Contragram

Newsletter 1

No 1
February - 1995

CONTENTS

Contrastive Linguistics in Belgium

Filip Devos

Since the sixties, contrastive linguistics has received a great deal of attention. In spite of the uncertain theoretical state of affairs, and in spite of the fact that contrastive analysis is not the only "true" method, it was claimed that contrastive research should be stimulated especially because of its value for teaching. Of central importance was the practical usefulness of the discipline for foreign language teaching. Consequently, the seventies and eighties were a very productive period, as is evidenced by the many projects, publications, journals and conferences often exclusively devoted to the subject. Research projects were to be found in, among others, Germany, Yugoslavia, Poland, Rumania, Sweden, Hungary and Bulgaria.

Belgium, too, has shared in this general growing interest, but to a lesser extent than the ideal multilingual fertile soil in this country might lead us to expect. Not only does Belgium have three official languages, but it is also an international forum for business and administration and, as a result of its geographical position, a country surrounded by quite different language communities and cultures. Moreover, the country can boast of a long tradition in foreign language teaching and learning.

Though Belgium should thus be an ideal place for contrastive linguistics to grow and flourish, the situation is remarkably less rich than one would expect. In their short Survey of contrastive linguistic analysis in Belgium (Leuven:Acco, 1979), Dirven and Afschrift point out that this is somehow a peculiar situation: "In spite of this intensive multilingualism, actually dominating the Belgian linguistic scene, the general picture of contrastive analysis in Belgium is less rich and diversified than [the] sociocultural background might lead us to expect, however. One of the reasons for the relative underdevelopment of con trastive analysis in Belgium may precisely be the great variety of areas, in which languages have to be contrasted and the lack of contact between researchers in the various areas" (1979:19-20). Contrastive studies are indeed rarely systematic and tend to concentrate on specific problems either from the point-of-view of language teaching and learning or from the point-of-view of translation. Many analyses are oriented towards the description of difficulties in French as they present themselves to the foreign learner. Van Roey, for instance, in his A contrastive description of English and Dutch noun phrases (Brussels:Aimav/Paris:Didier, 1974) argues that "although the importance of contrastive language analysis has repeatedly been emphasized of late years, it is probably fair to say that many contributions in this field have been of a rather sketchy nature" (1974:7). Equally illustrative of this point is the fact that Dirven and Afschrift (1979), the only bibliography of contrastive publications in Belgium so far, lists only about 170 titles, next to more than 960 unpublished dissertations written at higher institutes for interpreters and translators, especially on subjects dealing with translation, lexicography and terminology. It was also in the seventies that the "Contrastive Analysis Series" (CAS) was published in Leuven. It was a series of nine volumes containing articles on Dutch-French, Dutch-English or French-English topics. However, before the start of a contrastive grammar project at the University of Gent in 1988, there were no long-term research projects in Belgium.

 

[table of contents]

A Dutch - French - English contrastive grammar for foreign language teaching

1. Situating the project

Collaboration between the Departments of Dutch, French and English Linguistics of the University of Gent led to the start in 1988 of a research project on contrastive grammar for foreign language teaching. The project was set up by order of the Ministry of Education, and under the auspices of the Fund for Joint Fundamental Research (Belgium). The supervisors of the project were Prof. D. Willems (spokesperson), Prof. A.-M. Simon-Vandenbergen (co-supervisor) and Prof. J. Taeldeman (co- supervisor); the researchers were F. Devos (Dutch), R. De Muynck (French), M. Van Herreweghe and L. Martens (English). The full title of the project was: "Studie van de gemeenschappelijke en taalspecifieke grammaticale begrippen in het Nederlands, het Frans en het Engels voor het gebruik bij het onderwijs van die talen op de diverse niveaus in de diverse richtingen van het secundair onderwijs" ["Study of shared and language specific grammatical notions in Dutch, French and English. A grammar for foreign language teaching at the various levels of various types of secondary education"] (FKFO project no. 11.OD5187). The project ran from 1/10/1988 till 30/9/1991.

2. Aims

In general, the contrastive grammar project aimed at a more efficient organization of foreign language teaching at secondary school level and especially concentrated on the grammatical analysis of French and English, the two main foreign languages in Flanders. Thus the first objective was to stimulate cooperation and feedback between teachers of different languages, and more specifically, to provide materials for a multilingual organization of the teaching of Dutch, French and English grammar. This kind of multilingual approach is highly desirable: at the moment there is little cooperation and widely divergent and hence confusing systems of grammatical terminology are used. Obviously, such a fragmentation results in a lack of understanding on the part of the language learner and a waste of time and energy on the part of the language teacher, not to mention the desillusionment on both sides. A contrastive approach may help to eliminate these shortcomings.

3. Guidelines

The publication of a three-volume contrastive grammar is to provide language teachers with materials for an interdisciplinary approach. In order to achieve this aim the following guidelines have been and will further be followed in the course of the analysis: (1) setting up a system for making optimal use of the knowledge of the mother tongue (the language learner's competence in his native language) in teaching a foreign language; (2) bridging the gap between the different grammatical systems by starting from a common grammatical model with a simple and uniform terminology; (3) defining possible problem areas in foreign language teaching and learning processes, and giving an extensive description of the differences between the native and the foreign languages.

4. Tenets and criteria

A contrastive grammar should meet five criteria: (1) it should involve all levels of linguistic organization; (2) it should be bi- or multidirectional, which means that equal attention should be paid to all languages which are compared; (3) it should be nonselective, i.e. it should treat the similarities (the common rules) as well as the differences or contrasts (the language specific rules) between the languages; (4) it should be descriptive, and (5) it should contain a uniform terminology .

5. Contents

Our contrastive analysis focuses on three aspects: the sentence (sentence parts, functions and types), the noun phrase (construction and elements of the nominal group) and the verb phrase (tense and aspect of the verbal group). The results of the investigation are appearing in three volumes under the common title Nederlands, Frans en Engels in contrast (Dutch, French and English in contrast). So far, two 300-page volumes have appeared, one on the noun phrase and one on the sentence. The preparation and publication of the third and final volume in the grammar series, a contrastive analysis of the verb phrase, is one of the first objectives for future work.

The contrastive grammar project has clearly demonstrated the importance and usefulness of exhaustive lexical descriptions of verb constructions, in particular valency and verb complementation. Such a valency dictionary is the aim of a new project which we will discuss in the next section of this newsletter.

 

[table of contents]

A Dutch - French - English - German verb valency dictionary

Bart Defrancq

During the last 25 years verb and other kinds of valency seem to have featured among the most challenging topics for research in linguistics. As a result, extensive theoretical work was published the subject received a large variety of definitions, coming from all main linguistic disciplines: morphology, syntax, semantics, cognitive grammar, etc... However, the step from theoretical to practical work has on many occasions not been taken so that valency dictionaries are rather scarce.

The valency dictionary which is being compiled at the University of Gent in the framework of the G.O.A.-project (Geconcerteerde Onderzoeksactie - Concerted Research Action) Werkwoordconstructies in het Nederlands, het Frans, het Engels en het Duits. Een theoretisch model met practische toepassingen (Verb structures in Dutch, French, English and German. A model with practical applications) owes its originality to the fact that it will be the first contrastive valency dictionary to describe systematically and exhaustively the syntactic and semantic differences between verbs from different languages.

 In order to achieve this goal, the descriptive model we propose differs from that of the monolingual valency dictionaries in its contrastive approach of the lexicon, as a result of which the entries for each language are not only structured according to their syntactic characteristics but also according to their translations in the other languages. On the other hand, unlike what happens in translating dictionaries, the contrastive dictionary proposes a one-to-one comparison of entries. This means that the proposed translation of any entry in another language is maintained for the entire analysis. In case of non-equivalence of the proposed translation with the original entry, an asterisk points out the non-equivalence and reference is made to another entry. The example in Figure 1 from Dutch, French and English will illustrate these principles (the syntactic features are dropped to simplify the presentation):

 

Figure 1: Extract from the entry ZIEN-VOIR-SEE
ZIEN ____  VOIR ____  SEE ____
Hij zag zijn wagen niet meer.  Il ne voyait plus sa voiture  He didn't see his car any more.
ZIEN ____  VOIR ____  SEE* look out
Dit venster ziet op het park.  Cette fenêtre voit sur le parc.  *This window looks out on the park
ZIEN* begrijpen VOIR ____  SEE ____
*Ik begrijp wat u bedoelt.  Je vois ce que vous voulez dire. I see what you mean. 

 

The advantage of this description is that in addition to the information a normal dictionary would provide (translation and example) it provides also information about incorrect translations (entries with asterisk) and helps the foreign language learner to avoid mistakes. This is, in fact, one of the practictal applications of the theoretical model.

More essentially the description of the verbal lexicon will offer insights in the polysemy structures of the languages, which, on a practical level, should allow a rationalization of the lexicographic material in nuclear and peripherical meanings, and, on a theoretical level, should contribute to the study of language typology. In this respect one could set up a semantic classification of languages based on the presence of lexical items and their relationship with semantic features. French would then be one of those languages which lexically confuse "ability" and "permission". This is not only the case for pouvoir, but also for permettre (e.g. Ceci nous permet de travailler plus vite ("ability")). This statement applies also to English since can and allow have roughly the same features as pouvoir and permettre.

Dutch, however, separates these two fields by using two distinct lexical items for pouvoir-can: mogen ("permission") and kunnen ("ability"), as well as for allow-permettre: toelaten ("permission") and the compound verb in staat stellen ("ability").

This specialization in Dutch could be the trigger for other differences between the languages. It is for instance interesting to see that whereas Dutch displays a possible extension of mogen which refers to the person who grants the permission, French and English fall back on another lexical entry (Figure 2).

We might conclude that in French and English "ability" and "permission" are confused unless the person giving the permission is explicitly mentioned. Dutch distinguishes already on the first level and consequently doesn't need to distinguish on the second one.

This is of course only one rather obvious example of how semantic features organise the lexicon. A real semantic typology will need a much more detailed description of the lexicon, a task which the Research Group proposes to undertake.

 

Figure 2: Extract from the entry MOGEN-POUVOIR-CAN
MOGEN ____ 
Infinitive 		 POUVOIR  ____ 
Infinitive 		 CAN ____ 
Infinitive 
Mag ik u een vraag stellen?  Puis-je vous poser une question?  Can I ask you a question?
MOGEN ____ 
van NP 
Infinitive 		 POUVOIR* permettre  CAN*  allow
Ik mag van mijn vader autorijden. *Mon père me permet de conduire.  *My father allows me to drive. 

[table of contents]

 

Book notice

Beth Levin (1993)
English Verb Classes and Alternations: A Preliminary Investigation Chicago: The University of Chicago Press. xviii + 348 pp.
(ISBN 0 226 47533 6)

Dirk Noël

This "set of resource materials on the English verb lexicon" (p. xvii) is a "snapshot of ongoing research" (p. 18) on the relationship between the syntax and the semantics of English verbs. The roots of the research lie in the Lexicon Project of the MIT Center for Cognitive Science, devoted to the study of lexical organization and representation, which explains the use in the introduction of terms like lexical knowledge and ideal lexicon, "a lexicon that contains the minimum information necessary" (p. 1). If the syntactic behavior of verbs is semantically determined, which is Levin's starting assumption, then verb entries do not have to contain information on argument frames and diathesis alternations since these can be derived from the meaning of the verbs. Levin's aim is not to offer explicit answers to questions about the extent of the contingency relationship, the identification of relevant components of meaning, the necessary kind of lexical representation of verb meaning, and the formulation of mapping rules that determine the syntactic properties, but simply to facilitate future study of these questions.To this end she presents two lists: a list of diathesis alternations that she claims are relevant to a speaker's lexical knowledge of English (Part I), and a list of verb classes that are "semantically coherent" (p. 17) and (therefore) pattern in the same way with respect to diathesis alternations and other properties (Part II). Both lists overlap to a certain extent because they present the same material from different angles.

In Part I Levin lists three major groups of alternations - transitivity alternations, alternate expressions of arguments within the VP that do not affect transitivity, and alternations with "oblique" subjects - and five smaller groups. Each alternation is exemplified and described very briefly; often examples are given of verbs that contrary to expectation do not allow a particular alternation. To try and link up the syntax with the semantics the semantic classes of verbs that do and do not display the alternation are identified as much as possible. There are references to relevant discussions in the literature and comments on noteworthy features of the alternations.

Part II contains a host of "syntactically relevant, semantically coherent" (p. 22) verb classes of English, grouped in 48 categories like "Verbs of Putting", "Verbs of Removing", "Verbs of Sending and Carrying", "Verbs of Exerting Force: Push/Pull Verbs", etc. Many of these are subcategorized further, e.g. the first group is split up in "Put Verbs", "Verbs of Putting in a Spatial Configuration", "Funnel Verbs", "Verbs of Putting with a Specified Direction", "Pour Verbs", "Coil Verbs", "Spray/Load Verbs", "Butter Verbs", and "Pocket Verbs". For each group references to the literature are supplied where possible, class members are listed, if possible exhaustively, their alternation properties are exemplified using a representative verb, and in a comments section special properties of each class are signaled.

It does not have to be one's ambition to model minimal entries for a mental lexicon to be interested in this book. It can be looked upon as a catalogue of hypotheses on the relation between form and function and cannot be ignored by anyone who wants to investigate functional motivations for the formal behaviour of verbs.

 

[table of contents]

To the table of contents of other CONTRAGRAM issues