This book "supplies a vocabulary of English words and idiomatic phrases 'arranged … according to the ideas which they express'. The thesaurus, continually expanded and updated, has always remained in print, but this reissued first edition shows the impressive breadth of Roget's own knowledge and interests."
AUTHOR: Stassen, Leon TITLE: Intransitive Predication, paperback ed. SERIES: Oxford Studies in Typology and Linguistic Theory PUBLISHER: Oxford University Press YEAR: 2003
Nilson Gabas, Jr., MCT-Museu Goeldi (Brazil)
[The announcement to which this review is linked is that of the original hardback edition. --Eds.]
INTRODUCTION The book is a careful investigation of cross-linguistic variation in one of the core domains of all natural languages, intransitive predication.
Basing his analysis on a database sample of 410 languages, Leon Stassen (henceforth S) presents a universally applicable model for defining the domain of intransitive predication in natural languages. Intransitive predicates are defined as a 'cognitive space', regarded in terms of four domains: events (X cries), properties (X is happy), classes (X is a student), and locations (X is in the garden).
Taking these four domains in consideration, S offers a typology of the structural manifestations of predication in terms of the nature and number of the formal strategies used in its encoding (Part I of the book), together with a discussion of a number of abstract principles which can be used to explain the cross-linguistic variation embodied by the typology (Parts II and III).
In the final part, S brings together the research results in a universally applicable model, which can be read as a 'flow-chart' for the encoding of intransitive predications in different language types.
DESCRIPTION OF THE BOOK'S CONTENT (modified from pages 21-22) Based on the definition of the domain of investigation given in Chapter 1, the next three chapters (which together form Part One of the book) investigate the cross- linguistic encoding options for the four predicate categories in the domain. Chapters 2 and 3 deal with the criteria according to which various different types of predicate encoding can be identified across the languages of the sample. The result of this discussion is the establishment of a number of prototypical features of the encoding of different predicate categories, in the form of different ENCODING STRATEGIES. In Chapter 4 this result is applied to the database of the project, leading to a typology in which, for each sampled language, the encoding of the four predicate categories in the domain is specified in terms of sameness or difference in strategy choice. In this sense, the descriptive research result arrived at in Chapter 4 consists in the specification of the FORMAL STRATIFICATION of the domain for each language in the sample.
Part Two (Chapters 5-8) and Part Three (Chapters 9-13) deal with specific, residual, problems. A conspicuous feature of the encoding of the domain at issue is that, in very many languages, more than one formal stratification appears to be permitted. This phenomenon of SWITCHING are looked at from various angles in Part Two, and the principles which turn out to constrain the possibilities of switching are identified. The chapters in Part Three contain the exploration of a major typological distinction in intransitive predicate encoding -- namely, the difference between 'verby' and 'nouny' encoding of property-concept predicates. It is argued that this distinction can be explained in terms of a principle which S calls the Tensedness Parameter. The definition of this principle is given in Chapter 9.
Chapters 10 and 11 in Part Three involve a detailed investigation of the empirical validity of the Tensedness Parameter for all the languages of the sample. The general assessment is that this parameter provides a reliable basis for the prediction of 'verby' versus 'nouny' status of property-concept predicates across languages. S then concludes Part Three by offering a few speculations as to the nature of the relationship between tensedness and property predicate encoding.
The descriptive and explanatory results assembled in the first thirteen chapters are summarized in Chapter 14, which, on its own, constitutes Part Four of the book. This chapter presents a model of the encoding possibilities of the domain of intransitive predication across the languages of the world. S argues that this domain is, cognitively or conceptually speaking, the same in all languages, and that cross- linguistic variation in the formal stratification of the domain derives from the interaction of a limited number of universal principles of natural language, which are either formally or cognitively motivated.
CRITICAL EVALUATION The book is the result of a massive and thorough piece of work. S builds up his analysis based on empirical data and upon intense typological investigation, bringing the reader with an impressive study of the subject. His concept of STRATEGY (characterized as prototypical) is an important key to his work, which he takes as the basis for the postulation of the four strategies involved in intransitive predication. Each of these strategies would be motivated by a different factor.
According to S, the verbal strategy (the one associated with intransitive event predicates) would be identified by the application of the criteria of agreement, auxiliary and negation, and is motivated by considerations of semantic relevance and typological markedness (cf. Section 2.3).
Locational strategy (which prototypically encodes locational predicates) would be identified by the presence of a supportive, 'locative' item which has the morphosyntactic characteristics of a V, and has as its defining feature its specific semantic character. For S, Locative Predicates denote concrete location, and are therefore sensitive to considerations of locational specification (cf. Section 2.4).
Nominal strategy (the strategy prototypically associated with class predicates) would be identified by a nonverbal supportive item, usually a 'zero copula' or have the form of a lexical item which originates from nonverbal discourse-marking elements ('pro-copula', 'particle copula'). It could be defined as deriving from the universal encoding properties of identity statements (Sections 3.6 and 3.7).
The definition of the last strategy, property strategy, would follow a universal principle, called by S 'The Adjective Principle'. This principle basically states that predicate adjectives have no prototypical encoding of their own, and that they take either the verbal strategy, the nominal strategy, or a mixture of both, as their encoding strategy.
The most interesting part of the book comes from S's search for a principle that predicts the occurrence of one strategy (verbal) or another (nominal) as the property strategy of a given language. For him, languages which make use of the verbal strategy are called A- languages, and languages which make use of the nominal strategy are called B-languages. The distinction between them is based on a parameter which he calls 'Tensedness Parameter', stated as follows (p. 350-351):
(8) Definition of the Tensedness Parameter (a) If a language has a grammatical category of tense, which (1) is morphologically bound on verbs, and (2) minimally involves a distinction between past and nonpast time reference, then that language is tensed. (b) In all other cases, a language is non-tensed.
This Tensedness Parameter is employed in the construction of the following set of bi-directional universal statements (p.357):
(17) The Tensedness Universals of Adjective Encoding (a) If a language is tensed, it will have nouny predicate adjectives. If a language has nouny predicate adjectives, it will be tensed. (b) If a language is non-tensed, it will have verby predicate adjectives. If a language has verby predicate adjectives, it will be non-tensed.
The stipulation of the parameter as well as its outcome seem to hold very well against the range of languages under study, and makes the hypothesis and its corollaries justly reliable. Nevertheless, S's conclusions do not seem to be universal. At least one counterexample for the stipulation of the Tensedness Parameter and for the statement that no language has a specific encoding for the property strategy comes from Karo, a Brazilian language of the Tupí family.
Karo, according to S's parameter, is a non-tensed language (tense in Karo is optionally marked by means of different particles) that has different encodings for all four types of strategies. In Karo, encodings of the verbal strategy, the locational strategy, the nominal strategy, and the property strategy are found, respectively, in the following examples:
1a. owét o=wé-t 1SG-cry-IND 'I cried'
1b. wat owã yat ka'a 'a' pe' wat owã ya-t ka'a 'a' pe' 1SG.POSS mother stand-IND house CL.ROUND LOC 'My mother is in the house'
1c. wat iyõm agóa'pât nãn wat iyõm agóa'pât nã-n 1SG.POSS father shaman be-IND 'My father is a/the shaman'
1d. cúrem 'õn cú=tem 'õn big=ADVZ 1SG 'I am big'
Although the existence of languages such as Karo might disqualify S's Adjective Principle as universal, it does not place S's investigation into risk. It only (conceivably) entails additional revision(s) to it.
An additional - perhaps small - problem arises from the fact that further references about Brazilian Indian languages (the languages that I am most familiar with) are missing in the sample database -- S could have been slightly more careful in representing them. For instance, there are additional references of Macro-Jê languages available in the literature (Eduardo Ribeiro, p.c.), such as Wiesemann (1972, 1986), and Maia (1998) which, if added to S's database, could, in principle, give additional support for his ideas.
Although the lack of references holds up for other families of languages of Brazil as well, this fact does not seem to diminish the (vast) contribution S makes to the field of Linguistics through the publication of Intransitive Predication.
REFERENCES Gabas Jr., Nilson (1999) A Grammar of Karo, Tupí (Brazil). Ph.D. dissertation. Santa Barbara, University of California.
Maia, Marcus A. R. (1998) Aspectos Tipológicos da Língua Javaé. Munich: Lincom-Europa. 90p.
Wiesemann, Ursula (1972) Die phonologische und grammatische Struktur der Kaingáng-Sprache. Janua Linguarum, series practica, 90. The Hague: Mouton. 211 p.