Nafsan
South Efate
Fate, Erakor
Native to	NortheastVanuatu
Region	Efate Island
Native speakers	6,000 (2001)[1]
Language family	Austronesian Malayo-Polynesian Oceanic Southern Oceanic North-Central Vanuatu Central Vanuatu Epi-Efate Efate Nafsan
Language codes
ISO 639-3	erk
Glottolog	sout2856
Nafsan is not endangered according to the classification system of theUNESCOAtlas of the World's Languages in Danger

TheNafsan language, also known asSouth Efate orErakor, is aSouthern Oceanic language spoken on the island ofEfate in centralVanuatu. As of 2005^[update], there are approximately 6,000 speakers who live in coastal villages from Pango to Eton. The language's grammar has been studied byNick Thieberger, who has also produced a book of stories and a dictionary of the language.^[2]

Nafsan is closely related toNguna and toLelepa. Based on shared features with southern Vanuatu languages (including echo–subject marking, and the free and preposed 1st-singular-possessive morphemes), Lynch (2001) suggests it could form part of a southern Vanuatu subgroup that includes New Caledonia instead of the neighboring Efate languages.

Nafsan has a total of 20 phonemes consisting of 15 consonant and 5 vowel sounds.^{[T2006 1]}

As seen in the above chart, Nafsan's vowel phoneme inventory is that of a five-vowel system; this is one of the most commonly seen vowel inventories in any given language in the world and also especially evident in manyOceanic languages. There is a distinction between short and long vowels but it is currently in a process of change that makes its status unclear.^{[T2006 3]}

In Nafsan, it is typical that two contiguous identical consonants occurring in an utterance undergo a process ofdegemination to be realised as a single consonant. In (1), the two contiguous identical consonants /n/ result in the phonetic attachment ofdemonstrativene 'this' to the preceding wordnawen ne [nawene] 'this sand'.^{[T2006 4]}

High vowels in prepositions acting as a prefix often undergo a process of vowel centralisation to reduce theunstressed syllable. In (2), the high vowel /i/ in the prepositionki is reduced to [ə] when preceding the demonstrativenen 'that'.^{[T2006 5]}

The system ofnumerals in Nafsan is base-5 (quinary). Numbers for two to five are distinct numerals that are then seen repeated in slight variation for the numbers seven to ten. The pattern of the numerals can be seen in the table below.^{[T2006 6]}

Ralim iskei can be used as an example to see the method for displaying number ten and above in South Efate; the numeral for tenralim is followed by its multiplier, which in this case isiskei for one. The term for andatmat is added after the multiplier with an additional numeral to form a number such as thirty seven:^{[T2006 6]}

There are two ways of markingadnominal possession in Nafsan: through the use of a possessive pronoun (indirect possession), or directly on the noun (direct possession). Indirect possession is used for general possession, while direct possession is used for nouns that are closely associated items (e.g., body parts or products, kinship terms, etc.).^{[T2006 7]}

Indirect possession ismorphosyntactically represented through the use of the possessive markersni (of) orknen (of it), or of the presence of a possessive pronoun such asnakte (my/mine).^{[T2006 7]}

When possession is marked by a possessive pronoun, the pronouns follow the possessed NP:

ni possession: the prepositionni only occurs when the possessum is a noun. The NP follows the form of ‘possessed ni possessor’.

knen possession: This form is used as an inanimate referent, and often indicates a previously mentioned participant in the discourse. It is positioned following the referent noun.

Direct possession is used for inalienably possessed nouns. This is similar to other languages of Vanuatu that denote inalienable nouns as those that refer to relationships of part-whole association such as kinship terms, body parts or products, and associated parts (such as leaf/stem).^[3] These nouns take directly suffixed possession markers, however they can also occur without possessive marking when the possessor is encoded by a noun. The directly possessed (DP) suffix only attaches to the class of directly possessed nouns. For sg and 3p forms, an unpredictable vowel (V) may be inserted to aid DP suffixation.^{[T2006 10]}

If the directly possessed noun has no possessive suffix, the referent is presumed unknown or disembodied. Lack of possession also occurs when possession is encoded by the possessed noun preceding the possessor. As in the following example, the directly possessed nounrait (mother) is preceded by the nountesa (child).

Negation in Nafsan occurs in two ways. The first is the use of the intransitive verbtik (no, nothing), which can be used singularly or paired with the generic proclitici= (3sgRS). The second, more widely used method, is through the use of discontinuous negative particlesta ... mau. Nafsan also does not differentiate between the negation of predicates and the negation of whole propositions.^{[T2006 5]}

Tik is a verb translated as 'no' or 'nothing' and is used in similar contexts to its English translations.^{[T2006 4]} In the following example,tik is used in the same way as in the English translation.

Tik is also able to be made into a transitive verb through the addition of the transitive suffix-ki. When this occurs, the new gloss is 'to not have'. As such, in the following example there is no instance of a possessive verb being negated, instead the verb in the sentence (tik-ki) is negative in meaning.^{[T2006 5]} Another verb that is negative in meaning istap, meaning 'to not do something'.

The other way of negating predicates in Nafsan is through the use of two negative particles:ta(p)...mau. There is free variation between the use ofta andtap, ie the usage does not change according to any specific environment.^{[T2006 13]}Ta(p) is used preceding the proposition to be negated, andmau follows at the end of the sentence. The following examples show a positive sentence, which is then negated in the second example.

Sometimes, in the casual speech of young people predominantly, the second negative particlemau is left off, as seen in the following example.

Theta marker can also act a durative marker, so in negative sentences where both uses ofta are present it can result in two different readings of a sentence. In the first example below, reading the firstta as the negative one results in the whole proposition being negated. In the second example, exactly the same sentence, reading the secondta as the negative results in only the final verb (puet 'to take') being negated, thus creating a different meaning entirely.

There are three common forms of demonstratives in Nafsan:go 'that, near addressee',ne 'this', andnen 'that'. Similar to otherSouthern Oceanic languages, these forms serve both spatio-temporal and discourse deictic functions.^{[4][T2006 16]} The formgo 'that' is addressee-anchored referent to a spatial location nearer to the addressee from the speaker's perspective or something previously said by the addressee.^{[T2006 16]} This encoding is found in otherlanguages of Vanuatu, such asSke, that do not belong to the Central Vanuatu subgroup.^[5][4] There is an ability for syntactic ambiguity stemming from the two common functions encoded by demonstratives resulting in dual interpretations possible in some utterances. This is shown in example (16) and (17) where the nounkal 'digging stick' used with the demonstrativego can mean 'the digging stick near you' from the spatial sense of the form or 'the digging stick you talked about' from the discourse sense of the form.^{[T2006 16]}

In Nafsan, demonstratives have a noun-demonstrative word order which is typical of Austronesian languages according to the World Atlas of Language Structures.^[6] They typically appear in two locations within a sentence as shown in examples (18) and (19); as a modifier of the noun phrase and following a directional adverb, respectively. Otherwise, demonstratives must undergo nominalisation through the addition of the prefixte- (see 3.4.1) as they cannot occur as the only exponent of a noun phrase. While demonstratives can co-occur with lexical nouns and focal pronouns, they cannot do so with clitic pronouns in Nafsan.^{[T2006 16]}

The formnen 'that' frequently occurs incollocation with thesubordinatorkin to create the English equivalent 'that which' as seen in example (19). As the formnen 'that' has the potential to act as a demonstrative or a relativiser, the pause given between the two formsnen andkin indicates that it is likely the formnen 'that' is acting as a demonstrative that is modifying the noun phrase in this context.^{[T2006 18]}

The demonstrativesne 'this', andnen 'that' also often occur in collocation with the noun mal 'time' to create the formsmalne 'this time', as seen in example (20) andmalnen 'that time'.^{[T2006 16]}

The distinction between demonstrative formsne 'this', andnen 'that' is a result of the distance-encodingclitic=n. This clitic can occur with several word classes as shown in the table below. Furthermore, like spatio-temporal demonstratives, it also has the deictic function of acting as a referent to previous parts of a discourse as shown in example (21).^{[T2006 10]}

The presentative morpheme is a demonstrative in Nafsan which has no paradigmatic relationship with the demonstrators detailed above. The formkia is often collocated withinterrogatives such as, such asfei kia 'who here', and typically places emphasis on the preceding nominal or utterance, as shown in example (22).^{[T2006 16]}

This function of drawing attention to its preceding forms has been used alongsidefillersiwel, gawan, tkanwan which are all used to mean 'thus', 'that's the way', or 'like that', the latter of which is used at the end of the story as seen in example (23).

The emphatic purpose of this demonstrative is similar to those found in other languages of Vanuatu such as the formna- in Ske in example (24).^[7]

Nominalisation of demonstratives, verbs, possessives, ordinal numbers, quantifiers, and nouns occurs in Nafsan through the attachment of the determiner prefixte-. The productive process ofte-nominalisation allows for the derivation of a large class of demonstrative pronouns.^{[T2006 20]}

Example (25) reflects how addressee deixis is encoded in the demonstratives that have undergonete-nominalisation.

The prefixka- is attached to nominals greater than one in Nafsan to formordinal numbers which can then further gain the prefixte- to form a demonstrative as shown in example (26).^{[T2006 17]}

There are mainly two classes ofpronoun in Nafsan. The free pronoun and the bound pronoun.^{[T2006 22]}

The free pronouns incorporate three area, demonstrative pronouns, focal pronouns(function as both subject and object) and the oblique free pronoun (in either possessive or benefactive form).

The focal pronoun (Lynch, 2000), also known as an independent pronoun (Crowley, 1998), functions as both the subject and object in an argument. It allows the pronoun itself to be the NP on their own unlike the bound pronouns which have to be attached to a verb. Focal pronouns express singular and plural but do not distinguish dual number.

1a) subject role

1b) object role

The examples (1a)& (1b) show the 1st person singular pronounkineu performed as the subject and object correspondingly. And the following is a list of the focal pronouns in Nafsan.

Oblique free pronoun function inpossessive alsobenefactive case. For the possessive pronoun, it follows the possessed NP, generally made up of the preposition -nig ‘from’/ ‘of’.

2) Possessive pronouns follow the possessed NP

There are variation forms of the suffix -nig , when it combines with an unstressed syllable, the high vowel will become lower. E.g. (niger → neger)

In the benefactive, the argument shares the same possessive morphology, yet the possessive morpheme is used in the pre-verbal position to express the beneficiary. The following example shows how beneficiary expressed by a pre-verbal position.

Bound pronoun comprises subject proclitics, object suffix for direct object and direct possessive. For the subject proclitics, there is neither separate set of dual object, nor oblique form. The obligatory subject proclitic pronouns are being seen as the arguments of the verb. For the pronominal suffixes of bound pronouns, the plural form is used to express any number that is greater than one.

The proclitic subject pronoun cannot stand alone without attaching to the first element of the Verb compound. They are considered to beclitics since they can attach to any part of the Verb compound. Subject proclitics happened in three archetypes, realis, irrealis and perfect. The subject proclitic represents the subject argument since it is the only obligatory element in the sentence except for the verb.

Proclitic subjects distinguish realis and irrealis situation. The realis is unmarked, and the irrealis being marked in the subject to show the action is yet to be realised, including most of the future events but not all, all the imperatives and hortatives. There is a strong preference for the subject of desideratives, achievement and predicates to be using irrealis form.

4)realis and irrealis paradigm

The examples (4) show all realis form of pronouns in all cases except the subject of the verb mai ‘to come’ which is appeared in a desiderative complement.

When dealing with aspectual past (event that is over), regarding the speaking event and past time reference, the perfect form of proclitic is used. Generally, perfect proclitics directly followed by the perfective particle pe, yet it is not a necessary criterion. Notably, perfect proclitics never occur in imperatives. Perfect proclitics can be found in narratives that deal with long events like World War 2.

5) narrative

The example(5) shows the perfect proclitics being used to refer to those who are long dead in a narrative sentence.

Traditional stories in Nafsan often use perfect proclitic form as they are set in the past. The example(6) of an extract of a custom story telling also shows that perfective particlepe is not necessary to appear in perfect proclitic sentence.

6) Storytelling

There are two separate types of object suffix, can be distinguished by the roles they encoded and the host they attached to. One type is for direct objects, the direct object suffixes attached to the object of the predicator to encode it. The other type is for oblique objects, the oblique object suffixes encode typically the location and the case of semi-transitive verbs. Based on the semantics of the semi-transitive verbs in the oblique case, the oblique object suffixes apply to movement to, at, or from a location. There are list of distinctive bound suffix being used in two types of object in table.2.

Object suffixes encode the object of derived transitive verbs, ambitransitive verbs, ditransitive verbs and of the preposition-ki. To reference an object in Nafsan can be either by an object suffix or a lexical NP. Therefore, object suffix cannot appear in the Verb Complex while there is a referential lexical NP for object indication.

7) transitive verb/ preposition-ki

This is an example (7) showing how object suffix used in transitive verb. The intransitive verbpes-kerai takes the transitivising suffix-ki to become transitive which allows it to take the object suffix-k in the first use. However, to emphasize the object, the last clause used the focal pronounag ‘you(singular)’ instead of the object suffix.

8) ambitransitive verb

In general, ambitransitive verbs requires a transitive suffix before the addition of the object suffix. The example (8) shows that transitive suffix-e is added before the object suffix-r occurred.

9) ditransitive verb

The object suffix indicates the recipient when it is with a ditransitive verb. The example (9) shows when the suffix-r is used to encode the addresses.

The oblique suffix has a locational meaning. The oblique case can also be indicating temporal and spatial references. The example shows the suffix -wes encoded the day that the race was held.

10) oblique suffix

The direct possessive suffix can only be attached to direct possessed nouns and reflexive/reciprocal morpheme yet not being a clitic. The 3 person singular is the most common form of direct possessive pronoun being found, even though there is other direct possessive pronoun see table.2. The following example(11) shows the 3sg direct possessive suffix -r.

11) direct possessive suffix

Below is a table explaining the common abbreviations used in negation examples above:^{[T2006 30]}

Thieberger's field recordings have been archived withParadisec:

• summary of the collection of materials in Nafsan
• listing of all material available via theOpen Language Archives Community for Nafsan.

• Pages from:Thieberger, Nick (2006).A Grammar of South Efate: An Oceanic Language of Vanuatu. Honolulu: University of Hawaii Press.ISBN 9780824830618.

• Other references:

• Anon. 1868.Nalag nig Efat. Trans.D. Morrison. Sydney: Mason, Firt, nigar asler (Mason, Firth and Co).
• Anon. 1892.Tusi nalag Efate Niu Ebrites. Sydney: F. Cunninghame and Co.
• Anon. 1979.Natus nalag (213 pp).
• Bible. 1864.Nadus iskei nig Fat. Aneityum: Mission Press.
• Bible. 1866.Nafsanwi nig Iesu Krist nag Mark. Trans. D. Morrison. Sydney: Sheriff and Downing.
• Bible. 1874.Kenesis natus a bei nag Moses ki mtir i. Trans. Cosh, J. Sydney: British and Foreign Bible Society.
• Bible. 1875?Nafisan nafousien. Sydney: F. Cunninghame and Co.
• Bible. 1883.The Gospel according to Luke. Trans.Macdonald, D.D. Melbourne: M.L. Hutchinson.
• Bible. 1885.The Gospel according to John, Tus Nanrognrogona Uia ni Iesu Kristo nag Ioane i mitiria. Trans.Mackenzie, J., Macdonald, D.D. Sydney: F. Cunninghame and Co.
• Bible. 1919. Natus bei ni nafisan ni Efate. Sydney: Epworth Press.
• Bible. 1919.Tusi tab fao (New Testament). Trans. Mackenzie, J., Macdonald, D.D. Melbourne: British and Foreign Bible Society.
• Bible. 1923.Scripture History. Sydney: Epworth Printing and Publishing House.
• Bible. 1923.Nafakoron ni aliat. Erakor Efate, New Hebrides. Nouméa: Imprimerie A.-L. Laubreaux.
• Bible. n.d.Nawisien nig Nagmer Apostol. Sydney: F. Cunninghame and Co.
• Clark, Ross. 1973. Transitivity and case in eastern Oceanic languages.Oceanic Linguistics 12(1–2). 559–606.
• ––––– 1978. The New Hebridean outliers. In Wurm, S.A. and L.Carrington, (eds.),Second International Conference on Austronesian Linguistics: proceedings. Fascicle 2: eastern Austronesian. (Pacific Linguistics Series) Department of Linguistics, Research School of Pacific Studies, The Australian National University. 911–928.
• ––––– 1982. "Necessary" and "unnecessary" borrowing. In Halim, A. (ed.),Papers from the Third International Conference on Austronesian Linguistics. Vol.3: Accent on variety. C 76 ed. (Pacific Linguistics Series): Department of Linguistics, Research School of Pacific Studies, The Australian National University. 137–143.
• ––––– 1985. The Efate dialects.Te Reo 28.:3–35.
• ––––– 1996. Linguistic consequences of the Kuwae eruption. In J. M. Davidson, G. Irwin, B. F. Leach, A. Pawley and D. Brown (eds.),Oceanic culture history: essays in honour of Roger Green. New Zealand Journal of Archaeology Special Publication. 275–285.
• ––––– n.d. The Efate-Tongoa dialects (Ms).
• Codrington, Robert Henry (R. H.). 1885.The Melanesian Languages. Oxford: Clarendon Press.
• Crowley, Terry. 1998.An Erromangan (Sye) Grammar. Honolulu: University of Hawai’i Press.
• Lynch, John. 2000. South Efate phonological history.Oceanic Linguistics 39(2):320–338.
• ––––– 2000.A grammar of Anejom. Canberra: Pacific Linguistics.
• ––––– 2001.The linguistic history of Southern Vanuatu. Canberra: Pacific Linguistics, Research School of Pacific and Asian Studies, Australian National University.
• ––––– 2004. The Efate-Erromango Problem in Vanuatu Subgrouping.Oceanic Linguistics 43(2):311–338.
• Thieberger, Nicholas. 2004.Topics in the grammar and documentation of South Efate, an Oceanic language of Central Vanuatu. PhD thesis, Department of Linguistics and Applied Linguistics, University of Melbourne.
• ––––– 2006a.A Grammar of South Efate: An Oceanic Language of Vanuatu Oceanic Linguistics Special Publication, No. 33. Honolulu: University of Hawai'i Press.
• ––––– 2006b. The benefactive construction in South Efate.Oceanic Linguistics, Volume 45, no. 2, 297-310.
• ––––– 2007. The demise of serial verbs in South Efate. Diana Eades, John Lynch and Jeff Siegel (eds.),Language Description, History and Development: Linguistic Indulgence in Memory of Terry Crowley. Amsterdam: Benjamins. 237-251.
• ––––– 2011a.Natrauswen nig Efat. Melbourne: The author.ISBN 978-1-921775-50-5.
• ––––– 2011b.A dictionary of South Efate. Melbourne: The author.ISBN 978-1-921775-51-2.
• ––––– 2012. Mood and Transitivity in South Efate.Oceanic Linguistics. Volume 51, no. 2, 387-401.
• Thieberger, Nicholas and Chris Ballard. 2008. Daniel Macdonald and the 'compromise literary dialect' in Efate, central Vanuatu.Oceanic Linguistics, Volume 47, no.2: 365-382
• Thieberger, Nicholas, and members of the Erakor community. 2021.A Dictionary of Nafsan, South Efate, Vanuatu: M̃p̃et Nafsan ni Erakor. Honolulu: UHPress.

An Android app version of the dictionary can be downloadedhere.

• Payne, Thomas Edward. 1997.Describing morphosyntax: a guide for field linguists. Cambridge, U.K.; New York: Cambridge University Press.

• Topics in the grammar and documentation of South Efate, an Oceanic language of Central Vanuatu by Nicholas Thieberger
• Dictionary of South Efate by Nicholas Thieberger
• Nafsan (South Efate) DoReCo corpus compiled by Nick Thieberger. Audio recordings of narrative texts with transcriptions time-aligned at the phone level, translations, and time-aligned morphological annotations.

• Central Vanuatu languages
• Languages of Vanuatu

• Pages containing links to subscription-only content
• Language articles citing Ethnologue 18
• Articles with short description
• Short description is different from Wikidata
• Articles containing potentially dated statements from 2005
• All articles containing potentially dated statements
• Articles with ambiguous glossing abbreviations
• CS1 maint: multiple names: authors list