This book "fills the unquestionable need for a comprehensive and up-to-date handbook on the fast-developing field of pragmatics" and "includes contributions from many of the principal figures in a wide variety of fields of pragmatic research as well as some up-and-coming pragmatists."
Review of Speech Rate, Pause and Sociolinguistic Variation
“Speech Rate, Pause, and Sociolinguistic Variation: Studies in Corpus Sociophonetics” presents a series of corpus-based studies of speech rate and pause. The studies look at individual speaker differences, but are particularly focused on the influence of social variables such as region and gender. In addition, the author devotes considerable attention to phonetic and sociophonetic research methodologies and to the value of corpus-based research. A website accompanies the book and includes data sets, as well as many of the statistical algorithms and tools needed to do the types of research described in the book.
The book is organized into three sections. The first section reviews the study of speech rate and pause from a psycholinguistic and a sociolinguistic perspective. The author discusses the theoretical paradigms behind the studies, the methodologies used, and their findings. A detailed introduction to the author’s own corpus-based research and methods is included as well. In the second section, the author presents a number of sociophonetic studies of speech and pause rate and discusses the merits and shortcomings of the methodologies used in each. The final section includes a broad discussion of the study of speech rate and pause as a means of understanding sociolinguistic variation.
In Chapter 1, the author provides brief overviews of the study of speech rate and pause. He discusses the ways they have been studied by sociolinguists and psycholinguists over the last 70 years. Among the discussion is the research by sociolinguists Weinreich, Labov and Herzog (1968), which demonstrated structured homogeneity. Kendall also cites the work of psycholinguists Sumner and Samuel (2009) on dialect, who also concluded that individuals vary their pronunciation in different contexts. They therefore argued for a more complex understanding of dialect. Furthermore, the benefits of studying speech rate and pause as a means of understanding sociolinguistic variation are put forward as well. Kendall notes that speech rate and pauses are salient features of speech in all languages. In addition to the meaning conveyed by words, grammaticality, accent, speech rate and pause carry socially determined meaning in all languages.
Chapter 2 gives a comprehensive review of the literature regarding speech rate and pause. Kendall notes that social judgments are often made about a speaker based on his or her rate of speech. In the United States, for example, Southerners are perceived as slower speakers than people from other regions of the country. This slower rate of speech is often judged negatively by non-Southerners. The approaches used historically to study speech rate and pause are presented as well. Tannen (1984) saw speech rate differences as part of conversational style, while Schnoebelen (2009) and Silverstein (2003) looked at speech rate in terms of indexicality. All of this background situates the author’s own research and provides the necessary background for understanding how his work moves the field forward, both methodologically and in terms of actual findings.
In Chapter 3, Kendall introduces the reader to the corpora he has used in his sociophonetic research. He discusses the origins of the source data and the methods he has used to analyze data in different studies. One source of Kendall’s data is the Sociolinguistic Archive and Analysis Project (SLAAP), which is a joint project between the North Carolina Language and Life Project and the North Carolina State University Libraries. SLAAP is an archive for recorded speech data from numerous studies. In addition to archiving its data, SLAAP tries to standardize the management of the data and its representation. Data from the SLAAP archive can be viewed in vertical, column-based, and paragraph formats. All transcripts are organized by phonetic utterances, and transcripts are stored database tables. The other source of data Kendall discusses is Online Speech/Corpora Archive and Analysis Resource (OSCAAR), which is an archive and data management system at Northwestern University. The main difference between the two archives is that the recordings in SLAAP are mostly sociolinguistic recordings, while those in OSCAAR are primarily lab-based.
Chapter 4 gives the reader an example of Kendall’s corpus-based sociophonetic research. Kendall presents a study looking at the speech and pause rates of people from three regions of the United States (i.e. North, South, and West) using data from the OSCAAR archive. Each participant read the same text aloud to provide a sample. The measurements used and the methods of analyzing the data are presented and explained, along with graphs illustrating the results. Measurements of articulation rates by speaker and by utterance, and speaking rates by talker yielded very similar results: Westerners spoke more quickly than both Southerners and Northerners. The advantages and disadvantages of reading samples are discussed as well. Kendall acknowledges that conversational speech cannot be controlled in the same way that reading samples can, but nevertheless argues that it is preferable to control speech samples.
In Chapter 5, Kendall presents research using conversation-based corpora from SLAAP, as opposed to reading passages. The methods, analysis, and results of this research form the main portion of the book. The studies measure speech and pause rates at the utterance level, the pause level, and the speaker level. The approaches are compared to determine which method better accounts for the data. Kendall demonstrates that speech and pause rates are influenced by region, ethnicity, and gender. Southerners, in the data examined here, do indeed have longer pause rates, yielding slower speech than speakers from other regions of the United States. Kendall also discusses the cost and benefit of using very large data samples in terms of yielding better models and more accurate accounts of research findings.
In Chapter 6, Kendall continues to present results of the research described in Chapter 5. The questions of how to measure pauses, and the variability of the duration of pauses are presented as well. A variety of both speech rate and pause measurements were used and are documented with clear graphs and tables. Kendall shows how changes in sample size and measurement of different factors cause significant and insignificant differences in the models (e.g. mixed-effect vs. fixed-effect). This simultaneous presentation of findings and critiques of methodologies is consistent throughout the book.
Chapter 7 turns away from the focus on speech and pause rates as measurements of sociolinguistic variation. Instead, Kendall looks at the influence that interlocutors have on variation in the speech rates and pauses of individuals. The data presented specifically address the question of how accommodation influences variation in individual speakers’ rates of speech and their use of pauses. His research shows that speakers’ speech and pause rate vary based on their interlocutors, the gender of their interlocutors, and the number of participants in an interaction. One of the most interesting findings he reports in this chapter is that the “interviewers” in many of his samples appear to accommodate their interlocutors more than the interviewees. Here, as in previous chapters, he presents and discusses both his results and statistical methods.
In Chapter 8, the author presents a method of speech and pause rate analysis not previously mentioned in the book. This is the Henderson graph, originally used in psycholinguistic research in the 1960s. In Kendall’s words, the Henderson graph “is a representation of a speech event in which talk time is plotted on the x-axis while pause time extends along the y-axis” (192). The steepness of the slope of each turn reflects the speaker’s pauses, as well as other fluency measures. Kendall demonstrates that the use of Henderson graphs, as he uses them in this chapter, allows researchers to better measure and understand variability in speech timing and pause realizations.
Chapter 9 briefly summarizes the findings of the book and encourages further research using corpora, as well as the statistical methods presented above.
“Speech Rate, Pause, and Sociolinguistic Variation: Studies in Corpus Sociophonetics” will be useful to many groups of researchers. For those who want to understand the social factors influencing speech rate and pauses, several studies are presented which show how their variability can indeed be explained by social factors, such as gender and region of residency. For readers who want a better understanding of the statistics involved in this research, Kendall gives very clear explanations of which statistics he uses and why. Perhaps the biggest beneficiaries will be readers who seek to understand the value of large corpora in sociolinguistic research on variation. Kendall makes a strong argument in favor of corpus based research, in general, and demonstrates clearly how he uses large corpora in the study of sociophonetics, in particular. Throughout the book, Kendall gives the reader a history of statistical methods used in sociolinguistic variation research and presents new statistical approaches as well. The ways in which fixed and mixed effect regression modeling enable us to explain and understand the data are clearly shown. He shows how advances in statistical methodologies and computer access have made these methods accessible to all researchers. Readers working in other areas of sociolinguistic research, as well as psycholinguistic research, can benefit from the clear and straightforward explanations given here.
For the reader who wishes to learn more about how to study phonetics and sociophonetics, the author describes a variety of methodologies. Together with the accompanying website, the reader is presented with very detailed technical explanations of how speech rate and pauses are measured and analyzed. Whenever Kendall gives these technical explanations, he also provides a discussion of how and why these methods are useful and how they complement or contradict each other. He offers guidelines for readers in order to help them decide how to select methods that will best suit their purposes.
In providing a historical review of research on sociophonetics, the reader is introduced to studies from both psycholinguists and sociolinguists. Kendall clarifies how the paradigms of both groups of researchers influence their methodologies. He also provides numerous examples of how they complement and inform each other. He gives examples of the two groups’ nomenclature describing the same phenomena, and also shows cases where, in fact, they are asking different questions.
Kendall makes a very strong argument throughout the book for using corpora for sociophonetic research. He demonstrates how data samples taken from different groups of researchers, for different purposes, at different times, can nevertheless be combined to give large samples that can be analyzed together. The benefits of large data samples are clear: they allow complex statistical analyses which cannot be done on smaller samples accurately or reliably. The models generated by the use of these samples are robust. Kendall shows that different methods of analysis of the corpora validate each other’s results. While he makes the argument for corpora use in the study of sociophonetics, researchers in other areas of sociolinguistics can easily see the benefits of them as well.
“Speech Rate, Pause, and Sociolinguistic Variation: Studies in Corpus Sociophonetics” will be a valuable book for students and researchers of psychology, sociology, and linguistics. Readers will come away from this book understanding the hows and the whys of social science statistics, as well as the values and limitations of corpus-based research. Finally, the link to a website with datasets and explanations of statistical methods is very helpful, as it allows the reader up-close and in-depth access to the data and methodologies. As research scientists, we want our methods and results to be transparent and replicable, and Kendall makes this possible by allowing the reader so much access to his own data. This linkage, which was not possible just a few years ago, should be a model for other writers presenting their research.
Schnoebelen, Tyler. 2009. The social meaning of tempo. Unpublished manuscript. Palo Alto, CA: Stanford University.
Silverstein, Michael. 2003. Indexical order and the dialectics of sociolinguistic life. Language and Communication 23(3-4). 193-229.
Sumner, Meghan & Arthur Samuel 2009. The effect of experience on the perception and representation of dialect variants. Journal of Memory and Language 60. 487-501.
Tannen, Deborah. 1984. Conversational Style: Analyzing talk among friends, rev. edn. New York/Oxford: Oxford University Press.
Weinreich, Uriel, William Labov, & Marvin Herzog. 1968. Empirical foundations for a theory of language change. In Winfred P. Lehmann & Yakov Malkiel (eds.). Directions for Historical Linguistics, Austin, TX: University of Texas Press. 95-195.
ABOUT THE REVIEWER:
Susan Joffe is a PhD student in Linguistics at Bar Ilan University. She has an AB in Linguistics from the Univeristy of Michigan and a MS in TESOL from the Graduate School of Education at the University of Pennsylvania. Her dissertation research focuses on the interaction of identity, motivation, and second language proficiency among English speaking immigrants to Israel. Her areas of interest include bilingualism, sociolinguistics, corpus linguistics, language impairment, and research methodologies.