Date & time: 23 November, 13:00-14:30

Location: Utrecht University, Kromme Nieuwegracht 80, Stijlkamer van Ravesteyn 1.06

Maite Taboada

Simon Fraser University (Vancouver, Canada)

Coherence and cohesion in multimodal documents

In this talk I present preliminary results of an ongoing project on the discourse characteristics of multimodal documents. This is joint work with Christopher Habel, carried out under an Alexander von Humboldt Foundation Fellowship at the University of Hamburg.

A great deal of work in the last few years has focused on the relationships between text and material presented through other modalities, be it visual, audio, or a combination of the two. Research on document design and learning has been trying to elucidate what kind of impact multimodal material has on the reader. Much research has studied whether to use multimodal material or not, where to place it, and what effect captions or other verbal information surrounding such material have on the reader (e.g., Acartürk et al., 2008; Mayer, 2009).

Less frequently discussed is the nature of the relationship between graphical material and the text itself. The point of departure for this work is that multimodal documents, just like any other form of discourse, exhibit coherence and cohesion relations. In particular, we are examining coherence relations between text and graphical material (pictures, diagrams, figures and tables), and cohesive ties that establish cross‐reference between the two modes. In order to understand and categorize the types of relations between figures and text, we are making use of Rhetorical Structure Theory (Mann and Thompson, 1988), and discuss whether RST relations are sufficient for describing these types of relations.

Ours is a corpus study. Because we believe genre constraints the types of figures present and the way they are introduced, we study three different genres: newspaper articles (New York Times), magazine articles in a scientific magazine (Communications of the Association for Computing Machinery), and scientific articles (Journal of Computational Linguistics). The corpus consists of about 1,500 pages of material, containing over 700 figures, tables and graphs. We show that figures stand in both presentational and subject matter relations to the text they accompany, and that cross‐referencing varies widely across genres, with newspaper articles showing little or no reference to the graphical material, and scientific articles marking the reference to the figure explicitly in the text.

The Discourse op Dinsdag discussion group is intended for researchers working on discourse from a language use perspective, and offers a platform to discuss their work (in progress). For more information check our website http://www.let.uu.nl/vici.

Location: Utrecht University, Janskerkhof 13, Room 0.06

Arnout Koornneef

Utrecht University, UiL-OTS

Anticipating causal relations between events

Implicit causality is a property of some interpersonal verbs in which one or the other of the verb’s arguments is implicated as the underlying cause of the action or attitude. For example, people normally ascribe the underlying cause of the praising-event in a sentence fragment such as David praised Linda because to the object NP (e.g. because she had completed the assignment). In series of self-paced reading, eye-tracking and ERP experiments we showed that this pragmatic implicit causality cue is used to anticipate how the unfolding utterance will continue. Currently, we are pursuing a line of research in which we combine these insights from linguistics with knowledge from visual perception research. More specifically, using neuroimaging techniques (ERP, fMRI) this project investigates the hypothesis that two of our basic cognitive systems – vision and language – make use of the same cognitive components to anticipate causal relations between events.

Date & time: December 15; 15:30-17:00

Location: Utrecht University, Janskerkhof 13, Room 0.06

Arie Verhagen

Leiden University

Intersubjectivity, Subjectivity, and Common Ground

Abstract

In this talk I will explore conceptual connections between the notions of Intersubjectivity as developed in Verhagen (2005, 2007; see also http://www.arieverhagen.nl/Documenten/CoI.html) and some other, related notions, especially Common Ground (cf. Clark 1996), in an attempt to clarify what is special about the grammatical phenomena that I labelled “constructions of intersubjectivity”. An important motivation for such an attempt at clarification is the fact that the term “intersubjectivity” is used in the literature in relation to phenomena (e.g. deixis), that I would not (without further argumentation) consider as linguistically involving intersubjectivity, despite the fact that “mutually shared information” somehow does play a role in their conventional meaning. The discussion will result in a proposal that constructions of subjectivity and of intersubjectivity can sensibly be said to profile and manage different aspects of the common ground in different ways.

References

Clark, Herbert H. (1996), Using Language. Cambridge: Cambridge University Press.

Verhagen, Arie (2008), Intersubjectivity and the architecture of the language system. In: Jordan Zlatev, Timothy P. Racine, Chris Sinha, Esa Itkonen (eds.), The Shared Mind: Perspectives on Intersubjectivity. Amsterdam/Philadelphia: John Benjamins Publishing Company, 307-331.

Verhagen, Arie (2005), Constructions of Intersubjectivity. Discourse, Syntax, and Cognition.  Oxford: Oxford University Press.

Location: Utrecht University, Janskerkhof 13, Room 0.06

Fridolin Wild

Knowledge Media Institute, Milton Keynes, UK

The Geometry of Learning

Abstract

Latent Semantic Analysis (LSA) is a mathematical technique for computationally modeling the meaning of words and larger units of texts. LSA works by applying a mathematical technique called Singular Value Decomposition (SVD) to a term*document matrix containing frequency counts for all words found in the corpus in all of the documents or passages in the corpus. After this SVD application, the meaning of a word is represented as a vector in a multidimensional semantic space, which makes it possible to compare word meanings, for instance by computing the cosine between two word vectors.

LSA has been successfully used in a large variety of language related applications from automatic grading of student essays to predicting click trails in website navigation. In Coh-Metrix (Graesser et al. 2004), a computational tool that produces indices of the linguistic and discourse representations of a text, LSA was used as a measure of text cohesion by assuming that cohesion increases as a functionof higher cosine scores between adjacent sentences.

Besides being interesting as a technique for building programs that need to deal with semantics, LSA is also interesting as a model of human cognition. LSA can match human performance on word association tasks and vocabulary test. In this talk, Fridolin will focus on LSA as a tool in modeling language acquisition. After framing the area of the talk with sketching the key concepts learning, information, and competence acquisition, and after outlining presuppositions, an introduction into meaningful interaction analysis (MIA) is given. MIA is a means to inspect learning with the support of language analysis that is geometrical in nature. MIA is a fusion of latent semantic analysis (LSA) combined with network analysis (NA/SNA). LSA, NA/SNA, and MIA are illustrated by several examples.

On Wednesday morning, November 18, Fridolin Wild will give a tutorial during which he will demonstrate the R-package he developed for LSA. For more information, please contact Rogier Kraf (r.kraf@uu.nl).

The Discourse op Dinsdag discussion group is intended for researchers working on discourse from a language use perspective, and offers a platform to discuss their work (in progress). For more information check our website http://www.let.uu.nl/vici.

Location: Utrecht University, Janskerkhof 13, Room 0.06

 Sarah van Vliet

VU University Amsterdam

The use of proper nouns and pronouns in narrative discourse:

Towards a process model of reference maintenance

 Abstract

When narrators refer to the characters in their stories, they typically alternate between explicit referential expressions such as proper nouns (‘Henry’), and attenuated expressions such as pronouns (‘he’). In this talk I report an analysis of maintained protagonist references within a large corpus of elicited written Dutch narratives (van Vliet 2008). The analysis focuses on the consecutive choices between pronouns and repeated proper nouns, relative to a number of grammatical and discourse factors. On the basis of the results I present a model of reference maintenance in which the choice between proper nouns and pronouns is guided by both linear factors such as referential distance, and hierarchical factors such as episode structure.  I will discuss the role of referent salience (cf. Ariel 2001 inter alia) and attention fluctuation in this model of referential choice, and the question whether referential choice can best be explained in terms of speaker-based or addressee-oriented mechanisms (cf. Arnold 2008)

References

Ariel, M. 2001. Accessibility Theory: An overview. In T. Sanders & J. Schilperoord & W. Spooren (Eds.), Text Representation: Linguistic and Psycholinguistic Aspects. Amsterdam / Philadelphia: John Benjamins.

Arnold, J.E. (2008). Reference production: production-internal and addressee-oriented processes. Language and Cognitive Processes, 23 (4), 495-527.

Van Vliet, S. (2008). Proper Nouns and Pronouns: The production of referential expressions in narrative discourse.   Ph.D. diss., Tilburg University. LOT series 175.

Date & time: March 30; 11:00-12:30

Location: Utrecht, Kromme Nieuwegracht 80, Stijlkamer van Ravesteyn 1.06

Jennifer Spenader

University of Groningen

Coherence Relations, Anaphora and Presupposition

Abstract:

Consider the following sentences and what the preferred interpretation of “he” is:

(1) Samuel threatened Justin with a knife, and he blindfolded Erin with a scarf (too). (Parallel)
(2) Justin was threatened by Samuel with a knife, and he blindfolded Erin with a scarf (too). (Parallel)
(3) Justin was threatened by Samuel with a knife, and he was arrested. (Cause-Effect)

The verb meaning makes clear that “he” should be interpreted as Justin in all three sentences, but somehow (2) is awkward. Kehler’s (2002) theory claims parallel coherence relations between two clauses constrains antecedent-anaphor relationships to parallel interpretations, e.g. object antecedents for object anaphora, etc. Because (2) has a passive-active voice mismatch, interpreting the pronoun in (2) is more awkward than in (1).  Cause-Effect relations are claimed to permit freer anaphoric links,  thus the same voice mismatch in a cause-effect relation in (3) is claimed not to influence anaphoric relationships.

Kehler illustrates his theory with convincing natural examples, but recently his theory has been tested experimentally, with a number of conflicting results (e.g. Frazier & Clifton, 2006, Kehler et al. 2008, Kertz 2008, Kobele et al. 2008). As a possible explanation, Frazier & Clifton (2006) and Hendriks (2002) have both suggested that the presence of presupposition triggers like “too” might influence pronoun interpretation possibilities.

In this talk I will present the motivation and set-up of a series of planned experiments to clarify the contribution of presuppositional triggers to pronoun interpretation in Parallel and Cause-Effect coherence relations, and discuss potential outcomes and their explanations.

The Discourse op Dinsdag discussion group is intended for researchers working on discourse from a language use perspective, and offers a platform to discuss their work (in progress). For more information check our website http://www.let.uu.nl/vici.

Date & time: March 24, 15:30-17:00

Location: Utrecht University, Janskerkhof 13, Room 0.06

Sergey Avrutin

UiL-OTS, Utrecht University

The syntax-discourse interface and representation of anaphoric dependencies

Abstract

I will present a model of the syntax-discourse interface that is based on the different representations of functional and lexical elements in discourse.  Specifically, in the model I develop, functional categories such as D introduce a discourse place holder (‘a file card frame’) and lexical categories such as N introduce the referential substance (‘a file card heading’).  Anaphoric elements of three types are analyzed: SE elements (‘zich’), SELF elements (‘zichzelf’) and pronouns. The main idea is that the differences in morphology and feature specification of these elements result in their different discourse representations.  Their interpretation depends on the relationship between discourse entities (file cards) which are constrained by a set of rules/principles that I discuss.  A substantial part of the theory of anaphora, therefore, appears to be explainable at a beyond-narrow-syntax level.

The Discourse op Dinsdag discussion group is intended for researchers working on discourse from a language use perspective, and offers a platform to discuss their work (in progress). For more information check our website http://www.let.uu.nl/vici.

Meetings are planned 4-weekly on Tuesdays 15:30-17:00, in Utrecht. If you wish to participate, please let us know. If you’re interested in presenting your research, we invite you to come forward by sending an email to r.vanveen@uu.nl.