Muli Strait

Timothy Usher, Santa Fe Institute

The Muli Strait family consists of two closely related languages, Mombum, also known as Komolom (Komelomsch,) and Konorau (Koneraw, Konerawsch,) spoken respectively on Komolom Island at the south end of the Muli Strait (Marianne Strait) and on the south coast of the larger Kolopom island immediately the west. Loans from Muli Strait in Sawuy, Yakhai and Warkay suggest that the family originated in the Casuarina Coast region, and was pushed south by the expansion of Asmat. Upon entering the Muli strait region, the family acquired a number of loans from the Kolopom languages and from the unrelated Marind language to the east.

[under construction]

Geurtjens (1933: 393-433) 455 comparative terms for Konerawsch (Konorau) and Komelomsch (Mombum)

Drabbe (1950) 422 comparative terms for Mombùm of Mombùm village

Boelaars (1950) …

Drabbe (1954) …

Voorhoeve (1975: 99) 40 comparative terms for Mombum following Drabbe (1950) and 37 for Koneraw following Geurtjens (1933)

Voorhoeve (1975: 396-398) pronouns for Koneraw following Geurtjens (1933) and for Mombum following Drabbe (1950)

Voorhoeve (1980: 61-121) Proto-Asmat includes comparisons to Drabbe's (1950) Mombùm

Susanto (2001) 238 comparative terms for Konorau (Waɣan) and Komolom

Olsson (2017) … comparative terms for Mombùm

Geurtjens' (1933: 393-433) comparative vocabularies of Konerawsch and Komelomsch make it perfectly obvious that the two form a tightly-knit group, but nowhere does he propose a classification of the languages he surveyed.

Drabbe (1950: 548,) without reference to Geurtjens' (1933) materials, places Mombùm alongside Paniai Lakes, Asmat-Kamoro, Awyu, Wambon and Muyu (Lowland Ok) in a group he designates the suffixing languages (“suffix-talen.”) While the other three of Drabbe's groups have at least some grain of taxonomic truth to them, his operative criterion is synchronic and typological rather than genealogical in nature, and in this respect suffixiation is the norm throughout New Guinea and Australia alike. Even so, it it true that Mombum belongs immediately with Asmat-Kamoro rather than with Kolopom proper (“koppelwoord-talen”,) Marind-Yaqay and Boazi (“præfix-talen”) or Moraori and the West Morehead River (“gemengde talen”) languages, so Drabbe's classification is arguably not completely groundless, and it's possible that the imperfect but non-random congruence of these morphological categories with taxonomic reality is what moved him to adopt it.

Voorhoeve (1968: 3-5) posits Drabbe's (1950) Mombum as a fourth branch of his Asmat-Awyu-Ok family, though qualifies this, saying: “Mombum is a border case; its closest relationship is with Asmat. Since it does not show comparably close links with any other language it has been provisionally included ln the Asmat-Awyu-Ok Family.” In support of this assertion, he calculates 31% cognacy between 156 comparable terms for Asmat and Mombum respectively. It is interesting to note that the borders of Voorhoeve's Asmat-Awyu-Ok, while formally based upon lexicostatistics rather than typology, closely track those of Drabbe's suffixing group.

Voorhoeve (1975: 369, 396-398) adds Geurtjens' (1933) Koneraw to form a Mombum Family, which is said to belong to his Central and South New Guinea Stock alongside not only Asmat-Kamoro, Awyu and Ok but also Awin-Pa (Elevala River,) East Strickland (Strickland River,) Bosavi (Papuan Plateau) and Duna-Bogaya. There is no mention here of an exclusive relationship between the Mombum Family and Asmat-Kamoro. But in Voorhoeve (1980: 1) he goes further than his earlier (1968) claim, stating, “The Mombum Family is the closest relative of the Asmat-Kamoro Family; together they form a sub group within the CSNG Stock.” This assertion is then is supported by a modest number of comparisons to Mombum scattered throughout his reconstructions of proto-Asmat, with a note (p. 59) explaining: The Mombum and Koneraw languages…have been strongly influenced by the neighbouring languages of the Kolopom and Marind families, and this has obscured their relationship with the languages of the Asmat-Kamoro family to a considerable degree. Still, Mombum data often throw unexpected light on the reconstruction of proto-forms, especially those of the ancestor language of the Asmat-Kamoro family, and for this reason I have admitted [Mombum] data as well.“

Wurm (1982: 130-138) fails to consider Voorhoeve's (1958, 1980) assertions of a special relationship between Asmat-Kamoro and Mombum-Koneraw, instead following Voorhoeve's (1975) positioning of the Mombum Family as a primary branch of the very broad Central and South New Guinea Stock. Because Wurm's breakdown was repeated in tertiary compilations such as Ruhlen (1987,) Voorhoeve's insight slipped through the cracks of the macro-regional taxonomic literature. Yet Voorhoeve (2005) himself made no mention of Mombum or Koneraw when comparing his proto-Asmat forms with Healey's (1964) proto-Mountain Ok and (1970) proto-Awyu-Dumut; it's not clear if he'd backed away from his proposal or simply had no reconstruction of proto-Mombum-Koneraw and felt that single language comparisons were out of place in this treatment. Mombum is still shown as a member of Asmat-Kamoro is Voorhoeve's (2007) unpublished New Guinea lexical files.

Greenberg (1971: 829-834) places Komelom (Mombum) and Koneraw in a Jei subgroup of a Southern or Kiwaiic branch of his proposed Indo-Pacific macrofamily. This Jei subgroup is said to include Yei and the westernmost languages of the Morehead River family, the Kolopom languages including Moraori and the Bulaka River languages as well as those of Muli Strait. This regional grouping is nearly the set of languages which have borrowed from Marind, or even all the languages which appeared in Geurtjens (1933: 393-433) or Nevermann (1939) except for Marind itself. This must be seen as among the very worst errors in Greenberg's breakdown, with no attempt to identify loans and unduly influenced by the juxtapositions of vocabularies in the print literature..

Ross (1995: 144) lists the Mombum family among those whose “pronoun forms which could well reflect Proto Trans New Guinea. but they do not reflect the paradigm convincingly enough for inclusion here,” further stating (p. 153) that there is some small amount of evidence that it belongs with the Trans-New Guinea phylum without saying what that evidence is. Pawley, Ross and Osmond (2005: 70-75) consider Voorhoeve's (1968) placement of Mombum with an Asmat-Awyu-Ok family and assert that Mombum shares with Asmat and Awyu-Dumut the common innovation of having replaced the vowel /*a/ in pronouns with /*u/.

Here we advocate the point of view expressed and exemplified in Voorhoeve (1980,) which is that the Muli Strait (Mombum) family is most closely related to the Asmat-Kamrau Bay family along the coast to the north and the far northwest.

[under construction]

Proto-Muli Strait had 14 or 16 consonants and perhaps 7 vowels as follows:

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

It is not clear if semivowels [*w *j] should be treated as non-stop consonants or as sequential vowels as they are in nearby Bulaka River and Marind. Voiced /*b *z/ originated as /*w *j/ but may have already been occluded by the time of proto-Muli Strait.

Initial consonants correspond as follows:

Medial consonants correspond as follows:

Final consonants correspond as follows:

Vowels and vowel sequences correspond as follows:

These correspondences are exemplified as follows, with Mombum (Komolom, Komelomsch) forms drawn from Drabbe (1950: 561-574,) Susanto (2001) and Geurtjens (1933: 398-433) and Konorau (Konerawsch) from Susanto and Geurtjens.

All consonants except for bilabial voiced stop /*b/ and apical non-stop /*r/ can occur initially.

Initial bilabial nasal /*m/ is retained as such in both languages:

Initial apical nasal /*n/ is retained as such in both languages:

Initial bilabial voiceless stop /*p/ is retained as such in both languages:

Initial apical voiceless stop /*t/ is retained as such in both languages:

Initial laminal /*s/ is typically realized as an affricate [tʃ ts] in Konorau, a change found also in the insular Kolopom languages which surround the Konorau enclave:

Initial velar voiceless stop /*k/ is retained as such in both languages:

Initial bilabial voiced stop /*b/ is uncommon. It is retained as such in both languages.:

Initial palatal voiced /*z/ …:

It's conspicuous that Drabbe and Geurtjens agree that the sound is a voiced laminal [z̪ z dz] in both languages while Susanto's much more recent vocabularies give unvoiced [x θ]. Most likely, this is an ongoing change under the influence of Marind's western dialects in which the regular reflex of Marind /*j/ is aspirated to [hʲ] (q.v. Drabbe 1955. 148-151.) Maklew of the Bulaka River family likewise reflects /*j/ as /s/, Susanto's Konorau [θ] being in keeping with the fronting which characterizes apicals in both languages.

Initial bilabial voiced prenasalized stop /*mb/ is uncommon. It is retained as such in both languages.:

Initial apical voiced prenasalized stop /*nd/ …:

Initial velar voiced prenasalized stop /*ŋg/ ….

Apical non-stop /*r/ has not been observed to occur initially.

Initial velar non-stop /*ɣ/ … Apparent variations in the realization of velar non-stop /ɣ/ are attributed not to phonemic change, but to Geurtjens' inconsistent hearing of the sound in Mombum and Konorau alike. Drabbe stipulates that /ɣ/ approaches [x] in stridency, leading Geurtjens to hear [h] in some examples:

In one example, the initial consonant is ambiguous between /*ɣ/ and palatal voiced /*z/ (above):

Any consonant can occur medially.

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

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

When followed by high front vowel /*i/, the reflexes of medial /*n/ are sometimes realized as palatalized nasal [ɲ] in free variation with apical [n]:

Medial bilabial voiceless stop /*p/ is generally retained as such in both languages:

Medial apical voiceless stop /*t/ is retained as such in both languages:

Medial laminal /*s/ …:

Medial velar voiceless stop /*k/ …:

Medial bilabial voiced stop /*b/ is usually devoiced to [p] in Mombum:

One apparently irregular example in which medial /*b/ is a reduplication of intiial /*p/ (/*p *b/ are not distinguished initially, above) possibly indicates a loan into Mombum from Konorau:

Medial apical voiced stop /*d/ is generally retained as such in both languages:

Medial palatal voiced /*z/ … Although it's not clear that it contrasts with sequential /*i/ [j], its reflexes in Konorau are like those of intiial /*z/ (above) which is contrastive :

Medial bilabial voiced prenasalized stop /*mb/ is retained as such in both languages:

… /*nd/ …:

Medial apical voiced prenasalized stop /*nd/ …:

One apparent exception suggests a compound or a loan from Kimaghama in which this root is found as /itumbu/ (q.v. Geurtjens 1933: 400-401):

Medial /*nd/ when followed by high front vowel /*i/ is palatalized and affricated to [ndʒ] in both languages:

Medial voiced prenasalized stop /*ŋg/ …:

Medial apical non-stop /*r/ is retained as such in both languages:

Medial velar non-stop /*ɣ/ …

In one example, the medial consonant is ambiguous between /*ɣ/ and palatal non-top /*j/ (above):

All consonants except for laminal /*s/, velar voiceless stop /*k/ and palatal non-stop /*j/ have been found to occur word-finally.

Final bilabial nasal /*m/ is retained as such in both languages:

Final apical nasal /*n/ …:

Final bilabial voiceless stop /*p/ …:

Final apical voiceless stop /*t/ …:

Laminal /*s/ does not occur finally, because Asmat-Muli Strait final /*s/ has merged with /*r/ (below):

Velar voiceless stop /*k/ has not been found to occur finally, perhaps suggesting that it has changed to non-stop /*ɣ/ (below).

Final bilabial voiced stop /*b/ …:

Final voiced prenasalized stops /*mb *nd *ŋg/ are merged with nasals /m n/ in Konorau.

Final bilabial voiced prenasalized stop /*mb/ is retained as /mb/ [mb b] in Mombum and meged with nasal /m/ in Konorau:

One example is somewhat confused in this respect:

Final apical voiced prenasalized stop /*nd/ is retained as /nd/ [d nd] in Mombum and merged with nasal /n/ in Konorau:

When preceded by high front vowel /*i/, /*nd/ is realized as [ndʒ] in Mombum and as /n/ in Konorau:

Final velar voiced prenasalized stop /*ŋg/ is retained as /ŋg/ [ŋg ŋk ŋ] in Mombum and merged with apical nasal /n/ in Konorau, as is final /*nd/ (above):

Final apical non-stop /*r/ is sometimes followed with a subphonemic copy vowel:

Palatal non-stop /*j/ has not been found to occur finally. Some reflexes of velar non-stop /*ɣ/ (below) suggest that the two may have been merged in final position.

Final velar non-stop /*ɣ/ …:

… vowels …

High front vowel /*i/ is generally retained as such in both languages:

… /*i/ … The reflexes of these sequences contrast with those of intiial palatal non-stop /*j/ (above):

… /*i/ …:

In two examples, /*i/ is backed to [u] in Geurtjens' Konerawsch when following high back vowel /*u/ in a previous syllable:

High back rounded vowel /*u/ is generally retained as such in both languages:

… /*u/ …:

High mid front vowel /*e̝/ …:

High mid back rounded vowel /*o̝/ …:

Low mid back rounded vowel /*ɔ/ …:

Low front vowel /*æ/ …:

Low central vowel /*a/ is generally retained as such in both languages. In monosyllables, it is often lengthened to [aː], especially in Konorau:

Muli Strait nominative pronouns are reconstructed as follows, with attestations drawn from Drabbe (1950: 562, 574,) Susanto (2001: 4) and Geurtjens (1933: 430-431):

The third person plural, if there was one, is not known. First person plural /*ni/ is tentative, based as it is solely upon Geurtjens' <ni>, conceivably a misprint.

Like other languages of the region, the Muli Strait languages, especially Mombum, have borrowed many words from Marind:

The fact that Mombum has more loans from Marind suggests most to be of recent origin, as Komolom is closer to Marind territory than is the south coast of Kolopom where Konorau is spoken.

Some words which by their meanings cannot be original to Marind are nevertheless borrowed immediately therefrom:

The second of these may contain a novel phoneme in Mombum in clear contrast with initial sequence /*uV/ and reflecting Marind's western dialects' realization of Marind /*w/ as [hʷ ]; see Marianna Strait reflexes of initial /*j/ (above) for comparison

… Indonesian …:

… Kolopom …:

The following loans between Muli Strait and Sawuy, Yakhai-Warkay and Kaugat (Atohaim) respectively suggest that Muli Strait languages were once spoken in the Casuarina Coast region well to the north of Kolopom (q.v. Voorhoeve 1971: 89, 110-114):

In a few of theseinstances, the direction of borrowing is not known. However, several have established etymologies in Asmat-Muli Stait or elsewhere. “Breast” /*abuɣ/ is the regular reflex of Asmat-Muli /*awVɣV/. Asmat-Muli /*iuwuri/ “dog” loses its subfinal apical in Muli Strait /*i[u]nui/, but the loan to Atohaim shows it to have still been present in pre-Muli Strait. On the other hand, /*ɣaræp/ “bow” is itself a loan from Kolopom /*karep/, which must be more ancient than proto-Muli as its reflex is found also in Moraori, strengthening the suggestion that the linguistic geography of this region was rather different in relatively recent times.