Annotated Swadesh wordlists for the Alor-Pantar group (West Trans-New Guinea family).

Languages included: Takalelang Abui [alo-abt]; Atimelang Abui [alo-aba]; Bring Klon [alo-kln].

DATA SOURCES

I. General

Holton et al. 2012 = Gary Holton, Marian Klamer, František Kratochvíl, Laura C. Robinson, Antoinette Schapper. The historical relations of the Papuan languages of Alor and Pantar. In: Oceanic Linguistics, Volume 51, no. 1 (June 2012), pp. 86-122. // A sketch of historical phonetics of the Alor-Pantar language group.

Martis et al. 2000 = N. Martis, W. Kurniawati, B. Aritonang, H. Astar, F. Feirizal. Monografi kosakata di kabupaten Alor. Jakarta. // 200-item Swadesh wordlists collected in various villages of the Alor island. An unreliable source with a number of typos present.

Stokhof 1975 = W. A. L. Stokhof. Preliminary Notes on the Alor and Pantar Languages (Pacific Linguistic B43). Canberra: Australian National University. // 117-item wordlists for various languages of the Alor and Pantar islands, collected during Stokhof's fielworks 1972, 1974.

II. Abui

DuBois 1938/1987 = C. DuBois. Abui. In: W. A. L. Stokhof (ed.), in co-operation with A. E. Almanar. Holle lists. Vocabularies in languages of Indonesia. Vol. 11: Celebes, Alor, Ambon, Irian Jaya, Madura and Lombok. The Australian National University, 1987. Pp. 85-114. // Ca. 1200-item wordlist of Abui Atimelang, recorded in 1938 by Cora DuBois and published by W. A. L. Stokhof. Supplemented by DuBois' & Stokhof's notes and several short texts recorded by DuBois.

DuBois 1944 = C. DuBois. The People of Alor. A Social-Psychological Study of an East Indian Island. Minneapolis. // Ethnological studies on the Abui society, supplemented by a short wordlist (pp. 564-565) of the Atimelang dialect of Abui.

Kratochvíl 2007 = F. Kratochvíl. A grammar of Abui. Utrecht: LOT. // Descriptive Abui grammar, based on the dialect of the Takalelang village. Supplemented by an Abui-English wordlist.

Kratochvíl & Delpada 2008 = F. Kratochvíl & B. Delpada. Kamus Pengantar Bahasa Abui. Abui-Indonesian-English dictionary. Kupang, Indonesia: UBB-GMIT. // Dictionary of the Abui language, supplemented by a grammar sketch and an English-Abui index. Based on the Takalelang dialect.

Nicolspeyer 1940 = M. M. Nicolspeyer. De sociale structuur van een aloreesche bevolkingsgroep. Ph. D. diss. Rijswijk: Kramers. // Ethnological studies on the Abui society, supplemented by glossed texts (pp. 74-155) and a vocabulary (pp. 156-180) of the Atimelang dialect of Abui.

Stokhof 1984 = W. A. L. Stokhof. Annotations to a text in the Abui language (Alor). In: Bijdragen tot de Taal-, Land- en Volkenkunde 140 (1984), no. 1, Leiden. Pp. 106-162. // Edition of a short Abui Atimelang text, recorded by Cora DuBois in the 1930s. Supplemented by various grammatical comments.

III. Klon

Baird 2008 = Louise Baird. A grammar of Klon: a non-Austronesian language of Alor, Indonesia (Pacific Linguistics #596). Canberra. // A descriptive grammar of the Klon language (mostly based on the Bring dialect), supplemented with a glossary.

Rosetta Kelon = The Rosetta Project: Kelon Swadesh List. n.d. http://archive.org/details/rosettaproject_kyo_swadesh-1. // An anonymous circa 100-item Swadesh wordlist (or rather several joint wordlists) of the Klon language (the dialect is unknown). Apparently, the main source for the compilation was [Stokhof 1975].

Martis et al. 2000 = N. Martis, W. Kurniawati, et al. Monografi kosakata di kabupaten Alor. Jakarta. // 200-item Swadesh wordlists collected in various villages of the Alor island. Not very reliable because of a large number of typos.

NOTES

1. General.

I. Abui

The Abui language consists of several dialects, out of them Takalelang is the best documented one, see the grammar [Kratochvíl 2007] and the dictionary [Kratochvíl & Delpada 2008].

The published materials on the Atimelang (Ateng Melang) dialect are also sufficient for 110-item wordlist compiling. The main source is the large wordlist [DuBois 1938/1987] with the additional data from [DuBois 1944; Nicolspeyer 1940; Stokhof 1975; Stokhof 1984].

Makadai dialect: the only source is [Stokhof 1975], where on p. 54 the 117-item wordlist of Abui Makadai is offered. Unfortunately, only ca. 70 items of our 110-wordlist have been found in the source. In view of this we prefer not to allocate the Makadai data into a separate list, but quote these forms in the notes on the Abui Takalelang list. For transcription see notes on the Abui Atimelang list.

Fanating & Aila dialects: a couple of forms found in [Kratochvíl 2007]. We quote these in the notes on the Abui Takalelang list.

Grammar notes

Many of the Abui verbs possess two different stems. These stems differ in their final elements (see [Kratochvíl 2007: 83 ff., 210 ff.] for details) and can be labeled as "imperfective" (continuative & inceptive according to Kratochvíl) and "perfective" (completive according to Kratochvíl). Examples are: bˈuːk (imperfective) / bˈuːt-i (perfective) 'to drink', mˈoŋ (imperfective) / mˈón-i (perfective) 'to die'. Here the imperfective stems are the result of various sound changes in the final position (e.g., t# > k, p# > k, n# > ŋ). In these cases we quote the perfective form as a main representation of the verb and list the imperfective form in the notes.

For some reason, a more ambiguous case is represented by verbs with the final r / l alternation. We prefer to quote both stems in the main field for such verbs.

II. Klon (Bring)

The Klon (a.k.a. Kelong, Kelon, Kalong, Kolon) language consists of two dialects: Bring (villages: Probur, Probur Utara, Tribur) and Paneia (villages: Halerman, Margera, Manatang), see [Baird 2008: 3 f.].

Our Klon list is based on the Bring dialect data; Paneia forms - if available - are quoted in the comments.

The main source for the Bring dialect is the grammar [Baird 2008], supplemented with a glossary, plus the 117-item wordlists from the Probur village published in [Stokhof 1975]. It should be noted that the 200-item wordlists in [Martis et al. 2000: 69-74], titled as "Probur", actually represent the neighboring Abui language (the exact dialect is unclear), but they do indeed include sporadic Klon-like forms.

Forms of the Paneia dialect have been taken from [Stokhof 1975] (the Halerman village 117-item wordlist) and [Baird 2008: 228] (a very short list of specific Paneia words). In some cases, the anonymous Klon wordlist, published by the Rosetta Project [Rosetta Kelon], was used as additional proof.

2. Transcription.

I. Takalelang Abui

Normally we use the transcription offered in [Kratochvíl & Delpada 2008] instead of the Abui modern orthography. The approximate transliteration of Kratochvíl's alphabet is as follows:

[Kratochvíl & Delpada 2008]	GLD
p	p
b	b
f	f
w	w
m	m
t	t
d	d
s	s
n	n
r	r
l	l

c	č
j, ɟ	ǯ
y	y

k	k
g	g
ng	ŋ

'	ʔ
h	h

i	i [y after vowels]
ii	iː
e	e
ee	eː
u	u
uu	uː
o	o
oo	oː
a	a
aa	aː

stress	ˈ [before the vowel; in most cases the stress is not phonological; the secondary stress ˌ is unnoted]

V́	V́ [high tone]
V̀	V̀ [low tone]
V	V [middle tone, unmarked]

1) It must be noted that, according to [Kratochvíl 2007: 26], the short vowels are lax and generally lower than their long counterparts, i.e. i is rather [ɪ], e is [ɛ], o is [ɔ], u is [ʊ], a is [ɑ] vs. [iː], [eː], [oː], [aː], but we prefer not to complicate our transcription with this non-phonological distinction.

2) Phonemes č, ǯ and g occur only in recent Malay and Indonesian loanwords [Kratochvíl 2007: 25].

3) It is quite possible that the low tone is not phonological (excepting an iconic three-wise distinction: ʔo 'that [medial horizontal]' ~ ʔò 'that [medial down there]' ~ ʔó 'that [medial up there]'), but this question requires additional investigation.

II. Atimelang Abui

Glottal stop (ʔ), which appears by default before a vowel onset in Abui Takalelang, is normally unnoted in the sources of Abui Atimelang.

Both in Takalelang and Atimelang Abui there is a tendency to close the final vowel with some laryngeal sound (h, ʔ) or creaky phonation. Cf., e.g., Takalelang ʔíaʔ, ʔìaʔ for ʔía, ʔìa in [Kratochvíl 2007: 59]. The same tendency is reflected in [DuBois 1938/1987], where the ayin (ʕ) is often found at the end of vowel-final forms (e.g., waʕ = Kratochvíl's wa 'mouth', tukuʕ = Nicolspeyer's tuku = Kratochvíl's toku 'leg, foot' and even weʕ = Kratochvíl's wei 'ear'), and in [Stokhof 1975], where an aleph (ʔ) is sporadically used in the case of the vowel final. DuBois explains ([DuBois 1938/1987: 114]; [Stokhof 1984: 116]) that her ʕ means h. We omit these final characters in our transcription, but quote the original spelling in the notes (rendering these final ʕ & ʔ as ʔ).

In [Nicolspeyer 1940] & [DuBois 1944] some words are spelled with the final letter sequences aug and eug. In [DuBois 1938/1987], these words have counterparts with final auġ, aġ, eġ and oġ. What the sign "ġ" actually means is unclear, but it is noted by DuBois (see [Stokhof 1984: 116]) that the rare dotted "ġ", as well as the very rare "ḳ", represents a "back velar" (it is obvious, however, that there is no phonological opposition of velars vs. uvulars in Abui, cf. [Kratochvíl 2007: 26] for the Abui Takalelang phoneme k that is realized as [k] in the initial position and as [q] in the medial and final positions). The Atimelang forms with "ug/ġ" correspond to forms with h in the Takalelang dialect (e.g., taug/tauġ 'on top' = Kratochvíl's tah) or to final Vi (e.g. nauġ 'tear(s)' = Kratochvíl's nai). Stokhof [Stokhof 1984: 135] points out that DuBois' "ug" can paradigmatically alternate with DuBois' "h" (e.g., teug ~ teh-i 'to dig'; the same in Nicolspeyer's data: taug ~ tah-i 'to join'). We normalize "ug/ġ" as h, quoting original spellings in the notes.

In [Nicolspeyer 1940], [DuBois 1938/1987], & [DuBois 1944] the graphemes w and v are distinguished (their articulatory mechanisms are not explained). From the statistical viewpoint, the letter "v" normally occurs only as a root onset before e & i; the letter "w" - before any vowel. In [Stokhof 1975] only "w" is used; the same concerns our sources of Takalelang Abui (only "w"). The opposition w ~ v is not very consistent, cf. Nicolspeyer's vel 'to wash' ~ DuBois' wel-a-ng 'id.' or vice versa: Nicolspeyer's we 'blood' ~ DuBois' veʔ / ve 'id.'. We transcribe both "w" and "v" as w (quoting original spellings with "v" in the notes).

In [DuBois 1938/1987] y can be rendered as "j" or "dj". The letter "j" for [y] is a norm of the Oceanistic tradition, whereas the conventional function of "dj" is a ǯ-like affricate. The usage of "j" and "dj" in DuBois' data is not consistent, cf. DuBois' dja 'water', but ja dui 'drinking bowl', ja sua 'to fetch water' (= Nicolspeyer's ja 'water'). Apparently DuBois' "dj" denotes some non-phonological strengthening of [y]. We normalize "j" and "dj" as y, quoting original spellings with "dj" in the notes.

We simplify the vowel system of [Stokhof 1975] in the following way: "a" & "ɑ" are rendered as a, "e" & "ɛ" - as e, "o" & "ɔ" - as o, because these oppositions are apparently non-phonological. The same concerns [DuBois 1938/1987], both of whose "e" and "E" (i.e. [ɛ]) we render as e.

The Dutch vowel sequence "oe" in [Nicolspeyer 1940], [DuBois 1938/1987], [DuBois 1944] is rendered as u.

Vowel length and tone are unnoted in all available sources of Atimelang Abui. In [Stokhof 1975] vowels can be sporadically doubled, nevertheless we transcribe these as shorts (quoting Stokhof's orthography in the notes).

Stress is unnoted in [Nicolspeyer 1940], [DuBois 1938/1987], [DuBois 1944], but in [Stokhof 1975] the signs ˈ and ˈˈ before a vowel are used to mark "the relative prominence of that vowel". We treat both signs as a stress indicator (ˈ).

III. Klon (Bring)

We transliterate the Klon orthography as follows:

Traditional orthography	GLD
p	p
b	b
w	w
m	m

t	t
d	d
s	s
n	n
r	r
l	l

j	ȡ
y	y

k	k
g	g
ng	ŋ

q	ʔ
h	h

i	i
ii	iː
é	e̝
e	e
ee	eː
u	u
uu	uː
o	o
oo	oː
ò	ɔ
a	a
aa	aː

1) Normally, the phoneme ȡ occurs in recent Malay and Indonesian loanwords, but it is also attested in several indigenous words in the final position (according to [Baird 2008: 15-16], some older speakers still pronounce d instead of ȡ).

2) The sounds g and ŋ seem to be in a state of complementary distribution: g only occurs before a vowel (and also in initial consonant clusters?), ŋ only occurs before a consonant or a pause (but not word-initially).

3) Other than the aforementioned vowels, there is also a very frequent vowel ǝ, which is not denoted in the traditional orthography at all. It only occurs between consonants in the unstressed position and can be regarded as a reduced unstressed variant of the aforementioned vowel phonemes (as noted in [Baird 2008: 22, 24], "the vowel found in unstressed syllables is typically reduced to schwa").

4) Orthographical consonant clusters in the initial position or between a prefix and following morphemes actually represent the sequence CǝC. According to [Baird 2008: 25], this is also true of the majority of orthographic consonant clusters in the medial position, but not all of them (details are unclear).

5) e̝ and ɔ are marginal phonemes. It must be noted that the opposition e̝ / e eː {é / e ee} is actually [e / ɛ ɛː], but, for the sake of convenience, we prefer to transcribe the frequent phonemes {e ee} as e eː, not ɛ ɛː.

6) Stress is non- or quasi-phonemic, normally depending on the "heaviness" of syllables [Baird 2008: 22 f.]. Stress is not noted in traditional orthography.

7) Four diphthongs are attested: ai, ei, oi, ui [Baird 2008: 25].

8) We simplify the vowel system of [Stokhof 1975] in the following way: "a" & "ɑ" are rendered as a, "e" & "ɛ" - as e, "o" & "ɔ" - as o, "i" & "ı" - as i, "u" & "ʊ" - as u. Additionally, Stokhof's double (tense?) consonants are rendered as plain.

Database compiled and annotated by:
Abui (Takalelang, Atimelang): A. Kassian, August 2011 / revised July 2013 (minor corrections).
Klon (Bring): A. Kassian, July 2013.