flying creatures, *manu-lele and the crawlers, *manu-totolo. Rennellese has manu-manu-gege ‘flying/jumping creatures’, and manu-manu-totogo ‘creeping creatures’, the latter including reptiles other than turtles. Nanumea, a dialect of Tuvalu (Pn), has forms cognate with the Rennellese, manu lele ‘bird, flying insect’ and manu totolo ‘ground animal e.g. spider, lizard, land mammal’. Niue also has cognate terms but insects are excluded in both: manu lele ‘bird, can include flying fox but not flying insects’ and manu totolo ‘any walking or crawling creature but not insects’. Speakers of the Hula dialect of Keapara (Papuan Tip) distinguish manu-manu ‘flying insect’ and mani-mani ‘creeping insect’. According to Fox (1974), Lau (SES) divides the entire animal kingdom three ways, with wāwā ‘generic for all creatures in or on the ground’, manu ‘generic for all creatures that fly’ and iʔa ‘generic for all creatures in water’.

The following example illustrates how one particular term may serve as the name of a class, in this case ‘fly’, and also be used as the first element in binomials defining members of that class. Examples include:

Adm	Lou	laŋ-laŋ(a)	‘fly, flying insect’
Adm	Lou	laŋ-et	‘housefly’
Adm	Lou	laŋ-laŋa-n palawa	‘honey bee’
Adm	Lou	laŋ-laŋ ŋara	‘bluebottle, horsefly, March fly’
NNG	Sengseng	laŋ	‘fly’
NNG	Sengseng	laŋiyuyu	‘hornet with papery nest’
NNG	Sengseng	laŋoŋoŋ	‘insect that makes holes in wood’
MM	Tolai	laŋa	‘fly’
MM	Tolai	laŋa bulit	‘bee’ (bulit ‘gum’)
SES	Gela	laŋo	‘fly’
SES	Gela	laŋo mbeli	‘k.o. hover fly’
SES	Gela	laŋo ni bolo	‘k.o. large fly’ (bolo ‘pig’)
SES	Gela	laŋo ni uvu	‘k.o. red fly’
SES	Gela	laŋo pasa-pasa	‘k.o. large fly’ (pasa-pasa ‘to scold’)
SES	Gela	laŋo vuru	‘large k.o. fly with yellow wings’
Pn	Niuean	laŋo	‘common fly’
Pn	Niuean	laŋo fufu	‘mason wasp’
Pn	Niuean	laŋo meli	‘bee’
Pn	Hawaiian	nalo	‘the common housefly and other similar flies’
Pn	Hawaiian	nalo ʔaki	‘small stinging fly’
Pn	Hawaiian	nalo hope ʔeha	‘hornet’ (lit. ‘fly with stinging posterior’)
Pn	Hawaiian	nalo keleawe	‘hoverfly?’ (lit. ‘brass/copperfly’)
Pn	Hawaiian	nalo lawe-lepo	‘mud wasp’ (lit. ‘dirt-toting fly’)
Pn	Hawaiian	nalo meli	‘honey bee’
Pn	Hawaiian	nalo nahu	‘stinging fly’ (lit. ‘biting fly’)
Pn	Hawaiian	nalo paka	‘tick; ensign fly’ (lit. ‘lean fly’)
Pn	Hawaiian	nalo pilau	‘bluebottle fly’ (lit. ‘stink fly’)

In some Polynesian languages the reflex of POc *moko ‘k.o. snake’ (p.260) has become a class marker for a whole range of creepy-crawlies:

Pn	Niuean	moko	‘insect, grub, caterpillar; lizard, any creeping creature’
Pn	Niuean	moko elo	‘stinkbeetle’ (elo ‘to stink’)
Pn	Niuean	moko manini	‘common brown beetle’ (manini ‘a plant, Physalis angulata’)
Pn	Niuean	moko maŋa	‘earwig’ (maŋa ‘forked, divided’)
Pn	Niuean	moko moŋa-moŋa	‘cockroach’
Pn	Niuean	moko niu	‘rhinoceros beetle’ (niu ‘coconut palm’)
Pn	Niuean	moko tafa	‘blackbeetle’ (tafa ‘be itchy’)
Pn	Niuean	moko taliŋa	‘gecko’ (taliŋa ‘ear’)
Pn	Niuean	moko vā	‘large green mantis’ (vā ‘to mock’?)
Pn	Tongan	moko	‘k.o. lizard, light-brown in colour’
Pn	Tongan	moko hula	‘earwig’ (hula ‘protuberant’?)
Pn	Tongan	moko hunu	‘sea slug’ (hunu ‘to singe’?)
Pn	Tongan	moko-moko tuʔaniu	‘mantis, stick insect’ (tuʔaniu ‘midrib of coconut leaflet’)
Pn	Tongan	moko tola	‘shellfish’ (tola ‘snout’)
Pn	Hawaiian	moʔo	‘lizard, reptile of any kind’
Pn	Hawaiian	moʔo huelo-ʔawa	‘scorpion’ (lit. ‘lizard with bitter tail’)
Pn	Hawaiian	moʔo niho-ʔawa	‘scorpion, viper, asp’ (lit. ‘lizard with bitter tooth’; from a Bible translation)

In some Polynesian languages, small lizards and certain other creepy crawlies are lumped together in a higher order taxon. POc *mʷata-mʷata ‘worm’ is the diminutive of *mʷata ‘snake’.¹¹ Note also Dobu (PT) didi mʷana ‘common earthworms, small lizards’.

Similarities in behaviour form the basis of other groupings. In Tamambo (NCV), the reflex of POc *(ñ,n)opuq ‘stonefish’ refers to ‘centipede or scorpion’ as well as ‘venomous fish’. The semantic link is clearly ‘painful stinger’. The Polynesian language of Rennell has extended the meaning of kutu (from POc *kutu ‘head louse’) to refer to ‘louse, bird louse, mite, leech and tiny insects of various kinds’. The term has presumably come to mean ‘small blood-sucker’, when no more specific reference is necessary. In Marquesan the reflex of POc *quloc ‘maggot’, iʔo, now refers to bugs and worms.

Some languages have developed and named unusual groupings. For instance, the Motu have a term, kikiri kikiri, which refers to ‘all creatures which live on the ground but jump or fly short distances. Cicadas and hawk moths are kikiri kikiri. So are beetles that make noise’ (Nigel Oram n.d.). In Kwaio, there is a collective term, faʔa/fulu, for edible insects.

26. Stability and variability of terms⇫

A small number of POc insect names show far greater stability than others. In a survey of 25 dictionaries chosen on comprehensiveness and whose languages together covered all major subgroups,¹² reflexes of *kutu ‘louse’ were listed in 23, *laŋo ‘fly’ in 22, *ñamuk ‘mosquito’ in 21, *bebek ‘butterfly’ in 16, and *qapator ‘sago grub, edible’ in 13. In an email comment on seahorse terms Robert Blust wrote that ‘The general principle for flora and fauna seems to be that if it was economically useful or dangerous it was named, and the more useful or dangerous the more stable the name historically.’¹³ With insects, it seems that the more intrusive in everyday life, the more commonplace and either pesky or valued they are, the more likely they are to be known by highly stable terms.

In contrast, it seems that unstable terms may be unstable for a variety of reasons. We have a number of instances where it has proved very difficult to make any well-supported reconstructions, and others where two or more reconstructions that differ only very slightly in form have been made for an insect. Examples are:

butterfly	POc *[kau]bebek, *kali-bobo(ŋ), PPT *qara-bembem, Proto Northwest Solomonic *pepele
mosquito	POc *ñamuk, *simuk
sandfly	POc *niku-niku, *nonok, *ŋisi
mason wasp	POc *pupuk, *bubu
wasp	POc *mañipo, *(n,ñ)i(n,ñ)ipo, Proto Central Papuan *naniɣo
grasshopper	POc *sakʷa, *siko; POc *riŋa, PROc *liŋo
ant	POc *drui, *[driu]driu PWOc *didi
spider	POc *ka(k,g)a, *koko, PWOc *kuku, PEOc *koko-mici

The existence of taboo to limit use of a particular form to a particular area, has already been suggested for ‘butterfly’ in §3, and may explain why this creature can have a name that is very stable in most of the Oceanic region, yet be unstable in parts of Western Oceanic.

Clearly, onomatopoeia plays a role in the naming of the humming and buzzing insects, where there is frequent interchange of n, ñ and ŋ. Examples include POc *ñamuk, PWOc *ŋati-ŋati, both ‘mosquito’, POc *ŋiŋi(ŋ) (1) ‘buzz as a mosquito’, (2) ‘mosquito’, *(n,ñ)i(n,ñ)i ‘bee, buzzing insect’.

Irregular patterning of terms for insects has been attributed to the fact that many insects are ‘expressive’ creatures. That is, they are associated with particular characteristics that impinge strongly on the human sensory system and become a significant identifying feature for that insect. Characteristics include sound (crickets, cicadas, the buzzing insects), movement (grasshoppers, butterflies, spiders), ability to sting or bite (mosquitoes, sandflies and other stinging flies, centipedes, scorpions, ants) and possibly other aspects of appearance or behaviour. It has been claimed that it is their ‘expressive’ condition that underlies a tendency for speakers to play phonological games with their names, resulting in irregularities that may then defy reconstruction. This tendency has been observed cross-linguistically by Erik Fudge (1970) in words of certain semantic type including onomatopoeia or movement, and in the names of plants and animals in Balinese and the closely related Javanese by Adrian Clynes (1995:44).

27. The supernatural element⇫

Among my POc reconstructions, those carrying the *kali-/*qali- prefix are limited to *[kali]-bobo(ŋ) ‘butterfly’, *qalipan ‘centipede’ and *qali-popot/*(k,q)ali-totop ‘firefly’. Ethnographic evidence from a number of communities indicates that all three creatures retain some supernatural association requiring them to be treated with caution. I also have a number of reconstructions at POc level or lower which may carry the prefix or some trace of it. They include PPn *kalisi ‘cicada, cricket’, a number of ant terms possibly reflecting POc *kal(a,o), two lizard terms, PSOc *qala ‘green lizard, Emoia sp.’, PEOc *kalis(i,u) ‘skink, k.o. lizard’ (see ch.5, §5.3.5), and a number of terms for ‘spider’ which carry the prefix ka-.¹⁴ However, if my reconstructions do in fact carry a trace of the danger prefix, they have lost what was evidently an essential element of *kali-/*qali- words at a stage earlier than POc, that is, conversion from the typical disyllabic base form to a quadrisyllable one. The prefix had ceased to be productive in POc.

Nonetheless, it seems that in Polynesian languages there is a different term which can be used to mark a creature as supernatural. PPn *qatua (from PMP *qatuari), often glossed ‘god, deity’, is used primarily to signify that an object, whether human or non-human, has supernatural qualities. In Williams’ Maori dictionary, the definition of atua runs ‘god, demon, supernatural being, ghost; object of superstitious regard; anything malign, disagreeable; strange, extraordinary’.¹⁵ Like qali-/kali-, its range in Polynesian languages extends to inexplicable phenomena such as rainbows and waterspouts.¹⁶

Examples of its use in Polynesia include Maori atua-piko ‘rainbow’ (cf PT: Kilivila kali-pedoga ‘rainbow’), Tongan ʔotua-kui ‘whirlwind, waterspout’ (cf MM: Nakanai kali-vuru ‘tornado, waterspout’ and MM: Tolai kali-vuvur ‘whirlwind’), and a small number of insect terms, Samoan atua-loa ‘k.o. centipede with poisonous bite’ (cf POc *qali-pan ‘centipede’) Maori pepe atua ‘northern wattle moth’ (literally ‘supernatural butterfly’) (cf. POc *[kali]bobo(ŋ) ‘butterfly’) and two Rennellese terms, ʔatua segesege baʔe ‘a large grasshopper’ (lit. ‘leg-cutting supernatural’) and ʔatua-seu ‘large stinging stick insect’ (lit. ‘lame supernatural’).

A detailed account of the way in which atua is linked to natural species in one community is given by Firth (1967:233). In Tikopia, as in other Polynesian communities, atua applies primarily to a supernatural object, whether human or non-human in form. However, it is applied also to any natural species which should not be eaten. ‘The terms atua and kai (food) in this connection are in fact mutually exclusive. The [former] applies to small crabs, sea anemones, bêche-de-mer and other marine creatures on the reef, as well as to iridescent lizards, spiders and some insects.’ (p.233). He suggests ‘that the connotation of atua as inedible may be derivative from its significance of supernatural’ (p.234), (italics mine, MO). He concludes that ‘one point, however, may be stressed with certainty, that any object which is regarded as an atua may not be eaten, and anything which is fit for human consumption cannot be in itself an atua — though it may … become temporarily associated with atua’ (p.234). Among the creatures he lists as atua are two inedible crabs, kaviki ‘small pale land crab’ and karamisi ‘reddish-brown or yellowish crab, [which] lives in crotch of tree’ (cf PMP *qali-maŋaw_ ‘mangrove crab’). Also atua are moko, the black lizard, and morokau, the centipede. In Tikopia, then, atua is not affixed to the names of natural species as it is in other Polynesian communities and in the manner of qali-/kali-, but is rather the name of a category.

The association of certain insects with the supernatural has survived from PAn society right through to certain contemporary Polynesian societies. Although the qali-/kali- label has evidently lost its force in Oceanic languages, the existence of reflexes of *qatuan in a similar capacity is a revealing linguistic clue to certain beliefs of POc speakers.

Notes⇫