This book presents a new theory of grammatical categories - the Universal Spine Hypothesis - and reinforces generative notions of Universal Grammar while accommodating insights from linguistic typology.
AUTHOR: Klamer, Marian TITLE: A Grammar of Teiwa SERIES TITLE: Mouton Grammar Library [MGL] 49 PUBLISHER: De Gruyter Mouton YEAR: 2010
Nicholas J. Williams, Department of Linguistics, University of Colorado at Boulder
This book is the first published grammatical description of Teiwa, an endangered non-Austronesian language spoken by about 4,000 speakers in the northeastern region of Pantar, a small island in southeastern Indonesia. Teiwa is a member of the proposed (Timor)-Alor-Pantar (TAP) language family. Wider genetic affiliation of Teiwa and the TAP languages remains unclear, though some have suggested an affiliation with Trans New Guinea based on correspondence of pronominal forms (Ross 2005). ''A Grammar of Teiwa'' is the most in-depth study of a TAP language to date. In addition to a lengthy description of the grammar (423 pages), the book also includes two narrative texts and two songs (25 pages), an English - Teiwa/Teiwa - English word list of 1,312 items, and an index.
In Chapter 1 Klamer provides a detailed introduction to Teiwa and the surrounding languages, including information on the geography, history, and culture of Alor and Pantar (AP). This chapter surveys the claims regarding genetic affiliations of the AP languages, situates Teiwa in its areal and typological context, and describes the methods and data used for the study. The description is based on approximately four months of fieldwork conducted by the author (2003 - 2007) and about 2.5 hours of recorded material. A variety of native speakers assisted in transcription and data collection, making this description representative of the broader Teiwa speech community.
In the remaining chapters (2 - 11) Klamer describes the structure of Teiwa, starting with segmental phonology and ending with information structure. The majority of the book is devoted to clause level phenomena including the verb phrase, noun phrase, grammatical relations, serial verb constructions, etc. The remainder of this review follows Klamer's (1999) own model for reviewing grammars. Rather than summarize each chapter, I will highlight theoretically and typologically interesting aspects of the language.
The phonology of Teiwa is relatively simple with just 8 vowels and 20 consonants. The consonant inventory contains some typologically and areally rare segments, including a uvular stop /q/ and a pharyngeal fricative /ħ/. Teiwa also makes a contrast between /l/ and /r/, a feature shared by all the AP languages, but rare for Papuan languages in general (Foley 1986). As for vowels, a length contrast exists, but is limited to /i/ and /u/. Phonetic processes include the unrounding and centralizing effect of /ħ/ on immediately following stressed vowels (/ħu/, /ħɑ/, and /ħɔ/ are realized as [ħɨ], [ħɜ] and [ħʌ], respectively) and the assimilation of nasals with the place of the following stop or liquid. Consonant clusters are restricted to onsets in which the second consonant is a liquid (/l/ or /r/). Stress is trochaic, with a few exceptions due to quantity sensitivity (heavy syllables attract stress) and the existence of prosodic word compounds. Full, but no partial, reduplication is a productive part of the language.
Major word classes include nouns, verbs, adjectives, adverbs, demonstratives, and spatial deictics. Teiwa is a largely isolating language, with little inflectional or derivational morphology. Thus, evidence for these distinct word classes is limited. Nouns are defined by their inability to reduplicate (in contrast with verbs, adjectives, and adverbs). Only verbs can take the realis suffix and object prefixes. Adjectives modify nouns, can be used predicatively, and do not take the realis suffix. Adverbs cannot be used predicatively. Adverbs are limited and most ''adverbial notions'' are expressed by serial verb constructions.
Among these word classes, of particular interest are the numerous pronominal paradigms in Teiwa. These include the typical subject, object, and possessor pronouns, as well as dual pronouns and pronouns with more idiosyncratic semantics, such as 'x and they,' 'x alone,' and 'x as a group of pronouns'. Subject and possessor pronouns have both short and long forms. In addition to this multitude of pronominal paradigms, several cross-linguistically uncommon forms exist in each one, such as the '3p elsewhere' and 'distributive' pronouns. Teiwa distinguishes 1p exclusive and inclusive, common for AP languages, but rare in Papuan languages more generally. Finally, 3rd person pronominals are losing their distinction in number, with the singular form used most frequently for both singular and plural referents.
Demonstratives and spatial deictics are described separately, though there is no clear morphosyntactic evidence for their status as distinct word classes. Demonstratives in Teiwa include the adnominals 'a' 'proximate,' 'u' 'distal,' 'i' 'forthcoming,' 'waal' 'that (previously) mentioned,' and 'eran' 'that one,' as well as the pronominals 'xa'a' 'this one,' 'xu'u' 'that one,' 'laxa'a' 'this one here,' and 'laxu'u' 'that one there.' Adverbial demonstratives include 'i xa'a' and 'i xu'u' for '(over)here' and '(over)there.' The first two adnominals (a and u) have been analyzed as basically spatial, while i, waal, and eran have basic discourse functions. 'Ga'an', the third person singular object pronoun, is also frequently used as a demonstrative with a discourse function (referent tracking and ''emphasis'' (p. 137)). This analysis leaves one wondering what discourse functions a and u might have, particularly when many examples throughout the book show use of multiple demonstratives in a single noun phrase (e.g. [ga-manak ga'an u xu'u] -- '3s-master 3s distal that.one', (p. 177) or [a bali [eran waal u]] -- '3s see that.one that.mentioned that', translated simply as ''she sees that'' (p. 306)). In general, the analysis of demonstratives and other spatial deictics (not described here) is preliminary, as the author herself acknowledges. This is obviously a rich area for further research.
Basic word order in Teiwa is SV (in intransitive clauses) and APV or, basically, SOV, (in transitive clauses). This is relatively strict, apart from possible extraposition of objects (PAV for full-NP objects and AVP for pronominal objects). Teiwa clearly distinguishes subjects and objects, lacking the ''semantic alignment'' commonly found in other AP languages (Klamer 2008, Kratochvil 2007, Baird 2008, Holton 2009). However, while S and A arguments are all treated the same (as subjects), P arguments receive differential marking based on the animacy of the object. Animate objects are obligatorily expressed by verbal prefixes, with an optional lexical NP or full pronoun. Inanimate objects cannot be expressed by object prefixes. Instead, they are encoded with lexical NPs or pronouns. Teiwa verbs are restricted to intransitives and mono-transitives, with no ditransitive verbs. Expression of three participant events involves the use of serial verb constructions to introduce the additional participant. Somewhat unusual cross-linguistically is the use of the deictic verb 'ma' 'come' in these ''ditransitive'' constructions (verbs meaning 'take' are more common).
Other notable features of the grammar include an alienable/inalienable distinction, the marking of reality status on verbs, the frequent use of serial verb constructions, the lack of embedded structures, and the existence of devoted topic and focus markers. Each of these topics will be described very briefly.
The alienable/inalienable distinction is marked through the use of an optional prefix and long-form pronouns for emphasis with alienable nouns, while inalienable nouns occur with obligatory prefixes and short-form pronouns for emphasis of the possessor. Inalienable nouns are limited to body part and kin terms.
In chapter 7, Klamer describes the use of the only non-argument related verbal morphology, '-an', the realis marker. The use of this morpheme distinguishes ''reality status'' as a category distinct from mood and modality, although it interacts with mood in interesting ways. Most notable in this regard is the optionality of realis marking on past and future events, questions, and prohibitives, in each case giving the expression of the event a different quality if marked realis. Rather unexpectedly, negated verbs can be marked either realis or irrealis (irrealis is expected for negated forms). Klamer leaves the meaning of this variation as an open question.
Serial verb constructions (described in chapter 9) play an important role in the language, used to express various ''adverbial notions,'' to encode modality and aspect, and to introduce additional participants, among other functions.
In chapter 10 (''Clause combinations'') Klamer argues for a lack of subordination or embedding in Teiwa, including a lack of relative clauses (all surprising for a ''Papuan'' language and contra Foley 1986). In general, clause combination in Teiwa typically involves simple juxtaposition, with optional use of temporal and coordinating conjunctions. There is a lack of special verb forms in combined clauses and no ''switch-reference'' marking (also unusual for a Papuan language).
Finally, chapter 11 presents analysis of information structure in Teiwa, noting the existence of both a dedicated focus marker (la) and a topic marker (ta). While 'la' is used to mark new information, 'ta' marks a shift in topic, a ''topic discontinuity,'' often re-introducing something previously discussed. In this chapter Klamer also describes the discourse function of the demonstratives 'waal' and 'i'.
A Grammar of Teiwa is an excellent description of an endangered and previously undocumented language. To date, as noted above, it is the most detailed study of an Alor-Pantar language. This makes it an important contribution to the growing literature on these languages and Papuan languages more generally.
I have a few critical comments on two significant formatting issues with the book. First, elicited and text-based examples are not distinguished. Furthermore, the source of each example is never indicated, making it difficult to find out more about the context of the utterance. Since many of the examples presumably were taken from texts such as those in the appendix, it would be extremely helpful to the reader to link each example to the specific text from which it was taken. The second issue with the formatting has to do with the surprising frequency of typographical errors in the text (at the rate of maybe one every few pages, roughly -- I stopped noting them after the first chapter). While they do not detract significantly from flow of the book and offer no problem for understanding the text, they are an occasional irritation.
Aside from these formatting issues, this grammar is a most enjoyable read. While some areas of the grammar remain underexplored, Klamer does an excellent job of describing enough of the language in sufficient detail to allow further investigation based on the available texts. Open questions and underexplored areas are acknowledged as such. Despite the relatively short time the author spent in the field, she demonstrates deep knowledge of the language. This book will become an important resource for scholars working on Papuan languages and Alor-Pantar languages in particular. The data and analyses provided will be valuable to linguists more generally, especially typologists interested in topics as diverse as grammatical relations, serial verb constructions, verbal morphology (e.g. reality status marking), as well as demonstratives and spatial deixis.
Baird, Louise. 2008. A Grammar of Klon: A Non-Austronesian Language of Alor, Indonesia. (Pacific Linguistics 596). Canberra: Pacific Linguistics.
Dixon, R. M. W. (1979). Ergativity. Language, 55 (1), 59-138. (Revised as Dixon 1994).
Foley, William A. 1986. The Papuan Languages of New Guinea. Cambridge: Cambridge University Press.
Holton, Gary. 2009. Person-marking, verb classes, and the notion of grammatical alignment in Western Pantar (Lamma). Typological and Areal Analyses: Contributions from East Nusantara, ed. by M. Ewing & M. Klamer, 97-117. Canberra: Pacific Linguistics.
Klamer, Marian. 1999. Review of Grammar of Tukang Besi (1999) by Mark Donohue, (Mouton de Gruyter: Berlin/New York), LinguistList 10.1439, October, 1999.
Klamer, Marian. 2008. The Semantics of Semantic Alignment in eastern Indonesia. In: Mark Donohue & Søren Wichmann, eds. Typology of Active-Stative Languages, pp. 221-251. Oxford: Oxford University Press.
Kratochvíl, František. 2007. A Grammar of Abui, Leiden University Ph.D. dissertation.
Ross, Malcolm. 2005. Pronouns as a preliminary diagnostic for grouping Papuan languages. In: Andrew Pawley, Robert Attenborough, Robin Hide & Jack Golson, eds. Papuan pasts: cultural, linguistic and biological histories of Papuan-speaking peoples. Canberra: Pacific Linguistics. pp. 15--66.
ABOUT THE REVIEWER
ABOUT THE REVIEWER:
Nicholas Williams is a PhD student at the University of Colorado at
Boulder, USA. He is beginning a project to document the Kula language of
eastern Alor for his doctoral dissertation. His interests include language
documentation and description, Papuan and Austronesian languages,
interactional linguistics, conversation analysis, and linguistic anthropology.