30.229, Review: English; Discourse Analysis; Semantics; Text/Corpus Linguistics: Kimps (2018)
The LINGUIST List
linguist at listserv.linguistlist.org
Wed Jan 16 01:15:51 UTC 2019
LINGUIST List: Vol-30-229. Tue Jan 15 2019. ISSN: 1069 - 4875.
Subject: 30.229, Review: English; Discourse Analysis; Semantics; Text/Corpus Linguistics: Kimps (2018)
Moderator: linguist at linguistlist.org (Malgorzata E. Cavar)
Reviews: reviews at linguistlist.org (Helen Aristar-Dry, Robert Coté)
Homepage: https://linguistlist.org
Please support the LL editors and operation with a donation at:
https://funddrive.linguistlist.org/donate/
Editor for this issue: Jeremy Coburn <jecoburn at linguistlist.org>
================================================================
Date: Tue, 15 Jan 2019 20:15:20
From: Hanno Beck [hannobec at buffalo.edu]
Subject: Tag Questions in Conversation
Discuss this message:
http://linguistlist.org/pubs/reviews/get-review.cfm?subid=36435397
Book announced at http://linguistlist.org/issues/29/29-2443.html
AUTHOR: Ditte Kimps
TITLE: Tag Questions in Conversation
SUBTITLE: A typology of their interactional and stance meanings
SERIES TITLE: Studies in Corpus Linguistics 83
PUBLISHER: John Benjamins
YEAR: 2018
REVIEWER: Hanno T Beck, University at Buffalo
SUMMARY
In “Tag Questions in Conversation,” Ditte Kimps offers a major survey of the
state of the art in research on tag questions (TQs), together with an
impressive array of new contributions. The new contributions are made possible
by Kimps having built a large enough compilation of TQ corpus data to enable
statistical analysis, and her hard preparatory work certainly pays off.
For those not familiar with tag questions, below are a few examples, shown
here without extra symbols for pauses, intonation, etc.:
I think that was Sunday wasn’t it. (p. 52)
Miss Hoffman was still there was she. (p. 63)
And Othello of course calls on uh Desdemona’s father doesn’t he and then he
tells her the story of his life. (p. 147)
The tag question might appear sentence-finally (1, 2) or medially (3), and
might include negation (1, 3); but these are just a few of the most
superficial properties of TQs. Why do people use tag questions? What is the
speaker trying to accomplish? Like so many innocent linguistic questions,
these lead us into an entire world of tag question categories, properties, and
justifications.
Ditte Kimps, finding the existing corpus landscape insufficient for a deep
study of TQs, decided to piece together her own analytical domain, bringing
together three corpora and extracting relevant data from them. The result is a
large collection (n=1,452) of examples of TQs spoken in the English of the
region around London.
And even beyond this – most researchers who approach a phenomenon like TQs
become quickly aware of the major role played by prosody. Prosodic factors are
too important to omit, so Kimps insisted on working with only those TQ
examples for which recordings were available.
Other researchers have developed many criteria for classifying TQs; what is
new in this book? We can point to three things. One, thanks to the size of her
data set of tag question examples, Kimps can base her findings and claims on
firmer footing than mere isolated examples and/or introspective evidence. Two,
Kimps takes prosody seriously. We all know intuitively that intonation and
other prosodic features make a difference in tag questions, but to give full
inclusion to these auditory factors in a corpus is no small achievement.
Three, Kimps is not using a single lens through which to view/analyze TQ
phenomena, but regards the phenomena through two different perspectives, each
with a chapter of its own (Chapters Five and Six) – we are, in effect, passing
through the data twice instead of just once.
These choices, taken together, result in a powerful book that gives the reader
a better view of the overall TQ landscape while moving our understanding
forward.
In Chapter One, “Introduction,” Kimps introduces tag questions, outlines her
plan for the book, and describes her overall strategy.
Chapter Two, “State of the art and aims of study,” provides a sort of survey
of the existing tag question literature. The body of literature on this
subject is so large that it would not be practical to describe each previous
study individually. As an alternative, Kimps collects the basic parameters of
other studies – their source(s) of data, size of data set, language studied,
and what sort of features and functions of TQs were examined. This becomes a
multi-page table (pp. 8-13). The result, for the reader, is a unique
bird’s-eye view of what linguists have been working on in the TQ space – and
at the same time, the extensiveness of the chart serves to underscore just how
much prior work has already been invested in this topic. I, for one, had no
idea that so much attention had already been given to TQs, but this high-level
snapshot provided by Kimps corrected that impression in a hurry.
The author goes on to outline the “typologies” that have been proposed in the
literature, whether rooted in interactional features or stance-based features,
and notes still other types of study that have been undertaken at some point.
In Chapter Three, “Corpora, data and methodology,” Kimps describes in detail
the three corpora from which she built her data set, as well as the procedures
that she followed. This information will be invaluable for anyone seeking to
build a data set of their own.
Tag questions are not so frequent that examples can be easily collected.
Moreover, they appear most often in informal contexts, such as telephone
conversations, and less often in available text sources such as books,
magazines and newspapers. This means that extra effort is required if one
wants to assemble a large, clean set of TQ examples that can be studied and
analyzed. If one insists on sound recordings for every example, as Kimps does,
then the bar is set that much higher.
Chapter Four, “TQ properties,” looks at the various structural, contextual and
prosodic properties that have been proposed for tag questions, and the likely
significance of these properties. The specialized data set that Kimps built
enables her to scrutinize some of these proposals and claims more sharply than
has previously been possible.
The next two chapters form a vital pair. Here is where the author takes us on
two separate tours of the TQ world, the first from the perspective of “speech
functions” and the second from the perspective of “stance”. Indeed, the book’s
subtitle is “A typology of their interactional and stance meanings” – Kimps is
emphasizing that we can learn the most about TQs by taking multiple passes
through the data.
In Chapter Five, “Speech functions,” Kimps puts her data set to work to
examine what are the purposes for which TQs are used, and which TQ properties
are associated with which of these purposes? Tag questions that make offers,
requests, questions, demands, all sorts of attested types are found, tallied
and analyzed with respect to formal and prosodic properties.
In Chapter Six, “Stance typology of TQs,” Kimps separates stance as a distinct
topic apart from the speech functions examined in Chapter Five. What sorts of
things are related to stance? We can imagine a conversation as consisting of
two channels simultaneously. One is the channel of ordinary meanings and how
they are being communicated – obviously tag questions have specific roles to
play there. But there is a second channel as well, where the conversation
itself is being monitored, tended and guided – it needs to be productive for
all concerned, it should not offend or embarrass anyone, it should respect
interlocutors’ discourse goals, etc. This second channel concerns stance,
whereby interlocutors keep things on an even keel and enable all concerned to
maintain face. Tag questions have work to do at this level, helping to manage
not only what is said, but how it is to be taken.
In the middle of this chapter (pp. 130-133) a few pages are spent in
considering claims that TQs in English serve many of the same purposes as
so-called modal particles in German (Müller 2014) and other Germanic
languages. Kimps quotes quite a few researchers as noting this overlap in
function, for instance König & Gast (2012) and Diewald (2013).
There are unquestionably some overlaps; certainly tag questions in some
contexts can be used to cast doubt, to make concessions, to approve, to pat
interlocutors on the back, etc., precisely as modal particles can do. With her
large data set, Kimps can and does speak to this topic (p. 133). I draw
attention to this because it is a perfect example of a door to further
research that Kimps’ work has opened wider than ever before, and hopefully we
will see new developments in this area.
In Chapter Seven, “TQs across the three corpora,” Kimps discusses her three
source corpora and whether differences among them can tell us more about TQs.
If certain types of TQ are overrepresented or underrepresented in certain
corpora, what might that mean? Does knowing some demographic information about
the speakers involved with each corpus (age, gender, socioeconomic background,
etc.) give us further insight? This is where we might be able to detect
demographic generalizations, or to debunk myths. Lakoff (1975) famously
suggested that women use tag questions more often than men, and more often
with rising intonation; this claim has been well refuted in the years since
then (Kimps recommends Calman and Davidson (1998) for discussion of this).
Kimps’ study supports the conclusion that “the impact of gender on the choice
of TQ is minimal for stance types, and non-existent for TQ speech functions”
(p. 204).
Chapter Eight, “Conclusions and prospects for future research,” summarizes for
us the major insights gained from Kimps’ studies. The book as a whole drives
home the point that with a large data set we can pursue quantitative answers
to a large range of questions – don’t skip that – but if you were somehow
limited to reading fewer than ten pages of this book, you won’t do better than
reading Chapter Eight.
At the end of the book is a References section, an Appendix, and separate Name
and Subject Indexes. The appendix consists of eight additional tables, showing
a lot of the raw, absolute counts for various measures (for example, Table 54
shows “absolute counts of prosodic properties per speech function”).
EVALUATION
This book delivers on one of the most important promises of corpus
linguistics, which is that the use of corpora can not only give us data, but
can furnish us with new opportunities. Go almost anywhere and you might hear
an example of a tag question. But, to have a large enough collection of tag
questions all in one place, one that can then be searched, modeled, charted,
etc., is a benefit that we only enjoy thanks to corpus linguistics (and to the
labor of individuals).
Before now, some questions could hardly be asked – for instance, what role
does a speaker’s age play in the tag questions used and/or in their frequency
of use – because there simply was no sufficiently rich source of tag question
data to enable an answer. Now, however, answers can start to emerge.
Kimps knows an awful lot about tag questions, but her book is not only for
insiders. She took care to provide a generous supply of examples – literally
hundreds throughout the book. As a relative newcomer to this topic, I
appreciated those examples, and because of them was never left wondering what
was really being discussed.
I would have appreciated being able to hear a few selected samples of TQs,
particularly those that illustrated interesting prosodic patterns. For a study
that is so involved not only with structure but also intonation, perhaps the
book could have included some links to online recordings of a few illustrative
examples? As corpus studies continue to grow and more modes of source data
become available, perhaps the editors at John Benjamins will think in this
direction – audio examples, video examples, etc., as appropriate, to
supplement their books and to help bring them to life.
At the end of the book (pp. 221-222), Kimps suggests some themes for future
research. I would like to add another that seems to me very important, and
that is to build up more such data collections as hers, for additional
languages/dialects. Kimps cites a source (Tottie & Hoffmann 2006) saying that
TQs are about five times more common in the United Kingdom than in the United
States – why such a big difference? Would a collection of United States
English TQ data yield similar or different conclusions compared to the
London-region collection? Or to go beyond English, any number of other
languages that exhibit the TQ phenomenon could result in great compilations.
Kimps’ creation of her specialized data set marks a real step forward, so I
want to see additional such steps, and as soon as we have more than one such
data set, further questions can be pursued – which characteristics seem to be
invariable parts of TQs, which patterns observed to date turn out to apply
only to London English, etc.?
This is a volume in the Studies in Corpus Linguistics series published by John
Benjamins. The book is well-produced, well proofread, and well indexed, and we
ought to give credit to both the author and the publisher for the nice array
of charts and graphs in the book, many of which include color.
In any field, the presentation of data and statistics runs the risk of seeming
dull or confusing. As linguistics becomes more quantitative, this risk will
only grow. In the current book, however, the liberal use of charts and graphs
makes the presentation as interesting and clear as is reasonably possible.
Another point about data presentation, perhaps less obvious, is worth noting.
When we are glancing at a list of values obtained for some variable, what
stands out to us might be the most frequent, but what we really want to know
is the most frequent compared to its expected frequency – and that could be a
value that looks unremarkable to the eye. To make her chart displays optimally
useful, Kimps has designed most of them to include Pearson residuals.
(Technically, a Pearson residual is the observed frequency of a value, minus
its expected frequency, divided by the square root of the expected frequency.
The greater a value’s Pearson residual, the more that value departs from its
expected value.) In the majority of charts in this book, Pearson residuals are
given and ones that are large enough to be noteworthy are highlighted. This
enables Kimps to achieve a double purpose: display a range of data for several
variables, and simultaneously show the reader which data points carry the most
import.
What exactly is going on with tag questions in English? How does corpus
linguistics empower us to provide better analyses of such linguistic
phenomena? This book provides worthwhile answers and points us in the
direction of deeper understanding.
REFERENCES
Calman, Alison, & Davidson, Marilyn. (1998) The impact of gender and its
interaction with role and status on the use of tag questions in meetings.
Women in Management Review 13:19-36.
Diewald, Gabriele. (2013) “Same same but different” – Modal particles,
discourse markers and the art (and purpose) of categorization. In Discourse
Markers and Modal Particles: Categorization and Description, Degand, Cornillie
& Pietrandea, eds., pp. 19-45. Amsterdam: John Benjamins.
König, Ekkehard, & Gast, Volker. (2012) Understanding English-German
Contrasts. Berlin: Erich Schmidt.
Lakoff, Robin. (1975) Language and Woman’s Place. New York: Harper & Row.
Müller, Sonja. (2014) Modalpartikeln. Kurze Einführung in die germanistische
Linguistik – Band 17). Heidelberg: Universitätsverlag Winter.
Tottie, Gunnel, & Hoffmann, Sebastian. (2006) Tag Questions in British and
American English. Journal of English Linguistics 34:283-311.
ABOUT THE REVIEWER
Hanno Beck is a PhD student in the Department of Linguistics, The University
at Buffalo (State University of New York). He is currently teaching
linguistics and German, while pursuing research on the semantics and
pragmatics of the tough-construction.
------------------------------------------------------------------------------
***************** LINGUIST List Support *****************
Please support the LL editors and operation with a donation at:
The IU Foundation Crowd Funding site:
https://iufoundation.fundly.com/the-linguist-list
The LINGUIST List FundDrive Page:
https://funddrive.linguistlist.org/donate/
----------------------------------------------------------
LINGUIST List: Vol-30-229
----------------------------------------------------------
More information about the LINGUIST
mailing list