Kamrau Bay

Timothy Usher, Santa Fe Institute

The Kamrau Bay family consists of three closely-related languages, Buruwai, also known as Sabakor (Sebakor,) South Kamrau (Kamberau,) also known as Asienara, and North Kamrau (Kamberau,) also known as Iria, spoken around and to the west of Kamrau Bay at the southeast corner of the Bomberai peninsula in the Kaimana regency of Indonesia's West Papua province. Buruwai and Kamrau are the names of major rivers which flow into Kamrau Bay around which the languages named for them are spoken, since the people seem to have no clearly stablished names for themselves or their languages (Anceaux 1958: 116-117, Voorhoeve 1975: 369-370, Walker and Hesse 1988: 1, 2, 3, Visser 1989: 65-66, 68.) Voorhoeve (2005, 2007) calls this family Sabakor, but this is at best the name of only one of the languages and Sebakor also refers to a bay on the west side of the Bomberai peninsula in which Kalamang and Uruangnirin are spoken.

The internal classification of Kamrau Bay is as follows (Walker and Hesse 1988: 2):

Kamrau Bay

Buruwai

Kamrau

South Kamrau

North Kamrau

There has been quite a bit of confusion surrounding the use of the term Asienara, leading to the common misconception that there are only two rather than three languages in the Kamrau Bay family and that Asienara is another name for the Buruwai language. According to Walker and Hesse (1988 2,) Asienara refers to Southhern Kamberau, as spoken in the villages of Esania, Yarona, Edor and Kembala to the east of the lower Buruwai river, a completely different language from Buruwai aross the river to the southwest. Anceaux's (1958: 119-129) brief published vocabulary of Asienara is indeed the same language as South Kamrau. The same is true of Greenberg's (n.d.) lengthier excerpts from Anceaux's (n.d.) notebooks.

However, the vocabulary entitled Asienara in Voorhoeve (1975: 1975: 100) is in fact Buruwai, despite the fact that it too was drawn from Anceaux (n.d.) We can offer no easy explanation for this discrepancy besides the possibility that Anceaux may have had several vocabularies alleged to represent Asienara, with Greenberg exerpting from one and Voorhoeve from the other. Indeed, Voorhoeve's (2007) unpublished comparative vocabularies have some words representing Gaka and Guriasa villages attributed in part to Anceaux (n.d.) Anceaux's (1958: 111) map of the region includes inhabited portions of Walker and Hesse's (p. 3) South Kamberau territory, while both South Kamrau and Buruwai-speaking villages are listed as representing two different dialects of Asienara (p. 116, recapitulated in Voorhoeve 1975: 31.) Visser (1989: 38) writes, “It is revealing that the language spoken by the Sabakor people was called 'Asienara' (Anceaux 1958: 116) , which in Kamrau language means: people with curled hair;” however (p. 75) her survey was conducted with Voorhoeve . This incorrect conflation may also explain the seeming contradiction between Voorhoeve's (1975: 369, echoed by Wurm 1982: 138) comment that Iria and Asienara might be dialects of the same language.with his assertion as quoted by Walker and Hesse (1988: 2 after Voorhoeve p.c.) that Buruwai is a distinct language.

Walker and Hesse (1988: 2) give lexicostatistical figures between eleven Kamrau Bay villages as follows:

(In the chart above, Gaka stands for Gaka-Guriasa, Ubia for Ubia-Sermuku, Tanggar. for Tanggaromi and Kamr. for Kamrau.)

Anceaux (n.d.) vocabularies for Asienara and Iria (unobtained)

Anceaux (1956) (unobtained,) reprinted in English as (1958)

Anceaux (1958: 116-117) brief description of and (pp. 119-120) 10 comparative terms for Asienara (South Kamrau) and Iria

Greenberg (n.d.) comparative vocabularies for Iria and Asienara (South Kamrau) following Anceaux (n.d.)

Voorhoeve (1975: 100) 40 comparative terms for Iria and Asienara (Buruwai) following Anceaux (n.d.)

Voorhoeve (1975: 369-370) phoneme inventories and pronouns for Iria and Asienara

Voorhoeve (1980: 661-121) Proto-Asmat includes Iria words from Anceaux (n.d.)

Visser (1989) kin and miscellaneous terms for Sabakor (Buruwai) of Gaka village and Kamrau (North Kamrau) of Ubia-Sermuku village

Voorhoeve (2007) 227 comparative terms for Buruwai and Kamrau (mix of North Kamrau and South Kamrau)

Walker (1978) 110 comparative terms for Gaka village (Buruwai) and Obia-Seramuku, Waho and Wainoma villages (North Kamrau)

Walker and Werner (1978) (unobtained)

Walker (1983) 49 comparative terms for Kambola village (South Kamrau)

Walker and Hesse (1988) 209 comparative terms for Buruwai of Tairi village and Northern Kamberau of Waho village, 127 comparative terms for Buruwai of Yarona and Guriasa/Gaka villages, Southern Kamberau of Esania-Kuna and Yarona villages and Northern Kamberau of Tanggaromi, Bahomia, Ubia-Seramuku and Koi villages and 58 comparative terms for Northern Kamberau of Wamesa village

Matsumura (1985) 60 terms for Kamberau of Wanoma village (North Kamrau)

In addition to these, three unattributed typewritten vocabularies have been made available to us by SIL Indonesia dating to July and August 1956. They are glossed in Dutch and Indonesian and give 100 comparative terms for Gaka village (Buruwai,) Esania/Kuna village (South Kamrau) and Koi village (North Kamrau).

Voorhoeve's (2007) unattributed Buruwai and Kamrau vocabularies are not specified by location, but most likely largely largely represent the villages of Gaka and Ubia-Sermuku, since these are the Buruwai and (North) Kamrau villages that Visser and Voorhoeve had visited in 1986 (Visser 1989: 66, 75.) Walker and Hesse's Ubia-Seramuku vocabulary is North Kamrau but contains some unambiguously South Kamrau words, which may at least partly explain Voorhoeve's mix of the two varieites in his Kamrau vocabulary as the village itself probably includes some South Kamrau speakers.

The first to mention Kamrau Bay languages in the taxonomic literature was probably Anceaux (1956, 1958: 118-119, ) who wrote “Of some languages it is very clear that they are related, in some cases even closely related, e.g. Iria to Asienara…It is quite possible that further inquiries will prove that all these Papuan languages are related. There are many indications of such a relationship but a definite proof has not yet been given. Of all the Papuan languages of the Bomberai Peninsula Asienara and Iria are the most dissimilar, at least in their vocabulary, but even they have words that may be common with other languages.“

Voorhoeve's (1968: 4) Kamoro-Sempan-Asmat group does not include Anceaux's Asienara and Iria. The discovery of the special relationship between these families fell to Greenberg (1971: 840), who wrote, “It has not been previously noted that these two languages [Asienara and Iria] are to be connected with the Kamoro group to the east from which they are separated both by the Etna Bay subgroup of the Western New Guinea subfamily and by Irarutu, an AN language. Voorhoeve (1975: 369) then includes Iria and Asienara in the Asmat-Kamoro family, but fails to credit Greenberg for this discovery, although Greeenberg's proposed placement of the Etna Bay (a.k.a. Mairasi) family is debunked in detail (pp. 425-431.) Of Kamrau Bay's position within Amsat-Kamoro, Voorhoeve writes, “Iria and Asienara share nearly 80% cognates, the other percentages within the family range from 50 % to 70 %. There is no clear subgrouping; Iria and Asienara seem to be somewhat closer related to Asmat than to Kamoro.” By Voorhoeve (2005) he appears to have abandoned the notion of a special relationship between Kamrau Bay, which he labels Sabakor, and Asmat in particular, treating Sabakor, Kamoro, Sempan and Asmat as four co-equal members of the Asmat-Kamoro family.

[under construction]

Proto-Kamrau Bay had perhaps13 or 14 consonants and 5 vowels as follows::

Comparison with the inventory proposed in Voorhoeve (1975: 369-370) …

…

Apical non-stop /*ɾ/ does not occur initially; otherwise any consonant can occur initially or medially. Consonant clusters are found only in compounds.

Only a restricted set of consonants occur root-finally. These are discernible when suffixed with the article /*-ɾa/, where [ɾ] is modified or dropped following a consonant:

The combinations of final consonants and article /*-ɾa/ are analyzed by Voorhoeve (1980: 66) as class markers: “In this language the nouns seem to fall into three classes: those ending in -ra, in -ka or in -ʔa, and in -a or -da.” Analogous situations are found in West Kainantu (as well as Goroka) and in Baruya of Kratke Range, where Bee (1963) and Lloyd (1969) interpret a reduced set of finals as nominal class suffixes.

Initial consonants correspond as follows:

Medial consonants correspond as follows:

Root-final consonants in combination with article /*-ɾa/ correspond as follows:

Vowels correspond as follows:

These correspondences are exempified as follows. Our sources present an unusual situation in which nearly every one is needed to support at least part of the reconstruction, yet none are authoritative. The attestations of Walker (1978, Walker and Hesse 1988) are the best documented and cover most of the villages, but while generally reliable (notably excepting the pronouns) are pre-phonemic and some of the vocabularies are quite brief even by survey standards. Those of Voorhoeve (2007) contain many terms not found in Walker's surveys, however the curation is haphazard, the provinence of individual entries is often unclear and both Buruwai and (North) Kamrau hav been mixed with South Kamrau words, with South Kamrau itself going unrecognized. Greenberg's (n.d.) excerpts from Anceaux (n.d.) are more carefully curated and their source is clear, but represent only Asienara (South Kamrau) and Iria (North Kamrau) nor are all the terms found in other vocabularies included.

We are faced, then, with a conundrum inasmuch as it's not possible to fit all columns needed for a structured source-delineated presentation upon the width of a page. On the other hand, the phonological structures of the languages are understood well enough and the attestations are consistent enough for us to present our own phonemic versions of the words in question. Therefore we will depart from Newguineaworld conventions in presenting …, noting that the materials from which we draw are available in spreadsheet form in the Vocabularies section of our site.

Any consonant other than apical non-stop /*ɾ/ can occur initially.

Initial bilabial nasal /*m/ \is retained as such in all three languages:

Initial apical nasal /*n/ is much less common than initial /*m/ (above.) It is retained as such in all three languages:

Initial bilabial voiceless stop /*p/ is uncommon. It is retained as such in al three languages:

Initial apical voiceless stop /*t/ is uncommon. None of these examples have known cognates in Asmat-Kamoro except for /*timako/ “crocodile”, which is identical to Kamoro /'timakɔ/ Drabbe 1953: 101) and hence must be a loan in one direction or the other as Kamoro /t/ reflects /*s/ while /*t/ has become /k/ (q.v. Voorhoeve 1980.) It is retained as such in al three languages:

Initial velar voiceless stop /*k/ is uncommon because Asmat-Kamoro initial /*k/ has been lost in Kamrau Bay. It is either lost in North Kamrau or simply inaudible as the medial reflex of /*k/ is glottal stop /ʔ/ (below;) this change is shared with Mairasi of the Etna Bay family (Peckham 1991: 152-153):

Initial bilabial voiced stop /*b/ is retained as such in all three languages:

Apical voiced stop /*d/ does not occur initially in native words. Two apparent examples of initial apical voiced stop /*d/ are loans from Kowiai /domaʔ/ “straight” and /na-dua/ “flood tide” respectively (Walker 1985):

Palatal voiced stop /*ɟ/ does not occur word-intiially because Asmat-Kamrau Bay initial /*ɟ/ has been lenited to non-stop /*j/ (below.)

Initial velar voiced stop /*g/is fronted to apical /d/ in Buruwai:

One example in which initial /*g/ is unexpectedly retained as /g/ in Buruwai may indicate a loan from South Kamrau into Buruwai (cf. Asmat-Kamoro /*tarawu/):

Initial bilabial voiceless fricative /*ɸ/ is deoccluded to /h/ in South Kamrau and in North Kamrau:

Initial apical voiceless fricative /*s/ when followed by mid and low vowels /*e *a *o/ is dropped in South Kamrau and in North Kamrau:

When initial /*s/ is followed by high vowels /*i *u/ it is retained in South Kamrau and in North Kamrau:

Several examples have an /*s/ which is not droppped in South Kamrau and North Kamrau despite being followed by /*a/. Excepting “yellow”, which has no known Austronesian etymology and may well be a loan from Buruwai into South Kamrau, all of these are known to be loans from from Irarutu or Kowiai (below):

Initial bilabial non-stop /*w/ …:

… /*w/ …:

Apical non-stop /*ɾ/ does not occur initially.

Initial palatal non-stop /*j/ …:

… /*j/ …:

Any consonant can occur medially.

Medial bilabial nasal /*m/ is retained as such in all three languages:

In one word, medial /*m/ has irregularly changed to final /*m/ (below) in north Kamrau:

Medial apical nasal /*n/ is generally retained as such in all three languages:

When following an initial segment /*ma/, medial /*n/ is lenited to non-stop /ɾ/ in Buruwai:

Medial bilabial voiceless stop /*p/ is retained as such in all three languages:

Medial apical voiceless stop /*t/ is uncommon. It si retained as such in all three languages:

Medial velar voiceless stop /*k/ is deoccluded to glottal stop /ʔ/ in North Kamrau:

Medial bilabial voiced stop /*b/ is retained as such in all three subgroups:

Medial apical voiced stop /*d/ is found clearly in only three native roots, where it is retained as such in all three languages. “Heavy” is found in Asmat-Kamoro as /*jutoro/. “Garden” is conceivably a metathesis of North Kamrau /imoj/ [imod-a] loaned into Buruwai from South Kamrau:

A third example is a loan from Irarutu /waranda/ “knife” (Voorhoeve 1995, below); see also Matsumara's (1985) North Kamrau [beːda] “machete” which is a loan from Kowiai /beda/ “machete, bushkife” (Walker 1985, below):

Medial palatal voiced stop/*ɟ/ is occluded to \voiced apical stop /d/ in South Kamrau and in North Kamrau:

A number of verbs have root-final /*ɟ/. Historically, these reflect either palatal non-stop /*j/ or subphonemic giide [*j] following a front vowel /*i *e/ or a dipthongs /*oi/ etc.; however as they follow the same correspondence as root-medial /*ɟ/ they are presented as such at the proto-Kamrau Bay level:

Medial velar voiced stop /*g/, like intiial /*g/ (above), is fronted to apical /d/ in Buruwai:

One example in which medial /*g/ is unexpectedly retained as /g/ in Buruwai may indicate a loan from South Kamrau into Buruwai (cf. Asmat-Kamoro /*iti/) or is conditioned by following high front vowel /*i/:

Medial bilabial voiceless fricative /*ɸ/, like intial /ɸ/ (above,) is deoccluded to /h/ in South Kamrau and in North Kamrau:

Medial apical voiceless fricative /*s/ is retained as such in all three languages regardless of the following vowel, unlike intial /*s/ which is dropped when folowed by /*e *a *o/ (above):

Medial bilabial non-stop /*w/ …:

… /*W/ …:

Medial apical non-stop /*ɾ/ is retained as such in all three languages:

Medial palatal non-stop /*j/ is retained as such in all three languages:

Only a restricted number of consonants /*m *n *t *k *ɾ *ɟ/ occur word-finally, where they are discernible primarily by the results of their combinations with nominal article /*-ɾa/. First, let us examine the behavior of /*-ɾa/ when it is appended to a vowel-final root and is thus reflected as /-ɾa/ in all three languages :

Numbers take article /*-ɾa/ in Buruwai but not in South Kamrau or North Kamrau:

Final bilabial nasal /*m/ …:

Final apical nasal /*n/ …:

Final apical voiceless stop /*t/ has been identified in only one example, where it descends from Asmat-Kamrau Bay fricative /*s/:

Final velar voiceless stop /*k/ …:

Final palatal voiced stop /*ɟ/ … resulting in a correspondence identical to that of medial /*ɟ/ (above):

Final apical non-stop /*ɾ/ … It is thus nearly indisiguishable from the reflexes of final /*n/ (above) except through outcomparison to Asmat-Kamoro although the sporadic presence of nasalization in some examples of the latter:

A number of final consonants are ambiguous between /*m *n *ɾ/ because there is no outocmparison to Asmat-Kamoro or Asmat-Kamoro has lost the final consonant (that is they do not continue a reduced segment /*CV/) and the final vowel is not /*a/, the disposition of which distinguishes /*m/ in North Kamrau:

In four examples, Buruwai indicates a final vowel but South Kamrau and North Kamrau indicate a final consonant /*m *n *ɾ/ (below). This may reflect yet another allowable final consonant the quality of which cannot be identified with the data at hand:

One word follows the opposite pattern, with Buruwai inicating a final consonant while South Kamrau and North Kamrau indicate a final vowel:

… vowels ….

High front vowel /*i/ …:

High back rounded vowel /*u/ …:

Mid front vowel /*e/ …:

Mid back rounded vowel /*o/ …:

Low central vowel /*a/ …:

When /*a/ is followed in the next syllable by high front vowel /*i/ . it is raised and fronted to /e/ in North Kamrau:

…:

When /*a/ is followed in the next syllable by high back rounded vowel /*u/, it is raised backed and rounded to /o/ in North Kamrau:

One example in which Walker and Hesse's South Kamrau forms also change /*a/ to /o/ probably indicates a loan from North Kamrau into South Kamrau:

One word is not affected because it is a compound with /*manam/ “eye” (above):

However there are several unexplained exceptions to this rule:

Buruwai has a tendency to lose initial vowels …: