Timothy Usher, Santa Fe Institute, developed with Eline Visser
Kalamang, known also as Karas, is spoken by perhaps 100-200 people (2016) living in two villages, Maas (Sewa /seˈwa/) and Anatalisa (Tamisen /taˈmisen/) on Kalamang island, the largest of the three Karas islands, in Sebakor bay off the west coast of the Bomberai peninsula in the Karas subdistrict of Fakfak regency in Indonesia's West Papua province. Speakers use the term Kalamang /kalamaŋ/ to refer to the island and Kalamangmang /kalamaŋ-maŋ/ to refer to their language; they object to the designation Karas because its sense includes two neighboring islands on which the unrelated language Uruangnirin is spoken. Visser speculates the term Kalamang to have originated as /kala(s)-maŋ/ “Karas language” with /maŋ/ subsequently re-added to distinguish the language from the island (Visser 2016: 1-4.)
[under construction]
Robidé van der Aa (1879: 342-347) … comparative terms for Karas after Coornengel
Cowan (1953: 35-36) …
Anceaux (1956) (unobtained)
Anceaux (1958) … comparative terms for Karas
Greenberg (n.d.) comparative vocabulary of Karas after Anceaux (n.d.)
Smits and Voorhoeve (1998: 22-233) comparative vocabularies of Mas village after Bertels, Karas-Laut after an unspecified source and an unspecified variety after an unspecified source
Galis (1955) 28 comparative terms for Karas after a list provided by the colonial government
Gravelle (n.d.) survey vocabulary of Karamangmang provided in spreadsheet format by Paul Whitehouse via the Summer Institute of Linguistics
Donohue (2010) 198 comparative terms for Karas of Mas village
Visser (2016) Kalamang story
Visser (2016) grammar of Kalamang of Maas village
Visser (2016: 43-99) gives 14 to 18 consonants and 5 vowels for Kalamang of Maas as follows:
| m | n | ŋ | ||
| p | t | [tʃ] | k | |
| b | d | [dʒ] | g | |
| [f] | s | [h] | ||
| l | ||||
| w | r | j |
| i | u | |
| e | o | |
| a |
Velar nasal /ŋ/ generally occurs only finally. It is found medially and as the first member of a consonant cluster only in a small number of loans; e.g. /ˈpiŋan/ “plate”, /ˈtʃaŋkir/ “cup” from Malay /piŋgan/, /tʃaŋkir/.
Unlike Mbahaam and Iha of the Bomberai mainland, Kalamang has no phonemic prenasalized stops, proto-West Bomberai voiced prenasalized stops having been denasalized to their plain voiced analogs as they have been in Timor-Alor-Pantar. In both instances, this is likely due to the influence of local Austronesian languages. Voiced stops do not occur finally.
Laminal affricates /tʃ dʒ/ occur only in loans from Indonesian and as conditioned palatalized allophones of apical stops /t d/ and, in the case of second person possessive suffixes, of velar stop /k/. Realizations vary between [tʃ c ɕ] and [dʒ ɟ ʝ] respectively
Labiodental fricative /f/ occurs only in loans.
Phonemic unoccluded fricative /h/ occurs only in loans from Indonesian and Arabic. Phonetic [h] is found as in free variation with laminal fricative /s/ in some words, where it is considered to be archaic.
Only a subset of consonants occurs finally:
| m | n | ŋ |
| p | t | k |
| s | ||
| l | ||
| r |
Sequential vowels are realized either as bimoraic sequences or as diphthongs as shown. Diphthongs [uⁱ oⁱ] have been found only in one word each; the example given for [uⁱ] is a compound (/mu-in/ 3.pl.-possessor.) Of all possible combinations of heterogenous vowels, only /oa/ has not been found:
| i | u | e | o | a | |
| i | — | iu | iᵉ | io | iᵃ ia |
| u | uⁱ | — | uᵉ ue | uᵒ uo | uᵃ ua |
| e | eⁱ | eu | — | eo | ea (?) |
| o | oⁱ | oᵘ | oe | — | — |
| a | aⁱ ai | aᵘ | ae | ao | — |
Stress is contrastive and does not appear to be conditioned by environment.
Visser (2016: 14-15, 15-17) gives pronouns for Kalamang of Maas in three case forms as follows:
| nominative | possessive | possessive | |
| 1 sg. | an | aŋ-gon | -an |
| 2 sg. | ka | ka-in | -tʃa |
| 3 sg. | ma | ma-in | -un |
| 1 pl. | pi | — | -pe |
| 1 pl. | in | iŋ-gon | -p-in |
| 2 pl. | ki | ki-n | -tʃe |
| 3 pl. | mu | mu-in | -un |
| 1 dl. | pi-er | — | — |
| 1 dl. | in-ier | — | — |
| 2 dl. | ki-er | — | — |
| 3 dl. | m-ier | — | — |
The semantic difference between the two first person non-singular bases is not understood; speakers deny that it is inclusivity, with which they are familiar from Malay. Likewise, the semantic difference between the two possessive contructions is not understood. The free possessives are postposed to the nominal they modify and can cooccur with the possessive enclitics in a single contruction.
[under construction]
Visser (2016: 86) gives the following demonstrative forms. These suggest a proximal root /yua-/ or /wa-/ and a distal root /(i)me/ or /mi/.
| proximal | distal | distal+ | |
| nominal | yua(ne) | ime(ne) | — |
| local adv. | watko | metko | owatko |
| manner | wandi | mindi | — |
| temporal | yua(ne), wane | -me | — |
| other | — | me (topic) | — |
Kalamang has at least seven case markers (Visser 2016: 87-92).
| locative | -ko |
| lative | -ka |
| accusative | -at |
| number nominative | -a |
| number accusative | -i |
| comitative | -bon |
| instrumental | -ki |
Three other nominal suffixes have been identified, but have less clearly defined uses (Visser 2016: 87, 92-95).
| adjective marker | -ten |
| derivator/inalienable | -un |
| focus | -a |
[under construction]
Just two verbal morphemes have been identified thus far (Visser 2016: 98-9).
| completive | -i |
| volitional/realis? | =kin |
A number of Kalamang affixes and clitics have been identified, but their exact function is unknown. The possible analyses of these forms is as follows (Visser 2016: 105-11).
| aspect or discourse marker? | =et |
| detransitivser? | na- |
| transitiviser? | ma(t) |
| imperative? | -te, -de |
| movement? | di- |
| (grammaticalised) dative? | -di |
Visser (2016: 101) gives the following question words:
| who | naman |
| what | neba |
| where | tama- |
| why | tamandi (neba) |
| when | jol tama |
| how | tamandi |
| how many | puraman |
| with what | mebanggia |