Review of Ergativity, Valency and Voice
|EDITORS: Authier, Gilles and Haude, Katharina
SERIES TITLE: Empirical Approaches to Language Typology 48
PUBLISHER: De Gruyter Mouton
Besides the editors’ introduction, this volume contains eleven studies on languages with predominantly ergative features, with a precise focus on voice alternations and transitivity phenomena found in these languages. These articles are based on oral presentations given at the monthly seminar, “Ergativité: typologie, diachronie et cognition” (Villejuif -- Paris, 2005-2009), organised by Francesc Queixalós. As is well-known, ergative languages are very different, but despite this fact, the volume has an obvious guiding line; all the contributors are fieldwork linguists, and all the data presented here are first-hand data from more or less known ergative languages.
The editors’ “Introduction” (pp. 1-14) contains a short presentation of ergativity and of specific terminology relevant to the volume, mainly based on Dixon’s work (1972, 1994). The notions defined in this section are well known from literature on ergativity and include: morphological ergativity, syntactic ergativity, pivots, alignment splits such as pronominal and aspectual splits, etc. Voice alternations (the key notion of the book) “determine the number, formal encoding, and semantic role of verbal argument(s)”, “serve to describe an event from different perspectives, and to retain the same participant as the central argument through larger stretches of discourse”, and “ideally form a productive system” (p. 5). The editors define several voice alternation mechanisms described in this volume: voice-decreasing devices (e.g. passive, antipassive, middles, anticausatives, noun incorporation); devices that maintain the same number of arguments (e.g. symmetrical voice, inverse systems, lability and lexical alternations, and a related phenomenon, namely differential object marking); and voice-increasing devices (e.g. causatives, benefactives or applicatives). A short outline of each article is provided at the end of the introduction.
The first two chapters deal with Mayan languages. “Ergativity and voice in Mayan languages: a functional-typological approach” (pp. 15-49), by Colette Grinevald and Marc Peake, starts with a brief presentation of the Mayan family. Section 2 deals with the multiplicity of verbal markers encoding transitivity (i.e. Pan-Mayan characteristics), and then presents data from specific Mayan languages (i.e. Jakaltek Popti’, Tojol Ab’al). Section 3 summarises the specific features of ergative marking in Mayan languages, taking into account two different terminologies: the “primitives” A/P/S; and the person markers of ergativity, “set A” and “set B”. Finally, in Section 4, the authors highlight the role of markers in the identification of voice systems (e.g. active-transitive, passive, antipassive, agent-focus, and applicative). Their conclusion is that ergativity is a major Pan-Mayan trait, and that Mayan patterns of verbal ergative alignment (including the voice system) are typologically relatively rare.
In the chapter “Ergativity and the passive in three Mayan languages” (pp. 51-110), Valentina Vapnarsky, Cédric Becquey, and Aurore Monod Becquelin offer a comparative analysis of the passive in Yucatec, Ch’orti’, and Tseltal. The extended presentation of the main characteristics of these languages and of their features related to ergativity and voice ends with some generalising conclusions: transitivity is a very important feature in all Mayan languages, where the authors identify many transitivising and intransitivising derivations, with reflexes in phonology, morphology and syntax; the use of the passive is motivated by discursive, semantic and discourse-pragmatic factors, rather than by syntactic ones. Consequently, the passive in Mayan languages is not strictly related to ergative or accusative features.
In the chapter “A tale of two passives in Cavineña” (pp. 111-131), Antoine Guillaume offers a detailed analysis of two verbal suffixes with passive value (-tana and -ta) in the above-mentioned ergative language from the Tacanan family spoken in Amazonian Lowland Bolivia. The article contains a brief presentation of the argument-coding system in this language, an analysis of the two passive derivations, and a diachronic account of the emergence of the two different suffixes. The underlying idea is that, despite many claims found in the literature, the passive is rather common in Amazonian languages and in ergative languages in general.
Three other chapters tackle Caucasian languages. Gilles Authier’s article, “The detransitive voice in Kryz” (pp. 133-163), deals with an unwritten ergative language belonging to the Lezgic branch of the North-East Caucasian family. This language is special among East-Caucasian languages because it has a detransitive voice with a prominent passive reading, the use of which is restricted by semantic parameters and lexical properties of verbs. The existence of a passive structure in this language seems to be motivated by modal and aspectual parameters, not by syntactic features (such as an accusative pivot), and probably appeared quite recently, under the influence of Azeri. The development of the passive was probably favoured by the existence of other detransitive voices with comparable morphology in nearly all branches of the East-Caucasian family.
In “Laz middle voice” (pp. 165-197), René Lacroix analyses the morpheme i- in Laz, a South Caucasian language. This morpheme present in various syntactic contexts with Class A and Class B middle verbs (e.g. Subject-Object coreference construction, Subject-Dative coreference construction, object possession construction, antipassive, perfective aspectual constructions, lexicalised items, passive, impersonal middle, anticausative, etc.) corresponds to what has been called ‘middle voice’ with reference to other languages and, as shown towards the end of the chapter, even if there is a historical relation between the middle and the applicative i-, these two markers should be kept distinct in a synchronic description.
In “Ergativity in the Adyghe system of valency-changing derivations” (pp. 323-353), Alexander Letuchiy questions the ergative nature of the West Caucasian language, Adyghe, by analysing transitivity increase mechanisms (i.e. causative, benefactive, malefactive, and locative), and transitivity decrease mechanisms (i.e. potential, antipassive, facilitive, and difacilitive). The conclusion of this chapter is that, despite some important differences with respect to the prototypical situations found in other syntactically ergative languages (e.g. in Adyghe, derivations can change the status of any participant, except for the agent/transitive subject), Adyghe can be considered a syntactically ergative language.
Guillaume Jacques’s paper, “Argument demotion in Japhug Rgyalrong” (pp. 199-225), deals with a Sino-Tibetan (morphologically) ergative language spoken in China. One of the core features of the verbal system of this language is transitivity. Consequently, there are many transitivity-changing devices, such as generic, antipassive, lability and incorporation (used for the demotion of patients), and generic and antipassive (used for the demotion of agents). Other mechanisms, such as the de-experiencer prefix, are used to derive an intransitive verb from a transitive verb of perception, labile verbs, and incorporation.
In “The Katukina-Kanamari antipassive” (pp. 227--258), Francesc Queixalós investigates the antipassive in the above-mentioned language from Amazonia, which seems to be the only surviving language of the small Katukina family. After reviewing some basic patterns of this language (e.g. ergative alignment, word order and constituency, movement, elision, ostention/modification or replacement by a demonstrative, coordination, focalisation, constituent questions, relativisation, nominalisation, control, subject and object), the author gives prominence to the antipassive device, which seems to have mainly formal motivations (e.g. allowing the agent to participate in movement, ostension, coordination, focalisation, relativisation, nominalisation), alongside some functional motivations, which are harder to detect (e.g. the pragmatic promotion or demotion of the agent or the patient, indefiniteness, etc.).
In the chapter “Undergoer orientation in Movima”, Katharina Haude analyses the system of verbal morphemes in an unclassified language from Amazonian Bolivia, in which most of the transitive clauses (the direct ones) display an ergative pattern (i.e. are undergoer oriented), while inverse constructions exhibit an accusative pattern (i.e. are actor oriented); in the intransitive domain, unaccusative verbs are generally oriented towards the undergoer, whereas unergative verbs are oriented towards the actor.
Aurore Monod Becquelin and Cédric Becquey’s article, “Case patterns and verb classes in Trumai” (pp. 289-322), deals with the Trumai language, which belongs to the Upper Xingu group from Mato Grosso, Brazil. The authors question previous analyses put forth for this language, focusing especially on the claim that it is an ergative language (Dixon 1994), and, by using corpus data, demonstrate that ergative verbs are not dominant in this language -- a fact which is considered important by the authors in establishing the ergative nature of a language. The data show that this language does not display any predominant alignment in the lexicon. Trumai patterns with Austronesian languages in that there are two sets of transitive verbs, agent-oriented (i.e. “extended intransitive” in Dixon’s terminology), and patient-oriented verbs (i.e. “ergative”). However, what is special about Trumai is that the ergative/accusative split is lexically governed for highly transitive verbs, and this split never involves morphological marking on the verb.
The final chapter is “The evolution of transitive verbs in Basque and emergence of dative-marked patients” (pp. 355-379), by Céline Mounole. The author shows that differential object marking (precisely, dative-marked patients), unusual in ergative languages, was first attested in the 16th century, but fully developed in the 19th century as a consequence of language contact with Spanish, and depends on factors like animacy and referentiality. In contrast with the other articles in this book, Mounole’s article does not explore a valency-changing device or a voice mechanism, but rather the way in which the spread of the dative-marked patient affects the canonical transitive structure in Basque.
This book, edited by Gilles Authier and Katharina Haude, includes a large amount of first-hand data from different ergative languages. Alongside these very interesting data, the editors and the authors offer a complete and very interesting picture of the relation between ergativity and voice alternations, although the limits between syntactic and lexical valency alternations are sometimes quite squishy. The most common voice alternation mechanisms are clearly defined in the introduction of the book and the more restricted ones are defined in the studies that refer to the respective mechanisms. Most of the chapters contain a brief presentation of the language under scrutiny, the mechanisms of valency change present in the respective language, and discuss the (typological) relevance of the existence of these mechanisms.
Besides the large amount of data and the systematic presentation of the voice/valency-changing devices, the strongest point of the book is that it re-evaluates some of the well-known assumptions on ergative languages; some of the analyses included in the volume – and the most interesting for typology and general linguistics – enable one to re-think the linguistic typology of the languages under discussion in some situations (e.g. Monod-Becquelin and Becquey demonstrate that Trumai is not an ergative language, Letuchiy shows that Adyghe is not only morphologically ergative, but also syntactically ergative) or to revisit certain typological generalisations, which are proven not to work for many of the languages described in this book (e.g. the traditional assumption that passivisation is uncommon in ergative languages is shown not to hold – see Grinevald and Peake’s demonstration for Mayan languages).
Even if the terminology is explained in most of the cases, sometimes it is difficult to follow the different systems adopted by the authors. For example, for referring to the transitive subject, transitive object, and intransitive subject, respectively, some of the authors use Comrie’s (1978) A/P/S terminology, others adopt Dixon’s (1987, 1994) A/S/O system, while others prefer Creissels’s (2006) A/P/U notation.
In conclusion, the book reviewed is essential reading for everyone interested in ergativity, voice alternations and valency-changing mechanisms. This book targets a very large audience: it is of interest not only to researchers working on ergativity, but also to undergraduates, who can learn what ergativity is, how it can be related to other phenomena found in different languages, and how one can work with understudied languages that require not only the interpretation of raw linguistic material, but also an accurate description of it.
Comrie, Bernard. 1978. Ergativity. In Syntactic Typology: Studies in the Phenomenology of Language, Winfred P. Lehmann (ed.). 329-394. Hassocks, Sussex : Harvester Press.
Creissels, Denis. 2006. Syntaxe générale. Une introduction typologique. Paris: Hermès-Lavoisier.
Dixon, R.M.W. 1972. The Dyirbal Language of North Queensland. Cambridge: Cambridge University Press.
Dixon, R.M.W. (ed.) 1987. Studies in Ergativity (Lingua, 71), Amsterdam: North Holland.
Dixon, R.M.W. 1994. Ergativity. Cambridge: Cambridge University Press.
| ABOUT THE REVIEWER:
Adina Dragomirescu is a Researcher at “Iorgu Iordan -- Al. Rosetti” Institute of Linguistics of the Romanian Academy, Department of Grammar, and Teaching Assistant at the University of Bucharest, Faculty of Letters, where she teaches Romanian syntax, morphology, phonology and stylistics and Romance syntax. In 2009, she defended her PhD dissertation, “Ergativity, typology, syntax, semantics”, which was published in 2010 by Bucharest University Press. She is the co-author of 5 other books and has published around 75 articles and book reviews. Her current domains of inquiry are Romanian supine and motion verbs.