Over the past few decades, tools and methods of Corpus Linguistics (CL) have been increasingly adopted in sociolinguistic research. This is, in many ways, not surprising, since both CL and sociolinguistics share some common ground. Although CL is perhaps best understood as a methodological paradigm, it is similar to sociolinguistics in being primarily interested in actual language use (performance) and empirically attested evidence. The particular strength of a corpus analysis is that it is normally based on a much larger amount of data than traditional sociolinguistic research, which, in turn, can produce more reliable and generalisable results. It is mostly this advantage that contributed to the burgeoning of corpus-based sociolinguistic studies. Frigal & Hardy's book gives a comprehensive account of this growing and promising field of research.
As the title suggests, this volume is primarily intended as a guide for students, specifically upper level undergraduates and postgraduates enrolled on courses in sociolinguistics. It is divided into three major sections, each containing a few chapters. Section A, ‘Introduction to corpus-based sociolinguistics’, focuses on the key principles and tools of corpus-based research. Readers are introduced to different types of corpora and issues involved in building a corpus such as sampling, balance, and proportion. The final chapter in this section turns to the general applications of corpus tools and methods in sociolinguistics and critically evaluates the benefits and limitations of corpus work for the purpose of sociolinguistic research. Section B, ‘Survey of corpus-based sociolinguistics studies’, expands on this by discussing in great detail a range of corpus-based studies in six areas including: (i) dialects, (ii) gender, sexuality, and age, (iii) politeness and stance, (iv) workplace discourse, (v) language change, and (vi) web registers. To begin with, each chapter offers an introduction to the relevant topic and some key points raised in sociolinguistic research. Subsequently, a selection of corpus-based studies is discussed at length by focusing on tools, procedures and results, often visualised in tables and graphs. Section C, ‘Conducting corpus-based sociolinguistic studies’, is practically oriented and intends to guide the reader through the process of corpus-based research. Issues involved in the collection of different types of data and ethical considerations are discussed. These are subsequently exemplified by presenting two case studies. The first focuses on the use of corpus tools in multimodal analyses of linguistic variation, whereas the second offers practical guidelines on how to conduct corpus-based research on aspects related to diachronic change. In the last chapter, the authors discuss a selection of analytical concepts and tools, specifically keyword analysis, the Linguistic Inquiry and Word Count (LIWC) tool, and the use of data that was tagged with parts of speech (POS).
Frigal & Hardy's volume can be commended on a number of points. The authors bring together an extensive selection of corpus studies that address a wide range of sociolinguistic topics and in a variety of contexts. These are discussed in an accessible and synoptic manner. Links are made to key concepts and studies in sociolinguistics, and the contribution of CL is clearly specified. While the benefits of corpus work are emphasised throughout, the authors engage critically with the methodology and address its limitations, especially in sociolinguistic research. They rightly stress that CL has a great deal to offer for sociolinguists, but its application is, at the present moment, limited to studying language variation and change. There are many sociolinguistic areas—for example, diglossia, language attitudes, and language planning—for which a corpus-based approach seems less suitable or is simply not required. Another critical point, which the authors stress on a number of occasions, is the importance of the functional interpretation of corpus outputs. While corpus methodologies can quickly deliver quantitative results in the form of frequent lexical items or statistically significant keywords, these always need to be adequately interpreted by exploring the context and the reasons for their use. This often requires integration of other qualitative methods, for example.
Furthermore, many of the subsections finish with reflective breaks with follow-up questions that encourage the reader to recall key concepts from the section and to think about ways in which they can be applied in other research contexts. Also very helpful are the interviews with experts in the field who offer a range of practical hints and critical assessments of corpus-based research. Since web registers are becoming an important source of sociolinguistic data and a popular topic at undergraduate and postgraduate levels, I am certain students will appreciate section B6, which focuses on various aspects of blogs, Facebook, and Twitter communication. This is a welcome addition that was missing in previous similar publications (such as Baker Reference Baker2010).
There are a couple of minor limitations that need addressing, however. There appears to be an overenthusiasm regarding the use of computational tools such as Gender Genie (which does not seem to exist anymore), LIWC (Linguistic Inquiry and Word Count; Pennebaker, Chung, Ireland, Gonzales, & Booth Reference Pennebaker, Chung, Ireland, Gonzales and Booth2007), and the related sentiment analysis for the purpose of sociolinguistic research. These tools were not originally designed for systematic linguistic analyses and those who have some experience with them will know that they tend to produce rather crude or inaccurate results. This is because they are mostly based on dictionaries containing single words without taking into consideration the lexical environment that can substantially alter the prime meaning of a lexical item. For example, a sentence such as I am not happy or satisfied is considered by LIWC as having two positive words, although in this context happy and satisfied are not meant as such. Tools that were designed by or with the help of (computational) linguists, for example, GATE (Cunningham, Tablan, Roberts, & Bontcheva Reference Cunningham, Tablan, Roberts and Bontcheva2013) or WMatrix (Rayson Reference Rayson2008), address some of these limitations, and it would be useful to introduce them instead. Also, this book would benefit from a more detailed consideration of the statistical tests commonly used in socio- and corpus linguistics to help novice readers, especially undergraduate students, interpret the outputs of the cited studies. Moreover, some sections introduce many corpus concepts, tools, and examples at once, at the expense of depth. Thus, this volume would need to be used in conjunction with other textbooks in CL, for example, McEnery & Hardie (Reference McEnery and Hardie2012).
Overall, Frigal & Hardy's volume is an inspirational source of research ideas for sociolinguists interested in CL and for corpus linguists with sociolinguistic interests. It can be a useful and practical addition to other core textbooks required in sociolinguistics or corpus linguistics courses at upper undergraduate and postgraduate level.
