See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/292309492

Repertoires Revisited: “Knowing Language” in Superdiversity

Article · January 2011

CITATIONS READS

209 2,569

2 authors:

Jan Blommaert Ad Backus

Tilburg University Tilburg University
304 PUBLICATIONS   15,060 CITATIONS    78 PUBLICATIONS   1,379 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Computational modeling of bilingual argument structure acquisition View project

Common Ground: effects of familiarity View project

All content following this page was uploaded by Jan Blommaert on 10 February 2016.

The user has requested enhancement of the downloaded file.

 Paper

Superdiverse Repertoires and the Individual

by

Jan Blommaert & Ad Backus

[email protected]
[email protected]

March 2012
To appear in: Ingrid de Saint-Jacques & Jean-Jacques Weber (eds) ‘Multimodality and
Multilingualism: Current Challenges for Educational Studies’. Sense Publishers, Rotterdam;
2012

Superdiverse Repertoires and the Individual

Jan Blommaert & Ad Backus

Abstract:

Repertoire belongs to the core vocabulary of sociolinguistics, yet very

little fundamental reflection has been done on the nature and structure of
repertoires. In early definitions, repertoires was seen as a triad of
language resources, knowledge of language (‘competence’) and a
community. Due to developments in the study of language competence
and in the study of social organization, this triad can no longer remain
intact. In a super-diversity context, mobile subjects engage with a broad
variety of groups, networks and communities, and their language
resources are consequently learned through a wide variety of trajectories,
tactics and technologies, ranging from fully formal language learning to
entirely informal ‘encounters’ with language. These different learning
modes lead to very different degrees of knowledge of language, from very
elaborate structural and pragmatic knowledge to elementary
‘recognizing’ languages, whereby all of these resources in a repertoire are
functionally distributed in a patchwork of competences and skills. The
origins of repertoires are biographical, and repertoires can in effect be
seen as ‘indexical biographies’. This, then, allows us to reorient the triad
of repertoires away from communities towards subjectivities, and suggest
that repertoire analysis can be a privileged road into understanding Late-
Modern subjectivities.

Keywords: Repertoire, language learning, subjectivity, super-diversity,

globalization, competence, indexical biography, sociolinguistics

1
1. Introduction

The term ‘repertoire’ belongs to the core vocabulary of sociolinguistics.1 John

Gumperz, in the introduction to the epochal Directions in Sociolinguistics: The
Ethnography of Communication (Gumperz & Hymes 1972 (1986)) lists ‘linguistic
repertoires’ as one of the ‘basic sociolinguistic concepts’ (Gumperz 1972 (1986):
20-21) and defines it as “The totality of linguistic resources (i.e. including both
invariant forms and variables) available to members of particular communities”
(italics added). In his equally epochal Discourse Strategies, he reformulated this
notion, basically juxtaposing his original definition with the wider range of
phenomena programmatically addressed by Hymes (1972a (1986); 1975):

“Studies of language use are called for which concentrate on what Hymes
calls the means of speaking. This includes information on the local
linguistic repertoire, the totality of distinct language varieties, dialects and
styles employed in a community. Also to be described are the genres or
art forms in terms of which verbal performances can be characterized,
such as myths, epics, tales, narratives and the like. Descriptions further
cover the various acts of speaking prevalent in a particular group (…), and
finally the ‘frames’ that serve as instructions on how to interpret a
sequence of acts.” (Gumperz 1982: 155; italics in original; cf also Bauman
& Sherzer 1975: 7)

The narrower notion of ‘linguistic repertoires’ is here combined with the broad
and somewhat less precise notion of ‘means of speaking’. The job of the
Gumperz-Hymesian sociolinguists was to describe all of that, to put these things
in relation to each other, and to interpret them in terms of that other key notion

1This paper grew out of discussions within the Max Planck Sociolinguistic Diversity
Working Group. A preliminary version was presented at a colloquium on sociolinguistic
superdiversity held at the Max Planck Institute for Ethnic and Religious Diversity, Göttingen,
November 2010, as a plenary lecture at the 32nd Ethnography in Education Forum at the
University of Pennsylvania, February 2011 and as a lecture in the series The Future of
Educational Studies, University of Luxemburg, September 2011. We are grateful for the
comments provided by audiences at all of these occasions, in particular those of Jens-
Normann Jörgensen, whose incisive comments greatly improved the argument in this paper.
This paper draws extensively on a broader-aimed one, Blommaert & Backus (2011), and
anticipates further developments in this direction.

2
in sociolinguistics, ‘communicative competence’ – the knowing what and knowing
how to use language which Hymes pitted against Chomskyan ‘competence’
(Hymes 1972b is the locus classicus, see also Hymes 1992). ‘Repertoire’ so
became the word we use to decribe all the “means of speaking” i.e. all those
means that people know how to use and why while they communicate, and such
means, as we have seen, range from linguistic ones (language varieties) over
cultural ones (genres, styles) and social ones (norms for the production and
understanding of language). In the eyes of Gumperz, Hymes and their peers,
repertoires were tied to particular speech communities, the third key
sociolinguistic notion. Repertoires characterized communities within which the
sharedness of repertoire guaranteed smooth and ‘normal’ communication. This
collocation of repertoires and communities was a precipitate of, let us say,
‘traditional’ ethnography, in which the ethnographer studied a ‘community’ – a
group of people that could somehow be isolated from the totality of mankind and
be studied in its own right.

This is very much where the concept has stayed since then; there has not been
much profound reflection on the notion of repertoire.2 The term is commonly
used in sociolinguistics, usually as a loosely descriptive term pointing to the total
complex of communicative resources that we find among the subjects we study.
Whenever ‘repertoire’ is used, it presupposes knowledge – ‘competence’ –
because ‘having’ a particular repertoire is predicated on knowing how to use the
resources that it combines. The four decades of use of the term and its links to
other concepts, however, have seen quite some shifts and developments, notably
in the field of what one can broadly call ‘language knowledge’. This paper seeks
to engage with these developments and to bring them to bear on the notion of
repertoire. If patterns of language knowledge are better understood, we may be
in a position to be more precise in what we understand by repertoires. Likewise,
we have moved on in our understanding of ‘community’; and here, too,

2 The other key notions, in contrast, did attract a considerable amount of theoretical
reflection. Hymes himself questioned the idea of isolated and closed speech communities in
his essay on the concept of ‘tribe’ (Hymes 1968); more recent critiques of the traditional
concept of speech communities include Rampton (1998). Blommaert (2005, 2010)
announced the crucial role of repertoires in further work and spelled out its potential
relevance.

3
important new insights can be projected onto the concept of repertoire.
Repertoire can so be turned into an empirically more useful and theoretically
more precise notion, helpful for our understanding of contemporary processes of
language in society.

This is the intellectual motive for this paper. There is, however, a more practical
(or polemical) motive as well. In spite of significant advances in the field of
language knowledge, dominant discourses on this topic seem to increasingly
turn to entirely obsolete and conclusively discredited models of language
knowledge. The European Common Framework for Languages is naturally the
most outspoken case, but language and literacy testing methods predicated on
linear and uniform ‘levels’ of knowledge and developmental progression are
back in force. Such practices and methods have met debilitating and crippling
criticism from within the profession (see the essays in Hogan-Brun 2009; also
Spotti 2011); yet they remain unaffected and attract more and more support
among national and supranational authorities in fields of immigration, labor and
education. Something is seriously wrong there, and this paper can be read as yet
another attack on the linguistic and sociolinguistic assumptions underlying this
complex of tests and models.

In the next section, we will summarize the most important developments our
understanding of the structure of contemporary societies. Armed with these
insights, we will set out to describe patterns of learning “the means of language”.
Such patterns, we will argue, are widely different in nature and in ‘technology’,
they range from highly formal modes of patterned learning to highly informal
and ephemeral ‘encounters’ with language. These different modes of learning
and acquiring lead to different forms of knowledge, and this is the topic of the
next section. We will consider the repertoires that can emerge from the widely
varied modes of learning and highlight some less expected modes of ‘knowing
language’ as elements of repertoires. In a final concluding section, we will
connect such repertoires to the wider historical frame in which they operate:
Late Modernity and its particular forms of subjectivity. Let us now turn to some
central insights which we need to take on board in this exercise.

4
2. Superdiversity

Questions of what is shared and not in the field of cultural (including linguistic)
knowledge acquire a particular urgency and relevance in the context of
superdiversity. Superdiversity is a descriptive term, denoting the new
dimensions of social, cultural and linguistic diversity emerging out of post-Cold
War migration and mobility patterns (Vertovec 2007). The new migrations
characterizing the post-1991 order in many parts of the globe, as well as the
emergence of mobile global communication systems such as the internet, have
led to extreme degrees of diversity to which the application of notions such as
‘diaspora’, ‘minority’, but also ‘community’ and other basic terms from the social-
scientific register have become increasingly problematic. ‘Ethnic’ neighborhoods
have turned from relative homogeneity into highly layered and stratified
neighborhoods, where ‘old’ migrants share spaces with a variety of ‘new’
migrants now coming from all parts of the world and involved in far more
complex and unpredictable patterns of migration than the resident and diaspora
ones characterizing earlier migration patterns. And while social life is primarily
spent in such local neighborhoods, the internet and mobile phone afford
opportunities to develop and maintain social, cultural, religious, economic and
political practices in other places. Exiled political leaders can remain influential
political actors in their countries of origin, even when they live in Rotterdam,
Marseille or Frankfurt; isolated individuals can maintain intense contacts (and
live social and cultural life) in a transnational network; languages can be used
through such networks as well, while they are absent from everyday
communicative practices in the local neighborhood. In general, most of the
‘normal’ patterns of social and cultural conduct that were central in the
development of social-scientific theories have now been complemented with a
wide variety of new, ‘abnormal’ patterns, for which we are hard pressed to
provide adequate accounts.

The impact of superdiversity is therefore paradigmatic: it forces us to see the

new social environments in which we live as characterized by an extremely low
degree of presupposability in terms of identities, patterns of social and cultural
behavior, social and cultural structure, norms and expectations. People can no

5
longer be straightforwardly associated with particular (national, ethnic,
sociocultural) groups and identities; their meaning-making practices can no
longer be presumed to ‘belong’ to particular languages and cultures – the
empirical field has become extremely complex, and descriptive adequacy has
become a challenge for the social sciences as we know them.

The implications of this for sociolinguistics have been sketched in a growing

body of work (e.g. Blommaert 2010; Creese & Blackledge 2010; Otsuji &
Pennycook 2010; Jörgensen et al 2011; Blommaert & Rampton 2011 provide an
overview), and they revolve around: (a) an increasing problemization of the
notion of ‘language’ in its traditional sense – shared, bounded, characterized by
deep stable structures; (b) an increasing focus on ‘language’ as an emergent and
dynamic pattern of practices in which semiotic resources are being used in a
particular way – often captured by terms such as ‘languaging’, ‘polylingualism’
and so forth; (c) detaching such forms of ‘languaging’ from established
associations with particular groups – such as ‘speech communities’ or ‘cultures’;
(d) viewing such groups exclusively in terms of emerging patterns of semiotic
behavior with different degrees of stability – ‘speech communities’ can be big
and small, enduring as well as extremely ephemeral, since they emerge as soon
as people establish in practice a pattern of shared indexicalities; (e) and seeing
people as moving through a multitude of such groups in ‘polycentric’ social
environments characterized by the presence and availability of multiple (but
often stratified) foci of normativity.

All of this is grounded in sociolinguistic and linguistic-anthropological work (e.g.

Silverstein 2004; Agha 2007). It is clear that work on communication in
superdiverse environments is not well served with a priori notions of ‘language’,
‘community’, or ‘understanding’, but must proceed from observations of actual
usage, and that it must allow for tremendous variability in observation and
interpretation.3 The stability that characterized the established notions of
language can no longer be maintained in light of the intense forms of mixing and
blending occurring in superdiverse communication evironments (both in spoken

3In Blommaert & Backus (2011), we examine the compatibility of these insights with recent
developments in usage-based linguistics.

6
and written forms of language; for the latter see e.g. Juffermans 2010 and Varis &
Wang 2011), and established notions of competence are in need of revision in
light of the highly unequal patterns of distribution of communicative resources
resulting in the often ‘truncated’ and ‘unfinished’ character of communication
(see e.g. Blommaert 2010, chapter 4; Kroon, Dong & Blommaert 2011).

In what follows, we shall engage with the paradigmatic challenge of

superdiversity and revisit patterns of language learning and the repertoires that
are results of such learning processes. The attempt is to reconstruct the concept
of repertoire in a descriptively realistic manner, driven by our usage-based focus
and attempting to avoid as much of the traditional linguistic and sociolinguistic
biases as possible.

3. Language learning trajectories

In superdiverse environments, patterns of ‘learning’ languages are widely

diverse. ‘Learning’ is a somewhat uneasy term that requires qualification, and
this will become clear when we review some patterns below. We use the term
here for the broad range of tactics, technologies and mechanisms by means of
which specific language resources become part of someone’s repertoire.
‘Acquisition’ is another candidate as shorthand for this complex of phenomena
and processes, but the term suggests an enduring outcome (resources have been
‘acquired’ once and for all), while ‘learning’ does not (one can ‘unlearn’ or ‘forget’
what one has learned). Hence the pragmatic choice for ‘learning’.

3.1. The biographic dimension of repertoires

With the distinction between ‘acquisition’ and ‘learning’, we have already

introduced a major differentiating feature into our discussion: the fact that some
effects of learning are permanent and enduring (e.g. learning the grammatical
patterns of a prominent language in one’s repertoire), while others are
temporary and dynamic. Discursive and sociocultural features would typically be
temporary and dynamic, in the sense that their learning patterns closely follow
the biography of the person. When someone is six years old, s/he speaks as a six-
year old. At the age of twelve this pragmatic complex of speech practices has
disappeared and has been replaced by another complex; likewise at the age of

7
eighteen, thirty and sixty: with each stage of life we learn the modes of
communication of that stage of life, and we unlearn part of the modes
characterizing earlier stages. At the age of forty, we cannot speak as a teenager
anymore. We can speak like a teenager, i.e. imitate the speech forms we observe
in teenagers (or remember from our own teenage years); but we cannot speak as
a teenager, deploying the full range of communication resources that define
people as teenagers. At the same age, we cannot yet speak as a very old person –
learning these resources will happen later in life. We can speak as a middle-aged
person, and the resources we can deploy define us as such.

This must be kept in mind: the ‘language’ we know is never finished, so to speak,
and learning language as a linguistic and a sociolinguistic system is not a
cumulative process; it is rather a process of growth, of sequential learning of
certain registers, styles, genres and linguistic varieties while shedding or altering
previously existing ones. Consequently, there is no point in life in which anyone
can claim to know all the resources of a language. Actual knowledge of language,
like any aspect of human development, is dependent on biography. As for other
aspects, knowledge of language can be compared to the size of shoes. Shoes that
fit perfectly at the age of twelve do not fit anymore at the age of thirty – both
because of the development of one’s body size and because of fashion, style and
preference (few of us would feel comfortable in the types of shoes we wore in
the 1970s). Repertoires are individual, biographically organized complexes of
resources, and they follow the rhythms of actual human lives.

This means that repertoires do not develop in a linear fashion. They develop
explosively in some phases of life and gradually in some others. Let us give one
very clear example. A child, typically, experiences an explosion of literacy
resources in the first couple of years of primary schooling. Between the age of six
and eight/nine, a child passes through the intensely difficult exercise of learning
how to write and read (see Kress 1997 for a classic survey and discussion) – not
just technically (increasingly not just in longhand but also on a keyboard) but
also ideologically, by attributing particular values to writing and reading
achievements – the sociocultural norms of literacy (Collins & Blot 2003). The
outcome is that starting (typically) from scratch, a child learns to write

8
linguistically and sociolinguistically relatively complex texts, and read large
volumes of such texts. Once this revolutionary stage is over, literacy skills
develop more gradually and incrementally. In the same stage of life, children
learn another vast complex of linguistic and sociolinguistic practices: ‘school
language’, the discourse patterns of formal education. S/he learns how to talk
and write as a pupil, and s/he learns how to listen to and read from instructors,
follow up their instructions, and convert them into regimented, ordered forms of
discourse practice. The child learns genres, registers and styles that are specific
to formal educational environments and have hardly any validity outside school
– think of Latin, mathematics or physics as a discursive field, for instance. This,
too, is a massive achievement which marks their repertoires for life, allowing
more gradual expansion and development after that.

With every new stage of life we learn new linguistic and sociolinguistic patterns.
Becoming a teenager involves exploring the experiential worlds of love and
relationships, of sexuality, of popular culture and of identity opportunities that
deviate from those preferred and organized by school or parents. Those who
proceed to higher education learn how to speak and write in new ways there,
and for many this period of life coincides with first experiences as someone who
lives apart from his/her parents and has to navigate that new complex world of
opportunities and responsibilities. Becoming an employee in the labor market
involves similar dramatic jumps in learning, as one acquires the discourse
patterns of specific and specialized professions as well as those of a salaried
independent person and consumer, now capable of purchasing expensive items
such as cars or a house (and having to manoeuver complicated financial, legal
and insurance aspects of it). Becoming a parent likewise induces one into an
entire world of new discourses, just as becoming unemployed, chronically ill, a
widow or widower, or a retired person come with new and highly specific
linguistic and sociolinguistic resources.

3.2 Learning by degree

We learn all of these new skills and resources in a variety of ways. The most
visible ways are those of formal learning environments: school and college, but

9
also formal training sessions, evening courses, self-study on the basis of a set
curriculum, and so on. Such formal patterns of learning result in particular forms
of skills and resources: uniformly distributed ones over the collective of students
who participate into the same learning environment, regimented and
normatively elaborated, often also with a high degree of self-awareness that this
is ‘knowledge’ (as in “I learned German at school”). Such formal patterns of
learning always go hand in hand with patterns of learning in informal learning
environments – the family, peer groups, media and popular culture or just life
experiences. Aquiring specific registers in adolescent and adult life is only partly
an effect of formal learning; it is more often an effect of having acquired access to
certain communities and groups in society – from Metallica fans to computer
engineers in a telecom business, or from parents of young children to victims of a
car accident, or from Catholic priests to Chinese professional colleagues – and
having been exposed to the specific discourse patterns valid in such communities
and groups. Naturally, the internet has become a tremendously influential
provider for such informal learning environments over the past couple of
decades.

Evidently, this vast range of ways in which people come across linguistic and
sociolinguistic resources leads to an equally vast range of modes of learning. Let
us highlight just a few, aware that the vocabulary we must use for describing
certain phenomena lacks clarity and precision.

“Comprehensive” language learning

Full socialization across a lifetime in a language, including having access to the

formal learning environments for such language skills and resources as well as
having access to a wide range of informal learning environments will lead to a
“maximal” set of resources: different language varieties, different genres, styles
and registers, distributed over oral as well as literate modes of production and
reception, and dynamic in the sense that one is capable to rapidly learn new
forms and patterns – the gradual expansion and overhaul of one’s repertoire.

“Specialized” language learning

10
Particular stages of life come with access to specific and specialized skills and
resources. Becoming a university student, for instance, comes with access to
technical and specialized registers, genres and styles (e.g. the academic essay or
thesis), whose validity is entirely restricted to that part of life and that specific
environment. For people all over the world, becoming immersed in the academic
environment increasingly means that they learn such specialized skills and
resources in different varieties of academic English. Parts of any multilingual
repertoire, consequently, will often be “specialized” in the sense used here: one
can be fluent and articulate in academic genres and registers in English, but not
in the genres and registers of everyday life outside of academia (e.g. those valid
in supermarkets or in a medical doctor’s office).

Those two patterns of learning we would consider to be profound and enduring;

the second type usually is nested in the first one, as one specific pattern of
socialization encapsulated in more general patterns of socialization. They
account for what Hymes (1972b, 1992) understood by ‘communicative
competence’: the capacity to be a ‘full’ social being in the communities in which
one spends his/her life; the capacity for ‘voice’, i.e. to make oneself understood
by others in line with one’s own intensions, desires and ambitions, and this in a
wide range of social arenas (Hymes 1996). When we see people as ‘fully
integrated’ members of some group, it is because they have acquired such
elaborate forms of language skills and resources.

Apart from those elaborate patterns of learning, however, we need to consider a

number of others: more ephemeral and restricted ones. Let us turn to some such
patterns.

“Encounters” with language

In the context of globalization, people and linguistic resources are mobile;

consequently, one can come across particular bits of language, learn them in
particular ways, and use them. In contrast to the two previously mentioned
modes of learning, we are facing minimal modes of learning here: we learn very
small bits of language, not the elaborate sets of genres, styles and registers we

11
discussed above. Let us survey some of them; they may illustrate what is
undoubtedly a much broader range of ‘minimal’ forms of language learning.

-Age-group slang learning. In particular stages of life, people pick up

particular bits of language that typify and identify them as members of
age groups, professional groups and so on. Thus, most middle-aged
people still have a repertoire of ‘dirty words’ , obscenities and obscure
slang expressions learned during adolescence. Together they amount to a
whole discourse system, to be used in particular social arenas with peer
group members and an occasional outsider. While such complexes define
particular stages in life, they tend to become less frequently used in later
stages of life and ultimately live on as an obsolete, anachronistic discourse
system.

-Temporary language learning. People who frequently travel often learn

small bits of the local languages, sometimes sufficient to conduct very
short conversations within specific genres (e.g. ordering a meal in a
restaurant or saying that you don’t speak or understand the other’s
language), to perform more elaborate greeting rituals or engage in some
mimimal form of social bonding with local people. Often, such learned
skills and resources do not survive; they are gradually forgotten and
disappear from one’s repertoire. Yet they were learned and were part of
someone’s repertoires at some point in time.

-Single word learning. Many of us know single words from languages we

otherwise do not speak, write or understand. Isolated greeting formulae
from different languages would very often feature in the repertoire of
many people: ‘sayonara’ and ‘konnichi wa?’ from Japanese, ‘ni hao’ from
Chinese, ‘shalom’ from Hebrew, ‘salem aleikum’ from Arabic, ‘ciao’ from
Italian, ‘karibu’ from Swahili, and even ‘aloha’ from Hawai’an: they all
belong to a globalized vocabulary known to large numbers of people.
Similarly, terms related to the use of food or drinks (‘salud!’, ‘santé!’,
‘Gesundheit!’, ‘nazdrovje!’, ‘bon appétit’), expressions for yes or no (‘njet!’,
‘Jawohl!’) or curses and insults (‘cojones!’, ‘hijo de puta’, ‘cornuto’,

12
 ‘merde’, ‘asshole’, ‘sucker’, ‘Schweinhund’ etc.) are widely available
candidates for single-word learning. The point is that such terms are often
the only words we know in some language, but that they nevertheless
represent a minimal form of learning and a minimal form of knowledge. It
is not as if we don’t know these words.

-Recognizing language. There are many languages we do not actively use

or understand, but which we are nevertheless able to recognize and
identify, either on the basis of sound or on the basis of script. Thus, many
people in Western Europe would recognize Chinese, Arabic, Cyrillic and
Greek scripts, even if they are not able to read texts written in that script.
Some may even recognize Thai or Amharic script, and many would
recognize the particular visual image of Finnish and French in writing.
Similarly, people who live in immigrant neighborhoods may be able to tell
the language people are speaking, even if they don’t understand these
languages: these people are speaking Turkish, others Russian, others
German, others Arabic. Recognizing language is the effect of a learning
process – typically an informal one – and it results in the capacity to
identify people, social arenas and practices, even if one is not able to fully
participate in such practices. It is again a – minimal – form of language
knowledge which goes hand in hand with social knowledge. Recognizing
someone as a speaker of Turkish involves identifying that person as a
Turkish person, and it triggers a world of ideas and perceptions: ideas
about Turkish people, about their religion, culture and presence in a
particular place; insertion into widely circulating discourses on
multiculturalism, Islam, the wearing of the veil, and so forth. Recognizing
language is an important emblematic process in which language projects
social, cultural, ethnic and political categories and social and spatial
demarcations (recognizing Hebrew writing, for instance, can make one
realize that one has entered a Jewish neighborhood). Minimal knowledge
of language here connects to maximum knowledge of society.

The first two modes surveyed above are ‘transitory’ patterns of language
learning: bits of language(s) are learned but lose active, practical deployability

13
after some time. The two latter ones are usually not seen as ‘language learning’,
either because of the extremely small amounts of language learned, or because
no active competence in the language has been acquired. Yet in all of these cases,
such bits of language are part of our repertoires; they document moments or
periods in our lives when we encountered language(s). Encounters with
language account for the otherwise inexplicable fact that we often know more
‘languages’ than we would usually acknowledge or be aware of; that we
recognize sometimes very alien forms of language; that we achieve particular
small communicative routines without ever having been deeply immersed in the
language or having gone through an elaborate formal training and learning
process.

“Embedded” language learning

We sometimes learn bits of language that can only be used if another language is
used as well. Thus, there are forms of learning in which the finality of learning is
to perform code-switching in an appropriate way. Computer-related terminology
is often a case in point: all over the world, English vocabulary associated with the
use of computers would be used as an embedded vocabulary in discourses
conducted in other languages (Dutch IT engineers, consequently, would speak
Dutch with English vocabulary embedded). The school languages that are not
studied for achieving productive fluency in them– think of Latin and Greek, but
increasingly also German and French in Europe – would typically be languages
that only exist as embedded parts of instuctional discourses in another language.
A Dutch secondary school student learning Latin would use Latin only as part of
Dutch instructional discourses, consequently. One can also think of hobby
activities that involve exposure to other-language vocabulary: Yoga, Feng Shui,
Karate, but also Italian or Oriental cooking would produce discourses in one
language dotted by specific terms or expressions from another language. Thus
people practicing Japanese martial arts would go to the dojo for practice and
would listen to their sensei calling ‘mate!’ – even when that sensei is a full-
blooded Antwerp native who has no competence whatsoever in Japanese beyond
the specialised register of the sport s/he practices. Note that such specialized
embedded bits of language can be quite large, running into dozens if not

14
hundreds of expressions. These bits, however, do not make up a ‘language’ in the
sense of an autonomously functioning set of resources, they always need
scaffolding from another language.

The ‘minor’ forms of language learning typically occur in informal learning

environments: through everyday social contacts with others, traveling, media,
internet use, peer group memberships, exposure to popular culture, and so forth.
When such forms of learning coincide with formal learning programs, as with
‘school languages’, we see the emergence of different, specific registers across
the range of languages learned – ‘school languages’ become polycentric
sociolinguistic objects whenever they are ‘taken out’ of school and used to poke
fun at each other or to imitate teachers and stereotypical characters associated
with the language. This was the case with the ‘Deutsch’ Ben Rampton observed
in UK schools, where pupils used bits of school German to bark commands at
each other (Rampton 2006). An imagery of Second-World War Nazi stereotypes
was never far away, and the pupils drew on this rich indexical source by turning
school German into an emblematic resource for playful brutality and
oppresiveness. The same thing happens when language material from outside
school is ‘brought into’ schools and blended with the formally learned bits – as
when the formally learned RP accent in school English is replaced by a ‘cooler’
American accent in the schoolyard; or when a degree of competence in school
English is used as a platform to experiment with alternative forms of writing, as
in ‘boyz’ or ‘cu@4’ ; or when children in a Barbadian classroom get reprimanded
by their teacher for inserting Rasta slang into their speech (Van der Aa 2012).

Formally and informally learned language and literacy resources merge into
repertoires, and such repertoires reflect the polycentricity of the learning
environments in which the speaker dwells. The precise functions of such
resources can only be determined ethnographically, i.e. from within the group of
users, from below. Thus, as every parent knows, it is by no means a given that
the most normatively regimented varieties of languages – ‘correct’ school
varieties, in other words – carry most prestige and operate as a yardstick for
social interaction. The specific blend of different bits of language – the fusion of
grammatical correctness (acquired in a formal learning environment) with

15
fluency in an adolescent slang (derived from informal learning environments),
for instance – provides the actual resources deployed by people. Evidently, such
resources (or ‘features’, Jorgensen et al 2011) can be part of what is
conventionally defined as ‘one language’ – Dutch, English, German – but they
may also be derived from a variety of conventionally defined ‘languages’. The
repertoires of people absorb whatever comes their way as a useful – practical
and/or pleasant – resource, as long as such resources are accessible to them. The
complexity of polycentric learning environments (something that escalates as an
effect of the growing importance of new media, as mentioned earlier) ensures
that new ‘markets’ for linguistic resources become accessible: linguistic
resources that were until recently almost exclusively acessible through formal
education (e.g. normative varieties of English) now become available through a
multitude of other, often more democratically organized channels (see e.g.
Blommaert 2010, chapters 2 and 6; Block 2012).

This creates complex and layered repertoires; at the same time, it raises a wide
variety of issues regarding normativity and stratification in the social use of
language. While some resources (e.g. HipHop English) have become
democratically distributed resources, the normative varieties of English remain
accessible only through access to exclusive learning environments. This also
counts for literacy resources: whereas literacy historically was intimately tied to
access to formal schooling, we see that alternative literacies (such as ‘cu@4’) can
be easily and quickly learned through informal learning trajectories (Velghe
2011). This democratization of access to literacy resources has, however, not
removed the hierachy between ‘correct’ and ‘incorrect’ writing: it has highlighted
and emphasized it. The expansion of the modes of language learning has not
resulted in a more egalitarian field of language learning; it has led (and is
leading) to increased stratification and polycentricity.

4. Knowledge of language(s)

We have seen that repertoires are the result of polycentric learning experiences;
we have also seen that they involve a range of learning trajectories, from
maximally formal to extremely informal – in fact, that we often learn bits of

16
language(s) without being aware of it; and we have seen that they involve a
range of learning outcomes, from ‘full’ active and practical competence down to a
level where language(s) are just recognizable emblems of social categories and
spaces, a form of learning that does not require any active and practical
competence. All of those very different resources are part of our repertoires, and
all of them have or can acquire a multitude of functions.

Let us now turn to someone’s actual repertoire. For the sake of argument, we
shall discuss the repertoire of the first author of this paper. Pending the
development of a more accurate vocabulary, we shall be compelled to list
languages as named entities and to group oral and literacy skills. The
categorizations we will have to use in this exercise are necessarily clumsy and
inadequate; we hope to give an impression, though, of the diverse and layered
structure of a repertoire. We shall also describe the synchronic repertoire, i.e. the
resources that are active in our subject’s repertoire at present; past temporary
language resources will not be listed (our subject learned, e.g. particular bits of
several African languages in the course of his life, but cannot claim any active
competence in those languages now).

We shall proceed in three stages: first we shall list the different languages from
which particular resources have entered the repertoire, after which we shall
attempt to introduce distinctions in the actual skills and competences they
involve. Finally, we shall comment on the biographical basis of this repertoire.

4.1. Thirty-eight languages

Let us distinguish between four large categories of competence – the actual

practices and skills enabled by the resources we shall list.

a. The first level would be ‘maximum’ competence: oral as well as

literacy skills distributed over a variety of genres, registers and styles,
both productively (speaking and writing skills) and receptively
(understanding oral and written messages), and in formal as well as
informal social arenas. Resources from two languages qualify for
inclusion here: Dutch and English. Note that in both languages, our
subject would also be competent in at least some intra-language varieties.

17
In Dutch, several regional dialects and slang codes are known; and
English covers (at least receptive) competence in different kinds of
regional UK and US English, different international (‘world’) accents,
some Pidgin and Creole varieties of English, and specialized varieties such
as Rasta slang and HipHop slang.

b. The second level would be ‘partial’ competence: there are very well
developed skills, but they do not cover the broad span that characterized
the first category, of genres, registers, styles, production and reception,
and formal and informal social arenas. Thus, our subject can read
relatively complex texts, but not write similar texts; he can understand
most of the spoken varieties but not make himself understood in speaking
them; or he can use the language resources rather fluently as an
embedded language in another one. Six languages qualify for inclusion
here: French, German, Afrikaans, Spanish, Swahili and Latin. Knowledge
of intra-language varieties is minimal here (our subject would be able to
recognize various regional varieties of French but not of German, for
instance).

c. The next level is ‘minimal competence’: our subject can adequately

produce and/or understand a limited number of messages from certain
languages, confined to a very restricted range of genres and social
domains: shopping routines, basic conversational routines and stock
expressions. Eight languages qualify: Japanese, Chinese, Italian, Greek,
Finnish, Russian, Portuguese, Lingala.

d. Finally, there is ‘recognizing’ competence. Obviously our subject is

able to recognize all the languages listed in the three previous categories;
the fourth category, however, lists languages in which our subject has
only recognizing competence. The list is quite long: Turkish, Arabic,
Korean, Northern Sami, Gaelic, Berber, Polish, Albanian, Hungarian, Czech,
Serbo-Kroatian, Hebrew, Yiddish, Schwytsertüütsch, Xhosa, Zulu, Gikuyu,
Yoruba, Amharic, Thai, Tibetan, Tamil. We count twenty-two languages in
which our subject can recognize sounds and/or scripts.

18
We see that our subject’s repertoire combines resources from thirty-eight
languages. These resources are very unevenly distributed, as we know, and while
some resources allow him versatility and choice in a broad range of social
contexts, others offer him only the barest minima of access and uptake. All of
these resources – all of them – have their places and functions however, and all
of them reflect particular itineraries of learning during specific stages of life and
in particular places and learning environments. Let us have a look at these
functions.

4.2. Competence detailed

When we look at what our subject is really capable of doing with these resources,
the picture becomes extremely complex. If we divide the broad notion of
competence over a number of concrete parameters that reflect the capactity to
perform actual practices and the different social domains in which they can be
practiced, we notice that the resources of each language listed above are
differently distributed and functionally allocated within the repertoire.
Someone’s actual competence so becomes a patchwork of skills, some
overlapping and some complimentary, with lots of gaps between them. While
our subject obviously has a broad and diverse range of resources in his
repertoire, there is no point at which he can be said to be capable to perform
every possible act of language. Some of the resources offer a general and
multigeneric competence, while others are extremely specialized and only occur
in rigidly delineated contexts.

We will turn to the former in a moment; an example of the latter would be Latin,
listed above under ‘partial’ competence. Our subject can adequately deploy a
broad range of Latin linguistic resources (“his Latin is good”, as one says in
everyday parlance), but only and exclusively as an embedded language couched
in Dutch instructional discourse. The Latin he knows is his own old ‘school Latin’
– a specific register structured along lines of translation and grammatical
analysis – which is nowadays deployed only when he coaches and supervises his
children’s (and their friends’) learning of school Latin. It is not as if he does not
‘know’ Latin – the knowledge of Latin, however, is confined to a particular

19
generic space and tied to a very small range of communicative events
(‘explaining’ and ‘teaching’ Latin by means of Dutch instructional discourse).
Latin is a highly specialized resource in his repertoire, and is not used
autonomously but always in synergy with another language.

Let us now move to two other languages listed in the same category: French and
German. We will see that, compared to Latin, those two languages offer an
entirely different range of competences to our subject; we shall also see that
even between these two there are major differences in the distribution of actual
competences, which are an effect of the different trajectories by means of which
they entered our subject’s repertoire.

Let us first consider French.

FRENCH Spoken Spoken Written Written

production Reception production Reception

FORMAL Restricted: not Advanced: Absent: not Advanced:

able to give a capable of able to write able to read
formal speech or understanding formal genres most formal
lecture without most formal in French genres in
preparation and genres in French
scripting; partial French
access to courtesy
and politeness
norms; partial
access to
‘sophisticated’
registers

INFORMAL Advanced: Advanced: Average: able Advanced:

capable of having able to to write some able to read
conversations on understand informal texts most infomal
a wide range of most informal (e.g. email) messages in
topics in a spoken without French,

20
 vernacular messages in assistance including
variety of French French, some
including regional and
some regional slang
and slang varieties
varieties

And let us now compare this to German.

GERMAN Spoken Spoken Written Written

production Reception production Reception

FORMAL Absent: not able Average: able Absent: not Advanced:

to produce formal to understand able to able to read
speech most formal produce most formal
speech genres formal text genres
in German written text

INFORMAL Very restricted: Average: able Absent: not Average: able

only simple to understand able to to read most
routines and most spoken produce informal
responses . Standard informal Standard
varieties of written text varieties of
German. text.

While both languages were listed under ‘partial’ competence above, we now see
that the actual ‘parts’ in which our subject has real competence differ
substantially. Our subject has hardly any real competence in the production of
spoken and written German; while he has some competences in the production
of French. Note, however, that (a) these productive competences in French are
by and large confined to informal domains, and (b) that his productive
competence in spoken French is restricted to the use of a vernacular variety –

21
whenever he speaks French, he speaks a distinct Belgian variety of it, influenced
by the Brussels dialect as well as by a Femish-Dutch accent. Notwithstanding
these restrictions, it is not unlikely that French interlocutors who encounter our
subject informally and have a chat with him may find him relatively fluent in
French. This fluency is generically and sociolinguistically restricted – it is a
‘truncated’ competence (Blommaert et al 2005; Blommaert 2010 chapter 4).
That means that this competence is not generative: fluency in these informal
conversations does not automatically imply fluency in other genres and social
domains; competence in one sociolinguistic area does not imply fluency in any
other area, nor can it a priori be seen as an engine for acquiring such fluency.
Competences are as a rule sociolinguistically specific (a point very often
overlooked by language teachers). They cluster around particular social arenas
and become generative in those arenas (a process called ‘enregisterment’: Agha
2007; Silverstein 2004), but have no automatic applicability outside of them.

Apart from these important differences, we notice similarities. Receptive

competences of our subject are present in both French and German, even if the
receptive competences in French are more advanced than those in German. Our
subject can thus perform with relative adequacy the roles of listener and reader
in both languages, even if listening to vernacular varieties of German can be
challenging. In actual interaction events, this unevenly distributed competence –
receptive competence without productive competence – can give rise to various
kinds of surprises, misjudgments and misunderstandings, as when German
interlocutors are surprised that a very well understood German question is
answered in English, not German; or when Francophone colleagues assume that
our subject can adequately lecture in French because they have unproblematic
informal conversations with him (or, even worse, believe that his conversational
fluency indicates that he can write academic papers in French).

A full and comprehensive survey of what our subject can actually do with his
repertoire would of course require an analysis of every particular resource in his
repertoire – an excercise we cannot contemplate in this paper. The point should
be clear however: all the elements that together compose the repertoire are
functionally organized, and no two resources will have the same range and

22
potential. A repertoire is composed of a myriad of different communicative tools,
with different degrees of functional specialization. No single resource is a
communicative panacea; none is useless.

4.3. Repertoires as indexical biographies

How did these different resources enter into our subject’s repertoire? Let us
have a look at the very different trajectories we have to review here.

-Vernacular Dutch is our subject’s first language – his ‘mother tongue’ or

‘L1’ as it is usually called. His first speaking skills were gathered through
common socialization processes, and they were composed of a local
dialect. This dialect stayed with him for the remainder of his life, even
though the communicative network within which he could deploy it
shrunk dramatically in the course of his life. His family moved to Brussels
when he was 11; the initial social world of dialect was replaced by
another one, now dominated by a vernacular variety of Standard Dutch
with a distinct Brussels regional influence. These dialect backgrounds
account for the distinct accent he has when speaking Standard Dutch (and
every other language, for that matter). Currently, dialect is exclusively
used in a tiny family network, and only in informal domains. The dialect
never developed into adult repertoires nor into specialized professional
repertoires; consequently the range of social roles which our subject can
assume through that dialect is very limited.

-Note that the L1 was a dialect (or a complex of dialects); Standard Dutch
as well as French, German and English, but also Latin and Greek were
school languages. The fact that they were school languages accounts for
the fact that some – Latin and Greek – never really transcended the level
of school competences: the capacity to translate a fixed body of texts and
to perform indepth grammatical analysis of them; accompanied in the
initial stages of formal learning by a modest capacity to speak and write
French, German and English and a well developed capacity to read formal
texts in those languages. Swahili was the language in which our subject
specialized during his student years. It is in a sense also a school

23
language: he acquired the school competences mentioned earlier and a
modest productive and receptive competence in formal Standard Swahili.
Part of the training he followed also included an introduction into Arabic
and Yoruba, the results of which were later shrunk to the ‘recognizing
language’ level.

-Some of these school languages, however, acquired a life after and

outside school in complementary informal learning environments. Growing
up in Brussels as a teenager meant that our subject picked up local
vernacular and informal varieties of French. This accounts for his present
conversational fluency in informal domains. Our subject, however, never
found himself in formal social domains where French was the code, so
that part of competence never developed fully. During his student years,
texts in English, French and German belonged to the mandatory readings
in African Studies, as well as a modest amount of texts in Spanish and
Portuguese. This accounts for the fact that reading formal texts poses
little problems in English, French and German. And finally, advanced
studies made our subject enter the world of academic English, which
became the code for formal speaking and writing in the academic field, as
well as for a certain amount of informal social skills. These competences
are consequently highly developed. Swahili, finally, broadened and
deepened as our subject further specialized in that language, made
numerous fieldwork trips documenting urban vernaculars, and eventually
did some language teaching in Swahili.

-Our subject learned several languages in a purely informal learning

environment. Bits of Spanish were learned by attempts to read Neruda’s
poetry, later complemented by reading some academic works in Spanish;
bits of Italian through an interest in Italian cinema and mediated by
competences in Latin and French; bits of Russian through reading a Teach
Yourself booklet; some contemporary Greek mediated through the
Ancient Greek learned at school; Lingala by social contacts with
Congolese friends and colleagues; Finnish by a two-year visiting
appointment in Finland; Afrikaans by frequent contacts with South

24
 African colleagues and by fieldwork in an Afrikaans-dominant area;
isiXhosa and Northern Sami also through fieldwork exposure.

-Traveling was a major source of new language material, and almost all of
the languages listed above were at some point or another also languages
of the traveling destinations of our subject. Japanese and Chinese entered
the repertoire exclusively through traveling, later complemented by
personal contacts with friends and colleagues. The recognizing
competence for languages such as Tibetan, Serbo-Croatian and
Schwytsertüütsch is also an effect of traveling.

-Life in the neighborhood, finally, is the origin for much of what is listed
under ‘recognizing competence’. Our subject lives in a super-diverse
inner-city neighborhood, where e.g. Turkish, Arabic, Berber, Polish,
Russian, Albanian, Thai, Czech, Tamil, Hebrew and Yiddisch are frequently
used and publicly displayed. The lingua franca of the neighborhood is a
‘truncated’ form of vernacular Dutch; hence the superficial competence in
the languages of the local immigrants: they are a social and cultural
compass that guides our subject in identifying interlocutors in his
neighborhood.

We can see how the particular synchronic competences we reviewed in the

previous section have their historical roots in the distinct ways in which they
arrived to him or in which he arrived to them – the roots are routes, so to speak.
Each of the resources was learned in the context of specific life spans, in specific
social arenas, with specific tasks, needs and objectives defined, and with specific
interlocutors. This is why our subject can seem very fluent when he speaks or
writes on academic topics in English, while he can be extremely inarticulate
when he has to visit a medical doctor, a lawyer, grocer or a plumber in the UK or
the US. It is also why he can chat in vernacular French but not lecture in it, why
he can read German but not write it, and distinguish between Turkish and
Yiddisch without understanding a word of either language. And of course, this is
why certain resources did not survive in the repertoire. Our subject had to
devote a considerable amount of time studying Tshiluba as a student; not a

25
fragment of that language survived in the repertoire. The course was entirely
unexciting, the exam requirements undemanding, and the opportunities to
practice the language nil, the more since he and his fellow students discovered
humiliatingly that no Congolese actually spoke the kind of Tshiluba their 1950s
missionary-authored textbook offered them.

Each of these trajectories – all of them unique – contribute more than just
linguistic material to one’s repertoire. They contribute the potential to perform
certain social roles, inhabit certain identities, be seen in a particular way by
others (e.g. an articulate or inarticulate person, as in the example of informal
versus formal French), and so on. The resources that enter into a repertoire are
indexical resources, language materials that enable us to produce more than just
linguistic meaning but to produce images of ourself, pointing interlocutors
towards the frames in which we want our meanings to be put. Repertoires are
thus indexical biographies, and analyzing repertoires amounts to analyzing the
social and cultural itineraries followed by people, how they manoeuvered and
navigated them, and how they placed themselves into the various social arenas
they inhabited or visited in their lives.

5. Late-Modern repertoires and subjects

Let us by way of conclusion recapitulate what we intended to achieve in this

paper. We set out to describe patterns of learning “the means of language”, taken
here in their broadest sense as every bit of language we accumulate and can
deploy at a given point in life. Such patterns, we argued, are widely different in
nature and in ‘technology’, ranging from highly formal modes of patterned
learning to highly informal and ephemeral ‘encounters’ with language. These
different modes of learning lead to different forms of knowledge of language, and
while the diversity of such modes of language is tremendous, we must accept
that all of them matter for the people who have learned them. None of them is
trivial or unimportant. Even more, we can see how a subject consituted him- or
herself by analyzing the indexical biographies that are contained in the spectre of
language resources they can deploy.

26
The relevance of the latter point should be clear. While earlier authors on
repertoire emphasized the connection between (socio-)linguistic resources,
knowledge and communities, we shift the direction from communities towards
individual subjects. We have explained the rationale for that in section 2 above:
super-diversity compels us to abandon any preconceived and presumed stable
or absolute notion of community, and replace them with a more fluid view of
networks, knowledge communities and communities of practice – all of them
dynamic, in the sense that most of them are peculiar to particular stages of life,
and those that persist through life (as e.g. the family or regional forms of
memberships) change in shape and value during one’s lifetime. Repertoires in a
super-diverse world are records of mobility: of movement of people, language
resources, social arenas, technologies of learning and learning environments. A
relevant concept of repertoires needs to account for these patterns of mobility,
for these patterns construct and constitute contemporary Late-Modern subjects.

‘Community’ is not the only notion we have to revisit; the same counts for
‘language’. We have repeatedly flagged the uneasiness of our own vocabulary in
describing the repertoires of contemporary subjects; the fact is that we all carry
the legacy of modernist hegemonies of language and society, and that an
important part of our task consists of redesigning the analytical instruments by
means of which we proceed. If we look back at our subject’s repertoire, we have
seen that no less that thirty-eight languages are represented there. Yet, of course,
none of these languages is in any realistic sense ‘complete’ or ‘finished’: all of
them are partial, ‘truncated’, specialized to differing degrees, and above all
dynamic. This also counts for the ‘mother tongue’, that mythical finished-state
language spoken by the ‘native speaker’ of language-learning literature. The
Dutch now spoken by our subject is different from the Dutch he spoke at the age
of eight or of eighteen, not just linguistically but also sociolinguistically. He still
occasionally uses his dialect, but since this dialect lost its broad social scope of
application due to migration at the age of eleven, it never developed any of the
registers of adult life: the register of relationships and sexuality, of parenthood,
of money, death, cars and work. Whenever our subject speaks his dialect, he can
only speak it from within two social roles: that of the son of his mother and the

27
brother of his sisters. He can no longer use it adequately during infrequent
encounters with childhood friends or relatives – the dialect does not allow him
the voice he wants and needs in that stage of life and that social arena. The
repertoires change all the time, because they follow and document the
biographies of the ones who uses them. In that sense, repertoires are the real
‘language’ we have and can deploy in social life: biographically assembled
patchworks of functionally distributed communicative resources.

As for our subject: the thirty-eight languages he has assembled throughout his
life may put him on the high side in terms of scope of repertoire. His life is that of
a mobile subject, someone who travels extensively and whose ‘basis’ – the
locality where most of his life is organized – is itself deeply colored by globalized
mobility. While he may be seen as an exception, we may as well see his
repertoire as unique – a unique reflex of a unique biography. But when similar
exercises would be applied to other subjects, surprising results could be
obtained even among biographically more ‘average’ subjects. We tend to
underestimate the degree to which our lives develop along trajectories of
mobility, in which we encounter, leave, learn and unlearn social and cultural
forms of knowledge (such as languages) because we need to be able to make
sense of ourselves. In that sense, we can see ‘structure’, or at least ‘pattern’ in
repertoires that are otherwise entirely unique. The structures and patterns are
dynamic and adaptable, while they are driven by shared motives and intentions:
to make sense, to have voice wherever we are.

There is an angle to this that merits exploration. Voice, as we know, is subject to

normative judgment – one has voice when someone else ratifies it as such. In
that sense, our subject’s repertoire is a complex of traces of power: a collection of
resources our subject had to accumulate and learn in order to make sense to
others, that is, in order to operate within the norms and expectations that govern
social life in the many niches in which he dwelled and through which he passed.
The elements of the repertoire are resources he needed to deploy, practices he
had to perform, in order to be ‘normal’ in the polycentric and dynamic world in
which he lived. We have here a very Foucaultian view of the subject: the subject

28
as an outcome of power, as a complex of features of self-disciplining, as a subject
perpetually subjected to regimes of normality

Thus conceived, repertoires invite a new form of analysis. No longer seen as the
static, synchronic property of a ‘speech community’, we can now approach it as
an inroad into Late-Modern subjectivities – the subjectivities of people whose
membership of social categories is dynamic, changeable and negotiable, and
whose membership is at any time always a membership-by-degree. Repertoires
enable us to document in great detail the trajectories followed by people
throughout their lives: the opportunities, constraints and inequalities they were
facing, the learning environments they had access to (and those they did not
have acess to), their movement across physical and social space, their potential
for voice in particular social arenas. We can now do all of this in significant
detail, because we are no longer trapped by a priori conceptions of language,
knowledge and community.

Or are we? We noted in our introduction the increasing predominance of purely

modernist technologies of language ‘measurement’ through uniform testing.
Such practices have become a central element of administrative and bureaucratic
apparatuses all over the world, and they operate with exceptional power in fields
such as education, labor and migration. The Common European Framework for
Languages has in a very short time become an industry standard for measuring
language competence far beyond Europe, and it is applied as an ‘objective’ tool
for measuring progress in language learning, the benchmarking and
accreditation of language experts such as teachers and interpreters, the
‘readiness to integrate’ of new immigrants as well as the ‘degree of integration’
of recent residents.

We do not believe that we have to engage in a lengthy comparison and critique of

the assumptions underlying such standardized language measuring tools; we
believe our critique of them should be clear from the way we addressed
repertoires here. The conclusion of our critique is therefore obvious: such
measuring instruments are a form of science fiction. They have only a distant and
partial connection with (specific parts of) the real competences of people, the

29
way they are organized in actual repertoires, and the real possibilities they offer
for communication. If we apply the Common European Framework levels for
language proficiency, our subject would undoubtedly score a C2 – the most
advanced level of proficiency – for English, when the language test concentrates
on academic genres of text and talk. The same subject, however, would score A2
– the most elementary level of proficiency – if the test were based on how he
would interact with a medical doctor, a plumber, an IT helpdesk operative, an
insurance broker, and so on. So, ‘how good is his English’ then? Let it be clear
that this question can only be appropriately answered with another one: ‘which
English?’

REFERENCES

Agha, Asif
2007 Language and Social Relations. Cambridge: Cambridge University Press.
Block, David
2012 Economising globalisation in Applied Linguistics in neoliberal times. In David
Block, John Gray & Marnie Holborow (eds.) Neoliberalism and Applied
Linguistics: 56-85. London: Routledge.
Blommaert, Jan
2005 Discourse: A Critical Introduction. Cambridge: Cambridge University Press.
2010 The Sociolinguistics of Globalization. Cambridge: Cambridge University Press.
Blommaert, Jan & Ad Backus
2011 Repertoires revisited: ‘Knowing language’ in superdiversity. Working Papers
in Urban Language and Literacies, paper 67.
Blommaert, Jan & Ben Rampton
2011 Language and superdiversity. Diversities 13/2: 3-21.
Blommaert, Jan, James Collins and Stef Slembrouck
2005 Spaces of multilingualism. Language and Communication 25: 197-216.
Collins, James and Richard Blot
2003 Literacy and Literacies. Cambridge: Cambridge University Press.
Creese, Angela & Adrian Blackledge

30
2010 Translanguaging in the bilingual classroom: A pedagogy for learning and
teaching? Modern Language Journal 94: 103-115.
Gumperz, John
1972 (1986) Introduction. In John Gumperz and Dell Hymes (eds.) Directions in
Sociolinguistics: The Ethnography of Communication: 1-25. London: Blackwell
1982 Discourse Strategies. Cambridge: Cambridge University Press.
Gumperz, John and Dell Hymes (eds.) 26
1972 (1986) Directions in Sociolinguistics: The Ethnography of Communication.
London: Blackwell
Hogan-Brun, Gabrielle, Clare Mar-Molinero and Patrick Stevenson (eds)
2009 Discourses on Language and Integration. Amsterdam: John Benjamins.
Hymes, Dell
1968 Linguistic problems in defining the concept of ‘tribe’. In J. Helm (ed.) Essays
on the Problem of Tribe: 23-48. Seattle: University of Washington Press.
1972a (1986) Models of the interaction of language and social life. In John Gumperz
and Dell Hymes (eds.) Directions in Sociolinguistics: The Ethnography of
Communication: 35-71. London: Blackwell.
1972b On communicative competence. In J.B. Pride and Janet Holmes (eds.)
Sociolinguistics: 269-293. London: Penguin.
1974 Ways of speaking. In Richard Bauman and Joel Sherzer (eds.) Explorations in
the Ethnography of Speaking: 433-451. Cambridge: Cambridge University
Press.
1992 The concept of communicative competence revisited. In Martin Pütz (ed.)
Thirty Years of Linguistic Evolution: 31-57. Amsterdam: John Benjamins.
1996 Ethnography, Linguistics, Narrative Inequality: Toard an Understanding of
Voice. London: Taylor & Francis.
Jörgensen, Jens-Normann, Martha Karrebaek, Lian Madsen & Janus Möller
2011 Polylanguaging in superdiversity. Diversities 13/2: 23-37.
Juffermans, Kasper
2010 Local Languaging: Literacy products and practices in Gambian society. PhD
Dissertation, Tilburg University.
Kress, Gunther

31
 1977 Before Writing: Rethinking the Paths to Literacy. London: Routledge.
Kroon, Sjaak, Dong Jie & Jan Blommaert
2011 Truly Moving Texts. Tilburg Papers in Culture Studies, paper 3.
Otsuji, Emi & Alastair Pennycook
2010 Metrolingualism: Fixity, fluidity and language in flux. International Journal of
Multilingualism 7/3: 240-253.
Rampton, Ben
1998 Speech community. In Jef Verschueren, Jan-Ola Östman and Jan Blommaert
(eds.) Handbook of Pragmatics 1998: 1-32. Amsterdam: John Benjamins.
2006 Language in Late Modernity. Cambridge: Cambridge University Press.
Silverstein, Michael
2004 Cultural concepts and the language-culture nexus. Current Anthropology 45:
621-652.
Spotti, Massimiliano
2011 Ideologies of success for the superdiverse citizen: The Dutch testing regime
for integration and the online private sector. Diversities 13/2: 39-52.
Van der Aa, Jef
2012 Monitoring narrative collaboration in a Caribbean classroom’. PhD
dissertation, University of Jyväskylä.
Varis, Piia & Xuan Wang
2011 Superdiversity on the Internet: A case from China. Diversities 13/2: 71-83.
Velghe, Fie
2011 Lessons in textspeak from Sexy Chick: Supervernacular literacy in South
African instant and text messaging. Tilburg Papers in Culture Studies, paper 1.
Vertovec, Steven
2007 Super-diversity and its implications. Ethnic and Racial Studies 30: 1024-1054.

32

View publication stats