International Journal of Corpus Linguistics - Volume 13, Issue 4, 2008

The most frequent statistics in corpus linguistics are frequencies of occurrence and frequencies of co-occurrence of two or more linguistic variables. However, such frequencies in isolation may sometimes be misleading since they do not take into consideration the degree of dispersion of the relevant linguistic variable. Many dispersion measures and adjusted frequency measures have been suggested but are neither widely known nor applied. Another unfortunate aspect of such measures is that many also come with a variety of problems. I pursue three objectives with this article. First, I want to raise awareness of this issue and make the available measures more widely known, so I present an overview of many measures of dispersion and adjusted frequencies. Second, I propose a conceptually simple alternative measure, DP, explain and exemplify it, and compare it to previously discussed measures. Third and most importantly, I urge corpus linguists to explore the notion of dispersion in more detail and outline a few proposals which steps to take next.

This article describes a hybrid corpus linguistic approach to conversational storytelling, whereby one first identifies a feature of interest in a small set of narratives, then moves to a general investigation of this feature in large corpora of transcribed conversation, focusing only later on the distribution and functions of this feature within narrative portions of the corpora investigated. Interjections are conversational units par excellence with no syntactic relation to adjacent clauses, so that investigation of large corpora is particularly vital for determining their patterning and functions in conversation. I show that interjections play a number of important roles in the organization of conversational storytelling, first in justifying tellability, next in marking narrative climaxes, particularly within constructed dialogue, then in evaluating the narrative point, and finally in receiving and commenting on the storytelling performance. I further describe combinations of interjections and the interaction of interjections with exclamative clauses.

Each instance of language comparison requires observations on semantic equivalence. Meaning is by far the most popular tertium comparationis in contrastive and typological research. However, the question of how semantic equivalence is to be determined remains extremely difficult to solve. This paper presents an approach to detect semantic relatedness between a limited range of lexical items from different languages on the basis of monolingual data. Applying distributional similarity (Dagan et al. 1999) cross-linguistically, it identifies semantically related verbs governing embedded interrogatives by looking at the frequency of the question words (i.e. wh-items) that are used in the embedded interrogatives in monolingual corpora. Convincing results are obtained for six different language pairs: English-French, English-Dutch, English-Spanish, French-Dutch, French-Spanish and Dutch-Spanish.

Liking adverbials (LAs) play a very important role in discourse cohesion. This study examines the frequency and usage patterns of English LAs (110 in total) across five registers (Spoken English, Academic Writing, Fiction, News Writing, and Other writings) in the BNC. While the data analyses offer support for some previous research results, they also yield new findings regarding, among other things, variations in the adverbials’ frequency distribution and usage patterns across the five registers. For example, while fiction’s overall use of LAs is much lower than that of speaking and academic writing, its use of the sequential type of LAs is actually higher than the latter registers’. Also although news has the lowest overall LA frequency among the five registers, it boasts the highest use of LAs that express simultaneity (e.g. meanwhile). Besides discussing new findings, the paper also addresses their implications for language description, learning, and teaching.

This paper reports the extension of the key words method for the comparison of corpora. Using automatic tagging software that assigns part-of-speech and semantic field (domain) tags, a method is described which permits the extraction of key domains by applying the keyness calculation to tag frequency lists. The combination of the key words and key domains methods is shown to allow macroscopic analysis (the study of the characteristics of whole texts or varieties of language) to inform the microscopic level (focussing on the use of a particular linguistic feature) and thereby suggesting those linguistic features which should be investigated further. The resulting ‘data-driven’ approach presented here combines elements of both the ‘corpus-based’ and ‘corpus-driven’ paradigms in corpus linguistics. A web-based tool, Wmatrix, implementing the proposed method is applied in a case study: the comparison of UK 2001 general election manifestos of the Labour and Liberal Democratic parties.