This textbook introduces readers to the ways in which co-text and discourse context affect syntactic structure in English. It is aimed at intermediate to advanced undergraduates in linguistics and should appeal to any students who wonder why the English sentences they come across in the real world differ so much from what they encounter in the classroom. Even though some of these phenomena are treated in other textbooks aimed at undergraduates, too many of those descriptions are limited to syntax and semantics. When a typical textbook brings up the passive voice, for instance, the discussion is invariably about how passive clauses differ in structure from their active counterparts while maintaining, in most cases, the same truth conditions. This book concentrates instead on why speakers might prefer to use the active or passive for expressing themselves in a given situation.
The authors split the book into three broad sections that are followed by a glossary. The first section (chapters 1 and 2) serves as a foundation. Chapter 1 (pp. 3–10) explains how the book builds on content typically found in introductory syntax and grammar courses and also how the book is organised to facilitate learning. Chapter 2 (pp. 11–48) defines important concepts, shows readers how to use corpora to study language in use, and describes different approaches to the data.
The second section (chapters 3–5) is on information-packaging constructions, which the authors place under the title ‘Grammar in discourse’. Chapter 3 (pp. 51–87) is about phenomena associated with the beginnings of clauses, such as complement preposing, left dislocation and subject-dependent inversion. Chapter 4 (pp. 88–129) deals with passives and the order of intransitive prepositions with direct objects. Chapter 5 (pp. 130–58) handles phenomena that shift linguistic material towards the ends of clauses: existentials, presentationals, extraposition and it-clefts. Surprisingly, it leaves out postposing and right dislocation.
The third section (chapters 6–9), which carries the title ‘Grammar of discourse’, covers a host of other topics. Chapter 6 (pp. 161–89) explores coordinators and connective adjuncts, while chapter 7 (pp. 189–221) yokes together pronouns and ellipsis, which might have deserved their own chapters. Chapter 8 (pp. 222–53) is on parenthetical discourse markers and chapter 9 (pp. 254–88) considers genre through two topics of particular relevance to undergraduates: academic writing and digital discourse.
The book deserves praise on several counts. First, there is a welcome focus on pedagogy throughout. Learning outcomes and important terms are set out near the beginning of each chapter, and exercises are provided at two difficulty levels. This is good: too many textbooks proceed directly to difficult research questions when readers first need to test how well they understand the concepts. Here, readers are given ample opportunity to build their confidence. They are then encouraged to apply the techniques they have learned to fresh data and are also asked to develop and refine research questions of their own.
The illustrations, which were provided by Wanner's (then) doctoral student Lynn Zhang, should also be noted because in several instances they genuinely contribute to understanding by reinforcing the points under discussion without merely repeating the textual explanation. Figure 3.2 (p. 64), for example, uses overlapping speech bubbles to show how given or inferred information can feed into future utterances.
The book draws together strands of classic and more current research that are not always found together. The phenomena are always presented as topics that are still being explored, which is important when students are looking for research ideas. Ample reading recommendations are provided for exploration.
I was particularly glad to see a continued focus on academic writing, an area of expertise for both authors. Academic writing is important to the target audience not only as a topic of study but also as a set of skills to be acquired. Students who become acquainted with the CARS model (Swales Reference Swales1990) will not only learn something about genre analysis but also discover useful tips on how to structure the introduction of a dissertation. When they encounter Shaw's (Reference Shaw, Charles, Hunston and Pecorari2009) comparison of the use of connective adjuncts by novice and experienced writers of academic prose, they will find out that journal articles tend to employ words like therefore more sparingly than student essays do. And Biber & Gray's (Reference Biber and Gray2010, Reference Biber and Gray2016) studies of academic prose style upend the pat generalisations about complexity that are so casually passed around as truths: academic prose is often described as ‘elaborate’, but ‘compressed’ is a better label. Students will come away from all this with a much better idea of how to fulfil expectations in their own academic work.
Finally, Huddleston & Pullum et al. (Reference Huddleston and Pullum2002), henceforth CGEL, was chosen as the book's reference for terminology (p. 7), a welcome decision for several reasons. It is important, of course, for constructions to be introduced in the theoretically neutral way that CGEL advocates, but there is also the added benefit that students can move to this book directly from the recently revised Huddleston et al. (Reference Huddleston, Pullum and Reynolds2021), an introduction to English grammar that follows CGEL's terminology and finishes with a chapter on information packaging. The book thus positions itself nicely for students who want to start engaging in research.
But this leads to my main issue, which is that any students following that learning path will encounter several incompatibilities with the CGEL framework, as the authors’ commitment to it is only partial. A large part of what CGEL sets out to do is clear away older definitions that have been preserved elsewhere for tradition's sake. At its heart is a strict separation of category and function that guides its analyses. Anyone who is still in the process of coming to grips with describing English grammar (which will surely be the case for many readers of the current book) will be frustrated by any inconsistencies.
I am not referring here to the trivial substitution of one term for another. Students will, of course, have to get used to the fact that there are many overlapping terms in the literature, so it does them a favour to point out, for example, the equivalence of intransitive prepositions and particles (p. 95). It would be good if the book did this more consistently: many other potential stumbling blocks, such as the change from deixis to exophoric reference or from complement preposing to fronting, are not mentioned.
But there are inconsistencies that go deeper. CGEL treats before as a preposition regardless of whether its complement is a noun phrase, a clause, or nothing at all. This analysis is defended at length (CGEL: 598–601, 1011–14). Dorgeloh & Wanner, however, label before as a ‘subordinator (conjunction)’ when it takes a clause as its complement (pp. 4–5). This is not an unreasonable stance to take: it aligns with the views in Quirk et al. (Reference Quirk, Greenbaum, Leech and Svartvik1985: 998ff.) and Biber et al. (Reference Biber, Johansson, Leech, Conrad and Finegan2021: 78ff.). But subordinating conjunctions are not part of the CGEL framework. The problem comes to a particular head later on when innovative uses of because are discussed (p. 275): subordinating conjunctions are brought up again even though the research in question, Bohmann (Reference Bohmann and Squires2016), explicitly adheres to the CGEL framework in categorising the word as a preposition no matter what follows. Dorgeloh & Wanner's reversion to old terminology undoes this progress.
A terminological point that might cause more confusion is the use of ‘canonical’. In the CGEL framework, there are five types of canonical clause: intransitive, complex intransitive, monotransitive, complex monotransitive and ditransitive. Clauses are classified as non-canonical for reasons connected with meaning (e.g. negation or clause type) or information packaging. These non-canonical clauses involve additional layers of syntactic complexity. But Dorgeloh & Wanner (p. 4) treat non-canonical clauses instead as simply less typical arrangements of their canonical counterparts with the same truth conditions but different structures that are motivated by discourse (p. 22). If we follow them in using the term this way, then negative and interrogative clauses, for instance, would both incorrectly count as canonical. The confusion is compounded when the term is applied to passives: they use ‘canonical’ as a label for be-passives (pp. 97ff.) and ‘non-canonical’ for passives with different forms (p. 99). Huddleston & Pullum et al. do not use ‘canonical’ to mean ‘typical’ except in one slip corrected in the errata (CGEL: 1365). So Dorgeloh & Wanner use canonicity both too narrowly (canonical and non-canonical counterparts are held to have the same truth conditions, which reduces canonicity to a matter of information packaging) and too broadly (certain passive clauses are held to be canonical as well, which dilutes canonicity to typicality). This muddles things up for students unnecessarily.
There are also smaller inconsistencies with CGEL scattered throughout. For instance, when Dorgeloh & Wanner discuss dative alternation, they discuss ‘giving money to somebody’ (p. 33) as having two objects, while CGEL is clear that this is a monotransitive with a non-core complement PP (CGEL: 308–11). Theme and patient are held to be equivalent (p. 90), which CGEL explicitly argues against (CGEL: 231ff.). And the single term ‘existential’ is applied to both true existentials (e.g. ‘there was an old woman’) and presentationals (e.g. ‘in the castle there lived a handsome prince’) (pp. 130ff.), which Huddleston & Pullum et al. treat separately for reasons related to information status (CGEL: 1390–1403).
In other cases, the issues are not connected with CGEL. When faced with the problem of distinguishing corpora from other collections of text, Dorgeloh & Wanner tell us that the former are pre-analysed: ‘At a minimum, corpora are annotated for part-of-speech information’ (p. 31) (this is put more softly on p. 26, where it is said that linguistic corpora are usually pre-analysed). But this definition excludes all unannotated corpora from consideration and completely disregards the principles by which the text is collected. If there is something that separates corpora from other sets of computer-based text, it is that corpora involve texts that are selected by external criteria so that they are representative (Sinclair Reference Sinclair2005). When people collect texts for other reasons, they usually make their selections based on internal properties of the texts. A collection of American cookbooks published in the nineteenth century is a corpus; a collection of your favourite recipes for angel food cake is not.
Later, after providing the example ‘Among the things I gathered in a pile by the door were an empty pack of cigarettes …’, Dorgeloh & Wanner say that the were it contains is a light verb, ‘which means it usually does not stand alone’ (p. 52). The fact that auxiliary be cannot easily stand alone is incidentally true, but that does not make it a light verb. When a verb like make is used as a light verb, there is a non-verbal complement later on that carries most of the meaning (as in ‘She made a reservation for a room’) and there is a simpler alternative without the light verb (in this case, ‘She reserved a room’). The verb be does not fit this pattern.
It is important to stress that these issues could all easily be addressed in a second edition. Even in its current form, the book would still be a good choice for a course text for undergraduates for its various merits: its usability, the variety of literature it draws together, and the way in which it serves as an introduction to how linguists work. We need more books like this to help students make the jump to independent research.
