International Journal of Corpus Linguistics - Volume 10, Issue 3, 2005

The concept of semantic prosody was introduced to the public by Bill Louw in 1993, and has become one of the more important concepts in corpus linguistics. However, while other concepts such as collocation, colligation and semantic preference are relatively unproblematic, one cannot say the same for semantic prosody. At present, it is defined in at least three, distinctly different ways, and more significantly, these differences remain largely undiscussed. This article offers a detailed analysis of Louw's concept of semantic prosody (in Sections 1 through 3), and hopes to demonstrate that the concept and the arguments for it are unconvincing. Concluding the article (Sections 4 and 5) is a brief reflection on the paradox created by Louw's use of metaphor in his definition of semantic prosody.

Relational databases can be used to create large corpora that provide both very good search performance and a wide range of queries. This paper outlines how this approach has been used to create theCorpus del Español, which contains 100 million words of text in Spanish texts from the 1200s-1900s. The main databases are composed of n-grams tables (all unique 1, 2, 3, and 4 word sequences) and the associated frequency of all n-grams in each century (historical Spanish) and register (Modern Spanish). These tables are then joined to other tables containing part of speech, lemma, synonyms, and user-defined lists of words and lemma. There is essentially no limit to the amount of annotation that can be added in additional tables (with little or no impact on performance), and the SQL-based queries allow a wide range of searches that are not available with traditional corpora.

In the last ten years, more and more attention has been paid to the importance of raising the language awareness of language teachers. This is an area in which corpus linguistics has a unique contribution to make. With the help of a concordancer, linguistic features that may be overlooked can be made salient and intertextual information that is implicit in a single text can be made explicit. This paper reports on a study of how corpus evidence was used to address questions sent by English language teachers in Hong Kong to a dedicated website. More than one thousand grammar questions sent to the website over a period of eight years were examined. Three types of most frequently asked questions were identified. The paper discusses how corpus evidence was used to help teachers to notice features and patterns which have escaped their attention and to question long-standing assumptions and misconceptions. It shows how subsequent interrogation of corpus data stimulated by teachers' question often led to new insights into linguistic patterns and language use.

The attempt to balance corpora with respect to their future usage led to the introduction of the termexpectations(Králík 2001b). On the bases of several statistical inquiries of such expectations, the textual structure ofSYN2000,which is the synchronic part of the Czech National Corpus (CNC), was proposed and realised. The present article explains the original composition briefly and discusses two new inquiries concerning expectations(A-2001andC-2001).Important corrections for future work on the CNC are suggested. The expectations concerning newspapers changed radically during 1996–2001. Within the same period, an obvious rise of interest in fiction can be detected. The reasons for these developments can be traced to trends in Czech society. Thus, we have proposed a considerable reduction in the proportion of newspaper texts and a large increase in the proportion of fiction texts. According to new searches, more detailed percentages for specific subject areas are suggested.

The aim of much linguistic research is to determine the grammar and the lexicon of a certain language L. The spoken variant of L – in so far as it is considered at all – is generally taken to be just another projection of the same grammar and lexicon. We suspect that this assumption may be wrong. Our suspicion derives from our contrastive analyses of four corpora, two Swedish and two Danish (covering spoken as well as written language), suggesting that – in the dimensions of frequency distribution, word type selection, and distribution over parts of speech – the mode of communication (spoken versus written) is much more significant as a determining factor than even the choice of language (Swedish versus Danish).

In this study, a purpose-built corpus, containing both child-produced and child-directed speech, is used to conduct a longitudinal examination of antonym use among children from the age of two to five years old. Previous research has tended to approach antonym acquisition using either elicitation techniques or corpora of printed adult language. In contrast, this research focuses on the speech of preschool children in naturally-occurring interactions. The discourse functions of antonymy in child-produced and child-directed language are quantified and compared with those identified in adult, written English (Jones 2002). Despite its complexity, Ancillary Antonymy is found to be most common in child-produced speech, even from the age of two, perhaps because of its particular usefulness in structuring ideas and discourse. This study presents a detailed inter-corpus comparison, assesses the discourse functions of antonymy at different stages of childhood, and discusses the correlation between antonym use in child-directed and child-produced speech.