h   

hair

hair:   lock of hair

hair:   lose hair or feathers

Tiruray ufaw is assumed to be a GCPh loan; the relationship of Malagasy ofo to the Philippine forms is attributed to chance.

hand over

Chance or borrowing.

hand:   scrape or scratch with the hand

Chance. Dempwolff (1938) posited ‘Uraustronesisch’ *Ta(ŋ)kur ‘scrape or scratch with the hand’ (mit der Hand scharren). In addition to the above forms he included Tagalog taŋkol ‘hand movement’ (Handbewegung), but no such word appears in either Panganiban (1966) or English (1986), and I am unable to find corresponding forms in other Philippine languages.

hand arm

Chance. This is yet another example of the many forced comparisons in Dempwolff (1938). The Ngaju Dayak form clearly reflects PMP *leŋen ‘forearm, lower arm’. The loss of word-final -n in obligatorily possessed nouns has led to reanalysis in other languages of Borneo, as in Uma Juman Kayan, where the reflex of PMP *ipen ‘tooth’ in constructions such as ipen-naɁ ‘his/her tooth’ has be reinterpreted as ipeɁ-naɁ ‘his/her tooth’ (with obligatory morpheme-final glottal stop). The Fijian word is most likely a reflex of POc *lima ‘hand’, with a sporadic replacement of *m by its labiovelar equivalent, as with Mota limwa ‘hand; five’. Dempwolff ignored the irregularity in the vowel correspondences of both syllables, and proposed Uraustronesisch *liŋa ‘hand, arm’ despite the availability of far better-supported words for this meaning.

harbor bad feelings toward

hard

Based on a slight variant of this comparison Dempwolff (1934-1938) reconstructed ‘Uraustronesisch’ *makas ‘hard’, but *maka appears to be a Tongic innovation that replaced Proto-Polynesian *fatu ‘stone’, and so has little time-depth even within Polynesian.

hare-lip

hare-lip

harm:   evil, harm, disaster

Dempwolff (1938) proposed this comparison, which appears to be a collection of unrelated forms in all three languages. I am unable to find sona in Richardson (1885), and the closest form I can find to his so-sol is sol ‘uprooted (by storm)’ in Pigeaud (1938).

hatch:   to hatch

Chance. For Motu papa-ia cf. note to *bajbaj.

have an angry facial expression

Although both of these bases could reflect POc *fula, the absence of supporting evidence from other languages, and the occurrence of the base only with another word or with affixation raises suspicions that this rather striking resemblance is a product of chance.

having gaps

Dempwolff (1938) reconstructed *siraŋ ‘to form gaps’, but this comparison appears to be a forced collection of unrelated forms.

head covering

Dempwolff (1938) compared the Tagalog, Malagasy and Toba Batak forms given here, and proposed ‘Uraustronesisch’ *panduŋ ‘head covering; crest’. The similarity between these forms now seems better treated as due to chance, and while the Philippine forms are clearly cognate and share the monosyllabic root *-duŋ ‘shelter, protect’, the Cordilleran words appear to be Tagalog loans.

heap

heap of earth

heap:   hoard, heap

The Fijian and Polynesian forms may be cognate, though borrowing cannot be ruled out (Churchward 1959 does not list Tongan ʔafuha, and the anticipated form in any case would contain first-syllable //o//, not //a//). The relationship of the Malagasy and Sa'a forms to these and to one another is regarded as convergent (cf. Lau ʔafuta 'all, every', Arosi ʔahuta-na 'all, the whole of, all together', and similar forms in other Southeast Solomonic languages which indicate earlier *kafusa).

heartwood

hernia

Borrowing from Malay, except for Dairi-Pakpak Batak burut-en, which is best attributed to chance.

heron

hidden

high sea or tide

Osmond, Pawley and Ross (2003:95) cite an equivalent, but slightly longer comparsion in support of POc *loka ‘high sea or tide, heavy breakers’. However, the POc status of the reconstruction they propose depends crucially on the Lou example, and since Lou regularly loses all final vowels that were not protected by an obligatory suffix the similarity of Lou loka to the other forms cited by them is best attributed to chance.

hindrance

On the basis of this rather unpromising comparison Dempwolff (1938) posited ‘Uraustronesisch’ *cawad ‘to hinder’. I have been unable to find cawad in any Javanese dictionary available to me, or in Old Javanese.

hiss

Probably a chance resemblance.

hit:   collide, hit, strike

The Philippine forms are unrelated to those in western Indonesia. Of the latter only Bidayuh (Bukar-Sadong) and Malay, or Malay and the Batak languages permit a comparison, and this cannot safely be attributed to PWMP.

hit, thump

Chance (based on a common root *-bek 'sound of breaking, etc.').

hoard, heap

The Fijian and Polynesian forms may be cognate, though borrowing cannot be ruled out (Churchward 1959 does not list Tongan ʔafuha, and the anticipated form in any case would contain first-syllable //o//, not //a//). The relationship of the Malagasy and Sa'a forms to these and to one another is regarded as convergent (cf. Lau ʔafuta 'all, every', Arosi ʔahuta-na 'all, the whole of, all together', and similar forms in other Southeast Solomonic languages which indicate earlier *kafusa).

hold firmly

hole:   bore a hole, drill

hole

holes:   round (holes)

Chance. Despite the striking formal and semantic similarity of these forms the correspondences are irregular (Kadazan Dusun o points to *e, Tiruray u to *u). The Tiruray form shows irregularities associated with Danaw loans, but a cognate word is yet to be identified in a Danaw language (or, for that matter, any other language).

hornbill

The Malagasy form appears to be a chance resemblance, and the Toba Batak form could easily be a loan from Malay.

house:   beam, house end

hurl:   sling, hurl

Late innovation, with some borrowing for Malay, Toba Batak, Dairi-Pakpak Batak, Sundanese, Old Javanese; chance for Gilbertese.

hut

This comparison, first proposed without the Bikol member in Blust (1970), has long been problematic because the Philippine evidence points to *-q, but the Iban evidence is incompatible with this (Blust 2009:561-568). There is a very strong temptation to assume cognation here, but on balance the similarity of these forms is perhaps best attributed to chance.