Thielemann, Kosta - Approaches To Slavic Interaction

Approaches to Slavic Interaction
Dialogue Studies (DS)

Dialogue Studies takes the notion of dialogicity as central; it encompasses every type
of language use, workaday, institutional and literary. By covering the whole range of
language use, the growing field of dialogue studies comes close to pragmatics and
studies in discourse or conversation. The concept of dialogicity, however, provides
a clear methodological profile. The series aims to cross disciplinary boundaries
and considers a genuinely inter-disciplinary approach necessary for addressing the
complex phenomenon of dialogic language use. This peer reviewed series will include
monographs, thematic collections of articles, and textbooks in the relevant areas.
For an overview of all books published in this series, please see
http://benjamins.com/catalog/ds
Editor
Assistant Editor
Edda Weigand
Sebastian Feller
University of Mnster
A*STAR - Institute of High Performance Computing, Singapore
Editorial Advisory Board

Adelino Cattani
Fritjof Haft
Talbot J. Taylor
Kenneth N. Cissna
John E. Joseph
Wolfgang Teubert
Franois Cooren
Werner Kallmeyer
Linda R. Waugh
Robert T. Craig
Catherine Kerbrat-Orecchioni
Elda Weizman
Stefanie Molthagen-Schnring
Yorick Wilks
Universit di Padova
University of South Florida
Universit de Montral
University of Colorado at
Boulder
Marcelo Dascal
University of Tbingen
University of Edinburgh
University of Mannheim
Universit Lyon 2
Tel Aviv University
Hochschule fr Technik und

Wirtschaft Berlin
Valeri Demiankov
Geoffrey Sampson
Marion Grein
Masayoshi Shibatani
Russian Academy of Sciences

University of Mainz
University of Sussex
Rice University
Volume 20
Edited by Nadine Thielemann and Peter Kosta
College of William and Mary

University of Birmingham
University of Arizona
Bar Ilan University
University of Sheffield

Edited by
Nadine Thielemann
University of Hamburg
Peter Kosta
University of Potsdam
John Benjamins Publishing Company

Amsterdam/Philadelphia
TM
The paper used in this publication meets the minimum requirements of

theAmerican National Standard for Information Sciences Permanence
of Paper for Printed Library Materials, ansi z39.48-1984.
Library of Congress Cataloging-in-Publication Data

Approaches to Slavic Interaction / Edited by Nadine Thielemann and Peter Kosta.
p. cm. (Dialogue Studies, issn 1875-1792 ; v. 20)
Includes bibliographical references and index.
1. Languages in contact--Slavic countries. 2. Slavic languages--Social aspects. 3. Europe,
Eastern--Languages--Variation. 4. Conversation analysis--Social aspects.
I.Thielemann, Nadine, editor of compilation. II. Kosta, P. (Peter) editor of
compilation.
P130.52.S54A66
2013
491.80141--dc23
2013020207
isbn 978 90 272 1037 1 (Hb ; alk. paper)
isbn 978 90 272 7146 4 (Eb)
2013 John Benjamins B.V.

No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any
other means, without written permission from the publisher.
John Benjamins Publishing Co. P.O. Box 36224 1020 me Amsterdam The Netherlands
John Benjamins North America P.O. Box 27519 Philadelphia pa 19118-0519 usa
Table of contents
Contributors
Introduction and overview
Nadine Thielemann and Peter Kosta
ix
1
Part I. Multimodal, grammatical and paralinguistic resources

in talk-in-interaction
Talking out of turn: (Co)-constructing Russian conversation
Lenore A. Grenoble
17
Reanimating responsibility: The we-V2 (take-V2) double imperative

in Polish interaction
Jrg Zinken
35
Eye behavior in Russian spoken interaction and its correlation

with affirmation and negation
Elena Grishina
63
Hesitation markers in transitions within (story)telling sequences

of Russian television shows
Hanna Laitinen
85
Part II. Statistical analysis of Russian talk-in-interaction

Russian everyday utterances: The top lists and some statistics
Tatiana Sherstinova
105
Speech rate as reflection of speakers social characteristics

Svetlana B. Stepanova
117
vi
Part III. Displaying and negotiating epistemic and evidential

status and evaluation in interaction
How evaluation is transferred in oral discourse in Russian
Nicole Richter
133
This is how I see it: No-prefacing in Polish

Matylda Weidner
147
How can I lie if I am telling the truth? The unbearable lightness

of being of strong and weak modals, modal adverbs and modal
particles in discourse between epistemic modality and evidentiality
Peter Kosta
167
Part IV. Facework and contextualization in interaction

From (im)politeness to humor
Irony in the face(s) of politeness: Strategic use of verbal irony
in Czech political TV debates
Jekaterina Maara
187
Parliamentary communication: The case of the Russian Gosduma

Daniel Weiss
213
Impoliteness and mock-impoliteness: A descriptive analysis

Michael Furman
237
Humor as staging an utterance

Nadine Thielemann
257
Table of contents vii
Part V. Language alternation in face-to-face interaction of bilingual families

Bilingual language use in the family environment: Evidence from
a telephone conversation between members of a community
of speakers of German descent
Veronika Ries
Russian language maintenance through bedtime story reading?
Linguistic strategies and language negotiation in Russian-French
speaking families in Switzerland
Liliane Meyer Pitton
Index
281
295
317
Contributors
Editors
Nadine Thielemann
University of Hamburg, Slavic Department
Von-Melle-Park 6
D-20146 Hamburg, Germany
Nadine.Thielemann@uni-hamburg.de
Peter Kosta
University of Potsdam, Slavic Department
Am Neuen Palais 10
Haus 1
D-14469 Potsdam, Germany
peter.kosta@uni-potsdam.de
Authors (in alphabetical order)
Michael Furman
University of Ohio, Department of Slavic and Eastern European Languages and Literatures
Hagerty Hall 400
1775 College Road
Columbus OH 43201, USA
mikedfurman@gmail.com
Lenore Grenoble
University of Chicago, Slavic Department
405 Foster Hall
1130 E. 59th St. Chicago, IL 60637, USA
grenoble@uchicago.edu
Elena Grishina
Institute of Russian Language, Russian Academy of Sciences (RAN)
Volkhonka st., 18/2,
121019 Moscow, Russia
rudi2007@yandex.ru
Peter Kosta
University of Potsdam, Slavic Department
Am Neuen Palais 10
Haus 1
D-14469 Potsdam, Germany
peter.kosta@uni-potsdam.de
Hanna Laitinen
University of Helsinki, Russian Language and Literature, Department of Modern Languages
P.O. Box 24
FI-00014 University of Helsinki, Finland
hanna.k.laitinen@helsinki.fi
Liliane Meyer-Pitton
University of Bern, Linguistics Department
Lnggassstrasse 49
CH-3000 Bern 9, Switzerland
liliane.meyer@isw.unibe.ch
Jekaterina Maara
University of Zrich, Slavic Department
Plattenstr. 43
CH-8032 Zrich, Switzerland
jemaz86@gmail.com
Nicole Richter
Viadrina University
Postfach 1786
D-15207 Frankfurt (Oder), Germany
nrichter@europa-uni.de
Veronika Ries
University of Bielefeld, Fakultt fr Linguistik und Literaturwissenschaft
Universittsstrae 25
D-33615 Bielefeld, Germany
veronika.ries@uni-bielefeld.de
Tatjana Sherstinova
Saint Petersburg State University, Faculty of Philology, Laboratory of Experimental Phonetics
Universitetskaya nab. 11
St. Petersburg 199034, Russia
sherstinova@gmail.com
Svetlana Stepanova
Saint Petersburg State University, Faculty of Philology, Laboratory of Experimental Phonetics
Universitetskaya nab. 11
St. Petersburg 199034, Russia
stsvet_2002@mail.ru
Nadine Thielemann
University of Hamburg, Slavic Department
Von-Melle-Park 6
D-20146 Hamburg, Germany
Nadine.Thielemann@uni-hamburg.de
Matylda Weidner
Univeristy of Antwerp, Departament Taalkunde
Stadscampus S.R.215
Rodestraat 14
Be-2000 Antwerp, Belgium
Matylda.Weidner@ua.ac.be
Daniel Weiss
University of Zrich, Slavic Department
Plattenstr. 43
CH-8032 Zrich, Switzerland
dawe@slav.uzh.ch
Jrg Zinken
University of Portsmouth, Psychology Department
King Henry Building
King Henry I Street
Portsmouth, Hampshire PO1 2DY, UK
Joerg.Zinken@port.ac.uk
Contributors
xi

1. Introduction
This volume is designed to acquaint the reader with current research conducted
on verbal interaction between Slavic as well as bilingual interlocutors. Most contributions in this volume try to apply standard theories and methods within the
scope of linguistic pragmatics.
If pragmatics can be considered as a relation between language systems, language as performance of these systems, the language users and the way he/she
refers to the world he/she is manipulating, how can we account for pragmatics
as a potential third factor including pragmatic behavior of humans? We strongly
believe that pragmatics understood as performance of language systems has to
include and even favor linguistic analyses of face-to-face interaction and talkin-interaction featuring prominently in contemporary research on spoken language and discourse. However, most research following current methodological
frameworks such as conversation analysis, interactional linguistics, interpretive
sociolinguistics or politeness research is based on data stemming from Western
speech communities. This volume comprises contributions applying different
modern approaches to Slavic interaction and tries to fill in this gap. It addresses
both scholars interested in face-to-face interaction and talk-in-interaction more
generally as well as linguists specializing on Slavic languages. In order to address
the former, all examples and extracts are presented in a way which allows readers
unfamiliar with Slavic languages to follow them. This can easily be done by using
standard methods of transliteration and transcription that facilitate and simplify
the understanding of Slavic (even Cyrillic) texts and transcripts, thus making
these contributions available to Western and non-Slavic countries and readers.
The contributions reflect the broad variety of issues currently dealt with in
Slavic linguistics with reference to face-to-face interaction and talk-in-interaction.
Research topics range from the use of grammatical constructions and linguistic
units in interaction to rhetorical strategies and phenomena of (im)politeness.
These issues are analyzed based on data stemming from institutional and mediated or screened interaction as well as casual face-to-face interaction of Russian,
Polish and Czech as well as bilingual interlocutors (Russian-French, Russian-German). All of which are languages and speech communities that have so far been
rather underrepresented in the linguistic analysis of interaction.
The contributions thereby adopt pragmatic as well as linguistic perspectives
on interaction and frequently take approaches connecting or even integrating
both perspectives with each other. Many contributions are inspired by conversation analysis and politeness research and some are interested in socio-pragmatic
issues. They deal with the structure of interaction and investigate how specific activities or action formats are accomplished in face-to-face interaction raising the
question about the driving force for choosing a specific format be it an action or
a genre. A substantial number of contributions follow the framework of Interactional Linguistics located at the interface of linguistics and conversation analysis.
These contributions analyze linguistic units or constructions as resources utilized
by the interlocutors and emerging in interaction in order to handle frequently occurring problems concerning the organization of face-to-face interaction.
The illocutionary force or meaning of the utterance (cf. Kosta, in press) is thus
inferred from the verbal interaction and from the context or situation.
How does the following volume differ with respect to methods and theories
conducted and applied in the work on spoken discourse of Slavic (Russian, Czech
or Polish) linguists in East Europe? Whereas the Slavic Discourse analysis goes
back to different roots and traditions, some of them rooted in dialogue analysis
(Jakubinskij), others in the analysis of spoken standard and substandard casual
varieties (e.g. e.g. Pisarkowa 1975; Luba 2003 on colloquial Polish and Zemskaja
(ed.) 1979, 1983; Zemskaja, Kitajgorodskaja and irjaev 1981; Zemskaja and
Koester-Thoma (ed.) 1995 and Lapteva 1976 on colloquial Russian), the present
volume attempts the kind of research which has ever since the 70ies been conducted in Western style conversation analysis or similar approaches (interactional
linguistics, Gesprchsanalyse, Dialoggrammatik). Even though some Eastern approaches focusing on pragmatic aspects of face-to-face interaction between Slavic interlocutors have only very slowly adopted conversation analytic ideas (e.g.
ydek-Bednarczuk 1994; Warchala 1991; Borisova 2005; Yakovleva 2004), most
Slavic dialogue studies of the past ten years or so have largely been conducted
within the framework of speech act theory and a type of discourse analysis which
is strongly influenced by speech act theory (cf. the review article by Kosta and
Thielemann 2009). Mills (ed.) (1990) volume is one of the first and view attempts
which make research following these frameworks and developing them further
accessible to a Western audience. It further presents Western approaches to spoken discourse interested in politeness, speech acts and genres based on Russian
data. The current volume presents contributions, that further depart from this
tradition and comprises contributions analyzing dialogue in terms of jointly accomplished interaction. Typical features of oral language are explained in terms of
resources serving the organization of interaction. Subsequently, it reflects prominent and current research topics as well as offers approaches that analyze Slavic
interaction primarily from a mostly Western perspective, while still offering
multiple connections for scholars working on interaction from Western speech
communities.
2. Overview of the volume

The contributions gathered in this volume deal with structures and units in Slavic
interaction situated on both the micro and macro level of interaction. Several contributions are interested in the function of linguistic units and constructions as
well as multimodal or paralinguistic cues in talk-in-interaction. Other contributions analyze discourse structures such as action sequences and rhetoric strategies
or complex units such as genres. The contributions analyze how these phenomena
are used in face-to-face interaction. Their function or meaning is derived from
their usage in interaction. All contributions observe their objects in face-to-face
interaction mostly based on authentic naturally occurring data and explain their
usage in terms of the narrow linguistic context, the wider sequential context or
the very broad social context including features of the speech event and social
characteristics of its participants. The presentation of the contributions within the
volume follows a path starting on the micro level and proceeding to the macro
level. The contributions gathered in the first three sections deal with linguistic
units and structures as well as with multimodal cues on the micro-level of interaction and reveal their function in utterances, sequences and activity types or
genres. The contributions in the last two sections deal with linguistic, pragmatic
and rhetorical strategies on the utterance level or above. These strategies are analyzed as being socially meaningful. Their function is located on the macro-level
of interaction since they indicate the relationship of the interlocutors, their social
identity or the institutional frame in which the communication takes place.
The first section (Multimodal and grammatical resources in talk-in-interaction) offers contributions predominantly following the framework of Interactional Linguistics. Interactional Linguistics (Selting and Couper-Kuhlen 2001;
. All contributions (except from Weidners) have been presented and discussed at a conference entitled Approaches to Slavic Interaction held at the University of Potsdam, Germany
and organized by the Slavic department on 1618th March, 2011.
Deppermann 2007; Ochs, Schegloff and Thompson (eds.) 1996) analyzes linguistic units and structures as resources systematically employed by interlocutors in
order to handle recurrent tasks in interaction. Interactional Linguistics is located
at the interface of conversation analysis and linguistics. Its methodological procedure reconstructs members use of a specific linguistic unit. The analysis reveals
the way in which interlocutors systematically orient to and use a specific linguistic
unit or structure. The contributions by Grenoble, Zinken, Laitinen and Grishina
are interested in the interactive meaning and usage of linguistic as well as multimodal and paralinguistic resources in Russian and Polish face-to-face interaction
and in revealing the task accomplished by a specific linguistic construction or
multimodal cue.
Grenobles contribution (Talking out of turn: (co)-constructing Russian
conversation) is devoted to interlocutors use of co-constructions in casual conversations among Russian migrants in the US. Co-constructions are syntactic
constructions started by one speaker and completed by an obligatory element
or complemented by an optional element which is provided by another interlocutor. She shows how Russian migrants casual conversations are organized in a
collaborative way due to phenomena of such shared syntax (Lerner 1991). This
contrasts with an earlier analysis conducted on material from media talk which
showed that interlocutors in institutional talk which is characterized by contesting agendas, employ co-constructions in order to direct the course of the interaction (Grenoble 2008).
Zinkens contribution (Reanimating responsibility: The we- V2 (take-V2)
double imperative in Polish interaction) is also clearly situated in the framework
of interactional linguistics. He analyzes the use of serial verb constructions in
the management of domestic tasks within Polish families and focuses on double
imperatives consisting of we + V2 (take + V2). In comparison to other imperative formats available in Polish grammar (cf. Zinken and Ogierman 2011), these
hendyadic forms are used in order to express the affordance of an action that is
already under way. Double imperatives are used when consent concerning the
required action has already been reached and when unlike other expectations,
it has not yet been conducted. Based on data from interaction within families,
Zinken empirically shows that specific grammatical constructions are chosen by
the interlocutors according to the type of social action performed.
The chapter on Eye behavior in Russian spoken interaction and its correlation with affirmation and negation by Grishina extends the scope of resources
employed in interaction in order to accomplish specific tasks. Her analysis based
on data from a multimodal corpus comprising screened Russian face-to-face interactions reveals how specific patterns of eye behavior (direction of gaze, blinking, eye closing) contribute to the contextualization of turn transition and the
segmentation of utterances. Her core interest is in the contribution of these multimodal signals to the accomplishment of affirmative and negative reactions in
interaction. She shows that verbal utterances expressing affirmation are accompanied by a different gaze behavior than those expressing negation.
Laitinens chapter (Hesitation markers in transitions within (story)telling
sequences of Russian television shows) is devoted to a paralinguistic resource
in interaction and presents research on the distribution of hesitation markers in
telling sequences taken from Russian TV talk shows. She conceives of hesitation
as a pragmatic-cognitive category since hesitation markers testify to problems
in speech production, but also play an important role in organizing interaction.
The analyzed oral narratives point to the speaker re-orienting on different levels one concerning the genre and its internal organization, the other concerning
the turn and its structural organization. Oral narratives usually consist of multiunit turns in which hesitation markers occur at specific places, namely at the beginning of a turn-constructional unit (TCU), preceding a parenthetical sequence
within a TCU and near transition relevance places. According to Laitinen, at all
these places they can be interpreted as indicators of the speaker re-orienting. Her
analysis of the distribution of hesitation markers with reference to the sequential structure of the narrative shows that these occur most frequently at specific
transitions within telling sequences at the beginning or end of the narrative (e.g.
abstract orientation/evaluation; orientation evaluation/resolution). These are
places within the genre at which the speaker re-orients and possibly may restructure or complement the narrative due to troubles concerning tellability or uptake
by the interlocutors.
These contributions reveal how Russian and Polish interlocutors systematically employ specific linguistic constructions and multimodal or paralinguistic
cues in order to handle recurrent tasks located on the micro-level of discourse.
These can include the accomplishment of a specific request format, the smooth
and supportive management of turn-taking or the signaling of dissent and consent as well as the shift from one sequential element of a genre to another. The
analyses derive these functions from the usage of the constructions under scrutiny in actual speech events and the interlocutors understanding of them disclosed
in sequential uptake. Here, Laitinens contribution can be regarded an exception,
since she treats hesitation as a display of speakers cognitive re-orientation and
refers the distribution of hesitation signals to the sequential pattern of the oral
narrative.
While Zinken and Grenoble take a qualitative approach to the linguistic constructions observed, the contributions of the next section (Statistical analysis
of Russian talk-in-interaction) favor a quantitative approach to the analysis of
linguistic units in talk-in-interaction. Interactional Linguistics starts from the
assumption that spoken language, especially talk-in-interaction is shaped by its

occurrence in turns at talk (Deppermann 2007; Fox 2007). Apart from interactivity or linearity, frequency is also considered to be a factor that has an impact on
the emergence of structures specific to talk-in-interaction (Fox 2007:300301).
Statistical analyses concerning different aspects of spoken language are able to
show which linguistic structures or units emerge from frequent use as well as
reveal which characteristics of the usage event have a quantitatively measurable impact on the emergence and frequency of specific structures. Sherstinova
and Stepanova provide statistic analysis of Russian talk-in-interaction based on
a corpus collected from face-to-face interaction from several interlocutors representing their communicative days and investigate which linguistic units and
structural features concerning prosody occur most frequently in Russian everyday talk-in-interaction.
Sherstinova (Russian everyday utterances: the top lists and some statistics)
detects the most frequent utterances in spontaneous Russian talk and describes
them in terms of utterance length as well as in terms of their duration. She lists the
most frequent Russian utterances consisting of one, two, three and four words.
Among the former there are many discourse particles and interjections facilitating interaction which testifies to the frequent need of interlocutors to connect
their utterances forwards and backwards in interaction. Sherstinovas analyses
show that utterance length in syllables affects the duration of utterances. Her results foster the assumption of two registers for Russian a dialogue register typical
of shorter utterances (up to 15 syllables) which is relatively slower (450150 sec
average utterance duration) and a monologue register characterizing longer utterances which are uttered relatively faster (150 sec average utterance duration).
Stepanova analyzes the speech rate of spontaneous Russian talk-in-interaction (Speech rate as a reflection of speakers social characteristics). Unlike
Sherstinova, she is especially interested in the impact of social factors such as
sex and age on speech rate. Her analyses reveal that Russian women speak more
slowly than Russian men and that speech rate decreases with age. Her results
further strengthen the assumption that speech rate corresponds with verbal and
communicative competence, since interlocutors with a high level of linguistic
competence speak significantly slower than those on a lower level of communicative competence. Stepanovas contribution delves even deeper than the microlevel in checking which intra-speaker factors affect the speech rate. In showing
that prosodic features such as speech rate point to the age, sex or verbal competence of a speaker she reveals that prosody can potentially serve as an index for
socialidentity.
The contributions gathered in the next section follow different methodological approaches in order to analyze linguistic, prosodic and discursive resources
utilized for Displaying and negotiating epistemic and evidential status and evaluation in interaction. Interlocutors have a wide range of means at their disposal in
order to express the certainty with which a position is held, the kind of knowledge
on which it is based and the evaluative stance which is taken relative to it. Since
interlocutors can contest the epistemic or evaluative stances taken in face-to-face
interaction, the need to negotiate them may arise as well. The contributions deal
with lexical units expressing epistemic stance as well as evidential status but also
with prosodic cues and rhetorical strategies that convey a speakers evaluative
stance. Kosta and Richter take a more general perspective on the resources and
analyze their usage in the sequential or discursive context in order to reveal which
kind of epistemic, evaluative or evidential stance is conveyed. Weidner analyzes
doctor-patient communication that represents an institutional speech event with
unequally distributed access to (especially professional) knowledge. Her analysis
therefore has a socio-pragmatic dimension as well as dealing with the way epistemic issues are negotiated in face-to-face interaction.
Richter (How evaluation is transferred in oral discourse in Russian) examines the discursive means expressing speakers evaluative stance and focuses on
prosodic and rhetorical resources through which interlocutors signal their negative or positive attitude. Unlike the other contributions in the volume, her analysis
relies on experimentally elicited data from reading tasks as well as on quasi-spontaneous elicited dialogues. Unlike the other contributions in this section, she
excludes lexical items and concentrates on prosodic and discursive means indicating evaluation. Based on the data stemming from the reading task, she shows
that a higher overall pitch level is typical of positive evaluation, while negatively
evaluated utterances are uttered at a lower f0. Yet, speakers lack of involvement
as typical of non-evaluative utterances is characterized by a still lower pitch level.
Concerning the quasi-spontaneous dialogues, interlocutors display their negative
evaluative stance by a specific responsive question format. These are questions
that do not ask for information but demonstratively ask for confirmation or for
reasons why a stance is taken, which indirectly signals speakers dissent with a
preceding evaluative statement.
Kostas contribution entitled How can I lie if I am telling the truth? The
unbearable lightness of being of strong and weak modals, modal adverbs and
modal particles in discourse between epistemic modality and evidentiality shifts
to lexical items expressing evidential and/or epistemic meaning and deals with
the interrelationship of modality and evidentiality. Using the example of the Russian particle vrode by, his analysis reveals that this lexical unit can be characterized as a portmanteau word between epistemic and evidential meaning. Based
on data from Russian face-to-face interaction, he shows that the epistemic-evidential marker vrode by allows the speaker to point to the source of information
(e.g. direct perception, hearsay) and to express his/her low degree of certainty
concerning the underlying proposition. The particle vrode by enables the speaker
to distance himself/herself from the prejacent proposition including its truth conditions which allows for its use as a signal of relative consent in question-answer
sequences. Using vrode by the respondent does not take over responsibility for the
truth of the prejacent. S/he instead displays an epistemic agnostic stance.
Weidners contribution (This is how I see it. No prefacing in Polish) further
pursues this line of research in analyzing how a discourse particle contributes to
the negotiation of epistemic issues in Polish doctor-patient interaction. In doctorpatient interaction the need to negotiate the status of knowledge either obtained
by the medical professional or by the lay patient frequently arises since the participants have different access to knowledge due to their institutional roles. Weidner
demonstrates how the use of the semantically almost vacuous Polish discourse
particle no in turn-initial position is systematically employed to confirm that the
interlocutors prior utterance conveys something obvious. However, no-prefacing
introduces utterances which provide a specific perspective on the aforementioned
due to a distinct access to the knowledge item or experience. Patients introduce
responses displaying evaluation mostly grounded in their personal experience
and view, and doctors introduce reactions reflecting their professional perspective. Weidners contribution shows how a discourse particle contributes to the
management of socio-epistemic issues in interaction within an institutionally
framed speech event.
The second part of the volume comprises sections dealing with the social
meaning of linguistic, rhetorical and pragmatic strategies in face-to-face interaction. The contributions investigate specific phenomena in their discursive and
sequential context as well as in their wider social context and take into account
their potential to display social as well as institutional identities or to manage the
interlocutors relationship. The chapters gathered in the following sections link
linguistic as well as pragmatic strategies to meanings and functions located on
the macro-level of discourse pointing to the wider social context of the interaction. The research frameworks range from conversation analysis, interpretive sociolinguistics and contextualization theory to politeness and socio-psychological
pragmatics. These serve as the means to uncover the strategies employed by interlocutors in order to display their social identity, their group affiliation or their
(dis)ability to perform in an institutionally required way.
The section entitled Facework and contextualization in interaction from
(im)politeness to humor presents chapters associated with politeness in the widest sense. The underlying concept of politeness is therefore necessarily broad and
covers the use and exploitation of linguistic as well as pragmatic strategies in order
to influence, shape and reflect the social and interpersonal relationship between
the interlocutors in a specific way. This understanding incorporates politeness

as well as impoliteness and group-specific as well as culturally distinct ways of
managing interpersonal relationships in communication. The contributions deal
with these issues in institutional as well as in private face-to-face interaction. They
investigate the linguistic as well as pragmatic and discursive strategies used to
pursue different social aims e.g. displaying a specific identity for a wider audience in front of the screen, or for the interlocutors of a smaller community of
practice. Many of the chapters of this section deal with pragmatic strategies aimed
at a non-serious or non bona fide interpretation of utterances and analyze their
accomplishment as well as their use in specific speech events, genres and social
situations.
Maaras contribution (Irony in the face(s) of politeness: Strategic use of verbal irony in Czech political TV debates) is devoted to irony, a pragmatic strategy
featuring prominently in research on politeness. Relying on Brown and Levinsons
(1987) conceptualization of politeness as the management of speakers and hearers positive and negative face, she analyzes how participants of Czech TV talk
shows employ irony in order to deal with specific institutional tasks concerning
the participants faces. She analyzes interlocutors use of irony as motivated by
institutionally constrained conditions which can also be derived from the media-specific communicative situation: In mediated interaction politicians have to
impress the audience in the studio as well as in front of the screen, they have to
attack the political opponent, represent the position of their party or office; hosts
have to control the discussion etc. Maara presents a number of ways in which
politicians as well as hosts use and react to irony in order to suffice their institutionally required role and in order to handle tasks specific to mediated interaction
for an overhearing audience. The effect of irony thereby ranges from face attack to
face enhancement. The way in which irony is employed is a further indication of a
politicians personal style and its use depends on the format of the TV show.
Weiss chapter also deals with political communication in describing the
communicative behavior within the Russian parliament (Gosduma) as an indicator of the political culture (Parliamentary Communication: the Case of the Russian Gosduma). He presents linguistic, pragmatic as well as substantial evidence
for the assumed low prestige and self-esteem of the Russian parliament and the
rough political culture, among which we also find issues involving politeness and
impoliteness. Weiss analysis is based on the official transcripts of parliamentary
debates. In comparison to those of the British or other Eastern European parliaments these transcripts undergo only a very slight editing process and preserve
many phenomena such as hesitations, traces of repair or grammatical constructions typical of oral and even colloquial Russian. Alongside the use of a rather colloquial variety he focuses on fights for the floor leading to controversial sequences
10
which emerge due to the chairs strict limitation of speaking time. Weiss further
detects the flexible use of forms of address ranging from officially preferred to
rather privately used strategies as pragmatic features typical of current parliamentary discourse in Russia. Weiss concludes from these observations that communicative patterns in the Russian parliament only deviate slightly from everyday
language use. Similar to the mediated interaction analyzed by Maara, parliamentary discourse is as well multi-layered and multi-addressed. Here Weiss especially
stresses the role of the parliaments website and the TV channel of the parliament
as devices enabling politicians to communicate with their electorate. Again, this
communicative site, too, displays many expressions of discontent with the political and parliamentarian culture on behalf of politicians as well as on behalf of
voters comments. The observed features concerning communication within the
parliament as well as directed to the people, according to Weiss, testify to its weak
status and influence compared to other European parliaments.
The next chapters in this section leave the realm of political communication
but still deal with pragmatically ambivalent strategies concerning their potential
to influence the interpersonal relationship of the interlocutors. To a certain regard they connect backwards to Maaras chapter since they focus on (potentially)
humorous episodes. Furmans contribution (Impoliteness and mock-impoliteness: A descriptive analysis) is devoted to mock impolite sequences in a Russian
reality TV format and reveals the contextualization cues preventing them from
being detrimental to the interpersonal relationships of the interlocutors. Mock
impoliteness elsewhere also termed banter (e.g. Leech 1983) is conceived of as
a non-conflictive communicative mode relying on strategies usually expressing
impoliteness and leading to conflict, i.e. impoliteness in the service of social affiliation and solidarity. Using data from the Russian TV-show dom dva that comprises casual yet mediated face-to-face interaction, Furman shows how interlocutors
jointly accomplish mock impolite sequences in interaction. Comparing mock
impolite episodes with genuinely impolite sequences, he detects the features suspending an impolite understanding of utterances containing swear words, belittling and scorning the other etc. by the interlocutor. Here, paralinguistic cues (e.g.
gesture, laughter) as well as explicit statements qualifying the utterance as a joke,
may signal the non-serious intent of an otherwise impolite utterance. Additionally, Furman also finds that situational factors such as specific activity types (e.g.
flirting) or a privileged institutional role (e.g. that of the host) licensing mock
impolite behavior may prevent the interlocutor from seriously interpreting mock
impolite utterances.
Thielemanns contribution (Humor as staging an utterance) follows this line
of research by focusing on one specific contextualization cue by which Russian
interlocutors point to their non-serious intent. Based on an analysis of conversational joking from casual face-to-face interactions, the article shows how interlocutors point to the humor-specific meta-message This is play by staging an
utterance. This covers instances in which a speaker ascribes his/her utterance to
a real, virtual or fictional source and shifts his / her position in the participation
framework (Goffman 1992). Several genres and forms of humor (e.g. irony, parody, teasing, fictional scenarios) rely on the speaker dissociating herself/himself
from such humorously rendered chunks of discourse. Since animating an avatars
utterance in order to humorously distance oneself from it does not merely point
to the meta-message this is play but is literally playing the avatars word, the
findings further reinforce the notion that staged or animated speech is an iconic
contextualization cue.
Whereupon the preceding chapters provided analyses of how (im)politeness
and humor is used in order to contextualize specific social identities and relationships or in order to perform institutionally required roles, the next section offers
analysis of language choice and code-switching phenomena as discursive, social
practices employed by bilingual interlocutors as related to their social identity in
the private sphere of the family. The contributions gathered in the next section
all deal with Interaction in bilingual families and apply a conversation analytic
methodology in order to investigate the way in which code-switching and language alternations are employed in casual interaction and reveal the discursive
functions and social aims pursued this way. The contributions focus on language
choice and code-switching within turns, sequences and activity types and view
language choice and alternation as a jointly accomplished and locally negotiated
phenomenon. Concerning the data, they both investigate code-switching and
language negotiation within the family in a linguistic situation in which the Russian language has a minority status contrasting with that of the other dominantly
used language of the social environment (German and French, respectively).
Ries chapter (Bilingual language use in the family environment evidence
from a telephone conversation by speakers from the community of people of German descent) offers a qualitative analysis of code-switching phenomena in a
telephone call within a bilingual family of German descent migrated from Russia
(Russlanddeutsche). The interlocutors now living in Germany are attested to have
an equally proficient command of German and Russian so that code-switching in
order to bridge a momentary gap in the lexicon and the like is rather scarce. Indeed, Ries analysis shows that code-switching is predominantly employed for discursive purposes and is involved in the sequential management of the phone call.
Code-switching, for example, facilitates the introduction or change of a topic, and
serves the segmentation of utterances or sets apart side sequences. Conversational
11
12
routines such as opening and closing sequences are coherently organized in one
language. Here, Ries stresses that such habitual routines are realized in the language
of the interlocutors German environment which bilinguals can hardlyelude.
This further raises the question of language maintenance within bilingual
families and the social and linguistic practices chosen by bilingual parents in order
to achieve this aim. Meyer Pittons contribution (Russian language maintenance
through bedtime story reading? Linguistic strategies and language negotiation in
Russian-French speaking families in Switzerland) is devoted to one such strategy employed by bilingual parents in order support the language socialization of
their children. Her analysis is based on a corpus comprising taped bedtime stories in Russian-French bilingual families and describes how the telling is turned
into an opportunity for language learning by the Russian-speaking parent. Taking
a conversation analytic approach Meyer Pitton clearly shows that the parentally
initiated language lesson depends on the childs uptake. The Russian-speaking
parent applies different techniques (e.g. elicited repetitions, role play) in order to
make the child react in Russian. The children, nevertheless, challenge or negotiate the favored language choice which may lead to repair sequences correcting
language choice. Meyer Pitton therefore argues that language maintenance within
the family is a joint accomplishment. The contributions in this section illustrate
that locally employed strategies such as code-switching are driven by discursive
and social factors motivating language choice and utilized in the context of specific social practices. Phenomena observable and described on the micro-level
trace back to the impact of the social context concerning the interlocutors bilingual identity.
So, the arc of suspense opened by interactional linguistic analysis of linguistic
units as resources systematically employed in order to deal with recurrent tasks
concerning the organization of interaction or the accomplishment of social actions is closed by rather socio-pragmatic analyses. The latter reveal how linguistic
and pragmatic strategies affect the way interlocutors shape their interpersonal
relationships and display and maintain a specific social identity.
Acknowledgments
The authors would like to thank Michael Furman for carefully proof-reading
the paper and correcting their English, all remaining errors are the authors
responsibility.
References
Borisova, Irina. 2005. Russkij razgovornyj dialog: struktura i dinamika. Moscow: KomKniga.
Brown, Penelope and Levinson, Stephen C. 21987. Politeness: Some Universals in Language
Usage. Cambridge: Cambridge University Press.
Deppermann, Arnulf. 2007. Grammatik und Semantik aus gesprchsanalytischer Sicht. Berlin,
New York: Mouton de Gruyter.
Fox, Barbara. 2007. Principles shaping grammatical practices: an exploration. Discourse Studies 9 (3): 299318.
Goffman, Erving. 21992. Forms of Talk. Philadelphia: University of Pennsylvania Press.
Grenoble, Lenore A. 2008. Sintaksis i sovmestnoe postroenie v ustnom russkom dialoge.
Voprosy jazykoznanija 1: 2536.
Kosta, Peter and Thielemann, Nadine. 2009. Gesprchsanalyse. In Slavic Languages. Slavische
Sprachen. An International Handbook of their Structure, their History and their Investigation. [HSK 32.1], Sebastian Kempgen et al. (eds.), 10291047. Berlin, New York: Mouton
de Gruyter.
Kosta, Peter. In print. Third Factor Relevance between Semantics, Pragmatics and Syntax.
Ms.
Lapteva, Olga A. 1976. Russkij razgovornyj sintaksis. Moscow: Nauka.
Leech, Geoffrey. 1983. Principles of Pragmatics. London, New York: Longman.
Lerner, Gene. 1991. On the syntax of sentences-in-progress. Language in Society 20 (4):
41458.
Luba, Wadysaw. 2003. Polskie gadanie. Podstawowe cechy i funkcje potocznej odmiany polszczyzny. Opole: Uniwersytet Opolski.
Mills, Margaret. (ed.). 1990. Topics in Colloquial Russian. New York: Lang.
Ochs, Elinor, Schegloff, Emanuel A. and Thompson, Sandra A. (eds.). 1996. Interaction and
grammar. Cambridge: Cambridge University Press.
Pisarkowa, Krystyna. 1975. Skadnia rozmowy telefonicznej. Wrocaw: Zakad Narodowy im.
Ossoliskich.
Selting, Margret and Couper-Kuhlen, Elisabeth. 2001. Interaktionale Linguistik. Linguistische
Berichte 187: 257287.
Warchala, Jacek. 1991. Dialog potoczny a tekst. Katowice: Uniwersytet lski.
Yakovleva, Elena. 2004. Deutsche und russische Gesprche: ein Beitrag zur interkulturellen Pragmatik. Tbingen: Niemeyer.
Zinken, Jrg and Ogiermann, Eva. 2011. How to propose an action as objectively necessary:
The case of Polish trzeba x (one has to x). Research on Language and Social Interaction
44 (3): 263287.
Zemskaja, Elena. A. (ed.). 1979. Russkaja razgovornaja re. Lingvistieskij analiz i problemy
obuenija. Moskva: Russkij Jazyk.
Zemskaja, Elena A. (ed.). 1983. Russkaja razgovornaja re. Fonetika. Morfologija. Leksika. est.
Moskva: Nauka.
Zemskaja, Elena A., Kitajgorodskaja, Margarita and irjaev, Evgenij N. 1981. Russkaja razgovornaja re. Obie voprosy. Slovoobrazovanie. Sintaksis. Moskva: Nauka.
Zemskaja, Elena A., and Koester-Thoma, Soja. (eds.). 1995. Russische Umgangssprache. Berlin:
Dieter Lenz Verlag.
ydek-Bednarczuk, Urszula. 1994. Struktura tekstu rozmowy potocznej. Katowice: Uniwersytet
lski.
13
Part I
Multimodal, grammatical and paralinguistic

resources in talk-in-interaction
Talking out of turn

(Co)-constructing Russian conversation
Lenore A. Grenoble
This paper examines the use of co-constructions in spontaneous Russian conversations. Co-constructions are found when one speaker completes another
speakers utterance, that is, a co-construction is a syntactic unit created within a
single turn construction unit but by multiple speakers. Co-constructions argue
for the existence of a shared syntax and for projection of that syntax by the first
part of the construction. They underscore the interactional and collaborative
nature of conversation: radio interviews illustrate how co-constructions can be
used by interlocutors to influence topic, and more informal conversations illustrate their use in signaling heightened involvement and solidarity.
Keywords: conversational analysis, co-constructions, turn completions, radio
interviews, Odessan Russian
1. Introduction
In this paper I investigate the use and distribution of co-constructions in Russian
conversation: those instances where one person is speaking and a different person
completes that first persons turn. In other words, there is a change of interlocutors within one turn-constructional unit (or TCU). The present study follows a
framework based upon conversation analysis, which takes as the basic unit in
conversation the turn-constructional unit (or TCU), as defined by Sacks et al.
(1974). For the purposes of the current discussion, it is relevant that the TCU
was identified in part to analyze how turn taking occurs in conversation, or more
specifically, how conversational participants know when to start or stop talking. This system is based on talk in interaction, i.e., talk which is collaboratively
. The actual definition of a TCU has come under much scrutiny since the publication of
Sacks et al. (1974); see Selting (2000) and Ford (2004) for a thorough discussion.
18
Lenore A. Grenoble
constructed by more than one speaker and, in fact, the collaborative nature of
conversation is one of their key premises. The system is designed to account for
how interlocutors change turns without any, or with minimal, overlap. Specifically, their turn-taking systems map out the rules by which talk is organized as a
series of turns; each speaker is entitled to talk until he or she reaches a point of
possible turn completion, called a transition relevance place (TRP). The current
speaker may continue talking at a TRP, or there may be a change in speakers.
(Talking may also stop, of course.) Cues for possible TRPs include prosodic factors such as falling intonation, pauses, and semantic and syntactic completion.
That said, even though syntactic units are ultimately and infinitely expandable,
they do reach a point or points of possible completion, and the turn constructional
unit is generally understood to reach a transitional relevance place at a moment of
syntactic completion. In conversation, syntactic completion occurs when an utterance, in its discourse context, can be interpreted as a complete clause, i.e., with
an overt or recoverable predicate. This definition allows answers to questions, elliptical clauses and backchannels to be considered syntactically complete (Ford
and Thompson 1996:143). While the turn is a unit in conversation, it does not
stand outside of a relationship to the rest of the discourse, both the surrounding
linguistic text and extralinguistic factors.
Co-constructions illustrate two different but interrelated properties of conversation. First, and perhaps most obviously, interlocutors share a linguistic system that entails a shared syntax; turns can be collaboratively constructed by more
than one interlocutor. The resulting conversation can be analyzed as collaboratively built, both topically and structurally. Syntax and conversational structure
share a central organizational feature, projection, a formal means to organize
projection; clear cues for syntactic projection include rules of government, constituency, adjacency and serialization (Auer 2005). In conversation, there is an
interactional projection based on knowledge about the sequencing of activities.
Strong evidence for projection is provided by co-constructions or compound turn
constructional units, i.e. units which are produced collaboratively by two or more
interlocutors (Ford 2004; Ford and Thompson 1996; Helasvuo 2004; Lerner 1991,
1996, 2004; Selting 2000).
Conversation is inherently interactional. In order to understand the use of
co-constructions, we need to examine conversation as a collaborative social act:
cultural knowledge and identity are collaboratively constructed by interlocutors (Silverstein 2005); this knowledge and identity enters into the structure of
a conversation. After examining the linguistic structure of co-constructions, I
turn to an analysis of how they are used in different interactional settings. The
study of Russian greatly enhances our understanding of co-constructions because
it is a morphologically rich language and the morphology of the first part of a
Talking out of turn
co-construction restricts what follows it, in the second, co-constructed part.

Moreover, thanks to this rich morphology word order is not used in Russian to
signal morphosyntactic relations but rather information structure. This is in stark
contrast to languages like English, where the beginning of a sentence typically
contains what must be the grammatical subject. Russian is not bound by such
word-order constraints, the verb can precede the subject, and a large number of
nouns have homophonous morphology in the nominative and accusative cases,
so they can function as subjects or direct objects. On the one hand this opens
up the possibilities for co-constructions, while on the other the morphology can
restrict them in ways not seen in some other languages.
1.1
The data
Data for the present study come from field recordings made in 2010 in Brighton
Beach, New York, and from radio interviews on Radiostancija xo Moskvy
(Radio Echo Moscow) available at http://www.echo.msk.ru/. Both are comprised of
interviews, but the interviews are of markedly different nature due to the speech
settings and the participants communicative goals. The Brighton Beach recordings were conducted in order to record oral histories (and linguistic information
at the same time) for a separate study focusing on the use of Odessan Russian.
These particular conversations were among people who all self-identified as originating from Odessa. Brighton Beach is often referred to as Little Odessa, a nickname it has earned due to the high percentage of immigrants from that city living
there now. Historically, Odessa was a stronghold of Jewish culture and is often
stereotypically portrayed as a Jewish city, as is Brighton Beach. Coincidentally,
all consultants for these recordings were Jewish. Not all considered themselves to
be speakers of Odessan Russian; many explained that they spoke Contemporary
Standard Russian (CSR) but were familiar with some of the phrases and phonology of Odessan Russian. These claims were born out in an analysis of the recordings: many speakers consistently spoke CSR but were able to provide some
meta-commentary about Odessan Russian. Further study is needed to determine
whether co-constructions in Odessan Russian differ in any way from those in
other varieties. The analysis to date does not provide any support for regional differences in co-constructions.
There were several participants present for all Brighton Beach conversations.
They were relatively informal in nature; all participants (with the exception of
myself) had known one another for years and were well-acquainted. My student
assistant is herself a member of the speech community; she had helped set up
the interviews and was present during the recording session. The setting was casual; excerpts used in this paper were taken from an hour-long recording made
19
20 Lenore A. Grenoble
outdoors, on the boardwalk. These interviews are open-ended. Although I asked

specific questions about their lives growing up, one of the primary goals was just
to get people to speak.
In contrast, Radio Echo Moscow conducts political interviews on hot topics. The radio station broadcasts the interviews and then posts transcripts. The
transcripts are not completely accurate representations of the conversations as
they are cleaned up for overall readability, but they are quite good and, with the
sound track, very easily fixed to be an accurate representation of the interview.
The acoustic quality of the broadcasts is good enough to make judgments based
on auditory perception but not really good enough for acoustic measurements.
Native speakers of Russian find usage in some of the examples that is not representative of standard Russian; this stems from the nature of the spontaneous
interviews as well as from the speakers own varieties of spoken Russian. Because
these are radio interviews, the conversational setting is relatively controlled. The
interviews range from 24 speakers, and they have clearly defined roles: in each
conversation there is one or two interviewers and one or two interviewees. Their
conversational roles are predetermined by this setting. Roughly speaking, the interviewers ask questions and direct the topic of the conversation. In general their
role is to direct the conversation which, as we see here, has an effect on how coconstructions are used in the interviews.
2. The mechanics of co-constructions

co-construction is a turn constructional unit (TCU) which is produced collaboratively by two or more participants in a conversation (see Sacks 1992:647
655; Helasvuo 2004; Lerner 1991, 1996, 2004; Ono and Thompson 1995). In
other words, Speaker A is speaking; Speaker B completes As turn, or partially
completes it, and with the first speaker ending it. In conversations with multiple
participants, more than two people can collaboratively construct the turn. This
definition of co-constructions rests on the participation of more than one interlocutor in the TCU; in this respect they are to be distinguished from what have
been called afterthoughts (Chafe 1988:6) or increments (Walker 2004:147148),
where a speaker reaches a potential TCU (pragmatically, syntactically and prosodically) and continues, adding something to that (potential) TCU, i.e., what
is added is dependent upon the previous utterance. The change in speakers is a
defining difference. The second-part component of the completions are evidence
of what Lerner (2004:226) has called affiliating utterances. The second component, by virtue not only of being contiguous to the primary component, but
more critically, syntactically dependent upon it, maintains the overall flow of the
Talking out of turn
conversation. The syntactic format of the TCU is unchanged, despite the change
in speakers, and the speaker change itself is unmarked.
There are two basic types of co-constructions: extensions and completions
(Ono and Thompson 1995). In extensions, A speaks and reaches a potential TCU;
B speaks and continues As turn. Thus the preliminary component constitutes a
complete syntactic unit and a complete TCU; the affiliating utterance extends or
lengthens this component by adding onto it, creating a new (and longer syntactic
unit). This is seen in (1):
(1) Extension, from Brighton Beach, Alla and Irina (1:131:18)
1 A Moj otec byl ranen
2 i ego vakuirovali kak ranenogo
3 i on nas zabral s soboj (.)
4 I na paroxode.
1 A
2
3
4 I
My father was wounded

and he was evacuated as a wounded [person]
and he took us with him (.)
on the steamer.
With extensions, syntactically dependent material occurs after a transition relevance place, that is, after the possible end of a turn-constructional unit, as defined by Schegloff (1996); see also Thompson and Couper-Kuhlen (2005:495).
Extensions are simply tacked onto the prior utterance. Prepositional phrases often
serve as extensions, as in (1): they do not require a specific syntactic host but can
be added as long as they are semantically and pragmatically felicitous. Another
frequent strategy is to conjoin an additional phrase or clause with a conjunction.
(Whether these should be considered extensions of the first speakers turn or a
new turn construction unit is not always clear; for this reason, such extensions are
excluded from the present discussion. They in no way affect the analysis here.)
Completions are syntactically different. The first speaker does not reach
a possible transition relevance point, and the TCU is incomplete not just semantically, but syntactically when the second speaker starts speaking. The first
component typically projects the second component but is not a typical place for
speaker transition. A clear example is in (2), where the relative pronoun kotorye
they in line 4 introduces a relative clause, which speaker B completes in line 5.
The penultimate vowel /y/ is lengthened, which could be interpreted by other
interlocutors as a hesitating device while the speaker thinks of how to complete
the clause. Note that there is no pause between this speaker change, although a
relatively long pause occurs after line 5:
21
22
Lenore A. Grenoble
(2) Completion, Boardwalk interview (Boris and Alla)

1 B Nikto ne smotrit.
2 istorieskie ne vosstanavlivajut (0.9)
3 a vosstanavlivajut paru kvartalov (1.4)
4 kotory::e:=
5 A =dajut pribyl'. (0.5)
1 B
2
3
4
5 A
Nobody is paying attention.

the historical ones they dont restore (0.9)
but they restore a couple of blocks (1.4)
which
make a profit. (0.5)
Note that both extensions and completions do not constitute turn-constructional

units in and of themselves, independent of a TCU uttered by another speaker. Significantly, morphosyntax carries across speaker changes. For example, in line 4 in
example (2), the relative pronoun kotorye which is morphologically ambiguous:
it can reference either a nominative plural subject or an accusative plural inanimate object; the co-construction must be built around this morphosyntax (and in
fact treats it as subject of the verb dajut they give).
Not all cases of completion are of this kind; in my corpus, it is more frequent
to have a co-construction where the preliminary component is less restrictive: it
projects a second component, but does not dictate its form as strictly as in examples (2). Auer (2005) argues that projection is a central organizational feature
of syntax. Rules of government, constituency structure, adjacency and serialization all operate within the syntax of a conversation to enable interlocutors to anticipate what may come next (and what cannot come next). Syntax provides a
formal mechanism for organizing projection within an ongoing conversation. At
the conversational level, the turn-taking system (as described by Sacks et al. 1974)
is a resource for projecting the next turn. For example, as an utterance can be ratified, questioned, cancelled; a question can be answered, ignored, and so on. The
turn-taking system and syntax work together in projecting possible completions
of TCUs, which have an internal structure, a syntactic, semantic and prosodic
structure that projects termination.
The morphosyntax of the first part of an utterance carries across the speaker
change into the second part of the utterance. This is one feature which distinguishes co-constructions from interruptions: the second part builds upon the first
part. A clear example is (3), where the genitive noun vlasti power in line 2 is
adnominal to the noun idiotizm idiocy, the last word in line 1:
Talking out of turn
(3) Bankrotstvo JuKOSa (27.06.2006; A. Venediktov and V. Geraenko)

1 AV Vy upotrebili v raznyx intervju takoe krepkoe slovo kak idiotizm.
2 VG vlasti.
3 AV Vlasti, idiotizm vlasti,
4 ja dumal,
5 skaete sami,
6 ili mne pridtsja podskazyvat.
7 VG A ja ne bojus.
1 AV
2 VG
3 V
4
5
6
7 VG
In various interviews you have used such a strong word as idiocy.

of power.
of power, idiocy of power,
I thought,
you would say it yourself
or I would have to suggest it.
I am not afraid.
The first line is itself a possible TCU, and the speaker concluded it with a falling,
typically sentence-final intonation. The extension of the turn is ratified by the
first speaker (Venediktov) which he repeated, first in just the genitive and then
as a full phrase (vlasti, idiotizm vlasti of power, the idiocy of power) in line 3.
Technically, this could be analyzed in one of two ways: (1) vlasti is adnominal to
the noun in the preceding line, idiotizm; or (2) vlasti is adnominal to an elided
noun (idiotizm) which is not overt in line 2. That the first interepretation is more
accurate is indicated by the fact that Venediktov repeats the bare genitive vlasti in
line 3, and then follows with a repetition of the full noun phrase idiotizm vlasti,
followed by an explicit statement that he might have had to suggest these words
to Gerasimov himself.
This section provides a brief summary of some of the structural constraints
and possibilities for co-constructions in Russian; for more detailed discussion
and examples, see Grenoble (2008). Completions provide strong evidence for
projection in conversation and in syntax. This brief overview is sufficient to understand the interactional nature of co-constructions in conversation, where each
participant contributes and determines not only the topic, but the structure of the
overall conversation. This is taken up in Section 3.
3. The interactional nature of co-constructions

Co-constructions serve a number of different communicative and social functions in conversation. Conversation is inherently interactive and collaborative; in
23
24
Lenore A. Grenoble
this respect it differs from monologues and other genres where the addressee(s)
is a recipient of the speech but not an interlocutor. But conversation does more
than communicate information: it is social interaction. Cultural knowledge and
identity are collaboratively constructed by interlocutors.
Co-constructions fulfill a number of different communicative functions.
Sometimes they fill in missing words, when the second interlocutor helps the
first find a missing word, as in (4):
(4) Razvorot (31.08.2006; Sergej Buntman, Tina Kandelaki and Petr Lanskov)
1 TK no poluaetsja
2 prosto vs: e:::
3 SB prodal (.)
4 TK prodal.
1 TK
2
3 SB
4 TK
but it turns out

he just everything e
sold (.)
sold [everything].
The first speaker (TK) begins by extending the vowel in the particle e for a full
second with sustained, level pitch in line 2. (This particle is not glossed here as
there is no equivalent in English.) The vowel in vs: is somewhat elongated; this
is the first linguistic clue of hesitation. The very long continuation of the vowel in
e occurs at a point which could not possibly be a transition relevance place; the
clause is syntactically and semantically incomplete. Finally, after a full second,
the second speaker (SB) supplies the word she appears to be searching for (prodal
sold) in line 3. The suggestion is then ratified in line 4.
This may be the most obvious use of a co-construction: here it is used to help
the first speaker find a missing word. But note that this strategy entails the addition of information by the second speaker, and therefore potentially directs the
course of the topic. Thus co-constructions can be used to prod the interlocutor
to adopt or articulate a topic, point, or point of view that might not have been
introduced otherwise. Not surprisingly, this is a relatively common strategy in the
radio interviews, where the interviewers role is to direct the course of the conversation. Radio Echo Moscow is known for having controversial discussions and
disagreements, and co-constructions are one device for manipulating the conversation. This is exemplified in example (3) where the interviewee (Geraenko)
supplies the politically charged noun vlasti of power, expanding the interviewers
own statement in line 1. Here the interviewer acknowledges that he was trying to
get the interviewee to say this, an overt recognition of how the topic of the conversation is not only collaboratively constructed but also manipulated.
Talking out of turn
Consider the following excerpt from the same interview, where Venediktov
uses the co-construction in line 3 to influence the content of the conversation:
(5) Bankrotstvo JuKOSa (27.06.2006; V. Geraenko and A. Venediktov)
1 VG I Iosif skazal
2 poalujsta, vot plan Maralla Ukraine i Belorussii
3 oni strany, to est respubliki, gde, v osnovnom, la
4 AV vojna.
5 VG vojna. i agressija.
6 Poskolku opjat e v OON krome Anglii vstupila Kanada i Avstralija
7 xotja oni byli dominiony i tuda-sjuda
1 VG
2
3
4 AV
5 VG
6
7
And Iosif said

please, heres the Marshall Plan for Ukraine and Belarus
theyre countries, that is, republics, where, primarily, there was
war.
war and aggression.
Since again besides England, Canada and Australia stepped into the UN
though they were dominions and what have you.
The verb (la went or less literally was) in line 3 projects a feminine singular
subject. In line 4 there is a completion, with the nominative noun vojna war
provided as a subject for the verb. Note that this TCU, stretching across a speaker
change, follows the norms of Russian word order topic-comment structure. When
Geraenko resumes speaking in line 5, he repeats vojna, thereby ratifying it, and
extends it with i agressija. That said, it is unlikely that this is what he had originally
planned, inasmuch as the compound subject, vojna i agressija war and aggression; in Russian la does not collocate with agressija. Rather, some other verb is
required for this to be felicitous (e.g. proisxodila agressija or imela mesto agressija,
aggression happened or took place). Thus the phrase gde la vojna a more felicitous TCU than the TCU with the extension, gde la vojna i agressija. One way
to analyze this is to say that the distance between the predicate in line 2 and the
new subject in line 4 makes this combination possible, with the rationale being
something like the intervening change in speakers and proposed subject (vojna)
makes it more acceptable. In general research shows that the farther apart the
elements are, the more likely agreement is to not occur. That said, the distance
between the subject and verb is not great, and this explanation is shaky. A better
analysis is that Venediktov actually succeeds in redirecting the topic, perhaps only
slightly, but quite specifically to war.
25
26 Lenore A. Grenoble
Such manipulation of the topic of conversation is a noticeable feature of the

radio interviews. This is even clearer where the completion is rejected by the first
interlocutor, as in (6):
(6)
1
2
3
4
5
elovek iz televizora (02 September 2006; I. Petrovskaja and K. Larina)

IP Nu, vot, daj Bog, toby programma=
KL =vernulas=
IP =ne rekonstruirovalas do takoj stepeni=
KL =toby my e ne uznali=
IP =toby my e ne uvideli.
1
2
3
4
5
IP
KL
IP
KL
IP
Well, there, God willing, may the program

return
not be reconstructed to such a point
that we dont recognize it.
that we dont see it anymore.
This is an interesting excerpt because there are multiple speaker changes across
what is one single TCU. The completion in line 2 is rejected by the first speaker,
who simply continues talking and completes line 1 in line 3 with ne rekonstruirovalas not be constructed; both lines show verb forms which are projected by the
complementizer toby so that, in order in line 1. This complementizer requires
a verb in either the infinitive form or in what is called the L-participle in Russian grammatical tradition; this L-participle is seen in the verbs in lines2 and 3,
where projected by toby in line 1, and in lines 4 and 5, projected by toby in each
of these lines. Note that the TCU was potentially completed in line 2, prosodically, syntactically and semantically; this is a TRP. Petrovskajas rejection of this
completion reopens the turn unit, inasmuch as line 3 is syntactically incomplete.
At this point Larina provides her own completion in line 4, which is again rejected
by Petrovskaja who supplies a different completion in line 5. It is an interesting
example in that it illustrates failed attempts by Larina to divert the content of the
conversation.
In sum, co-constructions in the radio interviews not only supply missing or
forgotten information but are also used to direct the topic of conversation. (They
have the added rhetorical effect of articulating a thought even if it is rejected by
the other speaker.) The boardwalk interviews are markedly different from the
Radio Echo Moscow interviews in several key ways. First of all, they differ in
terms of participants. In the radio interviews, there is one or more guest and one
or more interviewer; guests are often selected for their controversial positions and
it is the job of the interviewers to elicit lively conversation. Still, they are relative
strangers. In the boardwalk interview, the interviewees are family members and
Talking out of turn
know each other well. Moreover, the student interviewer knows them as personal
friends. Thus there are very high levels of background and shared knowledge.
The family has been asked about their time growing up in Odessa, and specifically about the use of language there. The conversation is marked by heightened
involvement, laughter, and a general spirit of collaboratively constructing not just
the conversation but the memories. There are multiple examples where the first
interlocutor (in any sequence) ratifies a co-construction supplied by a second interlocutor, as in (7):
(7) Brighton Beach, Boris and Irina (38:238:26)
1 B interesno to tak (.)
2 nainae (.)
3 I zadumyvatsja.
4 B zadumyvatsja. to interesno.
1 B
2
3 I
4 B
its interesting this way (.)

you start (.)
to think.
to think. Its interesting.
It is not the case that one speaker has a preferred role in this respect: in (6) Irina
completes Borjas turn, while in (8) Borja completes Irinas:
1 I Nu govorili normalno
2 ne bylo vot takogo specifieskogo kak vot=
3 B =kolorita=
4 I =kolorita u nas
1 I
2
3 B
4 I
Well [we] spoke normally

we didnt have that specific=
=local color=
=local color.
These are both straightforward projections, although (8) is more restricted morphosyntactically. In (7), the verb nainae begin requires a dependent infinitive, while in (8) the partial phrase takogo specifieskogo that specific requires
a genitive singular masculine or neuter noun for completion. In both excerpts
there is some hesitation before the affiliating utterance: there is a slight pause after
nainae in (7) and takogo specifieskogo is followed by hesitation markers or fillers (kak vot) in (8).
27
28
Lenore A. Grenoble
A very elaborate co-construction is seen in (9). Whereas the previous examples have demonstrated the use of completions and expansions, this example also
illustrates the use of the lexical items in one utterance for the next interlocutor to
build upon the previous speakers utterance:
1 I v osnovnom to bylo intonacija
2 kotoraja vyraala vs=
3 B =nepravilnaja intonacija russ [kogo jazyka
4 I [i rukami govorili
5 B estiku [lirovat rukami kogda ty govori=
6 I [estikuljacija=
7 =oni vsegda=
8 B =to-to dokazyvali.
1 I in general it was the intonation
2 which expressed everything=
3 B =incorrect intonation [of the Russian language
4 I [and they spoke with their hands
5 B gesticu [late with your hands when you speak=
6 I [gesticulation=
7 =they always=
8 B =were trying to prove something.
Just as in (6) there was a rapid exchange of speakers, here too the speaker role
switches back and forth from Borja to Irina. But the overall topic is collaboratively
built, unlike in (6), where the speakers are sharing morphosyntax but in some
metaphorical way running two parallel conversations. Irina begins this excerpt by
discussing the intonation which is stereotypically associated with Odessan Russian (line 1); Borja picks up on the word intonacija intonation repeating it with
expanded modifiers, i.e. nepravilnaja incorrect and russkogo jazyka of Russian
(line 3). Irina continues with a new thought, that they spoke with their hands
(line4); Borja paraphrases this with the verb gesticulate but switches grammatical subject to 2nd person singular (line 5). This is repackaged as a noun in line
6, but a noun which is morphosyntactically independent from the surrounding
discourse. It is interesting that Irinas utterance begins by overlapping with the
verb in line 5, and it appears that the first three syllables of the verb in line 5 suggest enough for her to introduce the noun, which sets a topic frame. She continues
with the beginnings of a full clause (oni vsegda they always) which Borja completes in line 8, completing not only the clause but the TCU as well. The result is
an intricately interwoven, collaboratively constructed TCU.
Talking out of turn
Reviewing the boardwalk interview in its entirety, it can be seen as a series of

social and linguistic moves. The interview begins in a format fairly standard for
this register, with me (as interviewer) asking questions which individual interlocutors answer. This is linguistically encoded as relatively straightforward adjacency pairs. All interlocutors use Contemporary Standard Russian (CSR), and the
turns initially proceed with very little overlap. The longer we speak, the more one
speaker (Borja) speaks Odessan Russian. As the conversation progresses, my own
role becomes less of an interviewer and more of audience; the conversation is directed to me as a listener but I am a less active participant. The other interlocutors
show heightened involvement, with significant overlap, interruptions and turn
completions. Example (9), which begins 38 minutes into the conversation, is a
good example of this.
Finally, a completion need not be as semantically and syntactically complicated as in examples (6) or (9). Example (10) provides a nice illustration of how
collaborative the conversation is; here it is unlikely that the speakers had ever had
prior occasion to define doktor-moktor, but they share knowledge of the culture
in which such a phrase is possible, and share enough daily information to know
how the other would interpret this:
(10) Brighton Beach, Borja and Alla (21:5322:00)
1 B Medu soboj kogda my govorili by
2 my by to ponimali (.)
3 doktor-moktor znait to priniaetsja, to:=
4 A =to ploxoj doctor
1 B
2
3
4 A

among ourselves when we would speak

we would understand this(.)
doktor-moktor means this belittles, that:=
=its a bad doctor.
{laughter}
In terms of the syntax of projection in this example, there is little of interest. The
complementizer to that in line 3 simply requires a subordinate clause; the elongated vowel may be interpreted as enough of a hesitation that Alla completes
the turn, or perhaps she completes the turn so as to be a participant in the conversation. Note that the co-constructions in the boardwalk interview could be
characterized as supportive moves in the sense that they build the conversation,
expanding on what the current interlocutor says, without in any way constituting attempts to direct or redirect the topic of the conversation. Rather, they are
interactive, participatory moves which keep all interlocutors actively engaged in
the current topic, contributing to it.
29
30
Lenore A. Grenoble
Note that this excerpt comes at a point when the interlocutors have begun to
relax and are less conscious of how they are speaking; Borjas speech in lines12
shows influence from English, translating when we would speak, we would understand into a Russian conditional (my by govorili, my by ponimali) instead of
the imperfective past (my govorili, ponimali) as would be expected in CSR. This
excerpt shows no overlap, but the completion in line 4 is an indication of the other
interlocutors engagement. Example (10) precedes (9) in the conversation. As the
conversation continued, interruptions, overlaps and co-constructions increased
in frequency, signaling heightened involvement.
4. Conclusion
Conversation is inherently interactive and collaborative. This is one fundamental
difference between conversation and monologue, and it is mistaken to construe
conversation as a sequence of alternating speaker and hearer roles. Rather, the
structure of a conversation is collaboratively built. Both the lexicon and the syntax
project, and these projections can be built up and/or manipulated. A number of
studies of conversation in a variety of languages have shown that interlocutors anticipate an upcoming TRP with remarkable accuracy; both speakers and hearers
project the end of a turn. Exactly what kind of information is used in projecting
TRPs has been a matter of debate, and all of the cues mentioned here in Section2
(intonation, pauses, and semantic and syntactic completion) have been argued to
be crucial in such projections. In an experiment designed to test these competing theories, Ruiter, Mitterer and Enfield (2006) altered Dutch stimuli to mask
intonation patterns, with lexical information intact, or to mask lexicosemantics
but with intact intonational contours. They found that removing pitch information had no influence on projection accuracy, but removing lexical content did
have a significant impact and greatly altered the ability of listeners to project an
upcoming turn end. As the discussion of co-constructions has shown, the syntax
provides a frame for the second part of a co-construction, either an extension or
a completion, but the pragmatic effect of the co-construction is to add or change
lexico-semantic information and, therefore, topic.
The data presented here have shown a number of different types of co-constructions. One basic division is into extensions and completions. The extensions
in this corpus and there are very few most typically add a PP to the prior TCU.
This is seen example (1) with na paroxode on the steamboat. This is a relatively
Talking out of turn
straightforward extension, picking up on the template in the preceding utterance.

Another common device is to simply extend the turn by using a coordinating
conjunction (i and) and attach something onto the prior utterance. This device
frees the speaker from any morphosyntactic frame established by the first utterance, as long as it is syntactically complete.
Completions differ from extensions. A simple completion is illustrated in (2),
where the first interlocutor does not complete a TCU and does not reach a TRP.
Rather, his sentence is syntactically incomplete; the second interlocutor finishes
it. Completions provide evidence of projections: the first part projects the second
part. This is obvious in example (6) where the use of toby in order to in line 1
projects the use of the L-participle in each of the attempted completions (lines 2
and 3); this preliminary part quite clearly projects the syntactic form of the second part. In other words, the preliminary part sets up clear morphosyntactic constraints. It also establishes lexico-semantic constraints, and at the same time can
be used for building completions (see example 9). In fact, it is the overall feature
of projection which enables completions.
Beyond the mechanics of co-constructions, we need to consider their interactional dynamics. Conversation is a collaborative social interaction. Cultural
knowledge and identity are jointly constructed by interlocutors. Co-constructions serve both communicative and social purposes. The two sets of data show
different interactional purposes achieved through the use of co-constructions. In
both sets of data, there is a consistent usage: they often supply a forgotten word or
string of words. But in the radio interviews, they frequently function as a device
used to (re)direct the topic of conversation. Although the discourse roles are defined by the very nature of the radio interview there is at least one interviewer
and one interviewee both are vying for a position in the conversation and both
often want to make a particular point. These interviews are platforms for political
statements, explanations, exposs and so on. In the Brighton Beach interviews,
the interview dynamic is entirely different. The interviewees are reflecting on
their lives and language; there is no sense of taking a political stance or making
a point in an argument. Here the interlocutors have extensive shared knowledge
and shared life experiences, and co-constructions serve a different social function. They are used to enhance participation and to signal heightened involvement. They show solidarity and function to align one interlocutor with another.
Despite these differences, co-constructions are collaboratively built in both sets of
data and provide strong evidence for viewing conversation as a whole as a fundamentally collaborative, interactional, communicative and social enterprise.
31
32
Lenore A. Grenoble
Acknowledgements
I would like to thank the Humanities Division at the University of Chicago for
funding the research for this project and Jessica Kantarovich for all her help in
recording and analyzing the data from the Brighton Beach interviews.
References
Auer, Peter. 2005. Projection in interaction and projection in grammar. Text 25: 736.
Chafe, Wallace. 1988. Linking intonation units in spoken English. In Clause Combining in
Grammar and Discourse, John Haiman and Sandra A. Thompson (eds.), 127. Amsterdam: John Benjamins.
Ford, Cecilia E. 2004. Contingency and units in interaction. Discourse Studies 6 (1): 2752.
Ford, Cecilia E. and Thompson, Sandra A. 1996. Interactional units in conversation: syntactic, intonational, and pragmatic resources for the management of turns. In Interaction
and Grammar, Elinor Ochs, Emanuel Schegloff, and Sandra Thompson (eds.), 136184.
Cambridge: Cambridge University Press.
Grenoble, Lenore A. 2008. Sintaksis i sovmestnoe postroenie v ustnom russkom dialoge.
Voprosy jazykoznanija 1: 2536.
Helasvuo, Marja-Liisa. 2004. Shared syntax: the grammar of co-constructions. Journal of
Pragmatics 36: 13151336.
Lerner, Gene. 1991. On the syntax of sentences-in-progress. Language in Society 20 (4):
41458.
Lerner, Gene. 1996. On the semi-permeable character of grammatical units in conversation:
Conditional entry into the turn space of another speaker. In Interaction and Grammar,
Elinor Ochs, Emanuel Schegloff, and Sandra Thompson (eds.), 238276. Cambridge:
Cambridge University Press.
Lerner, Gene. 2004. Collaborative turn sequences. In Conversation analysis. Studies from the first
generation, Gene H. Lerner (ed.), 225256. Amsterdam/Philadelphia: John Benjamins.
Ochs, Elinor, Schegloff, Emanuel A., and Thompson, Sandra A. (eds.). 1996. Appendix. Interaction and Grammar, 461465. Cambridge: Cambridge University Press.
Ono, Tsuyoshi and Thompson, Sandra A. 1995. What can conversation tell us about syntax?
In Alternative Linguistics: Descriptive and Theoretical Modes, Philip W. Davis (ed.), 213
271. Amsterdam: John Benjamins.
Ruiter, Jan P. de, Mitterer, Holger, and Enfield, Nick J. 2006. Projecting the end of a speakers
turn: A cognitive cornerstone of conversation. Language 82 (3): 515535.
Sacks, Harvey. 1992 [19671968]. Lectures on Conversation, Gail Jefferson (ed.). Oxford:
Blackwell.
Sacks, Harvey, Schegloff, Emanuel A., and Jefferson, Gail. 1974. A simplest systematics for the
organization of turn-taking for conversation. Language 50 (4): 696735.
Schegloff, Emanuel A. 1996. Issues of relevance for discourse analysis: Contingency in action,
interaction and co-participation context. In Discourse Processing: An Interdisciplinary Perspective, Eduard Hovy and Donia Scott (eds.), 335. Heidelberg: Springer.
Talking out of turn
Selting, Margret. 2000. The construction of units in conversational talk. Language in Society
29: 477517.
Silverstein, Michael. 2005. Cultural knowledge, discourse poetics, and the performance of social relations. In LACUS Forum XXXI: Interconnections, Adam Makkai, William J. Sullivan
and Arle R. Lommel (eds.), 3554. Houston, TX: Lacus.
Thompson, Sandra A. and Couper-Kuhlen, Elizabeth. 2005. The clause as a locus of grammar
and interaction. Discourse Studies 7: 481505.
Walker, Gareth. 2004. On some interactional and phonetic properties of increments to turns
in talk-in-interaction. In Sound Patterns in Interaction, Elizabeth Couper-Kuhlen and
Cecelia E. Ford (eds.), 147169. Amsterdam/Philadelphia: John Benjamins.
Appendix
Transcription conventions (from Ochs et al. 1996)
Note: following standard practice in Conversation Analysis, punctuation marks are used to
indicate intonation, not grammar.
. Falling, or final, intonation, not necessarily at the end of a sentence.
, Continuing intonation, not necessarily at a clause boundary.
A rise that is stronger than that of the continuing intonation indicated by a comma,
but not the same rise as question intonation.
Sharp fall in pitch.
= No perceptible pause between lines. Equal signs are used in pairs, one at the end of a
line and the other at the beginning of the next line,
(.) A micropause, one that is perceptible but not easily measurable, generally lasting less
than 0.2 seconds
(0.9) Numbers in parentheses indicate the duration of a pause or silence, measured in
tenths of a second
: Lengthening of the sound preceding the colon. The more colons, the longer the
sound.
[ Overlap.
33
Reanimating responsibility
The we-V2 (take-V2) double
imperative in Polish interaction
Jrg Zinken
This study analyses the use of the Polish we-V2 (take-V2) double imperative
to request here-and-now actions. The analysis is based on a collection of approximately 40 take-V2 double imperatives, which was built from a corpus of
10 hours of video recordings of everyday interactions (preparing and having
meals, playing with children, etc.) taking place in the homes of Polish families.
A sequential analysis of these data shows that the take-V2 construction is commonly selected in situations where the request recipient could be expected to already be attending to the relevant business (e.g., because they committed to this
earlier in the interaction), but isnt. By selecting the take-V2 format, the request
speaker reanimates the recipients responsibility for the matter at hand.
Keywords: Conversation Analysis, double verb constructions, imperative,
interactional sociolinguistics, morality in interaction, requests
1. Introduction
Recent work in comparative Conversation Analysis has begun to investigate the
relationship between language-specific grammatical practices and the social
action environments that sustain them (Sidnell and Enfield 2012; Zinken and
Ogiermann 2011). In this vein, the present chapter describes a grammatical
practice in the context of requesting in Polish, a practice that can be glossed as
a double verb construction (Carden and Pesetsky 1977; Weiss 1993; Wulff 2006).
Excerpts14 provide examples of double verb turns taken from a recording of a
Polish family in their home. In each excerpt, the familys mum, Bogusia, produces
a turn-constructional unit (TCU) in which the first verb is a form of the verb
36
Jrg Zinken
wzi (take), and the second verb is a transitive action verb with the same morphological marking. I will refer to this as the take-V2 construction.
In Excerpt 1, Bogusia is worrying that she has prepared too much food. In
the double verb turn (lines 1011), she announces a solution of sorts, namely her
plan to combine two dishes.
Excerpt 1. wezm zcz (take.1S combine.1S)
[PP3-2 00:25]
01
02
03
Henio:
Niepotrzebnie dwa odmiany

unnecessarily two version
You shouldnt have made
zrobia tych
made.F.2S these.GEN
two versions of this
Bogusia: Nie ale jak mogam nie dwie

no but how doul.1S not two
No but how could I not make two
04
odmiany zrobi
jak jak
versions make.INF how how
versions when Magda doesnt
05
Magda nie je
z
pieczarkami
Magda not eat.3S with mushrooms
eat it with mushrooms
06
(0.6)
07
Bogusia: za duo tego

zrobiam no ju
too much this.GEN made.1S PRT already
I made too much of it well
08
trudno
difficult
tough
09
(0.6)
10->Bogusia: no nic
to
wezm
to (.) chyba
PRT nothing then take.1S this
probably
No matter then I take combine this
. As the data will show, however, the two verbs do not have to be formulated adjacently in the
turn, and V2 can be formulated before V1 (take).
. Subtitled videos of all data discussed here are available at www.joergzinken.org. The password to access the videos is trzcina. The subtitles are simplified relative to the glosses presented in transcripts.
The take-V2 double imperative in Polish
11
12
zcz
combine.1S
I guess
Henio:
nie=nie no to
ju
zostaw
no=no
PRT this already leave.IMP
No=no then just leave it
In Excerpt 2, the double verb turn (lines 45) is a criticism by Bogusia of the fact
that Magda has laid cutlery for too many people on the table.
Excerpt 2. wzia przygotowaa (took.2S prepared.2S)
PP3-2 [2:15]
01
Bogusia:
02
03
A
ile
nas
jest trzy
but how.many us.GEN is
three
But how many are we, three
troje [nas
jest
three us.GEN is
we are three
Magda:
04->Bogusia:
[No tak
PRT yes
Well yes
No to
a
ty wzia cztery
PRT then but you took.2S four
But then but you took prepared
przygotowaa
prepared.2S
four
05
In Excerpt 3, there is a disagreement between Bogusia and her husband, Henio,

about how to deal with bald patches on the lawn. Bogusia advises Henio in her
double verb turn that what needs to happen is for Henio to sow some additional
grass seeds (lines 23).
Excerpt 3. wzi nasia (take.INF sow.INF)
[PP3-1 23:19]
01
02
Bogusia:
Wyrwna
ziemi ziemi
straighten.out.INF soil
soil.INSTR
Straighten out the soil, simply
po prostu i
wzi
na:sia
simply
and take.INF sow.on.INF
with some soil, and take sow on (there)
37
38
Jrg Zinken
dosia (.) [tej dos-=

sow.more.INF this.GEN
sow more of this
03
04
Henio:
05
Bogusia:
06
[Najlepiej pod[nie
best
lift.INF
Its best to lift (it up)
[tego
this.GEN
this
samego
rodzaju trawy.
same.GEN type.GEN grass.GEN
same type of grass
Finally, in Excerpt 4, Bogusias double verb turn (lines 56) is a directive for
Magda to join her sister, who is playing piano in the background.
Excerpt 4. we zagraj (take.IMP play.IMP)
PP3-2 [28:00]
01
Bogusia:
02
03
Ona gra
to
co
ma
co
ma
she play.3S this what has.3S what has.3S
Is she playing this piece she is
rozczyta
ten utwr tak?
read.out.INF this piece yes
supposed to read out?
Magda:
04
Nie wiem
bo
(ja umiem to
gra
not know.1S because I can.1S this play.INF
I dont now, because (I can play this
na cztery rce)
in four
hands
four-handed)
05->Bogusia:
Nie=nie=nie id
we (.) y: ty
no=no=no
go.IMP take.IMP
you
No=no=no you go take eh
06
(.) zagraj
to
co
miaa na dzisiaj
play.IMP this what had.2S on today
play what you had for today
07
(0.8)
08
09
Magda:
moja herbata
my
tea
my tea
(5.0) ((Magda drinks tea))
10
11
Bogusia:
No ju
zostaw
j i
zagraj
to
PRT already leave.IMP it and play.IMP this
OK, now leave it and play this,
ten utwr
that piece
this piece
In each of these examples, wzi (take) and another action verb are formulated as
part of one turn-constructional unit, and they seem to jointly target one action
rather than two separate actions of taking something plus doing something else.
While at least in the first three examples, the targeted action involves taking
some object (the dishes, the cutlery, the grass seeds), this is a sub-phase integral to
the target action formulated with the second verb. In Excerpt 4, it is not clear that
practicing her piano playing would involve Magda taking anything at all.
It seems then that wzi (take) in the kind of usage exemplified in these fragments has an auxiliary rather than fully lexical function, a characteristic of double verb constructions more broadly (cf. Hopper 2002). Take-V2 or take-and-V2
constructions appear to be, or have been, widespread in the languages of Europe
(Coseriu 1966), and they have received some attention from researchers, particularly with an interest in historical linguistics and grammaticalization (Brinton
1988; Coseriu 1966; Ekberg 1993; Vannebo 2003). In that literature, take-V2 constructions are sometimes described as having a (quasi-) aspectual function. For
example, Old and Middle English had such a construction, which according to
Brinton (1988) served the function of delimiting a new situation in a chain of
reported events. Similarly, Ekberg (1993) and Vannebo (2003) describe the take
component in the take-and-V2 construction in Swedish and Norwegian as having an initiating meaning. Researchers working on Russian have interpreted the
take-and-V2 construction in that language as expressing the suddenness or unexpectedness of a V2 event (cf. Weiss 2007); a meaning that is more specific, but
maybe related to the more general aspectual function of delimiting a new event.
Similarly, Aikhenvald and Dixon (2006:45) reports a take-V2 construction in Brazilian Portuguese, pegou falou ((he) took (he) spoke), with the gloss he spoke all
of a sudden.
While previous work has been based on data that provide limited information
on the context of use, it is the aim of this chapter to examine the usage of the Polish
take-V2 construction in everyday interaction. The analysis focuses on imperative
cases such as Excerpt 4, in which the we-V2 (take-V2) construction enters into
the assembly of a request for here-and-now action. It is based on video recordings
of everyday interactions, which makes it possible to examine the sequential context in which a speaker selects this double imperative to make a request.
39
40 Jrg Zinken
The Polish we-V2 double imperative has been described as adding an emotional emphasis to the proposition (Krlak and Rudnicka 2006; led 2001). led
(2001:20) describes this construction as emotionally charged, and according to
Krlak and Rudnicka (2006:141), it expresses impatience and irritation with the
addressees behavior and orders him to change the present state of affairs. The
interpretations of the take-V2 construction as having an aspectual or an expletive
function both capture some of the usage properties of this format, as the cases
discussed below will show. However, with respect to the imperative format as it is
used in the context of requesting, the sequential analysis reported here brings into
focus a moral dimension of the take-V2 construction. Speakers select the we-V2
double imperative in situations in which, firstly, there are grounds for expecting the recipient to be already attentive to the relevance now of the task at hand,
for example, because the recipient had been contributing to that task before; but,
secondly, that expectation isnt met by the other persons present conduct. In a
nutshell, the double imperative format reanimates the recipients co-responsibility
for the relevant matter.
2. Data and method

The analysis is based on a 10-hour corpus of video recordings of naturally-occurring interactions, which were made by six Polish families in their homes. Families
were provided with a video camera, and were asked to let the camera run whenever and for how long it suited them. The recordings are of various joint activities,
such as preparing and eating meals together, playing with children, or doing arts
and crafts activities. In the 10 hours of recordings, there are approximately 40
cases of requests for here-and-now actions formulated with the we-V2 double
imperative. This means that this format is not very frequent for comparison,
the same corpus of recordings contains over 60 cases in the simple imperative
of requests for objects alone (Zinken and Ogiermann in press) but that it is
nevertheless reasonably common, occurring on average twice in half an hour in
everyday situations of joint activity.
Researchers in Conversation Analysis, Interactional Sociolinguists, and related fields have investigated how the details of grammar enter into the accomplishment of particular social actions, or action formation (e.g., Hakulinen and
Selting 2005; Ochs, Schegloff, and Thompson 1996; Selting and Couper-Kuhlen
2001; Sidnell 2009). At the same time, this research has emphasized the role of social interaction as the primary environment for the development of grammatical
structure. Everyday interaction data are therefore crucial to an understanding of
the semantics of grammatical structures (e.g., Enfield 2003). The present analysis approaches the we-V2 double imperative from this perspective, attending to
the mutual constitution of social situations and grammatical structures. All data
reported here were transcribed according to the conventions established in Conversation Analysis (see Schegloff 2007 for an overview).
3. Requesting here-and-now actions with the take-V2 format

One strategy for getting a grip on the kind of sequential environment in which
speakers select this double imperative is to compare cases where this format
is used with cases in which speakers make requests using a simple imperative.
Simple imperatives are the most common grammatical format for requesting
here-and-now action in Polish, at least in the more circumscribed domain of requesting objects (Zinken and Ogiermann in press). Speakers of Polish, but also
speakers of Italian (Rossi 2012) and of British English (Wootton 1997; Zinken
and Ogiermann in press) select the simple imperative format when there are sequential grounds for the request speaker to figure that the recipient will comply
with the request. A recurrent kind of sequential ground for figuring this is that the
prospective request recipient is already involved in a joint activity to which the
requested action will contribute, and has thereby taken up a co-ownership of the
wider project (Rossi 2012). Here is an example. In Excerpt 5A, a fragment from a
family dinner, one of the toddlers, Kilian, encounters a problem with his eating:
he has put down his fork on his plate in such a way that the fork drops into the gap
between his highchair and the table, and to the floor. This attracts the attention
and concern of all three adults: Kilians mum, Ela, her partner, Tadek, and their
adult daughter Asia, who had finished her meal and got up from the table prior
to Kilians mishap. Tadek admonishes Kilian by calling his name (Kilian., line 3),
and Asia comments on Kilians self-inflicted predicament: how are you gonna
be eating now, with your fingers, right? (czym bdziesz teraz je palcem tak?,
lines 56). In what follows, three requests are addressed to Asia in simple imperative format. Ela requests that she bring Kilian a new spoon: daj mu yeczk daj
mu Asiu yeczk (give him a spoon, give him, Asia, a spoon, lines 78). Tadek
adds another request to this, for Asia to take away the fork that has fallen down
(line 9). And after Asia brings Kilian a spoon (lines 1113), one of the other children, Gosia, wants one as well (I mi daj yeczk, and bring me (also) a spoon,
line 14).
41
42
Jrg Zinken
Excerpt 5A. Daj mu yeczk (give.IMP him a spoon)

[PP1-1 16:40]
01
((Kilians fork drops on the floor))
02
03
(0.4)
Tadek:
04
05
Kilian.
(0.8)
Asia:
czym
bdziesz jad
teraz
what.INSTR will.2S eat.PST now
How will you eat now,
06
palcem
tak
finger.INSTR yes
with your fingers, yes?
07->Ela:
daj
mu yeczk daj
give.IMP him spoon.DIM give.IMP
give him a spoon, give
08
mu Asiu yeczk
him Asia spoon.DIM
him Asia a spoon
09->Tadek:
tak ale widelec (mu zabierz)

yes but fork
him take.IMP
yes, but take his fork away
10
(1.0)
11
Asia:
prosz:
please.1S
here you are
12
(1.0)
13
((hands Kilian a spoon))
14->Gosia:
i
mi daj
yeczk
and me give.IMP spoon
and give me also a spoon
15
ty chciaa
duy widelec
you want.PST.2S large fork
you wanted a large fork
Asia:
With her comment in lines 56, Asia has, among other things, made public the
fact that she is now attending to Kilians mishap. Furthermore, Asia might also
be displaying her realization that it will in all likelihood be her who will have to
help Kilian out as she is already on the move, she is in the best position to either
pick up Kilians spoon, or bring another one. This is the context in which Ela and
Tadek select imperative formats for making requests which address this very matter. And when Asia brings Kilian a spoon, Gosia can address her in her locally
established role as spoon-bringer by choosing the imperative format. In each

of these instances, the request speaker (Ela, Tadek, Gosia) can figure from what
has just happened that the requested action extends a line of action to which the
request recipient (Asia) is already committed. In other words, the request, at the
time it is formulated, targets an action as part of a wider activity which is already
co-owned by Asia (Rossi 2012).
After rejecting Gosias request for a spoon (line 15), and commenting that the
fork is out of reach (line 16 below), Asia turns back to dealing with other matters:
we know from earlier in the recording that she has an appointment for which
she needs to leave swiftly after dinner. In the meantime, Kilian encounters a new
problem: he is having trouble scooping the food onto the spoon Asia provided
him with, and only manages to slowly push the food towards the edge of his plate,
and onto the table. As Asia is walking back towards the table, Tadek points to
Kilians plate, and addresses a request in the double imperative format to Asia:
we pom Kilianowi ( ) (take.IMP help.IMP Kilian ( ), line 23). Asia resists this
request on the grounds that she has to go (lines 2526), although she ultimately
complies with a new request to gather Kilians food into the center of his plate (no
to zgarnij mu z- na- z brzegu talerza na rodek, then gather for him fr- to- from
the edge of the plate to the centre, lines 2729).
Excerpt 5B. we pom Kilianowi (take.IMP help.IMP Kilian)
[PP1-1 16:40]
16
Asia:
o
tam
za
Gosi ju
nie dostan si
PRT there behind Gosia already not reach.1S REFL
Its there behind Gosia, I cant reach it
17
Tadek:
18
19
Asia:
20
21
(4.5)
jesc
adnie umiecha si do kamery
eat.INF nicely smile.INF REFL to camera
Eat nicely, smile at the camera
(0.4)
Asia:
No::?
PRT
Thats right
22
(3.2)
23->Tadek:
[we
pom
Kilianowi [(
take.IMP help.IMP Kilian.DAT
Take help Kilian
24
[((points to Kilians plate))
43
44 Jrg Zinken
25
Asia:
[ja tato
I dad.VOC
I dad
26
ju
musz
i
already must.1S go.INF
have to go now
27->Tadek:
no TO
A
ZGArnij
mu [z- naPRT then but gather.IMP him fr- onSo THEN BUT Gather for him fr- t-
28
na- na- z- n- n- brzegu

talerza
on- on- fr- o- o- edge.GEN plate.GEN
to- to- fr- t- t- the edge to the centre
29
na rodek
on centre
of the plate
30
(2.6) ((Asia gathers Kilians food))
The we-V2 double imperative request at line 21 comes at a point in time at which
Asia is dealing with some other, unrelated matters: presumably her preparations
for leaving the house. Tadeks request attempts to dislodge her from those matters, and lays claim to her cooperation in dealing with Kilians eating at the expense of Asias present course of actions. This is in contrast to the earlier simple
imperatives (in Excerpt 5A), which were selected in sequential environments in
which Asia was already attending to the relevant business. This raises a puzzle: If
Asia is engaged otherwise, and complying with Tadeks request to help Kilian
would require her to depart from her present course of action, on what grounds
does Tadek formulate his request in an imperative format, which projects only
one response: compliance? A possible ground here for expecting Asias compliance is the fact that the we-V2 request concerns a matter that Asia could and
maybe should be already concerned with. She had taken a central role in dealing
with Kilians eating just a few moments before, and it was her who gave Kilian the
spoon. On those grounds, she might be expected to pay further attention to how
Kilian is getting on with his new eating implement. The request concerns a matter
that had already been (also) Asias business, and thus she can be held co-responsible for supervising Kilians eating.
3.1
Reanimating responsibility
In many cases in the collection, the interactional environment in which speakers build requests in the we-V2 format is characterized by these two features.
Firstly, the requested action addresses some matter that has already occupied the
participants earlier in the interaction. Secondly, the prospective recipient of the

request is presently engaged in some unrelated course of action, from which they
need to become dislodged in order to comply with the request. There is a tension between these two features which provides for the peculiar moral force of
the double imperative, and the emotional undercurrent identified in earlier work
(Krlak and Rudnicka 2006; led 2001): Since the matter to which the request
relates had already been in the common ground, it should not be necessary to
dislodge the addressed person from some other course of action.
The next case further clarifies the characteristic relationship between the request recipient and the business that the request is about, namely that the request
is in the service of a matter that should already be treated as co-owned by the
person who becomes addressed with the double imperative. The family in this
extract has come to the end of their dinner. In line 1, Magda, one of the daughters, reminds her mum, Bogusia, about some biscuits that she had been promised
a few minutes earlier. Here, as she did earlier, Bogusia advises Magda to wait for
the biscuits until her dad would finish his meal (line 3). When Henio confirms
that he has indeed finished his meal, Magda swiftly moves to start clearing the
plates from the table, and Bogusia advises her younger daughter, Zuzia, to do the
same. The parents, Henio and Bogusia, remain seated at the table. This is the context in which Magda formulates the we-V2 request: pieguski mama we wyjmij
pieguski w tym czasie (cookies, mum take.IMP take.out.IMP the cookies in the
meantime, lines 1516).
Excerpt 6. we wyjmij pieguski (take.IMP take.out.IMP the cookies)
[PP3-1 12:07]
01
Magda:
02
pieguski markizy:
pieguski markizy
(choc chip) cookies, (oreo) cookies
(0.4)
03
Bogusia:
zaraz poczekaj
tatu zje
Right.now wait.IMP daddy eat.PFV.3S
Just a moment, wait, let dad finish
04
Henio:
ja ju
zjadem
I already eat.PFV.PST.1S
Im done
05
Bogusia:
zjade
Eat.PFV.PST.2S
Youre done?
06
((Magda gets up and stretches out arms))
45
46 Jrg Zinken
07
Magda:
Daj
Give.IMP
Give
08
Magda:
daj daj daj daj daj daj

Give give give give give give
09
((Henio passes plate to Magda))
10
Magda:
>daj =daj =daj =daj =daj =daj< =d:aj

Give give give give give give give
11
Bogusia:
Zuziu odnie
po
sobie talerz
Zuzia carry.away.IMP after self plate
Zuzia, take your plate (into the kitchen)
12
(.)
13
Bogusia:
odnie
po
sobie talerz
Carry.away.IMP after self plate
Take your plate (into the kitchen)
14
Henio:
i
Magdy
And Magda.GEN
And Magdas too
15->Magda:
pieguski mama we
wyjmij
pieguski
Cookies mum take.IMP take.out.IMP cookies
Cookies mum take take out the cookies
16
w tym czasie
in this time
in the meantime
17
(.)
18
Bogusia:
no dobrze dobrze wyjm

te
pieguski
PRT good
good
take.out.PFV.1S these cookies
Ok Ok Im taking out those cookies
As in the previous case, the request concerns a matter that Bogusia could be expected to be already paying attention to. After all, she has repeatedly promised
cookies after the main meal, and thereby taken on a share in the ownership of
the project of providing cookies. However, she relevantly is not attending to this
project when the time has come to do so. Instead, she idly remains seated at the
table. The double imperative format orients to Bogusia being late in her cooperation: Bogusia is letting an expectable orientation to a jointly owned project slip in
favor of some alternative, individual course of action (just sitting). This sense
of lateness specifically is evident in Magdas turn formulation. Magda requests
that mum bring the cookies in the meantime, that is, while Magda and Zuzia
are clearing the table, and thereby formulates her perception that a phase of the
meal is now ongoing in which Bogusias return to her earlier commitments has
become relevant.
3.2
Laying claim to a cooperative orientation
The observation that the double imperative appears to re-align the recipient to
something they should already be attending to places this format in the vicinity
of practices that have been described as upgrades of a previous request (Craven
and Potter 2010). For example, when a parent requests that a child do something
(Can you move along a little bit), but that request is not complied with, requests
sometimes become upgraded with a different format that ostensively eliminates
any scope for non-compliance (come on, move along). Such upgrades, in a
broader sense, address a similar problem: The request speaker wants the recipient
to cooperate on some matter, but the request recipient isnt cooperating.
However, we-V2 requests are unlike such upgrades in important ways. In
the cases we have seen so far, the double imperative does not so much pursue
compliance in the face of some overt disaffiliation with a request. In the present
examples, it is rather the case that cooperation on some matter had previously
been secured Asia providing Kilian with a spoon, Bogusia promising cookies
after dinner and that the request speaker, by selecting the double imperative
format, is laying claim to a continuation or a making good of that cooperative
orientation.
The next two cases can further elucidate the relationship between the double
imperative and upgraded requests, because these cases are the closest to upgrades
in the collection. Excerpt 7 comes from a recording in which Marta and Karol
are making a paper caterpillar with their daughters Gosia and Jana. The parents
are gluing bits of paper together, and the daughters have been complaining that
they want to do the gluing themselves. At the beginning of the fragment, Mum is
explaining to Gosia that she wants to get everything organized before she lets the
children have a go (lines 13). She then turns her gaze to her husband, addresses
him with the preface You know what Karol (wiesz co Karol, line 3) and suggests: maybe (to) do this on some paper sheet, what? (this infinitival formulation works as a complete TCU in Polish: moe na jakiej kartce to robi co?,
lines 45). Karol does not align with this suggestion, but instead treats the use of a
paper underlay as a matter of individual preference, advising Marta to get herself
a paper sheet if she wants one: no dobrze no to we sobie kartki (well ok, then
get yourself paper sheets, lines 67). However, Marta does not do this, but goes
back to work in silence, while Jana and Gosia engage Dad in a brief conversation
about the glue.
47
48 Jrg Zinken
Excerpt 7. we mi daj t kartk (take.IMP me give.IMP this sheet)

[PP6-1 8:50]
01
Marta:
no dobrze Gosiu poczekaj

PRT good
Gosia wait.PFV.IMP
OK Gosia, wait
02
zobaczymy najpierw jak nam

see.1P
first
how us.DAT
we see first how it
03
bedzie szo Wiesz

co
Karol
will
go
Know.2S what Karol
will go You know what Karol
04->
moe jaka moe na jakiej

maybe some maybe on some
maybe some maybe on some
05
kartce to
ro[bi co?
sheet this do.INF what
sheet (we) do this what?
06
Karol:
07
08
[M::=no dobrze no
PRT good
PRT
Mm well ok
to
we
sobie kartki
then take.PFV.IMP self sheet
then get yourself a sheet
Marta:
((continues her work in silence))

((7 lines omitted))
16
(1.8) ((Karol starts getting up))
17
Marta:
wiesz
co
Gosiu jak y znacz- y
know.2S what Gosia how eh thaeh
You know what Gosia when eh thaeh
18
Jana:
ma[ma a
my:?
Mum
and us?
Mum, and what about us?
19->Marta:
20
Jana:
[we
mi daj
t
kartk
take.PFV.IMP me give.PFV.IMP this sheet
Take give me this sheet
a
my:?=
and us?
((5 lines omitted))
26
Karol:
((K passes sheet of paper))
Fifteen seconds after Mums suggestion to use paper sheets has been dismissed,
Karol starts getting up, at the same time as Mum starts addressing Gosia (wiesz
co Gosiu jak y znacz y; you know what Gosia if eh thaeh, line 17). However,
seeing that Karol is getting up, she abandons that turn, and instead starts addressing Karol with a double imperative: Take me give that sheet (we mi daj t
kartk, line 19). When Karol arrives back at the table, he passes paper sheets to
Marta and his daughter Jana.
In Excerpt 8, the family has finished eating, and the children have got up from
the table. One of the children is apparently playing with the camera or cable, which
prompts admonishings from the parents in lines 12, and a suggestion from Ewa
to Tadek, in the same infinitival form as in the previous case: moe to wyczy ju
co (maybe to turn this off already, what?, line 4). In response, Tadek first initiates repair (to, this, line 6), but then without waiting for a confirmation to his
candidate understanding, starts formulating a dismissal of Ewas suggestion: niech
leci jeszcze no to ju to nie jest przecie (.) (let it run a bit more this isnt after all
(.), lines 6 and 8). Ewa pursues turning off the camera with a double imperative:
no wycz we (PRT turn.off.IMP take.IMP, line 9).
Excerpt 8. wycz we (turn.off.IMP take.IMP)
[PP1-1 27:10]
01
Tadek:
02
Ewa:
nie=nie=nie tego nie [(rb)

no=no=no
this not
(do)
no=no=no this not (do)
[nie ruszaj
not move
dont touch
03
(.)
04->Ewa:
moe to
wyczy
ju
co
maybe this turn.off.INF already what
Maybe (we) turn this off already, what?
05
(1.6)
06
Tadek:
07
Ewa:
08
Tadek:
09->Ewa:
to n[:=niech leci jeszcze

this? May
run still
This? Let it run still
[no
PRT
Yes
no to
ju
to
nie jest przecie (.)
PRT this already this not is
but
PRT this already this isnt after all
no WYcz
we.
PRT turn.off.IMP take.IMP
Turn it off take
49
50
Jrg Zinken
10
Tadek:
tak
yes?
11
Ewa:
NO
PRT
Yes
12
Tadek:
trzymaj
hold.IMP.IPFV
hold this
13
Ewa:
pniej w- wczymy
jak [tego jak
later
turn.on.1P when this when
Later we turn it on when this
((Tadek goes to turn off the camera))
These last two cases share a quality of upgrades: they pursue a course of action
that is initially dismissed by the addressed person. However, what makes these
cases unlike upgrades is that the relevant project (of using paper sheets as an underlay, or of ending the recording) is initially broached not so much with a request
that the other person do something, but with an action formulated in the infinitive which appears to probe whether a particular course of action would be built
on mutually agreed ground, something we might gloss as a suggestion. When a
stance towards some matter turns out in interaction to be shared, this adds to
the social affiliation between interactants (Enfield 2008). Of course, when some
stance turns out not to be shared you think we need paper sheets, I dont the
effect can be the opposite. So we again have here situations in which one person
was assuming or hoping that there would be a socially shared understanding regarding some future action. When this assumption is disappointed, the double
imperative is selected to force the issue. When Karol in Fragment 7 dismisses the
suggestion of using paper sheets as an underlay, Marta lets the matter rest, and
then comes back to it at an opportune moment, when Karol is getting up from
the table. In our collection, Excerpt 8 most resembles an upgrade in so far as the
double imperative pursues the other persons cooperation in next position after
it becomes apparent that the initial suggestion is being dismissed. In contrast to
the other cases discussed so far, and uniquely in the collection, the auxiliary we
(take.IMP) is not placed before the focal imperative verb. Ewa formulates a turn
construction unit (TCU) that is potentially complete as a request in a simple imperative form (no wycz, turn (it) off ), but that becomes extended with we
(take.IMP, line 9).
These observations lead me to suggest that giving expression to a particular
emotional tone of, say, annoyance or impatience, is not what the take-V2 double
imperative is at heart about, even though such emotional stances might often be
part and parcel of double imperative turns due to contentions surrounding the
particular matter at hand. If the we-V2 imperative is specifically attuned to bringing back the recipients attention to something they have already been concerned
with, then it is easy to see that a moral evaluation can often be relevant: The recipient should already be concerning themselves with the relevant matter.
3.3
Splitting and splicing
So far, we have seen that by selecting the double imperative format, the request
speaker displays that the request recipient should already be aware of the need
for the action to be requested, in a situation where the request recipient isnt displaying the relevant alignment. We have until now considered cases in which the
grounds for figuring that the recipient should be aware of the relevant matter were
quite obvious, because the recipients concern with the matter had earlier occupied the interaction in a separate event. In Excerpt 5, Asia had previously been assisting Kilian with his eating, in Excerpt 6, mum had previously promised cookies
after lunch, and in Excerpts 7 and 8, the request speaker had previously suggested
the relevant action to the recipient. I now want to turn to some less obvious situations. Sometimes, the grounds for projecting compliance with a request, in a situation where this request does not extend the recipients present line of action, lies
not so much in a separate event which evidences that persons commitment to the
relevant matter, but more broadly in their very participation in the event so far.
When a participant in an event is projectably splitting off a social unit to turn to
other matters, a speaker might select the take-V2 format in an attempt to splice
that person back into the concerns of the ongoing event. We have already encountered something like this in Fragment 5B, where there were two strands of unfolding events: Firstly Kilians eating, which required attention and supervision, and
was therefore a concern occupying a social unit, namely Kilian and those feeling
responsible for assisting him; and secondly, Asias personal commitments, which
required her to prepare her things and leave the house, and which were, among
those present, a concern for Asia only. Asia splits off from the strand of communal events relating to Kilians eating after having given him a spoon; Tadek
splices Asia back into these matters with the double imperative.
Excerpt 9 provides another example. The family in this excerpt has finished
supper, and the two older sons have already left the table. Their parents, Ala and
Piotr, are still sitting at the table, and Ala is holding their baby boy on her lap. Both
have finished their meals and are talking only intermittently. After a rather long
lapse in the conversation (line 6), Piotr makes a move to leave the table as well: he
puts his glass on the table with a decided movement, says OK (dobra, line 7), and
gets up. In response to this, Ala makes a request formatted with the double imperative: then take turn off the camera (to we wycz kamer, lines 910).
51
52
Jrg Zinken
Excerpt 9. we wycz kamer (take.IMP turn.off.IMP the camera)

[PP5-2 10:54]
01
Ala:
02
A
co
jutro
robimy?
And what tomorrow do.1PL.IPF
And what are we doing tomorrow?
03
(3.0)
Piotr:
04
A na dziesit idziemy
do kocioa
and on tenth
go.1PL.IPF to church
So at ten we go to church
05
(0.8)
Piotr:
06
i
zobaczymy ((starts playing with glass))
and see.1PL.PF
and then well see
07
(16.0)
Piotr:
08
09
((puts glass on the table)) khm: dobra

khm good
khm OK
((gets up))
Ala:
->khkh (to)
we
wycz
khkh (then) take.IMP.PF turn.off.IMP.PF
khkh then take turn off
10
t
kamer
this camera
that camera
11
((Piotr walks to camera and turns it off))
When Ala and Piotr are sitting at the table for sixteen seconds (line 6) without
having any business such as eating, and without saying anything to each other after the last exchange about plans for the next day had rather fizzled out (lines 15),
it seems evident that this joint event is drawing to an end. Piotr finally moves to
get up at line 7, and produces a dobra (ok), which does some work of accounting for, or rather, indexing his departure. He is about to split off the social event in
which he has been a participant. Alas take-V2 request in lines 910: we wycz
kamer (take.IMP turn.off.IMP the camera) reanimates Piotrs co-responsibility for necessities concerning the event he has been co-constituting (here, the
need to turn off the camera at the end of the joint meal).
Excerpt 10 provides a more complex case. Marta has just asked the children,
Gosia and Jana, what they would like for supper (co jecie na kolacj, what are you
eating for supper, line 2), but it is her husband, Karol, who answers. The answer
he formulates is, on the one hand, built to be authoritative. Karol takes the liberty
to answer in place of the children, and he adds accounts to support the choices
he is making for them. But on the other hand, his answer is somewhat confusing.
He first decides that the children will eat cheese, because there is plenty of it in
the fridge and the best-before date is coming up (lines 57). When this does not
receive any uptake (line 8), Karol first does some more work to underline the importance of eating the cheese, but then changes tack and decides that the children
will eat cold meats, because the family have been given lots and there is a danger
that they will have to be thrown away (lines 1416). After completion of his turn,
another silence unfolds during which Marta turns her gaze towards Karol. She
then produces a request in take-V2 format: no to we zrb kanapki z szynk (then
take make sandwiches with ham, lines 1820)
Excerpt 10. to we zrb kanapki (then take.IMP make.IMP sandwiches)
[PP6-3 17:20]
01
(1.4)
02
Marta:
.Hhm- y:: >co

jecie na kolacj<
ehh what eat.2P on supper
Hhm eh What are you eating for supper
03
Gosia:
((to Jana)) MKHA:::=
04
Jana:
=[(
05
Karol:
=[y: serki jedz na kolacj

eh cheese eat.3P on supper
Eh they eat cheese for supper
06
>bo
si koczy
termin
because REFL finish.3S date
because the best-before date
07
wanoci
serw.
expiration.GEN cheese.GEN
is coming up
08
(2.4)
09
Gosia:
k(h):(h): .H:::
10
Karol:
trzeba wyoy troch tych

need
take.out a.bit these
One needs to take out some of
11
serkw bo
jest ich w
cheese because is
they in
these cheeses because there are
12
[brd i: albo zamiast serka

mass and or
instead cheese
masses and or instead of cheese
13
Gosia:
[a::::::::
53
54
Jrg Zinken
14
Karol:
albo wdlina z- zjedza bo

tam
or
cold.meat eat.3P because there
or they will eat cold meats because there
15
rodzice przywieli wdlin

i
parents brought.3P cold.meat and
the parents have brought cold meats and
16
nam si zarazpowyrzuca j
us REFL in.a.moment throw.away it
we will be throwing it away soon
17
(1.6) ((eye gaze towards Karol))
18->Marta:
no to
we
zrb
kanapki
PRT then take.IMP make.IMP sandwiches
So then take make sandwiches
19
z
szynk ja obior te
with ham
I trim
these
with ham I trim the
20
truskawki
strawberries
21
(0.2)
22
Karol:
no to
ktra je- prz- przecie
PRT then which
but
So then whats the- but it is
23
(dopiero) szsta no
only
sixth PRT
only six
24
(0.2)
By answering Martas question about supper preferences in place of the children,

Karol has made himself a participant in an interaction in relation to which he had
so far had the status of an overhearer. Furthermore, by producing a turn that is
occupied with giving advice about what others should be doing (what the children should eat for their supper), Karol has created an interactional slot in which
it is now Martas turn to take the project of supper preparation forward. Karol
himself might withdraw again from participation in supper matters, into which
he had entered in a sideways fashion. Martas we-V2 imperative, then take make
sandwiches with ham (no to we zrb kanapki z szynk, lines 1920), contributes
to an attempt to pre-empt Karols projectable withdrawal from his just previously
usurped position as a participant in supper preparation matters.
4. Animating responsibility: Socialising children into an orientation

towards communal tasks
Finally, there are a few cases in the collection in which it is difficult to see on what
local grounds the request speaker might expect the addressee to already feel coresponsible for the relevant matter. I now want to consider two such cases, to see
how the social orientations carried by the we-V2 format are brought to bear on
these situations.
In Fragment 11, the family has come to the end of their supper. A minute
before the start of this fragment, Iza has started paying compliments to her father
for his short hair, and they and Izas brother, Bolek, have started imagining scenarios for how father and daughter could swap hair, and how Iza would look with
three-millimeter long hair. When there is a short break in that interaction, Ilona
attempts to enlist Iza for clearing the table, and she does so with two double imperatives. The first unit of her turn is designed to recruit Iza to the broader activity
of cleaning up: Izunia we pom posprzta (Iza, take.IMP help.IMP cleaning
up, line 7). The second unit requests that she throw away an empty plastic container, which Ilona places on one corner of the table as she formulates the request:
we to wyrzu (take.IMP throw.away.IMP this, line 8).
Excerpt 11. we pom posprzta (take.IMP help.IMP clear up)
[PP 2-5 22:28]
01
Bolek:
02
chcesz eby ostrzyc na

want.2S to
cut.INF on
Do you want to cut yours
trzy mili[metry bedziesz
three milimeters will.2S
to three milimeters, you will
03
Iza:
[a::: li::::czne
a::: beautiful
04
Bolek:
miaa takie same

have such same
have the same (hair)
05
Iza:
.h dobra
.h good
.h OK
06
(0.3)
07->Ilona:
Izunia we
pom
posprzta.
Iza.DIM take.IMP help.IMP clean.INF
Iza dear, take help to clean up
55
56
Jrg Zinken
08->
we
to
wyrzu
take.IM this throw.away.IMP
Take throw this away
09
ale ja i
tak tak nie bd
but I and so so not will.1S
But even so I will not look as
Iza:
10
licznie
tak wyglda
beautifully so look.INF
beautifully like that
11
Ilona:
Ja ju
dzikuj
I already thank.1S
I am finished (with my meal)
12
Jacek:
ty licznie
wygldasz z
takimi
you beautifully look.2S
with such.INSTR
You look beautifully with the
13
wosami
jakie masz
hair.INSTR such have.2S
hair that you have
((11 lines omitted))
25
Ilona:
(0.4)
26
27
Izunia
Iza.DIM
Iza dear
Jacek:
Dla ciebie w sam raz

For you
in sole time
For you just right ((on Boleks hair))
(0.8) ((Ilona presents container to Iza))
28
29
Ilona:
Wyrzu
to
throw.away.IMP this
Throw this away
30
Bolek:
Cztery razy dusze od

twoich tata
Four
times longer from yours dad
Four times as long as yours dad
31
Iza:
32
Ilona:
do mieci.
To rubbish
In the bin
33
((Iza takes the container and throws it away))
It is certainly the case here that over the previous two minutes, it has become appreciable for all those present that cleaning up is becoming relevant. Everybody
has finished their meal, and Ilona got up from the table three minutes before the
start of this fragment and had started clearing away some bowls. However, as
opposed to the cases considered earlier, it is not clear here that Iza in particular
should already be attending to the fact that cleaning the table has now become
relevant. Of course, it is a socialization goal for parents to get their children to
start taking part in activities such as laying the table or clearing it, which have
value for the social unit. The moral import of the we-V2 format might index to a
child (or anybody who is being socialized into the roles and tasks of a social unit)
that this is a matter for which you should take on some responsibility. This moral
dimension can become a resource for parents who need to dislodge their children
from other courses of action that provide a more dubious benefit (such as fooling
around with their father), and recruit them for some action that is beneficial for
the social unit.
Indeed, dislodging children from unwanted courses of action is often in itself
a primary concern for parents, and the main reason for a request can be to make
sure that the child does not do something else. We have already seen that the weV2 format dislodges the addressee from some alternative course of action, and
occasionally this might be the main objective of the request. In Excerpt 12, a family is in the middle of their dinner. The mum, Ala, is holding her baby boy on her
lap. The two older boys, Lesio and Patryk, are spending most of their time fooling around with one another, and have hardly touched their soups, much to the
dismay of their parents. Lesio in particular seems intent on some mischief, and at
line 6 he nudges Patryks glass of juice just as Patryk is taking a sip from it. In this
situation, where the fooling around between Patryk and Lesio is projectably turning to the less harmonic, Ala addresses Patryk with a request that he spread out a
cushion in the cot so that she could put the baby into it. Her request to Patryk is
in the double imperative form (line 8).
Excerpt 12. we mi roz t poduszk (take.IMP me spread.IMP this cusshion)
[PP 5-1 9:3210:03]
01
Les:
02
03
do mojego poko:ju
to my.GEN room.GEN
to my room
Ala:
04
05
a
teraz przeniesiemy si
and now
carry.over.1P REFL
and now we go over
:::
sh::
((Lesio nudges Patryks glass as
Patryk is drinking))
Ala:
LESio
57
58
Jrg Zinken
06
Les:
hn:: ((grinning))
07
Pat:
Je::zu::.
Jesus
08->Ala:
Patry::k Patryk. we
mi
roz
Patryk
Patryk take.IMP me.DAT spread.IMP
Patryk, Patryk, take spread (for me)
09
y:: t
poduszk tam
w eczku
eh:: this cushion there in bed.DIM
eh this cushion there in the cot
(0.8)
10
11
Ala:
12
tak ebym ja moga

po- Jzka
pooso that I can.CON pu- Josef.ACC put.doso that I could pu- put Josef donie t
poduszk ksik.
not this cushion book
not this one, the book (shape) cushion
(0.5)
13
((2 lines omitted))

16
Ala:
17->
>Tak na- na dwie =wanie (.)

so on on two exactly
like in, in two, exactly
nie:< wanie ro:z
j.
no
exactly spread.IMP it.ACC
no, I mean spread it (open)
18
(0.8)
19->Ala:
Roz j:
t
poduszk ksik
spread.IMP it.ACC this cushion book
spread it, that cushion book
20
((Patryks dad takes over and spreads the cushion))
Alas request has become relevant because the baby has fallen asleep. Of course,
this fact is something that anybody who is present might notice, and as such,
it is possible to treat this, and related matters such as the need to prepare the
cot, as something that should be noticed by others present. But there are no local
grounds here for Ala to expect Patryk in particular to be already orienting to these
matters. Her request seems to mostly work to extract Patryk from the increasingly
problematic situation that he has become embroiled in, and turn his attention to
something more productive.
In sum, in the last two cases there were no clear local grounds for the request
speaker to consider that the recipient could be expected to already be attending
to the relevant matter. In both cases, the double imperative aimed at getting the
addressee, a young child, to stop fooling around, and turn their mind to something new that is beneficial to the social unit to which the request speaker belongs. The request speaker here draws on the affordance of the we-V2 format to
extract the addressee from some course of action and enlist them to some other
course ofaction.
5. Conclusion
This study has analyzed the use of the Polish we-V2 (take-V2) double imperative
to request here-and-now actions. This double imperative is commonly selected
when the request recipient carries some co-responsibility for the relevant business, but isnt currently attending to it. By selecting the we-V2 form, the speaker
reanimates the recipients responsibility, and appeals to the recipients shared
ownership of the wider activity for which the requested action is instrumental.
These results support and extend the findings of previous research. The weV2 construction often carries an element of criticism, insofar as it indexes an
accountable non-involvement of the addressed person in some particular matter they should feel co-responsible for. This is in keeping with the impression of
scholars who have emphasized the emotional tone of the we-V2 construction
(Krlak and Rudnicka 2006; led 2001). The two characteristics of the specific
sequential environment that is indexed by the selection of the take-V2 double
imperative you could already be attending to this, but you arent carry a moral
lesson for the recipient: You should be attending to this. Furthermore, the use of
the take-V2 double imperative often creates a situation in which a person becomes
newly enlisted for some socially beneficial action. This is consistent with the literature describing the ingressive, aspectual (Aktionsart) function of the take-V2
construction (Brinton 1988; Ekberg 1993; Vannebo 2003).
Linguists are used to describing the meaning of grammar in relatively abstract
terms, in an attempt to find a gloss that would cover all usages of the construction that can be attested in the language. I have taken a somewhat opposite approach, describing the imperative take-V2 construction as it enters into a specific
kind of social action: assembling a request for here-and-now action. Ultimately,
such an approach leads us to a different view of language, namely, not so much a
set of self-contained structures, but as an ensemble of grammatical practices in
the service of assembling social actions. From this perspective, understanding
the meaning of a grammatical construction requires us to examine the mutual
dependency between the construction and the social situation that its use builds
off and creates. To researchers interested in the relationships between grammar
59
60 Jrg Zinken
and culture, the approach exemplified here suggests that close attention to speakers grammatical practices affords us insights into the constitution of social actions and of interpersonal relationships. Cultural values and norms are not just
expressed in the ways people use language. Rather, cultural values and norms exist
primarily in speakers participation in sequences of action constituted, inter alia,
by language-specific grammatical practices.
Acknowledgements
Thank you to Giovanni Rossi, Matylda Weidner, and Katarzyna Zinken, who
made many helpful suggestions in the course of writing this paper. Also, I am
grateful to two anonymous reviewers for their feedback on an earlier version.
The work reported here was supported by a Fellowship grant awarded by the Arts
and Humanities Research Council, AH/H03451X/1. This chapter was completed
during a stay at the Institute of Advanced Study at the University of Konstanz,
part of the universitys Cultural Foundations of Integration Center of Excellence, established in the framework of the German Federal and State Initiative
for Excellence.
References
Aikhenvald, Alexandra, and Dixon, Robert M. W. 2006. Serial Verb Constructions: A cross-linguistic typology. Oxford: Oxford University Press.
Brinton, Laurel J. 1988. The Development of English Aspectual Systems. Cambridge: Cambridge
University Press.
Carden, Guy, and Pesetsky, David. 1977. Double-verb constructions, markedness, and a fake
co-ordination. Paper presented at the Papers from the 13th Regional Meeting of the
Chicago Linguistic Society, Chicago.
Coseriu, Eugenio. 1966. Tomo i me voy. Ein Problem vergleichender europischer Syntax.
Vox Romanica 25: 1355.
Craven, Alexandra, and Potter, Jonathan. 2010. Directives: Entitlement and contingency in
action. Discourse Studies 12 (4): 419442.
Ekberg, Lena. 1993. The cognitive basis of the meaning and function of cross-linguistic take
and V. Belgian Journal of linguistics 8: 2141.
Enfield, Nicholas J. 2003. Demonstratives in space and interaction: Data from Lao speakers
and implications for semantic analysis. Language 79 (1): 82117.
Enfield, Nicholas J. 2008. Common ground as a resource for social affiliation. In Intention,
Common Ground and the Egocentric Speaker-Hearer, Istvan Kecskes and Jacob L. Mey
(eds.), 223254. Berlin: Mouton de Gruyter.
Hakulinen, Auli, and Selting, Margret. (eds.). 2005. Syntax and Lexis in Conversation: Studies on the use of linguistic resources in talk-in-interaction. Amsterdam/Philadelphia: John
Benjamins.
Hopper, Paul J. 2002. Hendiadys and auxiliation in English. In Complex Sentences in Grammar
and Discourse. Essays in honor of Sandra A. Thompson, Joan L. Bybee and Michael Noonan
(eds.), 145173. Amsterdam/Philadelphia: John Benjamins.
Krlak, Emilia, and Rudnicka, Kinga. 2006. Selected aspects of directives in Polish. Revista
Espaola de Lingstica Aplicada 19: 129142.
Ochs, Elinor, Schegloff, Emmanuel A., and Thompson, Sandra A. (eds.). 1996. Interaction and
Grammar. Cambridge: Cambridge University Press.
Rossi, Giovanni. 2012. Bilateral and unilateral requests: The use of imperatives and mi x? interrogatives in Italian. Discourse Processes 49: 426458.
Schegloff, Emmanuel A. 2007. Sequence Organization in Interaction. A primer in Conversation
Analysis. Cambridge: Cambridge University Press.
Selting, Margret, and Couper-Kuhlen, Elizabeth. (eds.). 2001. Studies in Interactional Linguistics. Amsterdam/Philadelphia: John Benjamins.
Sidnell, Jack. 2009. Comparative perspectives in Conversation Analysis. In Conversation
Analysis: Comparative Perspectives, Jack Sidnell (ed.), 327. Cambridge: Cambridge University Press.
Sidnell, Jack, and Enfield, Nicholas J. 2012. Language diversity and social action: A third locus
of linguistic relativity. Current Anthropology 53 (3): 302333.
led, Anna. 2001. O pewnej specyficznej funkcji sw we i wecie w mowie potocznej.
Poradnik Jzykowy 7: 1822.
Vannebo, Kjell I. 2003. Ta og ro deg ned noen hakk: on pseudocoordination with the verb ta
take in a grammaticalization perspective. Nordic Journal of Linguistics 26 (2): 165193.
Weiss, Daniel. 1993. Dvojnye glagoly v sovremennom russkim jazyke. In Kategorija skazu
emogo w slavjanskix jazykax. Modalnost i aktualizacija, F. Giusto Fici (ed.), 6797.
Mnchen: Otto Sagner.
Weiss, Daniel. 2007. The grammar of surprise: the Russian construction of the type Koshka
vzjala da umerla Suddenly, the cat died. In Proceedings of the 3rd International Conference on Meaning Text Theory, Tilmann Reuther, Leo Wanner and Kim Gerdes (eds.),
427436. Mnchen/Wien: Otto Sagner.
Wootton, Anthony J. 1997. Interaction and the Development of Mind. Cambridge: Cambridge
University Press.
Wulff, Stefanie. 2006. Go-V vs. Go-and-V in English: A case of constructional synonymy? In
Corpora in Cognitive Linguistics: Corpus-based approaches to syntax and lexis, Stefan Th.
Gries and Anatol Stefanowitsch (eds.), 101126. Berlin: Mouton de Gruyter.
Zinken, Jrg, and Ogiermann, Eva. 2011. How to propose an action as objectively necessary:
The case of Polish trzeba x (one has to x). Research on Language and Social Interaction
44 (3): 263287.
Zinken, Jrg, and Ogiermann, Eva. In press. Responsibility and action. Invariants and diversity in requests for objects in British English and Polish interaction. Research on Language
and Social Interaction.
61
Eye behavior in Russian spoken

interaction and its correlation
with affirmation and negation
Elena Grishina
The paper describes the most commonly recurring traits of eye behavior in Russian dialogue and their connection with the Russian words Da yes and Net no,
which are the main modes of expressing affirmation and negation in Russian.
The eye behavior is analyzed from three main points of view: (1) gaze grammar,
(2) eye closing, (3) blinking. The author arrives at the conclusion that (1) Da
yes and Net no are connected to a specific gaze pattern, (2) both are combined
with eye closing as a special (embedded) gesture, and (3) they form the basis
on which the usage of blinking as punctuation marks and as accent diacritics
is founded. The research has been conducted using data from the Multimodal
Russian Corpus (MURCO) and selected examples from natural interactions.
Keywords: confirmation, negation, gaze grammar, turn-taking, blinking
1. Synopsis
Eye patterns, accompanying spoken interaction, are largely independent of our
control. However, this does not mean that their eye behavior is entirely unpredictable and chaotic. The paper describes the main regularities, which are specific to
eye behavior in the Russian spoken interaction. Here I argue for a basic correlation between a speakers eye behavior and two basic dialogic reactions that form
the foundation of linguistic interaction. Namely, affirmation and negation, which
are expressed in Russian with the words Da yes and Net no.
There are two main types of eye patterns in Russian dialogue: (1) gaze movements and (2) eye closing. In keeping with this division, the paper includes two
main sections: Gaze grammar (Section 3) and Eye closing (Section 4), which
are closely connected and describe eye behavior in Russian dialogue and its correspondence to the usage of Da and Net from considerably different points of view.
64 Elena Grishina
In Section 5, the phenomenon of blinking is examined. Blinking is interpreted as the highest degree of the grammaticalization of eye closing. Specifically,
I argue that while eye closing is connected with Da yes and Net no as lexical
units, blinking functions as a grammatical (in the sense of absolutely regular and
unconscious) unit and serves as a syntactic or accent marker. At the same time
blinking preserves its etymological connection with Da and Net as the markers
of affirmation and negation.
2. Introduction
The Russian words Da yes an Net no are the principal means of expressing affirmation and negation in Russian. The concepts of assertion and negation are
two basic human reactions, which have an extraordinary influence on dialogue
structure and its development.
On the other hand, a spoken dialogue is a kind of multimodal event; that is to
say, it is a combination of linguistic phenomena proper and visual behavior phenomena: gesticulation, face and gaze behavior, body postures and so on (Cienki
2005; Cienki and Mller 2008; Mittelberg 2007; Poggi, DErrico, and Vincze 2010).
Therefore, it is quite interesting to analyze the types of visual events that regularly
collocate with the basic Russian expressions of assertion and negation.
The following are some terminology and notation conventions are regularly
used in the paper.
BBW boundary between phonetic words
Change () gaze movement within Inside or within Outside that does not
cross the boundary between the two zones; a change in Focus of Attention
Communication zone (CommZ) a tetragon, the vertices of which are the eyes
of two interlocutors
Cue an utterance, which has turn-takings (the changing of a speaker) at its
boundaries
EC eye closing
Focus of Attention the object of ones Gaze, any person or object that a
speaker looks at
Gaze two imaginary rays that extend from the communicants eyes (Uryson
2003)
Inside (to be inside CommZ) a communicants gaze is located within the
limits of the CommZ
Eye behavior in Russian spoken interaction
Listener-Speaker Transformation (LST) the moment within a dialogue when

a listener becomes the speaker (the combination final phonetic word + turntaking + initial phonetic word from the point of view of the next speaker)
Login () the change from Outside to Inside
Logout () the change from Inside to Outside
Outside (to be outside CommZ) a communicants gaze is located outside the
limits of the CommZ
Shift the general term for Login, Logout and Change
Speaker-Listener Transformation (SLT) the moment within the dialogue
when the speaker becomes a listener (the combination final phonetic word+
turn-taking + initial phonetic word from the point of view of the previous
speaker)
Transfer the general term for Login and Logout

EC
|| pause
Login
Logout
Change
3. Gaze grammar
3.1
Previous research
The term gaze grammar (Wiemann and Knapp 1975:90) implies that the eye
movements in a dialogue are not random but are instead highly correlated with
the linguistic context. Numerous provocative investigations have already been
conducted in this field; however, the vast majority of the research has been conducted using spoken English data and the results of deliberately performed experiments and studio recordings.
So far, the basic principles of gaze grammar, which are more or less conventional, are as follows.
3.1.1 Listeners gaze patterns
1. A listener looks at the speaker most of the cue time (cf. Argyle and Cook
1976; Argyle and Graham 1976; Bavelas, Coates and Johnson 2002; Fehr and
Exline 1987; Kendon 1990; Kleinke 1986); the use of the trackers shows that
approximately 9095% of the dialogue time a listener looks at the speakers
face (Gullberg and Holmqvist 1999, 2006).
65
66 Elena Grishina
2. A listeners eyes very often follow the speakers gaze when it is referential (fixed)
and is directed beyond the communication space of the dialogue (cf. Driver et al.
1999; Gullberg and Kita 2009).
3.1.2 Speakers gaze patterns

1. Gaze direction marks the turn-taking in a dialogue (cf. Duncan 1972; Bavelas
2005), and
gaze role in turn-taking is context-specific: when the overall level of gaze is low,
as in conversations between strangers (Beattie 1980) or when the discussion topic
imposes a high cognitive load on the interlocutors, gaze plays a more significant
role.
(Novick, Hansen and Ward 1996:1888)
2. Two patterns in particular occurred fairly frequently during turn-taking:

(1) the mutual-break pattern: as one interlocutor completes an utterance, he or
she looks toward the other. Gaze is momentarily mutual, after which the other interlocutor breaks the mutual gaze and begins to speak; () (2) the mutual-hold
pattern, is similar except that the turn recipient begins speaking without immediately looking away; although in many cases the turn recipient subsequently broke
gaze during the course of the turn.
(Novick, Hansen and Ward 1996:1889)
3. In addition, the direction of a speakers gaze depends on the communicative

structure of an utterance:
the beginning of the thematic part of an utterance is frequently accompanied by
gaze behavior that looks away from the hearer, while the beginning of the rhematic part is usually accompanied by gaze behavior that looks toward the hearer;
() In cases where the beginning of the theme coincides with the beginning of
a turn, the speaker always looks away from the hearer. In cases where the beginning of the rheme coincides with the end of the turn, the speaker always looks
toward the listener.
(Torres, Cassell and Prevost 1997:2)
4. Accented or emphasized linguistic items are punctuated by head nods;

the speaker may also look toward the listener at these moments (Cassell
1999:21).
Based on the quotations included above, we can see that it is generally accepted
that gaze patterns depend on (1) turn-taking and (2) discourse structure (the actual division of a sentence).
We can see that the previous works on the topic have dealt (1) with the English
data and (2) with a speakers gaze behavior at the end of his/her cue. Because English has fixed word order, the end of a cue quite often coincides with the rheme. As
a result, it is difficult to differentiate whether a speakers eye behavior at the end
of his/her cue is connected with the termination of the cue or with rhematic emphasis. My approach differs from previous approaches in that it is based (1) on the
fact that Russian (as opposed to English) has free word order; (2) on the idea that
the most interesting and important results are obtained when we analyze the process of a speakers shift to a listener from a speaker (and vice versa), and not from
a speakers gaze behavior at the end of his/her cue. The combination of these two
conditions strongly limits the possibility to compare my results to the results of
previous research. To achieve comparability, my transformational speaker/listener
method ought to be tested using data from a large multimodal English corpus.
Because of the regular initial phrasal position of the propositional Da yes
and Net no in spoken Russian, it seems natural to analyze the most common
gaze movements at turn boundaries and to compare resultant data to the most
frequent gaze movements at the beginning of the cues, which include the initial
Da or Net.
3.2
MURCO data
As mentioned above, the investigation of gaze patterns in a dialogue has been performed mainly on the basis of spoken English, which exhibits fixed word order.
Moreover, the main findings in the field are the results of data gathered in experimental settings, rather than in natural settings. That is why it is very interesting to
crosscheck my data using an alternative data source.
The Multimodal Russian Corpus (MURCO) is one of the modules within
the framework of the Russian National Corpus (RNC, www.ruscorpora.ru). It
includes transcripts of spoken Russian aligned with corresponding video clips
(cf. in detail Grishina 2009a, b). The bulk of the MURCO consists of transcriptions of Russian movies from the period between 1930 and 2009. Therefore, the
MURCO puts data at our disposal that are absolutely new both in kind and in
quality from the point of view of the gaze in dialogue studies. First, the MURCO
consists of Russian data and the MURCO data imitate communication in a reallife framework. The results of the research may be greatly biased in both of these
conditions.
To analyze the gaze patterns at cue boundaries we have annotated around 550
successive cues from dyadic conversations in two films from two different time
periods (1969 Diamond Hand vs. 2007 Election Day), by two different directors and with two different casts. These two films have been chosen at random.
However, since eye behavior patterns do not differ between the films this does not
introduce any bias into the conclusions.
67
68 Elena Grishina
3.3
Gaze grammar at cue boundaries
The investigation of cue boundaries (546 LST (Listener-Speaker Transformations)

and 557 SLT (Speaker-Listener Transformations)) shows that a speaker at the end
of a cue and a listener at the beginning of a cue behave rather differently in terms
of gaze direction. We can see that the stable position Inside at the cue boundaries is far more typical for SLT than for LST (38% [209] vs. 27% [152]). On the
contrary, Transfers at the beginning of a cue are far more characteristic for LST
than for SLT (32% [183] vs. 11% [62]). Therefore, our data confirm the fact that
gaze movement in a dialogue is closely related to turn-taking: at the end of a cue a
speaker prefers the stable position inside the zone of communication, while at the
beginning of a cue a speaker tends to change his/her gaze position.
3.3.1 Speaker-Listener Transformation
To analyze the SLT at greater length, we need to distribute all the cues into two
groups: (1) provoking cues which require the listeners immediate reaction (questions and imperatives of different kind), 499 cues in total; (2) neutral cues (all the
rest), 582 cues in total. Accordingly, we have grouped the gaze patterns as follows:
INSIDE = Inside + Login; OUTSIDE = Outside + Logout.
The data show that the gaze behavior of a previous speaker in these two cases
(provoking vs. neutral cues) is quite different. The speaker controls turn-taking
using the position of his/her gaze within CommZ only while the provoking cues
are also occurring (position INSIDE at the end of the previous cue (75%) and at
the beginning of the next cue (68%)). If a neutral cue takes place, the speaker has
free choice regarding gaze direction both at the end of the previous cue and at the
beginning of the following one (the choice between the INSIDE/OUTSIDE positions is absolutely equiprobable). Therefore, we may state that a speaker controls
the process of turn-taking with their gaze if they are concerned with the immediate reaction to the cue. If a cue is neutral, then the speakers control is unnecessary
and the gaze movements are free.
3.3.2 Listener-Speaker Transformation
If we analyze LST positions following the same procedure, we can see that at the
beginning of a cue, whether provoking or neutral, the INSIDE/OUTSIDE positions of a speakers gaze are more or less equiprobable and cannot serve as the
defining characteristic for the beginning of a cue.
But if we compare the probability of the INSIDE position at the end of the
previous cue (at the final phonetic word of the previous turn), we notice that the
. In the square brackets we adduce the absolute values of the parameters.
probability of being in the INSIDE position for neutral cues is much higher than
for provoking ones (58% vs. 45%). At the same time, the probability of being in
the OUTSIDE position for provoking cues is much higher than for neutral ones
(55% vs. 42%).
3.3.3 Compensatory pattern
To summarize, we may say that
1. If the speaker controls turn-taking with their gaze (in the provoking cues,
for example, see Figure 1), a listener is free to choose the gaze direction at
the end of the speakers cue (in point of fact, a listener prefers the OUTSIDE
position);
2. If the speaker does not control the turn-taking with his/her gaze (in the neutral cues, for example, see Figure 2), the duty to control it passes to a listener.
Provoking cue
Speaker
Listener
Figure 1. SLT. Last phonetic word of the previous cue. Gaze direction
Neutral cue
Speaker
Listener
Figure 2. SLT. Last phonetic word of the previous cue. Gaze direction
. Here and hereinafter some fragments of the Herluf Bidstrups drawings are used.
69
70 Elena Grishina
We name this the compensatory gaze pattern. The compensatory pattern implies
that if the speaker does not control the turn-taking process using the special interrogative X or other phonetic cues associated with the imperative and the direction
of his/her gaze, the listener is forced to do it themselves. Therefore, in cases of uncontrolled turn-taking, a listener checks the speakers intentions (is the speaker
about to finish speaking?) and informs the speaker about his/her own intentions
(Im going to speak). To fulfill these two tasks, the listener enters the Inside/Login
positions at the final phonetic word of the previous turn.
3.3.4 Contrast pattern
The examined data shows that in many of the cues (38%) a new speaker marks
the beginning of his/her cues using a specific pattern we term the contrast pattern.
The scheme at the start of a contrast cue gives rise to the following regularity:
1. If, at the end of the previous turn, the future speakers gaze is in the INSIDE
position (Inside or Login), then at the beginning of his/her cue the speaker
leaves the CommZ;
2. If, at the end of the previous turn, the future speakers gaze is in the OUTSIDE
position (Outside or Logout), then at the beginning of his/her cue the new
speaker enters CommZ (see Figure 3).
Since a listener is more often situated outside CommZ at the end of a provoking
cue (according to the compensatory principle), he/she more often enters CommZ
A
Listener
B
Speaker
A
Listener
B
Speaker
Figure 3. LST. First word of the next cue. Gaze direction
at the beginning of his/her answer. And vice versa, at the beginning of his/her
reaction to the neutral cue, the new speaker more often leaves CommZ.
Therefore, Transfer, or change of the gaze status quo, is the most preferable
way to let the previous speaker know about the other interlocutors intention to
say something.
3.4
Da/Net-cues
Since affirmation and denial are two main types of pragmatic and semantic reactions to a previous cue, especially to a provoking one, it is worthwhile to compare
the grammatical gaze patterns at the beginning of all answers to provoking cues
and at the beginning of those answers beginning with Da yes and Net no. If we
find significant differences between two data sequences, we can be sure that Da
and Net gaze behavior have linguistic relevance.
We have analyzed 142 assertions that include an initial Da in propositional
meaning and 158 negations with an initial Net. The results are as follows:
1. It is a specific trait of Da-cues that they maintain a stable Inside position
while traversing the turn boundaries (38% vs. 23% on average and 20% for
Net-cues)
2. A stable Outside position is rather uncommon for Da-cues at the cue boundaries (15% vs. 24% on average and 25% for No-cues)
3. A Transfer at the beginning of the cue is typical of Net-cues (53% vs. 40% on
average and 37% for Da-cues)
4. The data regarding the gaze positions INSIDE and OUTSIDE on (a) the final
phonetic word of a stimulus and (b) the initial phonetic word of a reaction are
practically the same for Net-cues and for all cues as a whole (48% and 45%;
47% and 51%).
5. It is typical for the Da-cues to maintain the INSIDE position on the final phonetic word of a stimulus and on the initial phonetic word of a reaction.
These results allow us to state that:
1. The compensatory principle does not apply to Da-cues: in spite of the fact that
the speaker looks at the interlocutor at the end of a provoking stimulus, the
interlocutor, on condition that he/she is going to agree with the stimulus,
looks back at the speaker (see Figure 4).
2. The contrast pattern is relevant for Net-cues to an even greater degree than for
all answers to provoking cues and for Da-cues.
71
72
Elena Grishina
Speaker
Listener
Figure 4. SLT. Da-cues. Gaze direction
3. The main gaze patterns at the beginning of Net-cues and the main gaze patterns at the beginning of all answers to provoking cues are the same. This
means that negation is the basic driving force in a Russian dialogue and that,
even if a cue does not include the word Net no at its beginning, it nevertheless includes the objection seme.
4. In Da-cues, the contrast principle is irrelevant and the compensatory principle does not apply. This means that in the opposition of Da-cues to Netcues, the first is the intensive (marked) term and the second is the extensive
(unmarked) one.
4. Eye closing (EC)

4.1
EC as full gesture
To the best of my knowledge, the complete corpus description of ECs as gestures

does not exist, therefore it is not possible to refer to previous research for the field
in question. However, the comparison of my data with the data from different
languages would be a fruitful and informative venture and this is an area where
more research is needed.
4.1.1 EC to become unseeing
The only significant movement in EC is the movement of the superior eyelids
downward until they connect with the inferior eyelids. Yet, this simple event leads
to a quite complicated series of consequences.
In a non-conversational situation the movement of eyelids downwards has
only one purpose, which is to shield ones eyeballs from any damage. If anything
moves towards ones face, the eyelids close over the eyeballs and protect them;
bright light has the same effect. If the moving object is big enough or moves
quickly, a person closes his/her eyes tight.
This physiological reaction is a kind of uncontrolled instinctive reflex and, as
such, should not be considered a gesture and has no linguistic meaning.
But the eyes are not merely physical objects and parts of the human body, but
also channels through which information reaches ones brain. As such, the downward movement of ones eyelids may stop the flow in information through these
channels. In this case, EC becomes a gesture which means to become unseeing.
There are three main cases in which a person wants to see nothing.
1. The information may be disagreeable or may damage ones brain, as when a
moving object may damage ones eyeballs.
2. The information may be superfluous. This situation takes place when an individual experiences certain emotions or cognitive processes and is completely
absorbed in them. In this case, the external information may distract ones
attention, so the individual closes his/her eyes to avoid distraction. In such
situations, EC means concentration (Krejdlin 2004: 377).
3. One may be too weak to apprehend the information because of post-stress
relaxation. In such situations, EC means relief . The speaker does not want
to receive any new information because he/she doesnt have enough power to
process new data.
4.1.2 EC to vanish
Our personal experience tells us that when animals and little children do not want
to be seen, they hide their heads. The idea is that if I cannot see anybody, then
nobody can see me. When adults feel shameful, disreputable, confused or embarrassed, they do practically the same: they close their eyes and thereby imitate the
childhood activity of disappearing from an unbearable situation.
Subsequently, lowering ones eyelids may stop the flow of data in both directions. On the one hand, the speaker terminates the apprehension of outside information (seme to become unseeing). On the other hand, the speaker imitates
his/her own disappearance from an awkward communicative situation: closing
his/her eyes he/she leaves the zone of communication and breaks contact with
his/her interlocutors (seme to vanish).
The seme to vanish is present not only in ECs that mean to be confused, but
also in ECs that mean to be surprised. The inner form of EC to be surprised is as
follows: (a) the speaker sees something that does not correspond to his/her expectations or to his/her standards; (b) the speaker surmises that something wrong has
happened to his/her eyes; (c) the speaker closes his/her eyes and thereby leaves
the CommZ; (d) the speaker opens his/her eyes and thereby enters the CommZ,
73
74
Elena Grishina
in the hopes of seeing something that suits his/her standards, his/her eyes being
refreshed by the Logout-Login process.
4.2
Da/Net-reactions and EC as embedded gesture
4.2.1 Gesture ligatures

Co-speech gestures are very often bound in ligatures. There are two types of
ligatures.
1. Vertical ligatures combine two or more gestures that are produced at different
areas of the body (that is, gestures involving the activity of different gestural
articulators), but have the same meaning. For example, some frequent vertical ligatures are the combination of gestures to lift ones brows (active organ
brows) + to shrug ones shoulders (shoulders) + to throw out ones under lip
(under lip) + to hold up ones hands (hands) + to hang ones head on one side
(head). This vertical ligature (and every gesture by each gestural articulator)
means to be puzzled.
2. Horizontal ligature is a standard sequence of two gestures that are produced
at the same area of the body (involve the same active gestural articulator),
but that have a different meaning. For example, the sequence of gestures to
raise ones hand + to point to oneself with ones hand means give me and is a
frequent horizontal ligature.
4.2.2 EC within vertical ligatures
EC very often occurs in conjunction with two vertical gesture ligatures: (a) vertical ligature that means Da yes, and (b) vertical ligature that means Net no; that
is, the nod of agreement and negative head shaking are very often combined with
EC. The analysis of 136 Da-cues and 160 Net-cues from MURCO (that is, those
cues which have Da and Net that contain propositional meaning in first position)
show the following distributions.
1. Answering Net, a speaker may or may not shake his/her head. If a speaker
shakes their head then they are more likely to close his/her eyes (frequency
50% vs. 40%, on average). Vice-versa, if a speaker closes their eyes then they
are more likely to shake their head (frequency 80% vs. 64%, on average).
2. Answering Da, a speaker may or may not nod in demonstrating agreement.
If a speaker nods, then they are more likely to close their eyes (frequency 43%
vs. 26%, on average), if they do not, then in all likelihood they will not close
their eyes (frequency 90% vs. 74% on average).
If we test whether a speaker can express affirmation or negation with EC only:

without nods, head shakings, and the words Da or Net, we soon find that this is in
fact, impossible. EC alone, without the accompanying gestures or words, cannot
express either affirmation or negation.
Therefore, in this case we encounter a new type of vertical ligature: in the
pairs EC + nod of agreement and EC + negative head shaking, EC is a gesture of
affirmation or negation, but this gesture cannot be used separately and takes on
meaning only when used in combination with the words (Da/Net) or a gesture of
affirmation/negation.
We propose that this kind of gestures be termed embedded gestures. Embedded gestures cannot be understood out of context and must be combined with
speech and full gestures. On the contrary, full gestures (see above, Section 4.1)
are self-sufficient and context-free; they dont depend on corresponding speech
or gesticulation.
This begs the question: why are such different (even opposite) reactions such
as affirmation and negation accompanied by the same embedded gesture, EC?
I argue that in the Da-answers, the embedded gesture EC is the backup gesture which metonymically duplicates the nod of agreement: the descent of ones
eyelids repeats the vertical descent of ones nodding head.
As for negation, it we show in Section 3 of our paper that the contrast pattern
is very specific for negation in Russian: at the beginning of negative cues, Russian
speakers very often break the gaze status quo. If at the end of the previous question a Net-answerer looks Outside, then at the beginning of his/her answer he/she
is likely to Login; and vice-versa, if at the end of the question a Net-answerer is
Inside and looks at a questioner, then at the beginning of his/her answer he/she is
likely to Logout. We may assume that the negative shaking of ones head is the contrast gaze pattern reduced to a conventional gesture, and it redupicates the Logout
of the gaze, which in these situations breaks off the communicative relation of
the interlocutors. If we analyze EC from this point of view, we understand that
in the case of negation, EC represents the seme to vanish, which was mentioned
in Section 4.1.2: closing his/her eyes, the speaker imitates his/her own disappearance from the CommZ, which matches with the negatively shaking head and the
break-off of the communication.
75
76
Elena Grishina
5. EC as empty gesture: Blinking

5.1
Investigated data
A thorough search of the literature reveals that the systematic investigation of

blinking and its role in the spoken interaction has been conducted only on English data (Loehr 2007; see below). Moreover much of this research is comparatively recent. Within Russian, only the 2009 doctoral dissertation by Prozorova
(2009) on the role of blinking in Russian Sign Language deals with gesture in
any detail. The current study represents the first comprehensive investigation of
blinking in spoken Russian and its relationship to linguistic content. The previous
section shows that full gestures can be modified and become empty ones (the embedded gestures are transitional phenomena between these two groups). In this
section, I analyze EC as an empty gesture (blinking). Specifically, I ask whether
blinking in spoken Russian is both irregular and non-linguistic.
The cinematographic data are inapplicable in the case of blinking. Blinking, as
it is absolutely uncontrolled, is closely connected with a speakers cognitive processes, and specifically with the process of speech production. Since the dialogue
in movies is not spontaneous, the instances of blinking in cinematographic speech
are quite rare and do not help a researcher to draw any conclusions. Therefore, I
generated a corpus of naturally occurring speech from the following sources:
2 fragments of TV-interviews (monologues):
speaker G, male, 182 words, 89 cases of blinking;
speaker L, male, 271 words, 12 cases of blinking;
1 fragment of a TV-interview (dialogue):
interviewer P, male, 349 words, 47 cases of blinking;
interviewee Z, male, 453 words, 116 cases of blinking;
2 fragments of conference presentations (monologues):
speaker K, male, 118 words, 71 cases of blinking;
speaker R, male, 198 words, 99 cases of blinking;
1 fragment of an informal polylogue in cafe:
participant I, female, 278 words, 133 cases of blinking.
5.2
Blinking and pauses
Blinking may coincide with the boundaries between the phonetic words (BBWs)
as well as within the phonetic words themselves. This fact, combined with the
individual differentiation in blinking density, which varies with speakers and
genres, demonstrates that blinking is entirely random and chaotic.
Let us investigate the following possibilities: (a) the concurrent action of

blinking and BBWs, and (b) the concurrency of blinking along with the words.
The idea that there is a strong correspondence between blinking and pauses
is corroborated by the fact that Russian Sign Language uses blinking to serve as
a marker at the boundaries of elementary discourse units (Prozorova 2009). Our
data show that the longest pauses (duration > 1 sec.) are accompanied almost as
a rule, by blinking.
To analyze our data, let us distinguish between two types of pauses:
1. phonetic pauses, that is, the break-off of phonation, which may be the result of
hesitations, syntactic boundaries, changes in focus of attention, and so on;
2. syntactic pauses, which take place when there is no break in speech, but when
the breakpoint occurs within the syntactic structure of a phrase.
The types of syntactic pauses are as follows:
a. the beginning of a cue
b. the end of a cue
c. the beginning of a new sentence (main, subordinate, independent), or a participle clause
(1) Dumaju to vs e ne takogo mastaba Vo-pervyx tam || poze naalos
All the same, I dont think it goes to such an extent. First of all, it began there
later
d. parentheses, qualifying constructions

(2) On togda predloil soverenno radikalnuju ve otmenit vstupitelnye
kzameny v vuz kak takovye
Then he suggested absolutely radical idea to cancel all entrance examinations to the universities as such
e. the points of change in speech strategy, the points of auto-correction

f. all kinds of citations
(3) || kak vy proiznosite || sejas v stjannom variante?
Eh || How do you pronounce || sejas when you pronounce it
contractedly?
The data show that blinking will coincide with a pause if blinking relates either
to a phonetic pause or to a syntactic one. In this case, the data show that in the
absolute majority of the cases blinking at BBWs coincides with pauses (84% [252]
of the cases of blinking at BBWs coincided with pauses, and only 16% [49] of
them didnt).
77
78
Elena Grishina
Moreover, closer investigation of our data shows clearly that the correlation
between blinking and syntactic pauses is far stronger than the correlation between
blinking and phonetic ones. Indeed, the number of phonetic pauses that dont coincide with syntactic ones is approximately 32% ([99 of 305 phonetic pauses]);
whereas the number of cases of blinking that dont coincide with syntactic pauses
is around 16% ([40 of 208 cases of blinking at BBWs]).
The above data suggests that ocular pauses are more specialized in purpose
and are used to iconically mark the syntactic boundaries in verbal speech, while
the functions of the phonetic pauses are more diversified.
Thus, the data show that blinking at BBWs in spoken Russian fulfills the same
functions as the following punctuation marks:

capital letter (the beginning of a cue, of a phrase)

period and the period-equated marks (the end of a cue, of a phrase)
comma (the boundaries of subordinate clauses, parentheses)
colon (qualifying phrase)
brackets (parentheses)
quotation marks (citations)
strikeout (changes of strategy, auto-corrections)
5.3
Blinking and stress
My analysis of the concurrence of blinking and the production of a phonetic word

shows that internal blinking (blinking which occurs within the limits of a phonetic word) correlates with the stressed syllable. To correlate means that the peak
of an act of blinking mainly takes place on the stressed vowel, but that sometimes,
depending on the individual characteristics of a speaker or on the linguistic peculiarities of a phrase, blinking may take place immediately before or immediately
after a stressed syllable (if a word is a compound, blinking is very often hooked to
the first stressed syllable, which has a secondary stress):
gòlfstr'im, gòlfstr'im, gòlfstr'im Gulf Stream, karibskom Caribbean
(Loc., Sg.), krùgovor'ot rotation, zalive bay (Loc., Sg.), vody water (Acc.,
Pl.), vodnye water (attr., Pl.), let years (Gen., Pl.), nazad back, sluanjax
hearings (Loc.), gòsd'ume State Duma (Loc.), drugoe another (neutr.),
vsèpobed'ajuee all-conquering (neutr.), òdnovrem'enno at the same
time, and so on.
The fact that blinking is connected with the stressed syllable in American English
was established in Loehr (2007). Our data confirm that this characteristic also
holds true for Russian.
In concluding, two observations should be made. First, the ocular accentuation of full words (nouns, adjectives, adverbs, verbs, pronouns) is more frequent
than the ocular accentuation of empty words (15% [200] vs. 1.4% [19]). Second,
4% of all words are accompanied by internal blinking cannot be described as
having ocular stress because blinking does not correlate with the stressed syllable in these cases. The exceptions are as follows: probl'emy problems (speaker
L), zapoln'jaetsja is filled, pustota' emptiness (interviewee Z), sravn'eniju
comparisonDat., o'bestve societyLoc., t'estovostju testingInstr., uprazdn'ili
abolishPast., Pl. (speaker K), suffiks'alnoe suffixalneutr. (participant I).
5.4
Shifts as punctuation marks and as accentual diacritic
In the previous sections, we have seen that blinking fulfills the roles of punctuation marks and accentual diacritics. However, the characteristics of the speakers
in Section 5.1 make it evident that blinking frequency is highly specific to the
individual. For example, speaker L has a near zero value of blinking frequency
(0.04), while the speaker K has the blinking frequency 0.6.
The question arises: if a speaker has a very low blinking frequency, does it
mean that they have no means for marking pauses and stresses in their speech
with special eye behavior?
Analysis of the data shows that syntactic and phonetic pauses and stressed
syllables are regularly accompanied not only by blinking, but also by Shifts (Login,
Logout, and Change) as well. For example, in extract (4) (speaker L) in the words
vosprinimajut, vsjakoe, usilenie and nix, the stressed syllables are accompanied
with Shifts and the pauses before the words potomu and potomu to are also combined with Shifts.
(4) || potomu oni vosprinimajut vsjakoe usilenie gosudarstva kak
socializm || ponimaete || potomu to dlja nix est tolko odin aspekt dlja
amerikanskix konservatorov usilenija gosudarstva to socializm
Extract (5) (speaker R) illustrates the possibility of combining blinking and Shifts
to mark phonetic and syntactic pauses and accentuation. In the words pytalsja,
vspomnit, rodilos, vspomnil, pomnju, togda, vystupal (in contrast to let, nazad,
Gosdume), the accentuated syllables are marked with the combination of blinking and Shifts. In the same way, pauses {1} and {2} are marked with the very same
combination of ocular events (in contrast to pauses {3}, {4} and {5}).
79
80 Elena Grishina
(5) Ja pytalsja vspomnit {3} kak rodilos EGE {1} I znaete to ja

vspomnil? {4} Kak let desjat nazad || na sluanjax ||mmm{5} v
Gosdume ||{2} ja ne pomnju kakuju problemu togda rassmatrivali vystupal Evgenij Abramovi Bunimovi
We can see that the data are highly varied for different speakers. Moreover, blinking is connected far more tightly with pauses than Shifts. Yet, it would seem that
Shifts have the ability to mark pauses on their own, without any support from
blinking.
Thus, we can assert that blinking and Shifts are functionally identical. At
BBWs, both blinking and Shifts mark the pauses (syntactic and phonetic); within
word boundaries, both blinking and Shifts mark stressed syllables.
Therefore, if the speaker is not disposed to blink for one reason or another,
the functions of blinking are executed with Shifts.
In concluding, we will analyze the distribution of Shifts between the pause
and stress positions. As for Changes, their distribution (68% [102] 32% [48]) is
almost the same as the average distribution of all Shifts (64% [207] 36% [116]).
Consequently, Changes are prima facie connected with changes in Focus of Attention; the coincidence of Changes and pauses/stresses takes secondary importance.
Login is attached to internal position (92% [72] vs. 64%, on average). All types of
events that include Logout (Logout proper, Login-Logout and Logout-Login) are
attached to BBWs (65% [63] vs. 36%, on average).
Subsequently, we can conclude that only Transfers (that is Shifts that are connected with entry into the CommZ or with exit from it) have linguistic value.
Login is preferred for marking a stressed syllable, while Logout is preferred for
marking pauses (phonetic and syntactic ones).
6. Conclusions
These data show that pauses are marked with blinking and/or with Logout, while
stresses are marked with blinking and/or with Login. The natural question to ask
with this result is why are the opposite actions (Login and Logout) accompanied
by the same event (blinking)?
You may recall that we faced the same question in Section 4.2 when was described EC as an embedded gesture it accompanied an act of negation and an act
of affirmation. I believe that this blinking exhibits the same pattern.
When blinking is used as a pause mark, EC marks the exit from the CommZ;
Logout means the same; a pause itself means the termination of a communication
act.
When blinking is used as an accent mark, EC functions as the metonymic

equivalent of a nod. A nod is one of the most common ways to convey emphasis
(see Section 3.1.2, the citation from Cassell (1999:21) and also Poggi, DErrico,
and Vincze (2010)). Therefore, on the phrase level a nod accompanies a word emphasis just as on the word level blinking accompanies a stress. As for Login, in the
Introduction to the paper we argued that this gaze pattern very often is emphatic
in its deployment. So, in both cases we see the coordination of different visual,
phonetic and semantic (in the broadest sense of the word) events:
blinking + Login + emphasis = stress
blinking + Logout + break-off = pause.
It means that Da/Net reactions, or affirmation and negation, are the basic driving
forces that influence eye behavior in Russian dialogue at all possible levels.
On the level of gaze grammar, Net-reactions heighten the contrast principle
in the gaze movements, while Da-reactions do not adhere to the compensatory
principle and thereby force the speaker to look at a listener.
On the level of syntactic gesticulation, Net-reactions become the basis for
the Logout pattern and make it possible to use visual modes of discourse to mark
pauses and syntactic boundaries, whereas Da-reactions form the basis for the Login pattern and enable the speaker to mark the stressed vowels. The same results
are achieved with blinking, which is the terminus of the grammaticalization of the
embedded gesture eye closing meaning Net no (phonetic and syntactic pauses),
or of the embedded gesture eye closing meaning Da yes (accent diacritic).
Acknowledgments
The author would like to thank Michael Furman for carefully proof-reading the paper and correcting her English, all remaining errors are the authors
responsibility.
References
Argyle, Michael and Cook, Mike. 1976. Gaze and mutual gaze. Cambridge: University Press.
Argyle, Michael and Graham, Jean A. 1976. The Central Europe experiment: Looking at persons and looking at things. Journal of Environmental Psychology and Nonverbal Behavior
1: 616.
Bavelas, Janet B. 2005. Appreciating face-to-face dialogue. Auditory-Visual Speech Processing. Auditory-Visual Speech Processing (AVSP05I, British Columbia, Canada July 2427).
81
82
Elena Grishina
Bavelas, Janet B., Coates, Linda and Johnson, Trudy. 2002. Listener responses as a collaborative
process: The role of gaze. Journal of Communication 52: 566580.
Beattie, Geoff. 1980. The role of language production processes in the organization of behavior in face-to-face interaction. In Language Production, Vol. 1, Brian Butterworth (ed.),
69107. London: Academic Press Inc.
Cassell, Justine. 1999. Embodied Conversation: Integrating Face and Gesture into Automatic
Spoken Dialogue System. In Spoken Dialogue Systems, Susann Luperfoy (ed.). http://
www.justinecassell.com/discourse/pdfs/Cassell.ECA_chapter.handout.pdf (latest access
25/10/2012).
Cienki, Alan and Mller, Cornelia. 2008. Metaphor, gesture, and thought. In The Cambridge
Handbook of Metaphor and Thought, Raymond W. Gibbs, Jr. (ed.), 483501. Cambridge:
Cienki, Alan. 2005. Image schemas and gesture. In From perception to meaning: Image schemas in cognitive linguistics, Beate Hampe and Joseph E. Grady (eds.), 421442. Berlin/New
York: Mouton de Gruyter.
Driver, Jon et al. 1999. Gaze perception triggers reflexive visuospatial orienting. Visual Cognition 6: 509540.
Duncan, Susan. 1972. Some signals and rules for taking speaking turns in conversations. Journal of Personality and Social Psychology 23 (2): 283292.
Fehr, Barbara J. and Exline, Richard V. 1987. Social visual interaction: A conceptual and literature review. In Nonverbal behavior and communication, Aron W. Siegman and Stanley
Feldstein (eds.), 225326. Hillsdale, N.J.: Erlbaum.
Grishina, Eelena. 2009a. Multimodal Russian Corpus (MURCO): general structure and user
interface. NLP, Corpus Linguistics, Corpus Based Grammar Research. Fifth International
Conference, Jana Levick and Radovan Garabk (eds), 119131. Brno: Tribune.
Grishina, Elena. 2009b. Multimodal Russian Corpus (MURCO): Types of annotation and annotators workbenches. In Corpus Linguistics Conference CL2009, Universuty of Liverpool,
UK, 2023 July 2009, http://ucrel.lancs.ac.uk/publications/cl2009/#papers ( latest access
25/10/2012).
Gullberg, Marianne and Holmqvist, Kenneth. 1999. Keeping an eye on gestures: Visual perception of gestures in face-to-face communication. Pragmatics and Cognition 7: 3563.
Gullberg, Marianne and Holmqvist, Kenneth. 2006. What speakers do and what listeners look
at. Visual attention to gestures in human interaction live and on video. Pragmatics and
Cognition 14: 5382.
Gullberg, Marianne and Kita, Sotaro. 2009. Attention to Speech-Accompanying Gestures: Eye
Movements and Information Uptake. Nonverbal Behavior 33 (4): 251277.
Kendon, Adam. 1990. Conducting interaction. Cambridge: Cambridge University Press.
Kleinke, Chris L. 1986. Gaze and eye contact: A research review. Psychological Bulletin 100:
78100.
Krejdlin, Grigorij E. 2004. Neverbalnaja semiotika. Moskva: RSUH.
Loehr, Dan. 2007. Aspects of rhythm in gesture and speech. Gesture 7 (2): 179214.
Mittelberg, Irene. 2007. Methodology for multimodality: One way of working with speech
and gesture data. In Methods in Cognitive Linguistics, Monica Gonzalez-Marquez, Irene
Mittelberg, Seana Coulson and Michael Spivey (eds.), 225248. Amsterdam/Philadelphia:
John Benjamins.
Novick, David G., Hansen, Brian and Ward, Karen. 1996. Coordinating Turn-taking with
Gaze. Proceedings of ICLSP-1996 3: 18881891.
Poggi, Isabella, DErrico, Francesca and Vincze, Laura. 2010. Types of Nods. The polysemy of a
social signal. Proceedings of LREC-2010. http://www.lrec-conf.org/proceedings/lrec2010/
pdf/596_Paper.pdf (latest access 25/10/2012).
Prozorova, Evgenija V. 2009. Markery lokalnoj struktury diskursa v russkom estovom jazyke.
Moskva: MSU.
Torres, Obed E., Cassell, Justine and Prevost, Scott. 1997. Modeling Gaze Behavior as a Function of Discourse Structure. First International Workshop on Human Computer Conversations. Bellagio, Italy.
Uryson, Elena V. 2003. Problemy issledovanija jazykovoj kartiny mira: Analogija v semantike.
Moskva: Jazyki slavjanskoj kultury.
Wiemann, John M. and Knapp, Mark L. 1975. Turn-taking in Conversation. Journal of Communication 25: 7592.
83
Hesitation markers in transitions

within (story)telling sequences
of Russian television shows
Hanna Laitinen
In this paper I argue that in Russian television shows a transition within telling
sequences may cause hesitation. The focus of this article is on investigating the
types of transition that invoke hesitation phenomena. The results reveal that the
most common transition types succeeded by hesitation markers, relative to all
transitions within telling, are between abstract and orientation or evaluation,
and between orientation or evaluation and resolution. In addition, this paper
suggests that Labovs (1972) model is rather well applicable in defining the location of hesitation markers not only in storytelling, but also in more general telling sequences.
Keywords: hesitation markers, transitions, telling sequences, Russian
1. Introduction
Talk-in-interaction is produced online in collaboration with interlocutors and is
aimed at them. These features, together with the cognitive capacity of human beings, invoke hesitation phenomena which are by their nature problems related to
speech production (e.g. Biber et al. 1999:10411067; Chafe 1980b). Such phenomena are rather frequent, as it has been estimated that approximately 6% of talk is
dysfluent (Fox Tree 1995:710; cf. also Biber et al. 1999:10531060). The hesitation
phenomena emerge in talk in different forms: filled and unfilled pauses, lengthening of syllables, monoton pitch, repeats, so-called thinking face, changes in the
direction of gaze and gestures, conventionalized lexical markers of word search,
and metacognitive and metalinguistic phrases (e.g. Clark and Fox Tree 2002:104;
Goodwin 1987; Goodwin and Goodwin 1986; Schegloff et al. 1977:367; in Russian Aleksieva 2007; Podlesskaja and Kibrik 2009:178187). In this article these
forms are called hesitation markers (HMs).
86 Hanna Laitinen
Since the 1950s, when Lounsbury (1954) launched hypotheses about the typical location of occurrence of hesitation (pauses) in speech, that is to say, when there
is high transitional entropy regarding the following units of speech (ibid.:99100),
particularly Anglo-American studies have suggested that HMs, especially filled
and unfilled pauses, typically appear in speech near the absolute beginning of an
utterance (Boomer 1965; Shriberg 1994:103). In this kind of position, including
the absolute beginning of a turn and of a sequence, these markers have been interpreted as signaling trouble on a larger scale than merely a local word search. This
involves either semantic or structural planning (e.g. Fischer 2000:160; Gardner
2001:88; Shriberg 1994:154) or a shift in orientation (Chafe 1980a; 1980b).
When investigating the location of HMs relative to the sequences in conversation, it was found that HMs, especially filled pauses, occur quite often in transitions of a different kind: they indicate the beginning of a new topic (Schegloff
1979:270272, 276; 2010) and of a new sequence (Schegloff 2010). Furthermore,
they are used in order to exit and re-exit the sequence (Schegloff 2009) and to
mark a thematic break (Fischer 2000:160).
HMs have been also found in storytelling (Chafe 1980a; 1980b). According to
Chafe (1980b:172176, see also 1980a:4349) hesitation markers typically occur
in certain phases of a story. These phases are the beginning of the story, transition
between images in a cartoon, shift to a new goal, shifts between episodes of the
story, introduction of a new character or set of characters of the story, and various kinds of changes in the story including change of location, time period, event
schema, and world. Chafes (1980a:47) findings indicate that a shift of worlds
causes maximum hesitation. He explains the phenomenon by the maximum reorientation that the shift of worlds requires. In other words, the author (ibid.:49)
proposes the hypothesis according to which the amount of hesitation increases
with the amount of reorientation the speaker must undergo in his or her own
consciousness and must transfer to the consciousness of the listener.
Recently, there has also been growing interest in hesitation phenomena in
Russian. The research has tended to focus on the taxonomy of different markers of hesitation phenomena (e.g. Aleksieva 2007; Podlesskaja and Kibrik 2009),
the interrelationship between hesitation and different types of repair (Podlesskaja
and Kibrik 2007; 2009), and the location of HMs relative to the structural features
of language (Korotaev, Kibrik and Podlesskaja 2009:318; Podlesskaja and Kibrik
2007:911; 2009:184187) rather than relative to the sequences in conversation
(cf. melev 2004), which may turn out very important when discussing the role of
HMs in talk-in-interaction.
This article explores the location of HMs in the transitions within telling
sequences (e.g. Jefferson 1978; Sacks 1974) in television shows. The theoretical
background is based, among others, on the model (theory) of Functional Syntax
(Mustajoki 2006). I also take into consideration psycholinguistic studies of speech

production phases (e.g. Jay 2003; Levelt 1989; Shriberg 1994). The framework for
the analysis is provided by conversation analysis (CA) (e.g. Sacks 1992a; 1992b),
interactional linguistics (e.g. Ochs et al. 1996), and Labovs (1972; cf. also Labov
and Waletzky 1967) model of the internal structure of narrative. The focus of this
article is on investigating the types of transition that invoke hesitation phenomena. I will show that the most common transition types are located either at the
beginning of the telling sequence or near its end. In addition, I will suggest that
the Labovian model is rather well applicable in defining the location of HMs not
just in the telling sequence of a story but also in more general telling sequences.
The data for the study consist of four fragments of Russian television shows
in length of 1 h 15 min. In all the interviews the focus is on the experiences of
the interviewees. Because of the genre and the foci in the shows, the turns tend to
become long: they become multi-unit turns (e.g. Schegloff 1996:61) and include
several telling sequences.
2. Hesitation as a cognitive-pragmatic category

The theoretical background for this paper includes Functional Syntax, with its
fundamental goal of describing language with the semantic structures as a point of
departure. In general, this means that when the speaker wishes to convey a particular meaning, he or she may use different constructions to express him or herself.
In other words, the description starts from the semantic categories. These semantic categories are expressed in various ways, and each means of expression provides additional information about the meaning (Mustajoki 2006:1824,8284).
As to hesitation, the speaker does not normally wish to hesitate, but rather
encounters a speech production problem or a trouble in interaction and produces
a HM. Therefore, I suggest that different kinds of speech production and interaction problems constitute not a semantic category, but rather a cognitive-pragmatic category, namely hesitation. I use here the term cognitive-pragmatic because, as
proposed earlier, the problem may occur mainly because of speech production,
which is a cognitive aspect, or because of interaction, which is a pragmatic aspect.
Although hesitation is conveyed by different means, interlocutors are able to recognize when the speaker is encountering a trouble related to speech production
or interaction, which may induce them to offer help (cf. Goodwin 1987; Goodwin
and Goodwin 1986; Lerner 1992:256257; 1996:261263).
Like semantic categories, the category of hesitation may occur in speech in
various ways. These HMs are lexical markers of word search in a given context
(e.g. nu well), metacognitive and metalinguistic phrases (e.g. kak to skazat how
87
88
Hanna Laitinen
to say), filled pauses that are often delivered in a monotone pitch and prolonged.
In addition, the same category includes prolonged syllables (with monotone
pitch), repeats of prepositions, pronouns, and conjunctions, and also, in some
contexts, unfilled pauses.
3. Stories and storytelling

Oral stories have been investigated in various linguistic sub-fields and in other
social science disciplines. In CA the focus has been on investigating how stories
are constructed in cooperation by interactants. It has been demonstrated that storytelling is sequentially organized (e.g. Jefferson 1978; Sacks 1974). According to
Sacks (1974), the three successive sequences are the preface sequence, where the
telling is offered to the interlocutors by the intending teller, the telling sequence,
and the response sequence, which provides the evaluation of the story by the recipient. The action of telling happens in the telling sequence, which consists of
several TCUs. The model of the sequential organization of storytelling has been
abstracted by using everyday conversation.
Apart from the sequentiality of telling, another important concept is tellability.
Tellability means the reason for telling. There must be some relevance or newness
of a reported issue in a particular situation (Sacks 1992b:1213, 172173,233).
Therefore, tellability has an influence on whether the story will be told at all and
how it will be told (ibid.; cf. also Ochs and Capps 2001:34).
The telling sequence may be mostly produced by one speaker (cf. however
co-telling, e.g. Lerner 1992; Norrick 2000:8387, 126132). Thus, as Routarinne
(2003:55) points out, the focus of the interaction-oriented researchers has not
very commonly been on the internal structure of the telling sequence. Nevertheless, when conversation analysts have been interested in how the telling sequence
is constructed, they have often applied to some extent Labovs (and Waletzkys)
model of the structure of narrative (e.g. Norrick 2000:2943; Routarinne 2003)
despite the problematic nature of some of its basic conceptions (cf. discussion
e.g. in Georgakopoulou 2011:389; Schegloff 1997). For instance, one problem
is that the model was generated not by observing authentic data, but by asking
informants whether they were ever in serious danger of being killed (Labov
1972:354). Therefore, as Schegloff (1997:98101) formulates it, the model lacks
interactional aspects.
Narrative has been defined by Labov and Waletzky (1967:20; cf. also Labov
1972:360) as one method of recapitulating past experience by matching a verbal
sequence of clauses to the sequence of events which actually occurred. Hence,
a minimum of two clauses must be temporally ordered, and these two clauses
form a minimal narrative (Labov 1972:360361; cf. also Labov 1997; Labov and
Waletzky 1967:25). This definition reveals that the scholars take into consideration both semantic and structural criteria. They also suggest both criteria for
the elements of the narrative (cf. Labov 1972; 1997; Labov and Waletzky 1967). I
focus on the semantic aspects because they seem to be less language-specific than
the structural criteria (cf. e.g. Mustajoki 2006:378390). In addition, semantic
criteria are also applicable in Functional Syntax, as already mentioned.
The only compulsory elements of the narrative are complicating action (CA)
and resolution (or result) (R). Nonetheless, there are other elements that are found
in more fully-formed narratives. These elements are abstract (A), orientation (O),
evaluation (E), and coda (Co). As suggested, abstract summarizes the whole story,
orientation provides background information for the story, and complicating action reports the series of events. Evaluation offers information on the attitude of
the narrator towards the narrative: in other words, it indicates the point of the story. Evaluation may be either external or embedded in other parts of the narrative,
even throughout the whole narrative. Finally, resolution offers a response to the
question of what finally happened, and coda indicates that the question of what
happened has already been answered. The function of the coda is to bring the
interactants back to the present moment from the world of narrative. The order of
different elements may vary slightly and they may be chained and fused with one
another (Labov 1972:362373; Labov 1997; Labov and Waletzky 1967:3241).
Before moving on to present the data and methods, I will briefly discuss the
Labovian model in the context of conversation analytic and other interactionoriented studies. As has been demonstrated, telling as part of conversation does
not always fulfill the definition of a narrative according to Labovs model (e.g.
Routarinne 2003:6063). Instead, telling has been seen as a more general action,
going beyond storytelling or narratives (cf. e.g. Ochs and Capps 2001:1820;
Routarinne 2003:66) and it may be intertwined with other actions as well (Ochs
and Capps 2001:40). Hence, as Routarinne (2003:63; cf. also Norrick 2000:29
38; Ochs and Capps 2001:3) argues, Labovs model should be seen as a prototype of storytelling, not as a definition. Nevertheless, the Labovian model and
the conversation analytic view of telling also have something in common. First,
abstract and preface are sequentially located in a similar place, although their
functions are reported to be different (cf. Routarinne 2003:5657). As already
mentioned, preface is the offer of a story, not a summary of a story like abstract.
However, these functions may intertwine, as one can conclude by observing the
example presented by Routarinne (ibid.:4041, 57). Second, evaluation can be
seen as a device of constructing tellability (cf. reportability, Labov 1997). Third,
conversation analysts have shown that coda or evaluation may emerge at the end
of storytelling as a result of an interactional problem, namely, a lack of response
89
90 Hanna Laitinen
sequence by the interlocutors (Goodwin and Heritage 1990; Schegloff 1997:102;

cf. also Jefferson 1978:228237; Routarinne 2003:5659).
4. Data and methods

The data for this study consists of four samples of television shows, totaling 1
hour 15 minutes (the length of the different fragments ranging from 14 minutes
to 24 minutes). The programs were shot and broadcast on Russian television in
the 1990s after the collapse of the Soviet Union. There are 17 informants in the
television interviews from different social backgrounds (including sex, age, and
occupation). In all the interviews the focus is on the experiences of the interviewees. This may induce telling. Some of the informants are interviewed in their professional capacity, and some are invited to the television studio because of their
conflicting attitudes toward the topic of the show. Therefore, one can suggest that
the data seems to be rather varied.
Consisting of fragments of television shows, my data can be characterized as
being less spontaneous than everyday conversation. However, television shows
have features that both decrease and increase spontaneity. Elements that decrease
spontaneity are, for instance, the institutional character of the interaction, including the different, asymmetric roles that the interactants adopt as they perform
their institutional task (i.e. the roles of host and guest), and the existence of an
agenda that both hosts and guests orient to (cf. e.g. Drew and Heritage 1992:22
25, 49; Lapteva 1990:927; Nuolijrvi and Tiittula 2000:6468, 135220; Popova
2002). In addition, the participants of the show usually do not speak the same
variant of spoken Russian as they do in everyday conversation, i.e. their speech
is not razgovornaja re, but rather a variant of public communication (raznovidnost ustnoj publinoj rei / publinoe obenie) (Lapteva 1990:1627; Popova
2002:34). Nevertheless, as Lapteva (1990:17, 22) points out, conversation in
television shows, as long as it does not entail reading aloud, is to some extent
spontaneous because its production coincides with the speech production phases
abstracted in psycholinguistics. Another increasing element of spontaneity is the
appearance of guests. They may not accept the agenda, turn-allocation, or roles
that have been offered to them (cf. Nuolijrvi and Tiittula 2000:164170, 175
180). Therefore, the genre of the data may have some influence on the hesitation
phenomena in general. For instance, the variety of different hesitation markers
seems to be smaller than in other kinds of data. In the Rasskazy o snovidenijax
corpus, which consists of dream stories told by children and youngsters, approximately half of them with a neurological impairment, there was a greater variety of
lexical markers of hesitation (e.g. sejas now, tot samyj this very, kak ego how
(to call) it) than in my data (Podlesskaja and Kibrik 2007; 2009).
I transcribed the data using the transcription system developed by Gail
Jefferson (cf. Atkinson and Heritage 1984:ixxvi) that has become conventionalized in CA. In the transcription the arrow () indicates a monotone pitch (cf.
Clark and Fox Tree 2002:104). The names and other identifiable references were
changed in the transcription. The transcription was conducted by ear, and approximately of it afterwards verified using the speech analyzer program Praat.
All the examples presented in this paper were checked with the help of that program. The translation into English in the third line is as literal as possible. However, I did not mark in it HMs or any prosodic features because these phenomena
are to a large extent language-related (cf. Janko 2008).
The genre of my data had some practical influence on the analysis. First, most
of the data was not broadcast live, and the programs were edited before being
aired. Therefore, in some cases where the possible HMs were located in the neighborhood of the edited spot, I was not able to discuss their functions comprehensively. In such situations I considered the HMs as possible indicators of hesitation.
Second, I did not choose the angle of view. Thus, I was not able to take systematically into account the multimodal aspects of interaction such as gaze and gestures
that might be related to marking hesitation (cf. e.g. Goodwin 1987; Goodwin and
Goodwin 1986).
Besides using the transcription system developed in CA, I used the apparatus and terminology of CA. In this paper the most important concepts are those
related to the organization of telling, and also projection and TCU. By projection I mean the possibility that an ongoing action and syntactic construction may
foreshadow its possible completion or another action (cf. Auer 2005; Sacks et al.
1974:702). TCU is the smallest interactional unit in its context with a syntactic
and perhaps also a prosodic and a pragmatic completion (cf. Sacks et al. 1974:702;
Schegloff 1996:59; Selting 2000).
Furthermore, I applied to some extent the psycholinguistic models of speech
production phases. According to psycholinguists, speakers tend to plan at first
the larger semantic meaning and then smaller units, such as words (e.g. Jay 2003
and suggested literature; Levelt 1989). Therefore, I assume that when the speaker
hesitates in transitions the problem is probably not the following word, but there
is trouble on a larger scale.
I identified the storytelling sequences in the data using the definition of minimal narrative by Labov. With this definition I found 16 storytelling sequences. In
addition, telling was defined as a sequence where some kind of change has taken,
will take, or might take place. The number of more general telling sequences was
65. These 81 samples were distributed in all four fragments of television shows
91
92
Hanna Laitinen
as follows: the first contained 15 samples, the second 28, and the third and the
fourth 19 samples each. In presenting the results I will discuss these two categories together.
In identifying HMs related to transitions, I leaned on psycholinguistic studies. If HMs indicate a problem related to transition, they have to be located initially relative to a TCU (cf. Shriberg 1994:154; Swerts 1998). This means that the
TCU should not project anything syntactically before the occurrence of HMs.
Hence, there are few options for the location. HMs may appear at the absolute beginning of a TCU, or they may follow either conjunction or particle (cf. Shriberg
1994:5051). Other locations that I assumed relevant to this paper were TCUmedial positions succeeded by a parenthetical sequence or the abandonment of
an ongoing TCU, because in both cases the speaker self-oriented to a shift (cf. Fox
et al. 1996:189; Routarinne 2003:7085).
5. Hesitation markers in the transitions within storytelling

and more general telling sequences in the data
In this section, I argue that the Labovian model can be usefully applied in locating
the HMs in transitions within the storytelling sequences, and also when telling
is seen more widely. I also demonstrate that the different elements of a narrative
may intertwine in a natural conversation, in this case in a television interview.
The following example illustrates the frequency of occurrence of HMs in the
storytelling sequence. The host (H) introduces the guest by telling a story about
how they became acquainted. This excerpt is from the beginning of the whole
television interview.
(1) Excerpt: First time
1 H: .hh ja pomnju: kakie kakoe ouenie u
menja bylo.=

I remember
what-pl what feeling
with me
was
I remember what kinds of what kind of feeling I had
2 =kogda ja posmotrel pervyj .hh (.) raz film :
when I watched
first
time film
when I watched for the first time the film
3 Bobrova.=
to byl : film .hh (.) Xleb i
sol?
of last name that was
film
bread and salt
made by Bobrov it was the film Bread and Salt
4 >> .hh (.) : ja soverenno ne oidal,=:: znait pered

I absolutely not expected
so
before
I had absolutely no expectations so before
5 filmom pokazyvali kakim-to igrovym tot, .hh

film
showed
some-cli feature that
a feature film it was shown that
6 korotkij multiplikacionnyj film?=
short
cartoon
film
short cartoon film
7 =ja soverenno niego ne oidal? .hh ot
nego i
byl
I absolutely nothing not expected
from it
and was
I had absolutely no expectations at all of it and I was
8 nastolko <poran> tem
to uvidel?
so
dazzled by that what saw
so dazzled by what I saw
9 .hh to >sejas
e stal razyskivat reissra.<=

that immediately prt began to look for director
that I immediately began to look for the director
10 =priexal k nemu na studiju?
came
to him to studio
came to him to the studio
11 > .hh i: :: sostojalos nae znakomstvo.--
and
took place our acquaintance
and we became acquainted
The host starts with an abstract (lines 13): .hh ja pomnju: kakie kakoe ouenie u menja bylo.=kogda ja posmotrel pervyj .hh (.) raz film : Bobrova. I
remember what kinds of what kind of feeling I had when I watched for the
first time the film made by Bobrov. It would seem that the TCU fulfills at the
same time the function of evaluation and orientation because it provides the circumstances for the story and also evaluates the whole forthcoming action. The
orientation continues in line 3, when the host mentions the name of the film. In
line 4, he hesitates (:) and moves on to the complicating action with an embedded evaluation: ja soverenno ne oidal,= I had absolutely no expectations.
Then he cuts off and again produces a HM. After that, he shifts to the orientation
(lines 46): znait pered filmom pokazyvali kakim-to igrovym tot, .hh korotkij multiplikacionnyj film?= so before a feature film it was shown that short
cartoon film. After completing this TCU, he repeats the interrupted evaluative
complicating action and continues it by first using complicating action fused with
evaluation (lines 78): =ja soverenno niego ne oidal? .hh ot nego i byl nastolko
<poran> tem to uvidel? I had absolutely no expectations at all of it and I was
so dazzled by what I saw and after that, purely complicating action (lines 910):
.hh to >sejas e stal razyskivat reissra.<=priexal k nemu na studiju? that I
immediately began to look for the director came to him to the studio. Then, in
93
94 Hanna Laitinen
line 11, he hesitates again (i: ::) and offers the resolution of the story: sostojalos
nae znakomstvo. we became acquainted.
This example demonstrates that the speaker might hesitate in some transitions
within storytelling sequences, but not always. On three occasions a HM occurs in
a transition: first, between orientation and complicating action (line4), second,
vice versa (line 4), and third, from complicating action to resolution (line11).
However, the excerpt also demonstrates that not all HMs occur in transitions (cf.
lines 13) and that the speaker may produce a shift in a narrative without hesitating (cf. lines 3 and 7). Nevertheless, in this excerpt, all the TCU-initial HMs and
a parenthesis-initial HM, the latter being at the same time succeeded by an abandonment of an ongoing TCU, occur in a location where the speaker produced a
transition. Thus, the Labovian model captures the HMs rather well in these three
types of position.
The second example suggests that different elements of narrative may also
be found in more general telling sequences. Like the first example, the second
example also shows that HMs may occur in transitions. After finishing the story
(cf. example 1), the host evaluates the guest as an artist.
(2) Excerpt (continued from the excerpt 1): Wonderful man
12 > H: .hh : ja obnaruil, : udivitelnogo

I discovered
wonderful
I discovered a wonderful
13 eloveka(-to). : : tj soverenno .hh
man-cli
absolutely
man of absolutely
14 neobyknovennoj kultury, mjagkosti, tonkosti,
unusual
culture
gentleness subtlety
extraordinary culture of gentleness of subtlety
15

16 >>

.hh (.) : udivitelnoj

uvstvitelnosti.
of astonishing sensitivity
of astonishing sensitivity
.hh : v obem.=: Andrej
Bobrov to :
in general
first name last name this
in general Andrej Bobrov he is
17 > nastojaij xudonik. (0.6) xudonik : kotoromu

real
artist
artist
to whom
a real artist an artist to whom God
18 otpustil oen edro. .hh talanta i
uma.-- gave
very lavishly
talent and intellect
gave very lavishly talent and intellect
Bog
God
After finishing the story the host starts to evaluate the animator and his production. Therefore, it may be assumed that all HMs occurring in the telling sequence
indicate to some extent a problem in evaluation. However, my aim is to explore
the position of the HMs more closely.
The sequence begins with a HM (:). It is located between the resolution
of the story (example 1, line 11) and the new evaluative sequence. Evaluation
from lines 12 to 15 resembles a complicating action because it is constructed as
a mental and perhaps also a physical action: ja obnaruil,--- I discovered---.
After this evaluation section, the host hesitates twice in a transition in line 16: :
v obem.=: HM in general HM and moves on to a static evaluation: Andrej
Bobrov to : nastojaij xudonik. Andrej Bobrov he is a real artist. Then the
speaker probably hesitates during a silent pause (0.6) in line 17 because soon after
that he again encounters trouble. Continuing, he offers an evaluation: xudonik
: kotoromu Bog otpustil oen edro. .hh talanta i uma.--- an artist to whom
God gave very lavishly talent and intellect---. It is slightly differently constructed
in comparison with the previous one. Now Andrej Bobrov is characterized as recipient. He is still represented as a passive actant (aktant), but now there is also an
active actant, who has given him something (cf. Mustajoki 2006:156173). Thus,
the evaluation again resembles a complicating action.
To conclude, similar to the storytelling sequences, in more general telling
most of the HMs occurring in the TCU-, parenthesis-, or syntactic abandonment-initial position are succeeded by a transition. In excerpt 2, HMs are only
twice produced in one of the above-mentioned locations without a transition between the elements of the narrative (cf. lines 13, 15). Furthermore, there are three
transitions that coincide with the occurrence of HMs: from resolution to evaluation that resembles a complicating action (line 12), from evaluation, resembling a
complicating action to (static) evaluation (line 16), and vice versa (line 17).
However, the two examples presented in this section also demonstrate, as
Labov (1972:363, 369) has pointed out, that the different elements of narrative
may indeed intertwine. Especially in example 2, the shifts with occurrence of
HMs are mostly not very substantial, possibly because of this phenomenon. Transitions seem to appear more commonly between sub-categories.
6. The distribution of hesitation markers in the transitions within telling

In my data the total number of HMs and strings of them was 625. Besides, there
were 373 markers that implied or might have implied hesitation among the other
functions that they fulfilled. HMs were found in the turns of all the 17 informants,
and 13 of them hesitated in the transitions within telling. There were 122 HMs
95
96 Hanna Laitinen
in transitions. In addition, 57 markers either carried or might have carried the

implication of hesitation among other functions. Therefore, approximately 20% of
all hesitation markers occurred in transition within either storytelling or a more
general telling sequence. Taking into account also the markers with possible implication of hesitation, the ratio was 18%. Table 1 displays not just HMs, but also
linguistic devices with a possible implication of hesitation occurring in the transition. Thus, the total number was 179 samples.
As to the results, I have made generalizations in presenting them because, as
illustrated in the previous section, different elements of narrative may be intertwined. In order to present Table 1, I have defined the basic function of each TCU.
As to orientation and evaluation, according to my findings they seem to be fused
rather often. They both tend to offer hints for the interpretation of what the point
of the telling sequence is. Thus, I have usually presented them together. I have
not illustrated HMs within transitions in storytelling and more general telling
sequences separately because, as the examples in the previous section revealed,
HMs seem to locate in rather similar positions in both types of telling.
According to the results (see Table 1), various kinds of transitions might
cause hesitation. The most troublesome transition seemed to be a minor transition within a category (25 samples of HMs, 35 samples of HMs & possible HMs).
Transitions from orientation or evaluation to complicating action also tended to
cause trouble (20 samples of HMs, 29 samples of HMs & possible HMs). In addition, when the speaker moved on from the complicating action to orientation
Table 1. HMs and possible HMs in transitions within (story)telling sequences
A O/E:
O/E C A:
O/E R:
O E:
E O:
C A O/E:
C A R:
R O/E:
R Co:
Other transitions occurring
seldom, < 10
Other minor transitions, for
instance Evaluative O O
HMs HMs &

All
possible HMs transitions
HMs / HMs & possible

All % HMs / All %
7
20
14
3
3
20
11
13
3
3
9
29
24
6
8
26
13
17
5
7
16
89
44
18
20
72
37
43
11
24
44%
22%
32%
17%
15%
28%
30%
30%
27%
13%
25
35
not counted
122
179
374
56%
33%
55%
33%
40%
36%
35%
40%
45%
29%
Table 2. HMs & possible HMs in transitions and in other locations of (new) beginning
HMs & possible HMs
in transition
All TCU-, parenthesis-, syntactic abandonment- %

initial HMs & possible HMs
179
449
40%
HMs in transition
TCU-, parenthesis-, syntactic abandonmentinitial HMs
122
271
45%
or evaluation, he or she appeared to encounter problems (20 samples of HMs, 26

samples of HMs & possible HMs).
Approaching the results relative to the total amount of transitions, the most
troublesome transition seemed to be between abstract and either orientation or
evaluation: 44% of such transitions were accompanied by HMs, and if we also
take into consideration markers with a possible indication of hesitation in a similar position, the ratio was 56%. The second most problematic transition according to the results was a shift between orientation/evaluation and resolution. HMs
occurred in 32% of such shifts. The ratio was considerably higher, 55%, with possible HMs also taken into account. There were also rather often problems in the
shift between resolution and coda (27% and 45%) and between resolution and
orientation or evaluation (30% and 40%). Thus, one can draw a conclusion that
especially the beginning and finishing of the telling sequence causes trouble.
The results with and without the markers involving only a possible indication
of hesitation were rather similar. The three most troublesome transitions were
the same with both categories taken into account. Relative to the total number of
occurrences of each transition type, the two transitions most frequently causing
problems turned out to be the same.
As to all the HMs and markers with a possible implication of hesitation that
are located in TCU-initial position or at the beginning of a parenthesis or a syntactic abandonment, 40% of them occur in a transition within either storytelling or a more general telling sequence (see Table 2). Furthermore, taking into
consideration only the pure HMs in the same position, 45% of them occur in
transitions.
7. Conclusion
In this paper, I first outlined, leaning on previous, mostly Anglo-American studies, the occurrence of HMs in transitions of different kinds and in storytelling.
97
98 Hanna Laitinen
This study demonstrated that Russian also shows a relatively strong correlation
between the HMs and shifts in telling.
The results indicated that especially initiating the telling and also finishing it
rather often causes trouble for the speaker. There are several possible explanations
for this. First, the result may be explained by the hypothesis launched by Chafe
(1980a:49). Indeed, the hesitation might increase with a reorientation in the consciousness of the speaker. When the speaker transfers from abstract to orientation
or evaluation, from orientation or evaluation to resolution, vice versa, and from
resolution to coda, the viewpoint of telling might also change quite significantly.
Likewise, the shift in consciousness seems to be fairly extensive in the case of
moving from orientation or evaluation to complicating action and vice versa.
However, the occurrence of HMs in transitions between sub-categories does not
clearly support the idea of Chafe.
Second, the shifts causing trouble may be explained by the concept of tellability. The speaker might encounter trouble because of hesitating over whether something is tellable or not, whether something should be said right now, or whether
one should give some background information immediately. This approach also
takes into consideration the interactional aspect of telling. Third, the problems
between the resolution and coda or evaluation might arise due to misunderstanding about the end point of the telling. As already suggested, coda or evaluation
may emerge at the end of storytelling as a result of a lack of response sequence
by the interlocutors (cf. Goodwin and Heritage 1990; Schegloff 1997:102; cf. also
Jefferson 1978:228237; Routarinne 2003:5659). Thus, the speaker might hesitate after producing a resolution if the interlocutor does not take the turn.
Additionally, this paper pointed out that the Labovian model can be usefully applied in investigating the HMs in transitions within storytelling and more
general telling sequences. As observed earlier, approximately 18% of all HMs or
markers with a possible implication of hesitation are located in transitions within
either storytelling or more general telling sequences. Considering HMs occurring in transitions, relative to HMs in all locations of (new) beginning, the ratio is
45%. Nevertheless, as the results indicate, most of the transitions (25 samples of
HMs, 35 samples of HMs & possible HMs) occurred within the same category in
Labovs model. Therefore, some elaboration is required for a more careful analysis. One possible solution could lie in using the tools of Functional Syntax for
describing semantic structure, as I did to a small degree.
As to the sufficiency of the data, although the total length of the television
shows might seem rather small, the total amount of HMs and the markers that
. I am grateful to Christian Sappok (p.c.) for suggesting this motivation.
implied or might have implied hesitation among the other functions that they
fulfilled was rather large. Besides, 122 of HMs and 57 of possible HMs occurred
in the transitions of telling sequences. The number of telling sequences was also
considerable (81 samples), and they were distributed, as noted above, in all four
fragments of television shows. The different kinds of transitions that cause trouble
were found in all four television shows and in the turns of various informants (13)
from different social backgrounds. Therefore, one can argue that we are dealing
with a rather widespread phenomenon in which the role of idiolectal factors is
not decisive.
Because of the multifunctionality of linguistic devices and the relatively large
number of possible HMs, the results concerning their distribution that I have presented in this paper must be viewed with some caution. However, the results were
rather similar when discussing the distribution of HMs in transition and when
taking into consideration markers that were possibly indicating hesitation among
other functions that they fulfill. Consequently, the findings enable us to conclude
that a shift in a storytelling or telling sequence appears to be both a pragmatic and
a cognitive action that may cause hesitation.
References
Aleksieva, Tatjana I. 2007. O javlenijax xezitacii v spontannom diskurse. In Mir russkogo
slova i russkoe slovo v mire, 275280. Tom 1. Sofia: Heron Press.
Atkinson, J. Maxwell and Heritage, John (eds.). 1984. Structures of Social Action: Studies in
Conversation Analysis. Cambridge: Cambridge University Press.
Auer, Peter. 2005. Projection in interaction and projection in grammar. Text 25 (1): 736.
Biber, Douglas, Johansson, Stig, Leech, Geoffrey, Conrad, Susan and Finegan, Edward. 1999.
Longman Grammar of Spoken and Written English. London: Longman.
Boomer, Donald S. 1965. Hesitation and grammatical encoding. Language and Speech 8:
148158.
Chafe, Wallace L. 1980a. The deployment of consciousness in the production of a narrative. In
The Pear Stories: Cognitive, Cultural and Linguistic Aspects of Narrative Production, Wallace
L. Chafe (ed.), 950. Norwood, NJ: Ablex.
Chafe, Wallace L. 1980b. Some reasons for hesitating. In Temporal Variables in Speech: Studies in Honour of Frieda Goldman-Eisler, Hans W. Dechert and Manfred Raupach (eds.),
169180. The Hague: Mouton.
Clark, Herbert H. and Fox Tree, Jean E. 2002. Using uh and um in spontaneous speaking.
Cognition 84: 73111.
Drew, Paul and Heritage, John. 1992. Analyzing talk at work: An introduction. In Talk at work:
Interaction in institutional settings, Paul Drew and John Heritage (eds.), 365. Cambridge:
Fischer, Kerstin. 2000. From Cognitive Semantics to Lexical Pragmatics: The Functional Polysemy
of Discourse Particles. Berlin/New York: Mouton de Gruyter.
99
100 Hanna Laitinen
Fox, Barbara A., Hayashi, Makoto and Jasperson, Robert. 1996. Resources and repair: A crosslinguistic study of syntax and repair. In Interaction and grammar, Elinor Ochs, Emanuel
A. Schegloff and Sandra A. Thompson (eds.), 185237. Cambridge: Cambridge University
Press.
Fox Tree, Jean E. 1995. The effects of false starts and repetitions on the processing of subsequent words in spontaneous speech. Journal of Memory and Language 34: 709738.
Gardner, Rod. 2001. When Listeners Talk: Response Tokens and Listeners Stance. Amsterdam/
Philadelphia: John Benjamins.
Georgakopoulou, Aleksandra. 2011. Narrative analysis. In The SAGE Handbook of Sociolinguistics, Ruth Wodak, Barbara Johnstone and Paul Kerswill (eds.), 388403. Los Angeles,
CA/London: SAGE.
Goodwin, Charles. 1987. Forgetfulness as an interactive resource. Social Psychology Quarterly
50 (2): 115131.
Goodwin, Charles and Heritage, John. 1990. Conversation analysis. Annual Review of Anthropology 19: 283307.
Goodwin, Marjorie H. and Goodwin, Charles. 1986. Gesture and coparticipation in the activity of searching for a word. Semiotica 62 (1/2): 5175.
Janko, Tatjana E. 2008. Intonacionnye strategii russkoj rei v sopostavitelnom aspekte. Moscow:
Jazyki slavjanskix kultur.
Jay, Timothy B. 2003. The Psychology of Language. Upper Saddle River: Prentice Hall.
Jefferson, Gail. 1978. Sequential aspects of storytelling in conversation. In Studies in the Organization of Conversational Interaction, Jim Schenkein (ed.), 219248. New York: Academic
Press.
Korotaev, Nikolaj A., Kibrik, Andrej A. and Podlesskaja, Vera I. 2009. Oslonenija kanonieskoj struktury: na styke mono- i polipredikativnosti. In Rasskazy o snovidenijax. Korpusnoe issledovanie ustnogo russkogo diskursa, Andrej A. Kibrik and Vera I. Podlesskaja (eds.),
219332. Moscow: Jazyki slavjanskoj kultury.
Labov, William. 1972. Language in the Inner City: Studies in the Black English Vernacular. Philadelphia: University of Pennsylvania Press.
Labov, William. 1997. Some further steps in narrative analysis. Journal of Narrative and Life
History 7 (14): 395415.
Labov, William and Waletzky, Joshua. 1967. Narrative analysis. In Essays on the Verbal and
Visual Arts, June Helm (ed.), 1244. Seattle: University of Washington Press.
Lapteva, Olga A. 1990. ivaja russkaja re s telekrana: Razgovornyj plast televizionnoj rei v
normativnom aspekte. Szeged: Jate Kiad.
Lerner, Gene H. 1992. Assisted storytelling: Deploying shared knowledge as a practical matter.
Qualitative Sociology 15 (3): 247271.
Lerner, Gene H. 1996. On the semi-permeable character of grammatical units in conversation: Conditional entry into the turn space of another speaker. In Interaction and
Grammar, Elinor Ochs, Emanuel A. Schegloff and Sandra A. Thompson (eds.), 238276.
Levelt, Willem J. M. 1989. Speaking: From Intention to Articulation. Cambridge, MA: MIT
Press.
Lounsbury, Floyd G. 1954. Pausal, juncture, and hesitation phenomena. In Psycholinguistics.
A Survey of Theory and Research Problems. Report of the 1953 Summer Seminar Sponsored
by the Committee and Linguistics and Psychology of the Social Science Research Council,
Charles E. Osgood and Thomas A. Sebeok (eds.), 98101. Baltimore, MD: Waverly Press.
Hesitation markers in transitions 101
Mustajoki, Arto. 2006. Teorija funkcionalnogo sintaksisa. Ot semantieskix struktur k jazykovym

sredstvam. Moscow: Jazyki slavjanskoj kultury.
Norrick, Neal R. 2000. Conversational Narrative: Storytelling in Everyday Talk. Amsterdam/
Nuolijrvi, Pirkko and Tiittula, Liisa. 2000. Televisiokeskustelun nyttmll. Televisioinstitutionaalisuus suomalaisessa ja saksalaisessa keskustelukulttuurissa. Helsinki: SKS.
Ochs, Elinor and Capps, Lisa. 2001. Living Narratives: Creating Lives in Everyday Storytelling.
Cambridge, MA/London: Harvard University Press.
Ochs, Elinor, Schegloff, Emanuel A. and Thompson, Sandra A. (eds.). 1996. Interaction and
Grammar. Cambridge: Cambridge University Press.
Podlesskaja, Vera I. and Kibrik, Andrej A. 2007. Samoispravlenija govorjaego i drugie tipy
reevyx sboev kak obekt annotirovanija v korpusax ustnoj rei. Nauno-texnieskaja
informacija 2: 223. September 14, 2011 <http://www.philol.msu.ru/~otipl/new/main/
people/kibrik-aa/files/Self-repair@NTI_2007.pdf>.
Podlesskaja, Vera I. and Kibrik, Andrej A. 2009. Reevye sboi i zatrudnenia. In Rasskazy o
snovidenijax. Korpusnoe issledovanie ustnogo russkogo diskursa, Andrej A. Kibrik and Vera
I. Podlesskaja (eds.), 177218. Moscow: Jazyki slavjanskoj kultury.
Popova, Tatjana I. 2002. Teleintervju v kommunikativno-pragmatieskom aspekte. Saint
Petersburg: SPbGU.
Routarinne, Sara. 2003. Tytt ness. Parenteesit ja nouseva svelkulku kertojan vuorovaikutuskeinoina. [Girls Talking. Parentheses and Rising Intonation as Narrators Interactional
Devices.] Helsinki: SKS.
Sacks, Harvey. 1974. An analysis of the course of a jokes telling in conversation. In Explorations in the Ethnography of Speaking, Richard Bauman and Joel Sherzer (eds.), 337353.
London: Cambridge University Press.
Sacks, Harvey. 1992a [19641968]. In Lectures on Conversation, Gail Jefferson (ed.). Oxford/
Cambridge: Blackwell.
Sacks, Harvey. 1992b [19681972]. In Lectures on Conversation, Gail Jefferson (ed.). Oxford/
Cambridge: Blackwell.
Sacks, Harvey, Schegloff, Emanuel A. and Jefferson, Gail. 1974. A simplest systematics for the
Schegloff, Emanuel A. 1979. The relevance of repair to syntax-for-conversation. In Discourse
and Syntax, Talmy Givn (ed.), 261286. New York, NY: Academic Press.
Schegloff, Emanuel A. 1996. Turn Organization: one intersection of grammar and interaction.
In Interaction and grammar, Elinor Ochs, Emanuel A. Schegloff and Sandra A. Thompson
(eds.), 52133. Cambridge: Cambridge University Press.
Schegloff, Emanuel A. 1997. Narrative analysis thirty years later. Journal of Narrative and
Life History 7 (14): 97106.
Schegloff, Emanuel A. 2009. A practice for (re-)exiting a sequence: and/but/so + uh(m) + silence. In Language in Life, and a Life in Language. Jacob Mey A Festschrift, Bruce Fraser
and Ken Turner (eds.), 365374. Bingley: Emerald Group Publishing Limited.
Schegloff, Emanuel A. 2010. Some other uh(m)s. Discourse Processes 47: 130174.
Schegloff, Emanuel A., Jefferson, Gail and Sacks, Harvey. 1977. The preference for self correction in the organization of repair in conversation. Language 53 (2): 361382.
Selting, Margret. 2000. The construction of units in conversational talk. Language in Society
29: 477517.
102 Hanna Laitinen
Shriberg, Elizabeth E. 1994. Preliminaries to a Theory of Speech Disfluencies. Ph.D.-dissertation,

University of California at Berkeley. September 14, 2011 <http://www.speech.sri.com/
papers/shriberg-thesis.pdf>.
melev, Aleksej D. 2004. Zapolniteli pauz kak kommunikativnye markery. In anr intervju:
osobennosti russkoj ustnoj rei v Finljandii i Sankt-Peterburge, Marja Leinonen (ed.), 205
222. Tampere: Tampere University Press.
Swerts, Marc. 1998. Filled pauses as markers of discourse structure. Journal of Pragmatics 30:
485496.
Part II
Statistical analysis of Russian

talk-in-interaction
Russian everyday utterances

The top lists and some statistics
Tatiana Sherstinova
This paper presents lists of the most frequently used Russian utterances in everyday interaction and gives a statistical description of Russian spontaneous
speech; including, frequency distribution of utterance length measured in words
and syllables and temporal statistics of utterances. Audio data for this research
were taken from the ORD Speech Corpus of Russian Everyday Communication.
Special investigation was conducted to analyze average utterance duration and
syllable rate of speech relative to utterance length in syllables. Additionally, we
propose the existence of two temporal registers in Russian everyday speech.
Keywords: Russian spoken language, everyday utterances, utterance length,
ORD speech corpus, everyday speech communication
1. Introduction
This paper continues the statistical description of Russian spoken interaction that
was started in (Sherstinova 2010).
Audio data for this research were taken from the ORD speech corpus of Russian everyday communication. All recordings were made under natural conditions, with digital voice recorders hung around the neck of subjects, and captured
natural spoken language communication during a single 24 hour period. A detailed description of the ORD corpus is presented in Asinovsky et al. (2009).
To date, transcripts and multi-level linguistic annotation have been made for
40 hours of the ORD recordings and represents speech from 20 male and 20 female subjects and their interlocutors. For this research we have excluded from
investigation all utterances containing overlapping speech, unintelligible speech,
unfinished (broken) remarks, utterances with inner pauses and/or hesitations,
and thus, the reduced corpus consists of roughly 15,000 utterances produced by
106 Tatiana Sherstinova
approximately 200 different people. These utterances formed the basis for our
investigation.
The principles of division of speech into utterances were described in
Sherstinova, Stepanova and Ryko (2009).
All statistics presented here have been obtained for the whole subset of 15,000
utterances regardless of age, gender and other social or psychological characteristics of speakers. Another analysis of speech rate was conducted on this same
data (Stepanova 2011). However, it took into account different social factors of
speakers.
2. Frequency distribution of utterance length in words

As an initial step, the frequency distribution of utterance length in words was obtained (see Figure 1). Based on our data, the average utterance length is 4.35words,
and standard deviation is 4.02words. The longest utterance in the ORD corpus
has 71 words. It turned out that the majority of all utterances consist of a single
word or word-like particle (25.26%). Two-word utterances make 15.58% of the
whole data, three-word utterances has the third rank making 12.45%. Four-word
utterances make 10,98%, five-word utterances 8.74%, etc.
4500
4000
Absolute frequency
3500
3000
2500
2000
1500
1000
500
0
10
13 16 19 22 25 28
Utterance length in words
Figure 1. Frequency distribution of utterance length in words
31
35
44
58
Russian everyday utterances 107
In general, the shorter the utterance length in words, the higher its frequency
in real communication. Subsequently, utterances consisting of up to three words
make more than a half of all speech interaction.
3. The top lists of Russian everyday utterances

Frequency lists were further obtained for utterances of the same length in words.
In this article we present high-frequency zones of these lists for most common
utterances consisting up to 4 words (see Tables 14). Making these lists we try
to keep reference to functional type of utterance marked by the following correspondent symbols: declarative utterance (//), interrogative utterance (?) and
exclamatory, greeting, vocative or imperative utterances (!). The percentage was
calculated within each group of utterances. Whenever possible, we tried to save
the transcripts division into rhythmic groups (or syntagmas) marked by a forward slash (/) (e.g., vs / ja ponjal thats enough, I got it). However, if multiple
division into rhythmic groups was possible for the same word string (e.g., da da
and da / da yeah yeah and yeah / yeah), we left in the list just one variant
summing all relevant frequencies.
It should be noted as well that when making these lists we excluded from
them all phrases that only occur in the speech of a single speaker. This left only
those which are more common. For example, we excluded from the lists an utterance of a triple occurrence Xoma, razdevajsja! (Khoma, take off your clothes!)
when a mother was trying to get her three-year old son to change his clothes after
walking, or a favorite quote of a song of another subject ta enina v okoke
(This lady behind the window), which he repeated several times, as well as other
non-common utterances.
As shown in the previous section, more than a quarter of all Russian speech
interaction is performed using one-word utterances. The top list of one-word utterances obtained on our material is given in Table 1. The total number of oneword utterances in our ORD transcripts is 3,648 (the percentages shown in the
right column were calculated out of this number).
We may see that many of them are formed by discourse particles, which have
mainly the pragmatic function of providing successful interaction: e.g., ugu,
aga, tak, da (all four meaning yep/yeah), vot (well/ here is), nu (well),
m (a hesitation marker), ponjatno and jasno (both meaning its clear),
koneno (sure), normalno (its all right), and many others.
Among interrogative utterances the highest rank has a?, m?, to? (all used
in the meaning of what are you speaking about? or asking for repetition), da?
Table 1. Most frequent one-word Russian utterances

Rank Utterance
Count
Percent
Rank Utterance
Count
Percent
1
2
3
4
5
6
7
8
9
10.5
10.5
12
13
15
15
15
17
18
19
20.5
20.5
22
23
24
26.5
26.5
26.5
26.5
30.5
30.5
595
436
142
104
100
93
74
73
65
55
55
50
47
42
42
42
29
26
23
22
22
21
18
17
12
12
12
12
11
11
16.31%
11.95%
3.89%
2.85%
2.74%
2.55%
2.03%
2.00%
1.78%
1.51%
1.51%
1.37%
1.29%
1.15%
1.15%
1.15%
0.79%
0.71%
0.63%
0.60%
0.60%
0.58%
0.49%
0.47%
0.33%
0.33%
0.33%
0.33%
0.30%
0.30%
30.5
30.5
34
34
34
36
38
38
38
40.5
40.5
44
44
44
44
44
51
51
51
51
51
51
51
51
51
62.5
62.5
62.5
62.5
62.5
11
11
10
10
10
9
8
8
8
7
7
6
6
6
6
6
5
5
5
5
5
5
5
5
5
4
4
4
4
4
0.30%
0.30%
0.27%
0.27%
0.27%
0.25%
0.22%
0.22%
0.22%
0.19%
0.19%
0.16%
0.16%
0.16%
0.16%
0.16%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.11%
0.11%
0.11%
0.11%
0.11%
ugu //
da //
tak //
aga //
vot //
a?
net //
m?
da ?
oj !
m //
a //
to ?
nu //
xoroo //
vs //
nu ?
davaj !
zdravstvujte !
ponjatno //
poemu ?
spasibo //
koneno //
privet !
a!
gde ?
kto ?
tak ?
kuda ?
ego ?
sejas //
aj !
molodec !
pojdm !
poalujsta !
net ?
ja ?
al !
bljad !
kak ?
gospodi !
na !
netu //
normalno //
o!
jasno //
vo !
vs ?
est //
zaem ?
kakie ?
net !
opa //
xm //
xoroo ?
allo !
mama !
kakoj ?
zameatelno //
prikolno //
(yes?), which is often used to demonstrate astonishment, nu? (well?), stimulating interlocutors to continue speaking, explaining or performing some other
action. This list is continued by standard question utterances: poemu? (why?),
gde? (where?), kto? (who?), tak? (so? or OK?), kuda? (whereto?),
and many others.
The most common one-word interjections are the following: oj! and aj!
(similar to English oh! and ah!), a! (ah!), o! (oh!), molodec! (well done!),
gospodi! (Lord!), and others.
Table 2. Most frequent two-word Russian utterances

Rank Utterance
Count
Percent
Rank Utterance
Count
Percent
1
2
3
4.5
4.5
6
7
8.5
8.5
10
11.5
11.5
16
16
16
16
16
16
16
54
33
29
15
15
12
10
8
8
7
6
6
5
5
5
5
5
5
5
2.39%
1.46%
1.28%
0.66%
0.66%
0.53%
0.44%
0.35%
0.35%
0.31%
0.27%
0.27%
0.22%
0.22%
0.22%
0.22%
0.22%
0.22%
0.22%
24
24
24
24
24
24
24
24
37
37
37
37
37
37
37
37
37
37
37
4
4
4
4
4
4
4
4
3
3
3
3
3
3
3
3
3
3
3
0.18%
0.18%
0.18%
0.18%
0.18%
0.18%
0.18%
0.18%
0.13%
0.13%
0.13%
0.13%
0.13%
0.13%
0.13%
0.13%
0.13%
0.13%
0.13%
nu da //
nu vot //
ne znaju //
nu ladno //
m da //
nu davaj //
da da //
i vs //
nu ponjatno //
o gospodi !
spasibo boloe //
ugu / ugu //
a ego ?
vot tak //
da ladno //
da net //
ne nado //
do svidanija !
nu to ?
a zaem ?
a poemu ?
v smysle ?
ne xou //
nu-nu //
tak / sejas //
tak to ...
ja ponjal //
a kuda ?
a ty ?
a to ?
vot to ?
vs / spasibo //
vs ravno //
da / koneno //
da u //
ili net ?
ne ponjal //
niego sebe //
Regarding greetings (salutations), there were no surprises here. At the top are
zdravstvujte! (hello!), privet! (hello! or hi!), and the telephone greetings
al! (hello!), and allo! (hello!).
Additionally, there were a considerable amount of utterances expressing evaluation, like xoroo (OK!), normalno (its all right!), zameatelno (great!),
prikolno (cool!), etc.
Examining at frequent two-word utterances (see Table 2), we noticed that
many of them have a similar structure, where each begins with a particle. The
most active particle here is nu (well), which initiates the most frequent utterances: nu da (well, yes), nu vot (an utterance making a boundary in speech
or interaction), nu ladno (OK!), nu davaj (well, (lets) do it), nu ponjatno
(well, I see), nu-nu (well, well), etc.
Particle da (yeah) is used in a similar function, forming utterances such
as da da (yeah yeah), da ladno (OK!), da u (well, yeah), da / koneno
(yes, sure), da net (although a literal translation of this utterance is yes (well)
no, meaning no!), etc.
About 10% of all two-word questions begin with conjunctions: a (and, but):
a ego? (and what?), a gde? (and where?), a zaem? (and what for?), a
poemu? (and why?), a ty? (and you?), a to? (and what?), a kuda?
Table 3. Most frequent three-word Russian utterances

Rank Utterance
Count Percent
Rank Utterance
Count Percent
1
2.5
2.5
4.5
4.5
7.5
7.5
7.5
7.5
24.5
24.5
14
6
6
4
4
3
3
3
3
2
2
24.5
24.5
24.5
24.5
24.5
24.5
24.5
24.5
24.5
24.5
24.5
2
2
2
2
2
2
2
2
2
2
2
da da da //
nu i to ?
nu ne znaju //
vot i vs //
vot tak vot //
vs / ja ponjal //
nu i vs //
ja ne znaju //
ja ne mogu //
a / nu da //
a / nu ponjatno //
0.78%
0.33%
0.33%
0.22%
0.22%
0.17%
0.17%
0.17%
0.17%
0.11%
0.11%
a da da //
a nu da //
da ty to !
da ty to ?
i to teper ?
kak vy moete ?
ne figa sebe !
nu / spasibo boloe //
nu vs / poka //
nu davaj / poka //
nu i xoroo //
0.11%
0.11%
0.11%
0.11%
0.11%
0.11%
0.11%
0.11%
0.11%
0.11%
0.11%
(and where to?), etc. The most typical negative utterances are the following: ne
znaju (I do not know), ne nado (we dont need it or dont do it!), ne xou
(I dont want it), ne ponjal (I dont understand).
The only complete utterance (defined here as having both a subject and a
predicate) presented in this list of most frequent two word utterances is ja ponjal (I understood). The total number of two-word utterances in our corpus is
2262.
Three-word utterances are less numerous and their frequencies are rather low
(see Table 3). Thus, the most frequent utterance, consisting of triple repetition of
the particle da (yes/yeah) occurs only 14 times. In this corpus of 1804 threeword utterances, it represents less than one percent. We see here mainly the same
words and particles that found to be the most frequent for one and two-syllable utterances as well: da (yes / yeah), vot (well), nu (well), vs (well / thats
enough), tak (yes), to (what), davaj (OK! / lets ), xoroo (OK!) and
others. Many of these utterances are very close in meaning with shorter forms.
Moreover, Table 3 shows that some of three-word utterances are a combination of the most frequent one and two-word elementary utterances presented
above. For example: nu ne znaju (well, I dont know), vs / ja ponjal (thats
enough / I got it), a / nu da (ah, Id rather agree), a / nu ponjatno (well, I see),
nu / spasibo boloe (well, thanks a lot), nu vs / poka (well, lets stop, bye!).
However, in contrast to utterances consisting of one or two words, three-word
utterances use the personal pronouns ja (I), ty (you), vy (you) rather
regularly.
Table 4 presents four four-word utterances that occur in the speech of different speakers: kak u tebja dela? (how are you?), a to to takoe? (well, what is
it?), tak / tak tak tak (so so so so), ja voobe ne ponimaju (I absolutely do
Russian everyday utterances
Table 4. Most frequent four-word Russian utterances

Rank Utterance
Count
Percent
1.5
1.5
3.5
3.5
3
3
2
2
0.19%
0.19%
0.13%
0.13%
kak u tebja dela ?

a to to takoe ?
tak / tak tak tak //
ja voobe ne ponimaju //
not understand). All the other four-word utterances occur just once in our data
or are peculiar only to individual speakers. The total number of four-word utterances investigated is 1565.
Interestingly, in our subset of the ORD corpus there were no utterances of five
or more words that occurred at least twice in the speech of different speakers.
4. Frequency distribution of utterance length in syllables

Frequency distribution of utterance length in syllables is shown on Figure 2. Here,
the utterances are classified according to the ideal (or full-style) number of
syllables.
Utterance length in our data ranges from 0 syllable (as in m// or m?) to
132 syllables. The average utterance length is 8.01 syllables and standard deviation
is 7.71 syllables.
The most frequent Russian utterances consist of one or two syllables and represent 11.0% and 11.7% of all utterances respectively. Three-syllable utterances
are ranked third (8.7%), four-syllable utterances are ranked fourth (7.7%), fivesyllable represent 6.88% and six-syllable utterances represent 6.50%. Utterances
longer than 20 or more syllables occur less than 1% of the time. Thus, more than
half of all spoken communication consists of short utterances with a length up to
6 syllables.
. According to Russian phonetic tradition, full pronunciation style is a well-articulated

pronunciation without reduction, simplification or omission of sounds and syllables. This style
is strongly correlated with a slow rate of speech, and is typical for public (mainly, declamatory)
speaking. It may be also called the ideal style, as opposed to that of real spontaneous speech
with multiple reductions. The ideal number of syllables in textual transcripts is automatically
calculated basing on rules of the full pronunciation style and may slightly differ from the real
number of syllables in the original speech.
111
2000
1800
1600
Absolute frequency
1400
1200
1000
800
600
400
200
0
10
15
20
25
30
35
40
45
50
55
60
79
132
Utterance length in syllables
Figure 2. Frequency distribution of utterance length in syllables
5. Average duration of utterances depending on their length in syllables

Special investigation was made to analyze average duration of utterances depending on their length in syllables (Sherstinova 2011). The main statistics of this distribution for utterances under 30 syllables are given in Table 5. Here we did not
analyze longer utterances, as their number in the ORD corpus is not sufficiently
significant.
One may notice that one- and two-syllable utterances, being the base of spoken communication, have the minimal variation when compared to other types
of utterances. Thus, the average utterance duration was 453 ms for one-syllable
utterances (n = 1608, SD = 208), while it was 563 ms for two-syllable utterances
(n= 1710, SD = 242). That means that their durations are the most predictable.
What concerns the dependency of average utterance duration on their length
in syllables, for our empiric data it is well described by the following linear function: y = 133,28x + 367,5, where x is a number of syllables, and y is an average
duration of such utterances in milliseconds. Figure 3 presents both empirical data
(marked by dots) and the theoretical line. Our proposed model fits the data well.
453
282
248
226
202
192
187
179
173
167
165
166
158
160
151
4500
Average syllable
duration, ms
Average syllable
duration, ms
1001
208
242
303
329
306
349
418
434
414
434
450
548
529
567
537
Standard
deviation
Standard
deviation
641
453
563
744
903
1008
1151
1312
1429
1554
1674
1816
1986
2055
2236
2268
Average utterance
duration, ms
Average utterance
duration, ms
188
1608
1710
1263
1118
1001
943
858
752
711
636
485
448
395
319
259
Count
Count
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Utterance length
in syllables
Utterance length
in syllables
Table 5. Descriptive statistics of utterance length
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
221
238
197
152
114
112
70
85
68
77
67
47
45
33
34
2484
2647
2715
2896
3023
3121
3353
3342
3572
3782
3974
3981
4061
4280
4343
562
537
631
539
658
661
735
602
725
765
743
702
832
594
1085
155
156
151
152
151
149
152
145
149
151
153
147
145
148
145
ms
4000
Average utterance duration
3500
3000
2500
2000
1500
1000
500
0
10
12
14
16
18
20
22
24
26
28
30
Figure 3. Average utterances duration and utterance length in syllables. Empirical data
and the theoretical model
6. Dependence of average syllable duration on utterance length

in syllables
This section is concerned with average syllable duration for utterances of different
length in syllables. The data we analyze here were presented above in Table 5 in
two columns (Utterance length in syllables and Average syllable duration, ms). The
correspondent dependency curve is shown in Figure 4.
Differences between neighboring averages from Table 5 were tested for significance by means of T-test (the normality of distribution was confirmed by the
Shapiro-Wilk W-test). For example, we tested the differences between: (1) onesyllable and two-syllable utterances, (2) two-syllable and three-syllable utterances, (3) three-syllable and four-syllable utterances, and so on.
The result is quite convincing. It turned out that for utterances under 15 syllables, the average syllable duration significantly differs between any two neighboring utterance types, at a significance level of 95%. Thus, we may assume strong
dependency of average syllable duration and utterance length in syllables. Based
on our data, the average syllable duration in utterances (whose length is more
than 15 syllables) does not differ significantly and is equal to approximately
to150ms.
500
450
Average syllable duration
400
350
300
Regular (dialogue) register
Speedy (monologue) register
250
200
150
100
50
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Figure 4. Average syllable duration relative to utterance length in syllables.

A hypothetical division into two registers
. The dependence of average syllable duration on utterance length in syllables is integrative

part of the so called Menzeraths-Altmann law, according to which the increase of a linguistic
construct results in a decrease of its constituents.
An important conclusion drawn from our research might be summarized as

follows: We can assume the existence of two registers of speech:
1. The regular (or dialogue) register is used for producing utterances whose
length does not exceed 15 syllables. Its distinctive feature is a strong dependence between average syllable duration (syllable rate) of utterances and
their length in syllable. Therefore, in this case the average syllable duration
is a function from utterance length in syllables, ranging approximately from
450ms to 150 ms.
2. The speedy (or monologue) register is used for producing long utterances
(exceeding 15 syllables). In contrast to a dialogue register, the average syllable
duration (or an average utterance rate) of a speedy register does not depend
on utterance length in syllables and is equal to approximately 150 ms.
We may suppose that 150 ms functions as a sort of constant, around which the
average duration of long Russian spontaneous utterances tends to occur. Finally,
we note that the average syllable duration for the whole analyzed part of the ORD
corpus is 160 ms, which is only 10 ms longer.
7. Conclusion
The main results of this research are the following:
1. The majority of Russian spoken interaction consists of one or a few word utterances, which contain one or few syllables.
2. The top lists of the most common Russian utterances are obtained. Discourse
particles occur in, or at times entirely constitute the most common Russian
utterances.
3. The dependency of both the average duration of utterances and their average
syllable duration on utterance length in syllables was studied.
4. The hypothesis on existence of two temporal registers of speech was
proposed.
Further investigations in this area may include studies on functional, rhythmical and syntactic typology of the most frequent Russian elementary utterances.
Quantitative analysis of utterance duration should be made on the basis of the
real number of pronounced syllables in each concrete case (e.g., the common
Russian greeting Zdravstvujte! (official Hello!) having in full-style pronunciation 3 syllables in real spontaneous speech is often reduced to 2-syllable utterance
like Zdraste! or a completely reduced one-syllable form like Drast!). Such
investigation allows us to prove or reject the proposed hypothesis concerning two

different temporal speech registers. Finally, it seems that interesting results may
be obtained by conducting similar research for different groups of speakers (men
and women, young men and adults, etc.).
Acknowledgments
The study was supported by the Ministry of Education and Science of the Russian Federation, project 8554 Speech Corpus of the Russian Language: Complex
Analysis of Oral Speech.
The author would like to thank Michael Furman for carefully proof-reading the paper and correcting her English, all remaining errors are the authors
responsibility.
References
Asinovsky, Alexander S. et al. 2009. The ORD Speech Corpus of Russian Everyday Communication One Speakers Day: Creation Principles and Annotation. In xt, Speech
and Dialogue. TSD-2009 [LNCS/LNAI 5729]. Matouek Vaclav and Mautner Pavel (eds.),
250257. Berlin/Heidelberg: Springer.
Sherstinova, Tatiana Y., Stepanova, Svetlana B., and Ryko, Anastassia. 2009. Sistema annotirovanija v zvukovom korpuse russkogo jazyka. In Materialy XXXVIII medunarodnoj
filologieskoj konferencii, Section Formalnye metody analiza russkoj rei, Pavel A. Skrelin
(ed.), 6675. St. Petersburg: St. Petersburg State University.
Sherstinova, Tatiana. 2010. Quantitative Data Processing in the ORD Speech Corpus of Russian Everyday Communication. In Text and Language. Structures Functions Interrelations Quantitative Perspectives, Peter Grzybek, Emmerich Kelih and Jn Mautek (eds.),
Advisory Editor: Eric S. Wheeler, 195206. Wien: Praesens.
Sherstinova, Tatiana. 2011. Repliki povsednevnoj russkoj rei v slogovom izmerenii. In
Materialy XXXX medunarodnoj filologieskoj konferencii. Section Polevaja lingvistika.
Integralnoe modelirovanie zvukovoj formy estestvennyx jazykov, Alexander S. Asinovsky
and Natalija V. Bogdanova (eds.), 229239. St. Petersburg: St. Petersburg State University.
Stepanova, Svetlana B. 2011. Russian spontaneous speech rate (based on the speech corpus of
Russian everyday interaction). In Proceedings of the 17th International Congress of Phonetic Science, Way-Sum Li and Eric Zee (eds.), 19021905. Hong Kong: City University of
Hong Kong.
Speech rate as reflection of speakers

social characteristics
Svetlana B. Stepanova
This paper presents an analysis of speech rate in spontaneous conversations

between Russian interlocutors conducted on the basis of recordings of 40 speakers and their interlocutors collected in the Speech Corpus of Russian Everyday
Communication One Day of Speech (ORD corpus). The results allow us to
compare the speech rate between speakers of different mother tongues using data drawn from other researchers work. For example, Russians speak on
average faster than Norwegians, but considerably slower than Spaniards and
Brazilians.
The impact of different factors on the rate of speech is illustrated by the following findings:
1. There is a statistically valid difference between mens (m) and womens (f)
speech rate: men speak substantially (from the statistical perspective) faster
than women.
2. With age we start speaking more slowly.
3. Informants whose level of verbal competence was assessed at a high level by
experts speak more slowly, while an articulation rate higher than the average is typical of speakers with a lower level of verbal competence.
4. Furthermore, we observed that the speech rate is dependent on a statistically significant way on the number of syllables of the utterance: the longer
the phrase, the faster the rate.
Keywords: speech corpus, speech rate, Russian
1. Introduction
The Russian speech rate has been studied by many Russian phoneticians. However, those studies are based either on recorded reading material (Krivnova 2007),
on limited samples of colloquial speech (200 sec for 7 speakers in Rozanova 1983),
or on taking the number of words per minute as the rate unit (Svetozarova 1988).
118 Svetlana B. Stepanova
Unfortunately, it is difficult to compare the results of those analyses with the results obtained on spontaneous speech material in other languages.
In the present study, we use material from the Speech Corpus of Russian Everyday Communication (ORD corpus) created at Saint Petersburg State Universitys School of Philology. The main aim of the ORD speech corpus is to fix Russian
spontaneous speech in natural communicative situations. A characteristic feature
of the corpus is that speech is recorded not in a laboratory (or in the presence of
the researcher or interviewer), but with the help of a technique designed by the
developers of the British National Corpus (http://www.natcorp.ox.ac.uk/corpus/
creating.xml, latest access 19/11/2012): on an average day, volunteer informants
carry a speech recorder on their neck all day long and record their own speech
and the speech of their conversation partners.
Currently, 350 hours of Russian spontaneous speech have been recorded in
this manner. At present, the ORD corpus contains recordings made by a demographically balanced group of 46 subjects (23 men and 23 women) representing
various social and age strata: students, military students, engineers, managers,
scientists, doctors, IT specialist, a merchant, a builder, a psychologist, a photographer, baby-sitter, a drawing teacher, etc. The subjects age ranges from 16 to 70
years. Although the recordings were made while maintaining absolute anonymity,
all subjects filled in sociological questionnaires, were given a psychological evaluation and kept diaries of their day of speech, noting the main conditions of communication. In addition to the subjects speech, 600 interlocutors with whom the
subjects conversed were also recorded. The interlocutors as well were of different
ages (from 3 to 68 years), professions, and occupations; they were in both formal
and informal relations with the subjects. About 40 hours of recordings have been
transcribed (on average, one hour each from the 40 informants) with multi-level
annotation using the professional tool ELAN (cf. Asinovsky et al. 2009:250257
for further information about the ORD speech corpus).
2. Research material
As is well attested, the perception of the rate of speech is affected by both the
tempo of articulating sounds and by the presence of pauses of certain duration.
At the first stage of the project, we decided to consider the articulation tempo,
regardless of pauses.
In the ORD Corpus, we selected utterances without pauses, sighs, laughter,
and overlapping words from more than one speaker at a time.
The total number of selected utterances is 13,459, their total length is
19,298seconds for a total of 5.36 hours. The rate was calculated as the number of
Speech rate as reflection of speakers social characteristics 119
syllables per second (cf. Pellegrino, Coupe and Marsico (2011:545) discussing the
advantages of this method). The syllable length was equated with the number of
vowels. The material was processed using the STATISTICA program.
3. Results
General characteristics of the speech rate
3.1
Taking all the material into consideration, the average rate equaled 5.31 syllables
per second (syll/s), see Table 1.
For different speakers the average rate changes from minimum 3.6 syll/s
(female-speaker, 63 years old) to maximum 6.7 syll/s (female-speaker, 20 years
old) see Figure 1.
Our results concerning the average speech rate do not contradict the observations made by other experimental phoneticians: Krivnova (2007:58), for example,
claims that the average duration of a syllable in Russian is 150200 ms, which corresponds to 4.766.67 syll/s.
Table 1. Average speech rate (m male, f female, k children)
Descriptive Statistics (Tempo.sta)
f+m+k
Valid N
Mean (syll/s) Minimum
Maximum
Std. Dev.
13459
5.31
13.19
1.93
0.63
Average speech rate

8.00
7.00
Speech rate
6.00
5.00
4.00
3.00
2.00
1.00
S24
S44
S9
S40
S35
S20
S7
S22
S19
S8
S43
S17
S36
S37
S38
S1
S11
S12
S4
S14
S26
S13
S5
S2
S3
S27
S42
S23
S6
S25
S10
S39
S41
S16
S28
S15
S21
S18
S30
0.00
Speakers
Figure 1. Average speech rate depending on the speaker
In their review of research on speech rate Verhoeven, De Pauw and Kloots

(2004:298) list the average speech rate in several languages:

in British English:
3.165.33 syll/s (Tauroza and Allison 1990)
in French:
4.31 syll/s (Grosjean and Deschamps 1973)
5.73 syll/s (Malcot, Johnston and Kizziar 1972)
in Northern Standard Dutch:
5.2 syll/s (Blaauw 1995)
in Norwegian ranges:
3.54.5 syll/s (Almberg 2000)
In Brazilian Portuguese:
6.57 syll/s
In Spanish:
7.81 syll/s (Rebollo Couto 1997)
Compared with these data, Russian speech rate is situated in the middle of this
list: Russians speak faster than Norwegians, but slower than Brazilians and
Spaniards.
This is probably connected to phonetic differences. Here, we agree with
Verhoeven, De Pauw and Kloots who argue that
When speech rate is expressed as number of words per minute or syllables per
seconds, the measures will reflect these structural differences, and language with
long words or syllables will necessarily have a lower speech rate measures then a
language with short words or syllables.

(Verhoeven, De Pauw and Kloots 2004:298)
Further, the country the speaker lives in has a significant effect of on the speech
rate: Verhoeven, De Pauw and Kloots (2004:306) mention that Dutch-people
speaking in Belgium show a slower speech rate than those living in the Netherlands. Our data show similar results: speakers from Saint Petersburg speak slower
than speakers from the provinces (see Figure 2).
3.2
The impact of gender on speech rate
The material also allowed us to check for statistically valid differences between
mens (m) and womens (f) speech rate. Our analysis shows that men speak (from a
statistical perspective) substantially faster than women (5.46 syll/s vs. 5.30syll/s).
Childrens speech rate (k) (neglecting gender) is the slowest 3.86syll/s (see Table2). The statistical significance of these differences has been checked using the
Box & whisker plot

Non-residents vs. SPb-residents
6,4
6,3
6,2
Speech rate
6,1
Mean
Mean SE
6,0
Mean 1,96* SE
5,9
5,8
5,7
5,6
5,5
Non-residents SPb-residents
Figure 2. Significance of differences in average speech rate: Saint Petersburg (SPb)

residents and non-residents
STATISTICA program. The t-test showed a statistically significant difference between the mean values in these groups (p < 0.001) (see Table 2).
A similar impact of gender on the speech rate has been observed in other
languages as well (cf. for Chinese and American English by Yuan, Libermann
and Cieri 2006 and Jacewicz, Fox and Wei 2010; for Dutch by Verhoeven, De
Pauw and Kloots 2004 or Quene 2008; and for English, French, German, Italian,
Japanese, Mandarin, Spanish, and Vietnamese by Pellegrino, Coupe and Marsico
2011:546). Characteristically, all authors observe that the difference between
mens and womens average speech rate is not substantial but statistically considerable. Quene explains mens faster speech rate in terms of social dominance: male
speakers may also express their social dominance by speaking somewhat faster
than female speakers (Quene 2008:1112).
Table 2. T-test for male/female/childrens speech rate
m vs. k
m vs. f
k vs. f
Mean
Group 1
Mean
Group 2
t-value
Valid N
Group 1
Valid N
Group 2
p
Variances
5.46
5.46
3.86
3.86
5.30
5.30
16.63
4.67
16.49
5067
5067
471
471
7921
7921
0.00
0.00
0.003
The average speech rate depending on the utterance length

Speech rate (syll/s)
8,00
6,00
4,00
2,00
0,00
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
The utterance length (syll)
Figure 3. Average speech rate depending on the utterance length
However, all these researchers recognize that the speech rate is to a great extent determined by the length of the generated utterance. Possibly, all other factors
affecting the tempo can be derived from the length of an utterance, as Liberman
argues: Maybe region, sex and age dont really influence speaking rate after all,
except indirectly via their influence on phrase length (Liberman 2006:1). Figure3, indeed, shows that the longer the phrase, the faster the rate: all material
considered, the average rate in single-syllable utterances is 2.59 syll/s, while in
ten-syllable ones it is 6.25 syll/s.
However, in even lengthier utterances, the speech rate does not change substantially and remains within the range of average values of 6.27.2 syll/s. This
may be explained by purely physiological reasons: a syllable has to be articulated
by the speech organs, and the movement rate of the latter cannot grow infinitely.
Further, speech perception also sets certain limitations subconsciously recognized by the speaker:
When the rate exceeds maximum, syllable recognition may be impeded or even
become impossible. These limitations may be explained by the facts that the process of interpretation requires time and that duration is one of the formal indicators of a stream element necessary for defining the latter (a section shorter than a
certain value cannot be an element.
(istovi 1976:14; transl. S. B.)
In order to take into account the phenomenon of anticipatory shortening, we

checked the average syllable length of the analyzed fragments. Our material
showed that the average number of syllables in mens utterances is slightly lower
than that in womens utterances (8.1 vs. 8.3) (see Table 3), which means that Russian men on average pronounce shorter phrases, but at a higher rate.
Here, our results differ from the observations made by Fenk-Oczlon and
Fenk (2010). According to them (Fenk-Oczlon and Fenk 2010:1538), for Russian
the number of syllables per clause is 5.68. This is substantially lower than our results. This divergence is probably due to the different character of the speech data
Table 3. Average number of syllables and words in a speech fragment

Descriptive Statistics (words/syll in phrase)
words-m
syll-m
words-f
syll-f
Valid N
Mean
Minimum
Maximum
Std.Dev.
5065
5065
7921
7921
4.4
8.1
4.4
8.24
1
1
1
1
44
82
45
88
4.0
7.5
4.0
7.7
analyzed and the different counting procedure. Fenk-Oczlon and Fenk (2010)
analyzed simple declarative sentences, whereas our analysis was based on speech
fragments between physical pauses. These fragments could include complex sentences or a few simple phrases, spoken without pauses and irregularities.
Furthermore, a more detailed examination of the material comparing the average speech rate of men and women in utterances ranging from 1 to 20 syllables
has been conducted.
Statistical analysis showed that in single-syllable utterances the speech rate
does not depend on the speakers gender and is almost twice as slow as the average rate in the whole body of the material 2.6 syll/s (see Figure4). Two-syllable utterances were pronounced slightly faster by women than by men (4.09 vs.
3.97 syll/s). However, the difference is not statistically significant. In all other 18
cases (utterances of more than two syllables), mens speech rate was higher than
8,00
Average speech rate (syll/s)
7,00
6,00
5,00
Female
Male
4,00
3,00
2,00
1,00
0,00
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Number of syllables
Figure 4. Average speech rate depending on the utterance length (according to speakers
gender)
womens. In 5-, 7-, 10-, 11-, 14- and 20-syllable utterances the difference was statistically significant.
8,000000
7,000000
Speech rate
6,000000
5,000000
Speech rate
St.Dev.
4,000000
3,000000
2,000000
1,000000
70
60
58
53
Age
44
40
37
32
27
22
19
16
0,000000
Figure 5. Average speech rate and standard deviation depending on the informants age
Box & whisker plot

<40 vs. >40
6,4
6,2
6,0
5,8
5,6
5,4
5,2
<40
>40
Figure 6. Significance of average speech rate differences (older vs. younger than 40)
3.3
The impact of age on speech rate
Unlike research based on other languages, our analysis does not point to a direct influence of speakers age on the speech rate. As mentioned in Section 3.1,
children (in our corpus age 312) display a much slower speech rate than adults.
However, from the age of 16 on, age does not seem to influence the speech rate
in a clearly patterned way (see Figure 5). There is a tendency towards a decrease
on speech rate with speakers age, but peaks of male speakers aged 53, 55 and
56 disturb it.
Nevertheless, should we divide our speakers into two groups (younger vs.
older than 40), a significant difference concerning speech rate can be observed
(see Figure 6): with age the informants start speaking more slowly, although never
reaching the slowness level of childrens speech rate (see Table 2).
3.4
The impact of other social factors
Statistical analyses did not reveal substantial differences concerning speech rate
between speakers with higher, incomplete higher and secondary education. He
terogeneous factors may account for this observation: the ORD corpus contains
only a small number of recordings from speakers without higher education. Further, there is a statistical spread of their average speech due to other factors.
However, the level of the informants verbal competence (LVC) has on average an impact on the speech rate. The level of verbal competence is determined by
a combination of the speakers social characteristics: a speakers level of education,
as well as professional or non-professional use of speech, play a crucial role in assessing his or her LVC, but a speakers personal involvement in social matters is
also to be considered. In determining a speakers LVC, we relied on an experts assessment and interpretation of audio extracts from the corpus. These results have
subsequently been checked with the background information from the speakers
questionnaire in order to find correlations between the experts interpretation of
speech samples and the real social characteristics of the speakers (Bogdanova et
al. 2008:5761).
It turned out that the informants whose LVC was assessed by experts as
high speak more slowly. An articulation rate higher than the average is typical
for speakers with a lower LVC (see Figure 7). Probably, an analysis that considers pauses and their length will yield different results.
It seems evident to point out that we speak differently with different interlocutors. Nevertheless, at this stage we can only conduct statistically reliable analyses
of the average speech rate in conversations with colleagues and friends. These
Categ. Box & whisker plot: speech rate

6,4
6,2
Speech rate
5,6
Mean
Mean SE
5,4
Mean 1,96* SE
5,2
5,0
4,8
Verbal competence level
Figure 7. Significance of differences in average speech rate values: Average (m) high
(h) low (l) LVC
Box & whisker plot
Friends vs. Colleagues
5,8
5,7
5,6
5,5
Mean
Mean SE
Mean 1,96* SE
5,4
5,3
5,2
5,1
5,0
Friends
Colleagues
Figure 8. Significance of differences in average speech rate values: With friends vs. with
colleagues (all material considered)
show that the analyzed speakers speak faster in a relaxed and friendly atmosphere
than in the office with their colleagues (see Figure 8).
4. Conclusion
The present study based on the Speech Corpus of Russian Everyday Communication (ORD corpus) has yielded new data on the rate of spontaneous Russian
speech. Although the ORD corpus is not as large as the corpora used by researchers of Dutch, English, Chinese or Japanese, the results obtained confirm many of
their conclusions: speech rate depends on such factors as gender, age, level of verbal competence, as well as on the length of the utterance. Our analyses show that
1. Men speak substantially faster than women.
2. Speech rate decreases with age.
3. Informants with a high level of verbal competence (determined by experts
assessment) speak more slowly, while an articulation rate higher than the average is typical of speakers with a lower level of verbal competence.
4. Finally, there is a statistically significant relationship between the speech rate
and the number of syllables in the spoken phrase: the longer the phrase, the
faster the rate.
However, other factors concerning the speaker (e.g. the communicative situation,
the kind of relation with the interlocutor, the emotional condition) can also significantly influence the speech rate. The ORD corpus contains several genres and
different styles of speech. There are casual conversations at home with relatives
(talk during meals, party talk, family chats, etc.) as well as professional and informal conversations with colleagues, communication in academia (lectures, practical lessons, and students informal conversations), communication with friends
in different situations and at different places, consultations with doctors, talk with
shop assistants, telephone conversations, etc. Recordings were made at home, in
the office, in public transport, on the streets, at universities, in a military college,
in coffee bars and restaurants, in shops, in amusement parks, and so on. Further
analyses of these data will allow us to check how the speech rate is affected by
communicative situations and, subsequently, by the communicative role of the
speaker.
Acknowledgments
The study was supported by the Ministry of Education and Science of the Russian Federation, project 8554 Speech Corpus of the Russian Language: Complex
Analysis of Oral Speech. My sincere thanks to all colleagues working on the creation of the ORD corpus and especially to Tatiana Sherstinova for her continuous
help in the automatic processing of annotated materials.
References
Almberg, Jrn. 2000. Kor fort snakkar vi eigentleg? Nordlyd 28: 6073.
Asinovsky, Alexander S. et al. 2009. The ORD Speech Corpus of Russian Everyday Communication One Speakers Day: Creation Principles and Annotation. In xt, Speech
and Dialogue. TSD-2009 [LNCS/LNAI 5729]. Matouek Vaclav and Mautner Pavel (eds.),
250257. Berlin/Heidelberg: Springer.
Bogdanova, Natalia V. et al. 2008. O korpuse ivoj rei: principy formirovania i vozmonosti
opisania. In Kompjuternaja lingvistika i intellektualnye texnologii: Trudy medunarodnoj
konferencii Dialog 2008 (7/14), Aleksandr E. Kibrik (ed.), 5761. Moscow: MGU.
Blaauw, Eleonora. 1995. On the Perceptual Classification of Spontaneous and Read Speech [OTS
dissertations series]. Utrecht: Led.
British National Corpus: <http://www.natcorp.ox.ac.uk/corpus/creating.xml>.
istovi, Ludmila A. (ed.). 1976. Fiziologija rei. Vosprijatie rei elovekom. Leningrad: Nauka.
Grosjean, Franois and Deschamps, Alain. 1973. Analyse des variables temporelles du franais
spontan II Comparaison du franais oral dans la description avec l anglais description et
avec le franais interview radiophonique. Phonetica 28 (34): 191226.
Fenk-Oczlon, Gertraud and Fenk, August. 2010. Measuring Basic Tempo across Languages
and some Implications for Speech Rhythm. In International Conference on Spoken Language Processing (Interspeech-2010), Makuhari, Chiba, Japan, 15371540.
Jacewicz, Eva, Fox, Robert A., and Wei, Li. 2010. Between-speaker and within-speaker variation in speech tempo of American English. Journal of the Acoustical Society of America
128 (2): 839850.
Krivnova, Olga F. 2007. Ritmizacija i intonacionnoe lenenie teksta v processe rei-mysli.
Abstract of PhD thesis. Moscow: MGU.
Liberman, Mark. 2006. Guys are a bit gabbier in Dutch, too. http://itre.cis.upenn.edu/~myl/
languagelog/archives/003682.html (latest access 13.11.2012).
Malcot, Andre, Johnston, R. and Kizziar, P. (1972). Syllabic rate and utterance length in
French. Phonetica 26: 235251.
Quene, Hugo. 2008. Multilevel modeling of between-speaker and within-speaker variation in spontaneous speech tempo. Journal of the Acoustical Society of America 123 (2):
11041113.
Pellegrino, Francois, Coupe, Christophe, and Marsico, Eggidio. 2011. A cross-language perspective on speech information rate. Language 87 (3): 539558.
Rebollo Couto, Letcia. 1997: Le rhythme en espagnol et en portugais: syllabique ou accentuel.
Travaux de lInstitut de Phontique de Strasbourg 27: 6390.
Rozanova, Natalia. N. 1983. Supersegmentnaja fonetika. In Russkaja razgovornaja re: Fone

tika. Morfologija. Leksika. est, Elena A. Zemskaja (ed.), 579. Moskow. Nauka.
Svetozarova, Natalia D. (ed.). 1988. Fonetika spontannoj rei. Leningrad: LGU.
Tauroza, Steve and Allison, Desmond. 1990. Speech rates in British English. Applied Linguistics 11 (1): 90-105.
Verhoeven, Jo, De Pauw, Guy, and Kloots, Hanne. 2004. Speech rate in a pluicentric language.
A comparison between Dutch in Belgium and in Netherlands. Language and Speech 47
(3): 297308.
Yuan, Jiahong, Liberman, Mark, and Cieri, Cristopher. 2006. Towards an Integrated Understanding of Speaking Rate in Conversation. In International Conference on Spoken Language Processing (Interspeech-2006).
Part III
Displaying and negotiating epistemic

and evidential status and evaluation
in interaction
How evaluation is transferred

in oral discourse in Russian
Nicole Richter
The paper deals with evaluation in spoken Russian. This phenomenon can be
conveyed by various phonetic and rhetoric means which I studied using an
experimental approach including an experiment on read speech and one on
quasi-spontaneous speech. Some phonetic-prosodic cues, e.g. different pitch
phenomena, are commonly said to be relevant for evaluation. As production
and perception are equally important, I tested whether native Russian speakers
were able to recognize the intended attitudinal meaning even in cases where
only the prosody was being presented to them. Noticeably from a rhetorical
perspective, questions are frequently used instead of answers for expressing
ones attitude. The structure of the answers given in these contexts has been
studied in detail.
Keywords: evaluation, Russian, prosody, stance, rhetoric, oral
1. Introduction
Interaction phenomena can be studied from different perspectives including research on rhetorical and phonetic strategies. Among the vast spectrum of conversational data, there is a wide range of emotive speech which consists of utterances
referring to pure emotions, but also contains speech data associated with valence
only. In the present article, I shall focus on the latter.
. I would like to thank the anonymous reviewer for very helpful comments on an earlier version of the paper.
134 Nicole Richter
2. Evaluation in phonetics and linguistics

Evaluation is a phenomenon associated with emotive communication that can
be described as a complex phenomenon transferred linguistically as well as nonlinguistically. As is true for all linguistic expressions, there is always a form and
a function attributed. So, evaluation includes pure linguistic aspects as well as
psychological characteristics. It may be characterized as a type of emotive speech
which is usually discussed when linguistic realizations of pure emotions, such as
anger or joy, are described, but also when expressive and emphatic speech is reported on. This is done in an oral as well as in a written discourse.
Evaluation has been properly described as including all types of verbal and
nonverbal choices that suggest an inferrable positive or negative evaluative stance
on the part of the speaker with respect to a topic, part of a topic, a partner, or
partners in discourse (Caffi and Janney 1994:354). This description shows that it
is the hearer who decides about the special reading in a given utterance. Downes
(2000) in his contribution focuses on felt experience when distinguishing between emotions, evaluations and intuitions. He describes evaluations as construals of experiences in context on binary scales between positive and negative: good
(plus value) vs [sic] bad (minus value), desirable vs [sic] undesirable, important
vs [sic] not important, lovely vs [sic] hideous etc. (Downes 2000:104). He further mentions so-called discourse level resources which include both lexical and
phoneticcues.
Markers of evaluation differ in written and oral discourse. For oral discourse
these may be: facial expressions, voice quality, prosodic cues, verbal expressions
and also several nonverbal cues. These markers can be uttered simultaneously and
thus occur as one complex statement or they may be uttered independently of
each other. This independent way of looking at several features of speech samples
is also necessary for linguistic-phonetic analyses of evaluative utterances.
2.1
Subjective or objective evaluation
Evaluation may be interpreted by the listener or reader as being subjective or

objective. According to Kue (2007), an evaluative utterance is subjective if the
speaker utters their evaluation as a personal experience or perception (persnliches Empfinden), whereas an objective evaluative utterance is said to be true independent of ones personal experiences (subjektunabhngig) (Kue 2007:145).
In the evaluation process, speakers or writers explain their subjective evaluation
by mentioning the circumstances under which an evaluative utterance evolved.
This is especially true for clarifying why a speaker agrees or disagrees with an
How evaluation is transferred in oral discourse in Russian 135
action or a person. We will later see in the rhetorical strategies found in the quasispontaneous speech data that some of the participants answers may be classified
as objective evaluation on the surface (cf. example (4)), but may also be interpreted (by the interlocutor) as being subjective.
2.2
Stances, attitudes, evaluation
Different disciplines employ several and differing terms for the concept of evaluation. In the linguistic literature, we mainly find three terms describing this phenomenon, these are: stance, attitude and evaluation. Apart from these three terms
used in linguistics, there are other terms known from emotion studies, psychology or philosophy which are not considered here.
A slightly different but nevertheless interesting approach for evaluation is
Labovs concept of evaluation for narratives. It is rather related to syntax and covers aspects that go beyond valence (cf. Labov 1972). In the present paper I focus
on the positive-negative scale which is especially helpful for the experimental part
of the study where we have the possibility to keep the three comparable features
constant.
The relevant characteristic for stance or evaluation is valence referring to impressions on a scale between pleasant and unpleasant or good and bad. Such an
impression results in a positive or negative attitude. As Downes (2000:104) notes,
emotions always involve evaluations. However, the categories activity and potency, especially known from emotion studies are not focused in the present article
because they are less relevant in this context.
3. Oral discourse
The term discourse has two meanings: discourse as an abstract set of rules and
procedures for the production of particular discourses, and discourses or groups
of statements themselves (Mills 2008:55, italics in the original). Discourse is a
phenomenon occurring whenever people interact with each other and may be
realised in different modes. Factors influencing a discourse situation are for example nonverbal, tactile or olfactory means. Disciplines studying discourse are
linguistics, literary studies, language acquisition, sociology, but also minor fields
such as political sciences or psychology. Defining oral discourse may not seem
necessary but has to be done especially because of its interdisciplinary character.
A definition that suits both oral and written discourse can be found in Leech and
Short (1991):
136 Nicole Richter
Discourse is linguistic communication seen as a transaction between speaker and

hearer, as an interpersonal activity whose form is determined by its social purpose. Text is linguistic communication (either spoken or written) seen simply as
a message coded in its auditory or visual medium. (Leech and Short 1991:209)
So, discourse may appear both in the written and the oral modality. An essential
characteristic of oral discourse is the involvement of at least two interlocutors
interacting simultaneously. One aim in discourse situations is the attempt of the
interlocutors to convince the other(s) of their own opinions. In oral discourse,
speakers try to convince one another by objective arguments and by evaluating
themselves, the situations and their communication partners. This is done using
the resources described in the following.
4. Realization of evaluation
Evaluative meaning may be transferred lexically as well as via paralinguistic
means. The linguistic realization of evaluation is achieved with the help of specific
evaluative lexemes that have an inherent attitudinal meaning such as adjectives
(uasnyj horrible), adverbs (krasivo nice) or also nouns (krasavica beautiful
woman) but it is also done prosodically and nonverbally. Downes (2000) calls the
linguistic means of evaluation resources and in this context he mentions discourse
level resources, lexical, phonetic and grammatical resources.
When discussing evaluation from a phonetic point of view, the term attitudinal intonation frequently occurs in the literature. Wichmann explains that
attitudinal intonation refers to any intonational cue which, together with nonlinguistic information in the voice, reflects speaker behaviour in a given situation,
either as intended by the speaker or inferred by the hearer or, both (Wichmann
2000:145). This explanation shows that evaluation does not necessarily depend
on successful transfer, as it may also be observed if it is only intended.
Thus, attitudinal prosody is a pragmatic device to convey emotive meaning.
We know from Downes that the resources for the emotional and the evaluative
overlap, in that emotion always involves some sort of evaluation (2000:108). This
is the reason for prosody studies on emotions show a number of similarities in
their findings.
5. Experimental vs. natural data

There are several ways to use speech data for the investigation of the research
question outlined above. One can use experimental speech data or natural speech
data. In the following, I will discuss these two approaches used in linguistics to
show that both can contribute to the investigation of prosody or rhetorical questions. In laboratory experiments, place, speakers and conditions are usually fixed.
The advantage lies in the comparability of the achieved data and the situation in
which they are produced, even if one could object that there is less naturalness. In
phonetic studies, this type of experiment is used quite frequently because comparability is a major tenet in phonetic analysis. Equivalence between speech material, participants and the setup of the recording is essential.
Natural speech data occur in recordings taken from speech corpora including
authentic speech, in so-called field experiments or in field studies (cf. Kerlinger
and Lee 2006). The advantage of this kind of data is their naturalness and authenticity, thus a compromise has to be made at the expense of data comparability.
For the actual phonetic analysis, there might also be some technical disadvantages concerning the sound quality of the recordings. However, a real advantage
of natural speech data is that it is produced independently of a linguistic research
question. Such a procedure can make the data more objective and authentic than
experimental speech. Natural speech can include discussions, interviews or everyday informal conversation.
With all the advantages and disadvantages just mentioned, it is hard to decide
for either laboratory or natural speech, because actually, it is difficult to know
if artificiality is a weakness or simply a neutral characteristic of laboratory experimental situations (Kerlinger and Lee 2006). An experimental situation alone
does not make speech less natural. Discourse produced in a linguistic experiment is also a discourse situation with real interlocutors, although special. In the
analysis one has to be aware of the kind of speech data you are dealing with. The
research question determines the type of data one is using. The present study
includes experimental speech data as well as quasi-spontaneous speech data and
thus offers material for a broader analysis.
6. Experimental setup
In the speech experiments, native speakers of Russian were asked to read small
dialogues and some of the participants later took part in an experiment aiming at
quasi-spontaneous speech (cf. Richter 2009 for further details). None of the participants had any prior training in speaking and none were professional speakers
in the sense of public speaking.
138 Nicole Richter
6.1
Participants and conditions read speech
The experiment on read speech was conducted with twelve female Russian native
speakers reading dialogues which were embedded in specific contexts. The age of
the participants ranged from 18 to 45 years. The participants read fourteen dialogues containing utterances embedded in an evaluative context. One sentence
from the answers of all dialogues was lexically, syntactically and segmentally
identical in all versions. The three versions were the following: positively evaluating and negatively evaluating and neutral. The invariable declarative sentence
was the one to be compared and analyzed prosodically. Only the context of the
dialogues was evaluative in its meaning. The analyzed sentences in the examples
(1a) and (1b) are underlined.
The declarative sentences of eight speakers were included in a perception test
during which independent hearers had to identify the intended reading by only
listening to the prosody without being presented with the lexical evaluation.
6.2 Data collection read speech
The recordings took place in a phonetic laboratory. The participants spoke into a
microphone in front of them, so that the distance was kept constant for all speakers. The digital recordings were done using a DAT-recorder. The speech samples
were digitized and normalized using the program Cool Edit. The pitch contours
were analyzed using the speech analysis programme PRAAT (see www.praat.org
and Boersma and Weenink 2004).
6.3
Participants and conditions quasi-spontaneous speech
The setup for the field experiment on spontaneous speech was as follows: the participants were given some thematic restrictions which was achieved using photographs, pictures and an explanation asking them to evaluate the given objects
or situations. The answering participant always had to take the opposite position,
so that both speakers expressed positive and negative stances during their talk
in equal parts. Two female participants were chosen after the perception test, as
this showed that the read evaluative versions of these two participants had been
recognized best by independent hearers. The hearers were also native speakers of
Russian. The two speakers achieved exceptional recognition rates between 79%
and 81%. Compared with data from similar tests in emotion studies these are
respectable results (cf. Paeschke 2003).
6.4 Data collection quasi-spontaneous speech

The recordings for this experiment also took place in a phonetic laboratory in
order to achieve a high sound quality. However, this time the distance between
speaker and microphone could not be kept constant all the time, as the conversations were rather lively which was anticipated and even requested. The speech
material can be described as semi- or quasi-spontaneous due to its thematic restrictions on the one hand and its spontaneity of the interlocutors on the other.
After a few minutes the participants interacted quite naturally which was supported by the fact that the two speakers knew each other before.
6.5
Speech material
The speech material was consistent for the reading experiment, where prepared
speech was used. These dialogues contained evaluative speech so that the positions, in these cases positively or negatively evaluating, could be well perceived.
However, each dialogue contained one sentence not showing any evaluative lexeme or attitudinal meaning. I give one such dialogue for illustration in (1).
(1) A: U vas tam na proloj nedele priexal novyj kollega. On otkuda?
You got a new colleague last week. Where is he from?
a. B: On sejas v Tveri ivt. On oen intelligentnyj elovek i mnogo
rabotaet. Ja ego xoroo znaju. My e vmeste v KOlu xodili.
He is now living in Tver. He is a very clever person and works hard.
I know him well. We went to school together.
b. B: Ne znaju, gde on ivt. On asto pereezaet. Znaju to s nim oen
trudno rabotat. My e vmeste v KOlu xodili.
I dont know where he lives now. He moves a lot. I know that its very
hard to work with him. We went to school together.
Apart from the actual material there were also fillers included in order to distract
the speakers from the intended purpose of the reading task and to avoid a training effect.
In the second experiment the material was produced spontaneously in the
conversations and will be illustrated in the analysis part in Section 7. The spontaneity can be observed in the flexible word order, variant morphological means as
well as in the differing and more heterogeneous prosodic form.
140 Nicole Richter
7. Analysis of prosodic and rhetorical features

Analyses of conversations can be done with the help of different linguistic frameworks from phonetics, prosody, conversation analysis or rhetoric. Below I will
pick out some prosodic features to illustrate how the speakers in the experiments
produced the evaluative utterances in question. I will then concentrate on the rhetorical strategies used by the speakers in the experiment on spontaneous speech.
7.1
Prosodic features
Fundamental frequency (Hz)
In the following, I will give some of the results of the prosodic analysis from the
reading task and then compare these with quasi-spontaneous speech data. Pitch
does play a role for perceiving and comprehending evaluation. We can draw this
assumption from several emotion studies and test it in our data in the reading
task of the experiment. Tonal cues responsible for evaluation are pitch intervals
and pitch level, both for certain syllables as well as for the whole utterance.
Figure 1 shows the realization of the same item, cf. (1) My e vmeste v KOlu
xodili. (We went to school together) read by one speaker taken from two different evaluative contexts (positive and negative). The line above shows the positively evaluating version, the plain line below the negatively evaluating version of
the target sentence My e vmeste v KOlu xodili. An observation concerning the
global pitch parameters is the higher pitch level and a bigger pitch intervall for
positive evaluation over the whole utterance. However, the pitch pattern relevant
for declarative sentences does not seem to be affected. The linguistic function of
pitch seems to be underlying and is therefore still very well perceivable. Higher
300
200
150
My
100
0
vmeste
v KOlu
Time (s)
xodili.
1.46209
Figure 1. Typical pitch contours of the item My e vmeste v KOlu xodili (positive
evaluation: line above until kolu, negative evaluation: line below until kolu; speaker 4)
pitch is even more prominent on the syllable carrying the main stress of the sentence KO- in both evaluative conditions. In the positive version of the accented
bisyllabic word kola (school), the falling pitch movement starts off with a high
peak on the first syllable and results in a very steep fall on the second, with a remarkable interval of 142 Hz.
When compared, the prosodic cues of the data in the experiment on read
speech tend to be quite similar to the cues in the more spontaneous speech data.
The tonal cues show bigger intervals for quasi-spontaneous speech, especially for
positive evaluation. Concerning pitch level, negative evaluation has lower mean
pitch than its positively evaluating counterpart. Pitch level is determined globally
and is a mean measured over the whole utterance.
An intra-speaker comparison shows that speech in two different experimental settings points at similarities but also yields differences in the prosodic form.
Independently of the evaluative poles, there are heightened f0 maxima and lowered f0 minima to be found in the (quasi-)spontaneous speech data. But, this observation can be described as an intensified variant of positive evaluation in the
more lively or unconstrained way of speaking.
Figure 2 shows the mean fundamental frequency values for all items used in
the perception test. For the items of these eight speakers the overall picture is unusually definite. The mean fundamental frequency being measured globally over
the whole utterance is highest for the positive evaluation and considerably lower
for the negative one.
Mean fundamental frequency for eight speakers (eight items)
250
Fundamental frequancy in Hz
245
POS
240
235
230
225
NEG
NEUTR
Mean f0
220
215
210
205
200
Figure 2. Mean fundamental frequency of the neutral and evaluative versions of eight
items spoken by eight speakers
142 Nicole Richter
The least mean value, however, was found in the neutral non-evaluative variant. It can be argued that in these cases less speaker involvement leads to less pitch
and thus to lower f0 values than in negative evaluation.
7.2
Rhetorical features
Since speakers jointly use rhetorical features and prosodic markings, the following section concentrates on these features. In the data, some quite prominent
and convincing communicative and rhetorical strategies are to be found. These
strategies will be analyzed from a rhetoric perspective, also taking into account
argumentation.
Speakers may express their agreement or disagreement by different means, so
that both often have to be inferred by the hearer in the communication process
(for disagreement cf. also Rathmayr 1989 and Pomerantz 1984). In the experiment presented here, the speakers regularly used questions as a reaction to their
interlocutors statement. This strategy has already been described in the literature,
especially for Russian data. A question in this context transfers a rejection without containing a lexical negation and thus can be interpreted as an indirect way
of rejection. Yokoyama (1990) discusses the phenomenon that questions used as
answers may be interpreted as rude or as friendly in different situations. Rudeness
according to her is not only related to the question form as such. She distinguishes
between metinformational and informational questions and suggests that the type
of question may influence the different perception concerning friendliness or
rudeness of the utterance.
In the next three examples I will illustrate the strategies employed by speakers
in my sample.
(2) A: Oj! Vot tono takuju e ja videla u Lenki, i sebe by toe kupila.
Oh exactly this one I saw with Lena as well and I would [also] like to buy
one myself.
B: Vot takuju uasnuju krovat?
Such a horrible (lit.) bed?
In example (2), the speaker refers to the object praised by the interlocutor and
introduces a negative evaluating attribute which shows that she disagrees with
the others opinion. By choosing a question containing a nominal phrase with the
demonstrative pronoun takuju and the attributive adjective uasnuju the speaker
states the evaluated object as if it was objectively described and known as such.
The description can be interpreted as an objective evaluation.
(3) A:

B:

kakoj nevynosimo-siniij ciferblat!

What an unbearable-blue clock face
Tebe ne nravitsja takoj cvet?
Dont you like this [such a] colour
Example (3) expresses a strategy where the speaker with the help of a question
switches from a factual to a more meta-communicative level (cf. Richter 2005:180).
So, the question Tebe ne nravitsja takoj cvet? refers to the position A has just stated. She is not delivering her own opinion, but asks for the interlocutors view and
is directly referring to the colour mentioned. She seems to request whether this
is truly the opinion of the other speaker and thus tries to irritate her interlocutor
(maybe hoping to convince her of another opinion). So, this is a rather indirect
transfer of an opinion. One aim for doing so could be to unsettle the interlocutor,
as reflecting on such a question could perhaps change her opinion.
Metinformational questions, as used in the examples above, appear not to
satisfy the ordinary conditions for questions, because in (2) and (3) the speaker
already knows the answers and aims at something beyond the stated proposition
(cf. Yokoyama 1990: 8). They can even be interpreted as indirect ways of imperatives that may serve as a technique of rejecting something.
(4) A:

B:

eto kostjum u ne. Von v krasnom, krasnom, krasnom kostjume

This suit of hers. There, in [this] red, red, red womans suit
to ty v nem krasivogo nala?
What do you find nice (lit.) about it?
Example (4) shows a question that asks for the actual reasons that lead the speaker
to her evaluation. It is similar to the metinformational use in example (3) but in
(4) the speaker asks for more details in the process of evaluating. Speakers do
not necessarily have to be aware of this process and the underlying reasons. The
question to ty v nem krasivogo nala? in this way distracts the interlocutor from
the actual conversation. The question addresses the interlocutor in a very direct
way which requires a close interlocutor distance (for interlocutor distance cf.
Yokoyama 1994).
So, the rejection in examples (2)(4) is not stated directly, but is conveyed
with the help of questions and in that way may trigger the interlocutors to think
differently about the attitudes they expressed earlier.
144 Nicole Richter
8. Summary and outlook

As has been shown, different techniques of evaluating and of rejecting were used
by the speakers in the experiments of the study. The techniques include several
tonal features but also special lexical items found in quasi-spontaneous speech.
Interestingly, from a rhetorical point of view I have illustrated the importance of
questions being used as a means of rejecting. This rhetoric technique has been
used quite regularly by the participants. In the contributions there was attitudinal
meaning on the one hand and prosodic attitudinal marking on the other hand.
Both devices were produced simultaneously.
The data suggest a promising investigation of prosody and rhetorical strategies used in the realization of agreement and disagreement which I am going to
extend. Research on spontaneous evaluative speech has to be pursued. The observations lead to the assumption that within spontaneous conversations rhetorical
strategies can be properly studied. Work on rhetoric and other pragmatic principles has to be extended not only for Slavonic languages but also should include
authentic speech data like interviews or talk shows taken from speech corpora.
References
Boersma, Paul and Weenink, David. 2004. Praat. Doing Phonetics by Computer, Version 4.2.19,
[Computer programme]. Online: http://www.praat.org, accessed on 28 July 2011.
Caffi, Claudia and Janney, Richard. 1994. Toward a pragmatics of emotive communication.
Journal of Pragmatics 22: 325373.
Downes, William. 2000. The language of felt experience: emotional, evaluative and intuitive.
Language and Literature 9 (2): 99121.
Kerlinger, Fred and Lee, Howard. 2006. Foundations of Behavioral Research. Melborne et al.:
Wadsworth/Thomson.
Kue, Holger. 2007. Positives Bewerten. Diskurssensitive Beispiele aus dem Russischen und
dem Tschechischen. Zeitschrift fr Slawistik 52: 138177.
Labov, William. 1972. The transformation of experience in narrative syntax. In Language
in the Inner City: Studies in the Black English Vernacular, William Labov, 354396.
Philadelphia: University of Pennsylvania Press.
Leech, N. Geoffrey and Short, Michael H. 1991. Style in Fiction. A Linguistic Introduction to
English Fictional Prose. London/New York: Longman.
Mills, Sara. 2008. Discourse. London/New York: Routledge.
Paeschke, Astrid. 2003. Prosodische Analyse emotionaler Sprechweise. Berlin: Logos.
Pomerantz, Anita. 1984. Agreeing and disagreeing with assessments: some features of preferred/dispreferred turn shapes. In Structures of Social Action: Studies in Conversation Analysis, J. Maxwell Atkinson and John Heritage (eds.), 57101. Cambridge et al.:
Rathmayr, Renate. 1989. Ein net ist noch lange kein net. Ablehnen und Insistieren. In
Slavistische Linguistik 1988 Referate des XIV. Konstanzer Slavistischen Arbeitstreffens,
Wolfgang Girke (ed.), 245270. Mnchen: Sagner.
Richter, Nicole. 2005. Sachliches und emotionales Argumentieren in Gesprchen: Gemeinsamkeiten und Gegenstze. In Beitrge der Europischen Slavistischen Linguistik (POLYSLAV)
8, Markus Bayer, Michael Betsch and Rafa Zimny (eds.), 174182. Mnchen: Sagner.
Richter, Nicole. 2009. Prosodie evaluativer uerungen: Experimentelle Untersuchungen zum
Russischen. Frankfurt/M.: Lang.
Wichmann, Anne. 2000. The attitudinal effects of prosody, and how they relate to emotion. In Speech and Emotion 2000, 143148. ISCA Archive: http://www.isca-speech.
org/archive_open/speech_emotion.
Yokoyama, Olga. 1990. Responding with a question in colloquial Russian. In Topics in Colloquial Russian, Margaret H. Mills (ed.), 116. New York et al.: Lang.
Yokoyama, Olga. 1994. Iconic manifestation of interlocutor distance in Russian. Journal of
This is how I see it

No-prefacing in Polish
Matylda Weidner
A number of conversation analytic studies have suggested that asymmetries of

knowledge and access to knowledge and experience contribute to the organization of the interaction. Building on previous research on the epistemics of social
relationships, this paper draws on audiotaped doctor-patient data and uses the
methodology of conversation analysis to examine one practice for managing
access to knowledge and dealing with issues related to epistemic status in Polish talk-in-interaction no-prefacing. My analysis shows that turn-initial no in
responsive turns takes from the local (sequential) context and the participants
respective epistemic frames involved and becomes the participants resource for
managing their socioepistemic relationships toward a collaborative, corroborative or competitive outcome.
Keywords: conversation analysis, Polish, knowledge in interaction, doctorpatient interaction, particles
1. Introduction
This chapter takes a look at one specific resource in Polish, namely the particle
no, and explores its potentials for managing access to knowledge and dealing
with issues related to epistemic status in doctor-patient interaction. A number
of conversation analytic studies have suggested that asymmetries of knowledge
and access to knowledge and experience are the relevant features that contribute
to the organization of interaction. As far as medical consultations are concerned,
for example, these asymmetries reveal themselves as the interaction develops moment-by-moment. The patients and the doctors differential states of knowledge
(Drew and Heritage 1992) are made relevant as the participants negotiate epistemic primacy and authority in interaction. As Raymond and Heritage (Raymond
and Heritage 2006) observe there are various practices of speaking through
148 Matylda Weidner
which participants can make relative access to knowledge and information relevant. This chapter draws on audiotaped doctor-patient conversations and focuses
on one such practice in Polish no-prefaced turns in responsive position. Being
an exceptionally frequent word in Polish (spoken discourse), no has received surprisingly little analytic attention so far. This chapter fills this gap and builds on
previous work on the epistemics of social relations to take conversation analytic
look at a particular practice in Polish no-prefaced turns. As I will show, the
particle can be used as a resource for indexing the patients and the doctors rights
to knowledge and position those rights vis--vis co-participants. No, whose basic
function is to offer some kind of confirmation of what the prior speaker has said,
can also exploit this confirmatory quality to add varying undertones to the turns
of which it is a part. My research reveals that depending on the local (sequential)
context and the participants respective epistemic frames involved no-prefaced
turns can contribute collaborative, corroborative or competitive outcomes. This
paper will focus on one of these interactional effects only. Accordingly, I will show
three examples, in which the relationship of knowledge between the doctor and
the patient is managed toward a collaborative outcome.
2. What is no anyway?
To begin with, let me offer some brief background information on the historical origins of the Polish no as well as a capsule overview of some of its usages, as
described in the dictionaries. Although a number of dictionaries of Polish list no
as their entry (Bako 2000; Doroszewski 1999; Dunaj 1996; Linde 18071814;
Szymczak 19781981), there seems to be no unanimity as to its precise meaning and function. In general no is most often referred to as a word that adds expressive, emphatic or intensifying quality to another word or another utterance.
Moreover, most of the dictionary entries point to the frequent deployment of no in
(spontaneous and colloquial) spoken discourse, where it conveys the relationship
between interlocutors or utterances. The dictionaries suggest that no has three
main functional usages to reinforce the imperative (Szymczak 19781981), accomplish a strong negation (Doroszewski 1999) or contribute a strong confirmation (Bako 2000). Finally, most of the dictionary entries mention the position of
no as either prefacing or following words or utterances.
Leaving the dictionary descriptions aside and moving toward more discourse-oriented studies, Pisarkowa (1975) notes that the function of no seems
to be purely formal, in that it is merely a preface to a beginning of an utterance.
This gives no a status of an entry signal (sygna wejcia) similar to a capital letter
in a sentence of a written text. Wierzbicka (1976) and Kryk (1992) take a more
No-prefacing in Polish 149
pragmatic approach toward no and discuss its function as a discourse particle

and an interjection. The distinction is based on the position of no in an utterance.
Whereas in the initial or medial position no achieves a slowing down effect and
is a typical discourse particle, in the final position no is similar to other emphatic
expressions and functions as an interjection. Kryk (1992) draws our attention to
the fact that no verges on the borderline between colloquial and the substandard
uses of Polish and that the discourse functions of this interjection and discourse
particle may in fact correspond to the functions of English well or yeah. SzulcBrzozowska (2010), whose comparative study of Abtnungsfunktion (roughly
translated as hedging) of modal and tone particles in German and Polish takes as
its objects of investigation ja and no (for German and Polish respectively), stresses
the fact that the semantic-pragmatic extension of the function of no in Polish is
highly dependent on context and intonation. She also notes that unlike the German ja, the Polish particle can occupy different positions in a sentence, thus emphasizing different parts of the utterances propositional content. My analysis of
interactional data partially resonates with Szulc-Brzozowskas comparative findings in that it lends support to the authors suggestion that no enters quite a wide
array of turn construction environments. Accordingly, I have identified four main
positions where this particle can occur TCU-initial, TCU-medial, TCU-final
and as an independent lexical TCU. In the present analysis, however, I will limit
the scope of my examination to cases where no occurs TCU-initially. Even more
specifically, this chapter will only show examples where no is positioned as a first
element in a [no + more talk] compound (for instance No wanie or No wiem).
Turn-initial position has long been appreciated as a structurally and sequentially significant place in the design of turns at talk and the organization of
conversation (Heritage 2002; Lerner 1996; Sacks, Schegloff, and Jefferson 1974;
Schegloff 1987, 1996). Schegloff stresses the systematically- and interactionallyrelevant reasons for why this is the case:
One important feature of turn construction (that is, constructing the talk in a
turn in a conversation) and the units that turn construction employs (e.g. lexical,
phrasal, clausal, sentential constructions) is that they project, from their beginnings, aspects of their planned shape and type.
(1987:71)
Previous research has provided ample illustration that items produced at turn
beginnings not only establish a relationship between the TCU that they initiate
and preceding turn(s), but also can be indicative of the upcoming turns stance
toward preceding talk (cf. Schegloff 2004). Even though due to space limitations I
will not offer any comprehensive review of this literature here, let me just point to
a number of studies that have contributed systematic observations about various
objects that can be positioned at turn beginnings. Some of the well-studied items
150 Matylda Weidner
are, for example: By the way (Schegloff and Sacks 1973; Schegloff 1987), oh
(Heritage 1984, 1984, 1998, 2002; Heritage and Raymond 2005; Heritage 2005,
2011), well (Pomerantz 1984; Schegloff and Lerner 2009), look (Sidnell 2007,
2010), and (Heritage and Sorjonen 1994; Bolden 2010), so (Bolden 2006, 2008,
2009). This list is by no means exhaustive, if only for the reason that it is limited to examples from English only. However, since the main aim of this chapter
is to move beyond a comparative examination, the observations presented here
are only meant to provide a starting point for the analyses on the ensuing pages.
Given that the function of the particle no in Polish seems to be directly related to
issues of knowledge in interaction in general and the interactants epistemic status
in particular, a brief overview of conversation analytic studies in this domain will
be offered next.
3. Knowledge in interaction
Last decennium has seen a particular expansion of research related to aspects of
knowledge in interaction within the tradition of conversation analysis. A common thread that binds these studies together is the issue of management of rights
and obligations related to knowledge and its centrality to the organization of sequences. It has been demonstrated that organized practices of speaking can be
vehicles for conveying the participants relative rights to knowledge and a way of
describing what they know. As far as this knowledge-related form of organization
is concerned, Heritage and Raymond (2005), show that interactants constantly
deal with their rights to identity-bound knowledge and by doing so produce and
reproduce the epistemics of social relations (Raymond and Heritage 2006). This
is important not only in ordinary conversation but even more so in institutional
talk. In medical interaction, for example, knowledge is clearly distributed along
an asymmetrical axis (Heritage and Robinson 2006; Perkyl 2002, 2006). Hence,
whereas doctors have the right and the necessary expertise to diagnose and suggest proper treatment, patients have the knowledge of their subjective pain experience and the right to describe this experience.
Since interaction doesnt occur in vacuum, interactants recurrently position themselves relative to the degrees of knowledge they have and their rights
to communicate this knowledge in different social configurations and situations
(Raymond and Heritage 2006; Schegloff 2007; Stivers, Mondada, and Steensig
2011). Research has shown that the primacy of those rights can be negotiated in
the course of interaction, so that we can in fact speak about systematic but adjustable asymmetries in the relative positioning of the interactants regarding their
rights to tell, inform, assert or assess. Additionally, the access to knowledge or a
particular piece of information can be individual or shared as well as vary on a

scale from direct to indirect (first-hand or derivative) or substantial to minimal
(see Stivers, Mondada, and Steensig 2011 for a comprehensive discussion of numerous issues related to epistemics). Hence, for instance, participants with a more
proximal and substantial knowledge within a given epistemic domain are granted primary epistemic rights to assess the matter at hand (Stivers, Mondada, and
Steensig 2011; Heritage 2011; Heritage and Raymond 2005; Drew 1991). Research
has also contributed empirical evidence that the speakers epistemic primacy
(that is their greater rights to knowledge relative to that of their co-interactants)
might overarch the unfavorable sequential positioning of their contributions. In a
now classical study of oh-prefaced responses to assessments, Heritage (2002) has
shown how an assertion of ones own independent access to a piece of knowledge
from second position, indexed by oh-prefacing, can undermine the first speakers
epistemic claims.
As I will show in the following sections, in Polish responsive turns that are
marked with turn-initial no seem to be a commentary on their speakers epistemic
status regarding their-side knowledge, their-side awareness and/or their-side experience. Given that the semantics of no does not fall into any fixed denotational
field, there is a sense in which its function can be deictic in nature. Hence, we will
see that depending on its specific placement, [no + more talk] can be a fugitive
commentary on the speakers state of mind (Heritage 1984) in the given local
context of the interaction. My analysis suggests that when no occurs TCU-initially, and more specifically where it is positioned as a first element in a [no +
more talk] compound, it indicates the speakers my side epistemic evaluation
of some preceding information. This evaluation may not always be completely
in line with the evaluation proposed by the other speaker. As a result, depending
on the context and the epistemic frames involved, no-prefacing can contribute
different undertones to the turn of which it is a part. In the ensuing paragraphs
I will show three instances of no taken from medical interaction, where patients
and doctors monitor their talk and co-construct their interaction based on who
knows what and who has primary rights to express this knowledge.
4. Analysis
The analysis in this chapter is based on data from an audiotaped corpus of doctor-patient interactions in a large Polish city. The data were transcribed according to conventions originally developed by Gail Jefferson (for an overview see
Atkinson and Heritage 1984). The transcripts contain three lines: the Polish
original, the morpho-syntactic gloss and the idiomatic English translation. The
152 Matylda Weidner
recordings yielded a total of 699 instances of no, which were subsequently grouped
into four categories TCU-initial, TCU-medial, TCU-final and independent
TCU. The placement of no TCU-initially has proven to be the most numerous category, with 321 instances in my collection. Even though this chapter will discuss
only three cases, the findings I will present are based on a thorough analysis of all
cases in the collection. Moreover, although this chapter explores one interactional
practice drawing on doctor-patient conversations, the observations put forward
in the ensuing pages seem to extend beyond this institutional domain and into the
realm of ordinary conversation.
5. [No + more talk] my side collaboration

The instances I will subsequently present show that [no + more talk] seems to be
a turn-initial indication of the speakers my side epistemic evaluation of some
prior information. My analysis of all cases of no-prefacing in my doctor-patient
data, only part of which will be presented in this chapter, shows that this evaluation may not always be completely in line with the evaluation proposed by the
other speaker either in the just prior turn or in the local sequential environment.
At times, it works to re-calibrate the speakers epistemic position into that of analready-knowing-recipient and contributes a sort of an epistemic injection into
an ongoing sequence. Other times, the speaker of [no + more talk] exposes the
obviousness of the co-participants prior contribution and asserts his/her own
epistemic primacy regarding a particular piece of information. In consequence,
a TCU-initial no can contribute different undertones to the turn of which it is a
part collaborative, corroborative and competitive. Only the first of these options will be taken up in detail here and illustrated with specific examples, which
will show three different versions of no-prefacing. Obviously, these examples
do not exhaust the range of possible formats, but space limitations mandate selectivity and thus the focus will be restricted to these collaborative no-prefaced
responsesonly.
. The term my side (telling) comes from Pomerantz (1980) and describes a particular
practice whereby the speaker provides for the recipient to volunteer information, perform a
service, make an offer, extend an invitation, remedy an offense, answer an accusation, and so
forth without his directly being asked, requested, accused, or complained against (ibid:193).
Importantly, the my side teller orients to the fact that he/she has limited or less than best access [to the information] relative to the recipients access (Pomerantz 1980:190). The use of the
term my side (evaluation) throughout this chapter is not related to the practice described by
Pomerantz.
One common compound in my data is no (wic) wanie no exactly. Excerpt(1) illustrates this format. The fragment comes from a neurological consultation (the patient has been diagnosed with encephalitis) and begins at a point
where the consultation has been reopened by the patient. Prior to where the transcript begins here, the doctor has already started preparing the patients documents and making arrangements for a follow-up visit. At this point, the patient
re-opens the consultation and inquires about the chance that his immune system
will become weaker after all the drugs that he received in the hospital (lines 766 to
770) (see Robinson 2001 for his discussion of how patients re-open consultation
closings to topicalize new concerns; Schegloff and Sacks 1973 for their discussion
of how making arrangement is closing implicative; West 2006 for her account of
coordinating closings in medical consulations). The consultation takes place at
the end of August and the patient already learned from the doctor that the earliest he will be able to get back to work and normal life will be mid-October (data
not shown). Moreover, the patient has just had encephalitis (the reason for his
hospitalization) and has been on medication throughout his hospital stay. Additionally, the doctor has just prescribed more drugs, which the patient is advised
to take after he leaves the hospital and for an extended period of time. All that
contributes to the patients concern that his recent encephalitis, together with the
co-occurring bad season (the patient will leave the hospital in the fall), will make
his immune system weaker. That is why he suspects his increased susceptibility
to various diseases, which validates the patients question at lines 766770. In her
subsequent turn the doctor addresses the patients concern and in doing so takes
into account the bad time of the year (lines 779 through 785).
(1) Encephalitis, Neurology
766 P: Czy: po tym jak ju (.) wezm=
if
PRP this how already take.1SG
Wi:ll (it) after I have (.) taken=
767 =ten to (0.2) .hhh e tak poem=
this it
that yes say.1SG
=this (0.2) .hhh so to speak=
768 =bedzie
mona, (.) e
bior (.)=
be.3SGFT can.3SG
that take.1SG
=will (it) be possible, (.) as Im taking (.)=
769 =ten (.) lek,=organizm
sie=
this
drug.ACC organism R
=this (.) drug,=(my) body will=
770 =potem uodpo:rni:
then immunize.3SG.FT
=get immu:ne: afterwards
154 Matylda Weidner
((8 lines omitted))

779 D: =Natomiast=
while
=And=
780 =(0.3) y::::: pniej, no^ to=

later
PART it
=(0.3) y::::: then, no^ then=
781 =bdziemy m:- (.) pniej=
be.1PL.FT
later
=we will th:- (.) then=
782 =bdziemy
sie zastana:wia. Troche=
be.1PL.FT R
wonder.INF
a bit
=we will be thinki:ng (what to do). =
783 =jakiego takich (.) kapsuek z=
some
such.PL
capsule.PL with
=some (.) cod-livel o::il=
784 =tra::nem, (0.2) bo
to bedzie=
cod-liver oil
cause it be.3SG.FT
=capsules, (0.2) cause it will be=
785 =akurat je:sie.
just
fall
= fa:ll.
786 P: No [wic wanie. Dlatego
mw]i=
PART so
just
thats why say.1SG
No [thats exactly it. Thats why Im s]aying=
787 D:

[Take taki niedobry o:kres.]

so
such not good period
[So (it will be) such (a) bad ti:me.]
788 P: =e:by to:,

to
it
=tha:t ,
It is worth to remember that it was the patient who first voiced his concern about
the correlation between the amount and the kinds of drugs he received in the
hospital and his potentially decreased immunity. However, up to this point he has
never explicitly oriented to the relationship between these aspects and the precise
timing of his discharge from the hospital. In her response, the doctor addresses
the patients concerns and offers a long-term treatment plan that establishes a
link between the patients decreased immunity (line 783/784) and the fact that
the patient will leave the hospital in the fall (line 785). It is to the latter that the
patient orients in his successive turn at line 786. The first TCU in the patients
turn, No [wic wanie. No [thats exactly it., not only confirms the doctors prior
observation, but also asserts the patients independent access to it. In order to
get a full grasp of the development of this sequence, let me provide some more
contextual information, which lies outside the direct availability of this fragment.
Earlier on in the consultation, the patient asks the doctor when he will be able to
go back to work. The doctor does not offer any final answer but instead tells the
patient to wait for the results of the second lumbar puncture (which will show the
extend to which the patients condition has improved). The patient, however, performs a simple mathematical calculation and proffers a conclusion that he will be
able to get back to normal life somewhere mid-October. The doctor validates this
observation. Now, going back to the patients turn at line 786, No [wic wanie.
No [thats exactly it., evaluates the doctors prior contribution (lines 779785)
from the patients own point of view. The patient responds to the doctors prior
turn (line 785), in which the doctor mentions Fall, with a confirmation of his own
observation (back from earlier on in the consultation, see Note 2, lines 386389),
thus exposing the obviousness of the doctors noticing. What the patient says can
be in fact glossed as yeah thats exactly what I have in mind. However, in this case,
even though the resonance of this response is that of confirming the obvious, the
patients my side evaluation of the doctors prior turn contributes to a meeting
. This is the sequence that provides the background information to the current sequence:
382 D: =ale (.) y no=
but
PART
=but (.) y no=
383 =mwi, (.) s
t
prac
tak nie=
say1SG
with this work.INST yes no
=like (Im) saying, (.) do not make =
384 =planowa. (.) Tak jak mwilimy.=
planINF
yes how say.1PL.PT
=any job plans. (.) Like (were) saying.=
385 =(.) Bo:
to: si [raczej przeci^ga.

cause it R
rather prolong.3SG
=(.) Cause: i:ts [ taking lo^nger.
386 P:

[To: ju
wtedy=
it already then
[I:t will already=
387 =bdzie (.) to ju

jest
poowa=
be.3SG.FT it already be.3SG half
=be (.) it is already mid=
388 =(.) padziernika, kiedy=

OctoberGEN
when
=(.) October, when=
389 =n[ajwczeniej_

earliest
=at [(the) earliest_
390 D:

[W te::j (0.2) chwili (.) tak.=

in this
momentLOC yes
[At thi::s (0.2) point (.) yes.=
156 Matylda Weidner
of minds. Note that just after the first element of the patients No [wic wanie.
No [thats exactly it. turn is out there, the doctor starts talking (at line 787) in
overlap with the patients current turn. Implied in the doctors overlapping turn
is the shared understanding that Fall is a bad season, with frequent outbreaks of
all kinds of infectious diseases (which points to the you and I both know it understanding from the patients turn). Hence, what we can see here is a [no + more
talk] turn, whose interactional contribution is that of collaborating in bringing
about this particular shared understanding.
Excerpt (2) shows an instance of another frequent variant of [no + more
talk] No oczywicie No of course. The fragment shows a small segment of an
interview between a neurologist and a patient with multiple sclerosis. This discussion takes place after the treatment has already been recommended and accepted by the patient. Before moving to the consultation to a close the doctor
inquires about the patients health insurance (the question being whether the patient receives pension). In response the patient provides more information than
the doctors question requested (Stivers and Heritage 2001). In his answer, the
patient tell the doctor that he does receive pension and mentions that he is aware
of the fact that he will never be able to work again (so that his pension is his only
source of income). At this point the doctor asks the patient about the nature of
his professional occupation (data not shown). When it turns out that the patient
is a carpenter (who still does some minor jobs, but only as a hobby), the doctor
starts bantering with the patient and says that she will buy some old furniture and
commission the patient to renovate them for her. The patient seems to be taking
the matter quite seriously and says that some other doctor (probably during a different hospital stay) has already commissioned him to renovate a large wardrobe.
Even though the patient did it at that time, now he admits that he would rather
renovate some small pieces of furniture (the example he gives is an old hanging clock). Importantly, the patient points to a specific reason for his reservation,
which is where the fragment below starts (line 744).
(2) Multiple Sclerosis, Neurology
744 P: =A
przy due meble
to sie (kurcze)=
and PRP big furniture.PL it R
chicken
=And with big furniture (you) have to=
745 =trzea gdzie nachodzi przy=
one has to
somewhere walk.INF.PRP
=walk (damn) a lot when doing=
746 =tym.
this.LOC
=them.
747 D: =No^ oczywi:cie, (.) no oczywicie.=

PART obviously
PART obviously
=No^ of cou:rse, (.) no of course.=
748 =No: (.) to nie jest, nie jest [t PART
it no be.3SG no be.3SG
=No: (.) its not, its not [i-
At lines 744 through 746, the patient gives a self-explanatory account of why he
doesnt like to work on big pieces of furniture. The patient is suffering from multiple sclerosis and was partially paralyzed at the time of his current admission
to the hospital. Now, a couple of weeks later and following an appropriate treatment, the patient is able to walk again, albeit slowly and only for short periods of
time. This fact is made relevant in this fragment of the interaction, when the patient says that bigger items, such as a wardrobe, require a lot more mobility (line
744745) (whereas in case of smaller objects, the patient can work on them while
seated). Given that the doctor has earlier asked the patient whether he would
consider renovating a (potentially big) piece of furniture for her, the doctor is
now put in an awkward position of maneuvering out of this delicate situation.
Immediately after the patients turn reaches a possible completion (line 746), the
doctor responds to the patients assertion. Note that the doctors No^ oczywi:cie,
(.) no oczywicie. No^ of cou:rse, (.) no of course. (line 747) is delivered contiguously, which in the context of this sequence (i.e. the issue of working on big
versus small furniture, and its relation to the patients health) may be in line with
Sacks (1987 [1973]) observation that:
() there is an apparent interaction between the preference for contiguity and
the preference for agreement, such that, if an agreeing answer occurs, it pretty
damn well occurs contiguously, whereas if a disagreeing answer occurs, it may
well be pushed rather deep in to the turn that it occupies.
The doctors response is aligned to the agreement-preferring design of the patients prior turn. Moreover, the design of the doctors turn displays an empathic
stance toward the proposition contained in the patients account and thus affirms
the patients positive face (Brown and Levinson 1987). Not only is the doctors
turn delivered contiguously, but also additionally marked with a pitch peak at
the beginning of it (Keevallik 2010 showed that affiliative actions are frequently
indexed with a pitch peak at the beginning). In line with my previous observation about the possible function of [no + more talk] as a practice for indexing the
self-evident nature of the co-participants contribution, here as well No^ oczywi:
cie, (.) no oczywicie. No^ of cou:rse, (.) no of course. asserts the obviousness
of the patients prior turn. When the doctor evaluates the patients assertion from
158 Matylda Weidner
my side point of view, she attends to the shared domain of knowledge that the
patients prior turn has invoked. It is an obvious fact that the task of renovating
bigger furniture is more physically demanding than working on smaller pieces.
Therefore, when the doctors turn confirms the obviousness of the patients prior
contribution (at line 747), her utterances are in fact affiliating with and supportive of the proposition contained in the patients assertion. In consequence, the
doctors No^ oczywi:cie, (.) no oczywicie. No^ of cou:rse, (.) no of course.
collaborates with the patients perspective. Moreover, by evaluating the patients
explanation as an obvious reason for his reservation, the doctors turn works toward minimizing the potential negative implications of her prior inappropriate
request for the larger context and the further development of this sequence. This
example illustrates that a particular turn design and its positioning in a sequence
contributes to re-establishing and maintaining a solidary relation between this
doctor and this patient, in what has become a rather delicate local context (see
Clayman (2002) for a relevant discussion of the relationships between sequences
and solidarity).
The third example shows yet another variant of the [no + more talk] practice
that occurs in a fairly complex neurological consultation. The patient presents
with a number of sensory-related symptoms such as problems with hearing, vision and vertigos. Even though over the entire duration of his hospital stay the
patient has undergone multiple examinations, there has still been no definitive
diagnosis as to what the underlying cause of these symptoms may be. That is why
in the course if this consultation the doctor and the patient have to deal with the
unknown status of the patients condition. Even though the test results have been
inconclusive so far and the doctor is still waiting for the patients MRI, the patient
feels that his problems with vision, hearing and sense of equilibrium have to do
with some anomalies in his brain. As the consultation unfolds, at one point the
patient finally discloses that he thinks that he may have a brain tumor. This belief
is based both on the patients own intuitive interpretation of the facts and the doctors reference to the magnetic resonance as crucial for the diagnosis. In consequence, the patients suspicion about his condition seems to reflect his knowledge
and understanding, namely that MRI of the brain points in the direction of tumor
as an underlying cause of the patients problems. In response to the patients concern, the doctor tries her best to reassure the patient that from her point of view
the patients current symptoms do not look like typical brain tumor symptoms. At
the same time the patient goes out of his way to say that of course he wouldnt like
there to be a tumor in his brain (the data fragments are too lengthy to represent
them here, but see Note 4 below for this particular sequence). The doctor uses
her knowledge and expertise to do her best to try to steer the patient into positive
thinking in anticipation of the test results. This, then, provides the larger context
for Excerpt (3) below.
(3) Tumor, Neurology
435 D: =tak czy owak wolaabym
dosta ten=
yes CONJ other prefer.1SG.COND get.INF this.ACC
=anyways (I) would like to get this resonance=
436 =rezonans (0.2) gdzie bdzie
napisane e=
resonance.ACC where be.3SG.FT write3SG.PT.PV
=(0.2) where it will say that (there) are=
437 =zmian (0.2)
w rezonansie (0.2) nie ma.
change.PL.GEN in resonance.LOC
not have.3SG
=no (0.2) abnormalities (0.2) in (the) resonance.
.
320 P:

[Nie:, ja ta[m niechciabym]=

no
I there not want.1SG.COND
[No:, I [wouldnt like]=
321 D:

[Nat^ychmia:st
immediately
[Ri^ght no:w
]
]
322 P: =niechciabym
a[dnego]
not want.1SG.COND none.GEN
=(I) wouldnt like [any ]
323 D:

[Odpu:k]aj,
knock.2SG.IMP
[Touch] woo:d,
324 P: =niechciabym
a[dne]go guza
ani nic,
ale=
not want.1SG.COND none.GEN tumor.GEN or nothing but
=(I) wouldnt like [any] tumor or anything, but=
325 D:

[No:]
PART
[No:]
326 P: =wo[labym prefer.1SG.COND

=(I)d [rather327 D:

[Na guza
to nie wyglda. Spokojnie.
on tumor.ACC it not look.3SG calmly
[It doesnt look like tumor. Calm down.
328 P: =Wolabym
cokolwiek eby co
tam
takiego=
prefer.1SG.COND anything that something there such
=Id rather there was at least something=
329 =niepokojcego chcia byo
ebym=bymy=
worrying.ADJ at least be.3SG.PT CONJ.1SG.COND
=alarming there so that I=so that we=
330 =wiedzieli
co
jest, od
czego.
know.1PL.PT what be.3SG from what
=knew what it is, where it comes from.
160 Matylda Weidner
((3 lines omitted))

441 D: Yh hyh hyh. Umwmy
si e ja: [y

say.1PL.COND R
that I
Yh hyh hyh. Lets say that I: [y
442 P:

[Ja=jusz[:
I already
[I=[:
443 D:

[Ja=
[I prefer=
444 =wol.
[Dobra:?
prefer.1SG good.ADV
=(it). [Alri:ght?
445 P:

[W ogle bym nic

nie wola.Nie wola.
in all COND nothing no prefer.1SG.PT
[I wouldnt prefer anything. I wouldnt.
446 D: =N::o to prawda, ale: (0.2) ale: ty

przed=
PART it truth
but
but you.2SG before
=N::o its true, but: (0.2) bu:t you said=
447 =chwil
powiedziae e:: y [y chcesz] eby=
moment.INST say.2SG.PT
that want.2SG
that
=(a) moment ago tha::t y [y you want] that=
448 P:

[No
wiem.]
PART know.1SG
[No (I) know.]
449 D: =co
tam
byo.
something there be.3SG.PT
=(there) was something.
In the opening lines of this fragment the doctor refers to her own personal preference (line 435) to convince the patient that he has to stay positive and hold off with
any premature judgment until the arrival of the MRI results. Moreover, whereas
the patient seemed to be expressing his preference for some concrete abnormalities in the MRI that could explain the cause of his symptoms (see lines 328330
in Note3), the doctor contrasts the patients position with her own explicit preference for absence of any abnormalities (lines 436 through 444). It is with this
position that the doctors [Dobra:? [Alri:ght? (at line 444) seeks agreement. The
patient starts speaking again after the doctors first TCU is possibly complete (lines
444 and 445 respectively), beginning his overlapping turn in transition space
. At line 444 the doctor uses the same word wole to prefer that the patient has used previously to describe his intuitive feeling about the presence of a brain tumor (see Note 3, line
328).
(Jefferson 1983, 2004), precisely at the same time when the doctor is producing her
agreement solicitation. As a result, instead of responding to the action solicited by
the doctors second TCU, the patient produces a different utterance, in which he
reverses his previously held position (see lines 328330 in Note 3). Here, at lines
442 and 445 the patient states that he is tired of the uncertainty concerning the
diagnosis and says that in fact he would rather be healthy (than have a tumor) and
have nothing to worry about. At the beginning of her subsequent turn, at line 446,
the doctor first evaluates the patients previous statement in a confirmatory manner, by saying N::o to prawda N::o its true, but then uses this confirmation to link
it retrospectively to the patients own previously expressed divergent view. Following the contrastive ale: bu:t (line 446), the doctor recalls what the patient said
earlier (namely that he would prefer that there was something in the MRI, so that
he knew what was wrong with him). Furthermore, the contradiction between the
patients earlier and currently held position is noticeably reinforced by the stress
on the personal pronoun ty you. In formulating what the patient said earlier the
doctor now seems to be claiming authority over who knows what and who said
what. These claims are additionally warranted in the doctors explicit reference to
the patients recent contradictory statements, conveyed in przed chwil a moment
ago (line 447) (Heritage 2012b). Also, it should be noted that the progressivity of
the doctors turn is temporarily suspended (which is interactionally realized by
y at line 447) at the point of maximum grammatical control (Schegloff 1996),
which in this case is just before the verb chcesz you want (line 447). Following
a sound stretch on e:: tha::t and a subsequent marker y, the patient takes advantage of this vulnerable break in progressivity and before the doctor manages
to continue, the patient comes in with No wiem. No (I) know. Now, this is not a
completely innocent entry into the doctors turn. At the point where the patient
has already been provided with enough resources to infer in which direction the
turn is going, his No wiem. No (I) know. takes issue with the doctors epistemic
claims conveyed in the design of her turn (where the doctor positions herself as
a more knowledgeable participant) (Heritage 2012a, 2012b). Even though in the
local context of this sequence both parties seem to have an equal epistemic access
(Heritage 2012a; Stivers, Mondada, and Steensig 2011) to the referent (which in
this case is a matter of who said what), the doctor gained her access by mere recipiency (she explicitly refers to this at line 446/447 with but it was you who said).
Therefore, the patients position regarding his own preferences is still primarily
his to hold, narrate and/or contest (Heritage 2011, 2012a, 2012b; Sacks 1984).
In this way the patients explicit confirmation of his superior epistemic status (as
in I know it better, after all it was me who said it) renders the remaining of
the doctors turn unnecessary. The patients interjacently placed bid for epistemic
162 Matylda Weidner
primacy, concerning his rights to his own preference, explicitly pushes back on the
doctors claims to these rights. This gives the patients no-prefaced turn a slightly
competitive undertone. However, at the same time the patients No wiem. No (I)
know., which is positioned at a very information-rich projection point, collaborates in bringing about the sense of the doctors turn. It encodes the patients my
side evaluation of the doctors prior contribution, an evaluation that exposes the
obviousness of the propositional content conveyed in the doctors turn. In doing
so, it proposes that both the patient and the doctor know that what the patient
said before (i.e. that he would rather have the MRI show some abnormality in
his brain) is not necessarily what the patient really wants. In this case, both the
sequential position of No wiem. No (I) know. and the epistemic frames involved
contribute to the interactional effect of this [no + more talk] turn.
In sum, Example (3) presents a complex case, where the patient and the doctor are faced with a difficult task of managing the delicate epistemological aspects
regarding the issue of what they know, what they would like to know and what
they would rather not like to know. In this context [no + more talk], realized as
No wiem. No (I) know., indexes an explicit reference to the superior epistemic
status of the patient relative to that of the doctor. However, as we have seen, the
patients qualification of the doctors prior contribution as already-known-information integrates two shades a competitive and a collaborative one and adds
them to the ongoing sequence. Finally, this fragment provides a vivid illustration
for one of Heritages (2012) observations, namely that the interpretation of the
clause cannot correctly be effectuated without a fine-grained grasp of epistemic
domains, and relative epistemic status within them.
6. Conclusion
This chapter has offered some insights into the interactional functions of the Polish particle no, which is an item that does not seem to have a clear and uniform semantic meaning on its own. The scope of the analysis presented in this chapter has
been limited to one specific environment that is turn-initial no that combines
with more talk in responsive turns. Due to space limitations I have only been able
to discuss three variations of the practice realized as no wanie, no oczywicie
and no wiem. My investigation of these and other no-prefaced turn-construction elements points toward a possibility that the main usage of these turn formats seems to be related to confirming the obviousness of the prior participants
contribution. However, these no-prefaced turns can also further exploit this confirmation to establish a referential link to the speakers state of knowledge, their
epistemic status. In this sense, [no + more talk] contributes an evaluation from
the speakers perspective, from the point where the speaker is at with reference to
his own state of knowledge. The three examples discussed here show instances of
the participants my side evaluations that contribute to collaborative outcomes.
However, my data illustrate that the Polish particle no can also do confirmation
in other sorts of local environments, where, depending on its placement and the
participants epistemic frames involved, it can add corroborative or competitive
nuance to the turns of which it is a part. As this chapter has demonstrated no
takes from the local (sequential) context and the participants respective epistemic frames involved and becomes the participants resource for managing their
socioepistemic relationships. In the context of doctor-patient interaction, where
patients and doctors co-construct the consultation based on who knows what and
who has rights to particular knowledge, no can be an important signal for how
they see and orient to this knowledge as the interaction unfolds. With this last
observation in mind, I would like to close this chapter with a suggestion that however semantically vacuous no may be at a first glance, its interactional function
proves to be far from being meaningless. It is in and through the detail of talk-ininteraction that we can see the unfolding of its meaningfulness. As such, then, this
chapter contributes new insights to how participants to interaction negotiate and
manage epistemic territories, by looking at one specific resource the particle no
and extending the scope of studies in this domain by yet another language.
Acknowledgements
The research, parts of which are presented in this paper, has been supported by a
scholarship from the Research Foundation Flanders, which the author gratefully
acknowledges.
My special thanks goes to John Heritage, whose advice and insight have significantly contributed to my thinking about the issues that this paper addresses.
. Given that this paper shows just one portion of a larger study, it leaves out entire areas of
findings concerning other instances of no in Polish. A brief mention of the fact that stand-alone
no does the same kind of job as the compounds discussed here do, will have to suffice for now.
164 Matylda Weidner
References
Atkinson, J. Maxwell and John Heritage. (eds.). 1984. Structures of Social Action: Studies in
Conversation Analysis. Cambridge: Cambridge University Press.
Bako, Mirosaw. (ed.). 2000. Inny sownik jzyka polskiego. Warszawa: PWN.
Bolden, Galina. 2006. Little words that matter: Discourse markers so and oh and the doing of
other attentiveness in social interaction. Journal of Communication 56 (4): 661688.
Bolden, Galina. 2008. So whats up?: Using the discourse marker so to launch conversational
business. Research on Language and Social Interaction 41 (3): 302337.
Bolden, Galina. 2009. Implementing incipient actions: The discourse marker so in English
conversation. Journal of Pragmatics 41 (5): 974998.
Bolden, Galina. 2010. Articulating the unsaid via and-prefaced formulations of others talk.
Discourse Studies 12 (1): 532.
Brown, Penelope and Stephen C. Levinson. 1987. Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press.
Clayman, Steven E. 2002. Sequence and solidarity. In Advances in Group Processes: Group Cohesion, Trust and Solidarity, Edward J. Lawler and Shane R. Thye (eds.), 229253. Oxford:
Elsevier Science.
Doroszewski, Witold. 1999. Sownik jzyka polskiego. Warszawa: PWN.
Drew, Paul. 1991. Asymmetries of knowledge in conversational interactions. In Asymmetries in Dialogue, Ivana Markova and Karl Foppa (eds.), 199212. Hemel Hempstead:
Harvester/Wheatsheaf.
Drew, Paul and John Heritage. (eds.). 1992. Talk at Work: Language Use in Institutional and
Work-Place Settings. Cambridge: Cambridge University Press.
Dunaj, Bogusaw. 1996. Sownik wspczesnego jzyka polskiego. Warszawa: Wilga.
Heritage, John. 1984. A change-of-state token and aspects of its sequential placement. In
Structures of Social Action, J. Maxwell Atkinson and John Heritage (eds.), 299345.
Heritage, John. 1984. Garfinkel and Ethnomethodology. Cambridge: Polity Press.
Heritage, John. 1998. Oh-prefaced responses to inquiry. Language in Society 27 (3): 291334.
Heritage, John. 2002. Oh-prefaced responses to assessments: A method of modifying agreement/disagreement. In The Language of Turn and Sequence, Cecilia E. Ford, Barbara A.
Fox and Sandra A. Thompson (eds.), 196224. Oxford: Oxford University Press.
Heritage, John. 2005. Revisiting authority in physician-patient interaction. In Diagnosis as
Cultural Practice, Dana Kovarsky and Judith F. Duchan (eds.), 83102. New York: Mouton
De Gruyter.
Heritage, John. 2011. Territories of Knowledge, Territories of Experience: Empathic Moments
in Interaction. In The Morality of Knowledge in Conversation, Tanya Stivers, Lorenza
Mondada and Jakob Steensig (eds.), 159183. Cambridge: Cambridge University Press.
Heritage, John. 2012a. The Epistemic Engine: Action Formation, Sequence Organization and
Territories of Knowledge. Research on Language and Social Interaction 45: 2550.
Heritage, John. 2012b. Epistemics in Action: Action Formation and Territories of Knowledge.
Research on Language and Social Interaction 45: 125.
Heritage, John and Geoffrey Raymond. 2005. The terms of agreement: Indexing epistemic
authority and subordination in assessment sequences. Social Psychology Quarterly 68 (1):
1538.
Heritage, John and Jeffrey Robinson. 2006. Accounting for the visit: giving reasons for seeking medical care. In Communication in Medical Care: Interactions between Primary Care
Physicians and Patients, John Heritage and Douglas W. Maynard, 4885. Cambridge:
Heritage, John and Marja-Leena Sorjonen. 1994. Constituting and maintaining activities
across sequences: and-prefacing as a feature of question design. Language in Society 23:
129.
Jefferson, Gail. 1983. Notes on some orderlinesses of overlap onset. In Discourse Analysis
and Natural Rhetoric, Valentina DUrso and Paolo Leonardi (eds.), 1131. Padua: Cleup
Editore.
Jefferson, Gail. 2004. A sketch of some orderly aspects of overlap in natural conversation.
In Conversation Analysis: Studies from the First Generation, Gene Lerner (ed.), 4359.
Amsterdam/Philadelphia: John Benjamins.
Keevallik, Leelo. 2010. Social action of syntactic reduplication. Journal of Pragmatics 42:
800824.
Kryk, Barbara. 1992. The pragmatics of interjections: The case of Polish no. Journal of Pragmatics 18 (23): 193207.
Lerner, Gene. 1996. On the semi-permeable character of grammatical units in conversation:
Conditional entry into the turn space of another speaker. In Interaction and Grammar,
Elinor Ochs, Emanuel A. Schegloff and Sandra A. Thompson (eds.), 238276. Cambridge:
Linde, Samuel Bogumi. 18071814. Sownik jzyka polskiego. Warszawa: Drukarnia XX,
Piiarw.
Perkyl, Anssi. 2002. Agency and Authority: Extended responses to diagnostic statements in
Primary care encounters. Research on Language and Social Interaction 35 (2): 219247.
Perkyl, Anssi. 2006. Communicating and responding to diagnosis. In Communication in
Medical Care: Interactions between Primary Care Physicians and Patients, John Heritage
and Douglas W. Maynard (eds.), 214247. Cambridge: Cambridge University Press.
Pisarkowa, Krystyna. 1975. Skadnia rozmowy telefonicznej. Prace Instytutu Jzyka Polskiego. Wrocaw: Zakad Narodowy Imiania Ossoliskich Wydawnictwo Polskiej Akademii
Nauk.
Pomerantz, Anita M. 1980. Telling my side: Limited access as a fishing device. Sociological
Inquiry 50: 186198.
Pomerantz, Anita M. 1984. Agreeing and disagreeing with assessments: Some features of preferred/dispreferred turn shapes. In Structures of Social Action, J. Maxwell Atkinson and
John Heritage (eds.), 57101. Cambridge: Cambridge University Press.
Raymond, Geoffrey and John Heritage. 2006. The epistemics of social relations: Owning
grandchildren. Language in Society 35: 677705.
Robinson, Jeffrey D. 2001. Closing medical encounters: Two Physician Practices and their Implications for the Expression of Patients Unstated Concerns. Social Science and Medicine
53 (5): 639656.
Sacks, Harvey. 1984. On doing being ordinary. In Structures of Social Action, J. Maxwell
Atkinson and John Heritage (eds.), 413429. Cambridge: Cambridge University Press.
Sacks, Harvey. 1987 [1973]. On the Preferences for Agreement and Contiguity in Sequences in
Conversation. In Talk and Social Organisation, Graham Button and John R. E. Lee (eds.),
5469. Clevedon, England: Multilingual Matters.
166 Matylda Weidner
Sacks, Harvey, Emanuel A. Schegloff and Gail Jefferson. 1974. A simplest systematics for the
Schegloff, Emanuel A. 1987. Recycled Turn Beginnings: A Precise Repair Mechanism in Conversations Turn-taking Organisation. In Talk and Social Organisation, Graham Button
and John R. E. Lee (eds.), 7085. Clevedon, England: Multilingual Matters.
Schegloff, Emanuel A. 1996. Turn organization: One intersection of grammar and interaction.
In Interaction and Grammar, Elinor Ochs, Emanuel A. Schegloff and Sandra A. Thompson
(eds.), 52133. Cambridge: Cambridge University Press.
Schegloff, Emanuel A. 2004. On dispensability. Research on Language and Social Interaction
37 (2): 95149.
Schegloff, Emanuel A. 2007. A Tutorial on Membership Categorization. Journal of Pragmatics
39: 462482.
Schegloff, Emanuel A. and Gene Lerner. 2009. Beginning to respond: Well-prefaced responses
to Wh-questions. Research on Language and Social Interaction (42): 91115.
Schegloff, Emanuel A. and Harvey Sacks. 1973. Opening up closings. Semiotica 8: 289327.
Sidnell, Jack. 2007. Look-prefaced turns in first and second position: launching, interceding
and redirecting action. Discourse Studies 9 (3): 387408.
Sidnell, Jack. 2010. Conversation Analysis: An Introduction. UK: Wiley-Blackwell.
Stivers, Tanya and John Heritage. 2001. Breaking the sequential mold: Answering more than
the question during comprehensive history taking. Text 21 (1/2): 151185.
Stivers, Tanya, Lorenza Mondada, and Jakob Steensig. 2011. The Morality of Knowledge in Conversation. Cambridge: Cambridge University Press.
Szulc-Brzozowska, Magdalena. 2010. Zur semantisch-pragmatischen Erweiterung der Abtnungsfuntion bei polnischen Modalpartikeln aus kontrastiver Sicht (DeutschPolnisch).
In Linguistik online.
Szymczak, Mirosaw. 19781981. Sownik jzyka polskiego. Warszawa: PWN.
West, Candace. 2006. Coordinating closings in primary care visits:producing continuity
of care. In Communication in Medical Care: Interactions between Primary Care Physicians and Patients, John Heritage and Douglas W. Maynard (eds.), 379415. Cambridge:
Wierzbicka, Anna. 1976. Particles and linguistic relativity. International Review of Slavic Linguistics 1 (23): 327367.
How can I lie if I am telling the truth?

The unbearable lightness of being of strong
and weak modals, modal adverbs and modal
particles in discourse between epistemic
modality and evidentiality
Peter Kosta
The major part of my contribution will concentrate on the close relation between epistemic modality and evidentiality and the notions of truth value, indirect speech acts and conversational implicature (cf. Kosta 2005; Kosta 2011b).
It is well attested in the literature that the epistemic modal adverb Russian
oevidno, Czech oividn, German offensichtlich, Italian ovviamente can have
different interpretation depending on the conversation situation, truth values
and scope relations (cf. Kosta 2011a; von Fintel and Gillies 2010; Kratzer 2010).
Even a bona fide epistemic modal can have two interpretations: a strong interpretation, which at least with necessity modals commits the speaker to
the truth of the proposition the modal scopes over (von Fintel and Gillies 2010),
and a weak interpretation, which is relativized to the content of some source
of information that may or may not be faithful to reality. In order to be able to
decide whether epistemic particles and modals are strong or weak we have to
differentiate between different sources of conversational backgrounds. Following
the findings in the research of notional category of modals in Kratzer (2010),
the proposed analysis of modals allows for one modal parameter to be fixed by
the context of use. It implies that that parameter is responsible for the variety of
interpretations modals can receive.
Keywords: epistemic modality, evidentiality, strong and weak modals,
conversational background
168 Peter Kosta
1. Introduction and prerequisites

It is well attested that meanings of particles, prepositions and adverbs are often
lexically ambiguous and, even in a given context of an utterance, they are not fully
disambiguated. The major part of my contribution will concentrate on the close
relation between epistemic modality and evidentiality and the notions of truth
value, indirect speech acts and conversational implicature (cf. Kosta 2005; Kosta
2011b). It is well attested in the literature that the epistemic modal adverb Russian oevidno, Czech oividn, German offensichtlich, Italian ovviamente can have
different interpretation depending on the conversation situation, truth values and
scope relations (cf. Kosta 2011a; von Fintel and Gillies 2010; Kratzer 2010). Even
a bona fide epistemic modal can have two interpretations: a strong interpretation, which at least with necessity modals commits the speaker to the truth of
the proposition the modal scopes over (von Fintel and Gillies 2010), and a weak
interpretation, which is relativized to the content of some source of information
that may or may not be faithful to reality. Those two types of interpretations have
also been identified in the recent literature on the connection between epistemic
modals and reportative evidentials (Izvorski 1997; Faller 2002; Matthewson et al.
2007; Rullmann et al. 2008).
In a situation where it is raining outside and speaker A is in a room, from
which he/she cannot watch or hear the rain, we must interpret the adverb obviously as an evidential adverb by indirect inference (cf. von Fintel and Gillies
2010). For instance, A observes that B enters the room with a wet umbrella and
dressed in a raincoat and says to B: obviously its raining outside. Contrary to
that, A wouldnt be able to utter the same sentence using a strong modal adverb
obviously or a modal must with the epistemic reading, if A observed the rain
watching from the window: A to B (observing the rain from the window by direct
evidence): *Obviously its raining, *it must be raining. The situation is the same
in all analyzed Slavonic languages and in German so that we can be sure that the
problem has a common linguistic (more likely formal semantic) explanation. As
a consequence, the expressions obviously, its raining outside, it must be raining
(which A utters while watching the pouring rain, looking outside the window)
are pragmatically marked violating the maxims of relation (inappropriate remark
in such a situation), quantity (too much and thus redundant information) and
modality (ambiguous information between epistemic and evidential reading). In
the same vein, one can use the epistemic adverb or the strong Modal Auxiliary in
the following interview situation: A says to B: Obviously, you want to challenge
me! Must you challenge me all the time? in a situation in which A is an adult,
and B is a naughty child, which wants to provoke A.
Strong and weak modals between evidentiality and epistemic modality 169
Contrary to this situation just introduced, the progressive form in present

cannot be used with the epistemic adverb (*Obviously, you are challenging me!),
and the progressive form without the strong modal even being stronger than the
utterance with a strong modal is not a licit utterance (*Are you challenging
me all the time?). It cannot, however, be the progressive form of the verb which
causes this problem (cf. Rothstein 2004) but the illicit combination of conversational backgrounds, epistemic modals and grammatical categories of the verb
(tense, mode and aspect) that influence the (un)grammaticality of the output. But
above all, the epistemic adverb obviously cannot be used for the prediction with
direct observation and the strong modal verb must has no provision for a strong
commitment.
In the literature on epistemic modality and evidentiality there is a broad consensus that these two categories are semantically close connected. Some linguists
actually think that evidentiality is a subcategory of epistemic modality (Kratzer
2010). The other groups of researchers try to draw a clearer line between these
two categories (see Wiemer 2008; Kosta 2011a; Mendoza 2008). As to spoken
conversation in Russian, the category of evidentiality has been analyzed to my
knowledge only by Marion Krause wit respect to the correlation between prosody,
the degree of evidentiality and emotionality. In her monograph she concentrates
above all on the interaction in conversation between epistemic particles and
prosody (cf. Krause 2007) without differentiating strictly between evidentiality
and epistemic modality. With regard to the interaction between lexical items and
prosody, the author was able to show that epistemic and emotional evaluations
can differ, essentially by means of prosodic shape of utterances (ibid.:114). This
is remarkable particularly because the rational and emotional components often
co-occur in one and the same item, since they are often derived from the same
reservoir of prosodic markers (ibid.:130, 140). It seems also to be the case that
prosodic contours have a certain influence on the concrete manifestations associated with epistemic strength (ibid.:115). And prosodic similarities are greatest for those statements, where epistemic strength moves either to the lower or
upper range of the scale (ibid.:126). There is no objection against the splendid
empirical analysis of the work, but, as already mentioned, the lack of differentiation between epistemic and evidential reference assessment. Moreover, in order
to be able to decide whether epistemic particles and modals are strong or weak
we have to differentiate between different sources of conversational backgrounds.
Following the findings in the research of notional category of modals in Kratzer
(2010), the proposed analysis of modals allows for one modal parameter to be
fixed by the context of use. It implies that that parameter is responsible for the
variety of interpretations modals can receive. One of the conclusions of Angelika
Kratzers observation in the article What must and can must and can mean that
170 Peter Kosta
was carried over to The Notional Category of Modality was that the interpretation
of modals is relative to a conversational background that might be made explicit by
adverbial phrases and/or particles of various kinds.
2. Basic notions
According to Kratzer (2012:21ff.) there are important differences between different adverbial phrases contributing conversational backgrounds for modals. This
is illustrated by (1a) versus (1b):
(1) a. Given the article in the Hampshire Gazette, Mary Clare Higgins must have
been reelected.
b. According to the article in the Hampshire Gazette, Mary Clare Higgins
must have been reelected.
(Kratzer 2012:21)
An assertion of (1a) would commit one to the truth of what the article says, and
continuing with (2) would be infelicitous:
(2) but I wouldnt be surprised if she wasnt. The Gazette is usually too quick
to draw conclusions from projected election results.
Kratzer (2012:ibid.) explains the contrast in stating that an assertion of (1b)

would not commit me to the truth of the report in the Gazette, and () could
continue with (2) without contradicting oneself (Kratzer 2012:ibid.). The difference between (1a) and (1b) she ascribes to two different ways of interpreting what are commonly called epistemic modals. This means that in (1b), the
accessible worlds are the worlds compatible with the content of the report. In
(1a), the accessible worlds are worlds that have counterparts of the article in the
Hampshire Gazette that have the same content as the actual article and came into
existence in the same way (Kratzer 2012:ibid.). As a consequence, following
Kratzer, one shouldnt assert (1a) unless one believes the evidence for the Gazette
report to be highly reliable. (1a) and (1b) show that even a bona fide epistemic
modal can have two interpretations: a strong interpretation, which at least with
necessity modals commits the speaker to the truth of the proposition the modal
scopes over (von Fintel and Gillies 2010), and a weak interpretation, which is
relativized to the content of some source of information that may or may not be
faithful to reality (Kratzer:ibid.).
Those two types of interpretations have also been identified in the recent lite
rature on the connection between epistemic modals and reportative evidentials
(Izvorski 1997; Faller 2002; Davis et al. 2007; Rullmann et al. 2008). For example,
Rullmann et al. (2008) construe the modal alternatives for the Sttimcets reportative modal ku7 as the set of worlds where a relevant report was made, rather
than the set of worlds where the content of such a report is true. The result is a
given the report, rather than an according to the report, interpretation, and ku7
comes out as a strong epistemic modal that doesnt allow the speaker to distance
herself from the content of the report. Sttimcets ku7 thus contrasts with the
German reportative sollen, which relies on alternatives where the content of the
relevant report is true, hence is weak.
2.1
Evidentiality
In linguistics, evidentiality is, generally, an indication of the existence and/or nature of the evidence, or the type of evidence in support of a given assertion. A
marker element is being considered a grammatical evidential if its expression is
part of the grammatical system of a given language. Even though all languages
have means to specify the source or the reliability of information, it is not the case
that all languages express them grammatically. The grammar can be expressed
in different forms (depending on a given language), such as affixes, clitics, weak
modals, auxiliaries or particles.
On the other hand, some languages express evidentiality only as part of
other grammatical meaning: The so-called out-of-control circumfix ka--a in
Sttimcets (Lillooet Salish) expresses an initially puzzling cluster of meanings,
including be able to, manage to, suddenly, accidentally, and non-controllable.
These are illustrated in (3ae) respectively taken from Davis, Matthewson and
Rullmann (2009):
(3) a.

b.

c.

be able to
w7=lhkalh=tu7 ka-ns-a ekw7na St=a snmulh2
IMPF=1PL.SUBJ=ADD KA-go-A right.over.there Lillooet=EXIS us
We can go to Lillooet by ourselves.
manage to
ka-qmt-s=kan-a ta=w7 tsq-n-an
KA-hit-CAUS=1SG.SUBJ-A DET=IMPF throw-DIR-1SG.ERG
I managed to hit the target.
suddenly
ka-lhxw-a ta=n-sqx7=a
KA-appear-A DET=1SG.POSS-dog=EXIS
My dog appeared suddenly.
172 Peter Kosta
d.

e.

accidentally
ka-kws-ts=kan-a
KA-fall-CAUS=1SG.SUBJ-A
I accidentally dropped it.
non-controllable
ka-ctq-a ta=tnamten=a
KA-rise-A DET=moon=EXIS
The moon rose.
According to standard formal semantic analyses of Indo-European systems,

modals are quantifiers over possible worlds whose quantificational strength is
lexically specified as e.g., universal or existential, while differences between epistemic, deontic and other modal interpretations are derived from implicit conversational backgrounds, rather than from lexical ambiguity (Kratzer 2010:26).
However, as Davis, Matthewson and Rullmann (2009:207) have previously
shown, the lexical specification of Sttimcets modals is the inverse of the standard model: differences in modal conversational backgrounds are lexically specified (as e.g., epistemic or deontic) but quantificational strength is not leading to
quantificational variability between existential and universal readings.
Cross-linguistically, the invariant job of an evidential is to classify evidence
for what is being said as direct or indirect (Kosta 2011a). There is, however, another well-known fact about epistemic must intertwined with the weakness claim
in much of the literature: must carries an evidential signal, in particular it signals
that the speaker has reached her conclusion via an indirect inference.
Billy is a weather enthusiast. Looking out the window seeing pouring rain,
she can report by asserting (4a) but not (4b).
(4) [Seeing the pouring rain]
a. Its raining.
b. ??It must be raining.
If, instead, she sees people coming in from outside with wet umbrellas, slickers,
and galoshes, then even if she knows that rain is the only explanation she can
report with either the modalized claim or its bare prejacent. Either will do:
(5) [Seeing wet rain gear and knowing rain is the only possible cause]
a. Its raining.
b. It must be raining.
This is not the place for us to survey the field of evidentiality studies in any
amount of detail (some relevant references are Aikhenvald 2004; Davis, Potts and
Speas 2007; Faller 2002; de Haan 2001; Matthewson, Rullmann and Davis 2007;
Types of sources of information
Direct
Indirect
Epistemic modals
Attested
Reported
Visual
Auditory Other sensory

Secondhand
Thirdhand
Inference
Results
Folklore
Reasoning
Figure 1. Willets taxonomy of evidential categories
McCready and Ogata 2007; Rooryck 2001a, b; Willett 1988). Suffice it to say that
evidential markers are expressions found in many languages that signal the source
of evidence a speaker has for the prejacent claim. Evidentials systems across languages is given in Figure 1 from Willett (1988). Von Fintel and Gillies (2010) have
enriched the graphic with an indication of where the evidential signal carried by
epistemic modals fits in.
It appears that seen as evidentials, epistemic modals are markers of indirect
inference, which is the rightmost branch of Willets system. It should be noted
that the literature on evidentials often makes a strict distinction between epistemic modality and evidentiality, but the facts we discuss here indicate that this
is too simplistic a position. We therefore try to modify our recent opinion on this
subject (cf. Kosta 2011a).
Direct evidence may come from direct perception or 1st person experiences,
like skin itching or headaches. Indirect evidence may come from reports, rumors,
legends, or inferences drawn from direct or indirect evidence. The cross-linguistically invariant job of an epistemic modal is not to classify evidence, but to assess
the truth of a proposition against a range of possibilities projected from a body of
evidence. There are two distinct semantic jobs to be done, then: classify evidence,
versus assess the truth of a proposition against possibilities projected from a body
of evidence. The two jobs often end up being carried by a single portemanteau item
that might then be arbitrarily catalogued as modal or evidential. That evidential
meaning components are in principle independent from modal meaning components, but can be bundled together with other meaning components in a single
lexical item, was emphasized in Izvorski (1997). Izvorski points out that with finite
verbs in the present tense, the Turkish perfect morpheme m is interpreted as an
indirect evidential. In non-finite environments and with future or past tense, m
only has a perfect, non-evidential, meaning. The evidential meaning component
174 Peter Kosta
cant be contributed by m itself, then, but seems to be a separate component

spelled out in a portemanteau with the present tense. In Quechua and Korean,
too, evidential meaning components can be attached to items that are commonly
categorized as tenses (Faller 2002).
In my article on the relation between epistemic modality and evidentiality, I
tried to show how epistemic modality and evidentiality interact in the languages
of the world (cf. Kosta 2011a). It is fair to say that the overwhelming majority
of scholars who have studied the relationship between epistemic modality and
evidentiality stated there a very close connection between the two categories.
However, this relation is usually assumed without much comment, and little discussion is devoted to the differences between these two categories in conversation
with real empirical data base.
A good example to start with is an influential book on mood and modality (see Palmer 1986). In this book, Palmer leaves no doubt about his position,
namely that evidentiality is part of the epistemic modality. According to him,
both conceptual categories are concerned with the level of commitment on the
part of the speakers statement to the facts expressed in the utterance. This means
that Palmer considers evidentiality not as a special class. There are at least four
ways in which a speaker may indicate that he does not evaluate what he says as a
fact, but rather:
i.
ii.
iii.
iv.
he speculates
he presents it as a deduction
that someone else has told him about this (hear-say)
that this is only a matter of appearance, based on evidence of some other perception (which may be fallible)
All four were concerned with the indication by the speakers (lack of) commitment to the truth of the proposition being expressed (Palmer 1986:51). Based on
Palmers own type definitions only (i) has to do with the epistemic modality pure
while (ii)(iv) has to do with evidentiality, namely inference, rumor and sensory
evidence. Palmer makes this claim even more explicit: It would be a futile exercise to try to decide whether a particular system (or even a term in a system in
some cases) is evidential rather than being based on facts and truth conditions
(1986:70). A similar position is taken by Frajzyngier (1985, 1987:211; see also
Palmer 1986). He also sees a direct correspondence between evidentiality and
epistemic modality: () it seems quite obvious that different ways of acquiring
knowledge are at different levels of certainty of the truth of the proposition ()
(Frajzyngier 1985:250). In this view, there is a link between the interpretation of
Visual <
auditory
<
no visual
Direct evidence
More believable
<
inference
<
<
indirect evidence
quotative
less believable
Figure 2. The hierarchy of evidentiality
direct and indirect evidence provided that the truth value of the proposition is in
play. I show this schematically in Figure 2.
Most linguists assume that there is a universal inventory of grammatical and/
or lexical categories in all natural languages, but each language encounters a different choice. Nearly a quarter of all world languages have means to express the
source on which a statement is based, and in these languages evidentiality must
be expressed obligatory (Aikhenvald 2004:1).
2.2
The case of must and vrode by
As a number of authors have pointed out, the English epistemic modal must has
evidential characteristics, too (von Fintel and Gillies 2010). Epistemic must excludes direct perceptual or irreducibly 1st person evidence, for example, as illustrated by (6) and (7):
(6) a. # Your nose must be dripping. I can see it.
b. You must have a cold. Your nose is dripping.
(7) a. # I must have a terrible headache. I feel lousy.
b. The baby must have a terrible headache. He is screaming and pressing his
hands against his temples.
English spells out evidential, modal, and temporal meaning components together
as the single lexical item must, resulting in what we call a present tense epistemic
modal.
We take the theory of Kratzer (2010) and von Fintel and Gillies (2010) for
granted and differentiate between two classes of particles in Russian. The first
class includes particles such as vrode, vrode by, kak budto, kaetsja, mol, deska,
-de, jakoby, slovno that correspond to weak modals (such as German sollen) and
strong evidentials (reportatives, quotatives or inferential). The second class of particles and adverbs corresponds to the meaning of strong modals (such as German
mssen). The latter class of evidentials is represented with adverbs and particles
of high degree of epistemic modality such as oevidno, koneno, bez somnenija,
176 Peter Kosta
po-vidimomu, kak soobaetsja, logino, estestvenno, potomu, dolno by, sledo

vatelno, nesomenno, po soobenijam, that are based on indirect inference resulting
from reasoning (by logical conclusions, results and conversational implicatures),
from reports and hear-say.
We will try to demonstrate how particles that show the portmanteau effect
between evidentiality and epistemic modality interact and are used in conversations using for Russian the ORD and other data of the spoken National corpus
of Russian. The particle vrode and vrode by can be used to mark the evidential
meaning in a context such as (8) where the speaker does not remember exactly
the situation how to get to somewhere but tries to mark the evidence that he was
already here before and that he now tries to remember the way how to get there:
In the following conversation (8) between the doctor (I44) and the grandmother
(1) of a three year old boy the doctor uses the particle vrode by in a conversational situation where the weak epistemic reading of the particle vrode marks a
lower degree of certainty about situations and facts, while the evidential reading
by direct perception is well attested. The weak modal meaning marks not only a
lower meaning of certainty (supported by the conditional by byl and the sequence
with specific indefinites kak-to) but is accompanied by the deictic meaning of
direct perception of the speaker (supported by the explicitly expressed performative verb ja pomnju I remember). In German one could try to translate it with
the indefinites irgendwie, irgendwo or with epistemic particles and epistemic
phrases wohl, ich vermute, es scheint, es soll P or even angeblich but these
are, of course, not real equivalents. In the faded meaning they correspond to the
Czech particle jako and the English particle like in spoken discourse.
(8) I44 ugu // nikto ne boleet ? da ? 1 net // nikto u nas ne boleet // *P 1 roditeli
rabotajut // I44 ugu // *P 1 slava bogu ! *S v vyxodnye sideli doma // I44
xoroo // *V I44 tak voobe () asto prostuaetsja / () Gria ? 1 da net net
net net / net // I44 m-m / ugu // 1 vot // dae po listam vot // tam uvidite //
I44 o ! skolko vy nakopili ! a ja u vas kak-to byl // 1 * N / nu tak to samoe/
to za skolko ? za tri goda // @I44 nu ponjatno // *P 1 za tri goda // I44
ja u vas kak-to byl // ja pomnju / to () vrode by byl / i nedaleko to bylo
ot () ploadi Pobedy // *V a tut vot poneslo menja / (e-e) // 1 a kuda vas
poneslo? (ORD S44-02)
I44 well // is no one ill? Yes? 1 no // nobody is ill here // *P(ause) 1 the
parents work // I44 well // P(ause) Thank God ! at weekends they stayed at
home // I44 right // *V I44 generally () does he usually take cold / () Gria ?
1 oh no no no / no // I44 hmmm-m / well // 1 you see // even in the documents there // youll see it // I44 Jeez, how much you collected there! And I
visited you once (I believe) // 1 * N / well, this was so to say / this must have
been ? three years ago // @I44 well, I understand // *P 1 in three years //

I44 I was somehow at your (house) // I remember / that () like I was / and
not far away this was from () Victory Square // *V and it brought me here
from there/ (e-e) // 1 and where bring it you ? (One Conversational Day
S44-02)
The evidential meaning of hear-say of this particle is attested in the following

example from the National corpus of Russian language:
(9) [Do, en, 1967] Da-da-da // [Mat, en, 1944] Pogoda meala / potomu to
dod so snegom // [Do, en, 1967] Aa // [Mat, en, 1944] I sneg ue pol //
S severa polo to vot // U nas gde-to vot-vot nebos // Vrode by do oktjabrja
obeajut /tot / [Do, en, 1967] Teplo // [Mat, en, 1944] Da/ zamorozki
v oktjabre // [Do, en, 1967] [obraaetsja k koke] A to ty / Monja/
perevernulas?
[The daughter, female, 1967] yeah-yeah-yeah // [mother, female, 1944] The
weather interfered with / because of rain with snow / [daughter, female, 1967]
Aa // [mother, female, 1944] And the snow is already gone / From the north
it has gone now // We have somewhere is about I suppose / It seems that
until October they promise / this / [A daughter, wife, 1967] Heat // [mother,
wives, 1944 ] Yes / frosts in October // [The daughter, wife, 1967] [refers to
the cat] And what do you / Monja / you capsized?
The particle vrode by has the same portmanteau function as in (8), namely as
evidential, pointing at the source of information (weather forecast), taking simultaneously the relatively low degree of probability what the weather will actually be
like on the aforementioned date of October based on promises more than prognoses of the weather forecast.
In the following example (10), the epistemic-evidential marker vrode by is
used in a more opaque or even vague meaning it seems, cf.
(10) I44 da // *P vy polzuetes vsjakimi / (e) vot rastvorikami ? akvamaris$ /
solin$/ *P fiziomer$ / nu vot ti solevye tuki ?
1 net // *P net / a a net // nado vot () napisat Aneke // I44 ugu //
1 Ane / tob ona / to / () delala // moet u nas tut tolko vot tri es // I44
(m-m) 1 potomu to i vrode by / to-to
I44 yes // *P Do you use every (e) kind, well, of chemical solutions?
Aquamarin$ / solin$ / *P fiziomer$ / well these salt stuff? solin$ /
1 no // *P no / well no no // one should well () prescribe to Anna // I44
yes // 1 to Anna in order that she / would do it // maybe we have only three
of it // I144 (hmm-hmm) 1 because it seems to me / that
178 Peter Kosta
Besides the modal epistemic function that marks a low degree of certainty, the
particle vrode often occurs in discourse at the beginning of a new turn thus fulfilling a quasi metalinguistic function of summing up the content of previous
utterances or speech acts: we can call this function a metalinguistic summative
function of vrode. In example (11) it can be translated with the English colloquial
particle like in the same discourse function:
(11) I1 a / to est to / vs ta e kista / kak by vot to to bylo //
1 vrode vs sdelano / i vs tak neploxo //
I1 and / thats that / all the same cyst / as if it was / what it was
1 like everything is done / and all so good / /
In the evidential function, the particle vrode can be used in long-distance calls
between the doctor and the patient, in which the doctor wants to have information about the condition of the patient. The informant can use the particle vrode
kak, if he/she is not sure or if he/she cannot give precise information about the
symptoms of the patient, either because this information is not accessible to the
direct observation or because the informant is not sure about his/her own competence or his/her own judgment or his/her own testimony. In using this particle
vrode kak he/she is distancing herself from the content of the proposition (including the truth conditions). These conversational backgrounds I call informational
conversational backgrounds as opposed to realistic conversational background. A
detailed notion of informational conversational background as opposed to socalled realistic conversational background is given in Kratzer (2012:33).
Informational backgrounds are not the only backgrounds that are not necessarily realistic. Other not necessarily realistic backgrounds may relate to norms of
various kinds, and among those, backgrounds representing the normal course of
events in the world of evaluation play a privileged role:
(12) [M1, mu] najz nol odin / da? [I40, mu, 41] najz nol odin / da / ritromecin
dvesti pjat desjat // [M1, mu] najz dvesti pjat desjat // aga / vot / a po malenkoj to ja mogu skazat ? vot tak vot vizualno (-) na rukax na nogax syp
vrode kak ona ( ) ne vidna / i vot na ee / (-) vidno na ekax [I40, mu,
41] ugu // [Telefonnyj razgovor o bolezni rebenka // Iz materialov korpusa
Odin reevoj den, podgotovlennogo gruppoj A. S. Asinovskogo, 2009]
[M1, male] Nise zero one / yes? [I40, male, 41] Nise zero one / yes / Erythro
mycin two hundred and fifty // [M1, male] Nise two hundred and fifty //
yeah / here / and how about the little one what can I say? Here and so visually
. In this respect the particle fulfils a similar metadiscursive function as the Czech particle
jako and the English particle like in spoken dialogues, cf. Kosta (1998).
(hmm) at the hands and on her legs a kind of like her () not visible/ and here
on the neck / (hmm) can be seen on the cheeks [I40, male, 41] Uh-huh//
(13) [I41, mu, 63] ugu // [1, en] otodvin moju zanavesoku // [I41, mu, 63]
da / ary ne [1, en] [nrzb] vrode ne oen arko // [I41, mu, 63] ary
ne budet // e voobe ne budet segodnja // raz takoe utro to ue ne razogreet // budet umerennoe takoe // [1, en] a to utro / ploxoe to li // [I41,
mu, 63] net / ono n samo po sebe niego / no ono uee takoe / t ii /
von / vidimost / ne million na million // (m) nu / takoe budet // oblanoe //
[Razgovor za zavtrakom s enoj // Iz materialov korpusa Odin reevoj den,
podgotovlennogo gruppoj A. S. Asinovskogo, 2009]
[I41, male, 63] Uh-huh // [1, female ] open my curtains // [I41, male, 63]
yes / no heat there [1, female] [nrzb] did not seem very hot // [I41, Male,
63], heat will not / will not do it today // once we have such a morning it
will not warm up anymore // will moderate this // [1, Female] and that the
morning / Is that bad // [I41, Male, 63] no / its n in itself itis nothing/
but it is already like this / thee se / you see / visibility / not a million by
million// well / it will be like // cloudy // [The conversation at breakfast with
his wife// Out of body materials of the One Voice Day, prepared by a group
of A.S. Asinovsky, 2009]
In this example we can see that the evidential particle vrode, vrode by can introduce affirmative positive clauses in which the degree of epistemic probability is
very low but it can also introduce clauses in which the negation has narrow scope
only over the predicate (namely hot) whereas the proposition is in the wide scope
of epistemic modality.
The logical notification for (13) would be (13):
(13) x[W(x) & y[Modeepist W(y) y = x] & ~B(x)]
We can see that epistemic modality can be out-scoped by negation so that the
truth values can still hold (W = weather, x = individual variable, = existential
operator, = allquantor, B = hot, the proposition means: if there is weather x, and
for all situations y it is true that it is probable that the weather x being in the situation y is not hot; about truth values and negation cf. also Kosta, Ms.).
The next function we will identify now for the particle vrode by is the function
of relative consent in minimal dialogues in adjacency pairs of yes-no responses.
I call this consent relative because it is cancellable (or at least defeasible) as the
knowledge of the respondent corresponds to a kind of epistemic agnostic stance,
thus the respondent does not take over responsibility for the truth of P. Here the
conversational frame entails evaluative speech acts on a person and the particle
meets the pragmatic function of a non-binding agreement. Because evaluative
180 Peter Kosta
speech acts are not capable of truth values (see Kosta 1993; von Fintel and Gillies
2010), the answer cannot be sued, contradicted or sanctioned by the questioner.
(14) [Programmnyj direktor, Mihail Kozyrev, mu, 41, 1967] V rubke [Kapitan,
Fedor Dobronravov, mu, 47, 1961] Nu / ego? Princessa cirka / ta dressirov
ica. Vrode niego / a? [Matros Petrovi, Aleksej Xardikov, mu, 40, 1968]
Nu / tak. [Kapitan, Fedor Dobronravov, mu, 47, 1961] Nu i ty / k nej?
[Matros Petrovi, Aleksej Hardikov, mu, 40, 1968] Nu vrode / da. [Dmitrij
Djaenko i dr. Den radio, k/f (2008)]
[Program Director, Mikhail Kozyrev, Male, 41, 1967] [Captain, Fedor
Dobronravov, Male, 47, 1961] Well/ What? The Circus Princess / this trainer.
Like nothing / a? [Sailor Petrovich, Aleksey Khardikov, Male, 40, 1968] So /
so. [Captain, Fedor Dobronravov, Male, 47, 1961] Well, you mean what / for
her? [Sailor Petrovich, Aleksey Khardikov, Male, 40, 1968] Well like / yes.
[Dmitrij Djaenko et al. Radio Day (2008)]
To resume on the semantic and pragmatic function of the modal particle

vrode by one can postulate the following conversational implicature (cf. Kosta
2011b:66f.):
vrode by
a. I want to say what someone else says (= reportative component)
b. I say yes to P
c. I dont say I know that P is true (epistemic component, agnostic
stance)
d. I think that other people can think the same
d. leads to e.
e. I think that P might ~ be true (epistemic, cancellable)
( 14)

3. Modal strength and conversational backgrounds

We tried to demonstrate how a specific particle can contribute to our knowledge
about its evidential and epistemic meaning if we analyze it in discourse including different types of conversational background. Our major hypothesis was that
evidentials and epistemic particles are highly ambiguous from the point of view
of their lexical semantics. Their discourse functions can, however, differ and they
can even contribute to our theoretical findings of modality. We analyzed the relation between epistemic modality and evidentiality taking a single particle vrode as
a kind portmanteau word between epistemic and evidential meaning.
Some epistemic modal expressions are stronger than others. By asserting any
expression in the list (15)(17), a speaker stakes herself to a stronger claim than
he/she would by asserting the next in the list:
(15)

a.
b.
c.
d.
e.
Dolno byt, idet dod.

Navernjaka, dod idet.
Po-vidimomu, idet dod.
Moet byt, idet dod.
Dod idet.
(16)

a.
b.
c.
d.
e.
Asi pr.
Je tm jist, e pr.
Pravdpodobn pr.
Mohlo by pret.
Pr.
(17)

a.
b.
c.
d.
e.
It must be raining.
It is almost certainly raining.
It, presumably, is raining.
It might be raining.
It is raining.
That much is clear. But what about the relation between one of these modalized
claims and the bare prejacent it is raining? Where does that belong on the list?
Is it stronger or weaker than strong epistemic necessity modals like must in
English? That is less clear and is not anymore the focus of my contribution. The
basic quantificational treatment of modality makes a clear prediction. Since must
is Englishs way of representing an operator of epistemic necessity, it quantifies
over possibilities compatible with what is known, saying that all of them are possibilities in which the prejacent is true. Whatever is known is true, and so the
actual situation is always among those compatible with what is known. Thus, for
any , must entails . The strong modalized claim is stronger than the bare prejacent. Following von Fintel and Gillies (2010) we claim that epistemic meaning
of must is stronger than is usually assumed in the literature due to its evidential
potential based on different conversational backgrounds. In order to be able to
decide whether epistemic particles and modals are strong or weak we have to
differentiate between different sources of conversational backgrounds. Following
the findings in the research of notional category of modals in Kratzer (2010), the
proposed analysis of modals allows for one modal parameter to be fixed by the
context of use. It implies that that parameter is responsible for the variety of interpretations modals can receive.
182 Peter Kosta
References
Aikhenvald, Alexandra. 2004. Evidentiality (Oxford Linguistics). Oxford: Oxford University
Press.
Asinovsky, Alexander S. et al. 2009. The ORD Speech Corpus of Russian Everyday Communication One Speakers Day: Creation and Annotation. In Text, Speech and Dialogue.
TSD-2009 [LNCS/LNAI 5729], Vclav Matouek and Pavel Mautner (eds.), 250257.
Berlin/Heidelberg: Springer.
Davis, Henry, Matthewson, Lisa and Rullmann, Hotze. 2009. Out Of Control Marking as
Circumstantial Modality in Sttimcets. In Cross-linguistic Semantics of Tense, Aspect,
and Modality, Lotte Hogeweg, Helen de Hoop and Andrey Malchukov (eds.), 205244.
Davis, Henry, Matthewson, Lisa and Rullmann, Hotze. 2007. A Unified Modal Semantics for
Out-of-Control Marking in Sttimcets. In Papers for the 42nd International Conference
on Salish and Neighbouring Languages [University of British Columbia Working Papers in
Linguistics 20], Kristn Jhannsdttir and Martin Oberg (eds.), 119160.
Davis, Christopher, Potts, Christopher and Speas, Margaret. 2007. The pragmatic values of evidential sentences. In Proceedings of Semantics and Linguistic Theory 17, Masayuki Gibson
and Tova Friedman (eds.), 7188.
De Haan, Ferdinand. 2001. The relation between modality and evidentiality. Linguistische
Berichte, Sonderheft 2001: 201216.
Faller, Martina T. 2002. Semantics and pragmatics of evidentials in Cuzco Quechua. PhD thesis
Stanford University.
von Fintel, Kai and Gillies, Anthony S. 2010. Must. . . stay. . . strong! Natural Language Semantics 18 (4): 351383.
Frajzyngier, Zygmunt. 1985. Truth and the indicative sentence. Studies in Language 9 (2):
243254.
Frajzyngier, Zygmunt. 1987. Truth and the compositionality principle: A reply to Palmer.
Studies in Language 11 (1): 211217.
Gazdar, Gerald. 1979. Pragmatics: Implicature, Presupposition, and Logical Form. New York:
Academic Press.
Hansen, Bjrn and de Haan, Ferdinand. (eds). 2009. Modals in the Languages of Europe. A Reference Work. [Empiricl Approaches to Language Typology 44]. Berlin/New York: Mouton
de Gruyter.
Hansen, Bjrn and Karlk, Petr. (eds). 2005. Modality in Slavonic Languages. New perspectives
[Slavolinguistica 6.]. Mnchen: Verlag Otto Sagner.
Izvorski, Roumyana. 1997. The present perfect as an apistemic modal. In Proceedings from
SALT VII, Stanford CA, CLS, 222239.
Kosta, Peter. 1993. Bewertung und Konnotation in Milan Kunderas Werk als axiologisches
und translationslinguistisches Problem. In Slavistische Studien zum 11. Internationalen
Slavistenkongre, Preburg 1993, Karl Gutschmidt et al. (eds.), 247271. Kln: Bhlau.
Kosta, Peter. 1998. Argumentation, Persuasion und der Turn-taking-Mechanismus. In Dialoganalyse VI: Referate der 6. Arbeitstagung Prag 1996 [Beitrage zur Dialogforschung],
Svtla mejrkov, Jana Hoffmanov, Olga Mllerov and Jindra Svtl (eds.), 115131.
Tbingen: Niemeyer.
Kosta, Peter. 1998. berlegungen zum phatischen Diskurs (und zur phatischen Partikel
jako) in der Konversation tschechischer Intellektuelle. In Slavische Sprachwissenschaft
und Interdisziplinaritt Nr. 4: Olexa Horbatsch zum Gedenken, Gerd Freidhof et al. (eds.),
109145. Mnchen: Sagner.
Kosta, Peter. 2005. Direkte und indirekte Direktiva als Strategien des (Miss-)Verstehens in
Dialogen tschechischer Frauen und Mnner. In Mediale Welten in Tschechien nach 1989:
Genderprojektionen und Codes des Plebejismus [Specimina Philologiae Slavicae, 142], Jiina
van Leeuwen-Turnovcova and Nicole Richter (eds.), 191198. Mnchen: Otto Sagner.
Kosta, Peter and Thielemann, Nadine. 2009. Gesprchsanalyse. In Slavic Languages. Slavische
Sprachen. An International Handbook of their Structure, their History and their Investigation. Ein internationales Handbuch ihrer Struktur, ihrer Geschichte und ihrer Erforschung
[Handbcher zur Sprach- und Kommunikationsforschung HSK 32.1], Sebastian Kempgen
et al. (eds.), 10291047. Berlin/New York: Mouton de Gruyter.
Kosta, Peter. 2011a. Modalit Epistmique et Evidentialit et sa disposition la base dictique.
In Sprachkontakte, Sprachvariation und Sprachwandel. Festschrift fr Thomas Stehl zum 60.
Geburtstag, Claudia Schlaak and Lena Busse (eds.), 257283.Tbingen: Narr.
Kosta, Peter. 2011b. Konversationelle Implikaturen und indirekte Sprechakte auf dem Prfstein. In Die Sprache in Aktion. Pragmatik. Sprechakte. Diskurs. Language in Action. Pragmatics. Speech Acts. Discourse, Michail I. Kotin and Elizaveta G. Kotorova (eds.), 5569.
Heidelberg: Universittsverlag Winter.
Kosta, Peter. 2013. Third Factor Relevance between Semantics, Pragmatic and Syntax. In
Academia.edu. http://www.academia.edu/3207898/Third_Factor_Relevance_between_
Semantics_Pragmatics_and_Syntax.
Kratzer, Angelika. 2010. The Notional Category of Modality. In Papers on Modals and Conditionals. Version December 2010.
Kratzer, Angelika. 2012. Modals and Conditionals. Oxford: Oxford University Press. (Oxford
Studies in Theoretical Linguistics).
Krause, Marion. 2007. Epistemische Modalitt: Zur Interaktion lexikalischer und prosodischer
Marker (am Material des Russischen und Bosnisch-Kroatisch-Serbischen). Wiesbaden:
Harrassowitz Verlag.
Matthewson, Lisa, Davis, Henry and Rullmann, Hotze. 2007. Evidentials as Epistemic Modals:
Evidence from Sttimcets. Linguistic Variation Yearbook 7: 201254.
McCready, Eric and Ogata, Norry. 2007. Evidentiality, modality and probability. Linguistics
and Philosophy 30 (2):147206.
Mendoza, Imke. 2008. Evidentialitt, Modalitt, propositionale Einstellungen und (nicht-)
spezifische Referenz. In Slavistische Linguistik 2006/2007. Beitrge zum XXXII. / XXXIII.
Konstanzer Slavistischen Arbeitstreffen in Boldern und Potsdam (03.09.06.09.2007), Peter
Kosta und Daniel Weiss (eds.), 319338. Mnchen: Sagner.
Palmer, F. R. 1986. Mood and modality. Cambridge: Cambridge University Press.
Rothstein, Susan. 2004. Structuring events: A study in the semantics of lexical aspect. Oxford:
Blackwell Publishing.
Rullmann, Hotze, Matthewson, Lisa and Davis, Henry. 2008. Modals as Distributive Indefinites. Natural Language Semantics 16: 317357.
Rooryck, Johan. 2001a. Evidentiality, part 1. GLOT International 5: 125133.
Rooryck, Johan. 2001b. Evidentiality, part 2. GLOT International 5: 161168.
184 Peter Kosta
Wiemer, Bjrn. 2006. Particles, parentheticals, conjunctions and prepositions as evidentiality

markers in contemporary Polish (A first exploratory study). Studies in Polish Linguistics
3: 567.
Wiemer, Bjrn. 2008. Lexikalische Markierungen evidenzieller Funktionen: zur Theoriebildung und empirischen Erforschung im Slavischen. In Lexikalische Evidenzialittsmarker im Slavischen [Wiener Slawistischer Almanach, Sonderband 72], Bjrn Wiemer and
Vladimir Plungjan (eds.), 549. Mnchen: Sagner.
Willett, Thomas. 1988. A cross-linguistic survey of the grammaticalization of evidentiality.
Studies in Language 12 (1): 5197.
Part IV
Facework and contextualization

in interaction From (im)politeness
to humor
Irony in the face(s) of politeness

Strategic use of verbal irony in Czech
political TV debates
Jekaterina Maara
This paper examines the possible strategies behind the use of verbal irony in
Czech political TV-debates (since 1993) and the various factors that influence
politicians behavior in those debates. Since TV debates among politicians usually involve the participants trying to prove their point, they are set up as verbal
duels and, therefore, foster either negative use of irony against the opponent or
positive use to enhance the speakers own positive face. The choice of whether to
use irony or not can thus become part of a larger strategy of how the politician
and maybe even his whole party want to present themselves in the media. The
reaction to ironic remarks, especially those that attack the positive face, can also
play an important part, since it can help save the victims face or hurt it further.
Keywords: pragmatics, verbal irony, politeness theory, political discourse,
face threatening acts
1. Introduction
Many studies of verbal irony focus on either formal and semantic aspects (e.g.
Groeben and Scheele 1984; Halvorsen 1976; Lapp 1992; Weinrich 1966) or conversational properties of irony and the question of how and why ironic utterances
are processed and interpreted (e.g. Gibbs and OBrien 1991; Gibbs, OBrien and
Doolittle 1995; Giora 1995, 1997, 1999; Sperber and Wilson 1981; Wilson and
Sperber 1992; Attardo 2000). The central question of this study is how ironic utterances relate to politeness management (Brown and Levinson 1978, 1987) between politicians participating in televised debate-shows. This approach views
irony in a larger context of interpersonal and inter-party relationships between
. All mentions of irony in this paper refer to verbal irony.
188 Jekaterina Maara
politicians participating in a debate, and possibly also between them and both
the moderator and the public (depending on the format, there may be multiple
audiences the audience present at the taping and the viewers at home). Since
the issues of formal aspects of ironic utterances (lexical, prosodic, etc.) and their
decoding and reinterpretation by the addressee(s) are not of central importance
here, they will only be mentioned in passing. Factors that will be analyzed instead
are concerned with the positive and the negative face of the politicians and their
party, face threatening acts (hereinafter FTA) that constitute an attack against the
positive or negative face, and acts of protecting ones face. Ironic utterances will
be examined as to (i) their intended effect, (ii) their actual effect and (iii) the steps
the victim of irony takes or in fact, does not take in reaction to an FTA.
The data for the analysis stems from various political debate shows that are
described in more detail below. Furthermore, the genre of political TV debates
and their language is discussed briefly, followed by a breakdown of factors that
influence the production and reception of ironic utterances in such a setting. Finally, several case studies are introduced to illustrate different ways of addressing
an opponents ironic utterance.
2. On forming and decoding irony

There are many diverse approaches to verbal irony. Since it is not the primary goal
of this paper to discuss previous accounts of irony, I will only mention them briefly. The general definition of what irony is goes not only beyond the scope of this
paper but is sometimes difficult to grasp even in papers that explicitly deal with its
formulation. A working definition of verbal irony poses considerable problems,
. It is by no means the intention of the author to claim that all ironic utterances are necessarily negative or uttered with the intention to hurt the conversational partner. For analyses that
deal with the positive aspects of irony see e.g. Gibbs (2000), Maara (2011). However, the goal
of most participants in political debate shows is to win the debate and show that they are the
stronger competitor and, therefore, they are more likely to employ negative irony. A politician
might use positive irony to show that he is witty and intelligent and to ingratiate himself to the
viewers and create the semblance of a connection between himself and the viewers (positiveface enhancing irony, see e.g. Leech 1983; Maara 2011). Additionally, since this paper analyzes
the effect of irony on the opponent, the positive-face enhancing irony that is directed at viewers
can hardly be taken into account, since its effect is not immediately perceivable. There are, of
course, cases in which a specific utterance gains popularity and thus, the utterances effect on
the broader public is perceivable, but such cases are not discussed here and deserve a separate
study.
Strategic use of verbal irony in Czech political debates 189
since irony is a complex phenomenon determined by multifarious factors from

different fields of linguistics.
A very basic approach was taken in studies that were concerned with the form
and semantic meaning of ironic utterances: they centered on the aspect of incongruity between the surface meaning of an utterance and its intended meaning. A
rather broad definition of irony was established according to which the addressee
and/or audience understand that the speaker does not mean what s/he utters
(thus, s/he violates one or more of the conversational maxims usually quality
or relevance, but manner and quantity can also be flouted in order to produce an
ironic utterance) and try to decode the intended, hidden meaning of the speakers
utterance (e.g. Grice 1975; Searle 1979; Leech 1983; Haverkate 1990; Groeben
and Scheele 1984; Halvorsen 1976; Lapp 1992; Weinrich 1966). Some definitions
(Halvorsen 1976; Weinrich 1966) went as far as only considering those utterances
in which the semantic opposite of what is meant is uttered verbal irony; this is,
however, by far not the only possible way of creating ironic utterances. Therefore,
the broader definition of meaning something different than what is being said is
most certainly preferable. It is, however, unsatisfactory, mainly since it applies to
many different speech acts that have nothing to do with irony. Consequently, another factor must be added, which is that of humorous intent: by using an ironic
formulation the speaker wishes to appear witty to others and/or to mock or ridicule somebody an addressee or a third person/group of people (e.g. Lapp 1992;
Utsumi 2000).
Other approaches to irony try to establish, whether an addressee is able
to recognize irony and how s/he is able to reinterpret the utterance to find the
intended meaning behind it. Such studies are mainly conducted in the fields
of conversational analysis, cognitive and psycholinguistics (e.g. Sperber and
Wilson 1981, 1986; Sperber 1984; Wilson and Sperber 1992; Eisterhold, Attardo,
and Boxer 2006; Gibbs and OBrien 1991; Gibbs, OBrien and Doolittle 1995;
Giora 1995, 1997, 1999; Utsumi 2000). The most influential study among these,
mainly also because it has sparked a vivid debate and inspired or influenced many
subsequent studies, was the relevance-theoretic approach to irony as echoic mention by Sperber and Wilson. They argue that irony consists of an echoic reinterpretation of another persons utterance, opinion, view, expectation or norm. By
being ironic, the speaker dissociates him/herself from the echoed utterance/opinion etc. and supplements them with mockery or scorn (Sperber and Wilson 1981,
1986; Sperber 1984; Wilson and Sperber 1992). Nevertheless, the notion of echo
is by no means universally valid since it cannot account for every ironic utterance
(shown by, e.g. Giora 1995; Utsumi 2000). The echoic approach has for instance
difficulties in showing how irony can be distinguished from other echoic interpretations of others opinions/utterances (Utsumi 2000:1781).
This paper does not attempt to provide a new answer to the question of what
irony is and how the addressee can interpret an ironic utterance. The cases of ironic utterances examined below are all easily qualified as ironic by the more basic
approaches (e.g. Grice 1975; Searle 1979; Groeben and Scheele 1984; Lapp 1992).
An explanatory model that proves helpful in qualifying an utterance as ironic
is provided by Yus (2000) who proposes a set of contextual sources that people
can access to process the given utterance and recognize its ironic nature. These
sources consist of (a) encyclopedic, factual information; (b) setting; (c) speakers
nonverbal behavior; (d) addressees knowledge of addressers biographical data;
(e) mutual knowledge; (f) previous utterances in conversation; (g) linguistic cues
(after Yus 2000:33ff.). The more contextual sources can be accessed, the easier it is
for the addressee to interpret the ironic utterance (Yus 2000; Zhao 2011; see also
Utsumi 2000 and his notion of ironic environment).
Distinguishing between irony, sarcasm and mocking goes beyond the scope of
this paper, and, therefore, all three will be called irony where they overlap. They
all constitute FTAs against the conversational partner, but they vary in strength,
e.g. sarcasm is understood to be more aggressive/negative (Barbe 1995:28; Gibbs
2000; Norrick 1993:135).
3. Televised debates and their language

The focus of this study are 6 excerpts from talk shows with two or more politicians present, or, in some cases, only a host/interviewer and one politician as the
interviewee. Most of the programs (with the exception of Kotel) mentioned here
are not recorded in front of a live audience. The debate show Sedmika, however,
offers at least some indication of an audience feedback in the form of a continuous
text message poll that periodically displays the percentage of viewers who agreed
with one or the other politician. The politicians do not appear to know about the
results during their talk, yet they are of course aware of the audience being polled.
This, of course, still does not come close to an actual live audience, but it offers at
least some added awareness of the public. Additionally, it transforms the staged
performance into a virtual sports fight that can only have one winner, who is
literally being determined by a count on the screen. It should be emphasized that
. Some of the descriptions of categories of contextual sources have been shortened, but still
render the information given in Yus (2000) faithfully.
. For a broader introduction to the role of media and public sphere, see Busch (2006),
Fairclough (1995, 1998); for an overview of the various genres of political discourse, see
Fairclough (1995, 2006).
political talkshows, while being staged to provide information about the participating politicians agenda and views, also have an impact in the entertainment
sector (Busch 2006:611). It is therefore often the case that the discussions are
staged in a way to be more entertaining and attract more viewers. Since politicians
have become used to the formats in which they can be verbally attacked and have
to defend themselves and their policies, they, in turn, have become accomplished
television performers and even media personalities (Fairclough 2006:34). The
media and a politicians style of presenting themselves and their media persona
have become important components of politics (Fairclough 2006:33f.).
In view of face threatening acts, an especially important aspect of televized
political discourse, is the staging of a debate as confrontainment (Hess-Lttich
2007). Depending on the format, politicians are more or less expected to participate in a verbal fight. Representatives of two opposing parties or ideas are presented
as antagonists and the discussion itself takes the form of a formalised battletalk
(Hess-Lttich 2007:1367). Hess-Lttichs paper examines the Swiss show Arena
(see also Luginbhl 1999, 2007), which finds an even more battle-like counterpart
in the Czech political talk show Kotel, to be described below. A setting in which
a politician must display his willingness to verbally fight for his ideas fosters the
use of verbal violence (see Luginbhl 1999, 2007). Irony can be a sub-set of verbal
violence, if it is used to attack an opponent, it can be a means of displaying intelligence and wit, or a measure to protect a politicians negativeface.
Hess-Lttich (2007) introduces the term show conversations to refer to different kinds of conversations that are staged for an audience. He uses this term to
subsume conversations that follow basic structural rules of communication or
indeed violate these rules in certain ways to achieve specific effects (Hess-Lttich
2007:1361). This type of language use includes dialogues on the theatrical stage,
poetically designed fictitious conversations, () panels, telephone conferences,
radio phone-ins, experts appearing before public audiences, political discussions
on television and liturgical rituals (Hess-Lttich 2007:1361), but also talk shows.
An important feature of show conversations is their frame or setting in a media
world, which in turn defines the multiple communicative orientations of speakers and addressees (ibid.). The last statement is especially important in the world
of politics, since politicians often represent not only themselves, but also their
party. When they are conveying an opinion while on a talkshow, they are expressing their own stance and that of their party; additionally, they address their immediate opponent or the interviewer as well as an audience/viewers (physically
present or watching at home). Therefore, a politician can carry out various acts
. In order to keep the two apart, I will refer to the people present at taping as audience and
those watching at home as viewers public refers to everybody who is not a direct (i.e.
of speech and achieve different perlocutionary effects with one and the same utterance (Khn 1983:248; see also Khn 1995). Politicians participating in a debate
are aware of this and can use it to bring a point across; they can use a conversational partner to deliver a targeted message to a third party (i.e. the audience,
the viewers) (Hess-Lttich 2007:1362). This notion will prove important in the
analysis below, since an utterance must be examined from various points of view.
Not only must we consider at whom an utterance is addressed, but we should also
determine what role it plays in which sphere of a politicians activities.
The data used in this paper stems from two sources. Transcriptions of two
shows were made for this study from taped TV shows or debates, some of which
can be found on youtube and similar sources. The transcription was kept as simple as possible and only those indications of unusual emphasis, stuttering etc. were
made that are crucial for the proper understanding of irony and the question at
hand. The other source is the Dialog Corpus that provides both transcriptions
and videos of the programs, thus providing important cues for the understanding
conversing) participant in the media discussion; this group may include people who have not
been watching the discussion, but have heard of it/read about it. This third group is important,
because political discourse must be analyzed beyond the conversation of a single talk show. It is
constantly formed and changed by new interactions and public appearances by politicians and
they influence the public opinion as long as they are medialized. After World-War II, televised
debates became increasingly important in political discourse, since they included the public in
politics much more than before (Busch 2006). In the Czech Republic, important political TV
programs are often televised on Sunday, assuring that a larger segment of the population has the
possibility to view them. Of course, if something is especially relevant or maybe even shocking
or surprising, it can be picked up by other shows (e.g. news shows) and aired over and over for
everybody to see; this has never been more the case than in todays world of YouTube and other
internet platforms that are capable of turning the most obscure events into internet sensations
with millions of viewers. Additionally, some statements can be placed in a broader context by
the voters or by other TV shows. All this means that a politician must always be aware of being closely watched (Clayman and Heritage 2002; Hess-Lttich 2007; Nuolijrvi and Tiittula
2011).
. The excerpts from the talkshows Duel, Kotel and Otzky Vclava Moravce quoted below
were transcribed for this paper (the videos on youtube were last accessed on 9/14/2011).
Excerpt (2) is from Klaus website www.klaus.cz (accessed on 8/15/2011).
. The DIALOG corpus (ujc.dialogy.cz) is a multimedia corpus of spoken Czech transcribed
from various TV programs broadcast on Czech television. The corpus was compiled by the
Czech Language Institute of the Academy of Sciences of the Czech Republic in cooperation
with the Institute of Formal and Applied Linguistics at the Charles University Faculty of Mathematics and Physics. The transcriptions are numbered by turn and indicate elements such as
pauses and fillers, indicators of hesitation, emphasis, overlapping speech, latching, etc. The
corpus is searchable and the examples can be viewed as transcriptions or as videos, which is
especially helpful in case of non-verbal indicators of irony.
of ironic discourse, such as intonation, emphasis, expressions, and gestures. Rather than chosing a specific show or pairing of politicians, the excerpts were chosen
to represent the various possibilities of intended attacks on the positive or negative
face and the possible reactions. It was impossible to find all kinds of interactions
in just one debate. Therefore, the excerpts are presented as case studies illustrating
different types of attacks or reactions to such attacks (e.g. attack on the positive
face, protection of positive or negative face, etc.).
4. Factors influencing the use of irony in political discussions

This paper will examine possible reasons for employing irony and reacting to an
ironic utterance in a specific way, i.e. when does a politician choose to counter an
attack by using irony and when not. The context of political debates being usually
well known, they present suitable material for the analysis of multi-addressed utterances. It is often also possible to determine the reason for an ironic utterance as
well as the reason(s) for reacting to it in a specific way, since the people involved
in a conversation have assigned roles: the public knows which party a politician
represents and therefore can predict which politicians tend to be opposed to each
others ideas. Consequently, if a politician makes an FTA against opponent, the
context is rich enough to allow a fairly specific interpretation based on knowledge
about the setting; background information on the participants, such as personality, conversational style, family, personal lifestyle, etc.; knowledge of previous
utterances in the present interaction, knowledge of previous interactions, party
lines, etc.
The politicians participating in a debate have two positive faces that they must
protect. They participate as themselves on one hand, but also as representatives
of their parties and their parties causes on the other. It can even happen that the
two positive faces are in conflict with each other, if, for instance, a politician must
defend something that his party did, which he himself could not avert. FTAs can
be directed at any positive face, the partys or the persons, but in the particular
interaction, it is, of course, always the person who must react to them. A combination of the two attacks against the positive faces (the politicians and the partys)
is possible. An FTA may be aimed primarily and directly at the opponent who is
present, but can have a generalized reading that extends on the whole party and
its campaigning skills (see example (6) below).
. In the case of TV show transcriptions, the cited excerpts are accessible and non-verbal
behavior can be observed on video.
As mentioned before, politicians must choose how to present themselves

and what strategy to use. For a discussion of rhetorical strategies the speakers
employ to negotiate their position in the debates, e.g. enforcement, which gives
the speaker more communicative rights while at the same time also enhancing
the adversarys obligations, see Kallmeyer and Schmitt (1996) and Gotsbachner
(2008). Enforcement in the form of irony can be used to provoke the opponent,
force them to answer or defend themselves even if they originally did not see
cause to do so; irony can twist the oponents words, or simply be an attack on their
positive face.
The presence of an audience can also have a varying effect. The participants
awareness of viewers in front of their television sets can be taken for granted, but
the audience present in the studio may also influence the behavior of a politician
substantially, and certain shows seem to be conceived around that concept. As
part of Hess-Lttichs concepts of show conversations and confrontainment, the
producers of each show with a live audience usually determine the role of said audience quite specifically, because the presence or absence of audience reactions/
participation help shape the overall image of the show.
Within the Czech political discussions data, the show Kotel (TV station
NOVA, since 1998) should be mentioned as having a very specifically involved
audience. The interpretational frame is already set by the term kotel cauldron itself. There are two associations with this word in Czech that are completely in line
with the way the audiences participation is devised. Kotel is a slang expression
used to refer to a sports stadium where sports games such as soccer and hockey
are played, and where fans behave rather loudly and emotionally. Additionally,
the term cauldron brings about associations with hell, which also fits the shows
heated atmosphere with emotions rising to a fever pitch. Audiences are brought in
and encouraged/instructed to engage in heckling and booing of the invited guest.
The host has to look stern and somewhat angry, and she asks the audience to
express openly and loudly how they really feel about the politician. There are supporters and opponents of the politician (who is alone with the host in the middle
of an arena-like set with rows of audiences ascending around the stage) and while
he or she is answering questions, they interject various exclamations or cheers.
Needless to say that such a show concept is clearly meant to have entertainment
value rather than be informative (this is in line with the TV stations reputation
. For instance, American daytime talk shows that focus on everyday problems are in part
characterized by the sort of audience they have: the people in the audience are engaged in the
discussions by booing, heckling, cheering, and adding their own opinion; all of this is part of
the standard image of such shows, and therefore, the audiences are encouraged to behave in
thisway.
for relatively cheap entertainment without much intellectual value). A politician

who has to face the audience in such a setting will behave differently and may
have to re-evaluate the initial plan of how he wanted to present himself, because
he may not be given a chance to act the way he would in a debate with a different
audience. In a show like this (even more so than in the example of Swiss Arena
discussed in Luginbhl (1999, 2007) and Hess-Lttich (2007)), confrontainment
is foregrounded and verbal violence fostered. In contrast to Arena, where several
guests are divided into two teams battling each other, Kotel seems to adopt a one
against all principle of battle a setting that puts more pressure on the guest.
Most examples considered for this paper, however, are selected from talk
shows that are more neutral in their set-up and tone. They usually involve two
or more politicians who contest each others claims and ideas about current issues.
For the purpose of this study, they are more interesting; while verbal violence/
FTAs are still part of the conversation, they are not the basis of the interaction
and, therefore, such interactions are closer to normal conversations (as opposed
to the trial setting in Kotel). However, even in such neutral settings politicians
have to choose a role to portray. They can decide to be more aggressive against
their opponents or more laid-back and calm. The use of irony is also part of the
persona they portray. The interactions take place in several spheres of influence
presented in Figure 1.
The complexity of irony does not end with the various spheres that influence
a politicians performance/behavior. If we establish the politician who holds the
Public
Media reality
(Audience)
TV show/debate
(Audience)
Host
Politician A
Politician B
Party
B
Party
A
Viewers
Figure 1. Spheres of influence on politicians performance in talk shows
floor as speaker (S) and the addressee as hearer (H) we encounter a first problem.
It is not always easy to establish the identity of the intended addressee. If we imagine an entity H (viewers, or audience, if present), in everyday conversations it is
clearly a third party who is not directly addressed, but can overhear the conversation and possibly decode the hidden message.10 On television, however, it is often
hard to draw the line between H and H, since a politician may be answering the
hosts question quasi-neutrally and be speaking into the camera (through which
he is directing his utterance at the viewers), but at the same time intending to attack another politicians positive face through the implicature of his message.
In the following paragraphs, several case studies are presented in which the
factors discussed in this chapter are analyzed. For each example, a hypothesis is
given of why the speaker employs irony, how his/her utterance affects the addressee (which face(s) are being threatened) and how the addressee decides to
react to this attack. The examples were chosen to illustrate a broad range of possible interactions. However, due to the lack of space this study makes no claim to
cover all possible configurations. Instead, this study presents a small number of
examples that were analyzed in more detail because they display a higher degree
of complexity.
Most of the examples include an adverse use of irony, which should by no
means imply that all ironic utterances are examples of verbal violence. For this
study, however, such examples presented a more interesting context. Cases where
irony is clearly used to simply reinforce a good relationship through the enhancement of the others positive face are fairly straight-forward and the reaction to
them can either be equally positive or a rejection of the reinforcement of the relationship. Cases of violent irony, on the other hand, present a challenge to the
addressee that s/he can either answer or consciously leave unanswered to prove
a point.
The goal of these case studies is to show that a broad context must be considered when assessing the effect of irony on politeness management. The examples
will highlight reasons for chosing to employ an ironic strategy, the possible reactions to the ironic utterance in question and will, where possible, examine the
broader context surrounding these choices (i.e. what part does the position of the
party play etc.).
10. For a more detailed illustration of the possible configurations between S, H, and H see
Maara (2011).
5. Irony in use: Case studies

In example (1), the role of the pretended vs. the intended addressee is crucial in
the process of establishing the speakers position in the conversation. The speaker,
Zeman, is at the same time seemingly answering the hosts question, while also
describing three kinds of politicians to the public. On the surface, his answer
seems unrelated to his opponent, Sldek, but it is clear that he is attacking Sldeks
positive face by putting him in the third group, that of ridiculous politicians.
(1) modertor 1: [] jak hodnotte chovn, eknme, sldkovch republikn
[] v parlament?
Zeman: [] no, a pokud jde o hodnocen republikn .. j bych samozejm
nechtl hned v vodu naeho setkn pana Sldka urazit, .. ale
myslm si, e jsou ti typy politik: e, politici, kte jsou oblben, e,
politici, kte jsou neoblben, a politici, kte jsou smn. A j bych
rd nepatil do t tet kategorie.
mod. 2: []
mod. 1: mimochodem jsem troku ctil, e pan Zeman by mohl ct, e tm
smnm politikem byste mohli bt i Vy.
host 1: [] how do you assess the behaviour of, lets call them, Sladeks
(Sladekian) republicans [] in parliament?
Zeman: [] well, concerning my assessment oft the republican party .. I
would not want to insult Mr. Sladek right from the very start of our
gathering, .. but I think, there are three types of politicians: politicians who are popular, politicians who are unpopular, and politicians who are ridiculous. And I would not want to belong to the
third category.
host 2: []
host 1: by the way, I had a feeling, that Mr. Zeman might have said that that
ridiculous politician could be you.

(TV show Duel, 1993: http://www.youtube.com/watch?v=b8PeV8U4H2c&)
This is part of Zemans first turn on the show, and he establishes himself as the
stronger speaker right away. Without mentioning his opponent or the opponents
party, he manages to insult him while asserting that he does not want to do so.
By doing this, he implies that it is impossible not to insult Sldek. As discussed
above, the classic definition of irony as a device to express something by stating
the opposite (see Section 2) is too weak to express the potential of this utterance.
More important than the opposite meaning is the ambiguity between the overt
and the hidden message and the possibility of both messages playing an important role in the final interpretation (cf. Hutcheon 1994). A possible interpretation
may be provided by the echoic interpretation theory (Sperber and Wilson 1981,
1986, 1998; Wilson and Sperber 1992; see above), which describes irony as an
echo of something that occurred or was mentioned before or belongs to the set
of beliefs/values of the opponent and which the speaker evaluates as negative. By
providing this mock-factual classification of politicians, Zeman seemingly offers
a fact instead of a personal opinion. In the surface meaning of his utterance, he
is not actually committing the FTA. This is a skilled employ of irony and seems to
provoke the desired effect in the further course of the debate. Sldeks first reaction to this is simply a smirk, and then, when it is his turn to speak his mind, his
opening statement consists of stating that he does not get many chances to speak
on TV and that he would, therefore, like to wish everybody happy holidays. While
this might have been acceptable as an initial turn, after an attack, it looks like an
evasive maneuver.
What seems to be of lesser importance is the distinction between overt and
hidden message and the marking or not of the latter. While Zeman does not clearly mark the irony in example (1), it is understandable to both participants and
probably the majority of the viewers watching the debate. Since politicians know
each other quite well and often interact in similar circumstances, irony need not
be marked in order for its meaning to be grasped. The context and the interlocutors inherent political positioning are usually sufficient to decode irony (see also
Barbe 1995; Pexman and Zvaigzne 2004; Sperber and Wilson 1986).
In theory, unmarked irony is potentially dangerous, since the victim can miss
the hidden meaning and humiliate him-/herself further by reacting to the overt
meaning. In medialized political discourse, this seems to happen extremely rarely.
Instead, the choice of reacting to irony or ignoring it is part of a larger plan to
position/present oneself in a specific way. However, the way a politician reacts can
still demonstrate his or her verbal skill or lack thereof. In example (1) Zeman does
not count on Sldek being humiliated by not being able to interpret the irony: the
primary aim of such a phrasing is displaying his abilities as a public speaker. It is
important to note that the attack is successful. After Zeman called Sldek ridiculous, the latter reacts evasively and shows his lack of similar rhetoric skills, thus
proving Zeman right.
What deserves additional attention in this example is the continuation in
which one of the two hosts picks up the implicit attack and specifically addresses
Sldek, unnecessarily, translating Zemans implicature into an overt, albeit hedged,
insult. This probably points to the fact that Zemans FTA was welcome and that
the host hopes it will spark an emotional reaction. That would mean that the host
positions the two in a fighting situation, which of course fits the programs title,
duel. Sldeks evasive behavior, on the other hand, seems to be a cesura of sorts,
in which he can step away from the fight and show that his positive face was not
affected and that he is still friendly and well spoken. Yet, he does not miss the
opportunity to counterattack after his happy holidays message and continues
saying that now he can come back to those other details. This understatement
can also be read as irony and symbolizes that Sldek wants to position himself as
being above fighting and better than that.
While many linguistic accounts of irony are based on the assumption that the
essential element of irony is somehow negative (e.g. Barbe 1995; Hartung 1998)
or, more specifically, expresses the speakers adverse attitude towards the occurrence or statement he echoes (Sperber and Wilson 1992:60), there seem to be
more positive attitudes towards irony as well, some of which may be applicable to
political debates. Brown and Levinson (1987:265) and Leech (1983:82ff.) argue
that irony is a means to express criticism in a more polite way, since the addressee
must first decode the irony, and the less direct impact makes it more palatable.
This approach is far too generalizing and I would in fact argue that it is mostly
wrong in assuming a softening effect at all. Critique expressed through irony may
actually have the capability to be more hurtful, because it contains a mocking
undertone. Kotthoff s (2003) study, on the other hand, has shown that another typical use of irony is a ritualized, playful exchange among friends (see also
Maara 2011). While political adversaries can hardly be called friends, the ritualized aspect of irony-exchanges may still be important in televized political discussions, since these programs are made for entertainment as well as information,
and because politicians can present themselves as having a well-developped sense
of humor and thus appear more likeable to the voters. Instead of only exchanging
irony when criticizing each other, a general playful tone and more light-hearted
responses may add to a politicians appeal. Another positive aspect of irony is the
fact that it emphasizes the speakers ability to react fast and cleverly, which displays his or her intelligence (Hartung 1998:20f.).
While it is more typical to use irony in an FTA against an opponentss positive
face, there are also cases where irony is used to either protect ones own negative
face or where it supports the violation of an opponents negative face. As noted
during the analysis of example (1), another aspect that merits a closer look in political interaction is the reaction to irony. Although the speaker might think that
s/he can predict the outcome, it is not always possible to calculate the reactions
of the target and the public. This is essential in connection with politics because
the reaction to irony may very well be co-determined by the respective politicians
partys greater plan, or at least his personal understanding and execution of it.
5.1
Irony and the negative face
Cases where the negative face is affected offer the possibility of both protecting
ones own and attacking the opponents face with the help of irony. Especially in
the latter case, it is mostly a dual attack: it is hard to think of a possible FTA against
the negative face of the opponent that would not also threaten his positive face.
The only possible scenario where these two types of FTA do not combine would
be that the speaker is employing self-irony as a device that tries to pressure the
conversational partner into saying something (e.g. agreeing to something, paying
a compliment etc.). Within politeness theory this may be still viewed as an intrusion into the negative space of the conversational partner. However, even here, the
positive face of somebody must be attacked (even if it is the speaker him/herself).
Therefore, the core element of irony within politeness theory would have to be a
positive FTA. Even if irony is used to soften criticism (originally proposed as the
core function by Brown and Levinson (1987) and Leech (1983), but rarely found
in the data examined in this study), it is still part of an FTA.
5.2
Attacking the negative face
In example (2) the question of unemployment numbers is discussed and pidla

starts saying that it was not his party who caused unemployment to increase when
Klaus interrupts him stating that the chart proves pidla wrong. He formulates
this objection as a rhetorical request, since a chart is based on facts and redrawing it is impossible. This rhetorical request has two effects. First, it is an attack on
pidlas positive face, stating that he seems to be ignoring the facts, and second,
the challenge presents an attack against his negative face, because it forces the
opponent to come up with proof for his previous statement. The more urging repetition of this request adds further power to the attack, since the hidden message
must be read as something like I will not let you get out of this, explain or lose
this part of the debate. Moreover, in the greater scheme, Klaus attacks the whole
party for operating on false facts, or for not being able to corroborate their claims
with actual data. pidla tries to answer the hosts question, Klaus does not let him
finish his answer, because he feels that pidlas answer is wrong in its core.
(2) Bobokov (modertorka): Take pojme k t nezamstnanosti. Chtl jste
na ni reagovat, pane pedsedo?
pidla: Ano. [] Musm ci jasn, e nen pravda, e se nezamstnanost
zvila dvakrt v dob naeho nstupu, protoe v okamiku, kdy
jsme
Klaus: Mete protestovat proti tomuto grafu, mete ho pekreslit tady ped
divky.
pidla: Myslm si, e ho nepotebuji pekreslovat, protoe
Klaus: Pekreslete ho
pidla: protoe, ven pane,
Klaus: J vm ho pedm, vy jste ho neml pedtm.11
pidla: protoe nezamstnanost v dob, kdy jsme nastoupili k stavn
odpovdnosti, byla 7,5%, v souasn dob je 8,8% nebo 8,6%, to te
si pesn nepamatuji.
Bobokov (host): So lets talk about (that) unemployment (issue). Did you
want to react to it, Mr chairman (pidla)?
pidla: Yes. [] I have to say (clearly), that its not true that unemployment
doubled during our administration, because when we,
Klaus: You can contest this chart, you can redraw it here in front of the
audience.
pidla: I think that I dont have to redraw it, because
Klaus: redraw it
pidla: because, dear Sir
Klaus: I can give it to you, since you didnt have it before.
pidla: because at the time our administration came into office, when we
took over that responsibility, unemployment was at 7.5%, and now
it is at 8.8 or 8.6%, I cant remember which one it is right now.

(TV show 7 6/10/2002: http://www.klaus.cz/clanky/198812)
Using irony in a situation like the one cited in example (2) has more than the immediate benefit of pressuring the opponent to answer/explain further, and, ultimately, admitting that he is wrong (the latter being the ultimate goal). An ironical
and rhetorical request presented in a moment like this (while the opponent is
claiming the opposite of what Klauss chart shows) is very condescending exactly
because it delivers a strong attack against the negative face of pidla. Not only
does Klaus not let pidla finish answering, he does not even leave him any room
to present arguments that might support his claim, since, if the chart says something different and is based on accurate data, there are no supportive arguments
and, thus, Klaus has won this particular round of the discussion already. All of
this is conveyed through this ironic remark. This is supported by the fact that this
11. Italics added for highlighting.

12. This is the only excerpt for which I do not have the film clip. Therefore, it is printed here as
it was transcribed on the web page and does not conform to the transcription conventions of
the other excerpts.
transcript was published on Klauss web page, which indicates that his evaluation
of his own performance on the show is very positive and he feels that his supporters would enjoy reading this again.
Another example of an effective use of irony as a device to hurt somebodys
negative face is (3), although the primary effect is still a hurt positive face. After
one participant in a debate between various politicians does not let the host stop
him from talking, the host employs a rather humiliating ironic comment in order
to stop the politician.
(3) Moravec: Vy, tm e jste piel o mandt, to chcete tady moderovat.
Moravec (host): (I get it,) because you lost your mandate, youre trying to
moderate the show (literally: it here).

(Otzky Vclava Moravce)
The choice to humiliate the politician must be based on the assumption on the
hosts part that hurting the positive face of the conversational partner strongly
will cause him to stop speaking and let the host pose his question and conduct
the debate in an orderly manner. This is a reactional FTA provoked by the politician, since he was rude first by not letting the host manage the debate, and was
thus violating the hosts face as well. Such an exchange shows a classic fight for the
floor. However, the roles cannot be equal, since the host has the right and duty
to manage the debate and he, therefore, should be in a stronger position. By not
letting the host speak, the politician was undermining the hosts power of control.
Being ironic and humiliating his guest is thus a device to regain the power of the
manager of the debate.
5.3
Protecting the negative face
Irony can be a strengthening factor in protecting ones own negative face. In example (4), which was transcribed from the aforementioned show Kotel, the host
is questioning the guest about his personal finances and naming the worth of his
wifes and, as she alludes to here, his lovers villas. The hosts tone is already very
sarcastic and, as mentioned above, the show is about humiliating and enraging
the politicians who come there. Irony can be employed by the host in order to
incite an emotional, uncontrolled response. The guests can use irony in order to,
on one hand, protect their positive face by not reacting aggressively, and on the
other hand, to protect their negative face by not referencing the question in their
answer.
(4) Jlkov: take mohl byste nm vysvtlit, eeh, kde tyto milionov nemovitosti
se vzaly?
[povzbuzujc vkiky divk]
Sldek: Ven pan Jlkov, j jsem velmi rd, e tento poad, jak jste ho
zahjila, si pln zaslou j jsem slyel nkde na nmst, e se tomu
snad k stoka (..) nebo kanl (..) nebo nco takovho, jsem
velmi rd ((potlesk, psknut, etc.)) jsem velmi rd, e tuto otzku
jste nastolila jako prvn, abychom ji mohli vydit a u se mohli
vnovat skuten tm vcem, kter zajmaj obany: pedvolebnm
otzkm, zdravotnictv, kolstv (([interrupted by the host])) a tak
dl.
Jlkov (host): so, can you tell us, uhm, where these properties that are worth
millions come from?
Sldek: Dear Ms Jlkov, I am very glad that this show, the way you started
it, fully deserves I heard somewhere that it is apparently called a
sewer (..) or a manhole (..) or something like that, I am very glad
[audience clapping, whistling, etc.] I am very glad that you dished
up this question first, so that we can get it out of the way and discuss
issues that people are actually interested in: pre-election issues,
healthcare, the school system (([gets interrupted by the host])) and
such.

(Kotel, 1998: http://www.youtube.com/watch?v=qRlh53r6drU)
He does not hurt the positive face of the host (or rather her show) just for the
sake of it. He has two reasons for doing it: first, he signals that he will not respond to inquiries about his personal life by not dignifying the question with an
answer. Thus, by not providing a relevant answer and by being ironic, he protects
his negative face. Second, his reply is tantamount to a declaration that he is above
her provocations and will not stoop to her level. This example also shows how
complex the evaluation of irony is not, because it is particularly difficult to decode, but because it is difficult to place it on a scale of politeness. On one hand, we
can say that irony is mitigating aggression, since instead of being directly angry
and aggressive, the politician saves his face and remains composed. On the other
hand, that does not mean that the criticism is less strong. On the contrary, the interviewee is mocking the concept of the show even more in saying that he is glad
it lives up to its reputation, because now he can show his superiority.
As mentioned before, the entire set-up of the show Kotel is very atypical for
political discussion shows, since it is trying less to provide information than it is
to show a politician being on trial, judged by common people. Therefore, FTAs
and irony are found more frequently on both sides and are even part of the hosts
persona. Since the invited politician has to play the role of a defendant, his irony,
while sometimes being aimed at the show or the host, often serves to save his
positive or protect his negative face. Since there are no other politicians present,
he may attack an absent politician or a party, but even then it would also serve the
purpose of saving his face, because all attention centers on him and he is the one
who was invited there to answer questions. Consequently, ironic remarks uttered
in such a context should be regarded differently than those used in a more neutral
setting, where all invited guests must answer questions and defend themselves.
5.4
Let the fight begin! Positioning oneself and others in the debate
Since politicians participating in a televised debate are aware of having an audience and try to present themselves in a certain way, they may also choose a
position within the debate, i.e. to be more or less aggressive, to attack the other
participants or not, etc. The way they decide to act can also influence the choice
of being ironic or not.
A very common way of employing irony to position oneself in a particular
way is to show superiority, i.e. the speaker enhances his/her own face while trying
to damage the opponents positive face. This shows that the speaker is the dominant personality, he will verbally fight his opponent, he is the quicker and cleverer one etc. Example (5) illustrates such an attempt well. The first ironic remark
(poor little ODS) in Zemans turn would fall under frozen irony and is a simple
act of belitteling the other party, by implying that they always complain that their
mistakes were somebody elses fault. According to him, the opposite was the case
and therefore, the ODS party should stop behaving like this. The second ironic
remark, an ironic advice, continues this teaching way of speaking by instructing
Beneov, who is a politician herself and knows about politics, on how politicians
should act with respect to their party.
(5) Zeman: [] nen pravda, e by sociln demokracie brnila chudince
des ((mnno ODS)) neprivatizovat banky. je pravda e v
pedseda, pan Klaus, veejn ekl, e banky nechtl privatizovat
z dvod, kter tak uvedl. take, mluvte jednm hlasem politick
strana m mluvit [jednm hlasem.]
Beneov: [((povzdech))] j si nemyslm e tady mluvm
njak e jinm hlasem, nebo o nem jinm ne mluv vclav
klaus, v tom si myslim e mezi nma nen dn rozdl.
Zeman: [] that means its not true that the Social democrats kept the
poor little ODS (political party) from the privatization of the
banking system. Its true that your chairman, Mr Klaus, declared
publically that he wasnt interested in the privatization of the
banking system for the reasons he mentioned. So speak in a unified voice, a political party must always speak in a unified voice.
Beneov: (sigh) I dont think that Im speaking (here) in some, uhm, other
voice, or about something else than what Vaclav Klaus was talking about, I think there is no difference between the two of us in
this regard.

(7 ili Sedm dn; 21.3.1999 (Corpus ujc.dialogy.cz)13)
Through these remarks, Zeman shows that he has a sense of humor and he positions himself as the stronger speaker. The effect of his behaviour can be seen in
her reaction. First, the sigh betrays a certain degree of exasperation, and second,
she responds to the overt meaning of what Zeman said. While this was also part
of his attack, it is not the main concern to speak in a unified voice. Instead, she
should have reacted by saying something about the issue on which they spoke in
a unified voice, but she fails to do so, and instead more or less repeats herself three
times. Zemans ironic advice seems to disrupt his opponents line of argument and
prevent her from reacting adequately.
The TV program itself can already add something to this positioning. In the
next excerpt, the broadcast shows short pre-produced clips before each topic is
discussed, summarizing the issue and each of the politicians (and their parties)
stands on it. Those clips can, however, do more than that, as shown in example (6): they can set the mood for the discussion and already try to position the
speakers in a fighting situation before they even started discussing anything. The
host then might try and parlay the set-up by indirectly asking the guests whether
they will fight or not.
(6) Paroubek: pana topolnka semelu kafemlejnkem kdy si vzpomenu.
((stih))
Topolnek: myslte e j se nechm semlt? ((smch)) ((stih))
modertor: tak jak to [bude pnov? kdo koho semele?]
Topolnek: [((smch))]
Paroubek: [no tak]
mod.: [pane] premire
13. Changes: Names written out on first turn instead of initials for clarity; turn-numbers removed; italics added as highlighting.
Paroubek: j sp bych navzal na ty nae stetnut pedchoz, myslim si (.)

e (.) j mm tak troku obavy jak vst ten rozhovor dnen protoe minule kdy i takov lidov noviny musely uznat e to bylo s
moj pevahou, tak pan pedseda to nazval prasrnou, kdy jsem
mu poslal takov mrn ironick dopis na ten jeho inzert ke sto
dnm vldy sociln demokracie nebo m vldy tak e nebylo
dstojn pro nj na to reagovat. tak aby se mu neudlalo tady
patn.
mod.: j bych jenom ehm kdy tady vznte pravdpodobn enickou
otzku jak vst tento rozhovor, tak bych vm s ohledem i na
minulou sedmiku em e navrhl, e bych ten rozhovor vedl j.
Paroubek: to by bylo dobe.
mod.: tak se tak na tom na zatku [dohodnme.]
Paroubek: [doufejme e] budete ve form pane
redaktore.
mod.: j se budu snait ehm jestli chcete reagovat na pana premira
[pane]
Topolnek: [ehm] [] j si myslim, e (.) ani ne, k tomu eknu jedin ee j
vbec nemm ten a myslim se e spoustu lid nem ten obecn
pocit, e m pan pedseda paroubek semele kafem- jako kafemlejnek (.) musm ct e ta posledn <my jsme spolu byli dvakrt> v njak debat, [] a mm pocit tm rovnocennho
vsledku e co myslim e je dleit ja myslim e ale nejde o to,
jak my dva soupeme. pod si myslim a ee nejsem v politice
tak dlouho abych tu iluzi definitivn ztratil, e ten souboj je sice
personifikovn do naich dvou osob ppadn a proto j vyzvm
pana paroubka pojte do prahy. pojme. personifikujme ten souboj i v tch volbch. to si myslim e je dleit? ale pece jenom je
to vce souboj program, idej, tch politickch subjekt ne souboj
njakch osob kter se budou petlaovat silou svch vrok. j
se, j se tch vrok nedopoutm, dopoutt nebudu a na tu hru
prost [nepistoupm].
mod.: [tak pane] pedsedo j myslm e kdybych tady zaptral
ve svch paprech tak bych nkolik pknch vrok z vaich st
nael.
Topolnek: musim ct e velmi velmi sloit budete ptrat v tom e bych
napadal pmo pana premira paroubka. tomu se [tedy]
mod.: [ta:k]
Topolnek: intenzivn vyhbm.
Paroubek: I could crush Mr Topolanek with a coffee grinder if I wanted to.

((cut))
Topolanek: Do you think I can be crushed? ((laughter)) ((cut))
host: so, what will it [be? Who will crush whom?]
Topolanek: [((laughter))]
Paroubek: [well]
host: Mr [prime] minister
Paroubek: Id rather talk about our previous gatherings. I think . that. I am
sort of anxious about how to conduct this conversation today,
because last time when even a paper like the Lidov Noviny had
to admit that I dominated that debate, (so then) Mr chairman
(Topolanek) called it insolence (the Czech word is derived from
pig) when I sent him a slightly ironic letter as an answer to his ad
for the anniversary of hundred days of the social democrats in
office, or my government, uhm, he did not deem it worthy of his
response, so Im concerned about the possibility of him getting
sick here today.
host: I would just, uhm, because you brought up the, probably rhetorical, question of how to conduct this debate, so I would like to
propose, also looking back at our last show, that I will lead this
conversation.
Paroubek: That would be good.
host: so lets agree on that from the beginning.
Paroubek: I hope youre in shape
host: Ill try. Uhm, if you want to react to what the prime minister said,
[mr]
Topolanek: [uhm] [] I think that (.) not really, I will say one thing to that.
uhm, I dont feel like, I think many people dont feel like, generally, that the prime minister can crush me with a coffee grinder.
I have to say that the last one, we had two debates, [] I feel
like we had an even result, uhm, which I think is important, but
I think its not important how the two of us fight. I still think, I
havent been in politics long enough to lose that illusion for good,
that the fight is personified by the two of us. And thats why I say
to Mr Paroubek, lets go to Prague. Come on, lets personify that
fight for the election too. I think that is important. But actually, it
still should be the fight of party programs, ideas, political subjects,
and not the fight of some individuals who will try to measure their
strength through their statements. I, I dont allow myself to make
such statements, I will never do that and I wont play this game.
host: well, mr chairman (Topolanek), I think that if I looked through

my files here, Id find some really nice statements coming from
you.
Topolanek: I have to say that your search would be quite tedious if you tried
to find me attacking the prime minister (Paroubek) directly. I
[really]
host: [so]
Topolanek: try to avoid that.

(Sedmika, 28.8.2005 (Corpus ujc.dialogy.cz))
Later, after Topolaneks turn, the host broaches the issue of (verbal) fighting again
by telling Topolanek that what the latter said is not true and that he had made
some verbally aggressive statements himself.
Furthermore, this example shows an interesting pattern of dealing with irony
on Topolaneks part. He positions himself as being above having to attack his opponent verbally and thus not having to use irony and similar devices. That means
he attacks his opponent by not attacking him. Since the corpus provides videos,
the analysis of Topolaneks behavior can start on the level of non-verbal communication. When his opponent (Paroubek) talks about Topolaneks ad, his own ironic
letter to Topolanek, and the latter calling the newspaper insolent for evaluating
the result of a previous debate as being in Paroubeks favor, Topolanek smiles, but
seems annoyed at the mention of the ad. His laughing at Paroubeks comments
shows that he has detected the irony and that he is being a good sport. Later,
when the host asks him to react to Paroubeks statement, he smiles and simply
says that hes just not that interested in answering (ani ne). The phrasing, using a
simple ani ne (approx. not really), shows again that he is above fighting; he takes
what was said lightly. He then goes on to explain why, saying that he believes
that a partys campaign should be about political ideas and not about two politicians fighting and showing who of them is cleverer verbally. He rejects the ironic
approach and at the same time enhances his own and damages the opponents
positive face, because he says that he believes in true politics, while the other one
personalizes it and does not represent actual causes and ideas. This is evidenced
by the repeated use of the word evidence (which comes up again later in the debate in a similar context). Topolaneks choice to not attack his opponent and to
not use irony or similar strategies goes beyond the scope of the debate. While he
pretends that he does not attack his political adversaries verbally, he is indirectly
of course doing just that and more: he attacks the whole party. He is implying that
Paroubek has to resort to personal attacks, which means that he does not have
enough sound ideas to be able to hold his own in a factual political debate, while
Topolanek can do just that and still come out the winner in the end. However, the
host tries to break this argumentation up by saying that he remembers Topolanek
formulating some insulting statements himself.
While Topolanek tries to position himself in the role of the honest and factabiding speaker, his opponent, Paroubek, chooses to assume a more fight-ready
role in the beginning. His ironic remark may be directed at the host, but the ironyvictim is his political opponent; he implies that Topolanek is a bad debater who
cannot stand to be the weaker one and therefore, the host will have to work hard
to control the discussion. This is a classical use of irony to damage somebodys
positive face. Paroubek puts himself into the position of the quasi-leader who
even has to give the host advice. Interestingly enough, the host seems to oppose
both speakers to some degree. It seems that this is his way of leveling the field, on
one hand, but on the other hand, we might ask whether the host himself has to
negotiate his role as well. While the fact that he is the host would suggest that his
role is predetermined, this seems not to be the case. The battle to be the strongest
speaker on the floor can, at times at least, step beyond the line of the hosts role.
Examples like (3) and the beginning of (6) are not rare; politicians struggling to
make their point often ignore what the host is saying and that he might be trying
to lead the talk into a different direction or limit their speaking time.
6. Conclusions
The discussion of these six case studies has shown that it is important to take a
broader context into account when assessing the impact of irony on politeness
management. The examples demonstrated that reasons for chosing an ironic
strategy vary. One reason might be the wish to present oneself as the stronger,
more controlling participant in a debate. Another reason reflects the opposite
intention, namely the attempt to protect ones own negative face from the attacks
of an opponent. In a similar way, the possible reactions to irony can also vary and
may depend on various factors. To be able to assess a situation properly, it is also
relevant to weigh the following criteria: the setting of the talk show, the attitude of
the host, the individual politicians styles of self-presentation and the identity they
have established for themselves (or that has been partially established for them
by their party or the public) over the course of their career, and over the course
of a particular show some politicians might choose to be more or less aggressive during a specific show than they are normally. It should also be kept in mind
that certain politicians have more talent for show conversations and/or are more
inclined towards displaying their confrontainment abilities than others.
The interpretation of ironic utterances in the context of political talk shows

usually does not require enormous intellectual capacities. One reason for this is
the multitude of contextual sources (Yus 2000) available to the viewer and the
participants in the show: the viewers are familiar with the issues, the background
and even the politicians style of speech, all of which make it easier for them to
decode the ironic utterances. Another reason is the fact that politicians participating in talk shows do not have just their political issues in mind while discussing
them, but are always aware of the fact that they must try and appeal to a larger
part of the population in order to be re-elected or elected into another office.
Their use of irony differs from everyday use in that it does not only involve the
conversational partner, but is equally aimed at the viewers and thus must be easily
decodable bythem.
References
Attardo, Salvatore. 2000. Irony as relevant inappropriateness. Journal of Pragmatics 32(6):
793826.
Barbe, Katherina. 1995. Irony in Context. Amsterdam/Philadelphia: John Benjamins.
Brown, Penelope and Levinson, Stephen C. 1978/1987. Politeness: Some Universals in Language
Usage. Cambridge: Cambridge University Press.
Busch, Brigitta. 2006. Media, politics and discourse: Interactions. In Encyclopedia of Language
and Linguistics 2nd Edition, Keith Brown (ed.), 609616. Oxford: Elsevier Publishers.
Clayman, Steven and Heritage, John. 2002. The News Interview. Journalist and Public Figures on
the Air. Cambridge: University Press.
Eisterhold, Jodi, Attardo, Salvatore, and Boxer, Diana. 2006. Reactions to irony in discourse:
evidence for the least disruption principle. Journal of Pragmatics 38(8): 12391256.
Fairclough, Norman. 1995. Media Discourse. London: Edward Arnold.
Fairclough, Norman. 1998. Political Discourse in the Media: An Analytical Framework. In
Approaches to Media Discourse, Allan Bell and Peter Garrett (eds.), 142162. Oxford:
Blackwell.
Fairclough, Norman. 2006. Genres in Political Discourse. In Encyclopedia of Language and
Linguistics 2nd Edition, Keith Brown (ed.), 3228. Oxford: Elsevier Publishers.
Gibbs, Raymond W. and OBrien, Jennifer. 1991. Psychological aspects of irony understanding. Journal of Pragmatics 16: 523530.
Gibbs, Raymond W., OBrien, Jennifer, and Doolittle, Shelley. 1995. Inferring meanings that
are not intended: Speakers intentions and irony comprehension. Discourse Processes 20
(2): 187203.
Gibbs, Raymond W. 2000. Irony in talk among friends. Metaphor and Symbol 15 (1/2): 527.
Giora, Rachel. 1995. On irony and negation. Discourse Processes 19 (2): 239264.
Giora, Rachel. 1997. Understanding figurative and literal language: The graded salience hypothesis. Cognitive Linguistics 8 (3): 183206.
Giora, Rachel. 1999. On the priority of salient meanings: Studies of literal and figurative language. Journal of Pragmatics 31 (7): 919929.
Grice, H. Paul. 1975. Logic and conversation. In Syntax and Semantics, Speech Acts, vol. 3,
Peter Cole and John Morgan (eds.), 4158. New York: Academic Press.
Gotsbachner, Emo. 2008. Durchsetzung von Deutungsrahmen in politischen Fernsehdiskussionen. Gesprchsforschung Online-Zeitschrift zur verbalen Interaktion 9: 269299.
Groeben, Norbert and Scheele, Brigitte. 1984. Produktion und Rezeption von Ironie. Bd. 1: Pragmalinguistische Beschreibung und psycholinguistische Erklrungshypothesen. Tbingen:
Gunter Narr.
Halvorsen, Per-Kristian. 1976. Semantics of Irony and Sarcastic Utterances. Working Papers
in Linguistics 7: 85103.
Hartung, Martin. 1998. Ironie in der Alltagssprache. Eine gesprchsanalytische Untersuchung.
Opladen: Westdeutscher Verlag.
Haverkate, Henk. 1990. A speech act analysis of irony. Journal of Pragmatics 14 (1): 77109.
Hess-Lttich, Ernest W. B. 2007. (Pseudo-)argumentation in TV-debates. Journal of Pragmatics 39 (8): 13601370.
Hutcheon, Linda. 1994. Ironys Edge. The Theory and Politics of Irony. London: Routledge.
Kallmeyer, Werner and Schmitt, Reinhold. 1996. Forcieren oder: Die verschrfte Gangart zur
Analyse von Kooperationsformen im Gesprch. In Gesprchsrhetorik. Rhetorische Verfahren im Gesprchsprozess, Werner Kallmeyer (ed.), 19118. Tbingen: Narr.
Kotthoff, Helga. 2003. Responding to Irony in Different Contexts: on Cognition in Conversation. Journal of Pragmatics 35 (9): 13871411.
Khn, Peter. 1983. Der parlamentarische Zwischenruf als mehrfachadressierte Sprachhandlung. In Sprache, Diskurs und Text [Akten des 17. Linguistischen Kolloquiums Brssel
1982; vol. 1], Ren Jongen et al. (ed.), 239251. Niemeyer, Tbingen.
Khn, Peter. 1995. Mehrfachadressierung. Niemeyer: Tbingen.
Lapp, Edgar. 1992. Linguistik der Ironie. Tbingen: Gunter Narr.
Leech, Geoffrey. 1983. Principles of Pragmatics. London, New York: Longman.
Luginbhl, Martin, 1999. Gewalt im Gesprch. Verbale Gewalt in politischen Fernsehdiskussionen am Beispiel der Arena. Bern, Berlin, Frankfurt, New York, Paris, Wien: Lang.
Luginbhl, Martin. 2007. Conversational violence in political TV debates: Forms and functions. Journal of Pragmatics 39: 13711387.
Maara, Jekaterina. 2011. Ironie, Freund oder Feind? Zur Auswirkung von Ironie auf das Hflichkeitsmanagement. To appear in Proceedings m*OST 2010. Welt der Slaven.
Norrick, Neal R. 1993. Conversational Joking. Humor in Everyday Talk. Bloomington: Indiana
University Press.
Nuolijrvi, Pirkko and Tiittula, Liisa. 2011. Irony in Political Television Debates. Journal of
Pexman, Penny M. and Zvaigzne, Meghan T. 2004. Does irony go better with friends? Metaphor and Symbol 19 (2): 143163.
Searle, John R. 1979. Expression and Meaning. Studies in the Theory of Speech Acts. Cambridge:
Simon-Vandenbergen, Anne-Marie, White, Peter R. R., and Aijmer, Karin. 2007. Presupposition and taking-for-granted in mass communicated political argument. An illustration
from British, Flemish and Swedish political colloquy. In Political Discourse in the Media, Anita Fetzer and Gerda-Eva Lauerbach (eds.), 3174. Amsterdam/Philadelphia: John
Benjamins.
Sperber, Dan. 1984. Verbal irony: pretense or echoic mention? Journal of Experimental Psychology, General 113: 130136.
Sperber, Dan and Wilson, Deirdre. 1981. Irony and the use-mention distinction. In Radical
Pragmatics, Peter Cole (ed.), 295318. New York: Academic Press.
Sperber, Dan and Wilson, Deidre. 1986. Relevance: Communication and Cognition. Oxford:
Blackwell.
Sperber, Dan and Wilson, Deirdre. 1998. Irony and relevance: a reply to Seto, Hamamoto and
Yamanashi. In Relevance Theory: Applications and Implications, Robyn Carston and Seiji
Uchida (eds.), 283293. Amsterdam/Philadelphia: John Benjamins.
Utsumi, Akira. 2000. Verbal irony as implicit display of ironic environment: Distinguishing
ironic utterances from nonirony. Journal of Pragmatics 32: 17771806.
Weinrich, Harald. 1966. Linguistik der Lge. Heidelberg: Schneider.
Wilson, Deirdre and Sperber, Dan. 1992. On verbal irony. Lingua 87 (12): 5376.
Yus, Francisco. 2000. On reaching the intended ironic interpretation. International Journal of
Communication 10 (12): 2778.
Appendix: Transcription conventions

Since the excerpts were taken from the corpus ujc.dialogy.cz, the transcribing conventions are
also those of that corpus. However, the original line numbering was removed since it was not
useful for this paper and the initials that mark the speakers in the corpus were written out as
full names to make the excerpts easier to follow.
[nevm]
[v t]
zase
a:le:
e ee eee eh ehm
em hm mhm ehe e-e
(.)
(..)
()
?
,
.
mon-
<uhodil>
(ale)
( )
((smch))
Two speakers talking simultaneously (overlap)

Syllable/word emphasized
Sound elongated
Sounds of hesitation and response
Short pause
Longer pause
Long pause
High-rising intonation
Low-rising intonation (slight rise or slight rise and fall)
Falling intonation
Unfinished word
Uttered while laughing
Presumed, but not fully comprehensible expression
Incomprehensible expression
Transcribers comment
Additions:
Square brackets [] indicate that the excerpt has been shortened for this paper.
Italic font was added by the author to highlight the relevant ironic sections.
Parliamentary communication
The case of the Russian Gosduma
Daniel Weiss
This study attempts to capture the characteristics of parliamentary debates in

the Russian State Duma (Gosduma) according to the following criteria: quality of data (e.g. editorial specifics of the transcripts), turn-taking system, forms
of address, Internet communication and TV coverage. The comparison of the
Russian findings with the British Hansard Report reveals a striking contrast: the
Russian data manifests a far broader range of variation in many respects, which
is obviously due to much more liberal or even completely lacking regulations.
It seems to be a fair assumption that these diverging attitudes can be partly explained by the different prestige of both institutions in their respective societies:
in particular, as can be seen in numerous examples quoted, the Gosduma resents the lack of attention, if not the disrespect it experiences from government
representatives, television, as well as the broad public.
Keywords: parliamentary discourse, turn-taking system, forms of address,
TV coverage, Internet communication, multi-layered communication
1. Introduction
European parliamentary discourse has already been the object of numerous individual and comparative studies, to quote but a few: Chilton (2004), Bayley (2004)
and Ilie (2010). Besides the British House of Commons, which for many reasons attracts the interest of most researchers and frequently serves as a basis for
comparison, we find for example papers devoted to the Italian, Spanish, French,
Swedish, Greek, Rumanian, German, Czech or Polish lower chambers (due to its
. For a comprehensive bibliography (with the exception of several German titles), see Ilie
(2010:5).
. Even Belorussian parliamentary speech has already been investigated in comparison to the
Polish Sejm, cf. uravleva and Potapova (2010).
214 Daniel Weiss
less powerful position, the second chamber (Senate, Bundesrat, etc.) is usually
neglected in these studies). Their Russian counterpart, however, the State Duma,
has with two notable exceptions (Sivenkova 2009, 2013) never been targeted
within such a comparative approach, nor has it been individually portrayed within
any larger individual analysis so far. The present study aims at partially filling this
gap by capturing the following aspects: the quality of the stenographic transcripts
available online (section iii), the turn-taking system (iv), forms of address (v),
the overall grid of multi-addressed and multi-layered communicative acts (vi),
and TV coverage (vii). The analysis in the sections iiiv is based on a comparison
with analogous data from the British Hansard Report. The period covered by this
paper encompasses five sessions from the end of the penultimate period (autumn
2007) to the end of the last period (autumn 2011), with the exception of section
(vii), which refers to recent developments. Said sections are preceded by a global
characterisation of the political background in Russia (ii).
2. Political background
From a Western perspective, present-day Russia is far from being an ideal democracy. This holds particularly true for the so-called directed democracy (upravlja
emaja demokratija) established in Russia during the last decade, whose specific
traits have not failed to also affect parliamentary discourse. By now, the latter has
to be examined against the background of a heavily disorderd system of checks
and balances, in which the Kremlin takes it all. To recall but a few milestones: the
concentration of power is noticable on the federal level where, since 2004, the 89
governors and regional presidents are no longer elected by the people, but instead
appointed (and dismissed) by the president; the monopolisation of public opinion has long reached all nationwide TV channels, and many influential papers are
likewise controlled directly or indirectly (through oligarchs connected with the
Kremlin) by the government; and finally, there is no independent justice system.
Since all these facts are widely known, there is no need to elaborate on them any
further; for a detailed overview of the drawbacks and shortcomings of Russian
democracy in its present condition see, for instance, Fleischmann (2010:7253),
for a general overview of political post-Soviet research Laitin (2003). It should,
. The early manual by Graudina and irjaev (1994) is mainly concerned with orthoepic
instructions for deputies, and Sakhno (2011) deals with the etymology of the designation
Duma.
. On the 25 of April 2012, the reintroduction of governors elections by the people (a law
initiated by Putin) was accepted by the Duma.
The case of the Russian Gosduma 215
however, be emphasised that the definition of democracy is a controversial issue

in political science, and it is by no means obvious that it should encompass freedoms of the press, association and assembly in addition to the competitive struggle for the peoples vote (P. Schlumpeter, quoted after Anderson 2010). From this
point of view, the Russian system does not present a case of derailed or even
failed democracy but simply reflects the very specific historical conditions from
which it originated (Anderson 2010).
As for the position of the Parliament of the Russian Federation, the Gosduma, it should be borne in mind that the Russian constitution was shaped as a
semi-presidential democracy, copied faithfully from that of France (Anderson
2010:405). This means that right from the beginning the position of the Duma
has been weaker than that of most other European parliaments. Its influence, as
well as its prestige, were additionally affected by the new electoral law and the
following elections in 2007. Not only did the government party Edinaja Rossija take the overwhelming majority of all seats (64.3%) with the three remaining parties dividing the rest between themselves (KP: 11,57%, Liberaldemokraty:
8.14%, Spravedlivaja Rossija: 7.74%), but the conditions for admission were also
completely unfair, since for a party to participate in the elections, a minimum of
50.000 registered members was required. Moreover, no party voting block (vybornyj blok) was allowed, and the minimal percentage to be achieved by a single
party was 7% of all votes. To make things worse, participation itself turned out to
be unfair, since one third of all 315 elected members of Edinaja Rossija (mostly
popular governors, ministers, mayors, etc.) subsequently gave their seats back in
favor of other party members. All this boils down to the conclusion that there
never was a Russian parliament so weak as the present one. The deputies themselves seem to be well aware of this, and members of oppositional groups voice
their deep concern, if not outrage, about it:
(1) . ., , . -
, ,
, ,
. , ,
:
,
. , ,
, 2 ,
.
(11/16 2007)
. All emphasis in citations mine, D. W.
216 Daniel Weiss
Kurjanovi N.V. The duma has now turned into an exclusively decorative
organ that decides nothing, has no impact on whatsoever and on the contrary
only obstructs the normal evolution of our society. Therefore there are good
reasons to add amendments to our Constitution: besides the introduction
of a mention of the Russian people, the point about the State Duma as a an
institution of power should be excluded. And this whole orgy that is now going
on on the eve of the State Duma elections only corroborates what I have said.
In a similar vein, another oppositional deputy calls the members of the Duma
deputaty-roboty robot deputies (see example (12) below), and a third one considers the imminent presidential elections a put-up affair (example (11)). Such voices
of frustration can be easily explained since the opposition sees no chance of assuming power in the next elections. But there are other reasons why frustration
has by no means decreased during the current period: for instance, the deputies
(not only from oppositional parties) keep complaining about the irreverance towards the parliament shown by ministers who refuse to appear at the Question
Time or do not give adequate answers to questions posed by deputies (3/23/2011),
and others denounce insufficient TV covering of Duma debates (see example (19)
and (20) below).
On the other hand, the Duma itself undertakes little to improve its prestige
among the electorate: its widespread absenteeism, the lack of attention during ongoing debates (cf. www.youtube.com/watch?v=QXFKVUfXyjw; 9/10/2012) and in
particular verbal injuries employed abundantly by certain deputies as well as real
brawls (mostly with irinovskij involved, cf. www.youtubcom/watch?v=mZGaaq
H2o6I; 9/10/2012) all these highlights are broadcast and sarcastically comm
ented on in the state TV to the amusement of the broad public. Absenteeism is
facilitated by the possibility of voting for absent deputies: as is shown in http://
www.youtube.com/watch?v=2DvP8Kvdrmw, one deputy recently managed to
push the voting buttons for nine colleagues within 20 seconds. All this is hardly
able to evoke the sympathy of the TV audience, cf. the voices from the Internet
community quoted in the last section of this paper.
3. The data
The transcripts of the Duma debates published on the Internet are the official
verbatim record, debates being reported in direct speech. But on what kind of
input data are they based? Are these stenograms carried out by a team of shorthand typists that work in shifts of 510 minutes, as this was the case in the British
Parliament during the period described by Slembrouck (1992:104)? The title of
the official Gosduma website wbase.duma.gov.ru/steno/nph-sdb.exe still lists these
transcripts as stenogrammy zasedanija shorthand notes of the session from,

which would support the stenotypist hypothesis, but the individual links are now
called transcript.duma.gov.ru/node/, which might suggest that we are dealing with
retroactively rendered transcripts made on the basis of audio-tapes. The latter
hypothesis was, however, denied by the assistant of a Duma deputy: stenotypists
still do their duty in the Gosduma, much as their colleagues in other Europan
parliaments. Deputies are offered the chance to review (which probably does not
mean to proofread) their own statements before publication (cf. article 39, paragraph 2 of the Houses regulations [reglament]). Unlike in the British Parliament, they obviously are allowed to read prepared speeches which they probably
hand over to the chair afterwards. There still remain some open questions: How
many stenotypists work simultaneously during one session, and where are they
seated? What linguistic or other guidelines are available for the editors? Are there
any instructions on how to proceed in cases of verbally abusive behaviour by individual deputies, such as cursing, calling names or offending other persons (in
other words, offenses against article 46 of the Regulations)? What is more, unlike
Slembrouck (1992), Ornatowski (2010) and Madzharova Bruteig (2010) we did
not have access to the primary data, viz. the spoken version of the debates in a
video archive, the only sources accessible being fragments broadcast on TV and
Internet videos of the most recent sessions (see ultimate section of this paper).
What we can judge, however, is the final result of the editing process. The
main question to be asked is to what extent the assumption that this is really a
word-for-word-report is justified. In other words: how much evidence do we find
for a systematic filtering out of spokenness and the elimination of grammatical
or lexical mistakes? My results, based on the analysis of 5 different sessions that
took place in 2007, 2008, 2009 and 2011, are most revealing when compared with
the British Hansard Report, i.e. the verbatim record of the debates of both the
House of Commons and the House of Lords such as described in Slembrouck
(1992). Tables 1 and 2 summarize my findings based on the minutes of the Gosduma, contrasted with Slembroucks description.
As can be seen, the Hansard Report is closer to the original in just one respect: it also records hecklings and other backchannel behaviour, which is never
. By all evidence, they are not located just in front of the speakers desk as for instance in the
Polish Sejm.
. Hecklings are also registered in Polish and Czech Parliamentary minutes. Moreover, according to Ornatowski (2010:226), the Polish Record indicates occurrences of applause, laughter, backchanneling, commotion in the chamber, hitting the pulpit, demonstrative leaving
and entering, and other verbal and non-verbal behaviours of MPs (including their timing and
intensity).
218 Daniel Weiss
Table 1. Traces of spontaneous speech: + = preserved, = eliminated

Half-pronounced words
Gaps / Hesitation markers (uh)
Overlaps
Tag questions
Incomplete utterances
False starts
Redundant (meaningless) repetitions
Reformulations
RU
UK
+
+
+
+
+
Table 2. Linguistic (stylistic, grammatical, lexical) editing and backchannel signals

Colloquial expressions / constructions
Morphological / syntactic mistakes
Stylistic akwardness
Standardised forms of address
Unified obligatory use of honorifics
Backbench discourse: laughter, cheering
Hecklings recorded
RU
UK
+
+
+
+
+
partially
+
registered in the Gosduma minutes. Most likely, the sound equipment in the auditory does not catch these signals because of missing microphones; the Chairs
answers to these hecklings, on the other hand, are always recorded. In all other
respects, the Gosduma transcripts deserve the label word-for-word report much
more than the Hansard report.
Evidence for a rather careless editing process simply abounds in my Russian
data. For instance, hesitations, repetitions and emotional expressions are preserved. The following examples illustrate the major types referred to in Tables 1
and 2:
(2) , ,

false start (11/16/2007)
All this leads to For example, yesterday the Deputy of the State Duma
Golvatjuk Andrej Mixajlovi was arrested
. In contrast to this, these traces of spontaneous speech are edited out in Czech minutes
(Madzharova Bruteig 2010:268).
(3) - , ,
. ,
, ,

colloquial conjunction + filler (11/16/2007)
Well, we agree that all should work together, but in our laws there is no mention of the Paraolympic Committee of Russia. Who will, generally speaking,
be in charge of the organisation of the Paralympics
(4) , , .
stylistic akwardness (2/6/2008)
I wish there would be reacting by the direction of the Chamber
(5) , , , , -.
[],
, ,
,

meaningless fillers (2/6/2008)
I think that the effects of this will affect not only aspects of foreign affairs,
so to speak, but also internal politics. Therefore I would like [] that they
conduct a poll among the departments involved on what will happen if we
recognize the independence of Abkhasia and South Ossetia in this vein.
(6) , .

colloquial verb omission (6/25/2007)
Everybody [is talking] about prices, but I [am talking] about revenues10
(7) , ,

syntactic mistake (crossover of two different constructions)11 (3/23/2011)
Troops located on Sudanese territory plan as a goal the seizure of power in
Tripolis
. As for verb omission in colloquial Russian, see e.g. Maara (2011) and Weiss (2011). Not
surprisingly, the elided elements are verbs of speech, cf. also:
So lets now [talk] about revenues (quoted after Sivenkova 2009:126).
10. It should be noted that there is a technical expression belonging to parliamentary slang
that likewise contains a zero, cf. ? Lit. Why from voice?. This seems to be used
when the point in question is being introduced spontaneously, i.e. without prior mention in
written form (e.g. not on the list of items to be treated during the session).
11. Note that this cannot be explained as a case of so-called superimposition (naloenie)
known from colloquial Russian (cf. Lapteva 1976:339351).
220 Daniel Weiss
Among grammatical deviations, we also find irregular word order typical of spontaneous speech as in fakt vopijuij revolting fact, deputata dejstvujuego of the
acting deputy, and wrong prepositions, e.g. do nekotoroj stepeni to a certain degree instead of v nekotoroj stepeni (all 2/6/2008). As for markers of spokenness,
many speeches contain colloquial idioms such as lapu na ui veaet leads [sb] up
the garden path12 and signals of spontaneous dialogicity, e.g. Sluajte, nu xvatit
izdevatsja nad zdravym smyslom! Listen, weve had enough of this mockery of
common sense (3/23/2011). No English equivalent of the particularities quoted
would be admissible in the Hansard Report. On the other hand, it goes without
saying that the Gosduma transcripts have nothing in common with a real scientific transcription of spoken language as it would be required for linguistic purposes
(see the detailed guidelines presented in Kibrik and Podlesskaja 2009).
To sum up, Russian Gosduma transcripts manifest only minimal traces of an
editorial processing. As a result, we find no visible grammatical, lexical or stylistic corrections and adjustments, and the semi-spontaneous character of the
oral presentation is preserved. The Hansard Report, on the other hand, reveals
a comprehensive and careful editorial processing: it eliminates all colloquialisms
and syntactic gaps and adjusts slips occurring in individual votes, in other words:
it translates the original into formal written English. As an anonymous informant of Slembrouck (1992:108) puts it: Hansard is essentially a verbatim record,
but speeches are automatically corrected by the reporters for English style and
grammar and are also made conform with House of Commons style.13 One important aspect is the removal of redundancy deemed irrelevant by the reporter
(Slembrouck 1992:110); in particular, this holds true for markers of the speakers
personal belief (for the latter as a redundancy marker see also Berlin 2011). As for
the overall relation between redundancy and relevance and their different assessment in oral and written discourse, see mejrkov (2011).
The reasons for the striking divergence between the British and the Russian
reports is certainly not connected with the different prestige of the two standard
languages in their respective societies: the Russian educational system does not
12. Similar colloquialisms may be found in the minutes of Czech parliamentary debates, cf.
Madzharova Bruteig (2010:282f.). Surprisingly enough, even elements of common Czech
(obecn etina), the once stigmatised Bohemian spoken interdialectal variety, occur. On the
other hand, the same author (ibid.:283) reports the elimination of repetition of pronominal
forms and phonetic markers typical of common Czech and some changes in word-order revealed by a comparison of the official record with the audio version.
13. In a similar vein, Chilton (2004:94) states that Hansards supposedly verbatim transcription () in fact corrects the form of interrogatives (and other features) to produce an idealised
model of the session that is supposed to have taken place.
ascribe less importance to existing linguistic norms than does education in the
UK. What really seems to matter, though, is the different prestige of both institutions: the British parliament can boast century-long traditions and a powerful
position in the political system, whereas the Russian Duma is of recent origin,
has become pretty powerless under the conditions of directed democracy and
enjoys a doubtful reputation among the population, to say the least. This may
be one of the reasons why it is less concerned with clarity, brevity, and processing ease in its official record.14 On the other hand, it should be pointed out that
the Hansard Report constitutes the utmost extreme in stylistic editing,15 with its
Polish and Czech equivalents located somewhere in between the Russian and the
British transcripts: for instance, the Polish stenographic record eliminates false
starts, incomplete utterances, redundant elements and grammatical mistakes, but
leaves colloquialisms unchanged and contains certain stylistic devations.
4. Turn-taking system: The role of the Chair

As in other parliaments, turn-taking in the Gosduma is strictly regulated, the
central role being performed by the Chair, who distributes the speech rights to
individual deputies. Most often, a successful exchange displays a three- or even
quadripartite structure, since the next speaker usually thanks the chairman and
optionally addresses him before addressing the House and delivering his/her
vote, cf.:
(8) : .
. ., . ,
. []
(2/6/2008)
Chairing person: Deputy Xajrullin.
Xajrullin A.N., Edinaja Rossija. Thank you, Oleg Viktorovi.
Moreover, the Chair optionally may give the floor by saying please:
(9) : . . ,
.
14. At any rate, it seems unlikely that mere historical stenographic traditions (cf. Jurkovskij
1949) should overrule the concern for a careful preparation of the transcripts for publication;
after all, similar historical traditions exist in other countries as well.
15. It should be pointed out that according to prescriptive handbooks even questions with
rhetorical or ironical expressions are not in order (Chilton 2004:95). The regulations of the
Russian State Duma do not contain any comparable constraints.
222 Daniel Weiss
. ., . . ,

(2/6/ 2008)
Chairing person: OK. Distribute [them] in the auditory. Aparina, please.
Aparina A. V., Communist Party. Thank you. Dear colleagues, before the elections and New Year
As these examples show, the range of optional components amounts to four different formulae. This already variable structure is extended in cases like the following, where one deputy speaks on behalf of another:
(10) . .
. ., .
. , !

(2/6/ 2008)
Chairing person. Deputy Kolomejcev Viktor Andreevi.
Xaritonov N. M., Communist Party. Xaritonov according a note by Kolo
mejcev. Dear Oleg Viktorovi, dear colleagues! My petition to the Chair of the
Legislation Committee Kraeninnikov is related to
In some sessions we find a host of unsuccessful exchanges. These failures are

caused by a rigorous time limitation marked by the switching off of the microphone, cf.
(11) ,
, ,

( .)
. .
[]
(2/6/ 2008)
the whole country and the whole world already guess who will be president in
our country, but let us at least create the semblance of a non-impeded, clean
and democratic election (microphone switched off)
Chairing person. Your proposal is clear. Deputy Obuxov.
(12) . ., , -
: ,
.
( .)
. .

(2/6/ 2008)
And I think that nowadays a robot deputy is very convenient for the government: it should just sit here, be silent and press [voting] buttons. (microphone
switched off)
Chairing person. Deputy Ivanov Sergej Vladimirovi.
Whereas in the first example the chair acknowledges the message delivered by
the interrupted speaker, in the second case he simply gives the floor to the next
speaker. In view of the evident sarcasm of both votes delivered by oppositional
speakers, one might be tempted to interpret the breakup as an act of punishment.
However, this turns out to be a false assumption since other statements with completely innocent messages are interrupted in the same way.
Sometimes the chair asks the sound operator for an extension and thus allows
the previous speaker to deliver the whole message, cf.:
(13) ,
,
( .)
. , . ..

Unfortunately, while clearing the aftermath of the earthquake several questions (microphone switched off)
Chairing person. Add some time, please. ..Under these circumstances all these
spendings
Sometimes we witness a real battle for the floor, with the speaker lamenting the
chairs unfair treatment and the latter adding to the overall stress by pressing the
current speaker to hurry up:
(14) . ., , ,
, - . , . .
, - , - ,
.
. ,
.
. .
(6/25/2007)
irinovskij V. V. Boris Vjaeslavovi, if I will not make it in time, extend
my speech time as well, or else you seem to be extending it quite selectively. I
think that one should not extend anybodys speech time. One minute is too little
for all of us. But this selection, when somebody gets an extension and another
person doesnt, is not very nice.
Chairing person: 20 seconds have already elapsed, Vladimir Volfovi,
irinovskij V. V., So I want to
In the case at hand, other deputies join the dispute, cf. ,

! Boris Vjaeslavovi, I ask you to support irinovskijs
224 Daniel Weiss
proposal to add time for everybody. You didnt allow me to finish my assignment to
the committee about cars for veterans! In this way, much precious time is lost on
arguing about the correct handling of the turn-taking system. This mechanic limit
of one minute during Question Time is dictated by the official regulations of the
Gosduma (reglament GD, razdel II, glava 5, statja 41, punkt 14) and therefore not
announced at the beginning of the session.16 On other occasions, however, the
duration of the individual statements on certain topics is subject to negotiation,
i.e. the chairman Gryzlov suggests a time limit the deputies have to agree with
(3/23/2011). The microphone guillotine is also used in such situations without
premonition.
Another type of faulty exchanges occurs when the deputy in question is absent, cf. . . . (2/6/2008) Chair
ing person: Deputy Gostev. Not present. Technical malfunction may also lead to
failure, cf. ,
, , . (2/6/2008) I have been informed that deputy Pepeljaeva pressed the button
for registration, but the button did not work, so I give her the floor. Besides this,
there is an illegal type of exchange represented by hecklings which are answered
by the chair, e.g. . ( .) .
? ! (6/25/2007) From the audience (not
audible). Chairing person. Why not? We will be able to ask this question!17
And finally, electronic exchanges such as the registration of requests to speak or
the results of a poll are completely standardized. The latter type is always initiated by the chairs stereotypical double formula , ,
. , , . Please turn on the voting
mode. Please show the results. (next, said results appear on the screen), after
which s/he summarizes by or Accepted / Rejected.
5. Forms of address
In comparison with the British Parliament, forms of address used in the Gosduma
exhibit a much smaller extent of regulation, or else: a much broader range of
variation. Two general remarks should be made: First, the 2nd pers.sg. ty which
16. In the Polish stenographic records, time limits are communicated in the beginning when
the chair explains the details of the agenda.
17. In her comparative analysis of Italian and British parliamentary debates, Bevitori (2004) has
shown that different turn taking regulations may lead to different national cultures of interruption. In my data, deputies interrupting the speech of their colleagues are seldom successful.
was quite common among members of the Communist Party in Soviet Union is
no longer customary,18 and second, the chairperson uses a different set of address
forms than all other MPs. The marker of the chairs position is the optional title
Deputat deputy, which is followed either by the mere last name or the full threepartite set composed of last name + first name + patronym. Thus, we find e.g. Aparina (cf. example (9)); this variant would be rude in everyday use, it was, however,
the official form of address at school in Soviet times), deputat Xajrullin, deputat
Pletnva (cf. example (12)), Ostanina Nina Aleksandrovna or Deputat Smolin Oleg
Nikolaevi; the latter variant with its inversion of first and last name represents the
passport version and is typical of official forms. The bipartite variant name+
patronym as in Vladimir Vollfovi (cf. example (14)) is often employed in ongoing exchanges. This sums up to 5 different patterns of addressing, two of them
having a marked institutional character (inversion and mere last name), which
all seem to be more or less interchangeable, without regard to negative politeness
and political preferences; whether personal factors, such as the degree of familiarity or reverence, gender, etc., play a role remains open.
As for the deputies, they may, as mentioned in the previous section, address the chaiperson or not. If they do, they usually use the honorific dear, cf.
Uvaaemyj Oleg Viktorovi in example (9). Moreover, they can optionally address
their fellow deputies by uvaaemye kollegi or dorogie druzja. And finally, they may
direct their speech to individual deputies, addressing them in the 3rd person as in
U menja toe budet vopros k kollege Kraeninnikovu I will also have a question
for colleague Kraeninnikov, vo-pervyx, to ne sootvetstvuet dejstvitelnosti to,
to Olga Georgievna govorila first, what Olga Georgievna said does not correspond to reality, Uvaaemye kolegi, ja xotel by podderat Pletnvu Tamaru Vasil
evnu Dear colleagues, I would like to support Tamara Vasilevna Pletnva. As
can be seen, the deputies have the same choice between the single, twofold and
threefold variants as the chair; in their speeches, however, we also find the less official word order first name + patronym + last name, cf. Xou obratitsja k Pavlu
Vladimiroviu Kraeninnikovu I want to direct this to Vladimir Kraeninnikov.
And finally, the third person can subsequently be replaced by direct addressing
in the second person, for instance when the vote contains a rebuke of a previous
utterance made by the addressee, cf.
18. Contrary to this, ty may be found in hecklings directed to other deputies in the Polish Sejm;
it then does not mark familiarity, but offense. Moreover, some deputies use the intermediate
variant with the polite address form Pan/Pani and the verb in 2nd pers.sg. in their votes, which
is considered much less polite than Pan/Pani with 3rd pers.
226 Daniel Weiss
(15) ,
, . ,
, . .
(2/8/2008)
And as for Nikolaj Fedorovi Rjabov Please, dont criticise us, start with
yourself. You should not say that the button did not work, the buttons work
fine with all of us. If you did not register in time just say so.
A significant detail is the use of the honorific tovari comrade when referring to
(not addressing directly) members of the communist party. This variant seems to
be mainly restricted to ironical contexts, cf. to kasaetsja vystuplenija tovaria
Fedotkina, to prede, em kritikovat knigu, nado ee proitat As for comrade
Fedotkins vote, he should first read the book before he starts critizing it.
A brief look at the forms of address admissible in the British Parliament reveals a striking contrast: honorifics are mandatory, no variation of names is possible, and deputies have to address each other in an indirect mode:
Briefly sketched, the use of the second person pronoun is restricted to exchanges
between an MP and the chair (Mr/Mrs (Deputy) Speaker, but ruled out in cases
where other MPs are addressed (the forms used then are either the honourable
Lady/Gentleman, honourable members or My honorable friend.19 Pronominalisation is in the third person, the assumption being that MP address each others
(Slembrouck 1992:113115)
in the Chamber through the chair.20
If an MP slips into the second person or refers to the Prime Minister by her name,
Hansard adjusts the mistake. Thus, the House of Commons symbolically proclaims its autonomy from other powers (Slembrouck 1992:114). One is tempted
to conclude that the lack of similar regulations in the Russian Duma is not only
due to its shorter tradition but also to its politically inferior status. Yet, a detailed
scrutiny of rules of address in other Europan parliaments may well reveal that it
is not the Russian recklessness but the British constraints that are most untypical.
However this may be, what results in the Russian case is a peculiar blending of
institutionalised and colloquial etiquette. The complex general rules regulating
the choice of the appropriate form of address in modern Russian as described in,
for example, Zemskaja (1987:218231) or Buchenau (1997), partly remain valid
19. As the author remarks, this variant is only admissible if the addressee belongs to the same
party as the current speaker. There is no similar constraint for Dorogie druzja Dear friends in
the Gosduma.
20. The same indirect mode is mandatory in the Czech parliament, see Madzharova Bruteig
(2010:292). In the Polish Sejm, however, direct addressing occurs as freely as in the Duma.
in the Duma, in other words: the behavioural rules of the latter differ less from
everday linguistic usage than those of the British Parliament.
6. Multi-addressed and multi-layered communication

Sometimes the final addressee and the whole status of the previous turn is questioned by the chair and has to be negotiated. In the next case, the outcome of this
process is a doubly addressed statement:
(16)
, ? ,
, .
. . , ,
, , .
(2/6/2008)
Deputy Xaritonov proposes to prepare an appeal to Kraeninnikov in order
to Is this going to be an assignment to the Committee? Please, Deputy
Xaritonov, explain this.
Xaritonov N. M. Oleg Viktorovi, I am directing this to Kraeninnikov and, as
it seems, to Pligin.
Such procedural uncertainty is often due to insufficient knowledge of the regulations in question:
(17) . . ,
, , , . ,
?

(2/8/2008)
Ivanov S. V. The point is that I already submitted the proposal for an invite to
the Question Time, but not for Kirienko, since he no longer is the director of
Rosatom, but instead the current director. So what, am I supposed to prepare
a written request to the Regulations Committee?
The last example also illustrates another point: a deputy can propose to summon
a member of the government for Question Time. This proposal must then be accepted (or rejected) by the House. The canonical form of such a proposal looks
as follows:
(18) . ., . [] ,
,
(2/8/2008)
228 Daniel Weiss
Ivanov S. V., Liberal Democratic Party. [] Therefore I submit the proposal

to invite the director of Rosatom for the next Question Time to tell the deputies
Among other officials to whom the Duma may address a parliamentary request
(zapros) are, e.g., the General Attorney, the president of the State Bank, the president of the Electoral committee, etc. (art. 47 of the regulations). On the whole,
we thus obtain a rather intricate system of direct and mediated and/or multi-addressed speech acts, which is represented in Figure 1.
Unlike the Duma transcripts, the turn-taking system and the address forms,
this typology of communicative acts does not contain any specific characteristics:
all types of exchanges occurring in it may be considered typical of parliamentary discourse in general. At least one possible exchange is, however, left out
because of a lack of data in the transcripts: individual deputies may communicate with each other in their hecklings, thus bypassing the chairperson as the
inevitable mediator.21 As for communication with absent addressees (representatives of the government, groups of the own electorate, etc.), this overview can be
compared with the schema outlined in a series of studies (Weiss 2006a, b, 2007)
where the dialogue with absent third persons in print interviews with Russian
politicians was analysed. The most essential modes of communication with the
Summary: the mediated parliamentary discourse
Foreign governments
President
Prime minister
Minister Z
Representative of the government

(Head of) Committee X
*
(Head of) Committee Y

*
*= protokolnoe poruenie
Chairperson
Sound operator
Deputy B
Deputy A
Deputy B
Whole assembly
A po kartoke C
Electorate: individual representatives whole groups (e.g. ernobylcy)
Figure 1. Types of direct and indirect communicative acts
21. Such an exchange encompassing four different hecklings is documented in Ornatowski

(2010:242).
world on the outside (and, as may be added, the most often discussed in the literature on parliamentary discourse) remain, however, unreflected in Figure 1 and
will now be briefly characterised: the Gosduma website and the TV coverage of
Dumadebates.
7. Communicating with the broader public: Internet and TV coverage

Similar to other parliaments, the Russian State Duma offers quite a rich array
of information on its official website. The search options include, e.g., not only
the entire list of current deputies with their passport data, but also their financial background and links to their statements held at the sessions so far. Data
on voting behaviour of individual deputies is also available.22 However, the list
does not contain electronic addresses of individual deputies, nor is the increasing importance of social networks reflected by the site.23 What is offered instead
is an anonymous contact address for requests (obraenija). All speeches, press
releases and other comments by the chair are published separately. In addition
to this, on-line broadcasts of particular sessions are regularly available since the
Spring session of 2012, and some deputies even explicitly address the Internet
audience in their speeches (e.g. O. Ju. Batalina, 6.20.12). But given that the major
part of the electorate still does not have access to the Internet, the traditional
mass media (TV and radio) still play a crucial role for the communicative exchange with the population. It seems to be a sound assumption that all Duma
debates are now being recorded on videotape.24 What becomes of these tapes is,
however, less clear: who has access to them and to what purpose? Are the Internet
broadcasts based on them? Are the short hand typists or the officials in charge of
the transcripts offered the delayed listening option? At any rate, they were not
publicly accessible in Internet archives before the 14th of September 2012, when
22. In most European countries such results are published regularly. Exceptions occur in federalist systems where one chamber represents not the people, but the units the federation is
composed of (states, cantons and the like). For example, in Switzerland the members of the
Stnderat (the equivalent of the US Senate) are reluctant to see their voting results published
because, as they argue, they often vote against the instructions of their own parties, but instead,
in the interest of their home region.
23. This is striking in view of the intense twitter activities of Medvedev and other members of
the government.
24. There is a closed-circuit television system that broadcasts the sessions (zasedaniia) of the
Duma to offices in the building, so it is entirely likely that there is a complete collection of video
recordings (Th. Remington, University of Atlanta, p.c.).
230 Daniel Weiss
the first video records appeared on the homepage of the Duma. As for the House
of Commons and the House of Lords in the UK, both live and archive coverage
(going back to the 1st July 2009) including debates and committee meetings are
available at (www.parliamentlive.tv/Main/About.aspx).25 On the other hand, according to the reply of the Gosduma chairman to the request in example (24)
below, TV stations can record whatever they like, since several cameras are installed on the balcony of the Duma. There is no separate TV channel reserved for
parliamentary purposes as is the case with, for instance, the German Bundestag,
the British Parliament (BBC) or its Italian equivalent (RAI), but on Sundays the
TV channel Rossija broadcasts the so-called Parliamentary hour (a reportage
that mainly highlights speeches by the Chair and the party leaders), and on Saturdays Radio Rossija emits the Parliamentary weekly. Remaining TV coverage
occurs in irregular intervals and consists mostly of summaries of and comments
on individual speeches. The so-called Governments Hour (pravitelstvennyj as;
the corresponding term in other countries would be Question Time) attracts more
attention, and speeches delivered by the governments representatives are often
broadcast in full lenght. This also allows the oppositional leaders to voice their
disagreement with the governments politics. Otherwise, the selectional criteria
are not very transparent: sometimes the importance of the given issue may be
decisive (e.g. the terrorist attack on Domodedovo airport or the controversy with
Poland over the war crimes committed in Katyn), but in other cases it is rather
the dramatic wording chosen by the deputy that seems to attract attention, and
the frequently broadcast excerpts of irinovskijs speeches show that scandalous
verbal and nonverbal behaviour of a deputy is most likely to be rendered in some
lenght, all the more as it can serve to discredit oppositional leaders. Insufficient
TV coverage is a frequent topic of parliamentary inquiries, cf.
(19) , , ,
24
.

.
(3/23/2011)
The Duma debates on Libya are of great interest to our society, therefore I
ask the chief officer in charge of mass media relations to discuss the question
25. However, it should be noted that even in the British Parliament, the establishment of
regular TV broadcasting was a painstaking process that lasted a long time. As Slembrouck
(1992:115) notes, An institution like Parliament is unlikely to yield immediately to contemporary media developments (cf. its slow and reluctant acceptance of the televising of parliamentary debates).
of broadcasting the debates with the channel Rossija 24 according to this

request by the Duma. The broad public simply remains ignorant of the positions of our political parties
In his reply, the chairman (Gryzlov) sees, however, no reason for an intervention
since all channels have the same right to broadcast Duma debates and the chamber has already decided not to exert any pressure to support or suppress such
transmissions.
Our next example illustrates once more the deputies concern about in
sufficient TV coverage, but additionally represents an attempt of actively interfering in TV programming, formulated in a bitterly sarcastic key:
(20) . ., . ,

, ,
. - ,
,

,
. ,

.

,
.
(6/25/2007)
Kain B. S., Communist Party. Dear deputies, last Sunday the TV channel
Rossija invited its audience to make up its mind and vote on who is more
important Gogol, Alexander Nevskij or the Academy member Saxarov. I
had a look at the Internet-site of this channel and found out that, according to
this terrific project, in autumn instead of the debates of the political parties in
the Duma warranted by the law, we will follow quite different debates; on the
struggle of Ilja Muromec versus the goalkeeper Lev Jain about who is more
important. I suppose Svanidze [a popular TV presenter] is already preparing
the role of Ilja Muromec The deputies have already accumulated many
questions about national television. I suggest to invite the deputy chairman of
the government Sergej Sobjanin, who is in charge of this domain, to Question
Time.
Sometimes, a deputy even comments on insufficient TV coverage of the governments activities:
232 Daniel Weiss
(21) . ., . ,
, ,
, . -
.
(2/6/2008)
Obuxov S. P., Communist Party. Currently, judging by communications from
federal TV stations, the only official who is actively carrying out his duties is
the first vice-premier Medvedev. At present, information about the activities
of the PM and the President is scarce on TV.
As for the broad public mentioned in example (19), part of the TV audience does
not seem to care too much about missing coverage of parliamentary sessions, as is
demonstrated by the following replies in an Internet forum:
(22) , ,
, ? ( (229)
: 1 )
Why dont they show Gosduma sessions on TV, the statements of different
parties, instead of presenting only already accepted bills?

. O
!
!
. (
)
, !

(otvet.mail.ru/answer/292647071; 9/10/2012)
Well, like this, during the next electoral campaign the deputies may take
their distance from the laws they supported themselves and shout: We voted
AGAINST that! Or on the contrary by shouting Our party was the one that
voted YES convince people that they should support these deputies. Anyway,
checking is impossible (the protocols of the votes are not published), so you can
only trust the words of those who lie more elegantly!
Besides the important point about missing data on voting behaviour of individual
deputies (see previous section), this fragment reflects a widespread attitude the
Duma itself is well aware of: its reputation in the electorate is far less flattering
than would be desirable.26 Internet forums simply abound in unfriendly remarks
like the following:
26. This attitude is, however, not limited to Russia. As Madzharova Bruteig (2010:275) reports, in spite of the rich possibilities of active participation offered to the citizens by the Czech
(23) [sic] , ,
. : ,
.

(otvet.mail.ru/answer/292647071/; 9/10/2012)
Just watch Comedy Club in its style, its contents and its vocabulary it is an
exact replica of Duma sessions. The overall principle is simple enough: here
and there, imbeciles are competing in imbecilism
This hostile note prevails in most postings. Certainly, similar voices would be easy
to find in European Internet forums, but in the eyes of Duma deputies they probably add to the overall impression of their own insignificance and the resulting
lack of self-esteem aired in numerous votes (see examples (1), (11) and (12)).
8. Conclusions
Our overview has revealed that communication in the Russian Parliament is in
many respects considerably less formalised than elsewhere, notably in the British
Parliament. Whereas some of the disorders illustrated above may be an incidental
outcome of the selection of data (e.g., deputies hesitate what status to assign to
their speech acts and whom to select as the ultimate addressee for their requests),
others seem to have a systematic character: turn-taking sometimes leads to chaotic scenes, which is partly due to the rigorous time limitation, and forms of address
vary significantly, oscillating between the official and the familiar key. Communication with the broader public likewise leaves a lot to be desired. The published
transcripts of Duma sessions manifest no grammatical and only minimal stylistic
editing and leave out substantial information (most types of back channel behaviour, hecklings). Moreover, the information available on the Gosdumas official
website has now improved but still does not provide all information that would
be required of a modern parliament, TV covering occurs irregularly, and there
was no publicly accessible video archive during the period covered by this paper.
The individual reasons for all these shortcomings may be manifold, but they add
to the overall picture of an institution that is constitutionally and politically less
influential, if not to say handicapped, compared with its European counterparts.
parliament, only one third of the people trust the Chamber of Deputies and even fewer have a
positive attitude to the Senate.
234 Daniel Weiss
References
Anderson, Richard D. 2010. When the center can hold: the primacy of politics in shaping
Russian democracy. Communist and Post-Communist Studies 43, Special Section: The new
Autoritarianism in the Former Soviet Union: 397408.
Bayley, Paul. (ed.). 2004. Cross-Cultural Perspectives on Parliamentary Discourse. Amsterdam/
Bevitori, Cinzia. 2004. Negotiating conflict: Interruptions in British and Italian parliamentary debates. In Cross-Cultural Perspectives on Parliamentary Discourse, Paul Bayley (ed.),
87109. Amsterdam/Philadelphia: John Benjamins.
Berlin, Lawrence N. 2011. Redundancy and markers of belief in the discourse of political hearings. Language Sciences 33 (2): 268279.
Buchenau, Klaus. 1997. Die Distanzanrede im Russischen, Polnischen und Deutschen und ihre
historischen Hintergrnde. Frankfurt a.M.: Lang.
Chilton, Paul. 2004. Analysing political discourse. Theory and practice. London: Routledge.
mejrkov, Svtla. 2011. Edited dialogues: redundancy replaced with relevance? Language
Sciences 33 (2): 280294.
Fleischmann, Eberhard. 2010. Das Phnomen Putin. Der sprachliche Hintergrund. Leipzig:
Leipziger Universittsverlag.
Graudina, Ljudmila K. and irjaev, Evgenij N. (eds.). 1994. Kultura parlamentskoj rei. Moskva:
Nauka.
Ilie, Cornelia. 2010. Introduction. In European Parliaments under Scrutiny, Cornelia Ilie (ed.),
125. Amsterdam/Philadelphia: John Benjamins.
Jurkovskij, Aleksandr M. 1949. Stenografija skvoz veka. Moskva: Kniga.
Kibrik, Andrej A. and Podlesskaja, Vera I. (eds.). 2009. Rasskazy o snovidenijax. Korpusnoe issle
dovanie ustnogo diskursa. Moskva: Jazyki slavjanskix kultur.
Laitin, David D. 2003. Post-Soviet Politics. Annu. Rev. Polit. Sci 2003: 117148.
Lapteva, Olga A. 1976. Russkij razgovornyj sintaksis. Moskva: Nauka.
Madzharova Bruteig, Yordanka. 2010. Czech parliamentary discourse. Parliamentary interactions and the construction of the addressee. In European Parliaments under Scrutiny,
Cornelia Ilie (ed.), 265302. Amsterdam/Philadelphia: John Benjamins.
ara, Jekaterina. 2011. Swiss Cheese and the Lazy Speaker: The Omission of Verbs in Russian and Czech. In Oslo Studies in Language 2 (1), Atle Grnn and Olga Klonova (eds.),
231242. Oslo.
Ornatowski, Cezar M. 2010. Parliamentary discourse and political transition. Polish parliament after 1989. In European Parliaments under Scrutiny, Cornelia Ilie (ed.), 223264.
Sakhno, Serge. 2011. Les dputs du Parlement russe pensent-ils? Autour du concept de
parlement : analyse en synchronie et en diachronie de certains termes de langues europennes. In Terminologie (I) : analyser des termes et des concepts, Jean-Jaques Briu (d.),
153190. Bern-Frankfurt: Peter Lang.
Sivenkova, Marija A. 2009. Otvety na voprosy v britanskix i rossijskix parlamentskix debatax.
Politieskaja lingvistika 1 (27): 124131.
Sivenkova, Maria A. 2013. On the metapragmatics of British, German and Russian political
questions and answers. In The Pragmatics of Political Discourse: Explorations across cultures, Anita Fetzer (ed.), 2146. Amsterdam/Philadelphia: John Benjamins.
Slembrouck, Stef. 1992. The parliamentary Hansard verbatim report: the written construction of spoken discourse. Language and Literature 1(2): 101119.
Weiss, Daniel. 2006a. Linterview politique dans les conditions du discours postsovitique. In
Le texte dans la Russie contemporaine. Actes du colloque de Grenoble des 12-13-14 mai
2005, Alexandre Bourmeyster and Isabelle Desprs (eds.), 1132. Grenoble: Universit
Stendhal.
Weiss, Daniel. 2006b. Der Dialog mit dem abwesenden Dritten in zeitgenssischen russischen
Politikerinterviews. In Iter Philologicum. Festschrift fr Helmut Keipert zum 65. Geburtstag [Die Welt der Slaven, Sammelbnde/Sborniki, Band 28], Daniel Buni and Nikolaos
Trunte (eds.), 8194. Mnchen: Sagner.
Weiss, Daniel. 2007. Der Dialog mit dem Publikum in russischen Politikerinterviews. In Sprache und Diskurs in Wirtschaft und Gesellschaft: Slawistische Perspektiven. Wiener Slawistischer Almanach, Sonderband 66, Ursula Doleschal, Edgar Hoffmann and Tilman Reuther
(eds.), 305323. Mnchen/Wien: Sagner.
Weiss, Daniel. 2011. Bezglagolnye konstrukcii russkoj razgovornoj rei: ix tipologija i status v lingvistieskom opisanii. In Slovo i jazyk. Sbornik statej k vosmidesjatiletiju Ju. D.
Apresjana, Boguslavskij Igor M. et al. (eds.), 139155. Moskva: Jazyki slavjanskix kultur.
Zemskaja, Elena V. 21987. Russkaja razgovornaja re: lingvistieskij analiz i problemy obuenija.
Moskva: Russkij jazyk.
uravleva, Natalja and Potapova, Olga. 2010. Polemika polemike rozn: osobennosti semantiki v belorusskom i polskom parlamentskix diskursax. Mova i kultura. Naukovij urnal
vip. 13: 176181.
Impoliteness and mock-impoliteness

A descriptive analysis
Michael Furman
In this article I offer an empirical analysis of impoliteness and mock-impoliteness in colloquial Russian conversation by examining the ways in which interlocutors produce and display an orientation to impolite and mock-impolite
utterances. The corpus consists of recorded video from naturally occurring
talk-in-interaction from the Russian reality television show Dom Dva House
Two. I argue that both, first-order, participant-constructed and second-order,
analyst-constructed approaches to politeness and impoliteness studies can, and
should, be used to inform one another. I use both approaches in order to analyze the specific ways mock-impolite and impolite turns are designed linguistically (through lexical items, turn structure and prosody) and paralinguistically
(through laughter, pauses and body language). Further, I examine how these
turns are discursively co-constructed by the interlocutors.
Keywords: mock-impoliteness, colloquial Russian, reality television,
impoliteness, speech act theory
1. Introduction
Scholars of impoliteness have rightly observed that studies of impoliteness have
traditionally been marginalized relative to those which focus on politeness
(Bousfield 2008:17). Recently, this imbalance has begun to be addressed; scholars
such as Jonathan Culpeper (1996, 2005), Derek Bousfield (2007, 2008), Marina
Terkourafi (2003, 2005, 2008) and others have been very active in developing and
enlarging the growing field of impoliteness. Indeed, the Journal of Politeness Research devoted a special issue to impoliteness in 2008, and so did the journal of
Intercultural Pragmatics in 2010.
The related topic of mock-impoliteness has yet to become a major area of
research within politeness and impoliteness studies. In this article, I analyze both
238 Michael Furman
impolite and mock-impolite utterances. Mock-impoliteness is defined in terms of

impoliteness, as it is parasitic on impoliteness: mock-impolite utterances contain
the same lexical content as impolite utterances but, above all, the socially conventionalized impolite meaning has been negated. Therefore, it is necessary to
distinguish between what is said (the locution) and the socially conventionalized
meaning (the illocution) of a given utterance. In its use of data from unscripted
television, this work can be compared with some recent studies on impoliteness
that analyze the dialogue in English-language documentaries and reality television (Culpeper 2005; Bousfield 2008). While there have been a few studies that
describe and analyze impoliteness in Russian (Zemskaja 1994; Larina 2009),
this is the first study to investigate the phenomena of mock-impoliteness in
colloquialRussian.
2. Previous research
2.1
Impoliteness
Research on politeness (as well as subsequent and current work on impoliteness

and mock-impoliteness) builds on J. L. Austins foundational work How to do
things with words. There he distinguishes between the three acts involved in performative utterances the locution (the literal content of what is said), the illocution (the socially conventionalized meaning of what is said) and the perlocution
(the effects or consequences of what is said). These insights have allowed linguists
and philosophers of language to analyze how language functions as a means to
accomplish social action.
In their seminal (if controversial) study of language used to perform social
action, Brown and Levinson (1987) posit a model speaker who, in performing a
face-threatening speech act (FTA), selects a given strategy from a repertory of five
superstrategies avoiding the FTA, off-record, bald on record, positive politeness,
and negative politeness in order to index the proper level of politeness. Culpeper
(1996) extends this model to the phenomenon of impoliteness, which he calls a
parasite of politeness (Culpeper 1996:355). Culpeper conceives of impoliteness
as the use of intentionally face threatening acts and lays out five super strategies
. Positive politeness, as defined by Brown and Levinson (1987), is the desire to have your
wants and desires also be the wants and desires of the group.
. Negative politeness, as defined by Brown and Levinson (1987), is the desire to have your
own will to be free from impingement.
Impoliteness and mock-impoliteness 239
(the inverse of the five super strategies put forth by Brown and Levinson) that
speakers use to make impolite utterances:
1. Bald on record impoliteness: performing the FTA in a direct, clear, unambiguous, and concise way even when face considerations are relevant.
2. Positive impoliteness: strategies designed to damage the addressees positive
face wants.
3. Negative impoliteness: strategies designed to damage the addressees negative
face wants.
4. Sarcasm or mock politeness: performing the FTA with politeness strategies
that are obviously insincere.
5. Withhold politeness: Not performing politeness work where it is expected.
As Culpepers initial model of impoliteness is based on Brown and Levinsons
theory of politeness, it has been subject to the same criticisms in particular, that
it is decontextualized and puts a strict emphasis on production to the exclusion
of perception. Culpepers revision of his model (2005) is in line with the general
switch in politeness and impoliteness studies from classifying utterances using
an analyst-constructed, second-order conceptions to participant-centered, firstorder conceptions.
2.2
Mock-impoliteness
One of the earliest works on mock-impolite behavior is Labovs (1972) study of

the phenomenon of sounding (the competitive usage of ritual insults) among the
youth of the African-American community in inner-city New York. Labov notes
that sounds have a rigid structure: they cannot be genuine insults but have to be
obviously false, as shown by ritualized cues. In addition, the use of a sound often leads to the occurrence of a second sound designed to top the first (Labov
1972:153). The two sounds thus form adjacency pairs where the production of
one sound leads to the expectation of another sound.
Subsequent approaches to mock-impoliteness preserve the notion that ritualized cues are key to separating impoliteness from mock-impoliteness. Thus
Culpeper defines mock-impoliteness as impoliteness that remains on the surface, since it is understood that it is not intended to cause offence (Culpeper
1996:352). Leech characterizes banter as mock-impoliteness meant to encourage
social harmony (1983:254). Kienpointner offers a typology of rudeness where
mock-impoliteness is conceived of as a form of cooperative and simulated rudeness (Kienpointner 1997:261). A more recent study by Bernal (2008) on insults
in colloquial Valencian Spanish divides impoliteness into two types: authentically
240 Michael Furman
impolite speech acts and inauthentically impolite speech acts. Bernal also argues
that addressees oriented to authentically impolite speech acts differently than to
inauthetically impolite ones; mock-impolite utterances are typically accompanied
by laughter or a joke, while authentically impolite utterances typically evoke protests and confrontation.
These studies point to the importance of conventionalized meanings. These
allow for a mock-impolite utterance to carry some sort of additional information that negates the conventional impolite meaning of the utterance; this, in
turn, allows for the interpretation of the utterances as mock-impolite rather
thanimpolite.
3. Methodology
Each of us have an understanding of what is and is not politic (Watts 2003) to
the interaction at hand, yet this understanding is discursively negotiated anew
through each interaction. Further, the structure of talk-in-interaction gives both
the analysts and the interlocutors the ability to monitor this negotiation and
display our understanding of it to our interlocutors. The field of conversational
analysis (CA) has taken the strict empirical stance that their understandings of
interaction are to be based only on publicly available information that is displayed
during talk-in-interaction. In general, CA describes the procedures and expectations with which participants produce and understand ordinary conversational
conduct (Heritage 1984:245). Sacks and Schlegloff argue for the sequential implicativeness of a turn, by which it projects a relevant next turn; they observe
that many conversational actions form a pair linkage, which they term adjacency
pairs (1973:295296). Among such adjacency pairs are the ritualized exchanges
of Hello and Goodbye, but notably also more complicatedly paired actions such as
question-answer, request-grant/rejection, etc. Sacks and Schlegloff do not make
the claim that adjacency pairs are inherently linked or invariably produced as succeeding actions. Rather, their claim is that there is a normative framework, such
that, after the hearers recognition of an utterance of a first pair part (a question,
request etc.), there is the expectation derived from the interlocutors habitus that
the next speaker will produce the second pair part (e.g., an answer to the question,
the granting of the request). This expectation is contingent upon the utterance of
the first turn of the adjacency pair; it allows the interlocutors, as well as analysts,
. It should be noted that, even though there is an expectation that the second member of the
pair will follow the first, this expectation is outside of the adjacency pair itself. It is not entirely
clear where in the process the recognition of the first member of the adjacency pair occurs.
to find that specific conversational events (e.g., answers to questions, refusals of

requests) are either present or absent. The fact that conversation is built up of
successive turns allows for the interlocutors (and analysts) to demonstrate their
mutual understanding of the task at hand without having to appeal to cognitive
projections of politic behavior. In this way the interlocutors are able to publicly
display and negotiate their understanding of utterances as being mock-impolite
or impolite. Following this, I take the position that adjacency pairs can be used
to do impoliteness and, accordingly, to do mock-impoliteness. Impoliteness and
mock-impoliteness can be created using any number of turns from a single turn
(in a single profane utterance or command) all the way to an indefinite string of
profane language over the course of innumerable turns.
In analyzing my Russian data from Dom Dva, I combine aspects of speech
act theory and the impoliteness strategies formulated by Culpeper (1996) with
tenets from Conversation Analysis (adjacency pairs and the use of natural talkin-interaction). The given approach acknowledges certain conventionalized understandings of utterances, while still allowing the participants themselves to
negotiate and discursively create the local meanings of the utterances. Specifically,
I adopt Culpepers (1996) impoliteness strategies as a set of super strategies that
the interlocutors both use and manipulate while ultimately using their publicly
displayed understanding of the utterances to classify the utterances as impolite or
mock-impolite.
Earlier I noted that Culpeper considers impoliteness to be parasitic on politeness. In much the same way, I argue that mock-impoliteness is parasitic on
impoliteness. That is, mock-impolite utterances necessarily contain linguistic or
paralinguistic features of a characteristically impolite utterance. I contend that the
contents of impolite and mock-impolite utterances are one and the same at the
level of the locution; consequently impolite and mock-impolite interpretations
arise from identical locutions. The difference between genuine and mock-impoliteness arises at the level of the illocution, the socialized meaning of the utterance.
I propose that, in addition to impolite illocutions, there exist also mock-impolite
illocutions. A mock-impolite illocution is one in which the socially conventionalized impolite illocutionary force has been negated. This negation is most often
accomplished through contextual factors (as in a host licensed by their authority
in a ritualized aspect of the show (Culpeper 2005); cf. the discussion of Clip 2,
below). It can also be explicitly mentioned (Im kidding, also present in my corpus) or signaled by paralinguistic cues (winks, laughter, gestures or general body
. Conversation analysts have also noted that adjacency pairs can also be utilized to accomplish particular actions such as rejections of requests, offers, proposals, agreements and disagreements (Pomerantz 1984), to name just a few.
242 Michael Furman
language) and/or prosody. Thus, the difference between an impolite illocution

and a mock-impolite illocution is in the intended illocutionary force, the desired
uptake and potentially, the perlocutionary effect rather than in the lexical content
or explicit form (the locution). Yet, whereas impolite illocutions are intended to
scorn, belittle or otherwise use further impoliteness strategies, mock-impoliteness illocutions carry no such intention and can be affiliative; that is, they can
promote social solidarity by signaling that the interlocutors share their own set
of social norms, which is in some ways separate from the larger conventions of
social interaction.
3.1
Data
My corpus consists of 18 clips of naturally occurring speech (8 mock-impolite

and 10 impolite ) gathered from over 40 hours of streaming content through the
website for the Russian reality television show Dom 2 (www.dom2.ru). Dom Dva
is a reality television show where the participants are explicitly trying to postroit
ljubov (build a loving relationship). The majority of the participants live on site
in housing built specifically for the show and have their lives continually filmed.
Some participants rotate in and out of the show weekly; however, the majority of
the participants have been on the show for a year or more.
In order to accurately transcribe the talk-in-interaction and note relevant
paralinguistic features, I captured the free streaming content from the website
www.dom2.ru (last accessed on February 28 2011) using the screen capture technology Snapz Pro X. This allowed the files to be saved to the hard-drive where
they could then be accessed and viewed repeatedly and, where necessary, discussed with native speakers of Russian.
4. Analysis
Turning now to the data, we can see the theoretical principles discussed above at
work. I first offer an analysis of impolite turns in order to establish a descriptive
account of impoliteness before analyzing the mock-impolite turns. The classification of these utterances is done in accordance with Culpepers impoliteness strategies (1996:356357). It is important to note that the analysis that follows is not a
. In making transcriptions from the streams, I utilized a set of transcription conventions
widely used by conversation analysts (see the Appendix). I chose this form of transcription
because it was better able to capture relevant nuances and details within the conversation than
other transcription conventions.
comprehensive account of impoliteness and mock-impoliteness in Russian in all

of their possible manifestations. Such an account would be exceedingly difficult,
if not impossible. Rather, the analysis is intended to describe particular examples.
Nevertheless, these examples do shed light on some of the possible strategies interlocutors can use to express and negotiate impoliteness and mock-impoliteness
in other contexts.
4.1
Impoliteness
Clip 1 features the impoliteness strategy of condescending scorn or ridicule. This

interaction has four participants, Sergej Ermakov (S), Tatjana (T), Inna (I), and
Elja (who does not speak in the clip below), although the conversation is largely
dominated by Sergej and Tatjana. They are seated at the kitchen table drinking
tea and snacking. The conversation at hand has focused on the potential outcomes
of the upcoming ladies vote, where there is the distinct possibility that Tatjana
will be voted off of the show. Right before the instance of impoliteness, Sergej has
been attempting to narrate a story; Tatjana has been continually interrupting him
to dispute his facts.
Clip 1: Past zakroj! Shut your trap!

(1)
1. S: , slyi? Svoju past zakroj! S toboj [razgovarivaju!]
2. T: [Oxrenel, to li?]
3. Pas[t u tebja!]
4. S: [Past zakroj!] Ty mo
5. : =Ser (throws hot tea in his face)
6. (They get up from table to confront one another.)
7. S: .hhhuuaa Ty kip[tjatkom, suka.]
8. T: [A ty menja] mom nazyvae?
9. S: Ty menja kipjatkom oblivae?
10. I: Vy , odureli?
11. S: Ty kipjatkom?
12. : =A ty kakogo xrena ja tebe!
13. (Sergej spits in her face, and she begins to kick him.)
1. S: do you hear me? Shut your trap! [Im talking with you!]
2. T: [Have you gone nuts or something?]
3. Youre the one [with the trap!]
4. S: [Shut your trap!]
5. You loser!
244 Michael Furman
6. T: = Ser (throws hot tea in his face)

7. (They get up from table to confront one another.)
8. S: You [threw] boiling water. [You bitch.]
9. T: [Why are you calling] me a loser?
10. S: =Why are you throwing boiling water on me?
11. I: What, are you two out of your minds?
12. S: Why are you throwing boiling water on me?
13. T: =And why the hell did you Ill get you!
14. (Sergej spits in her face, and she begins to kick him.)
Simply looking at the escalation of nonverbal violence, we can see that neither
Tatjana nor Sergej take very kindly to one anothers utterances. The illocutions
within the interaction demonstrate this escalation as well. The interaction begins with a directive from Sergej to Tatjana Svoju past zakroj Shut your trap
(line1). The traditionally intended perlocutionary effect of such a directive
namely, making the addressee be quiet is not achieved; instead, the directive is
met with a social sanction renel to li? Have you gone nuts or something?
plus the assertion past u tebja Youre the one with the trap (line 3). Sergej meets
this insult with two further illocutionary acts. The first (line 3) is a repetition of
his directive for her to shut up; the second (line 4) is an insulting declarative statement or exclamation ty mo You loser (or Youre a loser). It is this insult to
which Tatjana orients, as can be seen from her statement in line 8: A ty menja
mom nazyvae? Why are you calling me a loser? The perlocutionary effect of
(4) is escalation of the conflict, provoking Tatjana to throw burning hot tea in
Sergejs face (5). The fact that the statement in line (8) orients to the earlier utterance, not to Sergejs directly adjacent assertion that Tatjana is a bitch (line7),
allows the analyst to ascribe intentions to the interlocutors specifically, to pinpoint the likely reason why Tatjana threw tea in Sergejs face (as a response to his
declaration that she is a loser).
Typically, the preferred response to a question is an answer or a clarificational
question, yet Sergej responds to Tatjanas question A ty menja mom nazyvae?
Why are you calling me a loser? with another, non-clarificational question
Ty menja kipjatkom oblivae? Why are you pouring boiling water on me?.
Tatjana is likewise uncooperative. She refuses to answer Sergejs question or its
repetition (line 11) following the attempted intervention by the bystander Inna
. Sere is the colloquial vocative forms of Serea, a hypocoristic of the name Sergej.
. While loser is an adequate translation, the word mo has more negative connotations than
English loser.
(line 10); instead, she partially repeats her own question in a locution intensified
by profanity before breaking it off to threaten Sergej directly (line 12).
These questions without responses demonstrate to the analyst just what each
interlocutor believes to have been impolite. Specifically, Tatjana is concerned with
why Sergej has declared her to be a loser, corresponding to the impoliteness strategy of calling the other names (Culpeper 1996:358). For his part, Sergej orients to
why Tatjana threw water in his face, an action corresponding to the impoliteness
strategy of invading the others space (ibid.).
Both Jefferson (1984, 1986) and Larina (2009) have shown that overlapping
speech is not necessarily conflictive nor disorderly. However, the turns preceding
Clip 1 are overlapping to the point where even my native informants could not
readily make out what was said. Such an observation clearly points to the fact that
Sergej and Tatjana are not speaking at potential turn-completion units, but overlapping to the point where the conversation is unintelligible to native speakers of
Russian. Further, the perlocutionary effect of two of the turns (lines 4 and 12) is
to cause the interlocutors to commit acts of violence against each other (throwing
hot liquid, spitting, kicking); the interlocutors assume physically confrontational
postures in the course of the interaction (line 6); and one of the bystanders feels it
necessary to intervene with disapproval of their behavior (line 10).
Clip 2: man Deception

The following exerpt is taken from an interaction at the lobnoe mesto (execution
place), a ritualized outdoor gathering where new participants are introduced, existing participants are voted off of the show, and new romantic liaisons are all
publicly revealed. During this gathering, the participants sit in a circle, and one
of the hosts of the show (who does not live in the house with them) asks them
questions in an interview-like manner. Two of the participants, Venceslav (V) and
Alina (A), have just announced that they are now a couple, to the expressed surprise of the other participants and the host, Olja (O). According to the established
rituals of the show, they now need to show proof of their relationship typically
in the form of a public kiss. Alinas refusal leads to this sequence.
(2)
1. O:
2.
3.
4. :
A po moemu tebe prosto togo ne xoetsja net ^ Davaj

budem estnymi,
Alin. Zaem^ ty nas obmanyvae.
Kogo^ ja obmanyvaju?
. Kakogo xrena, while rude, is still a taboo substitution for kakogo xuja; that is, its not as far
as one could go (cf. English frigging, freaking, or effing).
246 Michael Furman
1. O:
2.
3.
4. A:
In my opinion you just dont feel like it right?^ Lets be

honest,
Alina. Why are you deceiving us?
Whom^ am I deceiving?
Oljas first turn in Clip 2 contains two unflattering propositions (1): Davaj budem
estnymi (Lets be honest), which implies that Alina not being honest; and (2)
zaem ty nas obmanyvae (Why are you deceiving us?), which presupposes that
Alina is deceiving them. These speech acts are examples of the impoliteness strategy of associating the other with a negative aspect, and both make this association
implicitly rather than explicitly. However, in responding to the presuppositional
question Why are you deceiving us?, Alina does not offer an answer the preferred second turn of a question but instead responds with another question,
Kogo ja obmanyvaju Whom am I deceiving?. This would seem to be an anomalous turn according to the norms of conversational interaction, but Alina is not
directly orienting to what Olja has said. Instead, she is orienting to the presupposition that she is deceiving the host and the other participants on the show. The
illocutionary force of her question is thus to challenge the question previously
posed to her.
The seemingly unexpected move in line 3 demonstrates why it is warranted
methodologically to use speech act theory and conversation analysis in tandem.
Conversation analysis does not allow for a reading beyond what is said. However,
as shown in Alinas turn in line 3, at times interlocutors orient not to what is
said, but to implicit associations and presuppositions, which cannot be dealt with
by conversation analysis alone. Without appealing to presuppositions or implicatures within an illocution, it is impossible to offer an adequate explanation of
thisturn.
Analyzing these and other interactions from my corpus using both conversation analysis and speech act theory allows us to draw four generalizations. First,
impolite turns and illocutions can lead to conflictive speech (Kogo ja obmanivaju?
Whom am I deceiving?). Second, they often do not follow the expected adjacency structuring: one directive may be countered with another, or a question may
be followed by a question. Third, impolite turns and illocutions can exhibit an
escalation in aggression (physical or verbal) through the course of a given interaction e.g., throwing hot liquids, spitting and kicking. Finally, impolite turns and
. Yokoyama (1990) notes that responding to a question with a question is a strategy used by
Russian store clerks in order invert the power structure so that they now have more power than
the customers.
illocutions can be met with social sanctions (e.g., Oxrenel to li? Have you gone
nuts or something; Vy odureli? What, are you two out of your minds?).
4.2
Mock-impoliteness
I began with the premise that mock-impoliteness must be parasitic to impoliteness. This is because, in order for an utterance to be produced and perceived as
mock-impolite, it must have the potential to be read as authentically impolite.
Therefore, there are certain similarities between mock-impoliteness and impoliteness. The following examples bear this out.
The following excerpt (Clip 3) has three different interlocutors Maa (M),
Tomas (T) and Inna (I), although it is mainly Maa and Tomas who speak. Maa
walks into the large dining area where Inna and two other participants are seated
at the table eating breakfast. Tomas is standing, fixing himself a drink in the kitchen, while Venceslav is seated on the sofa at the far end of the room. Maa has just
returned to the Dom Dva complex after having nearly been voted off of the show.
She returns to a roomful of the people who voted against her. In her opening turn,
she uses the conventional impoliteness strategy of inappropriate identity markers,
such as labeling friends enemies. Yet within the same turn she utilizes the illocutionary force indicating device (IFID) uu Im kidding to explicitly state her
intention and display to the interlocutors that this is not an impoliteness strategy
but a mock-impoliteness strategy.
Clip 3: Vragi Enemies

(3)
1. : Privet semja^(.3) .hhh
2. I: =privet
3. M: =Prjam vse kto pro10-Krome Venceslava (.3) Vse
4. vra-vragi(.) uu [druzja], druzja.
5. : [ahh] Privet Maa.
6. : Zdra^sste. Kak dela Tomas.
7. : =Super^, >Marija Paly<. Reili nas vyselit iz gorodskix
8. kvartir.
9. I: Da^.
10. : Serzno^ ty dumae silnok xvatit^ u tebja.
11. : Ne znaju, moet i xvatit moet i net.
10. Pro is used here in a parliamentary sense, as the adverb pro, aye, rather than as the preposition about or concerning. In this case pro is meant as in support of voting me off the show.
248 Michael Furman
12. I: Budut probovat skazali.

13. : Da^
1. M: Hey, family^(.3) .hhh
2. I: =Hey
3. M: Everyone who voted aye is here, except Venceslav.11 All are
4. en-enemies(.) Im kidding [friends], friends.
5. : [ahhh] Hey, Maa.
6. : Hello^. How are things, Tomas?
7. : =Super^ >Marija Paly.<12 They decided to evict us from the
8. city apartments.
9. : Seriously^ Do you think that youve got enough^ strength for
10. that.
11. : Dont know, maybe enough, maybe not.
12. I: Theyre going to try, they said.
13. : Yes^
Maa initially frames the entire interaction as affiliative by greeting those in the
room as her family. This greeting is oriented to and returned by Inna privet
(line 2), who is sitting at the breakfast table. Maa then makes two further utterances at potential turn-completion points. First, she observes that everyone present had been for her being voted off the show seemingly a revelation that she
is harboring a grudge; yet, she then declares that they are all her enemies (except
for Venceslav). Both of these declarations are ostensibly conflictive, as they are in
line with the impoliteness strategy of using inappropriate identity markers. However, Maa negates the conventional impolite illocutionary force in three different
ways. She immediately asserts that her illocutionary intent was non-serious (Im
kidding friends, friends). Further, each of her turns leading up to her greeting
to Tomas (line 6) are said with a smile and accompanied by exaggerated vertical
and horizontal hand gestures. All three of these actions serve to negate the impolite illocutionary force of labeling her interlocutors as vragi enemies.
The reading of vragi as a mock-impolite rather than impolite utterance is
supported by the history of the interactions preceding it. Maa was nearly voted
off of the show by the very people whom she is addressing, yet she was not voted
off of the show and was still living with those who voted against her. The delicacy
11. Venceslav did not vote in favor of her leaving the show.
12. It is interesting to note that in line 7 Tomas uses Maas (Marijas) full name, and also her
husbands surname Paly rather than her own, Kruglyxina. Name + surname address is not typical in colloquial Russian. In his previous turn (line 5), Tomas used a more expectable form of
address, the hypocoristic Maa.
of this situation, the unsuccessful attempt to vote her off the show and the addressees vulnerability to face attacks, coupled with the advisability of maintaining one anothers face amongst roommates, help to explain why Maa is careful
to be explicit about her intentions utilizing an IFID (Im kidding). Specifically,
their decision to vote for Maas expulsion from the show makes them more likely
to read Maas utterances as impolite, so Maa addresses what she perceives as
their predisposition by utilizing an IFID, smiling as she performs the ostensibly
impolite utterance and using exaggerated gestures to neutralize any impoliteness
that may have otherwise been perceived from her locution.
Importantly, Tomas and Maa simply go about their everyday business in the
discussion of the city apartment. There is a four-turn development in which Tomas orients not to the ostensible content of Maas declaration that he is an enemy
but to the negotiated idea that it was meant in jest; he offers the preferred response
to a standard greeting, a return greeting (privet hello, line 5), which suggests that
he has taken Maas first turn as simple greeting rather than an impolite utterance. Maa then displays her understanding that Tomas has rightfully interpreted
her turn as a mock-impolite utterance by continuing on with the interaction in
a matter-of-fact way. That is, she moves past the greeting and begins questioning
Tomas about his everyday affairs. Thus, by using publicly available information,
the analyst can state with some degree of confidence that Maa intended her illocution to be mock-impolite and that Tomas rightly understood her illocution
to be mock-impolite.
It is interesting to note that Maa utilizes a mock-impolite utterance to broach
the sensitive topic of her addressees voting to dispel her from the show. What
this seems to signal is that all of the interlocutors are above the out-group social
conventions that would discourage against the selection of such an utterance.13
Despite being voted against by her interlocutors only hours before, Maa can still
utilize certain in-group strategies such as mock-impoliteness. In this way, mockimpoliteness can also be used to signal social solidarity.
Looking at another interaction, we again see further uses that indicate that
lexical items traditionally perceived and produced as impoliteness can also be
used to express mock-impoliteness as well. That is, conventionally impolite illocutions can have their impolite meaning negated. The following excerpt (Clip4)
features an encounter between Sergej Ermakov (S), and Polina (P). Earlier in the
series, Sergej and Polina had a brief romance together, but are no longer together
at the time of this interaction. The beginning of the given interaction is not provided on the Dom Dva website. In the part that we see, Sergej and Polina are
13. Brown and Levinson state that in those situations where there is a strong threat to the
hearers face, one should withhold the face-threatening act (Brown and Levinson 1987:69).
250 Michael Furman
already together in a hallway; from there, they move into the kitchen, where sausages are cooking. Internal evidence shows that Sergej has been in the kitchen
shortly before the encounter. Just prior to the beginning of the clip, there has
been some sort of physical activity in which Polina has pinched Sergejs nipples.
This could be read as being violent and impolite, but Sergej does not orient to it in
this manner; instead he smiles and continues to joke with Polina in spite of what
seems to be some physical discomfort.
Clip 4: Suka! Little bitch!

(4)
1. S: Zaem^ ty za sisku menja uipnula.
2. Oh Ohhhh. Oh Ohhh. Suka! Za sisku uipnula!
3. ((Said between small spurts of laughter.))
4. ((Sergej approaches Polina from behind, puts his arms
5. around her neck and yells something as she smiles))
6. eh ha huh huh Aj pojdm ajku popm.
7. P: =No ja v lepkax^ kak s [bazara].
8. S: [Nu ego] zdes
9. tebe v lepkax ne xoditsja?
10. P: =Ty bolnoj to li?
11. S: =Tebe nuno kabluki srono potomu to u tebja opa potomu
12. to visit.
13. Ja ue ix tykal. Pododi. ((Polina jabs the sausages with a
14. fork.))
1. S: What^ did you pinch my nipple for! Oh Ohhh. Little bitch!
2. You pinched my nipple! ((Said between small spurts of
3. laughter.))
4. ((Sergej approaches Polina from behind, puts his arms
5. around her neck, and yells something as she smiles))
6. eh ha huh huh
7. Lets go have some tea.
8. P: But Im in sandals^ [like from the bazaar].
9. S: [Well how come you cant] walk around
10. in sandals here?
11. P: =Are you crazy?
12. S: =You need to put heels on right away because your ass
13. because its hanging out.
14. I already poked them. Wait. ((Polina jabs the sausages with
15. a fork.))
The mock-impoliteness strategy used here is that of scorning and belittling the
other through the word suka, diminutive of suka bitch. Typically, suka and suka
accompany speech acts whose illocutionary force is impolite, and Sergejs question and exclamation in (lines 12) superficially seem so and are conventionally
read as reproaches. However, this entire turn is uttered almost in laughter, and is
followed by playful behavior and an invitation to drink tea (lines 36). Moreover,
the perlocutionary effect of the speech act on the addressee is neither an escalation of violence, nor any evidence of conflict over the terms used.
In addition, Clip 4 features another adjacency pair that demonstrates traits of
mock-impoliteness. In response to Sergejs comment in lines 8 and 9 nu ego zdes
tebe v lepkax ne xoditsja? Well, how come you cant walk around in sandals
here? Polina seemingly questions his sanity Ty bolnoj to li? Are you crazy?
(line 10). In the following turn, Sergej does not orient to this as an insult, nor does
he offer the conventional second part of a question-answer adjacency pair. Rather,
he makes a declaration, with sarcastic intonation, pretending to agree with her
insecurity about wearing sandals in the kitchen: Tebe nuno kabluki srono You
need to put some heels on immediately (line 11), then offers a seemingly rude assertion about her anatomy, including the common vulgarism opa ass (line11).
Polina does not orient to either of these potentially conflictive utterances explicitly; the video clearly shows that there is no change in body language that might
indicate a change in the tenor of the conversation. Importantly, both the participants continue with the task at hand, i.e., tending to the sausage.
While we cannot use purely linguistic data to support the argument that the
last two turns are mock-impolite, we can see through the paralinguistic evidence
of their body language (playful shoving and touching of one another) and their
cooperative focus on the task at hand that these remarks are not understood to
be impolite. In addition, their subsequent interaction (not included in Clip 4) is
flirtatious rather than confrontational.
In the next clip, the interaction is between the host Olja (O) and a man named
Sergej (S), who has just come onto the show and is attempting to establish himself
as a legitimate participant. The ritual of the show, as mentioned above, is that
potential participants come and meet the already established participants in the
lobnoe mesto; there the host then asks them a number of pressing questions in
order to find out why they came and with whom they want to build a relationship.
For brevity, I have omitted the beginning of the interaction.
Clip 5: konomist Economist

(5)
1. O: A po professii kto?^
2. S: Individualnyj predprinimatel.
252 Michael Furman
3. O: =To est net obrazovanija ehh ha [haha]

4. S: [ax ejshhh] huh haha.
5. konomist ((smiling))
6. O: =Nu to xoroo konomist i prolyj borec. Zaem ty k nam
7. pril^?
1. O: And what is your profession?
2. S: Entrepeneur.
3. O: =That is you dont have an education [hahaha]
4. S: [ax ejshhh] huh haha.
5. S: An economist ((smiling))
6. O: =Alright, an economist and former wrestler. Why^ did you
7. come to us?
The straightforward, traditional reading of Oljas assertion that, if Sergej is an entrepreneur, he must not have an education (line 3) is that it is impolite; in contemporary Russian culture, education is a desirable trait to have, while a lack of it
is negative. Importantly, however, Oljas turn concludes with laughter, indicating
that its illocutionary force is non-serious, i.e., that her utterance is not impolite
but only mock-impolite. In the next turn, Sergej immediately and publicly displays his understanding of her intention by latching his turn with laughter to
hers before re-answering her initial question (4). While it is possible that this is
a case of institutionalized impoliteness,14 there are several features of this interaction that speak against this, apart from the laughter. First, the host smiles as
she delivers this potentially face threatening act, and Sergej smiles in response;
second, both interlocutors are relaxed in their body language and lean in toward
one another; and third, neither the host nor Sergej orient to these utterances as
impolite in any discernable way. Evidently, then, the perlocutionary effect of Oljas
potentially face-threatening but actually mock-impolite assertion is not to create
conflict but to provoke shared, friendly laughter. Moreover, in Oljas next turn she
moves on to a new topic without repairing the previous turn, which suggests that
she is satisfied that her illocutionary intent was understood in the way in which
it was meant.
We can further deduce that the mock-impolite illocution in line 3 plays an
additional role creating affiliation. As noted above, Sergej has just come onto
the show; according to the out-group social conventions typical of contemporary Russian society, he should be treated with a certain degree of respect and
not be subject to face-threatening acts. Yet, even in their very first encounter, the
host Olja seems to be asserting that he does not have any education; significantly,
14. This possibility was raised by one of the anonymous reviewers.
however, she clearly signals that this assertion is mock-impolite rather than serious (impolite). While this sort of behavior may not conform to larger social conventions, it is in line with the in-group social conventions of Dom Dva, where the
host is licensed with a certain authority to issue provocative, sometimes mockimpolite or even impolite utterances at the lobnoe mesto. Further, as mentioned
earlier, Oljas laughter functions as an invitation to Sergej to laugh and to partake
in the social norms for Dom Dva. In this way, laughter can be seen as a mark of
affiliation allowing Sergej to participate in their in-group norms of interaction.
Sergej accepts this invitation instead of taking issue with her ostensible insult.
This is further evidence that mock-impoliteness can and, in this particular instance, actually does perform a socially affiliative and inclusive function.
5. Conclusions
Using these examples as a basis, we can draw several conclusions. In contrast to
their impolite counterparts, mock-impolite utterances do not exhibit an escalation in conflict. We saw this in Clip 5, when the traditionally impolite words suka
and opa do not lead to conflictive speech; and in Clip 3, when Maa declares all
those who voted against her to be enemies, yet the interlocutors do not orient to
her statement as an impolite illocution. Mock-impolite utterances can contain
laughter produced by both the current speaker and/or other participants a feature missing in the genuinely impolite illocutions contained in this corpus. IFIDs
like the explicit identifiction of joking are also missing in the genuinely impolite
utterences discussed. In addition, all the mock-impolite illocutions are uttered in
contexts that favor an affiliative reading, e.g., flirting (Clip 4), same-sex bonding
(present in the corpus, but not analyzed here owing to space constraints), and
ritualized banter between the host and a new entrant to the show (Clip 5). The
analysis has also shown that mock-impoliteness can also be used to signal social
solidarity and be socially affiliative.
With only a few exceptions (e.g., Labov 1972 and Bernal 2008), mock-impoliteness has received little attention in previous research. In this article, I have
focused on this understudied phenomenon by examining how participants create
and orient toward mock-impolite utterances, as well as some of the functions of
mock-impoliteness in a specific social group. In addition, I have proposed and
demonstrated a methodology for distinguishing mock-impolite utterances from
genuinely impolite ones. For a better understanding (rather than simply theorizing) about the function and execution of impolite/mock-impolite utterances and
how mock-impoliteness functions, it is necessary to analyze conversational turn
sequences in detail, as has been done here.
254 Michael Furman
The methodology used here suggests that conversation analysis and speech
act theory can be utilized to inform one another. Conversation analysis does not
allow for a reading beyond what is said; however, mock-impoliteness often involves the presuppositions and implicatures that have traditionally been a concern of speech act theory. Conversely, speech act theory has largely concerned
itself with the production of utterances rather than their perception. By adopting
the assumption of conversation analysis that the utterance of one turn projects
the logical completion of the next, one is able to reach a deeper and more detailed understanding of the production and perception of the phenomenon of
mock-impoliteness.
Acknowledgements
I would like to thank the two anonymous reviewers for their productive comments and criticisms, Daniel Collins for his help in revising this paper and my
native informants who assisted me in the transcription of the data.
References
Austin, J. L. 1975. How to Do Things with Words. Cambridge: Harvard University Press.
Bernal, Maria. 2008. Do insults always insult? Genuine impoliteness versus non-genuine impoliteness in Colloquial Spanish. Pragmatics 18: 781802.
Bousfield, Derek. 2007. Impoliteness, preference organization and conducivity. Multilingua
26: 133.
Bousfield, Derek. 2008. Impoliteness in Interaction. Amsterdam/Philadelphia: John Ben-
jamins.
Brown, Penelope, and Levinson, Stephen C. 1987. Politeness: Some Universals in Language Use.
Culpeper, Jonathan. 1996. Towards an anatomy of impoliteness. Journal of Pragmatics 25:
349367.
Culpeper, Jonathan. 2005. Impoliteness and entertainment in the television quiz show: The
Weakest Link. Journal of Politeness Research 1: 3572.
Heritage, John. 1984. Garfinkel and Ethnomethodology. New York: Polity Press.
Jefferson, Gail. 1984. Notes on some orderlinesses of overlap onset. In Discoure Analysis and
Natural Rhetoric, Valentina DUrso and Paolo Leonardi (eds.), 1138. Padua, Italy: Cleup
Editore.
Jefferson, Gail. 1986. Notes on latency in overlap onset. Human Studies 1(2/3): 153183.
Kienpointner, Manfred. 1997. Varieties of rudeness; types and functions of impolite utterances. Functions of Language 4: 251287.
Labov, William. 1972. Language in the Inner City: Studies in the black English vernacular.
Oxford: Blackwell.
Larina, Tatjana V. 2009. Kategorija velivosti i stil kommunikacii: sopostavlenie anglijskix i
russkix lingvokulturnyx tradicij. Moscow: Rupopisnye pamjatniki Drevnej Rusi.
Leech, Geoffrey. 1983. Principles of Pragmatics. New York: Longman.
Pomeranz, Anita. 1984. Agreeing and disagreeing with assessments: some features of preferred/dispreferred turn shapes. In Structures of Social Action: Studies in Conversation
Analysis, J. Maxwell Atkinson and John Heritage (eds.), 57102. Cambridge: Cambridge
University Press.
Terkourafi, Marina. 2003. Generalized and particularized implicatures of politeness. In Perspectives on Dialogue in the New Millenium, Peter Kuhnlein, Hannes Rieser and Henk
Zeevat (eds.), 151166. Amsterdam/Philadelphia: John Benjamins.
Terkourafi, Marina. 2005. Beyond the micro level in politeness research. Journal of Politeness
Research 1: 237262.
Terkourafi, Marina. 2008. Toward a unified theory of politeness, impoliteness, and rudeness.
In Impoliteness in Language: Studies on its Interplay with Power in Theory and Practice,
Derek Bousfield and Miriam Locher (eds.), 4576. New York: Mouton de Gruyter.
Watts, Richard. 2003. Politeness. Cambridge: Cambridge University Press.
Yokoyama, Olga. 1990. Responding with a question in colloquial Russian. In Topics in Colloquial Russian, Margaret Mills (ed.), 116. New York: Peter Lang.
Zemskaja, Elena A. 1994. Kategorija velivosti v kontekste reevyx dejstvij. In Logieskij analiz
jazyka. Jazyk reevyx dejstvij, Nina D. Artjunovoj (otv. red.), 131137. Moscow: Nauka.
www.dom2.ru last accessed on February 28 2011.
Appendix: Transcription conventions

(.5)
The number in brackets indicates a time gap in tenths of seconds.
(.) A dot enclosed in a bracket indicates a pause in the talk of less than two-tenths of a
second.
=
The equals sign indicates latching between utterances.
[ ]
Squre brackets indicate the onset and end of a spate of overlapping talk.
.hh A dot before an h indicates speaker in-breath. The more hs, the longer the inbreath.
hh.
An h indicates an out breath. The more hs, the longer the breath.
sou:::nd Colons indicate that the speaker has stretched the preceding sound or letter. The
more colons, the greater the extent of the stretching.
(guess) Words within a single bracket indicated the transcribers best guess at an unclear
utterance.
word. A full stop indicates a stopping fall in tone. It does not necessarily indicate the
grammatical end of a sentence.
word, A comma indicates a continuing intonation.
word? A question mark indicates a rising inflection.
word^ A carrot indicates a strong rising intonation. More carrots, indicates a stronger rising inflection.
256 Michael Furman
words* An asterisk indicates a croaky pronunciation of the immediately following section.
word! An exclamation mark is used to indicate an animated or emphatic tone.
((word)) Segments within two parentheses indicate pertinent non-verbal action.
>word< Inward chevrons indicate that the talk they encompass was produced noticeably
quicker than the surrounding talk.
Humor as staging an utterance

Nadine Thielemann
In this paper conversational humor is understood as a discourse modality actively contextualized by speakers in order to indicate that an utterance is not
meant seriously. The paper focuses on the contextualization cue (Gumperz
1982) of animated speech which is accounted for in terms of shifted footing
(Goffman 1992), i.e. instances when speakers lend their voice to another character and stage utterances attributed to this character. Using the example of
data from Russian face-to-face conversations, it is shown that animated speech
plays a crucial role in establishing several forms of humor such as parody, irony
or teasing. Staging an utterance, speakers humorously detach themselves from
a rendered discourse and convey the meta-message This is play while at the
same time playing a role. It is therefore argued that animated speech can be regarded as a natural contextualization cue, where the indexed meaning is iconically motivated.
Keywords: humor, contextualization, reported speech, irony, teasing
1. Introduction
Conversational joking often does not fit in culturally well-entrenched genre categories such as jokes or wordplay. Discourse analytic approaches to conversational joking such as the contributions edited by Kotthoff (1996) therefore favor an
account within contextualization theory (Gumperz 1982:130152; Auer 1992):
When interlocutors joke in face-to-face interaction they use several contextualization cues (CCs) to establish a humorous framing or to convey the meta-message This is play (cf. Bateson 1972:177193) which indicate that they speak in
jest, and which allow their recipients to infer that the such marked stretch of talk
is not meant seriously. Conversational humor is then conceived of as a discourse
. Dynel (2011) reviews alternative concepts such as interactive frames and keying and discusses their ability to account for conversational joking.
258 Nadine Thielemann
modality or a meta-message which is actively contextualized by the speakers a

view that will also be adopted in this paper.
A similar proposal has been made by Zemskaja (1983), who identifies a sphere
of a humorous discourse modality (utlivaja tonalnost) in colloquial Russian
which is characterized by play on language (jazykovaja igra). She describes several
methods (priemy) how speakers create humor in face-to-face interaction employing elements from different linguistic subsystems such as grammar or prosody as
well as pragmatic or rhetoric cues. Among the last, she mentions style shifts or
clashes, the use of lexical elements of other stylistic registers (priem reevoj maski)
and quotations (priem citacii) or pseudo-quotations (psevdocitacija) by which the
speaker momentarily switches to another role.
These observations serve as a point of departure in this paper as it focuses on a
complex CC frequently used by Russian interlocutors in face-to-face interactions
which relies on the staging of an utterance. During conversational joking speakers
often lend their voice to another character, render speech of this character and in
doing so humorously detach themselves from what is said. Following the ideas of
Goffman (1975, 1992), we suggest the term animated speech to describe this CC.
In Section 2 the benefits of this concept in comparison to related concepts such as
reported speech will be discussed. The concept of animated speech is also able to
embrace several of the methods (priemy) mentioned by Zemskaja under different
labels such as speech mask or quotation (cf. Sannikov 2003:476479). Based on
data from Russian face-to-face interaction, we will then show in Section 3 how
animated speech is used to establish a humorous discourse modality in different
forms of conversational humor such as parody, irony, teasing and other culturally
less entrenched forms of humor. As contextualization cues (CC) are generally
claimed to have no referential meaning, we will subsequently discuss in Section 4
whether there is a natural link or a motivation between animated speech as a CC
and the meta-message This is play as the indexed meaning triggering a humorous interpretation.
2. What is animated speech?

When joking in casual conversations, speakers frequently mark a stretch of their
talk as stemming from another source and humorously detach themselves from
the rendered speech. These two features attribution of talk to another source
and humorous detachment are crucial for the phenomenon under analysis. In
this section we will suggest to account for this in terms of animated speech and
show why other concepts such as reported speech or quote have shortcomings with
regard to humorously rendered speech.
Humor as staging an utterance 259
Talk rendered by the speaker and attributed to another source is traditionally

treated as (direct) reported speech (in Russian uaja re cf. Kitajgorodskaja 1993;
Lejnonen 1998). Quote (Bolden 2004) or quotation (Mayes 1990) is sometimes
used as an alternative term. According to Coulmas (1986:2), direct reported
speech evokes the original speech situation and conveys, or claims to convey,
the exact words of the original speaker in direct discourse. Regarding conversationally reported speech this only holds true to a certain degree. As convincingly
shown by Tannen (2007:102132), reported speech simply serves as a means to
present talk as speech from another source or situation, and that it neither has to
offer a verbatim rendition nor a rendition of earlier talk at all. Speakers can render
speech from different sources whereupon a verbalized original need not necessarily exist e.g., when thoughts or written texts are reintroduced (ibid.:112119).
The rendered utterance may also resemble the presumed original to different degrees. Instances range from verbatim quotes to summaries or reformulations of
other utterances. Speakers can render their own utterances, utterances of their
interlocutors or of non-present third parties, or also utterances that have never
been verbalized which rather represent typical or stylized utterances, or utterances that can be attributed to a certain type of person (Kitajgorodskaja 1993:8086).
Especially in humorously animated speech, the rendered talk may embody ideas
attributed to any source (verbalize a clichs, stereotype, etc.) or even be purely
fictional (cf. Holt 2007). Tannen (2007) suggests the term constructed dialogue
to capture all these instances of casting ideas as dialogue (ibid.:119).
In introducing the notion of footing Goffman (1992:124159) provides an
alternative concept that is able to describe how speakers can detach themselves
from what they say. He distinguishes different production roles or alignments
of the speaker to what s/he renders: The principal is the originator of the ideas
conveyed by an utterance, someone whose beliefs have been told, someone who
is committed to what the words say (ibid.:144), the author has selected the
sentiments that are being expressed and the words in which they are encoded
(ibid.), and the animator is merely the talking machine, a body engaged in acoustic activity () an individual active in the role of utterance production (ibid.).
When talking, speakers can highlight or update one production role and shift
the footing of an utterance. When they merely become the animator of what they
say, they lend their voice to a figure who is present only in a world that is being talked about (ibid.:149). In other words, when speakers produce animated
speech they animate a virtual character (figure). At the same time they reduce
their responsibility for the rendered stretch of talk since it is attributed to another
source (principal) (Goffman 1975:512). This approach allows capturing reported
speech independent of the linguistic form, the length of the rendered talk or its
authenticity, and stresses the effect of speakers detachment. It has been applied to
account for reported speech (e.g., by Couper-Kuhlen 1999; Holt 2007) as well as
for talk that is marked as stemming from hearsay by Russian evidential particles
(e.g., by Grenoble 1993; Baranov 1994), and it is an apt concept to account for
humorously rendered speech.
According to Goffman, it is the communicative default that the speaker is
principal, author and animator of an utterance. Animated speech is a marked
speech event as this unity is dissolved. When speakers shift the footing and stage
a character to whom they lend their voice, they have different means at their disposal to mark this shift. These are the strategies which also serve to single out
reported speech summarized by Bolden (2004), Lejnonen (1998) or Kitajgorodskaja (1993) for Russian:
reporting frame such as [speaker] + [reporting verb], [speaker] or tot that,
kak so without a verbum dicendi
main clause phenomena (Mayes 1990:338345) such as interjections or discourse particles marking the start of the animated speech (cf. also Gnthner
2000)
style shifts, code switching etc. (cf. also Gnthner 2000), cf. prim reevoj
maski (speech mask) (Zemskaja 1983, 1995)
shift of the deictic centre; this may include a shift of the vocal deixis
(Couper-Kuhlen 1999:14)
In Russian a shift of footing can also be indicated by evidential particles (ksenopokazateli, Arutjunova 2000) such as mol or de (Grenoble 1993; Baranov 1994). The
treatment of these particles as evidentials is contested as they may, for example,
accompany rendered discourse lacking a (verbal) source (Sonnenhauser 2010; cf.
also Daiber 2010). Sonnenhauser, for example, describes them as delocutive particles which do not serve to convey reported speech, as is commonly assumed,
but rather to present speech and ascribe it to some other person (2010:405).
Nevertheless, these particles frame a stretch of discourse as speech and ascribe it
to another source (i.e. principal). As will be shown in the main section, just one
or few of these cues suffice to accomplish a shift of footing, animate a character
. N.B. The concept captures direct as well as indirect reported speech. Keeping in mind that
in conversations there is a continuum from direct to indirect reported speech as shown by
Gnthner (2000) on German data and that, as stressed by Lejnonen (1998) and Kitajgorodskaja
(1993), in colloquial Russian there can also be found hybrid forms mixing features of both; this
approach is even more appealing.
and stage an utterance from which the speaker can humorously detach himself
or herself.
Finally, in animated speech we do not deal with a neutral rendition of talk.
Interlocutors introduce reported speech in order to serve their own communicative purposes (Gnthner 1999:694, 696; Couper-Kuhlen 1999:17). Gnthner
(1999) argues that conversationally reported speech represents an instance of
Bakhtinian layering of voices or heteroglossia as one hears the animated voice
and the animators voice commenting on the reported speech. The reported
speech of the character blends with the reporters evaluation (ibid.:696). She
shows how speakers use mainly prosodic means to stage a character and to convey the animators meta-communicative evaluation of the reported speech, e.g.,
in complaints (ibid.:691696). The same holds true for humorously animated
speech, in which the animators voice and the animated voice overlap. This has
been captured by Sannikov (2003:475f.) in terms of bitextuality (bitekstualnost)
and by Zemskaja (1983:213) as a conversational instance of Bakhtinian polyphony. Kotthoff (1998) deals with (humorously and other) reported speech as staged
intertextuality and shows that the animators evaluative comment may be dissonant or consonant. The animator frequently indicates his or her humorous evaluation, e.g., by laughter, laughter particles and a smile voice articulation (cf. Chafe
2007:2551) or by exaggerating prosodic and stylistic features of the animated
figures speech. Thus in humorously animated speech the detachment is marked
in a twofold manner the speaker merely becomes the animator staging an utterance attributed to a figure and s/he produces a meta-communicative framing
which may reach from consent to humorous degradation or critical evaluation
(cf. Kotthoff 1998).
3. Forms of conversational humor drawing on animated speech

In this section we will show how animated speech contributes to the contextualization of a humorous modality and the production of different genres of conversational joking such as irony, parody or teasing. The analysis is based on data
from face-to-face conversations between Russian interlocutors stemming from
the Russian National Corpus, the ORD corpus and from recordings by the author.
The data selected for the analysis of humor from these sources mainly consist of
recordings from casual gatherings of friends or family members taped by one of
the interlocutors involved in the conversation. It will also be shown how this approach goes together with existing approaches especially to irony and sarcasm.
3.1
Parody
Conversational parody frequently relies on animated speech. In parody speakers

stage a character mainly by mimicking formal features of the assumed source
utterance. Kotthoff (1998:23) describes parody as evaluatively reported speech,
although she argues that the animators evaluation of the reported speech is more
or less consonant in parody. This is quite disputable, since parody is often employed in order to ridicule a staged character. As a result, Wilson and Sperbers
(1992:62f.) account of parody as an echoic allusion to linguistic form combined
with a dissociative attitude is probably more apt to capture conversational parodies. The argument that in parody the shift of footing is mainly accomplished by
a style shift, i.e. by mimicking linguistic form, is further supported by Zemskaja
(1983:180183) who describes how the speech mask strategy (priem reevoj maski) relies on a shift to another stylistic register in order to animate a specific type
of person. She adduces the example of a speaker using elements of prostoreie in
order to animate and parody a simple character (ibid.:183).
Example (1) illustrates that parody mainly relies on the mimicking of formal
features and that a dissociative attitude is involved as well. Sergej (Se) ridicules a
Ukrainian politician named Piskun for his poor linguistic abilities and his speech
habits by mimicking his use of suryk, a Russian-Ukrainian mixed language. The
animated speech consists of partially empty quotations (line 13) the focus is on
the linguistic form, not on the content. Sergej mimics mainly phonetic features of
suryk, when lending his voice to the virtual version of Piskun (line 8, 12 and 15).
The staged utterance contains a stereotype representation of an argumentative
pattern habitually used by the politician (This is all bullshit) (lines 8, 15). Apart
from code-switching, Sergej indicates the changed footing by a deictic shift including a significant shift of the vocal deixis. Animating the politician (p) Sergejs
voice becomes deeper.
(1) <<RU Fr 2005>>
1 Se: piskun.
piskun
2 takuju xujnJU
on
govorit vs vremja;
such
a bullshit hes talking all the time
3 [po televizoru=
on tv
4 O: [vot i]
well and
5 O: =kogo
oni dopustili do [ispolnenija]
who did they allow
to come into charge
6 Se:
[na surike.]
in surike
7 objazannostej. (0.5)
8 Se: p<<deeper voice>durnja
vs c>.
bullshit is all this
9 on skazal;
he said
10 [kak-to tam to-to somehow or other something
11 O: [<pp>mhimhimhim>]
<pp>mhimhimhim>
12 p<<deeper, smile voice>mni chovorjat
they tell me
13 po
and
tak
i
vot tak i
vot tak;
stuff and stuff
and stuff
14 pto ja vam i skau and i say to you

15 pdurnja
vs c>.
bullshit is all this
16 Ra: mhem[hemhem
mhemhemhem
17 O:
[mhemhemhe

mhemhemhe
Further, Sergej superimposes his own evaluative stance upon the animated politicians voice by a smile voice articulation which signals his non-serious attitude
which is reacted to by Raisa (Ra) and Olga (O). Their laughter reveals that they
share the same evaluative stance. The interlocutors join in ridiculing the animated
character.
3.2
Irony
Conversational irony provides empirical evidence for theories of irony that are
rooted in dramaturgy (Clift 1999:527) and account for it in terms of quoting, echoing or staging an utterance. According to Sperber and Wilsons revised
version of the echo-mention theory, in irony the speaker echoes an implicitly
. Clift (1999:526528) discusses Sperber and Wilsons (1981, 1992) echo-mention-theory,
Clark and Gerrigs (1984) pretense theory and Haimans (1990, 1998) approach to sarcasm as
theatre as belonging to this family of theories.
attributed opinion, while simultaneously dissociating herself from it (Wilson

and Sperber 1992:60). Kotthoff (1998:1217) treats irony as a specific form of
staged intertextuality in which the ironic stretch of talk is attributed to another
source, whereas the ironist superimposes his own dissonant or contrasting voice
upon it, so that there is an evaluative cleft: Irony stages two voices which evaluatively oppose one to another (Kotthoff 1998:23). Clifts (1999) proposal applies
Goffmans distinction of participation roles. The ironic speaker ostensibly denies
authorship of what he is saying () and lays claim only to animators status. ()
But in the very process of shifting footing, the ironist frames what is said, thus
becoming principal of an outside framing meaning (ibid.:533). This outer
frame conveys the ironists evaluative stance. But: Evaluation () is implicit in
the framing that characterizes irony (ibid.:538). Using the example of conversational data, Clift (1999:541f.) shows that ironic utterances are used in evaluative
slots, whereas the irony produces a mismatch between what is by default expected
to fill the slot and the item actually filling the slot: It is the stark disjunction
that contributes to the visibility of the frame and allows for the identification of
the irony (ibid.:542). This explains why the detection of irony need not rely on
conventionalized linguistic or paralinguistic cues (ibid.). It is also in line with
Kotthoff, who argues that in irony voice attribution is implicit (1998:23). In being ironic, speakers rely on their recipients to identify that the ideas conveyed by
the staged utterance originate from another principal and that the current animator holds a different (i.e. dissonant) view. That is why ironically staged utterances
frequently do without mimicking vocal or verbal features of the source utterance
and may be unannounced. Irony is a highly inferential mode of communication,
presupposing interlocutors with a strong common base of shared knowledge and
mutual knowledge of value judgments.
Ironically staged utterances may derive from different sources. Sometimes
speakers render ironically what an interlocutor has said a few turns ago. A prime
example of ironic echo-mention (Sperber and Wilson 1981) stems from an interview with Boris Nemcov, during which a guest in the studio addresses the strong
Turkish influence on the Crimea (a na samom dele tam oen silno vlijanie Turcii). Nemcov ironically alludes to this by echo-mentioning the assessment in his
reaction: z l o v e e e v l i j a n i e t u r o k ja by ne preuveliuval (I wouldnt
overdraw the o m i n o u s Tu r k i s h i n f l u e n c e ) [emphasis added]. Despite
the slightly changed verbal design, one can easily recognize that Nemcov ironically animates the opinion of the member of the studio audience not agreeing
with his evaluation.
. <http://www.echo.msk.ru/programs/opponent/612946-echo.phtml> (latest access
04/08/2011).
However, what is ironically rendered may not be attributable to any specific

person, but merely to a type of person, or people in general; it may be merely a
cultural aspiration or norm (Wilson and Sperber 1992:60). In the next example
the ironist animates a motto (m) well known from incentives and competitions
raising interest in ones home town Know and love your city (znaj i ljubi svoj
gorod). It is sufficient that the recipients detect the quote in order to discover the
irony. The interlocutors get lost during a walk through their home town, which
demonstrates that they do not know and love their city. Nevertheless, the ironist
marks her dissociative attitude additionally by attaching laughter particles to
thequote.
(2) <<ORDS01_02>>
1 I1: []
2 =a
voobe kuda my vyjdem,
and actually where will we arrive
3 esli pojdm vot TAK.
when we go there
4 (1.75)
5 da::;
well
6 mznaj i
ljubi svoj gorod.
know and love your city
7 <<p>xaxa[xa .x>
hahahaha .h
8 1:
[<<p>:::=

well
9 =my sejas (.) primerno we are now
about
Since Russian has particles labeled e.g., as evidential, quotative, delocutive that
signal that a stretch of talk is attributed to another principal and that the current
speaker is merely the animator such as mol and deskat (Grenoble 1993; Baranov
1994; cf. Sonnenhauser 2010), speakers may employ them as well to mark humorously animated speech. Arutjunova (2000:447) and Lejnonen (1998:225) prefer
the term ksenopokazateli, a label that stresses that speakers relate discourse from
another source. Both mention that speakers can use them when they polemically
summarize the position of an opponent, i.e. when they render it and dissociate
themselves from it at the same time. The particles mol, deskat and de combine the
. Cf. Sonnenhauser (2010) for a critical survey of the literature.
evidentiality of hearsay with differing dissociative or evaluative attitudes towards

the rendered discourse. Plungjan (2008) argues that the degree of speakers (i.e.
animators) dissociation from the rendered text is strongest with de. De combines
the rendition of another position with an ironic comment (ironieskij kommentarij) on it by the speaker (i.e. animator) (ibid.:302f.). These evidential or delocutive particles do not occur in the analyzed data. Far more often the newly
born quotative marker tipa is used in humorously animated speech. According
to Daiber (2010:87), tipa signals that a chunk of discourse is rendered which is
typical for the staged situation. Daiber (ibid.) stresses that quotations introduced
by tipa usually resemble direct reported speech. He expands Plungjans scale by
tipa (mol deskat de tipa) and argues that the reporting speakers dissociation is growing from the left to the right and that the rendered discourse they
accompany shifts from indirect to direct reported speech in the same direction
(ibid.:8384, 87).
Tipa is frequently involved in the humorous animation of another character
in the analyzed casual conversations. Irony is just one form of humor to which it
contributes. In the following example Viktor has just confessed that he blew off
an exam on music theory which elicits a comment by Jana. She ironically relates
what a friend (f ) of hers said in the same situation. Jana lends her voice to this
friend, animates him and stages his utterance. The animated speech is introduced
by a reporting verb (govorit) and by tipa (line 4). It is characterized by a deictic
shift and resembles direct reported speech. Jana tries to imitate what the friend
said but at the same time dissociates herself from his explanation. His absolute
pitch or creative talent is not the reason for his giving up (lines 7, 8). Tipa as well
as the laughter following signal her disaffiliation with his explanation. In joining
in, Viktor shows that he aligns to her evaluation of this explanation as odd.
(3) <<NKruJa Conversation about music (2006)>>
1 Viktor: - / ne / na solfedio zabil.
wait / no / I blew off music theory.
2 Jana: K? to glavnyj predmet? Ili /
how? but this is the main subject? well /
. In this context Dedais (2005) analysis of the Croatian parenthetical particle toboe as an
irony marker is very interesting. She summarizes the procedural meaning of toboe as follows:
(1) it redirects the attribution of the content of the utterance away from the speaker to someone else; (2) the speaker dismisses the truthfulness of such utterance; (3) the speaker does it
with an ironic overtone, which conveys mockery and disdain (ibid.:672). In terms of our account, toboe accomplishes a shift of footing and conveys the animators critical dissociation
from the rendered stretch of talk.
3 kak u menja znakomyj /

this is just like with an acquaintance of mine /
4 to saksofonist / na tenore to / govorit / tipa /
a saxophonist too / tenor too / says / like
5 fpril i srazu vrubilsja i ponjal /
i came, tuned in and understood /
6 fto slikom lgkaja dlja menja nauka /
that this subject is too easy for me /
7 ftem bolee / fs moim matematieskim skladom uma /
moreover / with my mathematically inclined mind /
8 ftvoreskoj naturoj i absoljutnym sluxom! [Smex]
my creative talent and absolute pitch [laughter]
9 Viktor: [Smex] Ne / prosto tam sovsem deti sideli
[laughter] no / there were just only kids around
10 i mne kak-to s nimi bylo ne v kajf.
and i didnt have fun with them.
3.3
Teasing
Staged utterances can also play a crucial role in teasing, a form of playful aggression. Teasing frequently relies on rendering the position (or action or behavior)
of the teased interlocutor in an exaggerated way in order to ridicule him or her in
a playful way (Gnthner 1996:8492; Kotthoff 1998:1316; Holt 2007:76f.). The
speaker may overdo the content as well as the verbal design of the position and ascribes this exaggerated position to the victim of the teasing, whose positive face is
damaged by this attribution (Gnthner 1996:92). The teasing speaker lends his or
her voice to a virtual version of the teased interlocutor and renders the exaggerated position. In teasing, overt exaggeration and speakers detachment by staging
an utterance are important to frame the attack as humorous (ibid.:87f.).
In the next example this is done by Vasilisa (Va) who mocks Oksanas (Ok) enthusiasm for a certain kind of tea. Oksana has just told that she especially went to
. Nevertheless, it is important to stress that teasing can as well be established in alternative
ways without the use of animated speech. Exaggeration, though, remains essential. In Drews
(1987) analysis of reactions to teases he observes that speakers rely on exaggeration in order
to signal that a stretch of talk is intended to be a tease. In his examples the words to depict
something/somebody/some action are exaggeratedly gross [] Each of those lexical choices is
therefore immediately recognizable as an exaggerated or extreme version, and by virtue of that
as not meant seriously to apply (1987:231). Teases are constructed as very obviously exaggerated versions of some action etc.; and/or by being in direct contrast to something they both
know or one has just told the other (ibid.:232).
a certain caf named le where it was served (line 6). This gives rise to a wordplay based on paronymy which accomplishes a script switch (Attardo 1994:195
229) from food and beverage to sex: Vasilisa humorously misunderstands Oksana
by adding the similarly sounding phrase in a negligee (v neglie) (line 7). Once
this script switch is realized, Vasilisa starts teasing Oksana by animating a virtual
version of her (Ok). The words she puts in the mouth of this avatar extremely
overdo her liking of the tea and further establish the sexual script. Vasilisa playfully insinuates that Oksana gave herself for the tea (ja otdalas dva raza za tot
aj) and alludes that it was paid for by sexual services offered by Oksana (so
mnoj rasplaivalis tim aem) (lines 11, 19/20). In addition, the teasing turns
are prosodically overdone by an extremely drawling articulation. In effect there is
exaggeration in prosody and in content.
(4) <<Ru B 2009>>
1 Ok: to moj ljubimyj aj;=
this is my favorite tea
2 =to moj ljubimyj aj;
this is my favorite tea
3 tam v le chodi[la radi togo aja.
i went to leger because of this tea
4 Af:
[(
5 tolko u nas (tuda) here only (there)6 v klub h [v le to the club to leger
7 Va:

[v neglie.=
in a negliglee
8 ?: =mhe[hehehehehehehehehehe[he
mhehehehehehehehehehehehehe
9 ?:

[<<laughing>v neglie>
in a negliglee
10 Ok:

11 Va:

[le.=
leger
Ok=<<drawling>ja
otdalas dva ra[za za tot aj>;

i gave myself two times for this tea
12 FF:

13 [hahahahahahaha
hahahahahahaha
14 Ok: [<<laughing>knajpa>;

the pub
[hahahahahahahaha
hahahahahahahaha
15 vo limberge;
in limberg
16 rjadom s uni kaf;
close to the uni cafe
17 le.
leger
18 nazyva[etsja;
its called
19 Va:

Ok[<<drawling>so=m
so mnoj rasplaivalis
by=m by me this tea was paid
20 [tim aem>.
for
21 FF: [mhehehehehehehe
mhehehehehehehe
22 Ok: ja tuda chodila i went there
23 []
Oksana as the victim of the tease reacts to it with a po-faced receipt (Drew
1987), which indirectly addresses the tease as a face attack. By not joining the
laughter and by offering a repair providing the exact location of the caf, she does
not affiliate with the other interlocutors.
3.4
Fictional scenarios
Apart from the performance of canned jokes, in which reported speech is an essential part, speakers also animate characters lending their voices to them in forms
of conversational humor that lack the label of a specific genre. In lack of a folk notion to denote the genre, these forms of humor have been described as comical
hypothetical (Winchatz and Kozin 2008), joint construction of humorous fictions (Kotthoff 2009) or as joking hypothetical scenario (Holt 2007). These are
forms of humor in which speakers often jointly with their interlocutors create a
fictional scenario. This scenario is not based on reports or past events, it is purely
fictional and a kind of ad hoc creation. In these genres the preceding discourse
serves as a point of departure from which the interlocutors, in co-constructing
a fictional scenario, stepwise move apart. The formerly serious topic is gradually
turned into some absurd, funny and unrealistic scene. The interlocutors involved
in the staging of a fictional scenario may switch their role and animate a figure
of the scenario by lending their voice to a virtual and created character. Thus
animated speech occurs in the joint construction of humorous fictions (Kotthoff
2009) when interlocutors cooperate in the development of a shared fantasy by

animating mostly fictional characters and freely adding comments of these virtual
characters (ibid.:204, 206). This rarely occurs in the comical hypotheticals analysed by Winchatz and Kozin (2008). The authors only provide one example when
a speaker jumps directly into playing an imaginary character, thus spontaneously initiating the CH [i.e. comical hypothetical, N. Th.] (ibid.:392). In contrast
to this, animated speech plays a crucial role in the joking hypothetical scenarios
analyzed by Holt (2007). She shows how speakers jointly expand episodes with
reported speech by using enactments, which transforms them into joking hypothetical scenarios. The enactment continues the sequentially preceding reported
scenario suggesting something that could happen, or is happening, but which is
not a serious proposal (ibid.:62). Characters animated in the joint construction
of humorous fictions (Kotthoff 2009) or in joking hypothetical scenarios (Holt
2007) may depict real life persons, stereotypical figures or fictional characters.
Animators frequently portray them by mimicking typical features of their speech.
As observed by Holt (2007:5457), enactments often lack an introductory reporting frame as they continue a reported interaction begun in a previous turn (57).
As Winchatz and Kozins (2008:392) example and our own data show, the same
holds true when there is no reported speech preceding the fictional scenario.
In the data analyzed, speakers employ animated speech in funny illustrations
that stage a position seriously presented in the preceding discourse in order to
contest and ridicule it. The following enactment then serves to reveal the absurdity of the position that is humorously contested by the scenario. This is in line
with Winchatz and Kozin (2008:389), who stress that the present conversational
topic, or a component thereof may give rise to a comical hypothetical. In example (5) Viktor (V) claims that the Discovery Channel secretly campaigns the
American way of life (lines 1, 2). Nadja (N) reacts to this by evoking a Discovery Channel documentary and animating a host (h) who invites the audience to
approach some little tigers (lines 10, 11). Subsequently Viktor himself joins the
scenario by taking over the role of the host and expanding the animated speech
(lines1317). The words he puts in the mouth of the virtual host illustrate the
absurdity of mapping a political and cultural antagonism onto wildlife comparing
crappy Siberian tigers with good American tigers (lines 13, 14, 16, 17).
(5) <<RU Char 2008>>
1 V: koneno disKAvri-=
of course Dicovery Channel
2 =to Oen skrytaja propaganda
thats very secret propaganda
3 amerikansko[go ] obraza Izni.

of the American way of life
4 S:

[net;]
no
5 S: est momenty there are moments

6 kogda diska[vri pokazyvaet PRAVdu.
when Discovery Channel shows the truth
7 P:
[(
)]
8 V: to EST-=
that means
9 N: =DA.
yes
10 h<<higher voice>davajte poprobuem podobratsja k

come on lets try to come closer
11 htim tigrjatam pobLI:e>.
to these tiger cubs
12 (1.0)
13 V: hvot usuRIJskie tigry;
those are the Siberian tigers
14 honi xu[vye.
theyre crappy
15 ?:
[((coughs))
16 ha vot ameriKANskie tigry;

and these are the American tigers
17 honi xoRO:ie.
theyre good
18 ?: .hh
19 N: mhemhemhemhem
mhemhemhemhem
20 P: to nimal plnt.
thats Animal Planet
21 (2.0)
The staging is solely indicated by a deictic shift including the vocal deixis (at least
with Nadja) which signals that a typical host of the Discovery Channel is animated. Kotthoff refers to these unannounced chunks of animated speech in which
voice attribution is implicit as pseudo-quotations (1998:23), as the exact source
or principal is not of primary importance and may rather be a type. With jointly
constructed fictional scenarios in mind, she later describes these instances as
category animations referring to Deppermann (2007:336), as they represent ways

of speaking that are typical of a certain type of speaker (Kotthoff 2009:206).
This is also the case in the next example offering an instance of an absurd illustration accomplished by animated speech. Oksana has just explained that Israeli
banknotes maintain their value as long as a metallic strip incorporated in the paper is intact. Then the National Bank would convert it. Subsequently Igor (I) animates a man (m) who wants his metallic strips to be converted, offering an absurd
explanation why he cut them off: money took too much space, so I just cut off the
strips (dengi mnogo mesta zanimali; tak ja tolko polosoki vyrezal) (lines3,
4, 6). The staged utterance is introduced by the quotative marker tipa indexing
that the animated speech represents what is typically said in such a speech event
(Daiber 2010:84). N.B., tipa additionally signals that the upcoming animated
speech is a category animation in the sense of Kotthoff (2009) and Deppermann
(2007). The smile voice articulation as well as tipa signal that the animator Igor
dissociates himself from the rendered pseudo-quote (cf. Daiber 2010:87).
(6) <<Ru Fr 2004>>
1 Ok: no- ty moe v nacionalnyj bank prijti well you can go to the national bank
2 Ok: [tebe pomenjajut.
they convert it for you
3 I: [tipa like-
m<<smile
voice>dengi mnogo mesta

money took too much
4 mzanimali; tak ja tolko polosoki vy[rezal;

space so i just cut of the strips
5 Iv:

[mhemhem
mhemhem
6 I: mteper mne vs obmenjajut>.

now please convert it all for me
7 Iv: mhemhem
mhemhem
Both examples show how the animation of virtual and typical characters and the
staging of their utterances cause a transition from serious to humorous discourse
evoking an absurd, fictional scenario contesting a serious presentation or explanation preceding.
In all these forms of humor, animated speech functions as a contextualization cue (CC) used in order to establish a humorous discourse modality. In the
next section we will address the specific nature of this CC and raise the question
whether there is a motivational link between the indexed meaning, the metamessage This is play, and the CC.
4. Is animated speech a natural contextualization cue?

Gumperz defines contextualization cues (CCs) as constellations of surface features of message form () by which speakers signal and listeners interpret what
the activity is, how semantic content is to be understood and how each sentence
relates to what precedes and follows (1982:131). CCs provide interpretive frames
that signal how an activity is to be interpreted. Animated speech as a CC is a surface feature which can signal that the rendered stretch of discourse is not meant
seriously. The animator detaches himself or herself from what s/he renders by
shifting the footing and animating a figure who utters the words. The CC points
to the meta-message This is play or I dont mean this.
Contextualization theory claims that CCs do not have referential (decontextualized) meaning of the kind we find in lexical items (Auer 1992:30). Gumperz
explains that they carry information although meanings are conveyed as part
of the interactive process. () the meanings of contextualization cues are implicit. () Their signaling value depends on the participants tacit awareness of
their meaningfulness (1982:131f.). CCs do not contain meaning. They index
how something is to be interpreted by guiding recipients inferences. Use and
interpretation of CCs is conventionalized within a speech community which may
cause misunderstandings in inter-group communication (ibid.:133135). Unlike
other CCs, the relation between animated speech and the indexed meaning, the
meta-message This is play, is not as arbitrary and conventionalized as initially
conceived in contextualization theory.
We will argue that there is a motivational link between the CC animated
speech and its meaning. Both are materially connected taking the words of
another person and playing with them by quoting, mimicking or exaggerating
them is literally playing with them. Animating a character and staging his or her
utterances does two things in one it is play and it conveys the meta-message
This is play. Auer allows for the inherent meaning potential of a CC to be
conventionalized (arbitrary) or natural (non-arbitrary), or a mixture of both
(1992:32). Moreover, many CCs have a [sic] inherent meaning base [] which
is only partly, or not at all, conventionalized but, in a specific sense of the word,
natural (ibid.:33). Auer adduces the lowering of F0 that signals the end of a unit
in talk-in-interaction as such a natural CC, as it is based on an iconic-metaphorical relationship between bottom/rest/termination (ibid.). Animated
speech relates to the meta-message This is play in an analogical way. In staging an utterance and animating a real, stereotype, virtual or fictional character,
speakers convey the meta-message This is play and play a role. This argument
is further nurtured by scholars who describe forms of humorously animated
speech relying on the stage metaphor (e.g., Kotthoff 1998; Clift 1999; Zemskaja
1983:180186). It is probably strongest with Haiman (1990, 1998) who elaborates the stage metaphor in explaining how sarcasm is marked. Like every actor,
the sarcast has a divided self, existing both as the performer and the persona the
character portrayed. The sarcast is a disdainful playactor who advertises his or her
insincerity by self-consciously keeping the former and the persona alive, distinct
and opposed (1998:61). According to Haiman, stage separators indicate that the
speaker is playing a role. The overt and separate metamessage is the stage separator which alerts the listener or reader that what is said is in some way meant in
jest or is pretended (ibid.:27). He further argues that linguistic, pragmatic and
paralinguistic devices encoding the sarcastic meta-message, the stage separators,
are iconically motivated (1990:181). Irrespective of the medium, irrespective
even of the code, the sarcastic metamessage (I dont mean this message or Im not
serious) is most frequently signaled by a very limited set of highly iconic gestures
(1998:28f.). These iconic gestures or stage separators indicating the meta-message
I dont mean this (Haiman 1998:3060, 1990:188202) resemble and partially
overlap with the strategies that indicate a shift of the footing and contribute to the
humorous animation of a figure. Haiman largely divides them into three groups:
formal indices of direct quotation or repetition, incongruity between segmental and suprasegmental texts and hyperformality (1990:181). All of them signal that an utterance is double-voiced and accordingly that the voice of the figure
and the voice of the animator overlap in it. However, Haiman is largely consumed
with those forms of animated speech in which the animator expresses dissent
with the position of the animated character.
5. Conclusion
Animated speech is a useful cover term to capture a widely used CC to indicate
a humorous modality in different forms of conversational humor such as parody,
irony or teasing. It further plays a crucial role in other forms of conversational
humor that lack a culturally well-defined genre label and which are still looking
for in-depth analyses and genre-like linguistic description. In each form of joking
other aspects may be highlighted when humorously rendering discourse. Parody focuses on humorously mimicking linguistic and prosodic form. Its success
depends on the recipients recognizing the source judging from the exaggerated
. Goffman (1975:516523) himself also frequently employs the stage metaphor in describing
utterances with a shifted footing, i.e. animated speech.
. Haiman (1990, 1998) chooses the term sarcasm in order to denote instances of volitional
irony. It is not restricted to aggressive forms of irony as elsewhere.
imitation of formal features and enjoying the performance. In ironically animated

speech the rendered content is crucial. Formal features of the source may be replicated or not. Frequently the source utterance is reformulated and its wording
is changed. The ironist relies on his or her recipients to infer the principal of the
staged utterance and his or her own dissonant point of view. In teasing as well as
in fictional scenarios there is no source or template. The staged utterance is purely
fictional and the animated character is a creation of the animator. Though it may
correspond with real-life persons and represent a certain social type or an exaggerated version of an interlocutor. All these genres of conversational humor share
the same CC which can be accounted for in terms of a shifted footing (Goffman
1992). In rendering others discourse and at the same time dissociating themselves
from it, speakers convey the meta-message This is play. In staging an utterance
they also become an actor, they indeed play a role and animate an avatar. One can
therefore assume a naturalistic or iconic relationship between the indicated metamessage (This is play) and the cue as such in the case of animated speech.
Data
ORD Odin Reevoj Den (One day of speech) corpus (2007), collected by A. S. Asinovskij and
his team, Saint Petersburg State University.
NKruJa Russian National Corpus, subsection: spoken interaction <http://www.ruscorpora.
ru/search-spoken.html> (latest access 21/02/2009).
Recordings of the author (RU 20052010).
References
Arutjunova, Nina D. 2000. Pokazateli uoj rei de, deskat, mol. K problemam interpretacii reepovedeneskich aktov. In Jazyk o jazyk, Nina D. Arutjunova (otv. red.), 437449.
Moscow: Jazyki Russkoj Kultury.
Attardo, Salvatore. 1994. Linguistic Theories of Humor. Berlin/New York: Mouton De Gruyter.
Auer, Peter. 1992. Introduction: John Gumperz Approach to Contextualization. In The
contextualization of language, Peter Auer and Aldo diLuzio (eds.), 137. Amsterdam/
Baranov, Anatolij N. 1994. Zametki o deskat i mol. Voprosy Jazykoznanija 4: 114124.
Bateson, Gregory. 1972. Steps to an Ecology of the Mind. San Francisco: Chandler.
Bolden, Galina. 2004. The quote and beyond: defining boundaries of reported speech in conversational Russian. Journal of Pragmatics 36: 10711118.
Chafe, Wallace. 2007. The Importance of Not Being Earnest. The Feeling behind Laughter and
Humor. Amsterdam/Philadelphia: John Benjamins.
Clark, Herbert and Gerrig, Richard J. 1984. On the Pretense Theory of Irony. Journal of Experimental Psychology: General 91: 121126.
Clift, Rebecca. 1999. Irony in conversation. Language in Society 28: 523553.

Coulmas, Florian. 1986. Reported speech: Some general issues. In Direct and Indirect Speech,
Florain Coulmas (ed.), 128. Berlin/New York/Amsterdam: De Gryuter.
Couper-Kuhlen, Elisabeth. 1999. Coherent Voicing: On prosody in Conversational Reported
Speech. In Coherence in Spoken and Written Discourse, Wolfram Bublitz, Uta Lenk and
EijaVentola (eds.), 1133. Amsterdam/Philadelphia: John Benjamins.
Daiber, Thomas. 2010. Quotativmarker im Russischen (/). Zeitschrift fr Slawistik
55 (1): 6989.
Dedai, Mirjana. 2005. Ironic Denial. Toboe in Croatian Political Discourse. Journal of Pragmatics 37 (54): 667683.
Deppermann, Arnulf. 2007. Playing with the Voice of the Other: Stylized Kanaksprak in Conversations among German Adolescents. In Style and Social Identities, Peter Auer (ed.),
325361. Berlin/New York: Mouton De Gruyter.
Drew, Paul. 1987. Po-faced receipts of teases. Linguistics 25: 219253.
Dynel, Marta. 2011. Joker in the pack. Towards determining the status of humorous framing
in conversations. In The Pragmatics of Humor across Discourse Domains, Marta Dynel
(ed.), 217242. Amsterdam/Philadelphia: John Benjamins.
Goffman, Erving. 1975. Frame Analysis: An Essay in the Organization of Social Experience.
Harmondsworth: Penguin.
Goffman, Erving. 21992. Forms of Talk. Philadelphia: University of Pennsylvania Press.
Grenoble, Lenore. 1993. Evidentiality and footing in Russian. In Harvard Studies in Slavic
Linguistics 2, Olga T. Yokoyama (ed.), 1541. Cambridge: Pusey Library.
Gumperz, John. 1982. Discourse Studies. Cambridge: Cambridge University Press.
Gnthner, Susanne. 1996. Zwischen Scherz und Schmerz. Frotzelaktivitten im Alltag. In
Scherzkommunikation. Beitrge aus der empirischen Gesprchsforschung, Helga Kotthoff
(ed.), 81109. Opladen: Westdeutscher Verlag.
Gnthner, Susanne. 1999. Polyphony and the layering of voices in reported dialogues: an
analysis of the use of prosodic devices in everyday reported speech. Journal of Pragmatics
31: 685708.
Gnthner, Susanne. 2000. Zwischen direkter und indirekter Rede. Formen der Redewiedergabe in Alltagsgesprchen. Zeitschrift fr germanistische Linguistik 23 (1): 122.
Haiman, John. 1990. Sarcasm and theater. Cognitive Linguistics 12: 181205.
Haiman, John. 1998. Talk Is Cheap: Sarcasm, Alienation, and the Evolution of Language. Oxford/
New York: Oxford University Press.
Holt, Elisabeth. 2007. Im eying your chop up mind: reporting and enacting. In Reporting Talk. Reported Speech in Interaction, Elisabeth Holt and Rebecca Clift (eds.), 4780.
Kitajgorodskaja, Margarita V. 1993. uaja re v kommunikativnom aspekte (na materiale
ustnych tekstov). In Russkij jazyk v ego funkcionirovanii, Elena Zemskaja and Dmitirij N.
melev (otv. red.), 6585. Moscow: Nauka.
Kotthoff, Helga. (ed.). 1996. Scherzkommunikation: Beitrge aus der empirischen Gesprchsforschung. Opladen: Westdt. Verlag.
Kotthoff, Helga. 1998. Irony, Quotation, and other Forms of Staged Intertextuality. Double
or Contrastive Perspectivation in Conversation. InLiSt 5 <http://www.inlist.uni-bayreuth.
de/issues/5/Inlist5.pdf> (latest access 4/08/2011).
Kotthoff, Helga. 2009. Joint Construction of Humorous Fictions in Conversations. Journal of

Literary Theory 3 (2): 195217.
Lejnonen, Marja. 1998. uaja re v ustnoj kommunikacii. In Liki Jazyka. K 45-letiju naunoj dejatelnosti E. A. Zemskoj, Marina Ja. Glovinskaja (otv. red.), 223232. Moscow:
Nasledie.
Mayes, Patricia. 1990. Quotation in spoken English. Studies in Language 14 (2): 325365.
Plungjan, Vladimir. 2008. O pokazateljach uoj rei i nedostovernosti v russkom jazyke:
mol, jakoby i drugie. In Lexikalische Evidenzialittsmarker in slavischen Sprachen, Bjrn
Wiemer and Vladimir Plungjan (eds.), 285312. Mnchen/Wien: Kubon & Sagner.
Sannikov, Vladimir Z. 22003. Russkij jazyk v zerkale jazykovoj igry. Moscow: Jazyki Slavjanskoj
Kultury.
Selting, Margret et al. 1998. Gesprchsanalytisches Transkriptionssystem (GAT). Linguistische Berichte 173: 91122.
Sonnenhauser, B. 2010. Rededarstellung im Russischen: mol und deskat. Zeitschrift fr slavische Philologie 67 (2): 377405.
Sperber, Dan and Wilson, Deidre. 1981. Irony and the use-mention-distinction. In Radical
Pragmatics, Peter Cole (ed.), 295380. New York: Academic Press.
Tannen, Deborah. 22007. Talking Voices: Repetition, Dialogue, and Imagery in Conversational
Discourse. Cambridge: Cambridge University Press.
Wilson, Deidre and Sperber, Dan. 1992. On verbal irony. Lingua 81: 5376.
Winchatz, Michaela and Kozin, Alexander. 2008. Comical hypothetical: arguing for a conversational phenomenon. Discourse Studies 10 (3): 383405.
Zemskaja, Elena A. 1983. Jazykovaja igra. In Russkaja razgovornaja re. Fonetika Morfologija Leksika est, Elena A. Zemskaja (otv. red.), 172214. Moscow: Nauka.
Appendix
Transcription conventions adapted from Selting et al. (1998)
[
[
=
(.)
(-), (--)
(2.0)
:, ::, :::
h, h
.h, .hh
h, hh
(h)
haha, hehe
hoho
akCENT
overlapping, simultaneous talk

latching between two units
micro pause
shorter pauses (0.250.75 sec.)
estimated duration of pauses longer than 1 sec.
lengthening of a syllable, according to its duration
hesitation marker
breath in, according to its duration
breath out, according to its duration
laughter particle inserted in speech
syllabic laughter, according to the quality of the vowel
strong accent, higher volume than surrounding speech
pitch contour on the syllable carrying the intonation contour

?
high rising
,
rising to a mid-level
-
constant
;
falling to a mid-level
.
deep falling
((coughs))
para- and nonlinguistic actions and events
<<laughing>> para- and nonverbal actions and events accompanying speech and their
scope
<<drawling>>
interpretive comments and their scope
( )
incomprehensible episode
(tak)
assumed wording
Part V
Language alternation in face-to-face

interaction of bilingual families
Bilingual language use

in the family environment
Evidence from a telephone conversation
between members of a community
of speakers of German descent
Veronika Ries
There are various studies investigating the speech community of people of German descent from Russia (Russlanddeutsche) and their language use. In my data,
I find various language contact phenomena, most remarkably code-switching.
I examine how the Russlanddeutsche alter their speech according to its possible
functions in discourse. Thus, my qualitative study focuses on the morpho-syntactic and interactional linguistic analyses of spoken data in natural contexts
and situations. In order to thoroughly answer the complex question How do
speakers use the languages available to them? I have taken the following sociocultural factors into account: the sociolinguistic aspects of the interlocutors taken from their biographies, especially information about the preferred language
of the speakers etc.
Keywords: language contact phenomena, family conversation, natural language
use, Russlanddeutsche
1. Introduction
The present contribution analyzes the natural and spontaneous language use of
the speech community of German descent from Russia (in the following: Russlanddeutsche). The major concern of this qualitative study is the analysis of authentic spoken data, focusing mainly on their morpho-syntactic and interactional
. The German term Russlanddeutsche is used in this context because there is no established
English equivalent.
282 Veronika Ries
linguistic aspects. The data, which were collected, taped and transcribed by myself,
present various language contact phenomena, the most substantial among them
showing code-switching. The instances of code-switching are usually restricted to
one sole statement; they can consist of different levels of (morphological) integration. Using my observations of the group of Russlanddeutsche and their natural
language use, I examined these language alternations with special regard to their
function in discourse. My analysis answers the following key questions:
How do speakers use the linguistic resources available to them (i.e. German
and Russian) during a family conversation?
What effects do language alternations have on the conversation?
A complete answer to these questions requires that the speakers socio-cultural
background be considered as well. For example, language use can be affected by
the dominant or preferred language of the speaker. Therefore, in addition to the
conversation data, the speakers linguistic history is also a relevant variable. One
crucial aspect is the identity of this group. The factor of identity is too complex
to be discussed within the scope of this article, but it must be considered as an
influential criterion of an ethnic group which has such unique identifying features. The Russlanddeutsche have in common the German descent of their ancestors and the German language, on the one hand, as well as their Russian/Kazakh
birthplace, the Russian language and different factors, on the other.
Here, I use the terms bilinguals and bilingualism to describe the speakers
and their language, whereas bilingual is used to refer to the use of two languages
(cf. Auer 1984). Furthermore, I use multilingualism to refer to specific dialects
and idioms used by the speakers.
2. Theoretical background
Many different studies describe the use of Russian and German by the Russlanddeutsche (for example, cf. Berend 1993, 1998; Meng 2001, 2003). Dietz, for
instance, found a discrepancy between speakers self-reported linguistic habits
and competence and their observed language use and competence (cf. Dietz
1995). Aspects such as fluency, however, are irrelevant for this present study, because I am interested in the natural and spontaneous language use and not in
linguistic competence. In the context of language use, questions such as Which
language do the speakers actually use? and Which factors are crucial in the
present situation? need to be answered. Meng has examined the language use of
Russlanddeutsche in several of her studies (for example, cf. 2003). In her study
on pre-migration language, she observed a change between the generations of
Bilingual language use in the family environment 283
a bilingual family communication (cf. Meng 2001, 2003). Meng argues that the
change between generations and the multilingual situation within the family results from the different first languages of the generations (cf. Meng 2001:82ff.).
Risse and Roll also show a correlation between generation and language use and
relate this to language competence (cf. Risse and Roll 1997:201). Other studies
examine how speakers negotiate the meaning of mother tongue (for example,
Rosenberg 1993, 1994).
The family is part of the so-called private domain (for an overview of the concept of domain, see Fishman 1967, 1970) and is characterized by privacy, intimacy,
and especially important in the context of Russlanddeutsche by the fact that
the family represents a space that is free from the socially normative influence of
German society. Consequently, the family is a place where the pressure to assimilate is absent. This suggests that a family environment allows individuals to speak
freely which, in turn, makes family dialogue an ideal context to observe natural
and spontaneous language. However, the family represents the first-socialization
for children (cf. Deprez and Persoons 1987:128; Segrin and Flora 2005:9). This
means that children acquire habits that are peculiar to their family especially in
the field of communicative behavior (cf. Fitzpatrick and Ritchie 1994:278). Due
to my longstanding friendship with the informants in this study, I am able to distinguish between a marked and an unmarked behavior.
A single unified group of Russlanddeutsche did not form until the beginning
of the 20th century. When German settlers initially came to the Russian Empire
in the 18th century, they were distinguished by religion, dialect etc. Subsequently,
the different groups of German settlers were divided into different villages by,
for example, their religion (Catholics, Protestants, Mennonites etc.). After World
War I and World War II, the German settlers became one group the Germans.
Due to their common nationality and owing to the animosity between the Soviet
Union and Germany during and after World War II, actions (like the forced deportation of entire villages (cf. Bohmann 1970:71)) were indiscriminately carried
out against all German settlers. As a result of the shared experiences of discrimination, a community spirit emerged and the people both saw themselves and were
seen as a collective. Besides these collective experiences, the experiences of individuals are incorporated in the speakers socio-cultural background information.
Studies of language alternations in conversations demonstrate the various
functions that these code-switches can fulfil. Appel and Muysken list six such
functions including referential and directive function (cf. Appel and Muysken
1987:118ff.). The referential function is caused by a gap in the lexicon or by a lack
of language competencies. The directive function enables the speaker to exclude
or include certain persons from/in the conversation (ibid.). Goffman and Auer
mention similar aspects regarding the functions and motivations for language
284 Veronika Ries
alternations. For example, both describe reported speech and communication partners as crucial factors in a conversation (cf. Goffman 2001:94f.; Auer
1995:120). A new function, namely segmentation, is analyzed and described in
this article.
3. Data and method

The audio corpus for this study consists of naturally occurring, everyday family
conversations from two extended families. All speakers in the corpus emigrated
from the former Soviet Union. At the time of recording, the persons were between
6 and 80 years old. These two families do not know each other. The recordings
were transcribed following the conventions of GAT the Gesprchsanalytische
Transkriptionssystem (Selting et al. 1998) and consist of approximately 90
hours of everyday conversations. Everyday conversations include talks at dinner,
family reunions as well as telephone calls. The recordings took place in Germany
at the families homes.
Throughout the descriptions and the analysis I refer to Peter Auers concept
of conversational code alternation (1995:116). Auer interprets bilingual phenomena within their sequential environment (1995:116; see also 2007:123f.).
According to Auer, the sequential environment covers the immediate preceding
turn and the subsequent utterance (ibid.). The bilingual phenomena could be a
reaction to the preceding turn and the subsequent turn shows how the speakers
interpret the respective utterances (ibid.).
The observation that language alternations are not random or ungrammatical (for example, cf. Fishman 2007; Myers-Scotton 1993) is very important as the
basis for my analysis. Thus, the very general question at the beginning is: Which
language is used and how do the languages appear in a given statement and in the
conversation examined?
4. Data analysis
The primary conversation analyzed here is a telephone call between the speakers
Aw and her father Fm. The dominant language of speaker Fm is German (and
Low German). Fms parents have consistently used German in spite of the language policy during and after World War II. Fm also preserved this tradition
. During and after World War II language prohibitions were pronounced the use of all
languages was forbidden, except Russian (for example, cf. Walth 1994:124).
when raising his own children and especially his first born Aw so that German
would be their dominant language. However, Russian soon became prominent in
Aws life since it was the primary language in school and in other social activities.
With the birth of Aws first child, she and her husband established German as the
primary language at home. However, owing to problems their child was confronted with, they chose Russian to be the first language for the second born child. At
the time of recording, speakers Aw and Fm are fluent in both languages German
and Russian. Speaker Fm states that he prefers German, because it is his native
language.
The weekend before this telephone call speaker Aw and her family visited
her parents, because her mother wife of speaker Fm had a birthday party.
Speaker Aw gave some presents to her mother and speaker Fm and wants to know
whether they liked them.
Example (1) Telephone call between Aw and Fm:
01 Fm:

02 Aw:

03 Fm:

04 Aw:

05 Fm:

06 Aw:

07 Fm:

08 Aw:

09 Fm:

10 Aw:

11 Fm:

12 Aw:

13 Fm:

[]
NAME
NAME
guten abend.
Good evening.
hallo?
Hello?
hallo guten abend
Hello, good evening.
nabend (.) hallo
Good evening. Hello.
na wie gehts bei euch?
So how are you?
gut
Fine.
gut?
Fine?
ja.
Yes.
na to vy smotreli to k emu tama?
Have you looked, who got what?
ja to tam?
Yes, but what is there?
nu (.) to tam to eh salz (.)
Well, there is salt.
a to est (.) nu da
Its there. Well, yes.
. Tama is a colloquial form of the adverb tam there.
286 Veronika Ries
34 Fm:

35 Aw:

36

37

[]
48 Aw:

49

50

51

52

53 Fm:
[]
58 Aw:

59

60 Fm:

61

62 Aw:

63 Fm:

64 Aw:
65 Fm:

66

67 Aw:

68 Fm:

na ja (.) und jetzt was gibts bei euch?

Well, and what (news) do you have?
ja nix (.) tak s=raboty prili (.)
Yes, nothing. We came home from work.
bisschen eingekauft (.)
We went shopping.
sejas pokuali vse
Now we finished dinner.
NAME4 govorit ona tam videla stroitelej nu
NAME4 said she has seen some construction workers there.
ne podxodila ona und eh to (.) a tak
Well, she didnt go there and so. But so,
my segodnja bystrenko- und kartoffelsuppe
today, we quickly-, and we cooked potato soup
gekocht mit fleischwurst pouinali
with sausage. We had dinner
i vs teper
and thats it now.
hm
na rabote stolko nabegalis (.) i vot
At work, we ran around so much, and so.
(.) na gut dann esli vs normalno [tak
Well then, if everything is all right, so
[ja ja
Yes, yes,
danke fr den anruf, alles in ordnung
thanks for calling, everything is fine.
ja bitte
Yes, youre welcome.
keine sorgen
Dont worry.
hm
wenn was nicht in ordnung ist werden wir
If anything is wrong, we will
anrufen
call.
ja gut einverstanden
Yes, fine, I agree.
gut tschss
Fine, goodbye.
1. As a first point, a representative case of code-switching is discussed. It is a

typical case and I would therefore like to begin with the brief analysis of a
clear example: It is the mentioning of the German lexeme salz (salt)
within an utterance in Russian (line 12). The area right before the lexeme
salz is important for the analysis: The filled pauses are an indication of a
word-searching-phase. Due to the fact that a German lexeme is used, its mentioning is interpreted as a temporary gap in the Russian lexicon. This gap is
then filled with a lexeme from the lexicon of the second language.
2. The first case of code-switching in the conversation concerns the opening of
the conversation. This is in German the official language of Germany, which
is therefore what you would expect. The first code-switch into the Russian
language is initiated by the daughter in her utterance to vy smotreli to
k emu tama? (line 10, Have you looked, who got what?). But why does
speaker Aw switch from German to Russian at this point of the conversation?
The question of speaker Aw na wie gehts bei euch? (line 06, So how
are you?) and the subsequent reply of speaker Fm gut (line 07, Fine.) form
a complete adjacency pair. Although this part is closed, the daughter repeats
the question in picking up Fms answer gut? (line 08, Fine?). With regard
to the dialogue, in fact, this question has the same function as the sequence
before the conversation opening. The answer of Fm is a short and confirming ja (line 09, Yes.). In line with the question, an initializing move in the
conversation, an answer is required. With the then following language change
(line 10), a change in topic also takes place. The new topic is the actual reason
for the telephone call, namely the birthday presents. Schegloff s description
of a telephone conversation opening serves as an orientation (cf. Schegloff
1986). Schegloff describes a set of four (essential) sequences:
a. the summons-answer sequence (the phone ring and hello)
b. the identification-recognition sequence (i.e. partners display each others
recognition of the other) (for example, hello Clara?/yeah)
c. the exchange of greeting tokens (hi/hi), and
d. the how are you-sequence (how are you/Im all right. how are you)
(ibid.)

Taking these sequences as a guide to canonical openings, it becomes clear that

the exchange in the how-are-you-sequence is missing. Thus, the expectations of the daughter are not satisfied and this results in her repeating a modified version of the same question. Neither a repetition nor a reformulation
leads to an exchange. The next possibility is a change of language. Because of
the contrastive mark of the language change, it is possible to get to the initially
planned topic without any introduction or entering passage. At this point, the
language change structures the conversation by closing the opening sequence
. This gap filling applies only in the actual situation. The gap in the lexicon is related to the
concrete conversation and not as a general result, substituted by a frequently used term from
their German surroundings.
288 Veronika Ries
and marking the topic sequence. Here, the change in language is a kind of
problem-solving-strategy, but the speaker is not aware of this strategic behavior. It is a linguistic behavior, which helps to bridge a problematic situation in
the conversation.
Another observation that can be added to the analysis is the following: Speaker Fm answers the questions of speaker Aw in a minimal way and without any
counter questions. Thus, the switch of language by speaker Aw definitely has
the purpose to get from the incomplete how-are-you-sequence to the initial
topic of the call; however, it is also an attempt to get speaker Fm to be more
talkative. At this point, the mentioned aspects (i.e. regarding the expectations
which are not confirmed) interact with the result that speaker Aw switches
from German to Russian.
3. The next example of code-switching is the switch to German by speaker Fm
with the question na ja (.) und jetzt was gibts bei euch? (line 34,
Well, and what (news) do you have?). At this point, a switch to the preferred
language at a special stage of the conversation can be observed: the preceding
topic is finished and the father takes the next turn. Thus, he is in the position
to begin a new topic as well as switch the code. With the introduction of a new
topic he also changes the language-of-interaction.
4. The answer of speaker Aw (line 3552) to the question of speaker Fm described above shows some language alternations. Is it just by chance or is
there a reason for these switches? Her immediate reply ja nix (line 35, Yes
nothing.) is also in German, and therefore in the same language chosen by
her father. The more extensive answer, which is a description of her day, is different: The day is divided into the single activities which are presented in an
alternated way by using the two languages in turn. Interestingly, Aw divides
up her day through the strategic selection of one language over another in her
narrative. The segmentation by language alternation starts right after her immediate answer ja nix (line 35) and looks as follows:
a. First activity: came home from work: tak s=raboty prili (line35)
b. Second activity: shopped: bisschen eingekauft (line 36)
c. Third activity: finished dinner: sejas pokuali vse (line 37)
And so on.
5. By means of detailed analyses other functions of language alternation beyond segmentation can be observed as, for example, in Aws utterance na
gut dann esli vs normalno (line 59, Well then, if everything is all
right, so.). Daughter Aw ends her descriptions and the topic is almost finished which is signalled by the following elements: pause, i vot (and so),
pause (line 58). The phrase na gut dann (well then) in German functions
as a marker indicating the beginning of a closing sequence. The phrase esli

vs normalno (if everything is all right) in Russian is not a continuation of the closing sequence with a random language change, but an offer for
another sequence and not the ending of the conversation. A Russian phrase
opens a sequence that returns back to the opening topics of the presents and
the health of her parents. Thus we have the following:
a. The German phrase na gut dann marks the beginning of the closing
sequence.
b. The Russian phrase esli vs normalno provides the opportunity to
leave the closing sequence and to return to the conversation.

The second language establishes a parallelism: The first language the language-of-interaction serves as the closing point for the conversation. The
second language serves to realize a side-sequence, so that there is no actual
interruption, but another level within the conversation.
6. My final section focuses on the opening and closing of the telephone call.
Both, the conversation opening and the conversation closing are in German
which begs the question; is this merely a coincidence? To be able to answer
this question, a step-by-step analysis of both sequences is needed. Independent of the language-of-interaction, the conversation development and the
actual chosen language, the conversation closing is in German. It is important
to take the general characteristics of opening and closing sequences into account in order to analyze this phenomenon:
a. Opening sequence: The opening sequence consists of certain steps like
greetings, how-are-you-phase, their exchange and so on. In the present telephone call, the greeting phase (line 0205) and the how-are-youphase (line 0609) are in German.
b. Closing sequence: According to Luke and Pavlidou, the fact that the
closing part of the telephone call is much less studied is linked to the
observation that the beginnings of closings cannot be readily found
(2002:11; see also Pavlidou 2002). The closing in the present telephone
call is analyzed as follows:
i. na gut dann (line 59): This phrase marks a pre-closing (cf. Pavlidou
2002:202), a beginning of the closing.
ii. ja ja danke fr den anruf alles in ordnung (line 6061):
This answer of speaker Fm indicates that he interprets the utterance
of speaker Aw as an offer to close the conversation.
iii. gut tschss (line 68): This final turn is the closure of the
conversation.
iv. Both, the pre-closing and the closing are in the German language.
290 Veronika Ries
A crucial finding is that these sequences constitute frames of conversations.

These frames are highly characterized by routines. Because of their frequency
in the German environment, these routines are ascribed to the German language. So habits are formed. Due to their frequency, the speakers are likely to
refer to these well-known linguistic patterns.
5. Conclusion
In the present telephone call bilingual language use can be observed. In particular,
language alternation is analyzed on the level of discourse, whereby different functions, behavior patterns and strategies are described.
The following language alternations were analyzed in detail:
1. A momentary gap in the lexicon of one language is ad hoc bridged by a lexeme of the lexicon of the second language.
2. Within the conversation opening sequence the second language is used for
the introduction of the first topic.
3. With the choice of a new topic the speaker switches to his/her preferred
language.
4. By means of a detailed description of the daily routine, the function of segmentation by language alternation can be observed.
5. One language marks the conversation closing and the other provides the opportunity for a side sequence.
6. The conversation opening and closing are marked by one definite language
choice.
It thus has been shown that the observed usage patterns of the languages are not
random at all (see Section 3). Furthermore, different functions of language alternations are described and analyzed in detail. A function like segmentation has not
yet been represented in the literature while other functions have already been
mentioned by Appel and Muysken (see Section 2). In general, the second language
serves as a resource around which the speaker is able to structure the conversation
at hand or to bridge problematic situations. With regard to speakers of the examined group (Rulanddeutsch-Sprecher in her terms), Berend (1998:3) writes that
code-switching is the unmarked or natural form of language use. Furthermore, she
stresses that their bilingual German-Russian competence is expressed in this way
(ibid.). This is exactly what has been observed and described in the data. During
the family conversation, the recorded speakers alternate between both languages
available to them in a natural and unmarked way. The availability of the second
language gives bilingual speakers an additional resource to structure conversations. The complex use of more than one language shows that a parallelism of the
two languages arises but that, in practice, their actual use is very flexible.
In the end, this detailed analysis gives valuable insights into an area which is
usually inaccessible to outsiders or out-group members due to its private nature.
Nevertheless, more research should be carried out in precisely these private domains, where natural and spontaneous language use is predominant (for example,
cf. Ries in prep.).
References
Appel, Ren and Muysken, Pieter. 1987. Language Contact and Bilingualism. London: Arnold.
Auer, Peter. 1984. Bilingual Conversation. Amsterdam: John Benjamins.
Auer, Peter. 1995. The pragmatics of code-switching: A sequential approach. In One Speaker, Two Languages: Cross-disciplinary Perspectives on Code-Switching, Lesley Milroy and
Pieter Muysken (eds.), 115135. Cambridge: Cambridge University Press.
Auer, Peter. 2007. The pragmatics of code-switching: a sequential approach. In The Bilingualism Reader, Li Wei (ed.), 123138. London: Routledge.
Berend, Nina. 1993. Sprachdrill oder kommunikative Integration: Zur Situation der Rulanddeutschen in der Bundesrepublik. In Sprachkontakte: Konstanten und Variablen, Ludwig
M. Eichinger and Nina Berend (eds.), 149160. Bochum: Brockmeyer.
Berend, Nina. 1998. Sprachliche Anpassung: Eine soziolinguistisch-dialektologische Untersuchung
zum Rulanddeutschen. Tbingen: Narr.
Bohmann, Alfred. 1970. Strukturwandel der deutschen Bevlkerung im sowjetischen Staats- und
Verwaltungsbereich [Menschen und Grenzen; Band 3]. Kln: Verlag Wissenschaft und
Politik.
Deprez, Kas and Persoons, Yves. 1987. Attitude. In Soziolinguistik: Ein Internationales Handbuch zur Wissenschaft von Sprache und Gesellschaft. (Sociolinguistics: An International
Handbook of the Sciencec of Language and Society.) [Handbooks of linguistics and communication science 3.2] Ulrich Ammon et al. (eds.), 125132. Berlin: de Gruyter.
Dietz, Barbara. 1995. Zwischen Anpassung und Autonomie: Rulanddeutsche in der vormaligen
Sowjetunion und in der Bundesrepublik Deutschland. Berlin: Duncker & Humblot.
Fishman, Joshua A. 1967. Bilingualism with and without diglossia: diglossia with and without
bilingualism. Journal of Social Issues 23: 2938.
Fishman, Joshua A. 1970. Sociolinguistics: A Brief Introduction. Rowley, MA: Newbury House.
Fishman, Joshua A. 2007. Who speaks what language to whom and when? In The Bilingualism Reader, Li Wei (ed.), 5570. London: Routledge.
Fitzpatrick, Mary Anne and Ritchie, L. David. 1994. Communication Schemata within the
family: Multiple perspectives on family interaction. Human Communication Research 20
(3): 275301.
Goffman, Erving. 2001. Footing. In Discourse Theory and Practice: A Reader, Margaret
Wetherell, Stephanie Taylor and Simeon J. Yates (eds.), 93110. London: Sage.
292 Veronika Ries
Luke, Kang Kwong and Pavlidou, Theodossia-Soula. 2002. Studying telephone calls: Beginnings, developments, and perspectives In Telephone Calls: Unity and Diversity in Conversational Structure across Languages and Cultures, Kang Kwong Luke and Theodossia-Soula
Pavlidou (eds.), 321. Amsterdam/Philadelphia: John Benjamins.
Meng, Katharina. 2001. Russlanddeutsche Sprachbiografien: Untersuchungen zur sprachlichen
Integration von Aussiedlerfamilien. Tbingen: Narr.
Meng, Katharina. 2003. Sprachliche Integration von Aussiedlern: Einige Ergebnisse, einige
Probleme. In Sprachliche Integration von Aussiedlern im internationalen Vergleich [Arbeitspapiere und Materialien zur deutschen Sprache 2/03], Ulrich Reitemeier (ed.), 3757.
Mannheim: IDS.
Myers-Scotton, Carol. 1993. Social Motivations for Code-Switching: Evidence from Africa.
Oxford: Clarendon Press.
Pavlidou, Theodossia-Soula. 2002. Moving towards closing: Greek telephone calls between
familiars In Telephone Calls: Unity and Diversity in Conversational Structure across Languages and Cultures, Kang Kwong Luke and Theodossia-Soula Pavlidou (eds.), 201229.
Ries, Veronika. 2013. Da kommt das so quer rein: Sprachgebrauch und Spracheinstellungen
Russlanddeutscher in Deutschland [Internationale Hochschulschriften, 582]. Mnster:
Waxmann.
Risse, Stephanie and Roll, Heike. 1997. Haben rulanddeutsche Sprache und Kultur eine Zukunft? Zur Lage der deutschen Minderheiten in den Nachfolgestaaten der Sowjetunion.
In Spracherwerb in Minderheitensituationen [Osnabrcker Beitrge zur Sprachtheorie, 54],
Jrgen Erfurt and Angelika Redder (eds.), 192217. Oldenburg: Red. OBST.
Rosenberg, Peter. 1993. Sprache, Identitt und Sprachgemeinschaft bei den Deutschen in
der ehemaligen Sowjetunion. In Sprachkontakte: Konstanten und Variablen, Ludwig M.
Eichinger and Nina Berend (eds.), 113148. Bochum: Brockmeyer.
Rosenberg, Peter. 1994. Variettenkontakt und Variettenausgleich bei den Rulanddeutschen:
Orientierungen fr eine moderne Sprachinselforschung. In Sprachinselforschung: Eine
Gedenkschrift fr Hugo Jedig, Nina Berend and Hugo Jedig (eds.), 123164. Frankfurt am
Main, Berlin: Lang.
Schegloff, Emanuel A. 1986. The routine as achievement. Human Studies 9 (23): 111151.
Selting, Margret, Auer, Peter, Barden, Birgit, Bergmann, Jrg R., Couper-Kuhlen, Elizabeth,
Gnthner, Susanne, et al. 1998. Gesprchsanalytisches Transkriptionssystem (GAT). Linguistische Berichte 173: 91122.
Segrin, Chris and Flora, Jeanne. 2005. Family Communication. LEAs communication series.
Mahwah, NJ: Lawrence Erlbaum.
Walth, Richard H. 1994. Strandgut der Weltgeschichte: die Russlanddeutschen zwischen Stalin
und Hitler. Essen: Klartext.
Appendix
GAT conventions used for the transcription
[ ]
[]
=
(.)
eh etc.
hm
?
,
.
overlapping talk
latching (concerning turns or units)
micropause
hesitation signals (filled pauses)
backchannel behavior
rising pitch
rising pitch to a midlevel
falling pitch
Russian language maintenance

through bedtime story reading?
Linguistic strategies and language negotiation
in Russian-French speaking families in Switzerland
Liliane Meyer Pitton
This paper proposes an interactional approach to the question of Russian

language maintenance through the activity of bedtime story-reading in Russian-French bilingual families in French speaking Switzerland. Reading stories
appears to be a language maintenance strategy commonly employed by the
Russian speaking parent. The ritual and recreational moment of story-reading therefore becomes an opportunity for language learning. Drawing upon a
language socialization perspective, this paper proposes an interactional analysis
of the language use in the activity of story-reading. It shows how the language
choice of the participants may be requested, negotiated and challenged during
the interaction. The analysis further informs us about the language choice pattern and the bilingual competences in these families. We will gain insight into
(Russian) language maintenance as a daily social and linguistic practice.
Keywords: language maintenance, language socialization, bedtime storyreading, bilingual interaction, bilingual families, Russian, French
1. Introduction
After the collapse of the Soviet Union in 1991, the mobility of the Russian language
speakers has greatly increased since it was no longer confined to the boundaries
of the former Communist bloc. As a consequence, Russian speakers may find
themselves in a situation of linguistic minority and are, therefore, faced with the
issue of the maintenance of the Russian language.
Language maintenance is usually studied as language use in correlation with
macro-sociological categories, embedded in a particular socio-political context.
296 Liliane Meyer Pitton
This approach does not pay enough attention to the fact that language maintenance has to be accomplished in daily interactional practices. This paper thus
proposes to view language maintenance within a language socialization framework, which considers language acquisition as a process embedded in everyday
activities and routines while putting an emphasis on its interactional accomplishment. The activity of story-reading in the Russian-French bilingual families examined in the present paper can, therefore, be considered as a site of language
socialization which constitutes for the Russian speaking parent an opportunity
for Russian language maintenance.
The proposed interactional analysis of three excerpts from story-reading
events will give us insight into the linguistic strategies employed by the Russian
speaking parent and the way the children adopt, challenge or negotiate the language choice. It will also shed light on the participants situated language choice
pattern and bilingual competences. Finally, it argues that language maintenance
has to be seen as a joint accomplishment enacted in daily practices.
2. Research background
Switzerland is not a traditional emigration country for Russian speakers and according to available statistics, Russian speakers constituted only about 0.3% of
Switzerlands resident population in 2010 (approx. 24,000 persons). Still, this
means that their proportion has doubled since 2000. In my research I am interested in one particular group of Russian speaking migrants, namely those living
with a French speaking (Swiss) partner in the French speaking part of Switzerland. In such binational-bilingual couples, language choice is not only an issue
of contact with the majority society outside the home, but it is part of private
. There are no recent statistics available, which give detailed information about the languages
spoken by the population. In order to be comparable to the numbers of the last census (see
Note2) I calculated the number by adding Russian, Ukrainian and Belarusian citizens as well as
the number of people born in these states and having acquired Swiss citizenship (BFS 2011).
. According to the last national census in 2000, Russian speakers (including Belarusian and
Ukrainian) made up 0.15% of the population, 9003 persons in absolute numbers (Ldi and
Werlen 2005:11).
. The data presented in this paper is part of my PhD thesis (Meyer Pitton 2013) which has
been funded by the Swiss National Science Foundation (www.snf.ch) as part of the research
group Multilingualism in social and vocational settings within the graduate program Pro*Doc
Language as social and cultural practice (20082011).
. This focus has been shaped by my personal interests and language skills.
Russian language maintenance through bedtime story reading 297
everyday interactions (Piller 2002; Lindenfeld and Varro 2008). The couple has
to decide on their family language policy, i.e. to negotiate the use of their respective first languages within the family and the strategies of language transmission
to their children (Piller 2001, 2002; Pavlenko 2004). In the specific context of my
research, the Russian language is relegated to a minority position in an environment where French is dominant. The labor of maintaining the Russian language
relies mostly on the Russian speaking parent alone. The aim of my research is
to investigate the questions if, how and why Russian language is maintained (or
not) in this situation. In the present paper, the focus is put on one specific activity
often mentioned by my informants as an opportunity for Russian language maintenance, namely the reading of (bedtime) stories.
3. Theoretical background
3.1
Language maintenance within a language socialization framework
Traditional research on language maintenance has focused on correlations between sociological categories and language use (Fishman 1966, 2000). Through
this approach, a myriad of factors have been identified that may be conducive to
language maintenance or language shift (Kloss 1966; Pauwels 2004). Even if daily
face-to-face interaction in families has been designated to be the cornerstone
of language maintenance (Fishman 2006), only a few researchers have actually
looked into the interactional details of everyday talk in order to establish which
communicative strategies are contributing to or, on the contrary, impeding language maintenance (Lanza 2007; Gafaranga 2010, 2011).
In this paper, I will argue that language maintenance should be studied within
a language socialization framework (Ochs and Schieffelin 1984; Schieffelin and
Ochs 1986) which conceives language acquisition as a socio-historically situated
and interactionally accomplished process. Furthermore, language socialization
should not be considered as a unidirectional process, where children/language
learners are passive receivers of predetermined rules. Rather, it should be conceived in an interactional and dynamic view of learning, characterized by active
observation and direct participation (Lave and Wenger 1993; Pontecorvo, Fasulo,
and Sterponi 2001).
Initially, most research within the language socialization paradigm was done
in mainly monolingual communities or settings. Meanwhile, the question has
been transposed onto multilingual settings, looking at the question of language
maintenance and shift (Bayley, Schecter, and Torres-Ayala 1996; Garrett 2007),
the role of bilingual practices, i.e. code-switching, in language acquisition (Garca
1980; Schieffelin 1993; Lanza 2007), and bilingual literacy practices (Hornberger
2004; Reyes and Moll 2008). The results underline the importance of considering language maintenance and bilingualism (as well as language shift and loss)
not only in their social, political and historical context, but also as being accomplished in interactional practices.
3.2
Bedtime stories as a site of language socialization
The choice of reading stories to young children may seem rather obvious to many
of us but it is not necessarily as common a practice as could be expected, as
several studies concerning language and literacy development have shown (Heath
1983; Barton 1994). Even a seemingly routine activity such as story-reading implies cultural representations about language, language acquisition and literacy,
as well as specific interactional practices. Through the activity of story reading,
children learn, among other things, about language and literacy. They learn about
grammatical structures and vocabulary as used in written language as the activity basically consists of written language read aloud as well as about narrative
structure. Children are socialized into using books and texts by observing adult
behavior, but they also participate in the activity of reading, or are summoned to
do so (Heath 1983; Barton 1994; Church 2010).
3.3
An interactional approach to language choice
Following the interactional focus of language socialization studies, language

maintenance in bilingual families should be approached through the analysis of
language choice in parent/adult-child interaction.
Such an interactional analysis can take its inspiration from the growing body
of literature applying conversation-analytic concepts and procedures to the study
of language alternation (or code-switching, which I will use interchangeably),
i.e. proposing to analyze it by looking at sequential patterns and interactional
cohesion (Auer 1984, 1998; Gafaranga 2000, 2001, 2005; Wei 2005). These studies address the shortcomings of traditional approaches to language alternation,
which either focus on the purely linguistic (grammatical) level of code-switching
(Poplack 2008), or explain it as a reflection of the social structure in language use
(Fishman 2000).
A conversation analytic approach to language alternation obliges the analyst
to consider language choice in its sequential development throughout the interactional episode. Applied to the question of language maintenance it is, therefore,
important to consider not only the linguistic input strategies of the parents, but
also the childrens response to it as well as the subsequent development of language choice.
In this paper, I am interested in analyzing the moments when an interaction
becomes a medium-related activity (Gafaranga 2010:243), or, in other words, a
language/medium negotiation sequence where language choice becomes the object of a repair (Gafaranga 2010, 2011), i.e. the conversational treatment of troubles or problems in speaking, hearing or understanding (Schegloff 2007:100).
Based on Jeffersons definition of embedded vs. exposed correction (Jefferson
1987), Gafaranga (2010) distinguishes between embedded medium repair and
explicit content repair as strategies to request the use of a specific language (or
medium). In the first case, previous language choice is marked implicitly, that is
without suspending the ongoing activity, e.g. through the use of another language
in the next turn. In the second case, language choice is made relevant through the
use of different discursive strategies, such as repeating the word or utterance in
another language.
4. Data and methodology

Following a linguistic-ethnographic approach, field research has been conducted
in a small town in French-speaking Switzerland, where a personal network provided me with access to several Russian-French speaking couples. Over a period
of approximately ten months (from June 2009 to March 2010), I was in contact
with these couples and conducted narrative interviews with eight Russian speaking persons about their individual language biographies as well as about the language use within the family. Seven of these eight couples also made recordings
of selected episodes in their daily family life with or without my presence. At the
. I will not review the abundant literature produced in the field of conversation analysis on
repair mechanisms but point out only the basic principles (see Schegloff, Jefferson, and Sacks
1977; Schegloff 2007). Repair is usually analyzed by differentiating who initiates and who accomplishes the repair, e.g. self/other initiated self/other repair, and in which turn respective
to the trouble source it is done.
. Linguistic ethnography generally holds that language and social life are mutually shaping,
and that close analysis of situated language use can provide both fundamental and distinctive
insights into the mechanisms and dynamics of social and cultural production in everyday activity (Rampton et al. 2004:2).
. All of them (seven women and one man) were born in the former Soviet Union and moved
to Switzerland between 1994 and 2005. All of them declare Russian to be their first or dominant
language.
time of data collection, four of these families took their children to a Russianspeaking playgroup, where I was present as a participant observer as well.
Story-reading, particularly bedtime story-reading, was mentioned by my
Russian speaking informants as an important practice to ensure sufficient linguistic input in Russian. In this context, childrens books and fairy tales in the
Russian language were mentioned as important material resource for language
maintenance. The event of reading bedtime stories is, therefore, considered as a
privileged moment of the day for the Russian speaking parent, where the creation
of a Russian language environment is possible. Consequently, the ritual and recreational moment of story-telling becomes an opportunity for language learning.
The importance of this specific communicative event is underlined by the
fact that three of the families with smaller children have recorded such an event.
Furthermore, I also observed (and partially recorded) several instances of storyreading practices in the Russian playgroup. The data discussed in this article come
from the recordings done by the three families at home. The length of the recordings varies from 11 to 38 minutes.
5. Analysis: When bedtime story reading becomes a language lesson

As mentioned above, the bedtime story is considered to be an ideal opportunity
for the Russian speaking parent to create a Russian environment for the child/
children. It is first and foremost presented as an occasion for unilateral language
input from the parent, even if the childrens participation in the activity is not excluded. The analysis of the following excerpts will show how children participate
in this activity, either self-initiated or upon the parents invitation. These excerpts
are chosen because they represent the very moments where language choice is
made relevant and the tasks of language maintenance and of language socialization become visible, i.e. are talked into being (Gafaranga 2010, 2011). They illustrate the interactional accomplishment of language maintenance in the activity
of story reading as it takes the form of a language lesson. In all three excerpts
the participants are introduced with pseudonyms to protect their anonymity. As
the recordings were done in audio-format, the interaction with the pictures in the
book and other multi-modal aspects cannot be included in the analysis except
when there are verbal indications for it.
. For information about the conventions for transcription and translation please refer to the
Appendix.
5.1
Repeat with me
In this first excerpt, Elena, a Russian speaking mother, is reading a first bedtime
story to her son Sebastian (4.10). The story is called Tri kotnka (three kittens).
It is about three kittens, a black, white and grey one, chasing different small animals and falling into substances that change their color. The excerpt presented
here is situated towards the end of the story when the three kittens have fallen
into water and turned again their original colors. Until this moment, Sebastian
has not uttered a word, but listened to his mother reading, laughed and (most
likely) looked at the pictures. As mentioned, the recording is in audio-format so
the interaction with the pictures can only be deduced from explicit mention, e.g.
vidi (you see) in line 4. Sebastians first verbal intervention occurs in line 2 of
the excerpt below.
(1) Tri kotnka (Three Kittens)
1 ELE: tri
mokryx kotnka poli do?moj (.)
Three wet
kittens went home.
2 SEB: (o)
Where?
3 ELE: nu
domoj oni idut,(.) po doroge oni obsoxli
Well home they go
on road
they dried
Well, home they go. On the road they dried,
4 vidi? solnyko i
stali kak byli
see.2.SG sun.DIM
and became as were
you see the little sun? And they became as they were:
5 rnyj seryj i? (..) belyj. povtori so
mnoj rnyj
black grey and
white repeat with me
black
black grey and white. Repeat with me: black.
6 (..)
7 ELE: rnyj kotnok govori rnyj?
Black kitten, say
black.
8 SEB: n- ( voir;noir)
Bla- (to see;black)
9 (.)
10 ELE: seryj
Grey
11 (..)
. 4 years 10 months.
12 ELE: se? ryj

Grey
13 SEB: ser
Gre14 (.)
15 ELE: -ryj nu
grome govori,
i? be:[lyj

well louder speak.IMP and white
-y. Well speak louder, and: white.
16 SEB:

[blanc
White
17 ELE: belyj
White
18 SEB: et blanc, quest-ce que a
veut di:re?
and white what-is-it that this wants say
And white, what does this mean?
19 ELE: nu
to cveta kotjat? (3)
ta:k?
well this colors kitten.GEN.PL so
Well, these are the colors of the kittens. So,
20 a
to- sejas my itaem druguju
and this now
we read
other.ACC
and this- now we read another ((etc))
The story reading proceeds solely in Russian until the moment when Sebastian
intervenes by asking a question in French (line 2). Elena answers in Russian, while
changing the word order of the utterance in line 1 which results in an emphasis
on the word domoj (home). She thus treats Sebastians question not as an implicit
request for language change, but as an acoustic problem concerning the last word
of her previous turn, which she resolves by repeating it in line 3. Further, Elena
continues the activity (story reading) in Russian, therefore implicitly correcting
Sebastians language choice. The result is a bilingual parallel format (Gafaranga
2010) where each participant sticks to his/her chosen language.
It is only a few turns later, in line 5, that Elena makes language choice explicitly relevant by giving an instruction to Sebastian in the form of the directive
povtori (repeat) and an item (rnyj black). The ensuing gap (line 6) marks
the absence of Sebastians turn which indicates a trouble. Elena treats his trouble again as a problem of comprehension/understanding rather than of language
choice, as she links the word black with the kitten (and the story just read) and
then asks Sebastian again to utter it (say, line 7).
After his mothers insistence (two directives), Sebastian repeats the item tentatively in line 8, while also mumbling something in French. Elena accepts his
repetition as a valid reaction to her request and goes on in line 10 with the next
word. She establishes an interactional format which consists of an item proposed

by her and a subsequent repetition by Sebastian.
Again, Sebastian does not take his turn in line 11, prompting Elena to reiterate the item in line 12. In line 13, Sebastian complies with the proposition but
again only tentatively and in a low voice. His mother completes the word by adding the missing ending and urges Sebastian to speak louder, before she proposes
the next, and last, item (line 15). This time Sebastian anticipates the task, his turn
even overlaps with the item, but he delivers his answer in French (line 16). Even
if Sebastian has followed the interactional format established by his mother, she
treats his answer as insufficient by repeating the item in Russian (line 17), correcting his language choice and thus making language choice again relevant.
Sebastian, however, does not follow his mothers language choice, but insists
on his own (line 18). Elena does not reiterate the correction of his language choice
nor does she continue the language lesson. She falls back into the parallel format
of the beginning, answering in Russian to his question in French (line 18/19), and
pursues the activity of story reading by suggesting a new book (line 20).
This excerpt illustrates how language choice can be made relevant and negotiated in interaction. It also shows how language choice pattern are developed
and how the bilingual competences of the participants can be interactionally
displayed.
We see that Elena pursues a Russian only strategy in interaction with
her son. However, she does not obstruct interaction when Sebastian responds
in French which he does consistently except when Elena insists on repeating
Russian words. The interaction, therefore, takes on a bilingual parallel mode
(Gafaranga 2010) where all self-initiated turns by Elena are in Russian and those
by Sebastian in French.
The excerpt also informs us about the bilingual competences of the participants. Elena is obviously proficient enough in French in order to react appropriately to Sebastians interventions in French. If we look at the presented excerpt
doing being language-blind (Gafaranga 2001:1901), i.e. only paying attention to
the interactional cohesion and not to the languages used, we would not necessarily notice the language alternation in line 13, nor in 1819. Sebastians question
(where) in line 2 is not treated by Elena as a problem of language choice (or as a
medium request, see Gafaranga 2010), but as a failure of acoustic understanding,
which she resolves by repeating the problematic part, a case of other-initiated self
repair. Similarly, she does not treat Sebastians question in line 18 as a problem of
language competence. Yet, if we look at the whole sequence, Sebastians question
in line 18 is somewhat surprising as it indicates that he apparently failed to grasp
the connection between the colors and the topic of the story. Sebastians turns
in this excerpt show that he has a partial understanding of what is going on in

Russian: in line 2, he asks where? (and not what which would have indicated a
more general problem of understanding), which indicates that he did grasp the
idea of place and movement. Also, his anticipated answer in line 16 (blanc) and its
repetition in line 18 points to the fact that he understands the color naming task
initiated by his mother.
Sebastians interactional preference for French seems to trigger the sequence
of language negotiation, the language lesson, as Elena tries to make him not only
listen to Russian, but actually speak it. In order to succeed, she makes use of directives and explicit instructions and insists on the format of a language lesson. Sebastian complies twice, even if not to the full satisfaction of his mother, who, each
time, corrects his answer. The third time he refuses her language choice but insists
on his own by repeating the word twice in French. The sequence of language negotiation results (again) in the parallel mode where each participant sticks to his
language choice, the mode which Gafaranga qualifies as a maintenance-oriented
strategy, if only a very weak one (Gafaranga 2010:266), as the parent does not
insist on the active use of her language choice by the child.
5.2
Not song but song
The next excerpt presents us with a bedtime story sequence recorded by Mihail,
a Russian speaking father, and his two sons: Konstantin (5.1) and Alexej (2.7).
The story is an old Slavic tale about a bun (Kolobok). An old woman bakes a bun
which becomes alive and escapes from the window sill to roll off on the road. On
his way he/it meets different animals who want to devour him. He/it then suggests
singing a song for them and manages to escape. This song is a repeated feature of
the story.
(2) Chanson (Song)
1 MIH: navstreu emu? seryj volk (2) kolobok kolobok ja
towards
him grey wolf
bun
bun
I
He run into the grey wolf: Bun, bun, I am going to
2 tebja sem
you
eat.FUT.PERF
eat you!
3 ALE?: non
No!
4 (..)
5 MIH: ne:t ne e menja seryj volk.(.) ja tebe

no
not eat me
grey wolf
I you
No, dont eat me grey wolf. I will sing you
6 pesenku
spoju (..)
song.DIM.ACC sing.FUT.PERF
a song.
7 ALE: chanson
Song.
8 MIH: pesenku.
da? skai
pesenku
song.DIM.ACC yes say.IMP song.DIM.ACC
Song, yes? Say song.
9 (1)
10 ALE: chanson
Song.
11 MIH: nu
ne chanson a
skai
pesenku
well not song
but say.IMP song.DIM.ACC
Well, not song, but say song.
12 (.)
13 ALE: pe
So
14 MIH: pesenku molodec. kak ego zovut?
song.ACC witty
how him call
Song, well done. How is he called?
15 (..)
16 ALE: kabo
((tries to say Kolobok))
17 MIH: kolobok. (.) a to?
Bun.
And this?
18 KON: (mo[i jai
dchir)
Me, I have torn.
19 ALE:

[vo
Wo
20 MIH: volk molodec

Wolf, well done.
21 (.)
22 KON: papa (jai;
cest dchir)
Dad (I have; its torn).
23 MIH: (moui) il faut pas- ne nado maly ne nado (2.5)
-yes it need not not need boy
not need
M-yes, you should not - shouldnt boy shouldnt
Alexej participates in the story by giving his comments: shouting no when an

animal wants to eat the bun (as in line 3 here) and saying song when the bun in
the story suggests singing a song (as in line 7). Even though he also employed the
French word chanson (song) before, Mihail reacts only in the second instance
which is visible in the excerpt (line 78). At this point Mihail suspends the story
line in order to initiate a side sequence (Jefferson 1972) in form of a language
lesson, making Alexejs use of chanson the focus of the interaction. He not only
flags Alexejs language choice (pesenku vs. chanson), but also gives instructions
(directive: skai) in providing an item (pesenku). The ensuing gap (line 9) marks
a transition relevance place (Sacks, Schegloff, and Jefferson 1974) as it is Alexejs
turn to repeat the item. Alexej finally responds by repeating chanson (line 10). In
line 11, Mihail initiates an explicit language repair by rejecting chanson as a possible answer (ne chanson) and giving an explicit instruction (a skai pesenku). By
juxtaposing the word in the respective languages and rejecting chanson, Mihail
makes language choice relevant.
In line 13, Alexej finally complies by uttering something that could be considered as an attempt to say pesenku.10 Mihail honors the attempt by repeating its full
form and giving a positive evaluation (molodec) (line 14). Instead of resuming the
story, Mihail continues the language lesson, but changes its format into a question-answer-evaluation sequence (line 1420). He now tests Alexejs knowledge
of the storys characters (most likely by pointing to the pictures in the book). This
side sequence to the story is then brought to a close by Konstantins intervention
(who apparently was not listening to the story but doing something on his own)
in line 18 and 22. Mihail reacts to this in line 23 and then resumes the story.
Again, in this excerpt we can observe the bilingual parallel mode as Mihail
does not react to Alexejs language choice until line 8. This is the moment when
language choice is made relevant by Mihail. Both participants (Konstantin, does
not seem to participate in this sequence) also display their bilingual competence:
Alexej reacts appropriately to the plot and Mihail is able to take up on his sons
French interventions. Mihails bilingual competence is shown even more clearly
in his exchange with Konstantin who appeals to him in French: Mihails response
in line 23 is first in French (il faut pas), but he then repairs his utterance by repeating the same in Russian (ne nado). This illustrates at the same time the self-
10. Please note that Mihail repeats the word in question (pesenka) always in its oblique case
(accusative) as it appears in the story. A real language lesson would rather ask for the item to
be given in its basic form, i.e. the nominative case.
control that has to be exerted by a bilingual parent in order to stick to his language
(medium) choice.11
5.3
Winnie pirate
The last excerpt is from a bedtime story recording done by Anastasija, a Russian
speaking mother, and her two sons Denis (6.6) and Dimitri (usually called by the
diminutive form: Dima) (3.3). Before this excerpt, Anastasija has already read two
stories to her sons, this excerpt is from the third (and last) story she reads to them
this evening. It is the popular Russian folk tale The bunny and the fox, where
the nave bunny offers accommodation to the sly fox, that had lost his house, but
then gets expulsed by the fox. The bunny is walking around crying and receives
help by different animals to chase the fox. These, however, get frightened by the
fox and abandon the bunny in its grief. The excerpt shown below is situated at the
moment after the first two animals had no success in chasing away the fox.
Anastasija integrates her sons into the story by organizing role-plays; but contrary to the usual way of doing this12 she does not make them play the characters
of the story, but lets them take on self-chosen additional characters in this case
Winni Pirat.
(3) Winnie pirate
1 ANA: (xx)po-prenemu zaplakal zajc- zajac, pol opjat

as before
cried
bunny went again
As before, the bunny started to cry and he went again
2 svoi doroki, i
tut
idut
navstreu kto
his ways.DIM and there go.3.PL towards
who
his way. And there, who comes towards him?
3 DIM?: winni pirat
4 ANA: winni pirat i?
Winnie Pirate and?
11. Gafarangas examples (2010, 2011) show how in the bilingual Rwandan community, which
he observed, the childrens medium request often ends in the change of the medium by the
adult to the childrens preference. This was however not the case in excerpt (1), and only partly
in excerpt (2).
12. As shown in the paper Gender socialization through literacy texts and their interpretations: A study of two folk tales used in a Russian preschool presented by Ekaterina Moore
(UCLA) at the conference Approaches to Slavic Interaction, University of Potsdam, March
1618, 2011.
5 DEN: net (x)

No.
6 ANA: snaala winni pirat
First
Winnie Pirate.
7 DEN: mm
8 ANA: davaj
dima davaj
winni pirat govori
give.IMP (name) give.IMP (name
) speak.IMP
Come on Dima, come on Winnie Pirate speak.
9 DIM: tu sais e
You know
10 ANA: nu
snaala zdravstvuj

Well, first: Hello.
11 DIM: vzdastvi
Hllo
12 ANA: zajac
Bunny.
13 DIM: vdv zajc
((hl bnny)) ((mumbled))
14 ANA: to ty plae?
what you cry
What are you crying?
15 DIM: sto ty plae
W(h)at are you crying?
16 ANA: a
ja plau potomu to lisa menja iz
doma vygnula
but I cry
because
fox me
from house chased
But I cry because the fox chased me from home.
17 ((sniff)) ne znaju
to mne delat

not know.1.P.SG what me do.INF
I dont know what to do.
18 DIM: eh? je vais appeler (xxi) pirata attends moja

I will call
pirata wait
mine
19 (4)
20 DIM: ding dong toc toc (...) chg
((sounds of ringing then knocking on door))
21 ANA: privet
Hi.
22 DEN: zdravstvujte poemu
vy prili?
Hello
why
did you come?
23 DIM: parce que
Because
24 ANA: potomu to
Because
25 DIM: potomu to lisa

vynde
Because
the fox cha
26 ANA: vygnala
chased away
27 DIM: vyngn:la zajca.
chasd
the bunny away.
28 ANA: iz
doma.
from home.
29 DIM: iz
doma.
from home
30 DEN: ladno
vygenuju
tu
tu lisu ja j[a
Well, I will chase away this that fox I I
31 ANA:

[(nu) to
Well, what ((etc))
The excerpt begins with Anastasijas invitation to her sons to introduce a new
character (line 2). One of them follows on this invitation in the next turn (line 3)
which is then confirmed by Anastasija (line 4).
The boys seem to recognize this moment in the story as the one to introduce
their own character (namely Winnie Pirate), as nothing indicates that the character now appearing will not be a regular one from the story. In line 6, Anastasija
takes over the role of the stage director, coordinating the appearance of her sons
roles. She then invites Dimitri (Dima) to speak (line 8). Dima starts in French in
line 9 and is immediately corrected by his mother, who gives him a Russian item
to start his intervention with (zdravstvuj). Her repair initiation does not explicitly
concern Dimas language choice, but the content of his utterance. By taking up
his mothers language choice (turn 11), however, Dima does not only repair the
content, but also his language choice.
Anastasija then goes on in her role as a prompter, continuing the format of low
voice item and subsequent repetition by Dima until line 15. In line 16/17 Anastasija takes over the role of the bunny. In line 18, Dima produces a somewhat mixed
language sentence, which is not corrected by his mother. The play continues as
Dima is seemingly ringing and knocking on a door (line 20), and Anastasjia again
takes up her prompter role (line 21). But instead of waiting for Dima to repeat the
suggested greeting (privet), Denis takes his turn in line 22 in Russian. In line 23,
Dima starts again in French (because), but reacts appropriately to his brothers
question (why). Anastasija initiates a repair on his language choice, giving him
the corresponding Russian word (line 24). The following turns (2429) take on a
similar format to line 1015 above, with Anastasija offering the Russian items that
are taken up by Dima. Contrary to the earlier sequence, Dima here produces his
own development of the sentence and does not simply repeat Anastasijas suggestions. She intervenes when he encounters a problem of either in how to articulate
a word (line 25/26) or how to continue the sentence (line 27/28).
Similarly to the preceding excerpts, this last one also shows moments where
language choice is made relevant in interaction (lines 1015, 2429), while on
other occasions the interaction is going on in a bilingual format (line 1820).
Anastasija succeeds in making her children not only listen and react to the Russian story, but also participate in it and, therefore, to speak Russian. However,
she continues controlling the narrative structure of the story as well as language
choice through directing and prompting the necessary interventions of her sons,
especially Dima.
5.4
Discussion
The three excerpts analyzed in Sections 5.15.3 show that the activity of story
reading is not only a way to provide the children with language input (e.g. grammatical structure; vocabulary; narrative structure), but also to trigger output, i.e.
to get the children to speak Russian. Furthermore, the children do not only react
to the parents summons, but they also participate spontaneously in the activity
by commenting or asking questions. However, their self-initiated contributions/
turns happen to be mostly in French (with the exception of Denis in excerpt (3)).
They only speak Russian as a reaction to the embedded or explicit language repair/correction, i.e. the language lessons, initiated by their parents.
The analysis illustrates the micro-interactional strategies the Russian speaking parents employ to ensure more than just (passive) language input. They create moments, when language choice is made relevant in the form of language
lessons that are integrated into the bedtime story. The children respond to it by
either adopting or negotiating the language choice of their parent. These negotiation sequences have different outcomes.
In excerpt (1), Elena introduces instruction/item repetition-sequences
in order to break the bilingual parallel format and to prompt Sebastian to utter
words in Russian. This strategy is challenged at the third item as Sebastian refuses
to comply to his mothers language choice insisting on French as his own choice.
As a consequence, the interaction falls back into the parallel mode.
In excerpt (2), Mihail introduces a language lesson similarly to Elena, making
Alexej repeat the word song in Russian after the boy has offered it in French. The
language choice is thus negotiated by Alexej, who first repeats the word in French.
But in the end, Mihails attempt succeeds to a certain degree as Alexej tries to
say something similar to the Russian word. Mihail then changes the format of
the lesson into the classical question-answer-evaluation sequence (Heath 1982).

Here, Alexej seems to have accepted the language choice; even if his pronunciation stays approximate, his attempts to answer are clearly in Russian, e.g. vo- for
volk (wolf ), which would be loup in French. Mihail therefore succeeds to make
Alexej align with his language choice. Konstantin, however, addresses his father
in French and receives a response in French, even if the response is followed immediately by its Russian equivalent.
In excerpt (3), Anastasija also gives instructions and items to Dima for him
to repeat. However, they are integrated into a role play, where the children take
over their own characters and become full participants in the activity. Even if
Dima spontaneously starts his turns in French, he always follows his mothers
language choice, repeating the offered items and even trying to elaborate the sentences himself. His older brother participates in the play in Russian, which adds
to the impression that the boys indeed consider this kind of activity as normally
being accomplished in Russian.
The interactional language choice pattern observed in these examples show
that the Russian speaking parents generally keep to the exclusive use of Russian
language (with the exception of Mihail (5.2), who, however, repairs his use of
French immediately). Furthermore, they also make their childrens language choice
relevant, as they correct their use of French either implicitly or explicitly.
Finally, in all three excerpts the bilingual competences of the participants become visible. All Russian speaking parents react appropriately to their childrens
contributions in French and are able to offer Russian equivalents. The children
show a sufficient understanding of Russian: their interventions are appropriate
regarding the topic and development of the story. However, the interaction often
falls back into a bilingual parallel mode (Gafaranga 2010, 2011), where the parent
speaks Russian and the children respond in French. The activity of story-reading
is, therefore, by no means a monolingual Russian interaction as the parents may
present or perceive it, but a bilingual Russian-French event.
6. Conclusion
In conclusion, the analysis of several excerpts from bedtime story reading events
has shown that even if this activity is considered as being important for the maintenance of Russian as a minority language, it is not a monolingual interaction, as
the other language (French) is regularly introduced by the children. It shows the
difficulty for the Russian speaking parent in a bilingual family to create an exclusive Russian speaking environment, even in a situation where he or she is alone
with the children. All parents presented in the excerpts, seem to make an effort
to find a balance between the normal development of the story-reading activity

and the introduction of more explicit language lessons.
We further observed that children are not passive, but active participants in
the activity of story-reading. They introduce themselves as interactants or are summoned by the parent to do so. Consequently, they also respond to the parental linguistic strategies by negotiating, challenging or accepting the language choice. As
a result we see that language maintenance strategies are not solely relying on the
right input of the parent, but also on the childrens response. Language maintenance appears to be a joint accomplishment, as mentioned in Section 3.1.
With the help of sequential analysis it can be demonstrated at which instances language choice becomes relevant, is negotiated, challenged or accepted
and, therefore, how language choice pattern and bilingual competences emerge
interactionally. The analysis further shows that the parents bilingual competence
require a certain control of their own language use lest they comply with the childrens language preferences.
Overall, we can conclude that language maintenance is not only a question
of macro-social changes and conditions, but has to be accomplished in everyday interactional practices. Its success or failure depends as much upon parental
strategies and efforts as on the childrens responses and the subsequent reactions
of the parents.
References
Auer, Peter. 1984. Bilingual conversation. Amsterdam: John Benjamins.
Auer, Peter. (ed.). 1998. Code-switching in conversation: language, interaction and identity.
London: Routledge.
Barton, David. 1994. Literacy: an introduction to the ecology of written language. Oxford:
Blackwell.
Bayley, Robert, Sandra R. Schecter, and Buenaventura Torres-Ayala. 1996. Strategies for bilingual maintenance: case studies of Mexican-origin families in Texas. Linguistics and Education 8 (4): 389408.
BFS, Bundesamt fr Statistik. 2011. Migration und Integration Indikatoren. Auslndische
Bevlkerung: Staatsangehrigkeit. Stndige auslndische Wohnbevlkerung nach Staatsangehrigkeit. Available at: http://www.bfs.admin.ch/bfs/portal/de/index/themen/01/07/
blank/key/01/01.html [Accessed November 4, 2011].
Church, Amelia. 2010. Opportunities for learning during storybook reading at preschool. In
Applied Linguistics Review, Li Wei (ed.), 221246. Berlin, New York: Walter de Gruyter.
Fishman, Joshua A. (ed.). 1966. Language loyalty in the United States: the maintenance and perpetuation of non-English mother tongues by American ethnic and religious groups. London:
Mouton.
Fishman, Joshua A. 2000. Who speaks what language to whom and when? In The bilingualism
reader, Li Wei (ed.), 89106. London: Routledge.
Fishman, Joshua A. 2006. Language maintenance, language shift, and reversing language shift.
In The handbook of bilingualism, William C. Ritchie and Tej K. Bhatia (eds.), 406436.
Oxford: Blackwell Publishing.
Gafaranga, Joseph. 2000. Language separateness: A normative framework in studies of language alternation. Estudios de Sociolingistica 1 (2): 6584.
Gafaranga, Joseph. 2001. Linguistic identities in talk-in-interaction: Order in bilingual conversation. Journal of Pragmatics (33): 19011925.
Gafaranga, Joseph. 2005. Demythologising language alternation studies: conversational structure vs. social structure in bilingual interaction. Journal of Pragmatics (37): 281300.
Gafaranga, Joseph. 2010. Medium request: Talking language shift into being. Language in Society 39 (02): 241270.
Gafaranga, Joseph. 2011. Transition space medium repair: Language shift talked into being.
Journal of Pragmatics 43 (1): 118135.
Garca, Eugene E. 1980. The function of language switching during bilingual mother-child
interactions. Journal of Multilingual and Multicultural Development 1 (3): 243252.
Garrett, Paul B. 2007. Language socialization and the (re)production of bilingual subjectivities. In Bilingualism: a social approach, Monica Heller (ed.), 233256. Basingstoke:
Palgrave Macmillan.
Heath, Shirley Brice. 1982. What no bedtime story means: narrative skills at home and school.
Language in Society 11 (01): 4976.
Heath, Shirley Brice. 1983. Ways with words: language, life, and work in communities and classrooms. Reprinted 1989. Cambridge: Cambridge University Press.
Hornberger, Nancy H. 2004. The continua of biliteracy and the bilingual educator: educational linguistics in practice. GSE Publications/Bilingual Education and Bilingualism 7 (2&3):
155171.
Jefferson, Gail. 1972. Side sequences. In Studies in social interaction, David Sudnow (ed.),
294338. New York, London: Free Press, Collier-Macmillan.
Jefferson, Gail. 1987. On exposed and embedded correction in conversation. In Talk and social
organisation, Graham Button and John R. E. Lee (eds.), 86100. Clevedon: Multilingual
Matters.
Kloss, Heinz. 1966. German-American language maintenance efforts. In Language loyalty in
the United States: the maintenance and perpetuation of non-English mother tongues by American ethnic and religious groups, Joshua A. Fishman (ed.), 206252. London: Mouton.
Lanza, Elizabeth. 2007. Multilingualism and the family. In Handbook of Multilingualism and
Multilingual Communication, Peter Auer and Li Wei (eds.), 4567. Berlin: Mouton de
Gruyter.
Lave, Jean, and Etienne Wenger. 1993. Situated learning: legitimate peripheral participation.
Repr. [1st edition 1991]. Cambridge: Cambridge University Press.
Lindenfeld, Jacqueline, and Gabrielle Varro. 2008. Language maintenance among fortunate
immigrants: The French in the United States and Americans in France. International Journal of the Sociology of Language (189): 115131.
Ldi, Georges, and Iwar Werlen. 2005. Eidgenssische Volkszhlung. Sprachenlandschaft in der
Schweiz. Neuchtel: BFS Bundesamt fr Statistik. Available at: http://www.bfs.admin.ch/
bfs/portal/de/index/themen/01/22/publ.html?publicationID=1737 [Accessed November
3, 2011].
Meyer Pitton, Liliane. 2013. Sprachloyalitt(en)? Sprach-Beziehungen und Sprachweitergabe als

Formen von Mehrsprachigkeit. Das Beispiel russischsprachiger Partner/innen in zweisprachig-binationalen Paaren und Familien in der franzsischsprachigen Schweiz. Bern: Philosophisch-Historische Fakultt, Institut fr Sprachwissenschaft (PhD Thesis, Unpublished
Manuscript).
Ochs, Elinor, and Bambi B. Schieffelin. 1984. Language acquisition and socialization: Three
developmental stories. In Culture Theory: Essays on Mind, Self, and Emotion, Richard A.
Shweder and Robert A. LeVine (eds.), 276320. Cambridge: Cambridge University Press.
Pauwels, Anne. 2004. Language Maintenance. In The Handbook of Applied Linguistics, Alan
Davies and Catherine Elder (eds.), 719737. Oxford: Blackwell.
Pavlenko, Aneta. 2004. Stop doing that, ia komu skazala!: language choice and emotions in
parent/child communication. Journal of Multilingual and Multicultural Development 25
(2&3): 179203.
Piller, Ingrid. 2001. Private language planning: the best of both worlds? Estudios de Sociolingistica 2 (1): 6180.
Piller, Ingrid. 2002. Bilingual couples talk: the discursive construction of hybridity. Amsterdam:
John Benjamins.
Pontecorvo, Clotilde, Alessandra Fasulo, and Laura Sterponi. 2001. Mutual apprentices: the
making of parenthood and childhood in family dinner conversations. Human Development 44 (6): 340361.
Poplack, Shana. 2008. Code-Switching / Sprachwechsel. In Sociolinguistics: an international
handbook of the science of language and society, Ulrich Ammon et al. (eds.), 589596.
Berlin: de Gruyter Reference Global.
Rampton, Ben et al. 2004. UK linguistic ethnography: a discussion paper. Available at: http://
www.ling-ethnog.org.uk/publications.html [Accessed March 31, 2011].
Reyes, Iliana, and Luis C. Moll. 2008. Bilingual and biliterate practices at home and school.
In The handbook of educational linguistics, Bernard Spolsky and Francis M. Hult (eds.),
147160. Oxford: Blackwell Publishing.
Sacks, Harvey, Emanuel A. Schegloff, and Gail Jefferson. 1974. A simplest systematics for the
Schegloff, Emanuel A. 2007. Sequence organization in interaction. A primer in conversation
analysis. Cambridge: Cambridge University Press.
Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks. 1977. The preference for self-correction in the organization of repair in conversation. Language 53 (2): 361382.
Schieffelin, Bambi B. 1993. Codeswitching and language socialization: Some probable relationships. In Pragmatics: From Theory to Practice, Judith F. Duchan, Lynne E. Hewitt, and
Rae M. Sonnenmeier (eds.), 2042. Englewood Cliffs, NJ: Prentice Hall.
Schieffelin, Bambi B., and Elinor Ochs. 1986. Language Socialization. Annual Review of Anthropology 15 (1): 163191.
Wei, Li. 2005. How can you tell? Towards a common sense explanation of conversational
code-switching. Journal of Pragmatics (37): 375389.
Appendix
Symbols used in the transcripts
.
,
?
belyj
di:re
vidi? solnyko
se-
ELE: be:[lyj
SEB: [blanc
(.)
(..)
(...)
(3)
(o)
(xxx)
( voir; noir)
((laughs))
falling, stopping intonation

continuing intonation
rising intonation
emphasis
lengthening
quieter speech
cut-off or self-interruption
overlapping talk
micropause (up to 0.3 seconds)
pause (between 0.30.5 s)
pause (between 0.60.9 s)
measured pause in seconds
words that cannot be clearly heard
words that cannot be identified
possible alternatives
non-verbal behavior, authors comment
Translation
SEB: et blanc, quest-ce que
Original
and white what-is-it that Translinear gloss (if considered necessary)
and white, what English translation
Note: In the original (line 1) as well as in the English translation (line 2 or 3) Russian is in italics,
French in regular font. There is no special marking of the languages in the translinear gloss.
Index
A
adjacency pair 240
affiliating utterance 20
agreement and disagreement
142
animated speech 258261
aspect (aspectual function of
double verb construction)
39
attitude 135, 143
attitudinal prosody 136
contextualization cue(s) 273

274
contrast pattern 70
Conversation Analysis 1731,
240241
conversational
backgrounds 169170, 172,
178, 180181
implicature 168, 176, 180
courses of action, or projects, in
interaction 41, 4346
B
backgrounds
conversational 169170, 172,
178, 180181
informational conversational
178
realistic conversational 178
bilingual
competence 303, 306, 311
parallel mode 302303, 306,
310311
blinking 76
(British) Hansard Report 214,
217218, 220221
D
deontic modal interpretation
172
direct perception 173
direct/indirect evidence 168,
170176
discourse particles 107109, 115
Dom Dva 242
double verb constructions
35ff., 39
C
Chafe, Wallace L. 86, 98
closing sequence 289
code-switching 282283,
286287, 290
colloquial 218220, 226
communicative
functions 24
situations 118, 127
compensatory pattern 6970
confrontainment 191, 194f., 209
contextual sources (of irony)
190
contextualization 273274
E
echoic mention 189, 198f.
embedded gestures 75
epistemic
adverb 169
modality 168169, 173176,
179180
necessity 181
status 151, 163
everyday conversation 284
evidential meaning of hearsay 177
evidentiality 168169, 171176,
180
existential and universal
readings 172
experimental speech data 136f.
expletive function of double
verb construction 40
F
face threatening act/FTA 188,
193, 199204
false start 218, 221
family conversation 283
footing 259260
frequency lists 107111
Functional Syntax 87
G
gaze grammar 65
Goffman, Erving 258260
H
Haiman, John 274
hecklings 217218, 224225,
228
hesitation phenomena 8586,
90
honorific 218, 225226
horizontal gesture ligatures 74
I
illocution 241
Impoliteness 238239, 243247
indirect speech acts 168
inference 168, 172176
interactional dynamics 29, 31
ironical 221, 226
irony 263267
K
knowledge (rights to knowledge)
150151, 161
L
Labov, William 8889, 98
language alternation 298299
language choice (pattern) 298
299, 302, 306, 309311
318 Approaches to Slavic Interaction
language socialization 57,

297298, 310311
locution 241
M
mens and womens speech rate
120121
Menzeraths-Altmann law 115
microphone (switched off)
218, 222224
Mock-Impoliteness 239240,
247253
modal auxiliary 168
morality 40, 45, 51
multimodal event 64
Multimodal Russian Corpus
(MURCO) 67
My side evaluation 152ff.,
162163
N
narrative 8889
natural speech data 137
negative face 199204
neutral cues 68
ORD corpus 105106, 117118
P
participation status 51, 54
pauses 118, 123, 125
perlocutionary effect 242
phonetic pauses 77
pitch level 141f.
portemanteau effect 176
portemanteau word 180
positive face 199f., 202ff., 204,
208
possible worlds 172
projection 91
projection, syntactic 22, 30
prosodic 139142
provoking cues 68
Q
quantifier 172
quasi-spontaneous speech
137141
Question Time 216, 224,
227228, 230231
R
redundant 218, 221
regulations (of the Duma) 217,
221, 224, 226227
reportative evidential 168, 170
rhetorical 140, 142
rumor 173174
Russlanddeutsche 283, 290
S
sarcasm 216, 223
segmentation 288
sensory evidence 174
show conversations 191
social dominance 121
speech production 87
speech production phases 91
speech registers 114115
spontaneity 90
Sttimcets reportative modal
171172
stance 135
strategic behavior 288
strong interpretation of modals
168, 170
strong modal 168, 169, 175
strong modal must 175
syllables 118, 122123, 127
duration 113116, 119
length 119, 122
rate 115
per clause 122
per second 119120
syntactic pauses 77
T
teasing 267269
televized political discourse/
debates 191, 199
tellability 88, 98
topic 2426
transcription conventions 33
transcripts 216218, 220221,
228229, 233
truth value 168, 175, 179180
turn completion 2122, 31
turn constructional unit (TCU)
2021, 9192, 9495
turn extension 2122, 3031
turn-initial position 149150
U
universal or existential
quantificational strength
172
upgraded requests 47, 50
utterance duration 112113
utterance length 122
in syllables 111115
in words 106
V
verbal irony 187f.
verbal violence 191, 195f.
vertical gesture ligatures 74
video archive 217, 233
voting behaviour 229, 232
W
weak interpretation of modals
168, 170

Thielemann, Kosta - Approaches To Slavic Interaction

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Thielemann, Kosta - Approaches To Slavic Interaction

Încărcat de

Drepturi de autor:

Formate disponibile

Approaches to Slavic Interaction

Dialogue Studies (DS)

A*STAR - Institute of High Performance Computing, Singapore

Editorial Advisory Board

Tel Aviv University

Hochschule fr Technik und

Russian Academy of Sciences

College of William and Mary

Approaches to Slavic Interaction

John Benjamins Publishing Company

The paper used in this publication meets the minimum requirements of

Library of Congress Cataloging-in-Publication Data

2013 John Benjamins B.V.

Part I. Multimodal, grammatical and paralinguistic resources

Reanimating responsibility: The we-V2 (take-V2) double imperative 

Eye behavior in Russian spoken interaction and its correlation 

Hesitation markers in transitions within (story)telling sequences 

Part II. Statistical analysis of Russian talk-in-interaction

Speech rate as reflection of speakers social characteristics

Approaches to Slavic Interaction

Part III. Displaying and negotiating epistemic and evidential

This is how I see it: No-prefacing in Polish

How can I lie if I am telling the truth? The unbearable lightness 

Part IV. Facework and contextualization in interaction

Parliamentary communication: The case of the Russian Gosduma

Impoliteness and mock-impoliteness: A descriptive analysis

Humor as staging an utterance

Table of contents vii

Part V. Language alternation in face-to-face interaction of bilingual families

Approaches to Slavic Interaction

Introduction and overview

Nadine Thielemann and Peter Kosta

Introduction and overview

2. Overview of the volume

Nadine Thielemann and Peter Kosta

Introduction and overview

Nadine Thielemann and Peter Kosta

assumption that spoken language, especially talk-in-interaction is shaped by its

Introduction and overview

Nadine Thielemann and Peter Kosta

Introduction and overview

the interlocutors in a specific way. This understanding incorporates politeness

Nadine Thielemann and Peter Kosta

Introduction and overview

Nadine Thielemann and Peter Kosta

Introduction and overview

Multimodal, grammatical and paralinguistic

Talking out of turn

Talking out of turn

co-construction restricts what follows it, in the second, co-constructed part.

outdoors, on the boardwalk. These interviews are open-ended. Although I asked

2. The mechanics of co-constructions

Talking out of turn

My father was wounded

(2) Completion, Boardwalk interview (Boris and Alla)

Nobody is paying attention.

Note that both extensions and completions do not constitute turn-constructional

Talking out of turn

(3) Bankrotstvo JuKOSa (27.06.2006; A. Venediktov and V. Geraenko)

In various interviews you have used such a strong word as idiocy.

3. The interactional nature of co-constructions

but it turns out

Talking out of turn

Reanimating responsibility: The we-V2 (take-V2) double imperative

Eye behavior in Russian spoken interaction and its correlation

Hesitation markers in transitions within (story)telling sequences

Speech rate as reflection of speakers social characteristics

This is how I see it: No-prefacing in Polish

How can I lie if I am telling the truth? The unbearable lightness

Parliamentary communication: The case of the Russian Gosduma

Impoliteness and mock-impoliteness: A descriptive analysis

Humor as staging an utterance