Liesbeth Degand

The impact of relational markers on expository text comprehension in L1 and L2

This article reports on an experimentinvestigating the impact of causal discoursemarkers (connectives and signaling phrases) onthe comprehension of expository texts in L1 andL2. Although several psycholinguistic studieshave investigated the impact of connectives andlexical markers of text structure oncomprehension (i.e. off-line), there is noconsensus on the exact effect of explicitdiscourse markers on text understanding; threedifferent findings are reported in theliterature: markers would have a facilitatingeffect, an interfering effect or no effect atall. The first goal of this article is toclarify this problem of contradicting resultsby limiting the scope of the study to causalrelations, and to one specific text type:expository texts. Furthermore, the naturalnessof the experimental texts was controlled,readers did not need specific backgroundknowledge to understand the texts and theexperimental method consisted of open answerquestioning. Our second goal is to investigateto what extent a supposed effect of linguisticmarking depends on readers proficiency in afirst or second language.The experiment consisted in the reading of short expository texts in two languages, Dutchand French, which both functioned as L1 and L2.The results indicate that readers benefit fromthe presence of causal relational markers bothin L1 and in L2. Implications for (theoriesof) text processing are discussed, as well asfor the further insights in readingcomprehension in L1 and L2.

Coding coherence relations: Reliability and validity

Abstract This paper tackles the issue of the validity and reliability of coding discourse phenomena in corpus-based analyses. On the basis of a sample analysis of coherence relation annotation that resulted in a poor kappa score, we describe the problem and put it into the context of recent literature from the field of computational linguistics on required intercoder agreement. We describe our view on the consequences of the current state of the art and suggest three routes to follow in the coding of coherence relations: double coding (including discussion of disagreements and explicitation of the coding decisions), single coding (including the risk of coder bias, and a lack of generalizability), and enriched kappa statistics (including observed and specific agreement, and a discussion of the (possible reasons for) disagreement). We end with a plea for complimentary techniques for testing the robustness of our data with the help of automatic (text mining) techniques.

Introduction: Grammaticalization and (inter)subjectification of discourse markers

In spite of the by now vast literature on the rise of discourse markers and the impressive amount of empirical information available on the history of individual markers in different languages, linguists still do not agree on the applicability of the term grammaticalization to the processes which have led to the emergence of the words which are generally labeled discourse markers, discourse particles or pragmatic markers. While there is consensus that the concept of (inter)subjectification does indeed characterize the changes that such words undergo in their historical development (cf. Traugott and Dasher 2002, Athanasiadou et al. 2006), it is less obvious and less generally accepted that we can always speak of grammaticalization. Some authors claim that “pragmaticalization” is a more applicable concept (Erman and Kotsinas 1993, Dostie 2004). Obviously, the crux of the matter is how one defines the terms and which criteria one considers relevant, essential or optional as defining features. Although these issues have already been discussed by linguists central to the field, notably Brinton and Traugott (2005), it appears that disagreement and confusion persist in the field, about the best terminology, about which words to include in the class of discourse markers, as well as about the best possible theoretical account for their development and functioning (see e.g., the contributions in Fischer [2006]). In this issue we wanted to have a clear focus. We decided to bring together contributions by scholars who have all written extensively on discourse markers in different languages (the contributions deal with English, French, German and Dutch), and to collect their views and reflections on the issue of grammaticalization from their empirical investigations and their theoretical angles. The contributors to this volume were therefore asked to address at least one of the following questions in their articles:

Toward Automatic Determination of the Semantics of Connectives in Large Newspaper Corpora.

We explored the possibility of using automatic techniques to analyze the use of backward causal connectives in large Dutch newspaper corpora. With the help of 2 techniques, Latent Semantic Analysis and Thematic Text Analysis, the contexts of more than 14,000 connectives were studied. The method of analysis is described. We found that differences that have been suggested in the literature via hand-based analyses between these types of connectives (e.g., on dimensions such as subjectivity, change in perspective, and factuality of the connected segments) also appear in our corpus of 16.5 million words.

Mapping prosody and syntax as discourse strategies: How Basic Discourse Units vary across genres

The aim of his contribution is to explore the identification of different types of basic discourse (BDUs), and the role they might fulfil in discourse production and interpretation. A sampled corpus of four discourse genres (political adress, radio news, conference talk and conversational narration) was annoted both for syntactic units and for prosodic units. The originality of this contribution lies in the mapping of syntatic dependency clauses and major prosodic units giving rise to four types of BDUs : congruent (one-to-one mapping), syntax-bound (one dependency clause cut off into several major prosodic units), intonation-bound (one major prosodic unit enclosing several dependency clauses) and regulatory (one discourse marker or adjunct with prosodic autonomy). Our corpus analysis revealed that prosodic and syntatic units combine in different ways depending on the genre at stake. We propose that each type of BDU represents a discourse strategy. We suggest this segmentation method shouls be used in the frame of discourse models, in order to provide the researcher with established basic units.

Causal connectives or causal prepositions? Discursive constraints

In this article, we draw a comparison between causal prepositions and causal connectives and present them as alternative realizations of the underlying causal situation. It is our aim to investigate under which constraints a language user tends to select either of both causal alternatives. It appeared from a quantitative corpus analysis that these constraints are primarily pragmatic in nature, since they have to do in the first place with the discourse domain and with the management of given/new information. This is also confirmed by an analysis of the grammatical and lexical constraints on causal prepositions and connectives

Towards Automatic Retrieval of Idioms in French Newspaper Corpora

The goal of this paper is to present a procedure for the automatic retrieval of idiomatic expressions from large text corpora. The procedure combines text segmentation techniques and Latent semantic analysis. Three indices were computed on the basis of the three-fold hypothesis that: (1) idiomatic expressions should have few neighbours; (2) idiomatic expressions should demonstrate low semantic proximity between the words composing them; (3) idiomatic expressions should demonstrate low semantic proximity between the expression and the preceding and subsequent segments. The result of this procedure shows that we have not yet reached a fully automatic retrieval of idioms from large corpora, but this first trial has shown that we are on the way. The procedure reduces the amount of data to consider to less than a quarter (23.8 per cent) of the original data, of which one-fifth (20.9 per cent) is idiomatic, and nearly 60 per cent (58.8 per cent) is phraseological in nature. In other words, this procedure drastically improves and facilitates hand-based retrieval. In addition, these first results already permit some linguistic exploitation of the retrieved idioms.

Advanced learners' comprehension of discourse connectives: The role of L1 transfer across on-line and off-line tasks

Discourse connectives are important indicators of textual coherence, and mastering them is an essential part of acquiring a language. In this article, we compare advanced learners’ sensitivity to the meaning conveyed by connectives in an off-line grammaticality judgment task and an on-line reading experiment using eye-tracking. We also assess the influence of first language (L1) transfer by comparing learners’ comprehension of two non-native-like semantic uses of connectives in English, often produced by learners due to transfer from French and Dutch. Our results indicate that in an off-line task transfer is an important factor accounting for French- and Dutch-speaking learners’ non-native-like comprehension of connectives. During on-line processing, however, learners are as sensitive as native speakers to the meaning conveyed by connectives. These results raise intriguing questions regarding explicit vs. implicit knowledge in language learners.

Historical and comparative perspectives on subjectification: A corpus-based analysis of Dutch and French causal connectives

Abstract In this article, we focus on the diachronic development of causal connectives and investigate whether subjectification occurs. We present the results of ongoing and previous corpus-based analyses of the diachronic development of Dutch want and omdat, and French car and parce que, all four causal connectives roughly meaning ‘because’. In addition, we try to show that “grammaticalization studies can gain from the systematic and principled use of large computerized corpora and the methods which have been developed within corpus linguistics” (Lindquist and Mair, Corpus approaches to grammaticalization in English, John Benjamins, 2004: x). Thats why we have combined two historical and two comparative corpus methods to chart the diachronic development of these four causals. Our study reveals that subjectification is not an integral part of the diachronic development of these causals: subjectification does occur in the rise of these connectives, but in the later stages of their development only parce que undergoes subjectification. Our analyses show that the four methods all have their own merits and limitations, but they are most effective when combined.

Annotating the meaning of discourse connectives in multilingual corpora

Abstract Discourse connectives are lexical items indicating coherence relations between discourse segments. Even though many languages possess a whole range of connectives, important divergences exist cross-linguistically in the number of connectives that are used to express a given relation. For this reason, connectives are not easily paired with a univocal translation equivalent across languages. This paper is a first attempt to design a reliable method to annotate the meaning of discourse connectives cross-linguistically using corpus data. We present the methodological choices made to reach this aim and report three annotation experiments using the framework of the Penn Discourse Tree Bank.