Weak and Strong Discourse Markers in Speech, Chat, and Writing: Do Signals Compensate for Ambiguity in Explicit Relations?

Abstract

Ambiguity in discourse is pervasive, yet mechanisms of production and processing suggest that it tends to be compensated in context. The present study sets out to analyze the combination of discourse markers (such as but or moreover) with other discourse signals (such as semantic relations or punctuation marks) across three genres (discussion, chat, and essay). The presence of discourse signals is expected to vary with the ambiguity of the discourse marker and with the genre. This analysis complements Das and Taboada's (2018) approach to discourse signalling by zooming in on the different types of discourse markers with which other signals combine. The corpus annotation study uncovered three categories of marker strength -weak, intermediate and strong -thus refining the concept of 'explicitness'. Statistical modeling reveals that weak discourse markers are more often compensated than intermediate and strong markers, and that this compensation is not affected by genre variation.

Lexical and Structural Cues to Discourse Processing in First and Second Language

“…These studies focus on single pairs of connectives (e.g., and vs. but ), and comparisons across various types of connectives are still lacking. More fine-grained distinctions between connectives would thus complement previous findings by further refining the “continuum from implicit to explicit discourse marking” (Crible, 2020 , p. 12) and may highlight conditions in which non-connective cues are more useful than others. Furthermore, how these interactions impact non-native readers remains entirely to uncover.…”

“…However, it should be noted that contrast is rarely included in discourse processing studies, which tend to focus on other types of relations such as concession or result. As Crible ( 2020 ) observed in corpus data, contrastive relations (along with elaboration relations) tend to be expressed by contextual cues more often than other relations, which means that the instruction of the connective might be redundant with the information already present in the segments themselves. As mentioned before, contrast is necessarily marked semantically, which makes it highly accessible to native speakers.…”

“…A further aspect that is only beginning to attract interest to deepen our understanding of discourse processing is the role of other types of cues from the semantic or syntactic levels, and how these cues interact with connectives. Corpus studies (Das and Taboada, 2018;Hoek et al, 2019;Crible, 2020) have shown for example that a wide range of contextual cues can contribute to the marking of discourse relations, from semantic relations to syntactic constructions and even punctuation signs. One of them is structural parallelism, which consists in repeating the same argument structure across the connected segments, as in Example (3).…”

The role of non-connective discourse cues and their interaction with connectives

“…First, we compare conceptually different relations, namely consequence (positive causal) and contrast (negative additive). Negative relations such as contrast tend to be reinforced by contextual cues besides the connective more frequently than positive relations such as consequence (Crible 2020;Das & Taboada 2018). Therefore, we expect to find a stronger effect of cues on the disambiguation of contrast than consequence relations.…”

“…There is thus an extensive body of research on the individual roles of connectives and other discourse cues in the marking and processing of coherence relations. By contrast, the interaction between connectives and other cues has only recently started to be systematically investigated from corpus-based (Crible 2020;Hoek et al 2019;Das & Taboada 2019) or experimental methods (Crible & Pickering 2020;Grisot & Blochowiak 2019). These studies point to a multifactorial phenomenon of co-occurrence between connectives and other cues, which is impacted by cognitive factors of complexity and information density.…”