Around 110 km just off the southwestern coasts of the island of Sumatra, lies an outlying island called Pulau Enggano. Home to around 4000 to 5000 people, around under a half of the island’s residents belong to the Enggano or E Lopeh people group. The language they speak is more enigmatic though, as despite the largely agreed-upon classification of the Enggano language as part of the Austronesian language family, there are some strong arguments that it might just be a language isolate which has received Austronesian influence. And if it really is an Austronesian language, the Enggano language would be a pretty divergent one at that.

Traditional cognate analysis suggested that Enggano is a language isolate, as it shared only a few cognates with other Austronesian languages, and retained only 21% of its words from the reconstructed proto-Austronesian forms. This low proportion of cognates is unusual of the languages in its geographical vicinity, and more consistent with those one would find in the Austronesian languages in Melanesia.

Owen Edwards’ 2015 publication would shatter this paradigm, by concluding that many of Enggano’s grammatical affixes, some numbers, and pronoun systems could be derived from proto-Malayo-Polynesian reconstructions, and that Enggano could be classified as an isolated branch within the Austronesian languages, pointing towards its unique divergence from the Austronesian languages in the region, and the possibility of a high rate of lexical replacement in the language. However, one criticism could be that Enggano borrowed these words and particles from the regional Austronesian languages, adding them to a foundation of unique Enggano words that do not seem to be related to other Austronesian languages. Nevertheless, major language catalogues still classify Enggano as an Austronesian language, such as Ethnologue’s shift towards this classification (previously language isolate) in 2015.

Having only made first contact with Europeans in the mid-16th century, the first known significant documentation of the Enggano language occurred in 1937 by Hans Kähler, with a reference grammar in 1940. However, previous studies with wordlists have been published throughout the later half of the 19th century, with the earliest one dating back to 1853. Contemporary research is primarily done by linguists such as Brendon Yoder’s 2011 thesis, Bernd Nothofer’s publications in the 1990s, in German and English, but studies are still generally few and far between. Over the course of Enggano’s documented linguistic history, several observations could be interpreted, such as the loss of dialectal variation in the Enggano language, with Kähler only recording one dialect spoken by Enggano speakers at the time of his studies. There is also an undated grammatical sketch by Terry Crowley, which was transcribed from his handwritten notes of the language, making particular reference to Kähler’s 1940 publication.

Enggano is one of the few languages in the world which exhibits a system of nasal harmony, with other notable languages generally clustered in the Amazon. This phenomenon is where every voiced plosive consonant and vowel in the word that follow a nasal vowel will become nasalised. That is, /b/ becomes /m/, and /d/ becomes /n/. Thus, the words e-uku (ribs) and ẽ-ũkũ (flatulence) take different meanings in Enggano, and form a minimal pair. This, according to linguist Owen Edwards, could have lead to a loss of phonemic distinction between nasal and plosive consonants. Other sound changes, such as development of allophonic variation, or sound mergers at some point in Enggano’s history could have contributed to the language’s unusually small consonant inventory compared with other Austronesian languages in the region. Other sounds may only occur in specific parts or specific native words as well.

Despite Edwards’ relative recency in covering the Enggano language, he made particular references to Kähler’s 1940 publication, as with the phonology of the language. There is a reported observation of vowel leak where a vowel before the glottal consonants /ʔ/ and /h/ would also be found after these consonants, forming a diphthong with the vowel that is supposed to follow the glottal consonant. These are the sounds of the Enggano language (Edwards, 2015):

At first sight, Enggano numerals do not resemble anything like those one would see in the Austronesian languages, which could indicate a history of unusual sound changes, or the innovation of their own number system. A particularly strong argument for the latter is the use of a base-20, or vigesimal number system, with sub-bases of 5. This contrasts with the decimal (base 10) number systems used in a majority of Austronesian languages, and even the quinary (base 5) number systems used in some Austronesian languages spoken in Melanesia. A strong argument for the former would involve comparing the numbers 1-5 with other Austronesian languages, taking note of the various changes that might have been involved. See below for a comparison between Enggano, Indonesian, and Batak Alas numerals.

English	Enggano	Indonesian	Batak Alas
one	kahaiʔ	satu	sadə
two	ʔaru	dua	duə
three	ʔakər	tiga	təlu
four	ʔaup	empat	əmpat
five	ʔarib	lima	limə
six	ʔakiʔakin (2 x 3?)	enam	ənəm
seven	ʔarib he ʔaru (5 + 2)	tujuh	pitu
eight	kĩpãʔĩõp / ʔãpãʔĩõp  ʔáopa hii ʔáopa (4 + 4)	delapan	waluh
nine	[kĩpãʔĩõp / ʔãpãʔĩõp] kaba kahaiʔ (8 + 1?)	sembilan	siwah
ten	kĩpãʔãũ	sepuluh	səpuluh
twenty	kahaiʔ kak (one person)	dua puluh	duə pulu
hundred	ʔarib kak (5 x 20) kahaiʔ ratuh (from ID)	ratus	səratus

If we take the ʔa– and ka- as prefixes of some sort, and consider the loss of final vowels in contemporary Enggano, Enggano numerals 1-5 start to resemble a bit like those in its purported Austronesian cousins. It seems that here, the /s/ sound could have become /h/ in Enggano, /d/ and /l/ to /r/, and /m/ to /b/ (possibly due to development of nasal harmony). 6 could have been a reduplication of 3, indicating a multiplication of 2 and 3.

When reading up on the numerals of this language, I found that the etymology for the number eight to be somewhat unusual. Both forms for the number eight appear to derive from the verb ‘to hug’, pã’ĩõp. Brendon Yoder’s 2011 analysis speculated that this is much akin to how the number 8 looks, which could suggest that this word could have been a more recent development, possibly after written Arabic numeral forms were introduced. It is perhaps the most parsimonious explanation behind the etymology of the Enggano ‘eight’. Another word I found, from Edwards’ 2015 paper, is ʔáopa hii ʔáopa, which could translate to 4 + 4, though it was from Kähler’s 1940 study or the 1987 word list, suggesting that it could be a more archaic term. 20, on the other hand, translates quite literally to ‘one person’, indicating the use of a digit tallying system in the Enggano, using the 20 fingers and toes on a person as tally points. 400, according to the 1940 grammar, would translate to “both our bodies” or e-kudodo-ka.

Today, however, Indonesian or Malay numerals are preferred, and according to this source, younger speakers have mostly forgotten Enggano numerals. Older speakers are not immune to this problem though, as the informants providing these data needed some discussion for larger numerals beyond 20.

Perhaps one of the more salient features of Enggano’s grammar is the use of case prefixes. This feature is a rarity amongst the world’s languages, let alone the Austronesian language family where nouns generally do not inflect for grammatical case. The three cases are the nominative, the oblique, and the locative, represented by e-, u-, and i- respectively. These affixes will still be affected by nasal harmony as well. One example is the sentence

• Kia ki-pudu e-koyo ẽʔãnã iʔioo u-bohe
• 3sg verb-kill NOM-pig that PREP OBL-spear
• He kills that pig with the spear

Edwards also noted that these affixes used in Enggano’s inflectional morphology could generally be traced to reconstructed proto-Malayo-Polynesian counterparts. Some are markedly similar, like pa- for the causative in verbs, much akin to *pa- in proto-Malayo-Polynesian for the causative as well. Others are not as intuitive, and would require thinking about the sound changes in the language. This includes ka- for the accidental in verbs in Enggano, and *taR- in proto-Malayo-Polynesian. Through studying the sound changes in Enggano, Edwards has provided a strong case for Enggano’s position as an Austronesian language despite the abnormally low lexical retention and cognates it shares with other Austronesian languages in the region and proto-Malayo-Polynesian. He also highlighted that it is these unusual sound changes that compounded the difficulty in identifying cognates in Enggano words.

Today, the Enggano language is still spoken by around 700 people as of 2011, but most, if not all of its speakers are at least bilingual with Bahasa Indonesia. Younger speakers tend to converse in Indonesian, leading to some concerns over the erosion of this language via language shift. One such example could be seen in how numerals are used in Enggano, as younger speakers tend to prefer to use Indonesian numbers to communicate numeral quantities. After a long history of isolation and limited contact with Sumatran languages that could have contributed to Enggano’s divergence from the rest of the Austronesian languages, the Enggano language is now classified as threatened or endangered by the Endangered Languages Project. With a scarcity in studies looking into the grammar of the Enggano language, it could be a race against time to properly study and elucidate how the Enggano language works, and how it should be classified.

Further reading

Crowley, T. (n.d.) “Grammatical sketch of Enggano”.

Edwards, O. (2015) “The Position of Enggano within Austronesian”, Oceanic Linguistics, 54 (1), pp. 54–109.

Kähler, H. (1940) “Grammatischer Abriss des Enggano”, Zeitschrift für Eingeborenen-Sprachen. 30, pp. 81–117, 182–210, 296–320.

Nothofer, B. (1986) “The Barrier Island Languages in the Austronesian Language Family”, Focal II: Papers From the Fourth International Conference on Austronesian Linguistics, 94, pp. 87–109.

Nothofer, B. (1992) “Lehnwörter im Enggano”, Kölner Beiträge aus Malaiologie und Ethnologie zu Ehren von Professor Dr. Irene Hilgers-Hesse, 1, pp. 21–31.

Nothofer, B. (1994) “The relationship between the languages of the Barrier Islands and the Sulawesi-Philippine languages”, Language contact and change in the Austronesian world, Berlin, Mouton de Gruyter, pp. 389–409.

Yoder, B. (2011) “Phonological and phonetic aspects of Enggano vowels”, MA thesis, SIL International.