From: Yury Lander <yulanderyandex.ru>
Subject: A Grammar of Teribe
AUTHOR: Quesada, J. Diego TITLE: A Grammar of Teribe SERIES: Lincom Studies in Native American Linguistics 36 PUBLISHER: Lincom GmbH YEAR: 2000 Announced at http://linguistlist.org/issues/14/14-1824.html
Yury A. Lander, Institute of Oriental Studies, Moscow
This monograph presents a full-length grammatical description of Teribe, a Chibchan idiom spoken in Northwestern Panama. Teribe is a dialect of a language occasionally called Naso, which constitutes a separate branch of the Chibcha family. Not so long ago Naso included also another dialect known as Terraba. The latter was used until recently by descendants of a group ''relocated'' to the present Costa- Rica. Now, however, the Terraba dialect seems to have disappeared, while Teribe is still actively employed. Thus, this grammar documents an idiom that still has about 1,000 speakers (according to the monograph) and is felt to be an important property of the Teribe people contrasting it with other cultures (as follows, for instance, from the fact that the author of the volume had to get a permit from the King of Teribe in order to study the language). That is why Quesada, the author of the grammar, was able to fulfill an investigation of a number of moot issues and provide a rather deep study.
It is worth noting that despite the fact that the Terraba dialect is now dead, it got more descriptions than Teribe. In reality, until the late 1990s, Teribe was honored just a few papers written by SIL scholars and describing single aspects of the dialect (sometimes inaccurately, according to the present grammar). Subsequent works include a detailed description of Teribe phonology by Oakes (2001), but first of all a number of studies undertaken by Quesada. The monograph reviewed here, however, is only partly based on those studies (and in fact, occasionally presents corrections to the author's former views). Anyway, this is the first grammar of Teribe and - as we will see shortly - perhaps the first grammar published in English that considers a Chibchan language so elaborately.
The body of the book consists of an introductory part and three chapters devoted to phonology, morphology and syntax. The exposition is supplied with sample texts and bibliography. The chapters of the book are further divided into sections, of which some constitute rather large and autonomous pieces. Below such sections will be treated separately.
Chapter 1 gives general information on the Teribe people and the genetic, areal and sociolinguistic context of their language. This chapter also includes a brief section on previous studies of Teribe and Terraba and an overview of the basic typological characteristics of Teribe.
Chapter 2 is devoted to the Teribe phonology. After presenting the list of phonemes, Quesada discusses most phonological oppositions illustrating them with minimal pairs. Then the author provides information on various phonological processes, suprasegmental phenomena (such as tone and stress) and introduces orthographical conventions accepted in the monograph. Remarkably, this chapter demonstrates that phonological means do not all play the same role in Teribe, since some contrasts that are arguably more complex than others (e.g., the aspiration contrast, which exists here for voiceless stops only, and tone oppositions) occupy a rather peripheral place in the Teribe system being in most cases neutralized.
Although Chapter 3 is entitled ''Morphology'', for the most part it does not deal with morphology per se (that is, with parts of words), since many grammatical notions are expressed in Teribe analytically. As a result, this chapter presents mainly the description of word classes and grammatical markers co-occurring with them.
Section 3.1 is concerned with nominals (i.e. nouns and various types of pronouns) and their satellites, basically those that are related to the expression of quantity - plural markers, a partitive particle etc. A typologically important feature of Teribe is the existence of numeral classifiers. They are too discussed in this section, although as we will see later, the attribution of numeral classifiers to nominal satellites is not without problems. Finally, already this section shows the author's interest in the interaction of the information structure and grammar, since it concludes with a detailed discussion of topic and focus markers.
Section 3.2 describes verbs, the word class that in fact displays the bulk of 'real morphology' in Teribe. Somewhat surprisingly, Quesada includes here a very brief description of non-verbal, 'asyndetic' predications, which are said to ''roughly correspond to copular sentences in other languages'' (p. 64). This becomes more understandable when one finds that the verb word class in Teribe is rather heterogeneous grammatically. In particular, Quesada shows (though not explicates) that putative verbs may differ in how many prototypical verbal features they display, with some lexemes being in a sense ''less verbal'' than others. Thus, for instance, for positional verbs the author states that ''there is no morphological evidence for the verbhood of these forms'' (p. 66), for movement verbs he argues that their behavior in certain aspects ''can be summarized as halfway between positional and intransitive verbs'' (p. 69), etc. Non-verbal predicates apparently constitute one of the extreme poles of this scale, thus their presentation in this section contributes to its complete picture.
Next, this section turns to verb categories, among which Quesada marks out aspect, person/number, position, and modality. To be sure, most of them are formally optional and/or require specific constructions (consequently, Quesada has to describe not only their functions but also the relevant contexts). A remarkable exception is aspect, which therefore lies in the heart of (some part of) the verb class and should be of definite interest for students of verb categories. Thus already the fact that there exists some aspectual variation which depends on syntax and information structure gives cause for reflection. Unfortunately, the monograph presents only the first approximation to this fragment of grammar, since it only makes formal statements and briefly discusses relevant meanings (although note that this topic is touched upon also in another chapter where inverse constructions are described).
Adjectives, which make the topic of Section 3.3, are perhaps the least interesting of content word classes in Teribe. They are claimed not to ''inflect'' like nouns or verbs (p. 85, but recall that the nature of ''inflection'' in Teribe is different from that in, say, European languages). Hence most of this section deals with adjective formation, the more so as non-derived adjectives in Teribe constitute only a small class (carefully described here), while most adjectives are derived by suffixation, or reduplication, or compounding.
The last section of the chapter deals with minor word classes, which include adverbs, postpositions, conjunctions, markers of question and negation, and particles. This section naturally involves a good deal of syntactic information (so that, for instance, adverbs are treated here together with complex adverbial phrases), hence it may serve as a bridge to the next chapter.
Chapter 4 provides a syntactic characterization of Teribe, which proves to require investigations into quite a few complex issues, such as the extremely puzzling encoding of grammatical relations, the interaction between a participant's topicality and its expression within the clause etc.
Section 4.1 of this chapter is said to be devoted to simple sentence. Nonetheless, here one can also find data on the structure of noun phrases, comparatives, possession as well as on certain virtually complex constructions (verb serialization and the relative construction). One topic shared by many parts of this section concerns identification of grammatical relations in Teribe and the related problem of the nature of inverse constructions. In addition, this section contains information on several ''valence and syntactic operations'' - causative constructions, external possession and dative shift, reflexives etc.
Section 4.2 focuses on complex sentences, among which the author contrasts between coordinate and subordinate (paratactic and hypotactic in Quesada's terms) constructions. In fact, this topic is not elaborated deeply in the grammar, so in most cases the author restricts himself to presenting general descriptive information on grammatical means involved in clausal combinations.
Finally, Section 4.3 deals with the reflections of the information structure in the organization of Teribe sentence. Note that this issue is touched upon in various parts of the grammar, yet this section provides a useful summary and discussion of the relevant facts presented in other sections.
Chapter 5 is in a sense an appendix to the grammar, since it contains five texts of different genres and length. All texts have morpheme-by- morpheme glosses together with a free English translation.
The monograph concludes with a bibliography.
Any language is interesting but it is a merit of a description that a language appears as interesting. Teribe as it is represented in this grammar looks indeed intriguing in some respects and indicative in others. This is achieved apparently by the author's scrupulousness and occasional keenness on concrete issues.
Typologically, perhaps the most notable feature of Teribe is the way it distinguishes between grammatical relations. It is not surprising therefore that this aspect of the Teribe grammar is mentioned in several parts of the monograph and honored with several subsections. In general, Teribe employs three means for distinguishing between core grammatical relations, namely (i) word order, (ii) morphological forms of pronouns (nominative vs. oblique), and (iii) ''agreement'', whereby verbs are suffixed with person morphemes (which originated from oblique pronouns) and normally are not accompanied by any corresponding lexical noun phrase. The last two means are used almost exclusively with pronominal participants, moreover, 3rd person nominative pronouns are null (except for the ''different subject'' plural pronoun), so it is word order that plays the major role in that part of the grammar. Quesada finds the following possible transitive constructions, certain non-indicative constructions left aside (here O refers to Undergoer, A to Actor, S to intransitive subject, -a and -s to personal suffixes referring to Actor and intransitive subject respectively, V to verb, while Nom and Obl stand for nominative and oblique forms of pronouns): 1. O(Nom) V-a, or (very infrequently) O(Obl) V-a 2. A(Nom) O(Obl) V In addition, there is an inverse construction: 3. O(Nom) V A(Nom), or O(Obl) V A(Nom), where the verb receives special aspectual marking and the Actor is usually (but not always) followed by the obviative marker. The intransitive clause looks simply as: 4. S(Nom) V, or (very infrequently) V-s
The system is straightforward, since it easily allows contrasting between Actors and Undergoers. Yet it turns out to be difficult to provide it with any appropriate typological characterization. Quesada spends much of his text in choosing between an ergative analysis and an accusative analysis. He argues in favor of the latter, but using a negative criterion: ''the members of the oblique paradigm cannot be used to code A nor S, only O'' (p. 109). This decision is not fully convincing, however. First, it is not at all apparent that negative criteria can be used for assigning a language accusative or ergative status (although they certainly can be counted as a matter of support of some analysis). Second, while following his argumentation, Quesada has to recruit the construction with agreement and oblique Undergoer, which he himself seemingly considers marginal. Third, the author apparently ignores the oblique origin of personal suffixes. Curiously, later Quesada attempts to show that ''the use of pronoun paradigms in Teribe has been subject to discourse grammar rather than to strictly sentence grammar'' (p. 118), so presumably the choice between the nominative and oblique form should not be involved in the syntactic characterization of the language.
Note that the view that the first two constructions only (without the ''marginal'' variant of the ''agreement'' pattern) constitute looks very similar to ''symmetrical voice'' systems, where transitive sentences have two (or more) structurally parallel variants, which only differ in whether the subject is Actor or Undergoer (see Foley 1998 among many others). With all this going on, neither Actor nor Undergoer are demoted to become obliques (as is the case in passive constructions in accusative languages and antipassive in ergative ones). All this applies to Teribe. In neither of the Teribe basic transitive constructions are there traces of elimination of Actor or Undergoer from the core of the clause. Furthermore, given that personal suffixes serve here as oblique pronouns, the two patterns form a symmetrical opposition: Actor Undergoer O V-a Obl Nom A O V Nom Obl Thus, Teribe at first glance seems to show a ''symmetrical voice'' system, and its main distinction from other similar languages (e.g., most Austronesian languages of Philippines and Taiwan) is that different alignments are reflected here not by voice morphology but by word order. For such systems, however, it does not make sense to speak whether they are ergative or accusative, since they are neither one.
The situation becomes more complicated when we take into account the inverse construction. However, this construction is an innovation, which still shows a number of irregularities and typological peculiarities. Quesada argues convincingly for the inverse analysis of this pattern, the appearance of which is motivated usually by the non- canonical alignment of two of the so-called topicality hierarchies, namely ACTOR > UNDERGOER and SPEECH ACT PARTICIPANT > PROXIMATE > OBVIATIVE. But many of the properties of the inverse construction are still unusual. Thus, for instance, the obviative Actor is nonetheless often marked with a topic particle, despite the fact that cross-linguistically inverse constructions prefer non-topic subjects. Further, case alignment in the inverse construction does not seem to be well-established (in fact, Undergoer - if pronominal - is expressed here either by Nominative form or by Oblique form, judging from examples given, depending on person). Recall now that the ''agreement'' pattern is restricted to pronominal Actors, so the inverse pattern complements it in that it allows non-pronominal actors to appear in a similar construction. Support for this comes from (1), where the ''agreement construction'' and the inverse construction merge. (This example is given in the section devoted to aspect and represents a pattern not mentioned by Quesada.) (1) wua-kz-a äya li dë ga shotwa-kz-a eat-SUD-3 devil TOP OBV CONN vomit-SUD-3 The devil ate [him] and vomited [him] at once. (p. 74) (CONN - linker, OBV - obviative, SUD - sudden aspect, TOP - topic particle, 3 - 3rd person Actor)
We thus get a clear picture of the distribution of constructions which is based neither on ergative nor on accusative scheme. There does seem to be a grain of truth, however, in Quesada's claim (p. 119) that there is ''a transition currently in progress in the language''. This follows from the occasional appearance of oblique Undergoers in the ''agreement construction'' and the inverse construction (note that here the two patterns again go together) as well as from the fact that personal suffixes now turn from bound argumental expressions into real agreement markers as this is evidenced by the occasional use of these suffixes in intransitive sentences (p. 84) and their appearance together with free actor noun phrases (1). Clearly, the direction of this reanalysis is towards the accusative type. This too resembles many Austronesian languages, which originally had ''symmetrical voice'' systems but later obtained features of the accusative type (although perhaps this reflects an even stronger typological tendency; cf. Maslova & Nikitina 2004).
While the Teribe system of identification of grammatical relations may look peculiar, some other features of the language are more typical - for the Chibchan family or typologically. Thus, for instance, the particular role of posture (positional) and motion (movement) verbs is observed in some other Chibchan languages, such as Cuna (Adelaar & Muysken 2004: 64-65). But Quesada's presentation actually highlights correlations between their grammatical properties and their place within the grammar. As has been said already, Teribe posture and motion verbs turn out to be among ''less verbal verbs'', and this is possibly somehow related to the fact that typologically these verbs are very inclined to grammaticalization (cf. Maisak 2005). In Teribe (as seemingly in Cuna) posture verbs often appear with their more prototypical confreres constituting what Quesada calls ''a verbal category of POSITION''. In addition, both posture and motion verbs are easily involved into serial chains (this allows the latter occasionally not to be marked for aspect, arguably the main verbal category in Teribe), which constitutes one of the canonical contexts of grammaticalization. That grammaticalization sources should be non-typical (in respect to other members of their category) seems to be an interesting implication, after all.
Another feature of Teribe that is representative of Chibchan languages is the existence of numeral classifiers. What is interesting about Teribe classifiers, however, is that they tend to ''float'' breaking away from the object counted: (2) Shwong ko plublun i-no-r k-ara dress color white see-PERF-1SG CL-one I saw one white dress. (p. 49) (CL - classifier, PERF - perfective, 1SG - 1st singular Actor)
In fact, examples of adnominal classifier-numeral complexes provided in the grammar either are given without context or are included into possessor noun phrases (which blocks their float). It has been argued elsewhere (Bach et al., eds, 1995) that A(dverbial)-quantification has bigger potential than D(eterminer)-quantification, so Teribe probably helps substantiate this claim.
There are other interesting facts that can be found in the monograph. Thus, for instance, Teribe displays very fuzzy borderlines between coordination and subordination (employing for most kinds of clause linking a single connector), instances of rather interesting polyfunctionality, sometimes typologically natural (as with the use of a single particle of adverbial origin for topics and relatives) but occasionally surprising (cf. the use of a demonstrative for marking CLAUSE boundaries), and even indefinite readings of personal pronouns (p. 97) etc. It is a pity that many of these facts are described in sections that are not devoted primarily to the phenomena in question. But it is, of course, a merit of the author that he managed to provide a detailed portrait of the language despite all these ''obstacles''. To Quesada's credit is also the fact that so much of the book is concerned with information structure, which is not usual for grammars.
As for the analyses proposed in this grammar, although they are not always fully convincing, it is important that for the most part Quesada provides enough data for readers to be able to make decisions themselves. This is not to say that this grammar is comprehensive - yet it is still detailed enough and may obtain wide employment for linguists and hopefully for the Teribe people - at least because the Teribes can be proud of this grammar, as can its author.
Adelaar, Willem F. H., with Pieter C. Muysken. 2004. The Languages of the Andes. Cambridge: Cambridge University Press.
Bach, Emmon, Eloise Jelinek, Angelika Kratzer, and Barbara H. Partee (eds). 1995. Quantification in Natural Languages. Dordrecht: Kluwer.
Foley, William A. 1998. Symmetrical voice systems and precategoriality in Philippine languages. Paper presented at the Workshop on Voice and Grammatical Functions in Austronesian Languages, LFG98 Conference, Brisbane, July 1998. [http://www.sultry.arts.usyd.edu.au/LFG98/austro/foley/fintro.htm]
Maisak, Timur A. 2005. Tipologija grammatikalizacii konstrukcij s glagolami dvizhenija i glagolami pozicii. [Grammaticalization paths of motion and posture verbs: a typology.] Moscow: Jazuki skavjanskix kul'tur.
Maslova, Elena and Tatiana Nikitina. 2004. Case marking patterns in diachronic perspective. Paper presented at the conference ''Syntax of the World's Languages 1,'' Leipzig, August 2004. [http://email.eva.mpg.de/~cschmidt/SWL1/handouts/Maslova1.pdf]
Oakes, Perry J. 2001. A description of Teribe phonology. SIL Electronic Working Papers 2001-003. [http://www.sil.org/silewp/2001/003/]
ABOUT THE REVIEWER
Yury Lander is a research fellow in the Institute of Oriental Studies, Moscow. He specializes in Austronesian and Caucasian languages and in the morphosyntactic typology of noun phrases and clauses. His current interests include the issues of polysynthesis and typological databases.