Journal of English Linguistics

This paper examines the discursive construction of refugees and asylum seekers (and to a lesser extent immigrants and migrants) in a 140-million-word corpus of UK press articles published between 1996 and 2005. Taking a corpus-based approach, the data were analyzed not only as a whole, but also with regard to synchronic variation, by carrying out concordance analyses of keywords which occurred within tabloid and broad-sheet newspapers, and diachronic change, albeit mainly approached from an unusual angle, by investigating consistent collocates and frequencies of specific terms over time. The analyses point to a number of (mainly negative) categories of representation, the existence and development of nonsensical terms (e.g., illegal refugee), and media confusion and conflation of definitions of the four terms under examination. The paper concludes by critically discussing the extent to which a corpus-based methodological stance can inform critical discourse analysis.¹

Many reference grammars cover the use of last and next, but none pays overt attention to when and why those words combine with ø or the before temporal nouns; for example, (a) I came to Boston ølast year and (b) I've been in Boston for the last year. Based on three theoretical notions of predicated time, extensivity, and the null article, and a corpus analysis of the tokens of (the) last/next from the Brown Corpus, the 1996 LA Times Corpus, and the Michigan Corpus of Academic Spoken English, this article presents a detailed account of when the determiners last and next combine with null or the; why last/next followed by singular temporal nouns occur with null, as in (a), or with the, as in (b); and why only singular temporal nouns, but not plural temporal nouns or non-temporal nouns, can combine with null + last/next.

As part of a larger perceptual dialectology study of linguistic diversity within California, this article focuses on a survey of Californians regarding the evaluation of language use within the state. Quantitative and qualitative analysis of two open-ended survey questions regarding where Californians “speak the best” and “speak the worst” reveals that Southern California is stigmatized by a majority of respondents as having the worst speech within the state and Northern California is valorized as having the best speech, due to the perceived differential access of speakers to educational opportunities. A small but socially significant “political correctness effect” also emerges, whereby some respondents indicate reluctance to evaluate others' linguistic varieties. The findings demonstrate both the need for greater attentiveness to ideological issues in research design and the importance of combining different theoretical and methodological traditions in the study of language ideologies and attitudes.