This article argues that English noun-plus-noun constructions (‘NNs’) originate both in the lexicon and in the syntax. It distinguishes between complement–head and attribute–head NNs, as well as between fore-stressed and end-stressed NNs. It argues that complement-head NNs are fore-stressed and originate in the lexicon while attribute–head NNs typically have end-stress and syntactic provenance. The latter are, however, potentially subject to diachronic lexicalization, which may moreover involve the adoption of fore-stress. Hence, lexical NNs may be fore-stressed or end-stressed while phrasal NNs must be end-stressed. Although further potential sources of irregularity are identified, it is demonstrated that the model's predictive power is fairly robust and that, where it fails to predict firm stress patterns, it predicts their variability.

According to Wright (1994a), subjectivity in the English progressive is typically associated with specific linguistic features. In particular, subjective progressives are said normally to occur in main clauses and to involve an adverb(ial) of the type always, a first- or second-person pronominal subject and a private or cognitive verb in the present tense. This study tests Wright's claim against a corpus of Early Modern English prose. The focus is on the kind of subjective progressives that are claimed by Wright to be most subjective of all, namely collocations of the progressive with adverbs such as always. It is shown that the ‘always progressives’ in the corpus are typically found in a subclause, in collocation with an activity verb, and that they commonly occur with different types of subjects and tense/mood combinations. The conclusion is therefore that Wright's predictions concerning typical linguistic contexts for subjective progressives are not borne out.

One aspect of the grammaticalization of the progressive is its spread in Modern English. Previous studies suggest that the progressive was initially restricted to animate or agentive subjects and spread to inanimate or nonagentive subjects only during the later stages of grammaticalization in Modern English. The article discusses the contextual variables – animacy and agentivity – that have been used in previous research. ARCHER – A Representative Corpus of Historical English Registers – is then used (a) to verify the hypothesis that progressives increasingly co-occur with inanimate/nonagentive subjects in the eighteenth and nineteenth centuries, (b) to test the reliability and comparability of previous research, (c) to verify whether the weakening of the contextual constraint was a condition for or a result of the spread of the progressive form in the nineteenth century, and (d) to find out whether there are any regional differences between American and British English in the loss of the contextual constraint.

In current accounts the Old English relative þe is analysed in two radically different ways. The traditional view, inherited from the nineteenth-century grammarians, views the word as a relative pronoun, while the generative analysis, derived from some remarks of Jespersen on the ModE relative that, takes it to be a subordinating particle. The generativist view is based on the word's lack of morphological variation, whereas the older approach examines more generally the grammar of the word, noting that the invariable þe shares the typical nominal categories of number and case, functioning both as a singular and a plural and representing all the four cases of OE nominal elements. A further indication of the word's nominal status is its referential function, distinguishing between specific and generic reference. Against these clear facts, the lack of overt inflection is a minor idiosyncrasy, paralleled by the OE generic man/mon, whose pronominal status is widely agreed. Þe may have been a subordinating particle in origin, but by historical OE times it retained this function in relative clauses only after relative adverbs, having been reanalysed elsewhere either as a relative adverb itself, or, in its most frequent relative use, as a pronoun.

Susan Hunston, Corpora in applied linguistics. Cambridge: Cambridge University Press, 2002. Hardback £40.00, ISBN 0 521 80171 0. Paperback £14.95, ISBN 0 521 80583 X. Pp. xii + 241.

Charles F. Meyer, English corpus linguistics: an introduction. Cambridge: Cambridge University Press, 2002. Hardback £45.00, ISBN 0 521 80879 0. Paperback £16.95, ISBN 0 521 00490 X. Pp. xvi + 168.

Gerald Nelson, Sean Wallis, and Bas Aarts, Exploring natural language: working with the British component of the International Corpus of English. Amsterdam: John Benjamins, 2002. Hardback £95.00, ISBN 90 272 4888 5. Paperback £55.00, ISBN 90 272 4889 3. Pp. xviii + 342.

Rodney Huddleston and Geoffrey Pullum, The Cambridge grammar of the English language. Cambridge: Cambridge University Press, 2002. Pp. xviii + 1842. Hardback, ISBN 0 521 43146 8. £120.

This article reviews Huddleston & Pullum (2002) from the viewpoint of a co-author of Quirk, Greenbaum, Leech & Svartvik (1985). (This author, however, makes no claim whatsoever to represent the views of the other authors of Quirk et al.) Particular attention is paid to some of the more controversial aspects of Huddleston & Pullum's analysis. It is argued that the two grammars, although similar in their comprehensively wide coverage of English, are not strictly comparable, in that Huddleston & Pullum's grammar is more theory-oriented and Quirk et al.'s grammar is more observation-oriented. These different orientations go with different strengths and weaknesses. In some areas Huddleston & Pullum's more up-to-date account has manifest advantages over that of Quirk et al., but there are also arguably areas where Huddleston & Pullum have not moved with the times.