Reviewed by Haitao Liu, Communication University of China
In the eyes of many linguists, linguistics is moving from the arts or humanities to the cognitive or life sciences. Linguistic methods are also changing from introspection to experimentation, and modern techniques and equipment are being developed.
These new methods are helpful to quantitatively and experimentally explore questions of linguistic structure and processing. However, these techniques are relatively recent and have been developed mainly in fields separate from linguistics. Therefore, many researchers and students in linguistics may be unfamiliar with these techniques. This volume is designed to provide an introduction to such a field.
The volume includes nine chapters. In Ch. 1, ‘Linguistics and the scientific method’, David Eddington compares empirical and nonempirical approaches to linguistics. He insists that valid explanations about real-world language processing must depend on adherence to scientific methodology and that only researchers who choose the method that is standard in scientific research can make progress in the field.
Wendy Baker discusses the pros and cons of the most common quantitative methods used in sociolinguistic studies—namely, field, experimental, and corpus methods. Patrick Juola sketches some mathematical basics of common connectionist systems and presents how these systems can be applied to a variety of linguistic problems. This chapter illustrates how connectionism can be a useful linguistic method and warns about the pitfalls of blindly using this method.
Exemplar-based models of language imply that linguistic behavior is a matter of comparison between a current expression and memories for tokens of similar expressions. One or more of these tokens are chosen as the basis for deciding how to operate on, interpret, or compose the current expression. After investigating Royal Skousen’s analogical model and the memory-based models of Walter Daelemans and colleagues, Steve Chandler concludes that exemplar-based models can consistently simulate real language behavior more accurately than either the rule-based models of generative grammar theory or connectionist simulations.
Experimental phonetics uses the scientific approach of testing certain variables while controlling other potential sources of variation in order to study the many factors that influence how speech is produced or perceived. Caroline Smith outlines several methods used to understand what the properties of speech sounds are and how they work as part of a linguistic system.
In the chapter entitled ‘Chronometric psycholinguistic techniques: Timing the lexicon’, which focuses on lexical decision making and priming, Gary Libben discusses technical issues that arise in the implementation of response time experiments and presents the most popular techniques that employ response time as the main dependent variable.
Bruce L. Derwing and Roberto G. de Almeida review ‘Non-chronometric experiments in linguistics’. They argue that human language is fundamentally a psychological phenomenon and that experimental psychological methods are therefore necessary to uncover its essential nature and character. Several experimental procedures, which can be used to explore the psychological reality of linguistic units and structures across a wide range of language and language types, are also described in this chapter.
Martin Meyer introduces the problems of ‘Neuroimaging of speech and language’. His article includes three sections: (i) a sketch of theoretical and technical principles and limitations of the most prominent brain imaging techniques (PET and fMRI); (ii) an overview of neuroimaging research on comprehension and production of spoken, written, and signed languages in the past fifteen years; and (iii) a perspective on how to complement fMRI and PET studies to provide more precise information about functional and anatomical connectivity within and between remote and adjacent language-related brain sites.
Usage-based linguists present a wide range of evidence to prove that linguistic structure is largely a product of humans’ collective and accumulated experience. Gathering relevant research from synchronic and diachronic research on grammar, from cognitive psychology and cognitive modeling, and also from computational linguistics, Joyce Tang Boyland provides a cross-disciplinary explanation of why usage is so central in the mental processes and representations of human language.
This volume does not have an index; therefore, it is not an easy task to find any particular topic. Additionally, quantitative linguistics in the volume is not very consistent with Quantitative Linguistics commonly understood by the International Quantitative Linguistics Association (IQLA, www.iqla.org/).