The word such represents a challenge to most grammatical treatments when it comes to assigning it to a particular word class. In this article I consider the syntactic as well as the semantic properties of such. I argue that its conventional dual treatment as a determiner and as a pronoun (Biber, Johansson, Leech, Conrad & Finegan, 1999; Quirk, Greenbaum, Leech & Svartvik, 1985) is unwarranted. I also discuss some more recent treatments (Altenberg, 1994; de Mönnink, 1996; Siegel, 1994) and argue that these too cannot fully account for the facts. Based on syntactic evidence drawn from the British component of the International Corpus of English (ICE-GB), I will argue that the analysis I refer to as the ‘uniform adjective analysis’ of such (Huddleston & Pullum, 2002) best accounts for the facts.

This article explores the orthographic labyrinth surrounding the phoneme often spelt <sh>, and looks for a path or network of paths through it. The aim is to offer a microcosm of what is still lacking in the literature, namely a highly detailed and explanatory account of all spellings of all phonemes, in which the various factors that have affected spellings (and more interestingly their relationships with spoken forms) are granted their importance. The purpose of such a full account would be to facilitate better understanding of how the orthography functions, which should in turn be of interest to linguists and teachers.

In accounting for the orthographic complexities concerning /∫/, reference is made to phonological matters such as palatalization and underlying forms, and also to general ones such as position in word, following phonemes and language of origin of loanwords. Rules are given, using formulaic notation, and exceptional spellings are listed and commented on.

So-called anticipatory it has been variously classified as a semantically empty prop it, as a referential pronoun, and as a category with an inherently cataphoric function. This article, which is based on a corpus study of actually occurring instances of anticipatory it, examines some of the arguments put forward for each of these classifications and – following Bolinger (1977) – argues for an analysis as ‘definite nominal’ with some referential force which can establish a referential link with a clausal constituent in the immediate context. As such, anticipatory it takes an intermediate position between prop it and referring it, all of which are linked by a scale of gradience specifying their scope of reference (wide vs. narrow). This view of anticipatory it, which allows for both anaphoric and cataphoric reference, can account for all informational types of it-extraposition as well as it-extraposition with complement omission, and provides a possible explanation for cases of it-omission.

There are two main strands to this paper. The first is that in Middle English – and early Middle English especially – there are many writing systems that are so complex as to seem disorderly. But a sympathetic and careful interpretation of these systems shows sophisticated underlying order. The second strand is related to the first: early Middle English writing systems are local and may be represented on maps. When complex systems are assigned geographical positions close to each other – and indeed close to where simpler, more economical systems are localized – a picture emerges that can appear haphazard and unlike the dialect continuum we would expect. We refer to this phenomenon as surface nubbliness. This masks the underlying regional dialect continuum we believe to have been present in the spoken language. But knowledge of how these written systems mapped symbol to sound may enable us to uncover a continuum at the level of sound substance.

The aim of this article is to establish whether the structure of morphologically complex words in English is flat or hierarchical and, in case of the latter possibility, whether it is left- or right-branching, and whether the branching direction is consistent throughout the lexicon. Six criteria (three phonological, one morphological, one lexical, and one semantic) are introduced as arguments bearing on the internal structure of words. These criteria are applied to the complete set of prefix-stem-derivational suffix sequences in English. The overall evidence clearly weighs in favour of the hierarchical right-branching model. Right-branching is argued to be a concomitant of translating meaning into form. No branching-direction preference emerges on the semantic criterion. Right-branching is relatively weak on the lexical criterion, stronger on the morphological criterion, and strongest on the phonological criteria. Left-branching is observed to play a certain role in words with unproductive prefixes and is understood as a symptom of demorphologization. The branching direction is found to be largely, though not entirely, consistent. The six criteria generate a low percentage of conflicting outcomes, suggesting that branching paradoxes are a rather uncommon phenomenon. The reason for hierarchical right-branching structure in English is claimed to lie in the processing advantage that it incurs. This explanation applies not only to the morphological but also to the syntactic and phonological levels. Processing ease thus leads to crosslevel harmony in the English language.

Renaat Declerck and Susan Reed, Conditionals: a comprehensive empirical analysis. Topics in English Linguistics 37. Berlin and New York: Mouton de Gruyter, 2001. Pp. xviii + 565. Hardback €98.00/sFr169/$108, ISBN 3 11 017144 9.

We would like to thank the following colleagues for kindly agreeing to referee papers for the journal over the past year.