AUTHOR: Christopher Joseph Jenks
TITLE: Transcribing talk and Interaction
SUBTITLE: Issues in the representation of communication data
PUBLISHER: John Benjamins
YEAR: 2011
Veronika Drake, Department of English, University of Wisconsin-Madison
INTRODUCTION
Christopher Joseph Jenks’ “Transcribing Talk in Interaction” consists of a
preface, six chapters, and four appendices containing a sample transcript,
different transcription conventions and a quick-start guide to transcribing. The
intended audience is post-graduate students and researchers new to
interactionist research. The threefold motivations for the book are: (i) a lack
of books on transcribing despite the considerable increase in
transcription-based research in the humanities and social sciences, (ii)
recognition that transcribing interactional data is a skill that needs to be
honed, and (iii) the need for an introductory text on transcription and
transcription issues intended for researchers and students new to interaction
based research. Jenks intends for the book to be used either in transcript-based
research or a class on analyzing spoken interaction.
SUMMARY
In chapter 1, “An introduction to transcripts of talk and interaction”, Jenks
provides an overview of what transcripts and their uses and benefits are. While
the actual recordings remain the primary data source, four main uses of
transcripts are identified: representation, assistance, dissemination, and
verification. Because spoken interaction is multifaceted, transcripts of
interaction provide additional -- often missed -- detail, such as gaze,
intonation, and restarts. Transcripts also make data available to other
researchers, and allow colleagues to verify an analysis. Another benefit is that
they can be used as empirical tools in conjunction with the recording, enabling
the researcher to recognize and analyze social actions. The author also
addresses whether transcripts can serve as accurate representations of talk and
interaction. While accuracy depends on the specific methodology used, Jenks
suggests that transcripts are, in fact, rather accurate representations of
interaction, especially when compared to what he calls intuitive data, because
by drawing on recorded data, researchers can examine a record of what actually
happened.
Chapter 2, “Theoretical issues”, covers topics such as transcripts as research
constructs, transcript variation, transcription politics, and transcription
ethics. Any stretch of talk can be interpreted and transcribed differently by
different researchers, due to their subjective predispositions, interests and
biases. For example, laughter can be understood as reprehensive or nervous at
the moment of transcribing, and the transcription of it may then show variation
across different transcribers. Jenks stresses that variation in transcription is
driven mainly by the investigatory aims of the methodology used, and he suggests
that variation may be more problematic if it occurs within a specific
methodology than if it occurs across disciplines.
An important issue in transcript-based research, Jenks points out, is
transcription politics. Jenks reviews issues concerning participants with power
differentials such as talk between doctors and patients as well as police
personnel and suspects. Citing Bucholtz (2000), Jenks notes that in such
interactions, the omission of a single word can make one participant seem nice
and caring or, conversely, rude and coercive. He further cautions that even in
ordinary talk, the way people are represented in the transcript influences how
these people will be seen, attesting to the transcriber’s power over the people
s/he is transcribing. In addition to power differentials, issues of
“vernacularization” (using IPA and/or “eye-dialect” as in “lukin” for “looking”)
and “standardization” are discussed, each giving rise to a set of potential
problems for the analysis. This chapter also covers issues such as informed
consent, confidentiality of participants, masking names and place names. Jenks
suggests using pseudonyms rather than abbreviations such as “speaker 1” or “S1”
during the transcription process so as to remember that transcription involves
real people.
In Chapter 3, “Transcribing talk and interaction”, Jenks introduces the reader
to the practicalities of the transcription process. This chapter covers software
and hardware choices, organizational and content issues such as layout, line
numbers, line breaks, spacing of documents, font type, speaker representation,
and transcription detail. He recommends standard audio file formats such as WAV
or MP3 and players that show wave forms. Because transcribing requires repeated
listening to short data segments, quality headphones and hot keys are
recommended for playback. Jenks also notes the benefit of USB foot pedals, which
free up both hands for typing.
Jenks’ suggestions concerning the organization of transcripts are based on
languages read left to right and top to bottom and chiefly concern readability
and portability of transcripts. Because portions of transcripts are often shared
with colleagues and submitted as parts of publications, using row over column
representation as well as wide margins of at least 3 cm is suggested. These
formatting tips help avoid the time-consuming task of reformatting and changing
line numbers when preparing transcripts for publication. For referencing
purposes, Jenks recommends using line numbers for the transcript as well as time
stamps, a title and a numbering system for excerpts.
Content issues in this chapter are limited to font type, speaker representation
and transcription detail. In contrast to proportional fonts such as Times New
Roman, non-proportional fonts such as Courier allow for even distribution of
characters, making it easy to align overlapping talk. Unicode fonts are a good
option for multilingual data and languages that do not use Roman alphabets.
Jenks reminds the reader that it is important to use pseudonyms to protect
speakers’ confidentiality, and that providing speaker labels often involves
information that could lead to a biased analysis. For example, in doctor-patient
interaction, using the speaker labels “doctor” and “patient” may bias the
analyst and reader of the transcript even in cases where such social roles may
not be relevant in the data. Providing an overview of five types of
transcription detail (narrative, orthographic, interactional, paralinguistic and
multimodal), Jenks concludes that the detail provided in transcripts depends on
the researcher’s interests, research goals, level of training, and methodology
used.
Chapter 4, “Transcribing interactional and paralinguistic features”, provides an
overview of three transcription conventions (i.e., the Jeffersonian system,
Santa Barbara School, Gesprächsanalytisches Transkriptionssystem), interactional
features such as turn-taking and pauses, as well as paralinguistic features such
as intonation and tempo. Jenks identifies the following three components that
are necessary for transcripts to “bear some resemblance to the data recording”
(p. 46): carefully timed pauses, turn-taking transitions, and some
paralinguistic features. These features illustrate the high level of skill and
precision used by interactants to manage talk, and in turn, accomplish social
actions. The discussion includes but is not limited to the following features:
overlapping talk, timed pauses, intonation contours, amplitude and laughter. The
chapter exemplifies the importance of each feature and the transcription
conventions used via data samples. Relevant studies illustrating the
interactional import of these features are provided.
While chapter 4 focuses on vocal features of talk, chapter 5, “Transcribing
nonverbal conduct”, deals with non-vocal conduct, which is as important as other
paralinguistic features in the organization of interaction. By transcribing
non-verbal conduct, researchers aim to further understand participants’
non-verbal behavior. Jenks provides an overview of six basic types of non-verbal
conduct (body posture, facial expressions, gestures, gaze, proximity and
actions) before he presents four methods for representing these in transcripts
(text descriptions of non-verbal conduct, video stills, drawings, and digital
renderings). Jenks recommends using visual media together with written
transcripts of spoken interaction. Using digital rendering and video stills
brings with it issues of confidentiality, though participants’ privacy and
confidentiality can be secured by digitally altering the video material (i.e.
embossing and/or blurring the images/videos). Jenks outlines three methods for
representing sequentiality of non-verbal conduct: using symbols or time stamps
to indicate when the non-verbal conduct begins and ends and sequencing of
several video stills.
In the last chapter, “Advanced issues”, Jenks moves beyond the basic
practicalities of transcribing by introducing the reader to advanced theoretical
issues (myopia, present and recall, outsourcing) and practical issues
(capitalization, apostrophes, conversational floors, translations, add-on
conventions, transcription software). The theoretical issues, Jenks shows, are
connected to involvement levels of the researchers. Over-involvement can lead to
both tunnel vision and emotional attachment, both of which can lead to a loss of
objectivity. Under-involvement can lead to outsourcing the transcription process
and the subsequent loss of the intimate knowledge needed for a sound analysis.
Jenks thus cautions the reader from outsourcing all of the transcription process
and urges the reader to strike a healthy balance between low and high levels of
involvement in the transcription process. Researchers are encouraged to actively
seek second and third opinions, in part, by presenting at data sessions and
professional gatherings.
EVALUATION
“Transcribing Talk in Interaction” is an excellent resource primarily for novice
interactional researchers as it introduces many of the foundational and
practical issues of transcription. As transcription is a core part of conducting
research on interaction, a solid grounding in the practicalities of transcribing
is crucial. A need for a compilation of the nuts and bolts of transcription work
is further evidenced by discussion threads on LangUse on some of the portability
issues Jenks lays out (i.e. margins, indentation, line numbers, etc.), even
among seasoned interactional researchers. Novices will certainly benefit from
the book as a resource in a concurrent class or interactionist research project,
as Jenks suggests. Jenks reminds the reader throughout the book that transcripts
are primarily aiding and assisting in the analytic process. That is,
transcripts, as central as they are, cannot be the sole basis for a sound
analysis of interactional data and should be used with the primary data recording.
Jenks’ book is hands-on, packed with useful suggestions about the practicalities
of transcribing, such as layout, font type, margin size. These practical
pointers are so important because they help novice transcribers avoid spending
hours trying to format transcripts and line up overlapping talk. What’s more,
these considerations are generally not included in introductory texts on
conversation analysis (CA) (Schegloff, 2007; Hutchby & Wooffitt, 2008). Other
excellent suggestions include using actual pseudonyms when transcribing so as
not to forget that transcribing involves real people, and starting to use
fictitious names from the get-go in the transcription process not only because
of confidentiality issues, but also because it will help researchers remember
using those pseudonyms at professional gatherings.
In addition to practical tips, Jenks covers several important underlying
transcription issues. Establishing the “transformational” nature of transcripts
in that we only see what the video camera is able to capture from a certain
angle, as well as the fluid nature of transcripts as they “are merely
second-hand interpretations of communicative events” (p. 4) counters the notion
that transcripts are finite products. Jenks makes clear that transcripts are
never finished but continuously refined by a transcription and research process
that is a “constantly evolving interpretative (cultural) process” (p. 4).
Finally, devoting a whole chapter rather than a subsection to non-verbal
resources and the complications of transcribing them is an added value to the
basics of transcription. It reflects not only a major trend in interaction-based
research, but offers great solutions for how to incorporate such data.
In chapter 4, Jenks includes excellent sources for further reading in his
discussion of which interactional features are crucial for transcripts.
Specifically, Jenks refers to seminal work illustrating the importance of those
interactional features, such as turn-taking, prosody and laughter. These studies
provide important avenues for further reading for novice interaction based
researchers and students, as these studies show why interaction based
researchers pay attention to the fine-grained details of talk-in-interaction.
Some additional content could have made the book even more valuable to its
intended audience. As an introduction to transcription, novices would benefit
from an availability of primary data samples -- either online or as a
supplemental data CD/DVD. That way, both the data samples and transcript
symbols used could be illustrated, making more accessible the nuances of
interactional resources such as rising and slight-rising intonation, up- and
downstep in intonation, tempo, amplitude and emphasis, all of which can only
fully be appreciated by actually listening to examples. This is especially true
considering Jenks’ emphasis on the crucial role of primary data and the
secondary role of transcripts in research. While I appreciate the difficulties
of making such clips available online, the benefits of such an addition would
far outweigh those difficulties. Excellent introductory transcription tutorials
are already available online, which explain symbols used in the transcription
process and then also provide excerpts as well as data clips to illustrate each
of them. For the Jeffersonian system used in CA, one such resource is the
tutorial housed on Schegloff’s homepage
(http://www.sscnet.ucla.edu/soc/faculty/schegloff/TranscriptionProject/), for
the GAT system (although the tutorial itself is only available in German), an
outstanding resource is the tutorial developed at the University of Freiburg
(http://paul.igl.uni-freiburg.de/gat-to/). These online tutorials are
freestanding resources and do not cover the practical and theoretical issues
Jenks discusses in this book. Adding such supplemental materials would have
combined both theory and praxis in one introductory resource on transcription.
Another concern is the lack of a rationale for interaction-based research.
Novice researchers and students find themselves wondering what exactly
“talk-in-interaction” is, how it differs from “talk and interaction”, what
“social actions” are and what the benefits of research focusing on naturally
occurring interactions are in contrast to other usage-based research approaches.
Even a short introduction to the rationale behind doing interactional research
(see Heritage, 1984; Hutchby & Wooffitt, 2008; Schegloff, 2007) would have
situated and framed the discussion and introduction to transcribing more
effectively. Of course, these points apply primarily when the book is read
without a concurrent engagement in transcript-based research or enrollment in
such a class.
As a text about transcribing in general rather than a specific approach to
transcription within a single discipline, the book casts a wide net, aiming to
be useful for fields as diverse as “sociology, education, anthropology,
psychology, linguistics” (1). While virtually all concepts discussed -- such as
social actions, turn-taking, overlapping talk, non-verbal conduct and prosody --
figure prominently in both CA and IL, the book avoids a clear commitment to
outlining transcription as it is done in these two fields. An approach as narrow
as CA or IL, of course, limits readership. However, by broadening the scope,
some issues discussed remain at the surface. The author cautions the reader
repeatedly that the final decision on, for example, how much detail to include
in a transcript, will depend on the precise methodology used, the intricacies
and relevancies of which cannot be addressed in this general introduction. An
example of why this approach might be problematic at some points is Jenks’
discussion of vernacularization vs. standardization. After outlining problems
with each choice, Jenks states that the two are not binary, and that some
researchers will only vernacularize if this is made relevant in the data. It is
not clear what discipline or approach would use which option and why. While this
may not be the main focus of this monograph, these issues are important for
novice researchers as it aids in the appreciation of which features are
important to their respective fields. A rationale for some of the major
different interactional research approaches could have then been referenced
throughout the book, which would have helped in addressing such decisions more
comprehensively.
Despite these limitations, Jenks meets his overall goal of providing a
“comprehensive account of current introductory issued in transcribing talk and
interaction” (1). This volume equips the reader with the necessary tools needed
to embark on a first transcription. Overall then, given the aim of the book to
be used in introductory courses on research using interaction data, Jenks’ work
is a welcome, hands-on resource for how to transcribe.
REFERENCES
Bucholtz, M. (2000). “The politics of transcription”. Journal of Pragmatics, 32,
1439-1465.
Heritage, J. (1984). “Conversation Analysis”. In J. Heritage (ed.), Garfinkel
and Ethnomethodology (pp. 233-292). Cambridge: Polity Press.
Hutchby, I. & Wooffitt, R. (2008). Conversation Analysis. Cambridge: Polity Press.
Schegloff, E.A. (2007). Sequence Organization in Interaction. A primer in
Conversation Analysis. Cambridge: Cambridge University Press.
ABOUT THE REVIEWER
|
