Words for ‘Bird’ in Eastern Indonesia

Nage (central Flores) Ana Wa Ta'a Co

In a recent book on Nage ethno-ornithology, I described the Nage term as a reference to a ‘covert taxon’ (Forth 2004a:36) in view of the fact that ana wa ta'a co —the least equivocal way of designating birds in general, and a category inclusive of all locally known aviformes—is a lengthy, descriptive phrase. I would now regard this point as moot, and am more inclined to accept ‘bird’ as a named life-form in Nage. Meaning ‘animal ( ana wa ; see Forth 2004b) that flies ( ta'a co )’, the Nage usage identifies birds by their characteristic form of locomotion. As some other creatures—most notably, insects—are also capable of flight, the category can be more exactly specified as ana wa ta'a co zéta lizu (‘animals that fly high in the sky’). As Nage remark, birds are further distinguished from flying insects by the fact that insects that fly also regularly move by ‘crawling’ ( laka ), and some by ‘jumping’ ( bedhi ). Birds can alternatively be described morphologically, as ‘animals with wings’ (Forth 2004a:36), but this is less common and moreover focuses on the feature essential to the characteristic form of locomotion.

Two other Nage usages referring to birds in a general sense are ana peti , a term that more specifically denotes small passerine birds and especially Estrildine finches, and peti kolo , a compound comprising the short form of ana peti and kolo , the name of certain smaller Columbiformes.3 However, these terms refer generally to birds only in particular contexts, where they are often better glossed as ‘bird of no particular kind’ or ‘any kind of bird’ (Forth 2004a:37-38). The restriction is demonstrated by direct questioning, which reveals that Nage do not normally consider eagles ( kua ), owls ( po ), or ducks ( bébe ae ), for example, as instances of ana peti or peti kolo , even though for Nage all are certainly ‘birds’, and all are recognized as instances of ‘animals that fly high in the sky’ ( ana wa ta'a co zéta lizu ).

Eastern Sumbanese Mahawurungu and Manggarai (western Flores) Kaka Lélap

These terms closely correspond to the Nage term for ‘bird’ in that both translate as ‘what flies’ or ‘flying thing’. Sumbanese ma - is a nominalizing particle, while hawurungu , like Manggarai lélap , means ‘to fly’. Since Verheijen glosses kaka as ‘animal’, the Manggarai expression might appear to correspond exactly to the Nage term. Yet however regularly it may be employed with the specific meaning of ‘animal’, kaka possesses the more general sense of ‘thing’, of which the meaning ‘living, animate thing’ is apparently derivative (Forth 2004b:60). A noteworthy feature of Manggarai kaka is its occurrence in numerous folk generic terms, not only for categories of birds but for other kinds of animals as well. Thus, kaka ketoq , denoting the Sunda pygmy woodpecker (Dendrocopos moluccensis), might be interpreted as ‘bird, or animal, that knocks, taps (on wood)’. Nevertheless, the term might equally, and perhaps more accurately, be construed as ‘tapping thing’, or ‘tapper’.4

Closely resembling Manggarai kaka in its range of applications is kokaq , a word from Rembong dialects spoken to the northeast of Manggarai, which Verheijen (1977) lists inter alia as ‘bird’, ‘animal’ (see kokaq reman , ‘wild animal’), and ‘object, thing’. Unlike kaka , however, kokaq is employed without modification in one Rembong dialect to refer to a particular category of fauna, and moreover a particular category of birds, the Estrildine finches.5 At the same time, the term occurs in four compounds labeling more specific bird categories (e.g., kokaq-but , denoting a quail), and a further four designating kinds of mammals (e.g., kokaq-béwéq , wild cat), a snake, and an insect. A very similar usage is suggested by bird terms from Tana Wolo, a district located just east of Manggarai and north of the Ngadha region (see Forth 2004a:xiv, Map 2), in regard to the lexeme koka , glossed by Verheijen as ‘Estrildine finches’. In Tana Wolo, six bird generics are labeled with compounds of koka , although these subsume quite different kinds from the Rembong kokaq compounds (namely, species of hawks of the genus Accipiter, the Large-billed Crow, Channel-billed Cuckoo, Savannah Nightjar, Sunda Pygmy Woodpecker, and wagtails). Rembong kokaq and Tana Wolo koka bear a palpable resemblance to names for the Helmeted Friarbird (Philemon buceroides) in other eastern Indonesian lauguages, including Nage, Ngadha, and eastern Sumbanese koka . Yet in view of their ornithologically very different referents, the Rembong and Tana Wolo terms are better understood as cognates of Manggarai kaka .6

The kinds of birds whose Manggarai names include kaka are as various as are the species associated with any of several comparable lexical sets comprising terms that incorporate manu and kolo (see Tables 2 and 4). Unlike these, however, and unlike Rembong and Tana Wolo kokaq and koka , Manggarai kaka is not identified with any more specific type of bird, nor does it unequivocally refer to birds in general, or some relatively large sub-set of these. Indeed, when used by itself, kaka is usually understood as ‘animal’ (Forth 2004b:59).

OPEN IN VIEWER

Table 2

Instances of kolo and cognates in eastern Indonesia.

Both Manggarai kaka lélap and eastern Sumbanese mahawurungu (‘flying thing’) include bats as well as birds, as does the Nage term ana wa ta'a co . Because in all three cases the life-form is explicitly defined with reference to a kind of movement, the fact that it subsumes the Chiroptera as well as Aviformes is perfectly logical. By the same token, it would not be entirely accurate to say that the Nage, for example, classify bats as birds (that is, within a taxon equivalent to modern English ‘bird’ or scientific ‘Aves’), but only that they group them conceptually with birds.

Lio Ule Bene or Ule Age and Rembong Anak Reman

The evidence for a term labeling an ethnotaxon ‘bird’ in Lio is less straightforward than what is found in other Bima-Sumba languages. Some evidence suggests that Lio ule can be used as a general term for ‘animal’ (Forth 2004b:55-57). In all other Flores languages, ule and cognates denote a category of worms, maggots, and grubs—a particular sense also associated with the term in Lio.7 Entries in Arndt's Lio dictionary (1933) further suggest that, in some contexts, ule alone can mean ‘bird’. Thus Arndt glosses ule léla dzére as ‘the bird flies suspended, hovers’ (s.v. dzére , ‘to hover’; léla is ‘to fly’, cf. Manggarai lélap ) and lists haba ule as ‘bird's nest’ (1933:132, s.v. haba , ‘(bird's) nest', cf. haba manu , ‘hen's nest'). Lio ule moreover occurs in compounds referring to at least six bird generics, including ule mesi , ‘heron’ (Stokhof 1983; mesi is ‘sea’) and ule miu , designating a bird that shrieks ‘miu’ (Arndt 1933). In these designations, of course, ule could just as well be interpreted as ‘animal’ instead of ‘bird’, in a way comparable to Manggarai kaka . Ule further appears in five compounds that apply to insects—as in ule ae , ‘dragonfly’ ( ae , ‘water’), and ule api , ‘wasp with a red body’ ( >api , ‘fire’). Yet with regard to the particular association of ule with birds, it is a point of interest that at least four of the five refer to relatively large flying insects.

The evidence for a Lio compound term denoting birds in general, and distinguishing these from other zoological life-forms, is equivocal. Two Lio phrases listed by Arndt that might possess this sense are ule bene and ule age . In one place, however, Arndt glosses ule age as ‘all kinds of creatures, worms, reptiles, birds’ (1933, s.v. ule ), thus suggesting a more general zoological reference. Drawing on more recent research, the anthropologist Sugishima (personal communication 2000) describes ule age as a term specific to birds in Lio. But neither Takashi Sugishima nor Arndt provides a separate gloss for age , and it would seem that the term may only occur in combination with ule .8

In ule bene , the other Lio term possibly specifying ‘bird’, bene translates as ‘weeds, bush’ (Arndt 1933). Although Sugishima never recorded ule bene (personal communication 2000), nor had a Lio informant I questioned in 2005 heard of it, Arndt's gloss of the second component accords with Sugishima's exemplification kuru bene , ‘wild plants’. In Nage, too, bene can mean ‘wild, undomesticated’. Thus, if it is a regular Lio usage, ule bene may be understood as a reference to creatures inhabiting wild vegetation, and thus, as a synecdochal usage specifying wild birds.9 Indeed, especially among the usages included in Arndt's dictionary, there is some suggestion that, in spite of comparative evidence indicating an original meaning of ‘worm, grub’, ule alone, as a reference to wild (as opposed to domesticated) creatures, may have a special association with birds. Expressed another way, the suggestion is that that wild birds are the focus of a Lio category of ‘wild creatures’ designated by ule .

Since the second element of Lio ule bene means ‘weeds’ or ‘scrub’, the expression partly parallels anak reman , listed by Verheijen (1977, s.v. anak ) as the term for ‘bird’ in the Rembong region of western Flores (see Forth 2004a:xiv, Map 2). Reman means ‘leaf (leaves), grass, weeds’, ‘undergrowth, scrub’, and ‘forest’ (Verheijen 1977), and like comparable terms in other eastern Indonesian languages denoting types of uncultivated vegetation (e.g., Nage witu , eastern Sumbanese rumba , and of course, Lio bene ), the Rembong term serves to distinguish wild from domesticated creatures (Verheijen 1977).10 Anak , on the other hand, is cognate with Nage ana (‘child, person, member’). Hence, Rembong anak reman further recalls the Nage term for animals in general ( ana wa , analyzable as ‘children, people of the wind’) as well as a contextual Nage use of ana alone in reference, especially, to birds (Forth 2004a:56).11 As with Lio ule bene , therefore, in the Rembong term anak reman one encounters a usage suggesting the general sense of ‘wild animal’ but which evidently refers more specifically to (wild) birds.

Insofar as the Lio and Rembong usages appear to refer ambiguously to birds and a more inclusive category of animals, another comparison may be found in certain Oceanic languages, where terms cognate with Central Malayo-Polynesian manu (or man , manut , etc.; see the third subsection of ‘Ambon-Timor Group of Languages’ below) denote ‘bird (in general)’ but are further applied to other animals (Blust 1982: 247-48; see also Table 3; notes 5 and 6). In these Oceanic languages, however, the ‘bird’ term has apparently been extended to certain non-birds, including mammals and fish, whereas there is no evidence for the Lio and Rembong phrases having originally referred exclusively to avifauna.

OPEN IN VIEWER

Table 3

Terms for ‘bird’ and ‘domestic fowl’ in various Malayo-Polynesian languages. NOTE: Names are reproduced in their original transcription. In some cases an older, Dutch-based orthography is employed, so that /j/ equates with modern /y/, /dj/ with /j/, and /oe/ with /u/. The exception is cognates of manu , where the /oe/ of Dutch lexicographers (as e.g., in Tanimbar ‘manoet’) has been replaced by /u/ (thus manut ).

OPEN IN VIEWER

Table 4

Manu compounds in eastern Indonesian languages. (For ease of comparison with Nage categories, English names have been substituted where an author gives only a Latin name.)

Lamboya (western Sumbanese) Mango; Ende (central Flores) Nake; and Bimanese (eastern Sumbawa) Nasi

Wielenga (1917:33) compares western Sumbanese (Lamboya) mango to eastern Sumbanese manginu , a term referring to a variety of small passerine birds, especially munias and other Estrildine finches (Forth 2000:176-77); but he also lists the Lamboya term as the equivalent of eastern Sumbanese mahawurungu , as a general term for ‘bird’. From this one may infer that, insofar as mango does refer to birds in general, it represents an extension from a primary or original denotatum, which was a variety of small birds, or what have been called ‘dicky birds’ (Hunn 1991:139). A relevant comparison is the contextual use of Nage ana peti , another term whose most specific referent is Estrildine finches, but which contextually is similarly applied to a wider category of avifauna.

In the first comprehensive lexicon recorded for the Ende region of central Flores, van Suchtelen (1921:316, 340, 389) lists the apparently monolexemic nake as the term for ‘bird’ in both the Ja'o and Nga'o dialects of Endenese. In the same vein, he gives sewo nake as ‘bird's nest’ (1921:316). Drawing on information published in 1896, the ‘Holle Lists’ (Stokhof 1983:164) also record nake as Endenese for ‘bird’.12 This early evidence finds some support in a recent dictionary of the Ja'o dialect by Aoki and Nakagawa (1993), who translate nake as ‘kind of bird’, a sense they also list for 'ana nake ( 'ana , ‘child’, denotes a small variety of something, as does its Nage cognate). These authors, however, further translate nake as ‘kind of parrot’ and ‘kind of hawk’, and with reference to the second gloss, they also record nake jata , ‘hawk’, and nake toro , ‘kind of bird of prey’ (the same definition given for jata toro ).13 The only compound incorporating nake found in van Suchtelen's list is nake buro , translated as ‘swallow’ in the Ja'o dialect (1921: 340).14 Surprisingly, Verheijen (n.d.) lists none of the foregoing names in the terminologies he records for various Endenese dialects. On the other hand, he does record nake teka as the Endenese name for the Sunda Pygmy Woodpecker (Dendrocopos moluccensis), and nake kowo as the term for ‘swallow’ (Hirundo spp.) in the Riti dialect of eastern Keo (Verheijen n.d.:14, 15).15 Also referring to Keo, Tule (1996) lists nake ‘ura , ‘rain bird’, as the name of an unidentified kind of small bird.

None of these more recent data confirms the status of nake as a general term for ‘bird’; but nor does any contradict van Suchtelen's gloss. The fact that nake seems to occur incidentally in Endenese folk generics for quite various kinds of birds (parrots, diurnal raptors, woodpeckers, swallows) is not contrary to the general sense of ‘bird’, any more than is the occurrence of English ‘bird’ in the names of a similarly limited number of diverse aviformes (see e.g., blackbird, frigatebird, lyrebird, mockingbird, ovenbird). In regard to the derivation of the term as a possible life-form label, it may be relevant that in other central Flores languages (including Nage, Ngadha, and Lio, but not certainly in Endenese) nake is a word for ‘meat’ and, in some languages, especially ‘game’ or ‘wild meat’. According to Arndt (1961:346), Ngadha nake refers to ‘game’ both in the sense of ‘wild meat’ and ‘game animal’, while in both Ngadha and Lio, nake occurs in compounds denoting aquatic creatures.16 Quite possibly, then, Endenese nake has come to be applied to birds synecdochally, by virtue of their identification as sources of wild meat or kinds of game—and thus in a way hypothetically comparable to Lio ule bene and Rembong anak reman , whose analysis similarly suggests a more general reference to wild animals.

Detracting from this hypothesis is the circumstance that people of central Flores do not consider all birds edible. Nor are those named with nake in Endenese (for example, raptors and parrots) at all typical of those that they do. Nage, for example, explicitly prohibit raptors as food (Forth 2004a:104). What is more, even the limited information available allows us to glimpse an alternative interpretation of nake . For two Endenese dialects, Verheijen (n.d.:19) lists ana ke kunu and ana ke jata as names of two kinds of Estrildine finches.17 As already shown, if not the sole referent, these sorts of small birds are focal to similar terms in several eastern Indonesian languages (including Nage ana peti , eastern Sumbanese manginu , and Rembong and Tana Wolo kokaq and koka ; see also Manggarai cik , ‘general name for almost all small birds’, Verheijen 1963; Komodo kesiq , ‘all small birds’, Verheijen 1982; Rembong sik , kokaq sik , ‘several kinds of small birds’, Verheijen 1977; Ngadha siu , Estrildine finches, Verheijen n.d.; Lio si , ule si , ‘kind of bird’, Arndt 1933; Sikanese ti , munias and various other small birds, Verheijen, n.d.). Although mostly focused on Estrildine finches, nearly all of these terms appear, in one way or another, to apply to a more inclusive category of small passerine birds, and, in the case of Nage ana peti , can even approximate ‘bird’ in a general sense. This comparative evidence, then, suggests that nake , ‘bird’, may not derive from a word meaning ‘meat’ or ‘game’, but may reflect a contraction of ana ke , or alternatively a short form of Aoki and Nakagawa's ‘ ana nake , a term they equate with nake . Noteworthy here is the latter authors’ separate listing—under kéé , ‘small’—of 'ana kéé as ‘kind of bird (very small)’ (1993:30). Whichever interpretation is correct, therefore, nake (or ana nake ) may be understood as another term denoting a more limited category of birds that has come to acquire something of the sense of ‘bird’ in general.

Essentially the same analysis possibly applies to nasi , the general term for ‘bird’ in Bimanese. It is a reasonable surmise that the term reflects a compounding of Bimanese ana (‘child, person’, cf. Nage and Endenese ana ) and a cognate of one or more of the series of names, listed just above, which are applied to small passerine birds in Flores languages (see especially cik , siu , and si ). By the same token, Endenese nake , ana ke , and Aoki and Nakagawa's 'ana kéé are obviously similar to ke , the term Fernandez (1996) gives for ‘bird’ in the Helong language, spoken on the island of Samau off the western tip of Timor. Not having access to other materials on Helong, I am unable to comment further on this name. It should be noted, however, that comparable Florenese terms like Manggarai cik , Sikanese ti , and so on, appear to be onomatopoeic in origin, and to that extent equivalent to English ‘chirp’, denoting the vocalization of a small passerine bird.

Palu'e Kolo

Although classified as a Bima-Sumba language (Wurm and Hattori 1981), the language of Palu'e, an island off the north coast of Flores, includes bird terms which in several respects are more reminiscent of bird terminologies from ‘Ambon-Timor’ languages spoken in more easterly parts of Flores. Consistent with this assessment is Vischer's statement, that Palu'e fauna includes ‘a large variety of birds collectively classified as kolo ’ (1992:18), an obvious cognate of the general term for ‘bird’ in Atoni (Dawan), a language of western Timor (see Table 1). In view of its widespread occurrence in eastern Indonesian ethno-ornithological lexicons, kolo is further discussed in the following section.

Sikanese Kenaha Horong (‘flying Thing’) or Horong (the Nominal Form of Horo, ‘to Fly’)

Although Sikanese belongs to the ‘Ambon-Timor’ languages, the Sikanese terms, both of which refer to birds in general (Pareira and Lewis 1998:79; Lewis personal communication 2001), are obviously comparable to the Nage, Manggarai, and eastern Sumbanese usages by virtue of their naming of the life-form with reference to its characteristic mode of locomotion. A further similarity lies in the fact that the Sikanese category, like the terms from Sumba and more westerly parts of Flores, includes bats as well as true birds. Also worth noting is the evident relationship between Sikanese horong (from horo , ‘to fly’), and the synonymous eastern Sumbanese hawurungu .18 Since ha - is a fused prefix, hawurungu is evidently related to wuru , a probable onomatopoeia referring to the whirring or flapping sound of a bird's wings in flight (Forth 2000; Onvlee 1984). Despite the apparent similarity, a connection between either of the eastern Indonesian terms and burung (or burong ), the term for ‘bird’ in Malay, Bahasa Indonesia, and several other Western-Malayo-Polynesian languages (see Table 3), is unlikely. According to Adelaar (2004: 50), burung derives from Proto-Malayo-Polynesian * buR'w , ‘to hunt’ (plus a fused suffix), an interpretation that in an interesting way recalls one interpretation of Endenese nake (‘bird’), as a word primarily meaning ‘game’—that is, the product of hunting.

Atoni (dawan) Kolo, Kédang Udang Olong

That kolo functions as a completely general term for ‘bird’ in the Atoni language of western Timor is indicated by both Fernandez (1996) and McWilliam (2002:308), who in a personal communication has noted how, in the contracted form kol , the term further appears in compounds referring not just to wild birds but also to the domestic fowl. Though often named simply as manu (see the following subsection), the fowl is more completely designated in Atoni as kol manu , while comparable binomials include kol ka , ‘cockatoo’, and kol ton , a kind of bird ‘that announces the onset of the rainy season’ (McWilliam, personal communication 2003). McWilliam further notes how Atoni kolo ‘may also refer to the idea of flying or flight’, as suggested by ot kolo , ‘flying automobile’, or ‘aeroplane’.

In other eastern Indonesian languages, kolo and variants mostly occur in compounds labeling two or more folk generics. Examples drawn from both Bima-Sumba and Ambon-Timor languages appear in Table 2. Among the Ambon-Timor languages, dialects spoken on Solor reveal as many as ten compounds of kolo . These are applied to such diverse categories as frigatebirds, egrets, sandpipers, terns, doves, the Brahminy Kite, and the Pied Bushchat (Verheijen n.d.)—a diversity that recalls Vischer's previously mentioned assessment of Palu'e kolo as a term ‘collectively’ classifying birds. In regard to the diversity found in Solor, it goes without saying that birds designated with compounds of kolo do not obviously constitute a single intermediate taxon. The Solorese and other compounds may be compared with the arbitrary occurrence of English ‘bird’ in names of such phylogenetically various kinds as ‘lyrebird’, ‘frigatebird’, and so on—a comparison suggested earlier with regard to Endenese nake . Nevertheless, one cannot simply infer from these numerous compounds that, by itself, Solorese kolo conveys the same general sense as does English ‘bird’, or functions as a completely general term for ‘bird’.

A cognate of kolo , namely olong , occurs in udang olong , reported as a general term for ‘bird’ in the non-Lamaholot language of Kédang, on the island of Lembata (R.H. Barnes personal communication 2001). Interestingly, udang alone also conveys this general sense, although the term has the more specific meaning of ‘pigeon’ (Barnes personal communication 2001). By itself, olong similarly denotes a ‘kind of spotted dove’ (cf. Nage kolo , Streptopelia chinensis; Manggarai kolong , Geopelia sp.). What one evidently encounters in Kédang, then, are names for doves and pigeons referring, at least in some contexts, to birds in general. In this, moreover, one glimpses a possible connection between the diverse referents of cognates of kolo encountered in various languages of Flores (though particularly ones belonging to the Bima-Sumba group, where the term almost always refers to pigeons and doves) and the Atoni use of kolo to designate the entire life-form category of ‘bird’.

Although the evidence does not reveal whether in Palu'e or Solorese—or indeed any of the Lamaholot languages of eastern Flores—unmodified kolo ever functions as a general term for ‘bird’, it is evidently on the basis of the widespread occurrence of cognates that Fernandez (1996) has reconstructed * kolon , ‘bird’, for the hypothetical proto-language he calls ‘Proto-Flores’. Actually, Fernandez's reconstruction would appear to relate to a grouping more inclusive than languages of Flores. One indication, of course, is the occurrence of kolo (‘bird’) in the Atoni language of Timor. Another is provided by six eastern Sumbanese compounds incorporating kulu , which, in a way recalling the Solorese usages, label such diverse bird categories as the small passerine Pied Bushchat, a rail, and a pigeon (see Table 2).19

Another possible indication of an association of kolo with a life-form category incorporating birds in general is the previously mentioned Nage compound peti kolo ( peti , it will be recalled, is the short form of ana peti , which designates Estrildine finches and, more generally, small passerine birds). As demonstrated elsewhere (Forth 2004a), although kolo specifically denotes members of the Columbidae, the binary expression can in context refer to any kind of bird, or birds of no particular or definite kind. At the same time, when this usage is considered within the wider framework of Nage binary idioms, it becomes apparent that one is dealing with an instance of dual synecdoche. Indeed, it is not ruled out that the use of kolo as a general term for ‘bird’ in Atoni, and perhaps other languages as well, reflects an extension of meaning comparable to what is suggested by terms like western Sumbanese (Lamboya) mango and, according to one interpretation, Endenese nake . This possibility of course casts doubt on Fernandez's reconstruction of * kolon . Further calling into question this reconstruction is the circumstance that cognates of kolo refer exclusively to pigeons and doves not only in languages of the Bima-Sumba group, but also in Ambon-Timor languages of eastern Flores (see, for example, the several Sikanese compounds of olong or olon , all of which refer to Columbiformes). Thus, if Fernandez is correct, one would need to explain why a term originally denoting a life-form category has become restricted to precisely the same set of birds—specifically, members of the genera Geopelia and Streptopelia—in languages belonging to two different eastern Indonesian sub-groupings.

On the other hand, the occurrence of eastern Sumbanese kulu and Solorese kolo , both in a series of compounds with quite various ornithological referents, does appear to support Fernandez's interpretation. Also relevant is Rembong kolong , occurring in two compounds denoting respectively button-quails and munias (Lonchura; see Table 2). A solution may be found in the possibility that kolo , kulu , and other reflexes of hypothetical * kolon derive from a polysemous Central Malayo-Polynesian prototerm that could be used to refer to birds in general but had its focus in certain Columbiformes. As will be shown presently, a similar polysemy inheres in cognates of manu in certain eastern Indonesian languages, where the term refers to both the domestic fowl (Gallus gallus) and to birds in general.

Rotinese Manu-pui, Tetum Manu, Kédang Manuq, Nuaulu Manu (Manue, Man), Tanimbar Manut, Kei Man (Manut), Wetan Maana (Maan), Leti Manu, Moa Mana

Rotinese manu-pui comprises manu (‘domestic fowl’) and pui (or puik ), a term that Jonker (1908:347) speculates may originally have referred to a particular kind of bird. He includes no separate listing for pui other than as a man's name (Jonker 1908:501), but an indication of its possible meaning is the probable cognate fuik , ‘wild’ (cf. Tetum manu-fuik , ‘wild bird’, Hull 2000). While Jonker's dictionary is based on the Termanu dialect of Rotinese, he further lists variants of pui as general terms for ‘bird’ in other dialects. Among these are Baä mpuik , Ti mbuik , and Dengka and Oe-nale mbui (Jonker 1908:756). Also noteworthy are several compounds comprising mbui in Dengka and Oe-nale, including mbui meo , a type of owl; mbui-nggeo , another kind of bird; and mbui-uda-oe , ‘rain bird’ (Malay burung hujan ; Jonker 1908). There is, then, some suggestion that mbui , and perhaps other cognates of pui , function in a way comparable to kolo in Solorese. On the other hand, Jonker's entries manu-bulu-lok (s.v. manu ), referring to a kind of small animal found in coastal sands, and manu-puik (s.v. puik ), a sort of marine fish, reveal compounds of manu denoting creatures besides birds—a situation reminiscent of uses of manu in some Oceanic languages (see Table 3, also Blust 1982:247-248).

The evidence for Rotinese suggests that, in this language, manu alone refers specifically to the domestic fowl and acquires the sense of ‘bird (in general)’ only when compounded with pui . Although widely occurring in eastern Florenese compounds denoting ornithological folk generics (see Table 4, especially terms from East Flores and Sikanese), it is unclear whether manu by itself serves as a general term for ‘bird’ in languages of eastern Flores. That it may sometimes do so finds some suggestion in Kédang manuq , although in this language another term is available for ‘bird’. Similarly, E.D. Lewis (personal communication 2001) describes manu as a word ‘for birds generally’ in the Tana ‘Ai dialect of Sikanese, where the same term more specifically denotes the domestic fowl. At the same time, in his Sikanese dictionary Lewis lists only the last sense for manu (Pareira and Lewis 1998), giving horong (or kenaha horong ) as the Sikanese term for ‘bird’.

By contrast, the evidence for cognates of manu denoting ‘bird (in general)’ is quite clear for Tetum as well as Tanimbar, Kei, Nuaulu, Wetan, Leti, and Moa, thus other Central Malayo-Polynesian languages included with languages of eastern Flores and Timor in Esser's ‘Ambon-Timor’ group.20 At the same time, regardless of the term used generally for ‘bird’, in most eastern Indonesian languages manu , or a related term, denotes the domestic fowl (see Table 3). While Nuaulu manu or manue refers to birds in general (Ellen 1993:121, 195), compounds incorporating the forms manu or man designate numerous folk generics, as do formally identical terms in eastern Flores languages and Tetum. Manu or man occur as a necessary or optional component of 14 Nuaulu compounds, which comprise about 30 percent of 54 Nuaulu bird names (Ellen 1993b; see Table 4). A comparable pattern is found in Tanimbar (Fordata), where the cognate manut (‘bird in general, more especially domestic fowl’, Drabbe 1932a) further occurs in at least five compounds denoting particular kinds of birds.21

What is most noteworthy about compounds of manu in Tetum, Nuaulu, and other languages of the Ambon-Timor group is the diversity of ornithological kinds they name (see Table 4). So various are they that it is difficult to envision their composing an intermediate folk taxon, or even a category defined by symbolic or utilitarian features. The same applies to compounds of kolo in some Ambon-Timor languages (e.g., Solorese) and to Endenese nake . At the same time, the evidence for compounds of manu reflecting the general sense of ‘bird’, even in languages like Sikanese where this meaning is not definitely attested, is considerably stronger than that for kolo (despite Fernandez's ‘Proto-Flores’ * kolon ), or indeed for nake .