International Journal of Corpus Linguistics - Volume 18, Issue 2, 2013

The aim of this paper is to demonstrate the functionality of the WebCorp Linguist’s Search Engine (WebCorpLSE) by comparing it with two other freely-available online data resources: Google and the Corpus of Contemporary American English (COCA). In particular, the paper looks at how WebCorpLSE can fill gaps in these sources. It begins by contextualising the WebCorpLSE project in relation to its parent project, WebCorp, and other ‘web as corpus’ initiatives. It then outlines the advantages of WebCorpLSE over Google search, before turning to a comparison of WebCorpLSE with COCA. Comparisons are made by matching extracts of textual and graphical output generated by each system. It is shown that WebCorpLSE provides information not available via the other search mechanisms, and that it can thus, in accordance with Mair’s (2007) argument for a combined approach to diachronic text study, usefully serve not only as a stand-alone tool, but as a complementary resource.

That complement clauses are a prominent feature of various registers including conversation and academic prose. In academic prose, that-clauses are of interest because they frame research findings, the writer’s central message to the reader. To achieve this persuasive purpose, that-clauses are employed to draw in various voices, including those of other researchers, research participants, research findings and the writer. This study extends prior investigation of complement clauses to examine their distribution across different sections of a corpus of research articles in social science. The social action of each section is partially achieved through what the different voices in the different sections of the article talk about, and the subtle variations in the stance of the author and other voices across sections. This study finds that use of reporting verbs is nuanced according to authors’ purposes in different sections, and also according to the source of the proposition in the that-clause.

Legal language is often said to be formulaic, but little research is available on the nature of frequently occurring sequences of words in different legal genres. This article investigates the lexical bundles found in four legal corpora: academic law, case law, legislation, and documents. Major differences are brought to light between the type of bundles that are found, and the roles they have in the text. Academic legal writing uses relatively little formulaic language. Case law uses noun phrase bundles relating to agents, documents and actions, as well as many extended prepositional phrases. Legislation and documents contain many noun phrase bundles, and verb phrase bundles with a deontic or referential function. The function of these different types of bundle as parts of a schematic frame or as slot-fillers is discussed.

This study attempts to examine the potential of selected corpus linguistics and computational stylistics methods in the investigation of translation universals in translational literary Polish. It deals with T-universals (Chesterman 2004), with emphasis on the simplification hypothesis, as manifested in the core patterns of lexical use (Laviosa 1998) and the levelling out hypothesis (Baker 1996). To that end, the purpose-designed corpora, each with approximately 350,000 tokens, of contemporary translational and non-translational literary Polish were compiled. The results confirm the simplification and the levelling out hypotheses but only with reference to the mean sentence length and variance for the mean sentence length. On the other hand, the results of multivariate analyses (Principal Components Analysis and Cluster Analysis) confirm the levelling out hypothesis that translations are more alike as compared with native texts.

Linguistic science of the past half-century has often been distorted through neglect of normal scientific standards of empirical falsifiability. An earlier paper in this journal used a quantitative literature survey to examine how far in practice the newer trend towards use by linguists of corpora and other empirical data sources had progressed. The result was ambiguous: a trend towards greater empiricism had occurred since about 1970, but around the turn of the century it appeared to have reversed, and the end of the period surveyed (2002) fell so soon thereafter that it was hard to guess whether this reversal was a blip or a long-term change. With a further decade of linguistic literature to examine, the present paper repeats the survey using a more systematic sampling technique, and this yields results that are much more clearcut than those of the earlier paper.