|AUTHOR: Koven, Michèle
TITLE: Selves in Two Languages
SUBTITLE: Bilinguals' verbal enactments of identity in French and Portuguese
SERIES: Studies in Bilingualism 34
PUBLISHER: John Benjamins
Jean-Jacques Weber, Departments of English and Education, University of Luxembourg
Koven's book provides a language-ideologically informed and ethnographically
grounded perspective on the Portuguese community in France. She studies the
bilingual identities of young luso-descendant women living in the Paris area,
and how they experience the link between language and identity. She is concerned
with describing her luso-descendant informants' experience of being a ''different
person'' in their two languages, in the sense that their varieties of French are
experienced and/or perceived as indexing the identity of young (sub)urban
Parisians, whereas their varieties of Portuguese are experienced and/or
perceived as indexing the identity of rural, provincial villagers. The fact that
they experience themselves differently in French and Portuguese deconstructs the
view of identity as fully formed independently of discourse. Koven's study thus
offers evidence in favor of a process view of identity as fully interlinked with
discourse, with the former at least partially emergent from the latter.
Koven brings together three perspectives in her exploration of these bilinguals'
subjectivities: the speakers', the discourse analyst's and the listeners'. She
analyzes how speakers explicitly report their feelings in speaking each
language, what discourse forms and patterns speakers use in their narratives of
personal experience, and how listeners perceive the speakers in each of their
In chapter 2, Koven presents her discourse-semiotic approach to bilingualism,
drawing upon the Bakhtinian notion of ''voicing'' as well as language ideological
ways of understanding indexicality in language and its link to identity. In her
analysis of how participants ''voice'' both selves and others in discourse, she
draws upon a wide range of research traditions, including sociolinguistics,
linguistic anthropology, psychotherapy, as well as experimental studies of
''personality'' and studies of bilingual memoirists. She emphasizes the importance
of observing three basic quality criteria: a semiotic theory of language which
allows an indexical understanding of how different personas get invoked in
bilinguals' use of their languages; an analysis of actual discourse, showing how
bilinguals use language to index the identities that they enact; and an
attention to local language ideologies and ethnographic contexts, in order to
reveal how people (tacitly) appeal to these language ideologies when inferring
their own or others' identities from discourse.
Chapter 3 provides an overview of the lived sociolinguistic contexts of
luso-descendants in France. Koven describes the background of her twenty-three
female luso-descendant speakers and five female luso-descendant listeners, all
of whom were university students at the time of the fieldwork. The speakers were
asked to tell narratives of personal experience in both of their languages, and
the listeners were at a later stage invited to react to recordings of the
original speakers' French and Portuguese narrative performances. All these
luso-descendants are members of the second generation, who have the capacity to
speak ''monolingual-like'' French but often feel linguistically insecure about
their Portuguese. Indeed, their variety/ies of Portuguese may reflect French
influences and/or index their parents' frequently rural or working-class origins.
The main part of the book is devoted to the analysis of the self-reports
(chapter 4), the analysis of narrative voicing (chapters 5-6) and of the
listeners' perceptions (chapter 7). While Koven is fully aware that the
metapragmatic self-reports cannot be taken as transparent reflections of what
speakers actually do in speech, she argues that they can provide useful insights
when combined with the analysis of the discourse forms used by the speakers and
of the listeners' perceptions of the speakers' speech. In particular, they
refract the speakers' language ideologies and their multiple understandings of
the role of language in their experiences of self, emotion and context.
Interestingly, speakers provide multiple, sometimes even contradictory accounts,
some more referential, some more contextual and others more psychological. For
instance, they frequently look upon their self as a fixed, pre-existing entity,
or they see themselves as taking on what they refer to as the French or
Portuguese ''mentality'' when speaking that language, or they perceive themselves
as having two ''personalities'', one in each language.
Chapter 5 lists the analytic categories for the study of narrative voicing.
Koven provides a systematic, qualitative and quantitative, analysis of patterns
of voicing across a large corpus of about 500 narratives. She distinguishes
between different speaker roles (as narrator, interlocutor, character), and
between the speech registers used within and across speaker roles. Chapter 6
presents the results of the analysis, painting a detailed picture of how
speakers enact bilingual selves differently in each language. For instance, they
make use of more interlocutor and character role speech and of a wider range of
registers (especially very colloquial or even ''vulgar'' registers) in French
Chapter 7 builds upon more traditional work in the area of language attitudes
but extends it by combining it with the study of language ideologies. In this
chapter, the results of the narrative analysis are triangulated with the
listeners' perceptions of speakers and their speech. Listeners, too, perceived
speakers differently in French and Portuguese: in French they frequently saw
speakers as the (stereo)type of the aggressive, assertive (sub)urban Parisian
youth, whereas in Portuguese they saw them as more reserved and ''hick-like''.
Thus the different verbal strategies that speakers use in each language invoke
and evoke different personas or ''types'' of people.
Chapters 8-9 add more ethnographic texture by focusing on two individuals and
investigating in detail how they use, experience and are seen through the
resources and contexts of their two languages. For the two speakers, Teresa and
Isabel, Koven provides a qualitative and quantitative comparison of the
discourse patterns in a story told once in each language, explores the women's
self-reports about how speaking each language affects them, and discusses the
listeners' reactions to recordings of both (re)tellings. The results of chapter
6 concerning the greater use of interlocutor and character role speech and of
more familiar or vulgar registers in French are confirmed here. One example is
Teresa's use of ''vous êtes qu'un con'' in the French telling, with an equivalent
expression completely missing from the Portuguese retelling of the same story.
Koven speculates as follows: ''Perhaps she [Teresa] feels she cannot fully evoke
personas of vulgar, racially marked insulters in Portuguese – because such
comparable personas are not available to her, culturally, or personally. She may
either not be able to or not feel comfortable making him and herself feel
equally 'vulgar' in Portuguese as in French'' (198) . As a result, Teresa comes
across as more outspoken and more ready to stand up for herself in the French
Like most of the luso-descendants, Teresa and Isabel are part of the landscape
of rural Portugal and the (very different) landscape of (sub)urban French youth
culture. One of the main findings of Koven's study is that the bilinguals'
sociolinguistic contexts do not easily translate from one into the other: in
particular, they did not feel able or ''entitled'' to enact the identity of the
aggressive and assertive (sub)urban youth in the Portuguese (re)tellings of
their stories. In other words, this identity of the tough Parisian youth
speaking in colloquial and even vulgar registers is available to them in French
but somehow they did not deem it appropriate for themselves in Portuguese. There
is thus a difference between ''what speakers can 'do' and who they can 'be' in
each language'' (72), and Koven concludes that it is this that underlies their
experience of being a ''different person'' in French and Portuguese.
Another important point that Koven's analysis brings out is the role of language
ideologies in these processes. She argues that it is because the
luso-descendants speak two named languages associated with ideologically
monolingual nation-states that they construe the identities linked with these
two languages as separate and compartmentalized. Hence the luso-descendants'
experience of a split self may actually to some extent be an effect of language
ideologies. Finally, Koven notes that the luso-descendants tended to erase the
sociolinguistic diversity within each of their languages, ''and yet, ironically,
in large measure what varied most for these women BETWEEN languages were the
registers they used WITHIN each language'' (246).
This is a brilliant study of the sociolinguistic contexts of transnational
luso-descendants living in the Paris area, and as such usefully complements
other recent sociolinguistic studies of luso-descendants in northern European
countries by Barradas (2007) and Weber (forthcoming). Barradas' study of
children attending Portuguese community schools in England reveals that learning
strategies and metalinguistic knowledge acquired by the children in these
schools tend to be transferred to mainstream schooling and translate into higher
overall educational achievement. Weber's study of luso-descendants in Luxembourg
can be compared more directly with Koven's book, as both authors analyze the
luso-descendants' language use as well as the ''meta'' dimension of their language
ideologies. Interestingly, Koven adds a third dimension of analysis: namely, how
their narrative tellings and retellings in both French and Portuguese are
perceived by other people. Another difference is that in Luxembourg
luso-descendants seem to be more hampered in their desire for upward social
mobility than in France because of the additional presence of Luxembourgish and
German in the linguistic environment of Luxembourg (in particular, in its
school-system). A symptom of this might be that Koven's informants are all
university students, whereas most of the informants in Weber's study are
students in vocational secondary education.
But beyond the intrinsic interest of the actual case study, Koven's book is a
ground-breaking study with important consequences both of a methodological and
of a theoretical nature (that is, if the two can be separated at all). From a
more methodological perspective, it is highly innovative in that it provides a
systematic analytic framework for operationalizing Bakhtin's notion of voicing,
and it uses multiple empirical - qualitative and quantitative - approaches which
not only provide convergent evidence but also rely upon sound ethnographic
fieldwork. From a more theoretical perspective, it significantly advances our
understanding of the role of language ideologies in the relationship between
language and identity, and hence it is or should be essential reading for
anybody interested in the study of bilingualism and the role of language in
experiences of self.
Barradas, Olga. (2007) Learning Portuguese: A tale of two worlds. In J. Conteh,
P. Martin and L. Helavaara Robertson (eds) _Multilingual Learning: Stories from
Schools and Communities in Britain_. Stoke-on-Trent: Trentham Books, 87-102.
Weber, Jean-Jacques. (forthcoming) _Multilingualism, Education and Change_.
Frankfurt/Main: Peter Lang Verlag.
ABOUT THE REVIEWER
Jean-Jacques Weber is Professor of English and Education at the University of
Luxembourg. His main research area is the study of language and education in
multilingual and multicultural contexts (such as Luxembourg). He is co-author
(with Kristine Horner) of ''The language situation in Luxembourg'' (Current Issues
in Language Planning 2008), and is now completing a monograph for Peter Lang
Verlag entitled _Multilingualism, Education and Change_.