SIL Electronic Book Reviews 2005-009
URL: http://www.sil.org/silebr/silebr2005-009
Grammatical relations: A functionalist perspective
Typological Studies in Language 35. Philadelphia: John Benjamins, 2001. Pp. 350. hardback $103.00, paper $35.95.
Reviewed by Anne Garber Kompaoré
Africa Inter-Mennonite Mission and United Bible Societies
I first want to preface by stating that this review is written from the perspective of a non-specialist of Functional Grammar. Therefore my remarks will reflect a certain lack of critique concerning specific hypotheses or analyses. Aside from the summary of its contents, I will especially comment on its accessibility to the ordinary working linguist, and (unfortunately) on the high number of typographical errors in certain articles.
This book is a collection of articles on grammatical relations from a functionalist perspective. It begins with an introductory article by the editor, followed by seven articles each written by a different author, each on a different language in diverse areas of the world. The differing topics and languages offer a very rich perspective on the differing grammatical structures of languages around the world. At the same time, it is quite a mental challenge to read a series of descriptions of diverse languages unfamiliar to the reader. It should also be noted that most articles assume a certain foundational knowledge of functional grammar and its terminology. There is an unequal treatment of defining of terms throughout the book: some authors define their terms clearly and are easy to follow; others assume familiarity, sending the uninitiated quickly to a reference book on functional grammar.
Givón provides an 85-page introduction to grammatical relations. This introduction is more like a summary, assuming prior knowledge of certain categories and terminology (for example, “Raising,” “Equi-NP,” etc.). He begins by stating that “this volume presents a functionalist perspective on grammatical relations without neglecting their structural correlates.” He treats the question of subjecthood and objecthood, opting for a prototype approach, outlining some formal properties of subjects and objects under the subtitles, “overt coding properties” and “behavior-and-control properties” in grammatical processes such as relativization, passivization, Equi-NP deletion, and reflexives. He emphasizes that languages tend to have a cluster of properties. He then compares the subject and object properties of languages worldwide to show that languages vary greatly in the manner that they mark subjects and objects, and to show how they behave in grammatical processes. Discussion of the notion of subject and object in ergative languages receives a separate subsection, as well as serial verb languages, and the topic of clause union.
Givón makes one mistake (to my knowledge) in his statement that Biblical Hebrew is a rigid VO language with flexible subject word order; in actuality, objects also have flexible word order, their position being governed by discourse or pragmatic considerations. One disappointing surprise was to uncover six typographical errors, including two missing words. 1
Each of the authors follows more or less the topics as presented by Givón. Sherri Brainard writes on “Ergativity and grammatical relations in Karao.” She makes fewer assumptions about prior knowledge and presents good explanations and definitions for each key term that she uses. She explores the question of whether the syntactic processes of relativization, passivization, etc. in Karao, an ergative language, are controlled mostly by the most agentive argument of the sentence; or whether the category of absolutive (found in ergative languages) controls syntactic processes; or whether there is a mixed pattern of control. The footnotes include good background information about the language.
Spike Gildea’s article on the “Evolution of grammatical relations in Cariban: How functional motivation precedes syntactic change,” was the most difficult for me to follow. He assumes complete familiarity with both functionalist terminology and the geographical area in which Cariban languages are found. If the reader has this knowledge this article is otherwise not difficult to follow. Gildea begins his article mentioning proto-Cariban in relation with Indo-Iranian languages. It is clear that Gildea assumed a readership knowledge of the Cariban languages. This should not have been done for a book of such wide language scope.
Using data from a number of Cariban languages, Gildea examines a verb form (referred to as t-V-se) which generally functions as a verb participle but also has other functions depending on the language: inverse voice, “pragmatically marked active ergative construction” (p.157), and ergative. He posits a verbal adjective function for the proto-Cariban verb form and then traces a possible evolution of the function of this verb form through various stages, pointing to data from specific synchronic languages as evidence. There is one typographical error (p.158, line 14).
In her article, “Grammaticalization, clause union and grammatical relations in Ecuadorian Highland Spanish,” Marleen Haboud examines how the verbal gerunds of Standard Spanish are used in an Ecuadorian dialect of Spanish influenced by the Quichua language. Comparing data from both dialects, she demonstrates how the grammatical function of the gerund has been re-analyzed in Eastern Highland Spanish, such that the gerund functions as the main verb, and the finite form, as an auxiliary verb (and is thus grammaticalized), allowing for further verbal collocations that would be either ungrammatical in Standard Spanish or else produce different meanings. Unfortunately there are at least six errors in this article, including missing words, spelling errors, clause-level errors, and misplaced words. 2Otherwise, Haboud’s article was clearly presented and easy to follow.
In his article, “The direct object in bi-transitive clauses in Indonesian,” Bambang Kaswanti Purwo examines the verbal suffixes -kan and -i in relation to the dative properties of ‘give’-like verbs, which are true bi-transitves, versus ‘buy’ -like verbs, which are mono-transitive. In a tantalizing manner the author invites the reader to try to sort the data out for himself before coming to a neat and simple conclusion. He proposes that “the zero or -i suffixes marks the norm direct object for verbs, while the -kan suffix marks the counter-norm direct object” (p.248). There are two typos, with missing plurals on pp. 233 and 243.
E. K. Osam examines “Serial verbs and grammatical relations in Akan,” a West African language. The major focus of his article is distinguishing the behavior of direct objects and indirect objects in bi-transtive and in serial verb constructions. Osam’s article gives a good presentation of facts and data. His discoveries could easily be compared with similar behavior in other West African languages. Unfortunately, it, too, is afflicted with typos, a total of at least five mistakes.3
In his article “Zero anaphora and grammatical relations in Mandarin,” Ming-Ming Pu focuses on the distribution of zero anaphora within a narrative text. He demonstrates a very simple analysis that seems to have been missed by others, based on cognitive factors, “discourse context, pragmatic and semantic information, and speaker/writer interaction” (p.282). He emphasizes the fact that texts are chunked into thematic units. In shifting from one thematic unit to the next there is more mental effort to access and process information, and thus the necessity to make referents more explicit. With the help of statistics for 75 pages of narrative text Pu shows that the distribution of zero anaphora is constrained to occur only where there is high local thematic continuity.
Pu’s article is well-written, supported by careful statistics, and would be of special interest to those working on discourse analysis. The last section, however, dramatically shifts theoretical approaches; in it Pu addresses Government and Binding analyses of the same phenomenon, concluding that a GB approach is inadequate for fully explaining zero anaphora in Chinese. Total familiarity with GB is assumed, leaving the non-initiated quite lost. There are at least four typographical errors.4
Noel Rude writes on “Dative shifting and double objects in Sahaptin,” a Pacific North American language. He first describes the morphological features necessary for distinguishing subjects and objects, and then covers all the tests in order to determine the behavior and control criteria for objecthood, before examining the problem of sentences which have two objects with identical case marking, as well as morphological causatives and object raising from nominalized complements and dative shifting. For these he applies several behavioral and control tests in order to ascertain which one is the true object. He concludes that in such cases the dative/agent is the direct object and not the patient.
Rude notes in the final sentence of his article that word order is not involved when the dative is promoted to direct object, since Sahaptin has flexible word order. In my opinion, it would have been profitable for him to mention the flexible word order early in his article, since his examples have so many variant word orders, making them difficult to follow. There are two typos;5 also, there is what appears to be an error in glossing (p. 331, examples (40) and (41), compare with (36)).
In conclusion, while this book is rich with analyses of grammatical relations from a wide variety of languages (it broadened my own knowledge), it would have been yet better if the editor could have ensured elimination of typographical errors, and better defining of terms in some of the articles. To best profit from this book, it is advisable to have some basic notions of functional grammar, and for the non-specialist to have a book on functional grammar on hand for reference.
Notes
Typographical errors: “by,” line 21 p. 34; “agent,” last line, p.39; incorrect words (“my” should be “by,” line 6, p. 29; “marged” should be merged,” p. 68; “is” should be “in,” line 6, p. 46.
Missing words: p. 205, line 10 “from”; p. 209, line 14, “an agent,” “PAT” should be spelled out as “patient”; spelling: p. 205, line 22, “-s”; p. 226, 3rd last line, “semantic-mail” should be “semantic main”; clause level: p. 206, line 19, “initiated by an agent” should be replaced by “from a point of reference”; words to delete: p. 222, line 17, the final “it.” There are also a few missing parentheses.
Three missing plurals on pp. 261, 267, 268, and on p. 275 there is a missing “and” as well as a missing t in grammaticalization.
Two missing plurals (pp. 309, 318), one missing –ed suffix (p. 313), and one wrongly capitalized word (p. 315).
One on p. 331 (verb conjugation) and one on p. 333 (missing quotes).