AUTHOR: Lindström, Jan
TITLE: Tur och ordning
SUBTITLE: Introduktion till svensk samtalsgrammatik
PUBLISHER: Norstedts Akademiska Förlag
YEAR: 2008
Madalena Cruz-Ferreira, independent scholar
INTRODUCTION
Jan Lindström's book title draws on a Swedish idiom, ''i tur och odning'', which
can be rendered in English as 'in orderly turns' for the purposes of this
review. The subtitle, 'Introduction to Swedish conversation grammar', explains
the word play on the term ''tur'', 'turn'.
This is the first book-length treatment of this topic, gathering together and
advancing findings from a significant body of publications on Swedish corpus
research involving universities in Gothenburg, Uppsala, Linköping and Helsinki,
the latter being Lindström's institutional affiliation. The book proposes a new
model for the analysis of patterns and functional units of spoken Swedish, and
will be of relevance to teachers and students, as well as to anyone with an
interest in language and communication .
In what follows, English renditions of Swedish quotations are my own, given in
square brackets.
SUMMARY
The book consists of seven chapters, followed by a cross-referenced glossary,
the list of databases used in the analysis, a bibliography, a subject index and
the set of conventions used in transcription.
Chapter 1, ''Inledning'' [Introduction], presents the book's contents and layout,
and announces the book's twin goal. First, to present a new model for the
description of spoken interaction, against the background of recent and
important work on grammar and language use, which nevertheless focuses away from
spoken usage, including Swedish usage. Second, to dispel the layman's conception
of spoken language as lacking structure and therefore grammar. This
misconception arises to a large extent from the contents of ordinary grammar
books, which can hardly be said to address the language used in everyday spoken
interaction. The descriptions that traditional grammars do offer thus tend to be
taken as the only acceptable norms of language use. Lindström's conversation
grammar offers instead evidence that ''talet består av mycket mer än bara det som
vanligen normeras i en grammatikbok'' (p. 13) [speech consists of much more than
what little is usually normed in a grammar book].
The chapter then addresses choices of transcription, particularly the
ever-vexing issue of how to do justice to the richness of speech within the
two-dimensional boundaries of print. Standard printed forms of language give the
same orthographic weight, most notably as identifiable 'words', either to single
words which have no correlate in connected speech or to word sequences which
speakers treat as single entities. Conversely, recurrent features of everyday
spoken utterances find no correlate in standard orthographies. Lindström opts
for a (very readable) pronunciation spelling, as it were, using the standard
Swedish alphabet to render visual counterparts of actual auditory input. For
example, what we hear as ''hörru de junte så'' (p. 15) has the well-behaved
spelling ''hör du det är ju inte så'' [listen, it's not like that at all], spoken
''att'' and ''och'' [infinitive 'to' and 'and'] are both ''å'', and ''ser du'' [you see]
is one single entity, variously pronounced ''serru'' and ''sörrö''. Other
transcription conventions are explained here, including filled pauses,
ingressives and several features of intonation and prosody, all usefully
gathered in the last page of the book.
Chapter 2, ''Orientering om samtalsgrammatik'' [Background to conversation
grammar], sets studies of Swedish grammar in a historical perspective,
highlighting differences in analytical focus, and so the different grammars that
the language has had described – and prescribed, including judgments which
remain familiar today, of written language as good language and related
recommendations that speech should emulate print. Lindström argues that
accumulated knowledge about written Swedish, minus the judgmental flourish,
affords a workable starting point for his own analysis of the language, which
complements it. Descriptions of spoken data are needed because it is as
unreasonable to expect speech to reproduce print as to expect the converse.
These are different kinds of knowledge about the same language that both need
exploring with tools that reflect their similarities as well as differences.
The chapter moves on to discussion of different types of conversation, which can
vary in setting and number of participants, the most common being dyadic
face-to-face interactions in everyday settings with family and friends. All
forms of conversation share a dialogic organization, that is, participants speak
in turns, respond to previous turns and anticipate coming turns (p. 29). The
'turn' can therefore be assumed as the basic unit of conversational analysis and
thus as a fully-fledged unit of grammatical description.
The bulk of the chapter is then taken up by a review of traditional levels of
description of Swedish grammar (e.g. phonology, semantics), focusing on how
their constructs fit or fall short of a description of the grammar of speech.
For example, the study of speech syntax makes little sense without regard to
speech prosody (p. 39) or even to visual signals like body language in
face-to-face interaction (p. 49). Traditional constructs like 'sentence' lose
the intrinsic analytical centrality which has been assumed of them, in that
syntactic usage itself can only make sense in context. Lindström observes that
utterances are best seen not as products which can be usefully analyzed on their
own, but as processes, as part of a cooperative construction of meaning that is
monitored on the fly, whose linguistic features adapt to, and are generated by,
the interaction itself, as participants assess what went on, process what is
going on and predict what will go on (p. 43).
Chapter 3, ''Talets enheter'' [Units of speech], starts with the assumption that
language use, and therefore its grammar, has ''en inre projektionskraft'' (p.51)
[an intrinsic projection power], which is the potential of language units to
combine in systematic ways. Linguistic units of relevance to conversation may
have no pre-determined linguistic form, since their boundaries and their
internal structure are permanently in progress during interaction. Lindström
chooses to use the term ''yttrande'' [utterance] as a convenient cover-word for
any speech unit which constitutes a grammatical, pragmatic and prosodic entity.
A turn constructional unit (TKE in the Swedish acronym, for
''turkonstruktionsenhet''), which has been a recurrent focus of research and
controversy in conversation analysis, is then an utterance whose linguistic
make-up and sequential occurrence (intuitively) make it a dialogic entity
(p.53-55). An utterance is a potential turn constructional unit much, Lindström
argues, as a speech sound is a potential phoneme (p. 55).
The remainder of the chapter discusses and re-defines analytical concepts like
word class membership, phrase, syntactic ordering and prosodic unit, from the
perspective of conversation grammar. That is, from the perspective of their
linguistic potential in an interaction. Paradigmatic and syntagmatic potentials
of linguistic constructions, for example, which are activated by single words,
build a net around those words such that ''det delinnehål som de enskilda orden
står för kan modifieras och berikas på ett oändligt sätt'' (p. 68) [the partial
content which particular words represent can be modified and enriched in
infinite ways].
Chapter 4, ''Diskursmarkörer'' [Discourse markers], deals with linguistic signals
which regulate an ongoing conversation and/or act as modifiers to utterances,
whether these utterances are one's own or others'. Discourse particles form a
class of their own which divides into subclasses depending on their
syntactic-functional contribution to the interaction. For example, ''vetdu'' [you
know] is an utterance particle whose function is to frame the utterance, in that
it occurs preferentially at utterance-edge, and ''liksom'' [like] is a focal
particle, calling attention to what, in the majority of cases, follows it.
Secondary discourse markers are words/expressions with a potential to serve
additional regulatory functions, although belonging to different word classes,
open as well as closed. Examples are ''förresten'' [by the way], ''vänta'' [wait],
''så där'' [like that]. Question frames like ''jag undrar'' [I wonder], conditioning
constructions like ''om man säger så'' [to put it that way] and even independent
utterances themselves are also used to regulate an interaction.
Having identified the turn as the analytical primitive in a model of
conversation in Chapter 2, Chapters 5 and 6 deal with the external and internal
patterns which characterize conversational turns, respectively. Chapter 5,
''Organisering av interaktionen'' [Organisation of the interaction], addresses the
systematic organization of conversation as a sequence of turns, that is, an
orderly sequence of participation in an interaction. Lindström details
turn-taking, turn-holding and turn-giving devices, whose accurate identification
and use depend on the interactants' online ability to interpret the syntactic,
prosodic and pragmatic cues which make up the physical manifestations of turns
(p. 123). The core role played by adjacency pairs in fluent conversation is
highlighted, as are the ways in which speakers signal preferred and dispreferred
continuations. The spoken structure of preference itself is also detailed (p. 141).
The analysis of repairs takes up a significant part of this chapter. Repairs are
practices through which participants attempt to resolve problems of speech
delivery, audibility and/or understanding. Repairs can be self- or
other-initiated (p. 149), but make use of specific linguistic resources, for
example, repetition, with copied or other intonation, use of dedicated words or
expressions (''eller'' [or/rather], ''va sa du?'' [pardon?/what?], and the
characteristic Swedish ''x-och-x'' [x-and-x] construction, which contains a
reduplication of a previously used word (x) in order to qualify it or prevent an
interpretation of it that the speaker just realized is possible but is
unintended. To give one example, the word ''många'' [many], used in this repair
frame (''många å många'', p. 163), in fact means 'not that many'. An interesting
observation is that the loci of interruptions, including those due to repair,
correspond to syntactic positions which require completion, thereby reinforcing
the expectation that participants provide proper grammatical closure to a
conversational turn (p. 171).
Chapter 6, ''Turdesign'' [Turn organisation], deals with the structure of turns
themselves. Lindström establishes a turn hierarchy, where subordinate turns
assist a successful interaction, for example in the form of parentheses or
fill-ins, and must exist in dependency to another turn (p. 184). The bulk of the
chapter addresses turn segmentation, that is, the linguistic features which make
it clear that a turn switch is in order (p. 201). For example, ''vetdu'' [you
know] with a mild rising intonation marks a turn end, whereas ''hördu'' [listen]
is a turn-initiating device. Lindström's detailed analysis of the internal
organization of turns uncovers a tight integration of interactional practices
and grammatical structure. Expectedly, the conclusion is that fluent
conversation takes place in the same patterned way as other uses of language,
which therefore are all amenable to description by a single, unified grammar.
For a discussion, in English, of the topic of this chapter, see Lindström (2006).
Chapter 7, ''Avslutning och kommande turer'' [Conclusion and upcoming turns],
briefly draws together the major insights about language garnered through the
book. A grammar is not just a description of forms of linguistic units and is
not restricted to traditional analytical units either. What is taken as core or
peripheral to grammatical analysis depends on the analyst's standpoint, not on
the language material itself. Lindström's take is that language material comes
to be, that is, makes sense, in context and must therefore be analyzed in the
context itself of its use.
REVALUATION
Conversation is arguably the most cooperative form of language use. It has
indeed been hailed as the prime form of linguistic interaction, most notably in
Dunbar's (1996: 123) finding ''strong support for the suggestion that language
evolved to facilitate the bonding of social groups, and that it mainly achieves
this aim by permitting the exchange of socially relevant information.'' The
grammar of what we say certainly deserves close attention, because most of what
we have to say is said in direct interaction with other people, whose spoken
feedback we in turn expect. Lindström's in-depth analysis, besides written in an
engaging style, with neatly presented text and examples, is further proof of the
insights that this topic has to offer.
Lindström convincingly argues that we cannot make sense of language if we
continue to disregard the grammar of core features of it like prosody or the
pragmatic weight of utterances, which are necessarily present in spoken
interaction. He furthermore shows how findings from interactional grammar in
fact fine-tune our understanding of traditional grammar. For example,
participant B's completion of participant A's truncated syntactic structure (p.
178) is evidence that we do not speak in 'sentences' or 'phrases', which we
perhaps already suspected, but also that 'sentences' and 'phrases' do make
interactional sense to speakers. Traditional constructs like modifier or
constituent boundary also find their place in Lindström's dynamic grammar model,
where the core issue is meaning making, not form building. For example, a
'directive' is recognized as such as it builds itself from interaction, rather
than from specific features of utterances. In short, grammar is best understood
in action. Participants interrupt themselves and others with repairs or
regulatory remarks, signal turn-keeping to buy time to complete their intended
meaning or signal turn-taking to help complete someone else's, tightly
monitoring the puzzle pieces of everyone's contributions, which are themselves
being formed as the conversation progresses. There is no blueprint to guide the
interaction, and conversation has no 'undo' function: what is said cannot be
unsaid and the successful outcome of a conversation must therefore be
constructed from what is being said, in real time. The cooperative nature of
conversation becomes clear from the observation that virtually any portion of a
spoken interaction, taken out of context, becomes meaningless.
Lindström's model is unifying. His methodology draws on usable traditional
grammatical constructs to build new proposals which feed back into those same
constructs' inadequacies to portray the actual richness of language uses. His
discussion of traditional word classes and of the rationale for new ones is
particularly illuminating in laying bare the fact that 'parts of speech' in fact
have little to do with speech at all. Although approaching his topic with
classificatory goals similar to traditional ones, Lindström is well aware of
what I would call the taxonomy hubris that befalls many like-minded analysts.
Identifying and naming things is quite entertaining and may indeed become
addictive, so Lindström's proviso that there is always a ''risk för [...]
godtycklighet'' (p.121) [risk of gratuitousness] in qualitative analytical
endeavors is most welcome. Taxonomies involve choices that solve and raise
problems that other choices might raise and solve, respectively. We know that
''all grammars leak'' (Sapir 1921: 38) and there's no reason to expect otherwise
of a grammar of conversation. Lindström sensibly supports his choices
throughout, thereby giving clear evidence of the complexity inherent to
interactional grammar. For the benefit of the reader, summaries of findings in
table format accompany discussion of each topic.
Being unifying, this model lays claims to universality. Although patterns of
conversational behavior are as language-dependent as any other patterns of
linguistic behavior, Lindström's Swedish model is flexible enough to be
generalizable to other languages. Several of his observations in fact remind of
findings in apparently unrelated research, for example, on codeswitching: the
locus of a language switch, like the locus of a turn switch, obeys strict
grammatical constraints of the language(s) in question.
The book contains a few suggestions for further study. Lindström observes, for
example, that syntactic subordination does not play a significant role in
conversation and he suggests that there may be typologically different kinds of
conversation, depending on factors like the social/hierarchical status of the
interactants or the physical context where the interaction takes place. Language
varieties certainly add to the picture, although the analysis of dialectal
variation is not one goal of this book. As expected in a first book on his
topic, Lindström follows mainstream practices of dealing with particular
languages in a general sense. His is a grammar of ''Swedish'', with data gathered
in both Finland and Sweden, in the same sense that we talk about the teaching of
''English'' or the phonology of ''Mandarin''. In this connection, I have two
suggestions. First, that one (or several) companion websites to this book would
be an invaluable resource, not only for audio access to the data but for updates
on host servers themselves. I was for example unable to access the one weblink
(Gothenburg Spoken Language Corpus) included in the extensive database used in
this book. Second, that the variety of Swedish contemplated in each corpus of
the database be specified. The corpora are identified by geographic location in
the book, but speaker location does not necessarily match language variety.
On a practical level, Lindström is understandably more concerned about the
impact of his findings among users of Swedish, presumably monolingual, in
everyday personal and professional settings where interaction is crucial. The
relevance of these findings for foreign language (FL) teaching/learning is
mentioned in passing, in one short sentence towards the end of the concluding
chapter (p. 278). I believe that their impact in FL settings cannot be
overstated. Partly because of significant immigration and associated language
integration issues, which are current particularly in Sweden, partly and more
importantly, because FL teaching continues to guide itself by the dictums of
traditional-minded grammarians. Most everyday interactions in a new country and
in a new language are conversations, though hardly the stilted textbook ones
where learners are required to enact a script that does not belong to them. It
is not only through the use of wrong word order or vocabulary that one makes
serious interactional mistakes in a new language, it is through faulty
perception, processing and use of prosodic and pragmatic signals.
Lindström's book adds to the timely surge of research in what I would call
Neglected Linguistics, the linguistics of real-life language – Carter & McCarthy
(2006) is one example, also corpus-based. I share Lindström's hope that his
book, by dealing with language which is recognizable from everyday usage, may
blow new life into the realization of what grammar is all about and so help
spring-clean ''grammatikens kanske en aning dammiga begrepp'' (p. 279) [grammar's
somewhat dusty conceptions].
REFERENCES
Carter, Ronald & Michael McCarthy. (2006). _Cambridge Grammar of English: A
Comprehensive Guide. Spoken and Written English Grammar and Usage_. Cambridge:
Cambridge University Press.
Dunbar, Robin I.M. (1996). _Grooming, gossip and the evolution of language_.
London: Faber & Faber.
Lindström, Jan. (2006). Grammar in the service of interaction: Exploring turn
organization in Swedish. _Research on Language and Social Interaction_ 39(1),
81–117.
Sapir, Edward. (1921). _Language. An Introduction to the Study of Speech_. New
York: Harcourt Brace.
ABOUT THE REVIEWER
Madalena Cruz-Ferreira is currently working on norms of language acquisition and
use in multilingual settings. Her research interests include (child)
multilingualism, multilingual phonology and prosody, the language of science and
linguistics pedagogy.
|
