Formulaic language is a term used by many researchers to refer to lexical units more than one word long. This fast-growing area of applied linguistics has seen a considerable number of publications in the past decade. A search in the Cambridge scientific linguistics and language behavior abstracts database with the keywords formulaic or chunk will pull up as many as 283 published works during the period from 1998 to 2008. Wray’s new book is a much awaited follow-up to its predecessor Formulaic Language and the Lexicon (2002), which won the 2003 British Association for Applied Linguistics book prize. This book extends the framework in Wray (Reference Wray2002) and presents cutting-edge research and new developments in the field. This volume, which compiles theoretical arguments and empirical studies, is aimed at theoretical linguists, SLA researchers, psychologists, computational linguists, corpus linguists, and professionals working in the area of communication disorders (such as Alzheimer’s and aphasia).
As its title indicates, the goal of this work is to investigate the nature and extent of formulaic language by examining a broad range of evidence as well as theories at the boundaries. Specifically, this book addresses the relationship between formulaic language use and practical challenges related to communication, such as face-to-face communication, communication under pressure, computer-mediated communication, and communication disorders. In other words, “formulaic language is investigated here by looking at what happens when language operates beyond its normal scope, where there are unusual constraints on communication that make formulaic language more evident” (p. 5).
This book consists of 22 chapters organized in four parts that provide the contextualization (Part 1), theory (Part 2), and evidence (Part 3) necessary for the discussion of the five questions “arising from published claims about formulaic language” (p. 5), which are the impetus of the book (Part 4). These five questions include the following:
1. Is formulaic language by default?
2. What determines the level of formulaicity?
3. How central is formulaic language in natural language learning?
4. How central is formulaic language when modeling learning for computers?
5. Does formulaic language constrain what we say and think?
Part 1 reiterates the heteromorphic model (i.e., a model that permits multiple part-mappings of the same information in the lexicon; Wray, Reference Wray2002), which challenges traditional boundaries between words and longer word strings and between novel constructions and formulaic language. In this model, linguistic items of various sizes (such as morphemes, words, phrases, sentential frames) are stored holistically in the mental lexicon. Linguistic input is thus interpreted according to communication needs so that larger items are only analyzed into subcomponents in which meaning cannot otherwise be assigned. A particularly interesting point here is the notion of need-only analysis, which is the underlying focus of this book.
Part 2 covers theoretical and empirical issues related to formulaic language. Wray examines existing approaches to the modeling of language patterns that include generative theory, functional grammar, corpus-driven models (e.g., pattern grammar and frame semantics), and construction grammar. The most impressive discussion in this part (or even in the entire book) deals with the tricky issue of identification of formulaic language. Interested readers will find that a much wider range of approaches (including frequency, phonology, form, spelling, and intuition) and rationales are proposed as compared to those in Wray (Reference Wray2002), and, most importantly, those identification methods are closely tied to specific research purposes. For example, a language teacher’s approach to identifying a formulaic sequence may be greatly different from that of a corpus linguist.
Part 3 reports a series of empirical studies previously published by Wray and her co-authors, which cover factual description of several commercial products. The text and sign support assistant (TESSA), for example, is a translation system designed for basic counter transactions at the post office (chapter 10). Talk aid using preloaded knowledge (TALK) is a software program that uses a voice synthesizer to relay text typed by individuals who cannot physically articulate language (chapter 11).
Expansive as its scope of topics is, this book’s usefulness might be limited because none of the chapters directly addresses classroom instruction of formulaic language or the implications of formulaic language research in compiling dictionaries for second language learners and curriculum development (although chapters 12 and 13 address formulaic language learning by beginning and advanced learners). In sum, this volume addresses the complexities in formulaic language research. It is a fun book to read, with fascinating examples from Harry Potter’s spells to flag signals and military bugle calls, from cryptical crosswords to stage actor’s script, from Maoist slogans in China’s cultural revolution to language teaching in Papua New Guinea. This book is full of the author’s genuine thinking and inventive proposals. The intertwining of theoretical discussion and practical implications makes it a must-read for those who are interested in formulaic language.
