George Aaron Broadwell

Modeling Sociocultural phenomena in discourse

In this paper, we describe a novel approach to computational modeling and understanding of social and cultural phenomena in multi-party dialogues. We developed a two-tier approach in which we first detect and classify certain sociolinguistic behaviors, including topic control, disagreement, and involvement, that serve as first-order models from which presence the higher level social roles, such as leadership, may be inferred.

Using imageability and topic chaining to locate metaphors in linguistic corpora

The reliable automated identification of metaphors still remains a challenge in metaphor research due to ambiguity between semantic and contextual interpretation of individual lexical items. In this article, we describe a novel approach to metaphor identification which is based on three intersecting methods: imageability, topic chaining, and semantic clustering. Our hypothesis is that metaphors are likely to use highly imageable words that do not generally have a topical or semantic association with the surrounding context. Our method is thus the following: (1) identify the highly imageable portions of a paragraph, using psycholinguistic measures of imageability, (2) exclude imageability peaks that are part of a topic chain, and (3) exclude imageability peaks that show a semantic relationship to the main topics. We are currently working towards fully automating this method for a number of languages.

The Historical Development of the Progressive Aspect in Central Zapotec

The Central Zapotec languages share the innovation of a progressive aspect prefix in /ka-/ which contrasts with a habitual aspect /r-/, which is the reflex of a Proto-Zapotec imperfective prefix */tyi-/. Early instances of this innovative progressive aspect prefix can be found in colonial Zapotec texts from the sixteenth century, and the geographic dispersion of Central Zapotec languages with this form argues that the innovation must date to some centuries prior to a.d. 1300, and that this morphological characteristic is a key marker that distinguishes Central Zapotec languages from other members of the Zapotec family.

Computing Affect in Metaphors

This article describes a novel approach to automated determination of affect associated with metaphorical language. Affect in language is understood to mean the attitude toward a topic that a writer attempts to convey to the reader by using a particular metaphor. This affect, which we will classify as positive, negative or neutral with various degrees of intensity, may arise from the target of the metaphor, from the choice of words used to describe it, or from other elements in its immediate linguistic context. We attempt to capture all these contributing elements in an Affect Calculus and demonstrate experimentally that the resulting method can accurately approximate human judgment. The work reported here is part of a larger effort to develop a highly accurate system for identifying, classifying, and comparing metaphors occurring in large volumes of text across four different languages: English, Spanish, Russian, and Farsi.

Student leadership in small group science inquiry

Background: Science educators have sought to structure collaborative inquiry learning through the assignment of static group roles. This structural approach to student grouping oversimplifies the complexities of peer collaboration and overlooks the highly dynamic nature of group activity. Purpose: This study addresses this issue of oversimplification of group dynamics by examining the social leadership structures that emerge in small student groups during science inquiry. Sample: Two small student groups investigating the burning of a candle under a jar participated in this study. Design and method: We used a mixed-method research approach that combined computational discourse analysis (computational quantification of social aspects of small group discussions) with microethnography (qualitative, in-depth examination of group discussions). Results: While in one group social leadership was decentralized (i.e., students shared control over topics and tasks), the second group was dominated by a male student (centralized social leadership). Further, decentralized social leadership was found to be paralleled by higher levels of student cognitive engagement. Conclusions: It is argued that computational discourse analysis can provide science educators with a powerful means of developing pedagogical models of collaborative science learning that take into account the emergent nature of group structures and highly fluid nature of student collaboration.

A New Passive in Kaqchikel

This paper contrasts two passives in Kaqchikel, a Mayan language spoken in Guatemala. The first passive, which we label the ‘standard passive’ is already well-attested in the literature. However, the second passive, which we label the ‘ki-passive’, has not been previously described. A verb in the ki-passive shows active morphology, with ergative agreement for a third person plural subject, as would be appropriate for a verb with an impersonal ‘they’ subject. In Kaqchikel, however, we argue that this verb form has evolved into a new passive. The paper compares the properties of the standard passive and the ki-passive, and argues that while they involve the same change of grammatical relations, the two passives differ in the discourse functions they assign to the agent and patient.

On the Phonological Conditioning of Clitic Placement in Zapotec

Clitic placement in San Dionicio Ocotepec Zapotec is determined by a combination of syntactic and phonological factors. The most important of these is alignment with the edge of a phonological phrase. In an optimality-theoretic account, it is possible to rank some phonological alignment constraints above syntactic alignment constraints in a way that correctly accounts for the data. Alternatives which seek to derive clitic placement through syntactic movement operations face insuperable problems with the range of positions in which these clitics appear.

Understanding Cultural Conflicts using Metaphors and Sociolinguistic Measures of Influence

In this article, we outline a novel approach to the automated analysis of cross-cultural conflicts through the discovery and classification of the metaphors used by the protagonist parties involved in the conflict. We demonstrate the feasibility of this approach on a prototypical conflict surrounding the appropriate management and oversight of gun-ownership in the United States. In addition, we present a way of incorporating sociolinguistic measures of influence in discourse to draw further insights from complex data. The results presented in this article should be considered as illustrative of the types of analyses that can be obtained using our methodology; however, no attempt was made to rigorously validate the specific findings reported here. We address open issues such as how our approach could be generalized to analyze cross-cultural conflicts around the world.

Discovering Conceptual Metaphors using Source Domain Spaces

This article makes two contributions towards the use of lexical resources and corpora; specifically making use of them for gaining access to and using word associations. The direct application of our approach is for detecting linguistic and conceptual metaphors automatically in text. We describe our method of building conceptual spaces, that is, defining the vocabulary that characterizes a Source Domain (e.g., Disease) of a conceptual metaphor (e.g., Poverty is a Disease). We also describe how these conceptual spaces are used to group linguistic metaphors into conceptual metaphors. Our method works in multiple languages, including English, Spanish, Russian and Farsi. We provide details of how our method can be evaluated and evaluation results that show satisfactory performance across all languages.