This chapter collects together terms that people use to describe each other and to express their own feelings. It includes terms used to describe
This is quite a mixed bag, both semantically and formally. Terms of temperament, emotion and wanting normally describe only human beings and sometimes higher animals, but terms used to describe physical qualities or to express an evaluation are applied to a wider range of referents that includes inanimates. Terms of temperament, emotion and wanting are quite often BPMs, as they denote human affective states. The range of meanings found in the headings in this chapter is somewhat restricted. For example, we tried to reconstruct terms for ‘lazy’ vs ‘hard-working’ and ‘generous’ vs ‘mean’, but could not do so because cognate sets are at best very local. One reason for this is certainly that speakers are creative in their use of metaphor, and so one term or metaphor is easily replaced by another. An associated reason is that many of our sources do not record the complex lexemes that result from metaphorical usage. Although BPMs occur for various aspects of wanting and desire, none are recorded in §11.5 because there is little conceptual similarity among languages.
Among the semantic fields in which BPMs are used, fewer are found in the description of temperament than in the labelling of emotions. The reason for this is almost certainly that temperament, like body shape (§11.2), is fairly stable through adulthood, and stable qualities tend to be labelled by single words. Emotions and desires, on the other hand, are essentially changeable, and are often described metaphorically. The boundary between temperament terms and emotion terms, however, is fuzzy. Saying that someone is brave may refer to someone’s temperament or to their current behaviour. In the latter case a BPM is expected, so that a BPM like POc *qate- (p,pʷ)atu(k) [liver- strong/firm] ‘brave’ (§11.3.2.1) was basically an expression used of someone’s immediate behaviour that was also extended to a stable propensity of temperament.
Languages vary as to whether or not they lexicalise a distinction between horizontal and vertical length. English does so by distinguishing between horizontal long and vertical tall, but the antonym of both terms is short. German and French equate vertical length, at least in its application to human beings, with size: German groβ, French grand ‘big, tall’ vs German klein, French petit ‘small, short in stature’.
Oceanic languages mostly agree in making no distinction between horizontal and vertical length. The term for ‘long’ also means ‘tall’ and the term for ‘short’ has both horizontal and vertical application. Both apply to the stature of human beings. The relevant terms are reconstructed in vol.2(197–199). In no Oceanic language for which data are available is human stature equated with size (§11.2.2).
It follows from the last statement that terms with the general meaning ‘big’ (POc *lapuat) and ‘small’ (POc *qitik, *riki(t,q)), reconstructed in vol.2(191–196), do not denote stature when they refer to human beings. To assert that someone is ‘big’ is typically to say that this person has social prestige, and this was probably also true in POc. The nature of that prestige depends on community structure. In chiefly societies, a ‘big’ man is a chieftain.
| POc | *lapuat | ‘big, large; chief’ | |
| Adm | Mussau | lapa-n | ‘important person, chief’ |
| Adm | Baluan | lapa-n | ‘chief; excellent’ (Schokkin 2014) |
| Adm | Titan | lápa-n | ‘leader, chief’ |
| Adm | Nyindrou | laba-n | ‘leader, chief’ |
| Adm | Papitalai | laba-n | ‘chief’ |
| MM | Mono-Alu | la-lafa | ‘chief’ |
| PEOc | *qa-lapʷa | ‘chief’ | |
| SES | Lau | alafa | ‘chief’ |
| SES | ’Are’are | a-raha | ‘chief’ |
| SES | Arosi | a-raha | ‘chief’ |
| MM | Tinputz | abuh | ‘big, huge; chief’ |
In non-chiefly societies a ‘big’ man is an important person, perhaps by virtue of leadership qualities, perhaps by virtue of accrued wealth.
| Adm | Titan | manr̃ean | ‘big; important’ |
| NNG | Takia | tubun | ‘big; important’ |
| NNG | Mangseng | pom | ‘big; important, prestigious’ |
| NNG | Numbami | bamo | ‘big, large; elder’ |
| NNG | Hote | bɛŋ | ‘big; deep; thick; wide; important’ |
| NNG | Patep | lɛvaʔ | ‘big; important’ |
| PT | Dobu | sinabʷa-na | ‘big, large; important’ |
| PT | Iduna | lakahi-na | ‘big; “big” man’ |
| PT | Sinaugoro | barego | ‘big; important’ |
| MM | Nakanai | uru | ‘big; senior; important’ |
| MM | Nehan | uleiki | ‘big, large; important’ |
| MM | Roviana | ululu-na | ‘big; important man’ |
| SES | To’aba’ita | baʔita | ‘big; important’ |
| SES | ’Are’are | paina | ‘big, large, great, loud; (man) in authority’ |
| NCV | Paamese | marīte | ‘big; important’ |
| NCV | Paamese | heitamen | ‘big; important’ |
| NCV | Lewo | keviu | ‘big; important’ |
A secondary but much less frequent association of ‘big’ is with age: ‘big’ equates with ‘older’.
‘Small’ in Oceanic languages is the antonym of ‘big’ with regard to size, but not in its extended meanings. The expected antonym of ‘chiefly, important’ would be ‘socially inferior’, but this usage appears to be very rare. Instead, ‘small’ seems far more frequently used of a child, that is, with regard to age. However, it is difficult to be sure about this, because a gloss like ‘young, small’ does not tell us that the word is used for small objects in general, as the translator may simply have ‘young child, small child’ in mind.
Oceanic languages typically distinguish between (a) the girth (circumferential thickness) of both human beings (English fat or stout) and non-human objects (e.g. a thick stem) and (b) the thickness of something with a flat surface, like paper, or the depth of a large body of water.
Two terms reconstructed for ‘thick’ in vol.2(201–202) concern us here, POc *ma-tolu and POc *[tubu]tubu[ka]. Glosses that disambiguate the meaning of *ma-tolu point in just one instance (Manam) to girth and in a majority of cases to the thickness of something flat, and it is reasonable to infer that this was its central POc sense.
| POc | *ma-tolu | ‘thick (of flat objects)’ (vol.2:201–203) | |
| NNG | Manam | ma-toli | ‘fat’ |
| NCV | Mota | ma-tol-tol | ‘thick, thick-skinned, callous; broad, thick (of speech)’ |
| NCV | Paamese | ma-te-tel | ‘thick; deep’ |
| SV | Anejom̃ | a-m-esej | ‘thick (of flat object)’ |
| Mic | Kosraean | mæ-tol | ‘thick, dense’ |
| Mic | Chuukese | ma-aɾɨ̄ɾ | ‘be thick (of flat objects)’ |
| Mic | Puluwatese | ma-aliyə̄l | ‘be thick (as paper), thickness’ |
| Mic | Woleaian | ma-alʉyelʉ | ‘be thick, close packed, dense’ |
| Pn | Ifira-Mele | mā-toru | ‘thick (of board etc.)’ |
| Pn | Rarotongan | mā-toru | ‘thick through, stout and solid’ |
| Pn | Rennellese | mā-togu | ‘thick, as a knife or a mat, crowded’ |
The POc term for a large girth was *[tubu]tubu[ka], reconstructed in vol.2:202.1 Additional reflexes have been found, and we give the cognate set as it now stands.
| POc | *[tubu]tubu[ka] | ‘thick (in dimension); fat (of vegetable, fruit, human being)’ (vol.2:202) | |
| NNG | Yabem | tʊp | ‘grow fat’ |
| NNG | Sio | tuᵐbu | ‘fat; gain weight’ |
| NNG | Numbami | -tuᵐbu | ‘grow, get fatter’ |
| MM | Patpatar | tubu | ‘fat’ |
| MM | Ramoaaina | tubu | ‘fat ; wide ; thick’ |
| MM | Ramoaaina | tubu-tubu | ‘very fat’ |
| MM | Tolai | tubu | ‘thick (in dimension)’ |
| SES | Lau | ūbu-ūbu-a | ‘thick, fleshy, in good condition’ |
| SES | Kwaio | ubu-ubu | ‘thick’ |
| SES | Arosi | ubu-ubu-ʔa | ‘thick’ |
| SES | Santa Ana | upu-pu-ɣa | ‘thick’ |
| Fij | Bauan | tubu | ‘grow or increase in size’ |
| PT | Misima | tabʷa | ‘grow well; (be) fat’ |
The term reconstructed for ‘thin’ in vol.2(202–203) is POc *manipis. Below are listed reflexes with disambiguating glosses, and it is clear that *manipis is the antonym of *matolu ‘thick (of flat objects)’ and thus not a term applied to human beings.
| PAn | *[ma]Lipis | ‘thin’ (vol.2:202–203) | |
| POc | *manipis | ‘thin (of flat objects), flimsy’ | |
| SES | Owa | manifi | ‘thin and transparent’ |
| NCV | Mota | mavin-vin | ‘thin; of speech, sharp (antonym of matoltol)’ (metathesis) |
| NCV | Ambae | manivi-nivi | ‘thin, shallow, low tide’ |
| NCV | Lewo | mani-nivi | ‘thin; shallow’ |
| NCV | Nguna | manive-nive | ‘thin (of an object, e.g. cloth, paper, sides of canoe etc)’ |
| Mic | Marshallese | maniy | ‘thin, flimsy’ |
| Mic | Ponapean | mɛnipi-nip | ‘thin (of flat objects such as paper)’ |
The data make it clear that Oceanic languages have a distinct term or terms for people who are thin/scawny/skinny, but no cognate set reflecting a POc term with this meaning has been reconstructed. Clark (2009) reconstructs a PNCV term that evidently had this sense.
| PNCV | *magura | ‘thin, lean’ (Clark 2009) | |
| NCV | Raga | magura | ‘thin’ |
| NCV | Uripiv | -mak-mak | ‘be thin’ |
| NCV | Port Sandwich | maᵑgü | ‘thin’ |
| NCV | Lewo | maula | ‘thin, bony’ |
| NCV | Namakir | mʷagir | ‘thin’ |
| NCV | Nguna | mʷagura | ‘thin, lean, slim (person), lean (meat)’ |
Oceanic terms for ‘strong’ tend also to mean ‘hard’. One POc term for ‘strong, hard’ was *toRas, reconstructed in vol.3(200–201) with the sense ‘a taxon of hardwood trees including Intsia bijuga’, where it is also noted that the POc term meant ‘hard, durable’. There is no evidence, though, that this term denoted human strength.
The three terms below, POc *kayu-kayu, PROc *[kayu]kayu-a and PEOc *kaila ‘strong, firm’ appear to be derived from the generic term for tree, POc *kayu (vol.3:71–73), used metaphorically for strength and hardness. However, this seemingly obvious derivation is a little problematic. If the reflexes of these three terms are compared with the terms for tree in the same languages, differences emerge. The *kayu-like terms for ‘strong, firm’ have a fortis initial k-, whilst in languages that have a fortis/lenis distinction2 the term for ‘tree’ has a lenis initial ɣ- (Bugotu, W Guadalcanal, Woleaian) or zero (Labu, Paamese). This appears to be evidence against the hypothesis that these terms are derived from *kayu. However, the history of the fortis/lenis distinction is not well understood, and in favour of the hypothesis is that POc *kayu-kayu and *[kayu]kayu-a were both formed from *kayu by two early Oceanic adjective-forming strategies. The first was CVCV- reduplication, giving POc *kayu-kayu ‘strong, tough, inflexible’ (vol.2:206–220). The second was the addition of *-[k]a to a noun (Ross 2000), giving *[kayu]kayu-a. The fortis/lenis and other formal discrepancies can perhaps be explained by assuming that speakers’ association of ‘strong, firm’ with ‘tree’ was lost in some languages, and as a result the root of the ‘strong’ term and the reflex of *kayu have undergone different phonological developments.
PEOc *kaila ‘strong, firm’ has a similar set of meanings, and was presumably also derived from *kayu, but the derivational mechanism has not been recorded in other terms.
| POc | *kayu-kayu | ‘strong, tough, inflexible’ | |
| NNG | Manam | kaikai | ‘strong’ (kai ‘tree; strong’) |
| NNG | Mangap | -keke | ‘stiff, strong, inflexible, proud, unwilling to serve’ (ke ‘tree’) |
| NNG | Sio | kaika | ‘strong’ (kɔe ‘tree’) |
| NNG | Labu | ka-ka | ‘hard’ (a ‘tree’) |
| PT | Muyuw | kei-kay | ‘hard, difficult’ (kay ‘tree’) |
| SES | Bugotu | ka-kai | ‘firm, steady, faithful’ (ɣai ‘tree’) |
| SES | West Guadalcanal | ka-kai | ‘strong’ (ɣai ‘tree’) |
| Fij | Wayan | kai | ‘wood, tree; strong, tough, powerful’ |
| Fij | Wayan | kai-kai | ‘hard, firm, rigid, stiff’ |
| PROc | *[kayu]kayu-a | ‘strong’ (Clark 2009: PNCV) | |
| NCV | Nokuku | (mel)kɔu | ‘strong’ (kɔu ‘tree’) |
| NCV | Tolomako | ɣau-ɣau-a | ‘strong’ |
| NCV | Unua | -xaiv | ‘hard, difficult, solid, strong, firm’ |
| NCV | Maskelynes | xai-xai | ‘strong’ |
| NCV | Paamese | keiho | ‘strong’ (a-ai ‘tree’) |
| NCV | Lewo | kawa | ‘strong; adult’ (la-ki ‘tree. stick, wood’) |
| NCV | Nguna | kasua | ‘strong, hard, difficult, loud’ (na-kau ‘tree’) |
| Fij | Bauan | kau-kau-a | ‘strong, hard’ (kau ‘tree’) |
| Pn | Tongan | kau-kau-a | ‘strong, sturdy, burly’ (kau ‘stalk, stem’) |
It is possible that Seimat [Adm] aila-n in the BPM patu ailan ‘he is obstinate’ means ‘hard, strong’. If so, then PEOc *kaila is elevated to POc status.
| PEOc | *kaila | ‘strong, firm’ | |
| SES | Sa’a | aʔaila-ʔa | ‘firm, strong’ (ʔai ‘tree’) |
| SES | Arosi | ʔaʔaira[ʔa] | ‘strong, firm’ (ʔai ‘tree’) |
| PMic | *kaila | ‘strong’ | |
| Mic | Kiribati | kaina(matoa) | ‘implacable’ (te-kaina ‘pandanus tree’) |
| Mic | Woleaian | kkaile | ‘strong, healthy’ (xaai ‘tree’) |
| Mic | Ponapean | kɛ̄l | ‘strength’ |
| Mic | Ponapean | kɛ̄l ayl | ‘strong, healthy, powerful’ (< PMic *kailaila) |
| Mic | Pingelapese | kēl | ‘strength’ (suhkae ‘tree’) |
| Mic | Puluwatese | kkel | ‘strong’ (yéé ‘house beam < POc *kayu ‘tree’) |
| NCV | Araki | ɣau-ra | ‘hard, solid; difficult, arduous, tough’ (< *kayu-ta; cf. ɣa ‘tree’) |
A further term for ‘hard, strong, firm’ is POc *(p,pʷ)atuatu ‘hard, strong, firm’, again an adjective formed by reduplication of a noun. The noun was *(p,pʷ)atu ‘outer shell, skull’ (Ch:- bodyparts, §4.2.1).
| POc | *[(p,pʷ)atu](p,pʷ)atu | ‘hard, strong, firm’ | |
| NNG | Takia | pat-pat | ‘strong, hard’ |
| PT | Dobu | patu-patu | ‘hard, of fat, sago’ |
| PT | Minaveha | vatu- | ‘strong, hard ??’3 |
| SES | Bugotu | patu | ‘hard, firm, taut’ |
| SES | Gela | patu | ‘hard, to make firm’ |
| SES | ’Are’are | pau-pau | ‘hard’ |
Two terms with an implication of human strength, both reconstructed by Blust (ACD), are recorded in vol.2(214): POc *paka(s) ‘have energy, strength’ and POc *laga(s) ‘spirited, energetic’.
Weak has at least two related senses in English when it is applied to human beings: ‘temporarily weak as a result of tiredness, sickness or hunger’ and ‘constitutionally weak’. It is the latter which relevant here. Just as ‘strong’ is associated with ‘hard’ in Oceanic languages, so ‘weak’ is associated with ’soft’. POc *[ma]lumu ‘soft, gentle, easy’ is reconstructed in vol.2(215). In its application to people, this term seems to have had two senses: with reference to the body, ‘constitutionally weak’, and with reference to temperament ‘gentle, easygoing’ (§11.3.1.1). In the listing below ‘weak’ is shown in bold when it occurs in a gloss.
| PMP | *[ma]lumu | ‘soft, tender, gentle’ (ACD) | |
| POc | *[ma]lumu | ‘soft (of objects); gentle, easygoing; (constitutionally) weak’ (vol.2:215) | |
| NNG | Bariai | marum | ‘soft, weak’ |
| NNG | Bukawa | malʊ | ‘peaceful’ |
| NNG | Yabem | malʊ | ‘calm, peaceful, good-natured’ |
| PT | Muyuw | manum | ‘gently’ |
| MM | Tinputz | mamarum | ‘weak (in body)’ |
| SES | Sa’a | malumu | ‘soft, gentle’ |
| PNCV | *ma-lumu | ‘soft, gentle, weak’ (Clark 2009) | |
| NCV | Mota | malum-lum | ‘soft, gentle’ |
| NCV | Nokuku | melum | ‘soft, slow, weak’ |
| NCV | Araki | m̫alum | ‘quiet, slow, weak’ |
| NCV | South Efate | mailum-lum | ‘quiet, slow; soft, weak’ |
| Fij | Bauan | malumu | ‘weak, faint, sick, soft’ |
| Fij | Wayan | malum | ‘weak, feeble (of a living thing)’ |
| Pn | Niuean | molū | ‘soft, humble, weak,’ |
| Pn | Samoan | malū | ‘soft, calm of sea, gentle’ |
| Pn | Tahitian | marū | ‘soft, gentle, easy’ |
Whilst Oceanic languages certainly have terms, some of them BPMs, for human propensities, many of these terms may refer to both temporary and permanent states. It is the permanent states that are properties of temperament, and some languages have a construction that encodes these. There is, however, no POc construction that can be reconstructed with this function.
Iduna, like a number of other Papuan Tip languages, makes copious use of to- ‘person who…’ (from POc *tau-; §2.2.1.2) in expressions like the following, where the item compounded with to-, typically a verb, encodes a quality construed as permanent.
| PT | Iduna | to-selakalaka | ‘boaster, arrogant person’ (-selakalaka ‘boast, brag, be proud’) |
| PT | Iduna | to-siveyawa | ‘healer’ (siveyawa ‘healing’) |
| PT | Iduna | to-taɣa-kʷeu | ‘deaf person’ (taɣa ‘ear’; -kʷeu ??) |
| PT | Iduna | to-talahaɣi | ‘person at enmity with s.o., not speaking to, eating with or visiting them’ (-talahaɣi ‘repudiate, hate, have nothing to do with s.o.’) |
| PT | Iduna | to-tubukoyo | ‘cripple’ (-tubukoyo ‘develop badly, not grow well’) |
| PT | Iduna | to-tunutunuɣina | ‘upright, honest person’ (tunutunuɣina ‘straight, upright, righteous, honest’) |
| PT | Iduna | to-nu-beʔu-beʔu | ‘lame person’ (beʔu ‘fall down’) |
In a number of these compounds the stem is itself a BPM formed with nua ‘think; mind’.
| PT | Iduna | to-nua-doɣa | ‘kind, generous person’ (nuanua-wadoɣa-na ‘gracious, generous’) |
| PT | Iduna | to-nua-hobu | ‘humble person’ (-nua-hobu ‘be humble’; -hobu ‘go down’) |
| PT | Iduna | to-nua-huya | ‘intelligent person, educated, wise, literate’ (nua-huya ‘wisdom, skill, knowledge’) |
| PT | Iduna | to-nua-kabubu | ‘kind, loving person’ (nua-kabubu ‘love, blessing, favour’) |
| PT | Iduna | to-nua-lolona | ‘miser, ungenerous person’ (-nua-lolona ‘withhold (food, possessions)’) |
| PT | Iduna | to-nua-luɣa | ‘double-minded person’ (-nua-luɣa ‘indecisive’; §10.8) |
| PT | Iduna | to-nua-sivebala | ‘agitator’ (-sivebalana ‘deter, prevent from going or doing, hold back’) |
| PT | Iduna | to-nua-vita | ‘downcast, depressed person’ (-vita ‘heavy’) |
Motu (PT) forms compound lexemes in a semantically parallel manner, but the attribute precedes tau-na, composed of tau ‘person’ and -na ‘its’. The attribute takes the prefix he-, indicating a state, condition or habit. he-abidae tau-na ‘a hospitable man’ (he-abidae ‘show hospitality’)
| PT | Motu | he-aɣi tau-na | ‘a braggart, conceited man’ (he-aɣi ‘boast’) |
| PT | Motu | he-ani tau-na | ‘a cannibal’ (ani ‘eat’) |
| PT | Motu | he-atotau tau-na | ‘one who pays another to betray, kill or save another’ (he-atotau ‘be held down by s.t. placed on top’) |
| PT | Motu | he-boɣahisi tau-na | ‘a compassionate, merciful person’ (he-boɣa-hisi [he-belly-pain] ‘pity, compassion’) |
| PT | Motu | he-ɣame tau-na | ‘beggar’ (heɣame ‘beg’) |
Although constructions reflecting POc *tau- ‘person who’ are widespread among Papuan Tip languages, there is little evidence that the construction was used for properties of temperament in POc. Instead, as noted in §2.2.1.2, the POc construction was used to denote people with skill in a particular occupation and sometimes people of a certain place or clan. A check of reflexes in Mussau (Adm), Nakanai and Teop (MM), Woleaian and Carolinian (Mic) and various Polynesian languages suggests that the extension of the construction to denote temperamental qualities was restricted to PT.
Several languages have a noun that is used with the sense ‘disposition, way of acting, habit’ and is modified by an adjective or stative verb to form a complex lexeme denoting a property of temperament.
Seimat (Adm) has such a term, oŋa ‘temperament or disposition’.
| Adm | Seimat | oŋa- solia-n | [disposition good] | ‘amiable’ |
| Adm | Seimat | oŋa- kalimen | [disposition bad, terrible, dreadful] | ‘greedy, self-centred, miserly, infamous, inhospitable’ |
| Adm | Seimat | oŋa- ewii | [disposition nice, calm] | ‘benevolent, generous, hospitable’ |
| Adm | Seimat | oŋa- lialun | [disposition bad] | ‘inhospitable’ |
Tongan aŋa reflects the PPn term *aŋa with widespread reflexes outside eastern Polynesia that are used in this way. Its resemblance to Seimat oŋa may be a matter of chance. Tongan in particular has hundreds of complex lexemes formed with aŋa ‘character, habit, nature’. A brief selection is given:
| Pn | Tongan | aŋa malū | ‘gentle, amiable’ (malū ‘calm of sea’) |
| Pn | Tongan | aŋa sauni | ‘vindictive’ (sauni ‘to avenge’) |
| Pn | Tongan | aŋa maka | ‘stubborn, obstinate’ (maka ‘stone’) |
| Pn | Tongan | aŋa hiki | ‘proud, overbearing’ (hiki ‘to lift, raise’) |
| Pn | Tongan | aŋa foaki | ‘generous’ (foaki ‘donate, bestow’) |
| PPn | *aŋa | ‘habit, custom, way of acting’ (POLLEX) | |
| Pn | Tongan | aŋa | ‘character, habit, nature’ |
| Pn | Niuean | aŋa | ‘habit, custom, behaviour’ |
| Pn | Samoan | aŋa | ‘conduct, way of acting’ |
| Pn | Tokelauan | aŋa | ‘habit, custom’ |
| Pn | Tuvalu | aŋa | ‘custom, way of acting’ |
| Pn | East Futunan | aŋa | ‘conduct, custom, usage, habit’ |
| Pn | East Uvean | aŋa | ‘nature, custom, usage’ |
| Pn | Tikopia | aŋa | ‘incline towards; inclination, orientation’ |
| Fij | Rotuman | aŋa | ‘usage, custom’ (Pn borrowing) |
In Tongan aŋa stands in contrast with loto ‘insides’ (< PPn *loto; vol.2:239). Where aŋa denotes a permanent feature of temperament, e.g. aŋa fiemālie ‘of contented and easy-going disposition’, loto denotes a transient emotion, e.g. loto fiemālie ‘contented, satisfied’.
In Wayan Fijian alo- ‘soul’ is similarly used for temperamental features as shown in the following examples, whilst the corresponding transient emotions are denoted by simple adjectives.
| Fij | Wayan | alo-kasa | ‘quick to learn, intelligent, having an absorbent mind’ (kasa ‘learned’) |
| Fij | Wayan | alo-kaikai, alo-qwāqwā | ‘determined, strong-willed, brave, stubborn, aggressive’ (kaikai, qwāqwā ‘strong’) |
| Fij | Wayan | alo-malumalum | ‘gentle, good-tempered, of quiet disposition’ (malumalum ‘soft, ripe’) |
| Fij | Wayan | alo-sewasewa | ‘frightened, intimidated, lacking courage’ (sewasewa ‘tiny’) |
| Fij | Wayan | alo-vinā | ‘kind, kind-hearted’ (vinā ‘good’) |
| Fij | Wayan | alo-vou | ‘young in spirit, young at heart’ (vou ‘new, fresh’) |
| Fij | Wayan | alo-wai | ‘moody, subject to changes in mood’ (wai ‘water’) |
| Fij | Wayan | alo-vaka-tāŋʷane | ‘manly, courageous’ (vaka-tāŋʷane ‘like a man’) |
Similar expressions occur in Bauan Fijian with the cognate term yalo- ‘soul, spirit, disposition’, e.g. yalo mālua ‘meek, lowly’ (mālua ‘quiet, gentle’), yalo vinaka ‘kind-hearted’ (vinaka ‘good’).
The examples collected from Iduna, Motu, Seimat, Fijian and Tongan show how each of these languages has devised a way to describe temperamental qualities, using a modifier with either a term like Tongan aŋa ‘character, habit, nature’ or a reflex of the POc construction with *tau… ‘person who’.
Modifiers that describe features of temperament are reconstructed in the sections below, but this appears to be an unstable semantic field, presumably because of the tendency to invent new metaphors to describe behaviour and temperament. As a result, few terms are reconstructable.
POc evidently inherited a pair of terms meaning ‘tame, docile, trained, well behaved’ that were applied both to animals and human beings. The pair were *laca(m) and *ma-naca(m). Pairs of property terms with the same root were apparently quite common in POc, one unprefixed, the other with either *ka- or *ma-, prefixes that often occur on stative verbs, both originating in the stative prefix *ka- (Zeitoun & Huang 2000:298). The seeming oddity about this pair is that the POc bare root had initial *l-. the prefixed root initial *n-. However, this is readily accounted for if the PAn root had initial *L-,4 which regularly became PMP *l- initially but *-n- medially. Thus PAn *Lajam became PMP *lajam and POc *laca(m), whilst PAn *ma-Lajam regularly became PMP *ma-najam and POc *ma-naca(m).
| PAn | *Lajam | ‘accustomed to, tame’ (ACD) | |
| PMP | *lajam | ‘accustomed to, tame’ | |
| POc | *laca(m) | ‘tame, docile, trained, well behaved’ | |
| MM | Sursurunga | las | ‘tame, used to, get used to’ |
| MM | Tolai | lā | ‘tame, domesticated (of animals), accustomed, acclimatised’ |
| PCP | *laca | ‘tame’ (Geraghty 1983) | |
| Fij | Bauan | lasa | ‘easy, contented, tame, accustomed’ |
| Pn | Tongan | lata | ‘feel at home or at ease, be comfortable, happy, contented’ |
| Pn | Tikopia | rata | ‘fix affections on; tame’ |
| Pn | Hawaiian | laka | ‘tame, domesticated, gentle’; [VT] ‘tame, domesticate’ |
POc *ma-naca(m) underwent an extension in meaning whereby ‘trained’ became ‘knowledgeable’ and then ‘know, understand, think about’, and as a nominal ‘knowledge, understanding, mind’. This extension is discussed in §10.3. Reflexes below are restricted to the ‘tame’ etc senses.
| PAn | *ma-Lajam | ‘tame, accustomed to’ (ACD) | |
| PMP | *ma-najam | ‘tame, accustomed to’ | |
| POc | *ma-nacam | [VI] ‘tame, docile, trained, well behaved; know, understand, think about’; [N] ‘knowledge, understanding, thought, wisdom’ | |
| NNG | Gedaged | mana-n | ‘tame, docile (mostly of animals), peaceful, obedient, trained’ |
| PT | Tawala | malagama | ‘tame’; [N] ‘experience’ (-l- for †-n-) |
| PT | Sinaugoro | marana | ‘tame, gentle’ (metathesis of *-c- and *-n-) |
| PT | Motu | manada | ‘even, smooth, gentle’ |
| MM | Nehan | mahanama | ‘tame, unafraid’ (metathesis of *-c- and *-n-) |
| PSES | *manasa | ‘tame’ | |
| SES | Gela | manaha | [VT] ‘tame’ |
| SES | Arosi | manata | ‘tame, trained, gentle, of man or animal’ |
| SES | Sa’a | manata | [VI] ‘be taught; quiet of animals, tame’ |
| SES | Lau | manata | ‘quiet, tame’ |
| SES | ’Are’are | manata | ‘behave oneself; tame of birds and animals’ |
| SES | Owa | manata | ‘tame’ |
It is tempting to try to relate the items below to *ma-nacam above, but they evidently reflect a different (PWOc) etymon.
| PWOc | *ma-napas | ‘tame (of animals); quiet, domesticated (of people)’ | |
| PT | Iduna | manava | ‘become tame (of wild animals); soft, pliable’ |
| PT | Iduna | (ve)manav(ina) | [VT] ‘tame, of wild animals’ (ve- ‘cause, make’, manavina ‘soft, pliable’) |
| MM | Nduke | manavasa | ‘tame’ |
| MM | Roviana | manavasa | ‘tame, subdued, at home, be used to’ |
It is reasonably clear from the glosses of the sets above that these terms referred to a temperamental quality for which there is no single English term. It describes a person who is domesticated, gentle and quietly spoken, and content with their lot. This quality was apparently positively valued. If it were otherwise, the glosses would almost certainly register the negativity.
POc *[ma]lumu ‘soft (of objects); gentle, easygoing; (constitutionally) weak’, reconstructed in vol.2(215) as a property of inanimate objects, is applied in various Oceanic languages to the human temperament with a sense resembling that of the items above, and this extension may well have occurred in POc. It was evidently also applied to the body in the sense of ‘constitutionally weak’, and the cognate set is listed in §11.2.4.
There is also evidence of a BPM containing this term.
| SES | Arosi | (ahu)marumu(ʔa) | [belly gentle] | ‘be sweet, gracious, gentle’ |
| NCV | Mota | lolo-malumlum | [insides-gentle] | ‘sort-hearted, of an easy mild temper; gently’ |
| NCV | Nokuku | lol-melum | [insides gentle] | ‘meek’ |
| Fij | Bauan | (yalo) mālua | [soul gentle] | ‘meek, lowly’ |
| Fij | Wayan | (alo) malu-malum | ‘gentle, good-tempered, of quiet disposition’ | |
| Pn | Pukapukan | (yau)mālū | [temperament gentle] | ‘meek’ |
In NCV, at least, the POc weather term *ma-drapu ‘still, calm, windless’ (vol.2:136) also has human application.
| PNCV | *madau | ‘quiet, gentle’ (Clark 2009) | |
| NCV | Raga | marou | ‘quiet, listless’ |
| NCV | Southeast Ambrym | merou | ‘calm, quiet, peaceful, sober’ |
| NCV | Port Sandwich | maⁿdrao | ‘be quiet, do gently’ |
| NCV | Paamese | merau | ‘weak; soft; do softly; do gently’ |
| NCV | Valpei | marav | ‘weak’ |
As POc reconstructions for ‘tame, accustomed to’ refer to a kind of domesticated amiability, so terms glossed ‘wild’ similarly refer to the behaviour often associated with undomesticated animals, i.e. ‘savage’, ‘fierce’. In places, terms may also be applied to the rainforest and to normally cultivated plants that have self-sown outside the gardens. While terms for ‘tame’ are also used to denote a human temperamental quality, it is less clear that this is true of POc *wasi.
| POc | *wasi | ‘wild, untamed’ | |
| PT | Gumawana | woiwoi | ‘wild’ (animals) |
| PT | Tawala | yahi-yahi | ‘untamed, wild, dangerous’ (y- for †w-) |
| PT | Kuni | aci | ‘wild’ (∅- for †w-) |
| PT | North Mekeo | aki | ‘wild’ (∅- for †w-) |
| PT | West Mekeo | aji | ‘wild’ (∅- for †w-) |
| MM | Maringe | asi | [VI] ‘run wild, go astray’ |
| SES | Bugotu | asi | ‘wild (of animals), fierce’ |
| SES | Gela | asi | ‘wild (of animals or forest)’ |
| SES | Longgu | wasi | ‘wild, undomesticated’ |
| SES | To’aba’ita | kʷasi | [VI] ‘(of animals), be wild, not domestic, (of plants) grow wild’ |
| SES | Lau | kʷasi | ‘wild, of animals or plants’ |
| SES | Kwaio | kʷasi | ‘wild, untamed’ |
| SES | Arosi | wasi-wasi | ‘wild, of any animal’ |
| SES | Sa’a | wasi | [VI] ‘wild, not tame’ |
| SES | ’Are’are | wasi | ‘wild, untamed’ |
| SES | Owa | wasi | ‘wild, untamed’ |
| NNG | Takia | kasik | ‘wild’ |
| PT | Kilivila | gasisi | ‘wild, savage’ |
Bravery is commonly expressed as a body-part metaphor (BPM) based on the liver or belly, with a modifying term apparently reflecting POc *(p,pʷ)atuatu ‘hard, strong, firm’, the latter derived from *(p,pʷ)atu(k) ‘outer shell, skull’ (§3.4.2). Hence we can tentatively reconstruct a POc BPM *qate- (p,pʷ)atu [liver- hard/strong] ‘brave’. In the languages of the set below, reflexes of *(p,pʷ)atu also function as a stative verb ‘firm, strong’. Reflexes of (p,pʷ)atu are formally and semantically similar to *patu ‘stone’ (vol.2:62), but in POc were distinct from it (§3.4.2.1). In Kwaio, however, which does not have a reflex of *(p,pʷ)atu, the fou of lae-fou ‘brave’ is the term for ‘stone’. One may infer that the Kwaio term is a reflex of *qate- (p,pʷ)atu which with the loss of a reflex of *(p,pʷ)atu has been reinterpreted as if it reflected *qate- patu [liver- stone]. Something similar has occurred in Gumawana and Iduna, where the modifying term taken alone has a meaning other than ‘firm, strong’.
Why not infer that *patu ‘stone’ occurred in the POc term and that all BPMs below reflect *qate- patu? Because POc *patu ‘stone’ would be reflected as Gumawana †vatu, Dobu †atu, Bunama †hatu, but these terms do not occur at all (Gela vatu ‘stone’ does occur). The morpheme patu in their BPMs thus probably reflects (p,pʷ)atu (cf. Dobu patu (VI) ‘harden, set’).
| POc | *qate-(p,pʷ)atu | ‘brave’ | |
| PT | Dobu | ʔate-patu | [liver hard] ‘brave, bold’ |
| PT | Bunama | ʔate-patu | [liver strong/firm] ‘courage, brave, confidence’ |
| PT | Minaveha | ate vatu | [liver strong/firm] ‘brave’ |
| SES | Gela | lio-patu | [heart-hard] ‘daring, brave’ |
| PT | Gumawana | ate- i-patu | ‘brave’ |
| PT | Iduna | ase-vatu | ‘courage, boldness’ |
| PT | Iduna | ase-ʔase-vatu- | ‘brave, courageous, ungenerous.’ |
| SES | Kwaio | lae-fou | ‘brave, unashamed’ |
A construal of bravery as strength or hardness — the two are typically encoded by a single lexical item in languages of Melanesia — of the seat of the emotions is widespread.
| NNG | Takia | ilo- dabai | [insides- strong] | ‘confident, courageous, strong in character’ |
| NNG | Takia | tini- dabai | [body- strong] | ‘confident, courageous, strong in character’ |
| PT | Motu | boga auka | [belly/liver hard] | ‘brave’ |
| NCV | Lonwolwol | bɛ-k mə-təte | [insides-my it-strong] | ‘I feel strong, energetic, courageous’ |
| NCV | Lonwolwol | ja təte | [body strong] | ‘courageous’ |
| NCV | South Efate | kerkerai | ‘strong, hard, brave’ | |
| Mic | Marshallese | pen pɯruo- | [firm/strong/hard heart-] | ‘brave’ |
| Fij | Bauan | yate dei | [liver firm/unwavering ] | ‘brave’ |
| Fij | Wayan | ate dei | [liver firm] | ‘brave’ |
| Fij | Wayan | alo kaikai | [soul strong/hard] | ‘determined, strong-willed, brave, stubborn, aggressive’ |
Relatively few Oceanic dictionaries have an entry for ‘cowardly’. Terms found tend to reflect *matakut ‘fear’ (§11.4.1), as the examples below suggest.
| PT | Dobu | mata-matauta | ‘cowardly’ |
| SES | Gela | matagu-pou | ‘coward’ |
| NCV | Southeast Ambrym | metau | ‘cowardly, timid’ |
| NCV | Paamese | tā-metau | ‘coward’ |
| SV | Anejom̃ | e-mtac | ‘fear; be afraid, fearful, cowardly’ |
| Fij | Bauan | mata-mataku | ‘be always afraid, timid, cowardly’ |
We have three reflexes of the BPM *qate- lapuat [liver big]. Oddly, this metaphor has the reverse meaning of that proposed by Blust for PMP where [big liver] stands for ‘brave, proud, arrogant’ (see §9.6).
| NCV | Mota | vara-lava | [liver large] | ‘one who is easily frightened, makes much of nothing’ |
| Fij | Bauan | yate levu | [liver large] | ‘cowardly’ |
| Fij | Wayan | ate levu | [liver large] | ‘cowardly’ |
Marshallese has eccelok acin [without liver] ‘he is not brave’.
A rather widespread BPM for ‘obstinate, stubborn’ has a modifier of the same meaning as the BPM for ‘brave, courageous’ (§11.3.2.1) namely ‘strong/firm’, but a different body-part. POc *qate- ‘liver’ was evidently used in the BPM for ‘brave’. The evidence of Admiralties and NCV languages below suggests that *qate- was replaced by *bʷatu(k) ‘head’ (§3.4.2) in the BPM for ‘obstinate’. This may be a further illustration of the hypothesis that more physically expressed qualities such as ‘brave’ form a BPM with *qate-, while mental states do not (§9.4). Instead of *lalo-, however, in this case the more specific *bʷatu(k) occurs. Clark (2009) reconstructs a BPM here, PNCV *bʷatu kayua [head strong] ‘wilful, stubborn’, reflected by the Big Nambas and Paamese terms below.
| Adm | Seimat | patu ailan | [head hard/strong] | ‘he is obstinate’ |
| Adm | Nyindrou | batun boto-on | [head hard/strong] | ‘stubborn’ |
| NNG | Takia | ilo patpat | [insides hard/strong] | ‘hard-headed, strong-minded, wilful’ |
| NNG | Takia | bube- sakar | [liver- hard] | ‘hard hearted, stubborn, uncompassionate’ |
| NNG | Mapos Buang | yu niggɔχ | [head strong] | ‘stubborn’ |
| PT | Kilivila | i-minimani daba-la | [it-tough/strong head-his] | ‘he is stubborn’ |
| SES | Gela | lio vatu | [insides stone] | ‘obstinacy; stubbornness’ |
| NCV | Mwotlap | pʷet-maymay | [head hard] | ‘stubborn’ |
| NCV | Big Nambas | pt-hua | ‘stubborn’ | |
| NCV | Paamese | vati-keiho | ‘person who is determined and inconsiderate of the feelings of others’ | |
| SV | Sye | -oŋkoŋko | ‘hard, strong, stubborn’ | |
| SV | Lenakel | nɨkii- r-ausɨk-ausɨk | [heart- it-REDUP-strong/hard] | ‘stubborn’ |
Another BPM with scattered WOc occurrences is ‘ear blocked’.
| NNG | Mutu | taliŋa- zizi | [ear- blocked] | ‘stubborn’ |
| NNG | Mangap | talŋa- ŋuŋun | ‘stubborn’ | |
| NNG | Bariai | taŋa- balbal | [ear ??] | ‘obstinate, stubborn’ |
| PT | Iduna | taɣa- -kulu | [ear blocked] | ‘(X is) obstinate, deaf to instruction’ |
| PT | Iduna | taɣa-kulu | ‘obstinacy’ | |
| PT | Iduna | taɣa-taɣa-kulu-na | ‘obstinate’ |
In Eastern Oceanic languages ignorance is sometimes expressed by a BPM [mind dark], a complex lexeme that also means ‘forget’ in some EOc languages (§10.6). In some languages this BPM has both senses. Again, the terms for ‘night’ reflect either POc *rodrom ‘be dark, be night’ or POc *boŋi ‘night’ (vol.2:295–298).
| SES | Lau | lio ro-rodoa | [mind dark] | ‘ignorant, puzzled’ |
| NCV | Mota | lolo-pʷoŋ | [inside-night] | ‘ignorant, stupid, unenlightened; forget’ |
| NCV | Mwotlap | lɔl-pʷoŋ | [inside-night] | [VT] ‘forget, ignorant’ |
| NCV | Nokuku | lolo- ōra | [inside- night] | ‘forget, ignorant’ |
| Mic | Kiribati | nano-ro | [insides dark] | ‘ignorant, uncultured’ |
Sometimes a verb ‘be dark’ is used alone as a metaphor for ‘ignorant’.
| SES | To’aba’ita | rō-rodoʔa | ‘be dark (of a place); be ignorant about (of a person)’ (rodo ‘night, be dark’) |
| Fij | Wayan | ŋīŋīlō | ‘dull, poor in light; get dark; be dusk, nightfall; ignorant, lacking in knowledge or wisdom’ |
Oceanic languages tend to have a term which means ‘mentally impaired’ and translates, depending on context, as ‘ignorant’, ‘stupid’, ‘foolish, silly’ or ‘mad, crazy, insane’. They reflect a number of POc terms which must have different shades of meaning that are now lost to us.
POc *kila has only one Oceanic reflex, and it is a reasonable inference that it disappeared early across much of Oceanic because the merger of PMP *g and *k rendered Oceanic reflexes similar to those of POc (VI) *kilala ‘know’, from PMP *kilala, with the opposite meaning.
| PMP | *gila | ‘wild; insane’ (ACD) | |
| POc | *kila | ‘ignorant’ (?) | |
| NNG | Kaulong | kila | ‘ignorant of, unfamiliar with (s.o., s.t.)’ |
The presence of reflexes of POc *[ŋa(q)u]ŋa(q)uŋ ‘stupid, ignorant’ in an Admiralties language and a few NNG languages warrants its reconstruction. Possible medial *-q- is shown, as it allows the reconstructed form to accord with POc canonic structure, and would have been lost in each of these reflexes.
| POc | *[ŋa(q)u]ŋa(q)u | ‘stupid, ignorant’ | |
| Adm | Titan | ŋow | ‘crazy, silly’ |
| NNG | Manam | -ŋao | ‘dull, stupid’ |
| NNG | Takia | -ŋao-ŋ | ‘ignorant, unconscious, stupid, confused’ |
| MM | Patpatar | ŋa-ŋawa | ‘ignorant, confused’ |
The presence of medial *-p- in POc *bʷa(p)u below is unambiguously supported only by Iamalele -v-, and is contradicted by the absence of -v- in Longgu. In the other witnesses *-p- becomes zero in this context.
| POc | *bʷa(p)u | ‘ignorant, stupid’ | |
| NNG | Kairiru | -pʷau | ‘ignore’ |
| PT | Iamalele | bavu-bavu- | ‘ignorant’ (-v- for †-∅-) |
| MM | Sursurunga | bau | ‘stupid, restless, uncontrollable, self-willed’ |
| SES | Longgu | bʷeu | ‘naughty, silly, stupid, crazy’ |
| SV | Anejom̃ | a-pʷa-pʷau | ‘slow-moving, unable to avoid injury, stupid, unintelligent’ |
| PMic | *pʷau-pʷau | ‘silly, stupid’ (Bender et al. 2003: Proto W Micronesian *pʷai-pʷai) | |
| Mic | Marshallese | pʷəypʷəy | ‘crazy, silly, foolish’ |
| Mic | Ponapean | pʷeypʷey | ‘stupid, silly, idiotic, simple, dumb’ |
| Mic | Mokilese | pʷeypʷey | ‘stupid’ |
| Mic | Satawalese | pʷayipʷay | ‘silly, disrespectful’ |
| POc | *[bʷa]bʷaŋ | ‘a fool; foolish, stupid, insane’ (ACD: *bobo ‘foolish; a fool’) | |
| NNG | Manam | boaŋ | [VI] ‘be insane’ |
| NNG | Manam | boaŋ-boaŋ | [ADJ] ‘insane’ |
| NNG | Bariai | buo-buo | ‘be confused about s.t.’ |
| NNG | Kaulong | poŋ | [VT] ‘be ignorant of, not know’ |
| PT | Tawala | būa | ‘fool; mad, silly’ |
| PT | Tawala | bū-bua | ‘foolish’ |
| PT | Dawawa | bua | ‘stupid, mad’ |
| PT | Sinaugoro | babo | ‘stupid, foolish, ignorant’ |
| PT | Motu | bōbo | ‘a fool’ |
| MM | Tolai | bobo | ‘a fool; foolish, stupid, ignorant’ |
| MM | Patpatar | ba | ‘crazy, retarded, foolish, worthless; good-for-nothing’ |
| MM | Tinputz | babɔn | ‘crazy, stupid’ |
| SES | Lau | bue | ‘ignorant, uncivilised, pagan, heathen’ |
| PCP | *wale | ‘ignorant, unskilled, stupid, mad’ | |
| Fij | Bauan | wale | ‘not worthwhile, very ordinary’ |
| PPn | *wale | ‘mad, ignorant, unskilled’ (POLLEX) | |
| Pn | Tongan | vale | ‘foolish, silly, ignorant, unskilled’ |
| Pn | Tongan | loto-vale | ‘ignorant’ (vale ‘foolish, silly’) |
| Pn | Tongan | vale faha | ‘stupid’ (faha ‘mad, insane’) |
| Pn | Niuean | vale | ‘mad, ignorant, unskilled’ |
| Pn | East Futunan | vale | ‘mad, ignorant, unskilled’ |
| Pn | Pukapukan | vale | ‘be senile, lose one’s memory’ |
| Pn | Samoan | vale | ‘idiot; worthless’ |
| Pn | Tokelauan | vale | ‘idiot, fool, lunatic; mentally ill person’ |
| Pn | Tikopia | vare | ‘stupid, foolish, silly; trivial’ |
| Pn | Māori | ware | ‘ignorant, lowly, careless’ |
Despite the fact that Oceanic languages typically have a term meaning ‘intelligent, clever, wise, understanding’, it has proven impossible to reconstruct a corresponding POc term.
Occasionally a BPM [mind daylight] occurs, the antonym of [mind night] ‘ignorant’ (§11.3.4), and it is possible that there was a PNCV BPM of the form *lolo- marani [insides daylight] ‘clever, intelligent’ (reflecting POc *ma-raqani ‘be(come) light’, vol.2:318–219).
| NNG | Mengen | lo- matana | [inside- light] | ‘knowledge, understanding’ |
| SES | ’Are’are | pau- makata | [head- bright] | ‘intelligent, wise’ |
| NCV | Mota | lolo-marani | [mind-daylight] | ‘be intelligent; remember, understand, know’ |
| NCV | Lonwolwol | lo- merɛn | [mind- daylight] | ‘clever’ |
The term for ‘wise, knowledgeable’ is often derived from the verb ‘know’ (§10.2).
| PT | Dobu | sina-sinapu- | ‘wise’ (sinapu ‘understand, know’) |
| MM | Roviana | tuma-tumai | ‘wise’ (tuma ‘know, understand’) |
| NCV | Paamese | kile-ile | ‘educated; knowledgeable’ (kilea ‘know’) |
Only for PPn has a form been reconstructed.
| PPn | *poto | ‘wise, clever’ (POLLEX) | |
| Pn | Tongan | poto | ‘clever, skilled’ |
| Pn | Samoan | poto | ‘clever, smart, intelligent’ |
| Pn | East Futunan | poto | ‘clever, knowledgeable’ |
| Pn | Anutan | poto | ‘wise, expert’ |
| Pn | Tikopia | poto | ‘skilled, adept, knowledgeable’ |
As discussed in chapter 9, emotions are typically described as emanating either from the liver (POc *qate), as seat of emotions and thoughts, or from a quasi-body part, POc *lalo-, *lalom ‘inside; seat of thoughts and emotions’, but other body parts sometimes occur in their place. Particular feelings are expressed as a BPM that specifies the nature of the feeling.
Although a degree of common conceptual patterns can be identified in BPMs for broadly identified emotions like happiness, sadness and anger, no reconstructions are made.
Languages closer to New Guinea seem to make the greatest use of BPMs for emotions, and languages further east use them less, petering out almost entirely in Polynesia, where the body-part component of the BPM tends to be lost and the modifying component tends to become an adjective in its own right (§3.4, §3.5, §3.8).
A small number of broadly identified emotion terms have been reconstructed as single lexemes, and we turn to these first.
Although a language will typically use numerous metaphors to describe different degrees or kinds of fear, languages from all major subgroups have reflexes of POc *matakut, probably indicating that it was the term most general in meaning. Both an intransitive and a transitive form are reconstructable. The intransitive form, *matakut, is widely reflected. The transitive form, *matakut-i-, has fewer reflexes, and it is possible that it reflects parallel innovations in various Oceanic languages. The reason for this inference is that the prefix *ma- is a stative formative that was originally incompatible with the transitive marker *-i-.
| POc | *matakut | [(VI) ] ‘be afraid’ | |
| POc | *matakut-i- | [(VT) ] ‘to fear (s.t.)’ | |
| Adm | Seimat | ma-matau | [VT] ‘fear, be afraid of’ |
| Adm | Seimat | ma-mata | [VI] ‘be afraid, timid’ |
| NNG | Kove | mataur-i- | ‘afraid’ (r reflects *R) |
| NNG | Mengen | matau, matau-e | [VI, VT] ‘fear’ |
| NNG | Manam | mataʔu | ‘be afraid’ |
| PT | Lala | makau | ‘afraid, fear’ |
| PT | Molima | matauta | ‘afraid’ (expect mataʔuta) |
| PT | Dobu | matauta | ‘afraid’ |
| PT | Saliba | mataus-i- | ‘be afraid’ |
| MM | Roviana | mataɣutu | ‘afraid, fearful’ |
| SES | Bugotu | mataɣu | ‘to fear, be afraid’ |
| SES | Gela | mataɣu | ‘to fear, be afraid’ |
| SES | Kwaio | maʔu | ‘afraid, shy’ |
| SES | To’aba’ita | maqu | [VI] ‘be afraid, fear’ |
| SES | To’aba’ita | maqulā | [N] ‘fear’ |
| SES | Ulawa | māu | ‘to fear, be afraid’ |
| SES | Arosi | ma-māʔu | [VI] ‘to fear’ |
| SES | Arosi | māʔus-i- | [VT] ‘to fear’ |
| NCV | Mota | mataɣ-taɣ | ‘to fear’ |
| NCV | Tamambo | matahu | ‘be frightened’ |
| NCV | Tirax | mtaxit | ‘be frightened’ |
| NCV | Nguna | mataku | ‘afraid’ |
| PSV | *a-metaɣ | [VI] ‘be afraid, fear’ (Lynch 2001c) | |
| SV | Sye | emetet | [VI] ‘be afraid, fear’ |
| Mic | Carolinian | mesaxu, -a | [VI, VT] ‘have fear, be afraid’ |
| Mic | Woleaian | metagu | ‘be afraid, anxious’ |
| Pn | Rennellese | mataku | ‘be afraid, cowardly, fear’ |
| Fij | Wayan | mataku | [VI] ‘be afraid’ |
| Fij | Wayan | matakuði- | [VT] ‘be afraid of’ |
| Pn | Niuean | mataku-taku | ‘to fear, be afraid’ |
| Pn | Samoan | mataʔu | ‘fear, hold in awe’ |
| Pn | Pukapukan | mataku | [VI] ‘fear, be afraid, frightened’ |
| Pn | Tikopia | mataku | ‘afraid, frightened, fearful’ |
| Pn | Tokelauan | mataku | ‘be afraid, frightened’ |
| Pn | Tahitian | mataʔu | ‘apprehension’ (s.t. bad might happen) |
A number of NNG and PT languages have BPMs to express fear, but their components differ from one language to the next. For example:
| NNG | Mangap | kuli- i-mozōro | [skin- it-scatter] | ‘really frightened’ |
| NNG | Mangap | mata- koikoi | [eye- evasive] | ‘fearful’ |
| NNG | Takia | ilo- i-rer | [insides- it-tremble] | ‘afraid, frightened, fearful’ |
| PT | Tawala | nugo-helele | [mind-fearful] | ‘nervous, anxious, afraid, shaking with fear, surprised, have pounding heart’ |
The next term is remarkable among emotion terms in being reconstructable right back to PAn with very high consistency of meaning. The emotion, for which there is no single equivalent English term, is valued as an instrument of social control, most effective in small communities. As described by Charles Valentine with reference to Nakanai speakers,
The feeling is described as a kind of acute embarrassment which is occasioned by public exposure, violation of modesty, recognition of deception, social exclusion, and certain other forms of interaction in which the subject feels threatened by the inappropriateness of his relations with others. (Valentine 1963:445)
| PAn | *ma-Seyaq | ‘shy, embarrassed; ashamed’ (PAn *Seyaq ‘shyness, embarrassment; shame’) (ACD) | |
| PMP | *ma-heyaq | ‘shy, embarrassed; ashamed’ | |
| POc | *maya(q) | ‘shy, ashamed’ | |
| NNG | Manam | maia, maya, maya-maya | ‘ashamed’ |
| NNG | Gedaged | ma-mai | ‘shame; embarrassment, confoundedness, abashment’ |
| NNG | Numbami | me-meya | ‘shy, ashamed’ |
| NNG | Yabem | maya | ‘shame, feeling of honour, self-esteem’ |
| NNG | Bukawa | maya | ‘shame, embarrassment’ (experienced by people who have been found out; Hogbin 1947, village of Busama) |
| NNG | Kove | ma-maia | ‘ashamed’ |
| PT | Ubir | ma-mai | ‘shy, ashamed’ |
| PT | Dobu | (o)maia-maia | ‘shame, shyness, be ashamed, shy’ |
| PT | Molima | (wo)maya-maya | ‘shame’ |
| MM | Ramoaaina | mai-mai | ‘shame; ashamed’ |
| MM | Maringe | ma-maja | ‘ashamed’ |
| SES | Gela | mā | ‘ashamed; feel reverence’ |
| SES | Lau | māsia | ‘shame’ |
| SES | Sa’a | masa, masa-masa | ‘shy, ashamed, respectful’ |
| SES | Sa’a | masa-ŋa | ‘shame, confusion, shyness’ |
| SES | Arosi | [ma]masa | ‘ashamed’ |
| NCal | Iaai | m̥e-m̥æ | ‘ashamed’ |
| Mic | Kiribati | mā-ma | ‘shame, timidity, shyness, bashfulness’ |
| Mic | Kiribati | ma-mā-ma | ‘be ashamed, shy, confused’ |
| Mic | Woleaian | ma | ‘shame, embarrassment’ |
| Mic | Woleaian | mā, ma | ‘ashamed; disgraceful; feel shameful’ |
| Fij | Bauan | mā(duā) | ‘ashamed, bashful’ |
| Pn | Tongan | mā | [VI] ‘feel shame, be ashamed’ |
| Pn | Pukapukan | (aka)mā | ‘be ashamed, embarrassed, shy, timid’ (aka ‘cause, become’) |
| Pn | Samoan | mā | ‘be ashamed, embarrassed’ |
| Pn | Tahitian | (haʔa)mā | ‘embarrassment or shame’ (opposite of mātau ‘comfortable with a situation’) |
| Pn | Māori | mā | ‘shame, abasement; shy, ashamed’ |
| NNG | Takia | miai | ‘shame, ashamed’ |
| PT | Kalauna | (veu)maiyiyi | ‘a mixture of anger, shame, self-pity and resentment’ (exorcising one__s own shame by casting the shame back at the one who shamed by forcing him to accept a gift etc.; Michael Young, pers. comm.) |
The English term love is polysemous, with one meaning, ‘romantic love’, given an importance in Western societies that is not generally paralleled elsewhere. In Tahitian maʔa-maʔa is translated by Levy as
to be crazy, bizarre (incl. being romantically in love). This is considered somewhat bad and abnormal. (1973:305)
Across Oceanic languages, terms for ‘to be lovers’5 are distinct from terms that denote caring about someone, and terms used to refer to the latter include an emotion akin to compassion. This semantic frame is labelled SORRY here, in recognition of the fact that the word sore encodes this frame in both Papua New Guinea Tok Pisin and Vanuatu Bislama. Motu hebogahisi ‘pity, compassion’ is instructive, combining boga ‘belly, seat of desire and affection’ with hisi ‘pain’. Other glosses combining what in English are distinct emotions include Gedaged ilo- pani [inside- give] ‘sympathise with, love, pity, yearn for, feel for, commiserate with, mourn for, be homesick for s.o., s.t.’; Tolai māri ‘to love, pity, have compassion for’, Roviana tataru ‘to pity, love’, To’aba’ita tatakomia ‘have a feeling of deep affection for s.o., s.t., such as sorrow, pity, compassion, mercy, love or admiration’, Arosi tabai ‘to love, pity’, Rotuman ruu ‘to love, value, care greatly for, feel solicitude’. English speakers, on the other hand, are more likely to include ‘love’ as an extension of ‘like’, ‘admire’ and a range of terms for ‘desire’.
As noted in §10.3, the SORRY frame is one of the meanings of widely distributed reflexes of POc (VI) *drodrom, (VT) *drom-i ‘think, worry; love, be sorry for, long for’. A dedicated SORRY verb, POc *qarop, *qarop-i- ‘feel pity, empathy, be sorry for’,6 was inherited from PMP but is reflected only in SES and Pn languages. The Arosi reflex and all Pn reflexes reflect an apparent PEOc *qarop-a. The suffix *-a probably reflects the POc nominaliser *an.
| PMP | *qarep | ‘like, be fond of’ (Dempwolff 1938: *haḷəp) | |
| POc | *qarop, *qarop-i- | ‘feel pity, empathy, be sorry for’ | |
| SES | Bugotu | (r)arov-i | [VT] ‘to pity’ |
| SES | Gela | arov-i | [VT] ‘to pity’ |
| SES | Longgu | arov-i- | ‘feel sorry for and sad for s.o., how you feel for s.o. else who has some trouble’ |
| SES | Arosi | ʔaroha | ‘love, pity’ |
| PPn | *qarofa | ‘love, pity, compassion’ (POLLEX) | |
| Pn | Tongan | ʔofa | ‘love, be kind to’ |
| Pn | Niuean | ofa | ‘to love (obsolete)’ |
| Pn | Rennellese | ʔagoha | ‘pity, feeling of love’ |
| Pn | Samoan | alofa | ‘love, affection, mercy, pity’ (stresses social bonding and obligation) |
| Pn | Tokelauan | alofa | ‘love, affection, kindness’ |
| Pn | Tokelauan | loto-alofa | ‘kindhearted, friendly; kindness, friendliness, hospitality’ |
| Pn | Tikopia | arofa | ‘sympathy, affection, love’ |
| Pn | Tahitian | arōfa | ‘compassion, pity, empathy; feeling when separated from s.o. dear’ |
| Pn | Hawaiian | aloha | ‘love, affection, compassion, mercy, pity, kindness; greeting’ |
| SES | Gela | aroha | ‘sit sad, lonely, pitiable’ (-h- for †-v-) |
| Pn | Pukapukan | aloa | ‘love, kindness, charity’ |
Two further cognate sets, each with few reflexes, permit the reconstruction of SORRY morphemes. The first, POc *dolom, is restricted to NWS and SES languages and may represent local innovations, but Nehan is the northernmost NWS language, so it is unlikely that the set is due to borrowing. It is likely that *dolom was a noun, as the Gela and Longgu transitives in -vi (for †dolom-i-) appear to reflect a PSES formation.
| POc | *dolom | ‘love, pity, sorrow, compassion’ | |
| MM | Nehan | dolomo | [N] ‘sorrow’ |
| MM | Nehan | (uel)dolomo | [V] ‘be sorry; show compassion’ |
| SES | Gela | dolo | [N] ‘pity’ |
| SES | Gela | dolo-vi | [VT] ‘love, pity’ |
| SES | Longgu | dolo | [VI] ‘love’ |
| SES | Longgu | dolo-vi | [VT] ‘love someone’ |
The set below is restricted to WOc. If the Iduna items reflect the same etymon, then its PWOc form was *(q)uduqu, but there is no three-syllable reflex ito confirm this.
| PWOc | *(q)udu, *(q)udu-an | ‘be sorry for, pity, be merciful’ | |
| NNG | Bariai | udu-an | ‘feel sorrow or pity for s.o. you care about; miss s.o.’ |
| MM | Bola | du | ‘care for’ (loss of initial u- unexpected) |
| MM | Tinputz | rūru-an | ‘pity, have mercy’ |
| MM | Tinputz | ur-uru-an | ‘forgive, pity’ |
| MM | Teop | uru | [N] ‘pity’ |
| MM | Teop | ur-uru | [V] ‘have mercy, love’ |
| PT | Iduna | duʔu | [N] ‘love, affection, love-gift, share (of food, property)’ |
| PT | Iduna | -duʔu | [V] ‘feel affection; long for (person of opposite sex)’ |
A number of languages have BPMs for the SORRY meaning, but their components usually differ from language to language. Examples include:
| Adm | Nyindrou | ade- hadru in ta- | [liver- true in OBL-] | ‘love, cherish, adore’ |
| NNG | Gedaged | ilo- i-pani- | [insides- it-give-] | ‘sympathise with, love, pity, commiserate with, mourn for, long for’ |
| NNG | Takia | bube- i-pani- | [liver- it-give-] | ‘love’ |
| PT | Bunama | ʔate-muya-muya | [liver-pain-pain] | ‘pity, compassion’ |
In many of these BPMs, however, the modifying component has no meaning independent of the BPM in which it occurs.
BPMs for ‘happy’, ‘cheerful’, ‘joyful’, ‘glad’, ‘carefree’ often have ‘good’ as their modifying component. This is sometimes a reflex of POc *puia ‘good’ or PEOc *leka ‘good, pleasant’ (§11.6.1).
| NNG | Takia | ilo- uyan | [insides- good] | ‘happy, pleased, thanks, greeting’ |
| NNG | Yabem | tɪtaʔ ŋayam | [belly good] | ‘contented, happy’ |
| NNG | Numbami | tae-wia | [guts-good] | ‘happy’ |
| NNG | Mengen | la-u-pe | [insides-my-good] | ‘I am happy’ |
| NNG | Mapos Buang | aʁe- nivəca | [insides- good] | ‘happy’ |
| PT | Kilivila | ibwaina lula | [belly good] | ‘he is happy, at peace, relaxed’ |
| PT | Dawawa | nua-vere | [insides-good] | ‘happy’ |
| PT | Lala | lalo nama | [insides good] | ‘happy’ |
| PT | Motu | lalo namo | [insides good] | ‘happy’ |
| SES | Kwaio | noni leʔa | [body good] | ‘grateful, happy, proud’ |
| Pn | Tokelauan | loto-fiafia | [inside-happy] | ‘be humorous, happy’ |
While some Pn languages use a BPM with fiafia ‘happy’ as its modifying element, many of them use fiafia alone, as in the set below.
| PPn | *fia-fia | ‘happy’ (POLLEX) | |
| Pn | Tongan | fiefia | ‘happy’ |
| Pn | Niuean | fiafia | ‘joy, delight, pleasure, be happy, joyful’ |
| Pn | Samoan | fiafia | [V] ‘enjoy’; [N] ‘gladness, enjoyment’; [ADJ] ‘happy’ |
| Pn | East Uvean | fiafia | ‘happy, joyful’ |
| Pn | Tuvalu | fiafia | ‘happy’ |
| Pn | Tikopia | fiafia | ‘glad’ |
It is just possible that POc had a single word, *puiawa-, for ‘happiness’ (or ‘happy’), apparently derived from *puia ‘good’ (§11.6.1). However, this derivation involves positing an otherwise unknown morpheme *-wa-, and possible reflexes are few.
| POc | *puia-wa | ‘happiness’ (or ‘happy’?) | |
| PT | Gumawana | uyawa- | ‘happy, pleased’ |
| PT | Dobu | uyawa- | ‘joy, gladness; rejoice’ |
| SES | Owa | piawa | ‘happy; calm (of ocean)’ |
No simple lexeme can be reconstructed for ‘sad’. Instead. many Oceanic languages use a BPM for ‘sad’, and related meanings like ‘depressed’, ‘miserable’ and ‘unhappy’, and here there are some consistent patterns. One pattern combines ‘insides’ or ‘mind’ with a term meaning ‘heavy’. The fact that no *qate- terms have been collected for ‘sad’-like emotions is possibly because these are seen as passive, non-violent. As Bugenhagen (2001:96) has argued for Mangap, *qate- terms occur most often with rash, impetuous emotions.
| NNG | Yabem | ŋalɪlʊm wapaʔ | [insides heavy] | ‘his heart is heavy, full of sorrow, dispirited’ |
| NNG | Mangap | lele- i-pata | [insides- it-heavy] | ‘worried, concerned, troubled, sad’ |
| NNG | Mapos Buang | ayo maɣin | [insides heavy] | ‘sad, unhappy’ |
| NNG | Takia | ilo-mulua-n | [insides-heavy-3SG] | ‘worried, feel badly, sad’ |
| PT | Motu | lalo-metau | [insides-heavy] | ‘unwilling’ |
| PT | Dobu | nua- i-mʷau | [mind- it-heavy] | ‘downhearted’ |
| PT | Kilivila | mʷau nano | [mind heavy] | ‘sad’ |
| PT | Gumawana | nuo- i-mou | [mind- it-heavy] | ‘sad’ |
| PT | Iduna | -nua-vita | [-mind-heavy] | [VI] ‘heavy-hearted, sad’ |
| SES | To’aba’ita | manata- e kuluʔa | [mind- it heavy] | ‘unhappy’ |
| Fij | Bauan | loma-bībī | [insides heavy, difficult, painful] | ‘sad’ |
The Polynesian terms typically use a modifier reflecting POc *mapat ‘heavy’ (vol.2:213), without body part.
| Pn | Tongan | mafas-i-a | ‘be weighed down, burdened, literally or figuratively’ |
| Pn | Samoan | mafat-i-a | ‘be hurt, affected, physically and mentally tired’ |
| Pn | Tahitian | fātaī | ‘to be depressed, yield to discouragement’ |
Terms for ‘sad’ often overlap with those for ‘angry’ (§3.6). A second pattern uses ‘insides’ + ‘bad’ for ‘sad’ and a range of ‘feeling bad’ emotions that includes sadness and anger. In some cases the ‘bad’ word is a reflex of POc *saqat ‘bad’ (§11.6.2):
| NNG | Takia | ilo-saian | [insides bad] | ‘sorrowful, angry, sad, feel badly’ |
| NNG | Mutu | lolo i-saɣat | [insides it-bad] | ‘sad, upset’ |
| SES | Longgu | taʔa kutu | [bad belly] | ‘be sad, anxious, worried, feel sorry’ |
| SES | ’Are’are | rae- e taʔa | [liver- it bad] | ‘angry’ |
In other cases another term for ‘bad’ is used:
| NNG | Takia | tiŋae- saian | [guts- bad] | ‘angry, furious, very annoyed’ |
| NNG | Manam | ilo i-goala | [insides it-bad] | ‘sad’ |
| NNG | Mapos Buang | aʁe nipaya | [insides bad] | ‘sad, unhappy, irritated, angry’ |
| PT | Dawawa | nua-gewa | [mind bad] | ‘extremely sad’ |
| PT | Misima | nua-nak | [mind-bad] | ‘sad’ |
| MM | Nakanai | ilo-ruru | [insides-wrong] | ‘mournful, sad, disturbed, upset’ |
| MM | Nehan | bala uasa | [stomach bad] | ‘upset, sad, incorrect’ |
| MM | Maringe | diʔa nañafa | [bad heart] | ‘sad, regretful, feel bad, sorry’ |
| NCV | Lewo | sine- vioa | [guts- bad] | ‘sad, unhappy, disappointed, sorry, upset’ |
| PT | Iduna | nua- gi-koyo | [mind- it-bad] | ‘upset, angry, annoyed’ |
| NCV | Paamese | ti- tīsa | [guts- bad] | ‘angry’ |
A less frequent but widely distributed BPM for ‘sad’ or ‘angry’ is ‘insides’ + ‘sick/painful’:
| NNG | Takia | ilo- madai | [insides- painful] | ‘angry, bitter’ |
| NNG | Lukep | lo- matamata | [insides- sick] | ‘despondent, depressed’ |
| PT | Tawala | nugo-totogo | [mind- sick] | ‘sad’ |
| PT | Misima | ati-lomʷan | [liver-painful] | ‘be sad; feel sorry for’ |
| MM | Patpatar | bala ŋuŋut | [stomach painful] | ‘angry, sore at s.o., disgruntled’ |
| NCV | Lonwolwol | lo- makenken | [insides- painful] | ‘sad’ |
| NCV | Lonwolwol | lɔ- merā | [insides- sore] | ‘angry’ |
No reconstructions are proposed. Languages typically have a number of terms for various kinds of anger varying with the intensity of emotion and the construal of the triggering event. The following BPMs are from Yabem (NNG):
| NNG | Yabem | tɪtaʔ gɪôʔ auʔ | [belly.his grown.over PERF] | ‘he is full of rage, cannot think clearly because of rage’ |
| NNG | Yabem | tɪtaʔ kɪmʷatiŋ tau | [belly.his knot itself] | ‘he is burning with rage; is angry, irritated’ |
| NNG | Yabem | tɪtaʔ kɪpʷa | [belly.his explode] | ‘his blood is up, swells with rage’ |
| NNG | Yabem | tɪtaʔ kɪbuli auʔ | [belly.his twisted PERF] | ‘’his heart is enraged, irritated, takes offence at s.t., feels scandalised by s.t.’ |
| NNG | Yabem | tɪtaʔ ŋamakiʔ | [belly.his bitter] | ‘he is angry, bitter, irritated’ |
| NNG | Yabem | tɪtaʔ ŋandaŋ | [belly.his hot] | ‘he is furious’ |
| NNG | Yabem | tɪtaʔ seʔ | [belly.his bad] | ‘he is discontented, displeased, angry, dismal, sad’ |
Anger often overlaps with sadness insofar as BPMs of the pattern ‘insides’ + ‘bad’ and ‘insides’ + ‘painful’ mean either ‘angry’ or ‘sad’ or both. Examples are given in §11.4.5.
The examples below, although ranging in meaning from ‘cross’ to ‘indignant’ to ‘furious’, all contain metaphors that relate to heat or fire or its consequences.
| Adm | Nyindrou | drine- i buku jih | [stomach- it burn fire] | ‘get hot with anger’ |
| NNG | Takia | bube- yai inani | [liver- fire cook] | ‘very angry’ |
| NNG | Takia | ilo- wananan | [insides- hot] | ‘indignant, cross, angry’ |
| NNG | Bukawa | ataʔ ŋade | [liver -his hot] | ‘angry’ |
| PT | Motu | lalo-siahu | [insides-hot] | ‘angry’ |
| MM | Nakanai | la hate-la mamasi | [the liver-his salty] | ‘he is angry’ |
| MM | Patpatar | bala mamahien | [stomach hot] | ‘very angry’ |
| SES | Lau | lio e sasu | [voice it smoke] | ‘angry’ |
| SES | ’Are’are | rae- e kora | [liver- it embers] | ‘angry’ |
| NCV | Araki | lolo koru | [insides burnt] | ‘angry’ |
| NCV | Lonwolwol | lɔ- mafrī | [insides- flaming] | ‘angry’ |
| NCV | Lonwolwol | ləl-faŋfaŋ | [insides-on.fire] | ‘angry’ |
Terms from three languages support a PEOc reconstruction:
| PEOc | *lole | ‘be confused’ | |
| SES | ’Are’are | rore | ‘be confused, talk confusedly’ |
| SES | Sa’a | lole | ‘be confused, dazed’ |
| Pn | Māori | rore | ‘intoxicated; entangle’ |
Some languages, like Maori above, express mental confusion by describing the mind as tangled (as of vegetation), while two SES languages below describe the mind as closed or blocked:
| NNG | Takia | ilo- i-balkaluk | [insides- it-tangled] | ‘confused’ |
| PT | Minaveha | nua- vi-tupatupa | [mind- it-dense] | ‘confused’ |
| MM | Tolai | puruai | ‘be tangled, confused, puzzled’ | |
| SES | Longgu | bono-bou | [head-blocked (uncleared of bush)] | ‘confused’ |
| SES | To’aba’ita | qasifono | [VI] ‘be very confused’ (qasia ‘particle intensifier’, fono ‘be closed’) | |
| SES | Kwaio | filu | ‘tangled, confused, perplexed’ |
Not unexpectedly a number of terms for ‘surprised’ are perceived as associated with fear and used also for ‘alarmed’ or ‘shocked’. In both the Pn and Mic cognate sets given below there is also an association with being woken suddenly.
A recurrent metaphor incorporates words for jumping or flying using reflexes of POc
FIXME: what to do with next line, protoform without protolang *Ropok ‘to fly’ (§6.3.2.1. Micronesian and Polynesian terms occur without body part.) NNG: Lukep kate rō [liver flew] ‘excited to the point of forgetting what one was doing’ PT: Kiriwina i-yowa lopo-la [it-flew belly-his] ‘he leapt in surprise’ PT: Kukuya viau novo ‘frighten, ambush, surprise s.o.’ (viau ?)
| PPn | *ofo | ‘wake up, be startled’ (POLLEX) | |
| Pn | Tongan | ofo | [VI] ‘to be surprised, wake up’ |
| Pn | Niuean | ofo | ‘to surprise, be surprised’ |
| Pn | East Futunan | ofo | ‘wake up’ |
| Pn | Samoan | ofo | ‘startled, surprised’ |
| Pn | Tokelauan | ofo | ‘amazement, surprise’ |
| Pn | Māori | oho | ‘start from fear, surprise; wake up’ |
| Pn | Hawaiian | oho | ‘leap up, as startled birds’ |
Nakanai uses the same metaphor with a non-cognate term for the verb:
| MM | Nakanai | la-hate-la raga | [liver leaps] | ‘he is startled’ |
| POc | *(lalo-) -rutu | ‘surprised’ | |
| NNG | Takia | ilo i-rut | ‘surprised, fearful, trembling inside’ |
| PMic | *rut(i,u) | ‘become aware, wake up, be surprised’ (Bender et al. 2003) | |
| Mic | Kiribati | uti | ‘to arise, awake’ |
| Mic | Marshallese | ruc | ‘wake up, arouse’ |
| Mic | Kosraean | lut | ‘be surprised, startled, amazed’ |
| Mic | Puluwatese | rɨ | ‘to be startled, surprised, alarmed’ |
| Mic | Carolinian | rɨ | [VI] ‘to be surprised, shocked, startled’ |
| Mic | Satawalese | rrɨ | ‘surprised’ |
| Mic | Woleaian | rʉsʉ | ‘be frightened, scared’ |
We have included desire and wanting in this chapter because in some contexts (being envious, homesick) it has a strong emotional basis.7 In others, of course—lacking food (§§4.3.3.1–2) or sleep (§4.6.2), sexual desire (§4.2.2.2)—it denotes a physical rather than emotional need.
The most striking result of our searches for ways in which POc speakers expressed ‘want’ and ‘desire’ is the absence of any consistency of expression in modern languages outside the Central Pacific (Fij + Pn) group. A major reason for this is that, as a result of the ubiquity and frequency of ‘want’ in the world’s languages, ‘want’ words tend strongly to undergo grammaticisation and to end up as particles in the slots otherwise reserved for tense, aspect and especially mood markers. The endpoint of this tendency is that the ‘want’ morpheme undergoes extension of function and becomes a future or irrealis morpheme, accompanied by the innovation of new ways of expressing ‘want’. Thus in Takia (NNG) the ‘want’ morpheme is the enclitic *=[w]o, which occupies the first of a series of postverbal enclitic slots (Ross 2008) that are otherwise occupied by aspect or mood morphemes:
| NNG | Takia | ŋ-le=o | [I-see=want] | ‘I want to see’ |
| NNG | Takia | ŋ-le=da | [I-see=imperfect] | ‘I am seeing’ |
| NNG | Takia | ŋ-le=ya | [I-see=realis] | ‘I saw’ |
| NNG | Takia | ŋ-le=wa | [I-see=irrealis] | ‘I shall see’ |
In Longgu (SES) the ‘want’ morpheme is preverbal tali, in the same structural position as certain aspectual morphemes and the negator (Hill 1992):
| SES | Longgu | tali inu | [want drink] | ‘(I) want to drink’ |
| SES | Longgu | tazani tate | [just appear] | ‘(I) have just appeared’ |
| SES | Longgu | vusi aŋi | [almost cry] | ‘(I) am almost crying’ |
| SES | Longgu | se lae | [not go] | ‘(I) am not going’ |
A number of languages of the Southeast Solomons, Fiji and Polynesia are like Longgu: they have a pre-verbal particle or prefix meaning ‘want’, but the SES and Wayan Fijian forms show no relation to each other, nor to PCP *via below.
| SES | ’Are’are | siri | ‘hanker after, long for, desire’ (siri hana ‘very hungry’, siri koʔu ‘thirsty’) |
| SES | Arosi | gasi- | ‘to desire, desiderative prefix to any verb’ (gasi-gono ‘thirsty’, gasi-maura ‘sleepy’, gasi-ŋau ‘hungry’) |
| SES | Kwaio | māli- | ‘prefix’ (māli-faŋa ‘hungry’ and māli-goʔu ‘thirsty’) |
| Fij | Wayan | mata | ‘preverbal particle, want, desire, feel need to do V’ (mata kani ‘hungry’, mata som ‘thirsty’, mata moðe ‘sleepy’) |
Data supporting the reconstruction of PCP *via ‘want to’ are given below.
| PMP | *pian | ‘want, desire, wish or long for’ (ACD) | |
| POc | *pia(n) | ‘want to’ | |
| PCP | *via | ‘desiderative particle or prefix’ | |
| Fij | Bauan | via | ‘auxiliary verb expressing desire’ |
| Fij | Wayan | via | ‘preverbal particle: marks an act as done for fun, pleasure rather than for serious purpose; want to do’ |
| Pn | Tongan | fie | ‘preposed verb; want, desire, wish, be willing’ |
| Pn | Niuean | fia | ‘desire, want’ |
| Pn | Samoan | fia | ‘pre-verbal particle: wish, like, aspire to’ |
| Pn | East Futunan | fia- | ‘verbal prefix indicating wish, desire’ |
| Pn | Pukapukan | wia | ‘want’ |
| Pn | Pileni | fie | ‘preverbal adverb indicating a wish or need’ |
| Pn | Tokelauan | fia- | ‘prefix indicating a wish, a liking’ |
| Pn | Tikopia | fia, fifia | ‘want, desire, wish (normally followed by common verbs or nouns, giving unitary concepts)’ |
| Pn | Tahitian | hia-ai | ‘desire food, drink’ |
| Pn | Māori | hia | ‘desire, want’ (prefix on small group of words such as eat, drink etc.) |
| Pn | Hawaiian | hia | ‘desire, want, delight in’ |
| Pn | Marquesan | hia moe | ‘sleepy’ |
| SES | To’aba’ita | fiia | [VT] ‘expect, anticipate s.t.; have a feeling that s.t. will happen; expect s.o. to do s.t.’ |
| NCV | Lewo | ve | ‘want’ |
As noted in §§4.3.3.1–2 and §4.6.2.1 Central Pacific languages express the concepts of being hungry, thirsty and sleepy as sequences of ‘want to’ + verb:
Since PCP *via apparently reflects PMP *pian, the reconstruction of POc *pia(n) can be inferred. However, it seems to have been displaced in non-Central Pacific Oceanic languages by a variety of lexical strategies.
A range of languages use reflexes of POc *mate ‘to die’ (§4.2.1.2) to express an intense need, particularly for such things as food, water, or betelnut, paralleling the English expressions ‘dying for a smoke’ etc.
| NNG | Takia | you=o -mat | [water=for -die] | ‘thirsty’ |
| NNG | Gedaged | -mat | ‘long, yearn, crave, desire, lust after’ | |
| NNG | Mengen | mate-ka- | [die eat] | ‘want food, be hungry’ |
| NNG | Kakuna | mate-kana | [die eat] | ‘hungry’ |
| NNG | Uvol | mete-ana | [die eat] | ‘hungry’ |
| SES | Lau | mae-li gwou | [die-TR water] | ‘long for a drink, be thirsty’ |
| SES | Kwaio | mā-li faŋa | [die-TR food] | ‘long for food, be hungry’ |
In a number of languages, desire is strongly identified with the reflex of POc *lalo-, *lalom ‘inside; seat of thoughts and emotions’ or whatever has replaced it as the term for ‘mind’ (§9.4). The following example from Bugenhagen’s Mangap-Mbula grammar (1995:223) illustrates how this works (or in some languages, once worked):8
| Nio | lele-ŋ | be | aŋ-la | pa | ᵐbeŋ | |
| I | insides-my | NF | I-go | at | night |
A more literal translation is ‘My inside/thought/desire (was) that I would go at night.’ This Kalokalo (PT) sentence has a similar structure (Guderian & Guderian 2002).
| nuanua-gu | ya-na-egimʷaneye-ya | |
| want-my | I-will-sell-it |
In some languages the ‘mind’ noun has been (half-)transformed into a verb, as in Lewo (NCV), where the transitive verbal suffix -ni is attached to the ‘mind’ noun sine- ‘guts’ and its possessor suffix, in this instance -la ‘their’ (Early 1994a).
| sine-la-ni | ∅-sape | Palua | ∅-va | e | wa | |
| guts-their-TR | it-say | Palua | he-iRREALIS.go | to | ship |
Ivens (1937) shows that the Gela ‘mind’ term lio- is similarly used. It takes a possessor suffix like any inalienably possessed noun, but is accompanied by the verbal morphemes that one would expect with a verb like ‘want’.
Finally, another strategy for expressing ‘want’ is to use the verb ‘say’, a natural extension of internal speech and thought. In Bariai (NNG) keo serves as both ‘say’ and ‘want’, but its sense is disambiguated by the construction that follows it (Gallagher & Baehr 2005).
| Ti-keo | pa=gid | taine | ngan | ti-la | dadanga-i. | |
| they-say | to=them | female | to | they-go | garden-at |
| Na-keo | ga | sabale | gaisala | eao | Ø-nam. | |
| I-say | that | tomorrow | morning | thou | thou-come |
Bowern (2011:151) notes a similar development in Titan (Adm).
Because what is experienced may be animate, inanimate or abstract and because evaluation is in the mind of the experiencer, evaluative terms, particularly those for ‘good’ and ‘bad’, may be applied to both animates and inanimates. As value terms they stand alone, but they are also a component of BPMs (§9.5) or they follow terms used specifically to mark a quality as customary or habitual (§11.3).
Terms for ‘good’ have been difficult to reconstruct. Two reconstructions are proposed: POc *puia and PEOc *leka.
With regard to *puia, originally an alternant *uia was reconstructed. The reflexes that require this are those from Takia and its neighbours and Hote, shown under ‘cf. also’. These appear to be instances of an idiosyncratic sound change in a frequently used item. However, the Loniu, Titan and Sio reflexes are those that would have occurred if *p were between two vowels. and it is reasonable to infer that in these languages, at least, the reflex of *puia once behaved as a stative verb and took subject prefixes (e.g. i- 3SG). The Gitua and Labu forms and the second Nehan form reflect *puaia, and this was perhaps a POc alternant.
The form *puia is unusual because it contains a sequence of three vowels, which is very unusual in POc (otherwise occurring, as far as is known, only in *kaiu ‘tree’). Indeed, it is tempting to reconstruct either †*pʷia or †*puya in order to adhere to a POc canonic shape, but †*pʷia is eliminated by Loniu, Kaiep, Manam and Nehan reflexes, since *-u- is retained and *puya is hard to justify in the light of so many reflexes of *-i- rather than *-y-.
The presence of *-u- in *puia is also unexpected, as non-Oceanic cognates reflect PMP *ma-pia, i.e. the root is *pia.
| PMP | *ma-pia | ‘good’ | |
| POc | *puia | ‘good’ | |
| Adm | Loniu | huya-n | ‘good’ (medial reflex of *p-) |
| Adm | Titan | wia-n, uya-n | ‘good’ (medial reflex of *p-) |
| Adm | Lou | pia-n | ‘good’ |
| Adm | Baluan | pia-n | ‘good, well, all right’ |
| NNG | Ulau-Suain | ya-ñ | ‘good’ |
| NNG | Kaiep | uya-n | ‘good’ |
| NNG | Manam | (ia)uia | ‘good’ |
| NNG | Dami | bia | ‘good, right, correct, righthand side’ (b- for †p-) |
| NNG | Mutu | poia | ‘good’ |
| NNG | Gitua | pʷaya | ‘good’ (for †pʷia) |
| NNG | Lusi | poea | ‘good’ |
| NNG | Mengen | pe | ‘good; right hand’ |
| NNG | Kilenge | pa-pue | ‘good’ |
| NNG | Mangap | pe | ‘good, well’ |
| NNG | Sio | wia | ‘right hand’ (medial reflex of *p-) |
| NNG | Kaiwa | vie | ‘good’ |
| NNG | Numbami | wia | ‘good’ |
| NNG | Labu | haya | ‘good’ (for †hia) |
| SJ | Sobei | fia | ‘good’ |
| PT | Gumawana | uya(wana) | ‘happy, pleased’ |
| MM | Nehan | uia | ‘good’ |
| MM | Nehan | uaia | ‘correct’ |
| MM | Tinputz | vi(h) | ‘good’ |
| PNCV | *vuia | ‘good’ (Clark 2009: *wia) | |
| NCV | Mota | wia | ‘good, of the right sort, without anything unusual’ |
| NCV | South Gaua | we | ‘good’ |
| NCV | Mwotlap | wɪ | ‘good’ |
| NCV | Tolomako | wia | ‘good’ |
| NCV | Lonwolwol | wu | ‘good’ (realis: bu) |
| NCV | Port Sandwich | voi | ‘good, well, pleasant’ |
| NCV | Nakanamanga | pʷia | ‘good’ (realis form?) |
| NCV | Nguna | wia | ‘good’ |
| NCV | Lelepa | wia | ‘good’ |
| NCV | South Efate | wi | ‘good’ |
| NNG | Takia | uya-n | ‘good’ |
| NNG | Bilibil | uya-n | ‘good’ |
| NNG | Matukar | uya-n | ‘good’ |
| NNG | Mindiri | uya-n | ‘good’ |
| NNG | Misim | (ma)ui | ‘good’ |
Reflexes of PEOc *leka ‘good, pleasant’ may also refer to temperaments, as in Pukapukan (yau)leka [temperament good] ‘calm, gentle, mild, quiet’.
| PEOc | *leka | ‘good’ | |
| SES | Lau | lea | ‘good’ |
| SES | Kwaio | leʔa | ‘good, well’ |
| SES | Dori’o | leʔa | ‘good’ |
| SES | Kwara’ae | leaʔ | ‘good’ |
| PPn | *leka | ‘pleasant’ (POLLEX) | |
| Pn | Pukapukan | leka | ‘pleasant, sweet, good, delicious’ |
| Pn | Tikopia | (tau)reka-reka | ‘fine, splendid, handsome, beautiful’ |
| Pn | Rarotongan | reka | ‘pleasant’ |
| Pn | Māori | reka | ‘pleasant, sweet’ |
| Pn | Hawaiian | leʔa | ‘pleasant’ |
The POc term for a negative evaluation of various kinds was *saqat ‘bad’.
| PMP | *zaqat | ‘bad’ (ACD) | |
| POc | *saqat | ‘bad’ | |
| NNG | Takia | saia-n | ‘bad’ |
| NNG | Yabem | seʔ | ‘bad, evil’ |
| NNG | Tami | sakat | ‘bad, spoilt’ |
| MM | Bali | zaɣata | ‘bad’ |
| MM | East Kara | (mo)sat | ‘bad’ |
| MM | Notsi | caka | ‘bad’ |
| MM | Tabar | caka | ‘bad’ |
| MM | Label | saka | ‘bad’ |
| MM | Patpatar | sakana | ‘bad; evil; ruined; worthless’ |
| MM | Minigir | saka(i) | ‘bad’ |
| MM | Tolai | ka(i) | ‘bad’ |
| MM | Ramoaaina | akə | ‘badly’ |
| MM | Teop | hata | ‘bad’ |
| MM | Torau | saka(ala) | ‘bad’ |
| MM | Vaghua | sata | ‘bad’ |
| MM | Varisi | sakata | ‘bad’ |
| SES | Lengo | ðaɣata | ‘bad’ |
| SES | Longgu | taʔa | ‘bad’ |
| SES | To’aba’ita | taʔa | [VI] ‘be bad, no good; feel bad, physically, mentally, emotionally’ |
| SES | Lau | tā | ‘bad, evil’ |
| SES | ’Are’are | taʔa | ‘wrong, bad, evil; be dying, be in a bad condition’ |
| SES | Arosi | taʔa | ‘bad, poor, of poor quality’ |
| NCV | Hiw | sa | ‘bad’ |
| NCV | Nese | sat | ‘bad’ |
| NCV | Nakanamanga | sa | ‘bad’ |
| NCV | Namakir | (a)haʔ | ‘bad’ |
| NCV | Nguna | sā | ‘bad’ |
| SV | Sye | sat | ‘badly; problem, trouble’ |
| NCal | Xârâcùù | xyā | ‘bad’ |
| Fij | Bauan | ðā | ‘bad, evil’ |
| Pn | Samoan | sā | ‘forbidden, sacred’ |
| Pn | West Futunan | sa | ‘bad’ |
A number of reflexes point to a final *-i. Whilst the Arosi and Bauan reflexes under ‘cf. also’ self-evidently reflect *saqat plus the transitive suffix *-i, the gloss ‘bad’ indicates that this is not the source of *-i in the items listed below and that they perhaps reflect an alternant *saqati.9
| POc | *saqati | ‘bad’ | |
| NNG | Malalamai | sati | ‘bad’ |
| NNG | Mutu | saɣati | ‘bad’ |
| NNG | Gitua | saɣati | ‘bad’ |
| NNG | Kove | sasi | ‘bad’ |
| NNG | Aria | sasi | ‘bad’ |
| NNG | Atui | ses | ‘bad’ |
| NNG | Avau | ses | ‘bad’ |
| PT | Sudest | ðari | ‘bad’ |
| MM | Lihir | caket | ‘bad’ |
| PNCV | *saqati | ‘bad’ (Clark 2009) | |
| NCV | Tamambo | sati | ‘bad; dead (euphemism)’ |
| NCV | Tangoa | sati | ‘bad’ |
| NCV | Nduindui | hati | ‘bad’ |
| SV | Anejom̃ | has | ‘bad’ |
| SES | Arosi | taʔa-i | [VT] ‘to spoil’ |
| Fij | Bauan | ðāti- | [VT] ‘to hate s.o.; deem s.o. bad’ |
POc *jika ‘be soiled, weakened’ appears to have been a stative verb used primarily of inaminates and meaning something like ‘be unfit for use’. But some languages extend their reflexes of *jika to describe negative emotions or behavioural qualities:
| PT | Motu | kara dika | [conduct bad] | ‘sin’ |
| PT | Motu | gaiho dika | [character bad] | ‘inhospitable, mean’ |
| PT | Motu | lalo dika | [insides bad] | ‘miserable’ |
| MM | Maringe | diʔa na̰ñafa | [heart bad] | ‘sad, sadness’ |
| SES | Gela | lio dika | [disposition bad] | ‘sad, sorry; to hate’ |
| SES | Bugotu | dika hehe | [bad heart/mind/wish] | ‘grief, to grieve, be sad; bear ill will’ |
| POc | *jika | ‘be soiled, weakened’ | |
| NNG | Lukep | sia(na) | ‘bad’ |
| NNG | Maleu | sia(ŋe) | ‘bad’ |
| NNG | Manam | ziʔa-ziʔa | ‘dirty, soiled’ |
| NNG | Bam | jik-jik | ‘dirty’ |
| NNG | Wogeo | -jika | ‘(wood) rotten’ |
| NNG | Kairiru | -jieq | ‘(wood) rotten’ |
| PT | Motu | dika | ‘bad, badness; calamity; guilt’ |
| MM | Marovo | cie-na | ‘bad’ |
| MM | Vangunu | sie-na | ‘bad’ |
| MM | Kokota | dia | ‘bad’ |
| MM | Maringe | diʔa | ‘bad’ (borrowed from Bugotu) |
| SES | Bugotu | dika | ‘be bad, evil, wrong’ |
| SES | Gela | dika | ‘bad, inferior’ |
| Fij | Bauan | ðika(i) | ‘be destroyed, be weakened’ |
| Fij | Bauan | ðika(a), ðika(va) | ‘destroy s.t.’ |