EDITORS: Marianna Di Paolo and Malcah Yaeger-Dror TITLE: Sociophonetics SUBTITLE: A Student’s Guide PUBLISHER: Routledge (Taylor and Francis) YEAR: 2010
Seetha Jayaraman, Dhofar University, Sultanate of Oman.
The book provides students of linguistics and phonetics with useful tips on conducting research, from the basics of data collection, to methods of recording and selecting equipment appropriate to the type of data being analyzed. In other words, the book caters to the needs of researchers who are involved in interdisciplinary research on sociological aspects in the phonetic analysis of speech, especially through conversations and interviews. An acoustic analysis of such authentic data combines the techniques of acoustic analysis of natural speech and sound segments with that of voice quality for the purpose of sociophonetics.
The book has a collection of 15 research papers contributed to the Workshop on Sociophonetics in 2004. The authors are experts in the field of speech prosody.
Chapter I is an introduction to sociophonetics and to the volume by the editors, Marianna Di Paolo and Malcah Yaeger-Dror. The chapter presents the definition and use of the term 'sociophonetics,' and relates social and phonetic factors to language studies. Variation and social context are closely associated with language and provide the basic tools for language studies, viz., language acquisition and language analysis. The scope of the book extends beyond language acquisition. It addresses the social and indexical meaning of speech variation and thus uses a novel approach to speech perception and speech production. This approach looks at the findings of experimental studies as diagnostic rather than as simple analytical modeling. The articles support themselves on earlier findings and recommend different research methodologies. The editors discuss the layout of chapters, all of which provide a compilation that outlines the development of the various relevant subfields to the present.
Chapter II, also by Marianna Di Paolo and Malcah Yaeger-Dror, on field methods, discusses gathering data, creating a corpus, and reporting. It also deals with different field methods used for the collection and compilation of data. The chapter outlines the theoretical framework of sociolinguistic studies by Weinreich et al. (1968) as the basis for social variation of age, gender and region to explain speech variables. It suggests different ways of contributing to research through various methodologies of data sampling, and sampling using different sources and instrumental techniques. Some of the methodologies recommended are collecting authentic data, and specifying the goals or features being studied within the actual speech of the target community. The analysis that follows leads to accurate results because a few key factors are taken into account like speaker characteristics (in the case of interviews), the interviewer’s characteristics, planning the recording, and minimizing the observer’s paradox, since the aim is to study normal or natural speech. This approach emphasizes the need to include the interviewer’s role in the responses recorded, depending on the speaker’s style. Social setting is another aspect being considered in this approach. The importance of sampling rate and the number of subjects, depending on the statistical analysis and the group being observed, are also focused upon.
The method of 'structured elicitations,' which is found to be very effective in sociophonetic studies, is stressed (e.g. using minimal pairs, phonological and semantic disambiguation, and vowel categorization for phonological studies). A second approach is 'apparent time and real-time studies,' which assumes that there is a certain language change across generations. Selection of tokens is as important as the selection of the size of speakers. It is advised to choose tokens that are representative of the same phonological variables of a single recorded sample for a common variable being studied. Important factors that help in creating a suitable corpus and the common problems that can be faced in making the corpus from the data are briefed. Finally, methods of reporting results of a study in a systematic manner are listed.
In Chapter III, about making a field recording, Christopher Chieri discusses the important aspects of recording data, such as sampling rate, sample size, the process of digitizing and compressing data, and compatibility criteria of software.
Chapter IV, which addresses transcription, by Margaret Maclagan and Jennifer Hay, shows how to use transcription of passages and segmentation, as well as phonetic, paralinguistic, and non-linguistic factors, when undertaking acoustic analyses of sound files. Various levels of fillers like pauses and hesitations, in addition to supra-segmental features, are also detailed. Aural punctuation is transcribed through diacritics and special symbols, like pauses as a dot '.', a short dash '-', or a long dash '__', depending on the length of the pause and the rate of speech. Similarly, interruptions and overlaps are indicated by '//' in transcription. Pauses can be meaningful or they can indicate hesitation on the part of speakers. Fillers, on the other hand, are indicated by the phonetic sounds emitted and are transcribed accordingly. For instance, laughter is transcribed as [laugh]. Intonation in longer stretches of speech is transcribed with the conventional diacritics of IPA transcription for tonal and pitch movements.
Chapter IV continues by looking at ways of preserving recorded data on CDs and DF Cards and converting files to a suitable format for further research. The preliminaries of acoustic analysis are also discussed in this chapter.
Chapter V, 'Issues in using Legacy Data,' by Paulina Bounds, Naomi Palosaari, and William A. Kretzschmar Jr., familiarizes the reader with different ways of using legacy data, which are recorded in the past and stored as archival collections to be retrieved and analyzed by researchers. These are important for studying endangered languages or less-used languages and can be maintained using advanced audio-storage technology. Therefore, digitizing this type of data is a good method of storing them for future research.
In Chapter VI, 'Analyzing Stops,' Paul Foulkes, Gerard Docherty, and Mark J. Jones give useful tips on analyzing stop consonants along their acoustic parameters of voice onset time (VOT), burst duration (BD), aspiration, voicing, duration, and formant transitions from a study of spectrograms. Different approaches to acoustic measurements of stop consonants are discussed exhaustively and comprehensively.
In Chapter VII, 'Liquids,' Eleanor Lawson, Jane Stuardt-Smith, James M. Scobbie, Malcah Yaeger-Dror and Margaret Maclagan give an overview of the phonetically and phonologically complex speech sounds /l/ and /r/. In articulatory terms, the two sounds differ markedly in degree of occlusion. Acoustically, liquids are characterized by formants, antiformants (for /l/), and bandwidth and formant transitions to the following vowel. The articulatory variants of liquids, that is, the perceptive variation in the quality of liquids, are explained through spectrograms. This chapter also views earlier findings on formant patterns put forth by different phoneticians through acoustic similarities and by providing spectrographic evidence to the male/female distinction in the articulation of these sounds. Sociolinguistic and phonological variation in the articulation of /l/ and /r/ are also explained in detail.
In Chapter VIII, entitled 'Analyzing Vowels,' Marianna Di Paolo, Malcah Yaeger-Dror, and Beckford Wassink analyze vowels while taking into consideration factors like syllable structure, syllable type, and phonetic and phonological environment (i.e. preceding and following sound segments). These aspects have been studied and reported by earlier researchers (Laver, 1969, Lehiste, 1970, Kent & Read, 1995, and Ladefoged, 2001). In line with these studies, this chapter gives methods of measuring formant frequencies of vowels. The importance of measuring the first two formants (F1 and F2) in vowel identity and vowel context, and measuring these two formant trajectories to identify diphthongs is discussed and is well supported by spectrographic illustrations.
Chapter IX, 'More on Vowels: Plotting and Normalization,' by Dominic Watt, Anne Fabricius, and Tyler Kendall, focuses on the normalization procedure of vowels and plotting F1 versus F2 values for comparing the male/female distinction in the articulation of vowels. Taking into account the specific nature of vowels, acoustic space and linguistic analysis, F1-F2 are plotted as per the method established by Joos (1948) and Peterson & Barney (1952), and vowel plots are interpreted differently for different varieties of English. The difference in articulation is visible between male/female formant frequencies, which are inaudible. Two sets of F1/F2 are plotted separately for male and female speakers. The objective of normalization discussed in this chapter is to quantify the formant frequencies of vowels in order to bring out the difference in their production between speakers and speaker groups. Normalization of vowel formant frequency serves as a model of cognitive processes involved in vowel production. Overall, this chapter emphasizes experimental methods and computing values.
Chapter X, 'Analyzing Prosody: Best Practices for Analysis of Prosody,' by Malcah Yaeger-Dror and Zsuzsanna Fagyal, re-emphasizes the importance of speech prosody and the acoustic correlates of fundamental frequency, duration and intensity/amplitude and the variations of these characteristics caused by social factors like ethnicity and gender. It also gives a brief account of earlier sociolinguistic studies on intonation and the utterance level prosody of pauses and hesitations, along with pitch contours and intonation based on English. The latest works of Golato and Fagyal (2008) and Levey (2008) add to the latest findings on bilingualism and its influence on speech. Features such as duration and stress, pitch variation, pauses, hesitations and other timing factors determine the variability in speech rhythm. Sociophonetic studies on fundamental frequency and intonation, based on language contact prosody, involve conversation analysis of bilinguals (e.g. English-Spanish bilinguals in Gibraltar (Levey, 2008) and North Africans in France (Fagyal, 2010)) and use the approach of comparative prosody. While some conversation analyses adopt the view of intonation holistically, others look at specific tones (e.g. Golato and Fagyal, 2008).
Chapter XI, 'Acoustical Analysis of Voice Quality for Sociophonetic Purposes,' by John H. Esling and Jerold A. Edmondson, treats the physiological aspects of speech, mainly those dealing with voice quality and voice dynamics, reiterating the view of Abercrombie (1967) and Laver (1970), and supporting the extensive studies done by Catford (1964) on phonation types and physiology of speech. Most of the views expressed are reconfirming all the earlier findings on mechanisms involved in speech production.
In phonetic description, identifying and describing vowel quality is as important as describing voice quality. Laver (1970) refers to changes in sound quality as relating to voluntary movements of the larynx and vocal tract, while Abercrombie (1967) considers voice quality as conveying the least linguistic meaning in speech. The sociophonetic research findings discussed in this chapter about voice quality are in line with those of Catford (1964), who outlines the non-phonological function of phonation and glottal features, which are indicative of the speaker’s sex, health, social class, place of origin, etc.
Chapter XII, 'Experimental Speech Perception and Perceptual Dialectology,' by Cynthia G. Clopper, Jennifer Hay, and Bartlomiej Plichta, shifts the focus from production to perception, and the experimental methods used to interpret the results of synthesized speech samples, such as Linear Predictive Coding (LPC) analysis and other new software available like AKUSTYK.
There are many software packages available for manipulating and varying (e.g. time, pitch, etc.) pre-recorded speech samples, such as the Linear Predictive Coding analysis/resynthesis (LPC AR) method, which is based on the Source Filter Theory of speech production. Bartlomiejz Plichta’s AKUSTYK is a more recent software package (http://bartus.org) designed for sociophonetic research and is accessible globally. Its advantage is that it is very flexible and enables division and analysis/synthesis of samples in individual frames independent of each other. Thus, it allows modifications of parameters (e.g. formants, bandwidth, etc.), in addition to generating speech continua automatically.
Chapter XIII, 'Working with Children,' by Ghada Khattab and Julie Roberts, brings us to the subject of challenges faced in recording baby-talk and youngsters' speech and analyzing it using Linear Predictive Coding (LPC) and Fast Fourier Transform (FFT) spectra. Although working with children is interesting, their noisy recordings and immature vocal track make acoustic analysis difficult for research on sociophonetics and language acquisition. Grouping the samples according to age and transcription of utterances on the basis of audio recordings are challenges faced when working with children. Activities and interaction with parents or caretakers can be used to overcome reluctance on the part of children to respond to questions. Furthermore, the authors address the latest advances in technology, like Language Environment Analysis (LENA) and Digital Language Processor (DLP), designed by Infoture, which represent useful techniques for handling audio samples. Finally, the chapter also lists websites for recording technology and discusses the role of social context.
Chapter XIV, 'Ascertaining Word Classes,' by Betty S. Phillips, relates studies on social and phonetic variables extensively to grammatical word classes and factors leading to variation in terms of historical description of pronunciation in different dialects using the Centre for Lexical information (CELEX) as a valuable frequency list. This is a database on the relative frequency of lexical items in English, which is compiled based on a corpus of about 17.9 million words. In sociophonetic studies, the emphasis is both on the social and phonetic aspects of language variation. Therefore, it is important to create reliable data, associate factors which influence lexical variation, and classify word classes based on phonetic variation. CELEX is an efficient method of ascertaining the frequency of occurrence of lexical items in a given language.
Chapter XV, 'Checking for Reliability,' by Cynthia G. Clopper, discusses the reliability of acoustic measurements of segments and their auditory correlates/coding, or transcription. Phonetic transcriptions of vowels, diphthongs, and aspirated plosives and their acoustic measurements, as well as problems in checking for reliability, are discussed. A data set is considered reliable when it can be replicated with different samples or when it gives the same values when repeated with the same utterances. Reliability is also certified if the difference between the values of measurements is 0, or close to 0, with the same samples.
In Chapter XVI, 'Statistical Analysis,' Jennifer Hay deals with key concepts in statistical techniques/analysis of acoustic parameters in speech using phonological phenomena and different models like 'classification and regression trees' (CART) or 'analysis of variance' (ANOVA). Statistical technique refers to laying out the data and measurements in readable forms, such as tables, graphs, or figures, depending on the size, categories, and details of the data being analyzed (preferably in the form of spreadsheets). This method gives all the relevant information category-wise or speaker-wise (e.g. numbers, percentages) in order to compare the values at a glance and interpret the variables measured.
The book is meant to serve as a guide to students working on phonetics and phonology, in addition to experimental phoneticians, phonologists, cognitive scientists and conversation analysts. Each chapter is based on a different viewpoint and presents the results of articulatory, auditory and acoustic approaches to analyses based on different varieties/dialects of English. Although empirical work has already been done, as seen through the results presented earlier, the book presents a blend of all sociophonetic aspects of conversational English (involving the typology of voice quality of speakers like creaky, harsh, and so on).
Aside from the techniques of qualitative analysis of production and reception of sociophonetic data, the book also discusses the quantitative analysis of language in relation to social interactions.
The last four chapters give a concise review of the statistical analysis of speech versus age and social contexts. A noteworthy feature of the book, which will help readers immensely, is the list of websites referencing audio-visual analyses that support the studies cited. As the book is meant for readership by students, the short summaries at the end of Chapters VI to VIII and the exercises and notes which appear at the end of each chapter further supplement the reader’s guidance towards further reference and comprehensibility of the theoretical analyses. Acoustic measurements and graphic representations of experimental evidence, for example, those of durational values, add to the reliability of the methodologies adopted for the studies and to the validity of the results of such studies.
The guide certainly gives useful tips on techniques of analysis and a comprehensible account of research methodologies of empirical data. It is a valuable textbook for students and researchers alike.
In conclusion, the book is a rich source of information on methodologies of sociophonetic analysis, with information ranging from the basics of data collection to the most advanced computational techniques.
The editors have done a remarkable job of putting together a variety of papers on phonetic, acoustic, and statistical analyses and linking them effectively to social factors such as speaker age, purpose (i.e. normal speech or formal interviews), role, and context. This gives a new perspective to phonetic analysis for both fresh researchers and veterans in the field. It is a good reference guide for research on speech corpora encompassing a wide range of topics from collecting speech samples to computing and validating results.
Abercrombie, D. (1967). Elements of General Phonetics. Edinburgh: Edinburgh University Press.
Catford, J.C. (1964). Phonation Types. In D. Abercrombie, D.B. Fry, P.A.D. MacCarthy, N.C. Scott and J.L.M. Trim (eds.). In Honour of Daniel Jones: Papers Contributed on the Occasion of his Eightieth Birthday, 12 September 1961. London: Longmans, Green & Co. 26-37.
Fagyal, Z. (2010). Rhythm Types and the Speech of Working-Class Youth in a Banlieue of Paris: The Role of Vowel Elision and Devoicing. In D. Preston and N. Niedzielski (eds.). A Reader in Sociophonetics. Berlin: Mouton De Gruyter. 91-132.
Golato, A. and Fagyal, Z. (2008). Comparing Single and Double Sayings of the German Response Token Ja and the Role of Prosody: A Conversation Analytic Perspective. Research on Language and Social Interaction 41(3): 1- 30.
Levey, D. (2008). Language Change and Variation in Gibraltar. Amsterdam: Benjamins.
Joos, M. (1948). Acoustic Phonetics. Language 24(2). Language Monograph 23: 5-136.
Kent, R. & Read, H. (1995). Acoustic Analysis of Speech. New Delhi: AITBS Publishers.
Ladefoged, P. (2001). A Course in Phonetics, 4th Edition. Fort Worth: Harcourt.
Laver, J. (1970). The Synthesis of Components in Voice Quality. In Proceedings of the 6th International Congress of Phonetic Sciences. Prague: Publishing House of the Czechoslovak Academy of Sciences. 523-525.
Lehiste, L. (1970). Suprasegmentals. Cambridge, MA: MIT Press.
Peterson, G.E. and Barney, H.L. (1952). Control Methods Used in a Study of the Vowels. JASA 24(2): 175-84.
Weinreich, U., Labov, W., and Herzog, M. (1968). Empirical Foundations for a Theory of Language Change. In W. Lehmann and Y. Malkiel (eds.). Directions for Historical Linguistics. Austin: University of Texas Press. 95- 188.
ABOUT THE REVIEWER
ABOUT THE REVIEWER:
Dr. Seetha Jayaraman is a faculty member at Dhofar University at Salalah,
Sultanate of Oman, where she teaches English language skills to
undergraduate students. Her research interests include sociolinguistics,
comparative linguistics, articulatory and acoustic phonetics of English,
French and Indian languages, comparative phonetics, and L2 acquisition and
mother tongue influence.