English Inversion: A Ground-Before-Figure Construction (review)

The present volume is another study in a line of recent monographs that aim to account not only for the structural, but also the discourse-functional and pragmatic characteristics of full inversion in English. What distinguishes Rong Chen’s approach from previous treatments by Betty J. Birner (The discourse function of inversion in English, New York: Garland, 1996) and Heidrun Dorgeloh (Inversion in English: Form and function, Amsterdam: John Benjamins, 1997) is that it is anchored within the framework of cognitive linguistics and largely based on the theory of cognitive grammar (Ronald Langacker, Foundations of cognitive grammar, 2 vols., Stanford: Stanford University Press, 1987/1991). C introduces a cognitive model called Ground-before-Figure (GbF) and intends to provide a unified explanation for the semantic, phonological, syntactic, and pragmatic behavior of inversion.

Ch. 1 discusses the different types and peculiarities of full inversion, reviews previous research, and explains basic tenets of cognitive linguistics. In this section, C repeatedly uses the notion of markedness, but nowhere in the text is this concept defined or explained. It remains unclear whether it refers to structural or discourse-functional aspects, especially when inversion is compared with other types of focus constructions in English, such as cleft sentences, which are discussed in a rather oversimplified manner.

Ch. 2 introduces the analytic model used in the book, discusses the phonology of inversion, and compares the GbF model to the view of inversion as a means of information-packaging (Birner 1996). Inversion is seen as a linguistic instantiation of the GbF model: the preverbal constituent of inversion represents the ‘ground’, the postverbal subject-NP the ‘figure’. Adopting a linear spatial orientation, this model also intends to capture the function of inversion as a focus construction. C argues that inversion serves to first direct the hearer’s attention to the ‘ground’, which also contains a ‘landmark’, that is, the previous discourse context. This landmark then serves as a signpost, navigating the hearer to the ‘figure’, which is placed in the focus of the hearer’s attention.

Inversion is understood to be a radial phenomenon, one with prototypical occurrences and extensions from this prototype. According to C, the prototype is characterized by the semantic features of both the pre- and postverbal element: It typically contains a locative element in the form of a PP in preverbal position, and a form of be or other stative verbs such as stand, sit, lie, or hang in postverbal position (also referred to as locative inversion). However, Birner (1996) and Birner and Gregory Ward (Information status and noncanonical word order in English, Amsterdam: John Benjamins, 1998) have convincingly identified two discourse-pragmatic constraints on the felicitous occurrence of inversion, both relating to the information status of sentence constituents. They argue that (i) the fronted constituent and (ii) the verb (other than be) that can appear in full inversion have to represent evoked or inferable information and must not contribute discourse-new information. Thus, a seemingly semantic restriction on verbs occurring in inversions (only intransitive verbs of stance and motion are allowed) actually follows from a discourse-pragmatic constraint.

Ch. 3 examines the various syntactic restrictions on inversion that have been proposed, involving issues such as polarity, transitivity, embeddedness, auxiliarity, and weight, while Ch. 4 discusses the various discourse functions of inversion in description, narration, and exposition. Ch. 5 summarizes the major findings and contains suggestions for further research, such as how the GbF representation can be applied to other languages. The volume is rounded off by an endnote section, lists of references and sources of the corpus examples, as well as a subject and author index.

C’s analysis is largely based on a corpus of 1,132 real-life examples of inversion in written (97%) and spoken (3%) English, and the data confirm some well-established facts and tenets about inversion, for example, that full inversion is virtually restricted to the written mode and strongly conditioned by context, and...