Semilingualism, Double Monolingualism and Blurred Genres - On (Not) Speaking
a Legitimate Language(1)
"When discussions of educational treatments for children point to a linguistic
deficit, we as researchers are obliged to thoroughly consider the ways in which
the institutional effects of our labels may contribute more to the malady than
to the proposed remedy of the learners."
(Jeff MacSwan 2000, 37 f.)
"For humans there exists neither a complete control of language nor a
completely homogeneous speech community. Never and nowhere will we find a perfect,
homogeneous monosystem; always and everywhere we will just find imperfect heterogeneous
polysystems. The relationship of humans towards their language is not one of
perfect monolingualism, but just the opposite, it is one of imperfect polylingualism
and one of polylingual imperfection."
(Mario Wandruszka 1979, 313; translation mine, V.H.)
1. Introduction: bilingualisms and other blurred complexities
Whenever I do advanced teacher training on intercultural and multilingual topics,
a - often heated - discussion on German language proficiency of migrant children
and adolescents regularly turns up. One of the keywords continuously dropped
in these discussions, is the term 'semilingualism'. One can hear statements
like "These children still can't speak German properly, I don't understand
that. And they mix languages. German, interspersed with Turkish and the like!
And when they can speak it on a quite reasonable level, they often can't write
properly. They are real semilingual." And with more information about the
kids' native languages some specialists conclude that if their mothertongues
are likewise interspersed with German, as well, these children even suffer from
'double semilingualism'.
Of course, what it is all about here, is a particular kind of bilingualism,
even if divided in half. Bilingualism is manifold: For some it is when a speaker
has a native like command in both languages; for others it is any kind of using
two languages side by side. Whatever holds true, bilingualism covers quite a
whole range of forms of competences and usages of two languages. And whatever
is meant by native like command, we have certainly no tertium comparationis
for that.
Bilingualism sounds neutral, as if really two languages were peacefully kept
under one roof, be it individually or societally. It is obvious, however, that
the usage of two and more languages will be at least complementary, that is,
e.g., an academic knowledge in one language will not automatically be reciprocated
by the same knowledge in the other language. And even further away from this,
we find languages, side by side, weighing differently, comprising different
legitimations of usage and support. For some, bilingualism is hence solely understood
as mastering the right language at the right time in the right place. This implies
that the usage of these languages must be kept apart, neatly separated according
to clear-cut domains such as school, family and peer group. No blurring of domains!
No semilingualism, no mixing. The real ideal is a double monolingual speaker(2).
Bilingualism, as it develops in Germany and other societies with a migratory
background of multilingual development does not at all bow to these norms and
ideals. What are created are blurred genres of bi- and multilingualism in every
respect - the typical (post)modern complexity.
This article deals with school failure of migrant pupils and their alleged
'semilingualism', their use of particular blurred forms of bilingualism (or
rather multilingualism - because it will turn out that more than just two languages
will be at issue), and their non-use of the legitimate forms of language. Obviously,
there is a correlation to be found between school achievements, command of languages
and the legitimate use of language. Furthermore, there is an invisible connection
between the German teachers' perception, the public discourse on migration and
integration, the socioeconomic givens of a migration society like Germany, and,
in particular, the role German as a Second Language plays therein.
I will thus begin with stating the already known on this issue, the macro-framework,
so-to-speak, and will then continue with taking a brief look at the historical
and ideological trajectory of that notorious term 'semilingualism' and what
it stands for today. The bulk of my paper, however, will be dedicated to the
ethnographic micro-perspective, based on empirical data from authentic and situated
encounters among migrant adolescent of Turkish descent - adolescents who exactly
feed the frustrations of those teachers quoted in the first paragraph and who
fall under the label of being semilingual. The data will give some counterevidence
to the teachers' perceived language deficits of their pupils, and will hopefully
help to discredit the notorious usage of the semilingualism-concept. But the
data will on the other hand frustrate all those who expect 'proper' bilinguals
instead of semilinguals. What they will find is blurred genres in a twofold
way: Firstly, bilingualism as intertwined usage, blurring perceived borderlines
between languages; and secondly, a new and blurred understanding of bilingualism
not only including all languages used, but also language creativity in the bi-
and multilingual communities of German society, not caring about any consideration
as to its legitimate role. What follows from this is certainly not an 'anything
goes'-plea for the role of languages at school, but rather a recognition and
re-evaluation of the multifoldness, richness and creative potential of migrant
language use. "The handling of German officials with the languages of migrant
people bears traces of destructing capital", as Gogolin states (Gogolin
2001, 11). And this should be voiced clearly from the very beginning: Recognition
and appreciation of the non-legitimate languages does in no way replace the
need for language promotion in school and beyond school. And promotion and encouragement
surely pertains to all languages and even to those users not going to school
anymore(3).
2. Migrants' achievements: state of disaster
When talking about immigrants, about social and linguistic integration into
society and the role languages play therein, we have to look at the structural
givens in a society like Germany with respect to existing migrant communities(4).
Exact figures and facts about the whole migrant population are difficult to
get. What is statistically collected are numbers of residents without a German
passport, foreigners (Ausländer). In fact, the majority of these foreigners
have somehow a migratory background, primarily in labour migration, but also
as refugees.
The population segment of Turkish descent is close to 2 million out of the
roughly 7.5 million people in Germany without a German passport. But many quasi-immigrants
not figuring in these statistics have an Aussiedler-background (resettlers).
These include once emigrated and deported Germans to Eastern Europe and ex-Soviet
Central Asia which have been re-migrating to Germany. They are Germans by passport
but because of their former environment, education, cross-marriages etc. in
particular the youth is much closer to being 'foreigners' than, e.g., young
Turks in Germany, of which the vast majority below 18 has been socialized in
Germany(5). Thus being migrant means having a migrant background;
it does, however, not correlate with ethnicity. Indeed, it rather comprises
categories like German Turks, Italogermans, and Russian Germans - but whatever
hyphenated ethno-labels we attach, none will realistically reflect the migratory
trajectory of the particular group we talk about.
The majority of migrant children and adolescents perform poorly in school.
E.g., pupils of Turkish descent leaving school without a qualified graduation
beat Germans by more than 100 percent; and there are more than three times as
many pupils of Turkish background as compared to German pupils who do not receive
a vocational training even after finishing school successfully. At the same
time, migrant childeren are clearly underrepresented in all types of higher
education. In percentage there are four times as many German girls with grammar
school degree as there are Turkish girls, and the relationship of seven to one
between German and Turkish boys is even more dramatic.
Thus, even after four decades of migration there is still no decreasing divergence
between the demographic weight of migrants and their representation within the
higher layers of the educational system. One could sadly summarize that the
higher the educational level, the less graduates from migrant background; and
vice versa, the less qualified, the more second and third generation migrants
we will find there.
Also after leaving school the present and future perspectives on the labour
market remain alarming. The unemployment rate among the non-German population
in the old Bundesländer (federal states, West Germany) is twice as high
as that among the German population. The correlation between the low educational
achievements of migrants and their position within the labour market is obvious.
It's the result of a perpetual vicious circle of structural discrimination,
a tight labour market, and low proficiency in education which leads to a petrification
of a low socioeconomic status. School does not compensate for these deficiencies
but rather perpetuate them. And some experts have clearly called it what it
is: Bildungskatastrophe, a disaster and scandal in education(6).
This brief survey may help to illustrate the highly critical situation. Of
course, there have been profiteers as well out of this situation. Over the years
a whole industry of migrant helpers and analysts have emerged. Hundreds of institutions,
foundations, study programs and inquiries have been initiated to ameliorate
the situation, to make proposals to the better and to qualify the specialized
professionals to aid a migrant clientele. One of the loudest voices within this
concerted enterprise was and still is that of integration specialists in the
realm of German as a Second Language (Deutsch als Zweitsprache, DaZ), a financially
and ideologically flourishing branch of integration industries, with many qualified
and underqualified and constantly underpaid helpers on temporary terms of employment.
Although German as a Second Language has been taught to migrant pupils for
the past thirty years without a serious change to the better in the socioeconomic
and educational status of migrants, myriads of voices keep on postulating proper
and good command and knowledge of the German language as the indispensable precondition
('unabdingbare Voraussetzung') for social integration into German society.
However, I do not know of any study that has proven that the fulfilment of
this indispensable precondition has yielded any better results in terms of socioeconomic
and educational status. I certainly do not plead for neglecting this qualification
(cf. footnote 1), I just doubt its correlation with respect
to socioeconomic success. More input in terms of highly qualified expertise
is certainly needed. But it's neither enough nor legitimate to take an individual
capacity like a particular command of a language as the only criterion for such
a complex matter as integration. This is, however, exactly what transpires from
the notorious discussion of proper German proficiency as 'indispensable precondition'
of integration. Furthermore, waiting for the fulfilment of the 'indispensable
precondition' legitimates the postponement of any other positive action and
returns the ball to the 'victims' rather than to the responsible authorities.
Some studies have shown how mechanisms of unintended structural discrimination
have a much stronger effect on the educational selection process than qualifications
in German as a Second Language (cf. Bommes, Radtke 1993;
Gomolla, Radtke 2002). As a matter of fact, command of
a language certainly puts the focus of responsibility into the individual, outside
of social-psychological and socioeconomic structures. It's thus the individual
alone who is blamed for failure.
Attributing problems to the individual which are mainly structural problems
is notorious of deficit theory. Deficits must be compensated for. If the deficit
is due to a linguistic lack or deficiency in the individual then the locus of
change is exactly there. Deficit theory, of course, has had and still has many
proponents in the field of pedagogics and sociolinguistics (Dittmar
1976). Already in the sixties the US-American sociolinguist William Labov
inquired into the allegedly deficient English as spoken by nonstandard speakers
of Afro-American descent in the inner cities ('Black English') to argue against
the deficit view. His findings on the "Logic of Nonstandard English"
opened up a new agenda of research (cf. Labov 1972).
Hundreds of studies on minority languages, ethnolects, contact languages and
bilingualism followed in the wake of this research, leaving its initiators far
behind. An urban ethnography of communication inquired into the use of languages
and varieties within ethnic minority and migrant groups and diaspora communities
and showed that these languages and varieties and forms of communications were
in many ways not deficient in comparison to the prescribed forms of school and
majority languages (cf. e.g. Durán 1981; Sebba
1993; Rampton 1995). They were just different in
form and expression, covering different functions and were closely related to
the life world of the respective communities.
Yet, the educational debate over the language problems and language needs of
migrant and minority pupils remained prescriptivist in the sense that one or
another language or variety of language had an inherently higher value than
others and that it ought to be imposed on the whole of the speech community
to maintain standards of communication (cf. Crystal 1986,
2) - at least so in the prevailing public discourse of policy makers.
3. Semilingualism - an ideological construct of non-explanation
The pedagogical and linguistic debate about language and migrants was eagerly
received in public when it started to focus on deficient codes of speaking such
as 'Ausländerdeutsch' (foreigners' German), 'Gastarbeiterdeutsch' (guestworkers'
German) or 'Pidgin-Deutsch' (pidginized German) and the like (cf. Hinnenkamp
1990), giving ample evidence that the language issue was at the heart of
the matter, that is for societal integration and success. Imported from the
Scandinavian debate the term 'semilingualism' (Swedish 'halvspråkighet';
German 'Halbsprachigkeit') began its triumphal procession as an allegedly apt
and exhaustive notion in the public discourse on migrants' linguistic enactments
of the host language. Interestingly enough this was the first '-ism' to be coined
in the migrants' language debate, long before, e.g., 'bi-' and 'multilingualism'
came to the public mind. And when the professional and lay public also took
the existence of a first language ('mothertongue') into consideration, again
it was not bilingualism that was focused upon, but the 'bilingual' splitting
or doubling of semi-ness into 'double semilingualism' (German 'doppelseitige
Halbsprachigkeit').
Although 'semilingualism' became such a popular notion, nobody really knew
what it meant, even less so what it implied. A minimalist definition comprised
something like "having only partial knowledge or partial understanding
of the language, or of the two languages, in question; lacking mastery of either".
As a result of language testing, it was argued that semilingualism was expressed
through a limited vocabulary, an incorrect grammar, and difficulty with expressing
abstract concepts. However, its first function was not to explain anything,
but to explain things away. It meant a diagnosis of a linguistic impairment,
including its case history and its locus of therapy. Hardly anything of its
implications had to be spelled out. It became one of these notions with an atmosphere
of academic and lay consensus, a fuzzy concept with many ingredients of deficiency,
but as well useful for an appeal to responsibility to do something against its
detrimental consequences. Thus, semilingualism and double semilingualism became
also a kind of weapon, a political weapon to put pressure on those institutions
and authorities of society which left children in this state of low proficiency
and linguistic 'inbetweenness'.
Of course semilingualism did not stand alone. The concept found itself in good
company with other terms functioning as explain-away-concepts such as 'split
identity' or 'cultural diremption' - one feels a puff of Durkheimian anomy.
If we take a look at the history of the semilingualism-concept, we find it
first mentioned by the American linguist Leonard Bloomfield. His description
of the language of the North American Indian White Thunder was often quoted:
"White Thunder, a man around 40, speaks less English than Menomini, and
that is a strong indictment, for his Menomini is atrocious. His vocabulary is
small, his inflections are often barbarous, he constructs sentences of a few
threadbare models. He may be said to speak no language tolerably" (Bloomfield
1927, 437)
The first to advance the study between ethnic minority-group speakers and semilingualism
was Hansegård (1968). His study was based on comparisons of the linguistic
performances of Finnish immigrant children living in Sweden and Swedish monolingual
children. The immigrant childern's ability in both languages showed signs of
considerable retardation. For Hansegård the term denoted a lack of competence
in the language of an individual in any of the following areas: (a) the size
of the repertoire of words and phrases that are understood or actively available
in speech; (b) linguistic correctness; (c) degree of automatism; (d) the ability
to create or neologize; (e) mastery of the cognitive, emotive, and volitional
function of language; (f) a richness or poorness in individual meanings (whether
reading or listening to a particular linguistic system "evokes lively and
reverberating semantic images") (Hansegård 1968, as cited in Skutnabb-Kangas
1981, 253). Furthermore he claimed that poor performance in these features
would become permanent among the immigrants, thus leading to social stigmatization
and a life-long handicap to the psychological, social and moral development
of the bilingual.
This classification, however, seems quite arbitrarily chosen. What is the scope
for 'correctness' (b)? How to measure (c)? - By speed of production? According
to which criteria are the language functions in (e) the only relevant ones?
Where does 'mastery' begin? How to find out about "lively and reverberating
semantic images" (f)? Isn't all that highly suspicious of prescriptivism?
When we look at some of the data in Chapter 4, we will find that none of these
criteria will be found. Yet, according to their poor school proficiency, those
kids are regarded as candidates for the 'semilingualism'-label (see below).
The academic popularity of the semilingualism-thesis is however not only due
to the Scandinavian Hansegård, but mainly to the work of the Canadian
bilingualism-researcher Jim Cummins. Cummins' became one of the main defenders
of the 'semilingualism'-hypothesis. The theoretical foundation of the concept
was based on his "Threshold Hypothesis" (Cummins
1976). This hypothesis claimed that the level of linguistic competence attained
by a bilingual child in its first and second language may affect his or her
cognitive growth in other domains. Cummins believed that there were two thresholds
of language learning and that attainment beyond the lower threshold "would
be sufficient to avoid retardation, but the attainment of a second, higher level
of bilingual competence might be necessary to lead to accelerated cognitive
growth" (Cummins 1976, 24). For him, children with
low levels of proficiency in both their first language (L1) and second language
(L2) may suffer "negative cognitive effects". Once mastery in one
language has been obtained, the child has moved beyond the first threshold.
"Positive cognitive effects" result when a child develops high proficiency
in both languages. Cummins pointed out that the studies that showed a negative
effect were associated with linguistic minorities, where the minority language
was being replaced by the socially dominant one, what he called "subtractive
bilingualism", whereas the studies that showed a positive effect were associated
with "additive bilingualism", a situation in which majority-language
children acquire a second language. In Cummins' words:
"Subtractive bilingualism, where L1 is being replaced by L2, implies that
as a bilingual in a language minority group develops skills in L2, his competence
in L1 will decrease. It seems likely that, under these circumstances, many bilingual
children in subtractive bilingual learning situations, may not develop native-like
competence in either of their two languages." (Cummins
1976, 20; my italics, V.H.)
A second hypothesis that became important in this context was Cummins' "Interdependence
Hypothesis" (1979). This hypothesis argued that
there was an interdependence between first and second language, such that when
the use of the L1 was promoted by the child's linguistic environment outside
the school, then a high level of L2 achievement would also be likely to occur
at no cost of L1 competence. Also L1 and L2 skills are seen to be interdependent,
i.e., they are manifestations of a "common underlying proficiency."
High levels of L1 proficiency help L2 acquisition, and conversely, high proficiency
in L2 has positive effect on L1 development.
Interdependence of the two languages on the one hand, and subtractive bilingualism
on the hand, will thus account for semilingualism. This was the basis of Cummins'
early adaption of the term 'semilingualism', which proliferated through the
academic and professional world of migrant studies and became part of the linguopedagogical
discourse. In the wake of these hypotheses and the discussions they evoked,
evidence and counterevidence for the implied assumptions were found. Anyway,
what remained was that it was typical for children from migrant social background
to grow up in a social and linguistic context of deprivation and subtractive
bilingualism. This verdict became a quasi automatism and made any kind of problems
in language proficiency subsumable under this category. The needs and demands
of migrant communities were not taken into account, let alone any form of bilingualism
which did not present double monolingualism (neatly formalized as L1 and L2),
such as forms of complementary bilingualism(7).
Although the term was eventually abolished because of massive academic criticism,
at least in academia, it crept in again through the backdoor. "There appears
to be little justification for continued use of the term 'semilingualism' in
that it has no theoretical value and confuses rather than clarifies the issue,"
wrote Cummins himself (1994, 3814). But then he continued
as follows:
"However, those who claim that 'semilingualism does not exist'; appear
to be endorsing the untenable positions that (a) variation in educationally-relevant
aspects of language does not exist, and that (b) there are no bilinguals whose
formal language skills are developed only to a relatively limited level in both
L1 and L2." (Cummins 1994, 3814)
In a way, this is a strange kind of argumentation: Those who claim that semilingualism
does not exist imply that there is no variation in language and that there was
no limited level in both languages of the bilingual child. A limited level relative
to what? Relative to which language at which point of time in the learning and
acquisition trajectory of a child? Certainly there will be 'limited levels'
at different points of time in any normal language development. Data on such
oscillating development abound (e.g. Mahlstedt 1996).
When Cummins' first argument is expressed affirmatively, it implies that those
agreeing with semilingualism admit that there is variation in language. Then
put in other words: variation implies semilingualism. The understanding of Cummins'
use of 'variation' is very restricted here. As one of his critics note:
"From one perspective, the assertion that variation implies semilingualism
appears strikingly similar to the basic claims of classical prescriptivism,
where linguistic differences are construed as related hierarchically, and the
speech of the educated classes is regarded as better or more developed in certain
respects than the speech of the poor (or, in the case of Cummins's theory, linguistic
minorities in the United States). As with prescriptivism, the characteristics
of 'better speech' are taken to be precisely those characteristics that so-called
semilinguals lack." (MacSwan 2000, 16).
Variation (i.e. the phenomenon of varieties in languages) in the linguistic
sense, is, of course, part of any language (in terms of historical, regional,
social, ethnic, etc. differentiation), and it comprises the competence of any
normal speaker; as it is part of a language's infinite possibilities of adapting
one's speech or style to different social contexts, varying participation frameworks
or whatever kind of registers are negotiated or possible in the course of the
communication process. How much variation ofvthis kind there is, we will see
in the speech samples to be represented and analyzed in the next chapter. But
certainly, these variations will rather give evidence of competence, not of
deficiency.
* * *
Let me summarize this far: 'Semilingualism' and 'double semilingualism' are
both labels-taken-as-concepts, and concepts-taken-as-labels with a kind of subsumptive
passe-partout function for qualifications and evaluations of bilinguality matters
for migrants and in particular migrant children who get confused with learning
two languages, their language L1 at home, and later, when getting enrolled in
kindergarten and school, the new language L2. In arguing in favour for the concept
(even if declining the term 'semilingualism' as such - see Cummins-quote above),
it is empirically poorly substantiated (mostly by tests) and in spelling it
out on a more theoretical basis, we come across metaphors like "subtractive
bilingualism" instead of a profound argumentation.
Surface criteria of semilingualism as enumerated by proponents like Hansegård
and Cummins thus stand on very unsteady ground. It seems that the notion has
been influenced by the deficit hypothesis put forward by Bernstein (1971)
in which the social class-determined notions of restricted and elaborated code
account for different linguistic behaviour.
The alleged deficiencies in L1 and L2 are also detrimental in the long run.
They would become permanent (the technical term for this is 'fossilization')
and, in social life, would hence lead to stigmatization and thus remain a life-long
handicap to the psychological, social and moral development of the bilingual.
'Semilingualism' is only a term, a notion. It started off as a kind of descriptive
hypothesis, trying metaphorically to grasp deplorable symptoms of poor bilingual
achievers; then it became somehow substantiated through theoretical constructs
like Threshhold Hypothesis and Interdependence Hypothesis and spread into academia;
from there it was anxiously absorbed by what I called 'migration industries'
above. From here it developed into a passe-partout-concept for all kinds of
linguistic deficiencies, applied only, though, to bilinguals from migrant background.
And most important: It remained a purely cognitive concept, with social and
psychological reverberations in the long run - and, we could add, socioeconomic
ones as well. Cognitive locations have the important effect that they are individualized
and only visible by surface symptoms. Interactional, social and socialpychological
reasons for deficient behaviour are to a large degree left outside of perception.
Last but not least, I have mentioned a kind of ideological basis for the successful
spread of the semilingualism-concept which, in a way, has become part and parcel
of the concept itself. This ideology was fed by three interrelated ingredients,
namely (a) precriptivism with its implication for normative use of a language;
(b) the role of a language as a legitimate one: only the majority, dominant
language as a means for social integration is what really counts (German in
Germany); and (c) the essentialist idea that languages and language use have
to be neatly separated and are not allowed to blur.
4. Language as blurred genre
4.1 Polyphonia
The linguistic reality is a bit more complex than the concept of semilingualism
will make us believe. Take a middle sized city like Augsburg , around lunchtime
when school is out hundreds of pupils pour into the trams. A multilingual jumble
of voices arise. What comes to my ears is not only German, Turkish, Greek, Russian
and other languages, but I can also hear mixed conversations in German and Turkish,
German and Greek, or German and Russian, in which languages are switched in
a breathtaking speed. I listen in amazement to the pupils' virtuosity until
my academic interest sobers me up, knowing that these adolescents' and kids'
linguistic productions are hardly estimated, and not respected in the schoolclasses
they just left. Monolingualism and German alone is what counts there. It's neatly
separated languages that are accepted as respected bilingualism. Though the
overheard conversations literally speak another language; one that is multiply
varied and diverse, mixed, polyphonic and multilingual. Linguistically such
languaget could be seen as a blurred genre, a mix; some call it 'hybrid language'.
For others, however, it might just be another instance of 'semilingualism'.
In the paragraphs to follow I want to present a couple of examples of such
blurred or mixed codes of language from migrant adolescents Turks. But not to
add evidence to the semilingualism-concept, rather to show that these forms
and codes undermine every single argument put forward in favour of semilingualism.
Furthermore I will show that the emergence of these blurred codes can finally
only be understood within the context of migration history and migrants' community
life worlds. Regarding these codes solely as a cognitive phenomenon would certainly
not help us to learn anything at all about the relevance these codes play in
the life of school kids with a migratory background.
The speech samples to be described are in no way homogeneous. But they have
many features in common. The most prominent is the switching between those two
languages, we call 'German' and 'Turkish'. The degree of switching and mixing
is highly differentiated, so are the various functions within their interactional
logic. What we find is a whole spectrum of bilingual patternings up to genuinly
new and autonomous forms, which do not belong to either of the languages involved.
But that is not the only source of variation. Also jargonized and dialectal
features, and furthermore, ethnolectal stylizations, form part of the resources
which are exploited for the switching and mixing of the language varieties involved.
All of these I will call 'mixed language varieties'.
4.2 The data
The basis for my analysis are conversations recorded between bilingual adolescents
of ethnic Turkish background whose parents or grandparents had originally immigrated
as 'Gastarbeiter' (guestworkers) to West Germany. The conversations were audio-
and partly videotaped in informal gatherings. It was mostly one of the participating
parties that did the recordings. In most situations the adolescents did not
know what kind of data the researcher was looking for(8).
The majority of my informants were between 15 and 18 years old; a critical
age, around the time of leaving German Hauptschule (that is the minimum graduation
within the German tripartite school structure) and looking for work or apprenticeship
on the labour market. With some of the informants I led in-depth interviews
about their usage of 'mixed language'. Most speakers in my data are male. Only
one of my recordings contains female 'protagonists'. But the gender bias is
solely due to the chain of my informants (male adolescents ask other male adolescents
to make the recordings etc.). In female groups there is a similar range of switching
and mixing as can be seen from some of my own data and in particular from the
data of the Mannheim project (cf. Kallmeyer, Tandogan-Weidenhammer,
Keim 2000).
The excerpts I present here are solely those which include 'heavy' mixing and
switching. Not all the data recorded display such language alternations. Many
stretches of talk are conducted more or less monolingually, mostly in Turkish.
Ignored were such data with single item insertions and those where the switches
were due to adressing monolingual participants.
A word has to be said about 'migrant adolescents'. The adolescents whose language
will be focused on in the following paragraphs are certainly no migrants themselves.
But regarding them as 'immigrants' or 'migrants' is a progress that acknowledges
at least their status being borne out of the migratory context (of their parents
or grandparents). Furthermore, these adolescents have so far been mainly the
object of this discourse, hardly subject of it. The semilingualism-debate is
one such instance of 'incapacitation'. Changes in perception and recognition
are coming slowly (cf. Terkessidis 2000). It's mainly
the articulations of 'migrant adolescents' themselves that promote this change,
a change into the direction of further de-objectifying the migratory discourse
and giving respect to autonomous forms of expression.
4.3 On the inherent logic of switching and mixing
The following excerpt is a sequence out of a discussion between the two 16
year old friends Ercan (E) and Hakan (H) about an "Inititiativkreis"
(Interest Group), a kind of social club for 'foreign' adolescents. Both speakers
grew up in Germany and go to the final class of German Hauptschule. The conversation
takes place at Hakan's home:
The excerpt combines a number of mixed language phenomena. Furthermore in German
as well as in Turkish we find some typical spoken language elisions (such as
in "nerde" line 1 or in "nich" line 18, 22) as well as dialectal
elements ("des" [dæ:s], line 15); in German there are also some
typical youth language expressions like "kolpingmäßig"
line 10, 11; "lernstudiomäßig", line 28, 29; or "cool"
line 31. The numeric relation of German to Turkish is roughly 2 to 1. But we
have to keep in mind that the suffixing principle of an agglutinating language
like Turkish may pack much more information into one word. Also some formal
aspects of language alternation are worth to be looked at: Twelve out of about
26 turns of speaking lines are monolingually German as opposed to five or six
turns in Turkish (including the insertion "Initiativkreis" in line
1). Longer sequences tend to alternate languages. Here we find German dominance
as in lines 13f. or Turkish dominance as in line 17. Looking at the mixed data
like this remains purely formal and normative, however. With a couple of lexemes
it is in no way clear to which language they should be attributed; proper and
quasi proper names like "Initiativkreis" and "Königsplatz"
fall out of count (lines 1, 3, 7). But what about insertions such as "Mitte"
and "Einbahnstraße" (lines 8 and 9)? By their suffixation we
can see that and how they are integrated into Turkish. Although the four word
sequence part "O Einbahn- straßenin tam Mittesinde" consists
of two German content words as opposed to two Turkish functional items, it is
nonetheless a genuine Turkish sentence.
As we can see, a formal approach is quite limited. Take alone my way of transcribing,
intended to make the reading easier, that is German in recte, Turkish in italics,
is quite problematic as it proceeds purely technical, namely referring to language
assignment, thus supporting a view of double monolingualism: The speakers speak
either German or Turkish, whereas in fact they 'language' in and with and across
both languages.
What is much more promising is taking a look at the switches themselves like
those between line 3 and 4, or between line 11 (7 ff. respectively) and 12,
17 and 18; or answering the question why there is a switch at all in line 17
or in line 28 or 29. Bilingual speakers have both languages available, as we
know. We could explain away some instances with word searching difficulties
and other competence related explanations. Also we know of such preference principles
like following prior speaker's choice (Auer 1988). But
there are no really satisfying answers we find here. Is "Richtung Stadt
böyle" the adequate connection to "bu Initiativkreis" (line
1 f.)? Rather not. In fact, most of the sequence where the "Initiativkreis"
is located is negotiated in Turkish (up to line 11). Then we have a language
switch. At least in what follows there is a shift in language dominance (lines
12 to 16). The non-Turkish sequence line 4 to 7a is clearly an insertional sequence
in which H tries to explain that the "Inititiativkreis" is in fact
already known to him. It constitutes a formal opposition to E's description
of where the "Initiativkreis" is located: It is an autonomous sequence
refuting E's overexpliciteness (line 4), followed by a checking question (line
5), H's confirmation of that (line 6) and its reconfirmation (line 7a) before
both return to the status quo ante in line 7b ff. We thus find traces of a codeswitching
structure in this conversation, one that emphasizes e.g. oppositional formats
of conversational sequences.
We come across similar methods of opposing formats in the sequence line 15
to 18 through H's contradiction by his questioning of what the "Initiativkreis"
has to offer (line 15), E's answer (line 16) und H's ridiculising conclusion
about that (line 17). For E's mildly formulated protest against this he remains
with his prior choice, i.e. German. That is, just for H's concluding statement
"orda para kaybediyor yani" (line 17) he makes use of this section's
contrastive language, i.e. Turkish; furthermore marking it by laughter which
might also emphasize it as modally contrastive.
Opposing patterns may be just one among many keys to explaining language alternation.
The next example is a transcript from a conversational exchange which a student
of mine has spontaneously recorded at a busstop. Here we come across the two
15year old adolescents Ferhat (F) and Ahmet (A) who wait for the bus and make
nasty comments about the busdriver and the bus service in general.
The first impression is that both languages are more or less equally distributed
(50 Turkish as against 58 German words). Again we find monolingual as well as
bilingual sequences. A type of alternation we could spell out as "Speaker
1 speaks language A, speaker 2 language B" we find e.g. in line 1 to 5
or line 6 to 9. Only in Turkish are lines 5 and 6; (almost) in German are the
last lines of the excerpt. Also within a turn we find those typical one-word-insertions
like "acht" in line 6. In line 10, 11, 13, 20 and 25 we find switches
within the respective turns, which to my opinion syntactically as well as functionally
are quite comprehensible. If we take line 13 as an example, we can immediately
see that we have (a) the Turkish bracketing of a German quote and (b) the ensuing
commenting of the quoted occurrence, a rhetorical and thematic differentiation
by means of the two languages German and Turkish. Or line 20, for example, is
an obvious case of adressee specification: In the Turkish part A refers to the
Turkish student (in 3rd person) who makes the recording; in the German part
- spoken in an accelerated manner - the immediate adressee is his mate F (in
the 2nd person). Even if not every single switch is explainable, it is quite
obvious that most of them nicely comply to phrase boundaries [//] such as in
line 10: "Otobüsün dolu olmasina çok gicik olyom // hey
Mann ge + voll" or in line 11 "Ja weisch (+) // girdik (h) {giris/giriyoz=simdi}
içeriye".
Not all alternations, however, stick to these phrase boundary rules: In line
25 "Zwei zu bir miydi?" (Was it two to one?) the switch 'respects'
no other boundary than just that between words. Its rhetorical function remains
obscure. And if we once more take a look at the language alternation distribution
according to speaking turn and speaker, more puzzles emerge: The first 5 speaking
turns seem to be characterized by the different language dominance of the interlocutors:
F. speaks Turkish, A replies in German. This might correspond to individual
language preference or to the degree of (un)certainty in the respective languages.
In the next turn, however, (line 6) A. switches to Turkish, thus complying to
F.'s language choice. Surprisingly, F. continues in German (line 7), likewise
in the turns to follow (line 8 to 10), we find the above pattern inversed before
both speakers use both languages within their turns (line 10, 11 and 13).
To conclude: Thus far we have not really found a reliable pattern of interturn
language alternation unless we declare the maxim "Don't use prior speaker's
language!" as at least valid for parts of the conversation. We would thus
have discovered a further pattern based on opposition - on formal speaking turn
opposition. There is hardly narrative or dramaturgical logic behind that, more
the potential of playing with oppositional resources. We have to ask, what is
it that stimulates the adolescents to conduct their conversational interchanges
in such manifold code alternating ways - sometimes quite comprehensible in terms
of our knowledge of codeswitching, and sometimes in many ways quite surprising?
4.4 Artful mixtures: Playing with languages and playing with normativity
In the next excerpt to be presented, the aspect of mixing will become particularly
clear, as we will be able to see, how bilingual competence will be made use of
as resource for language plays and for extempore poetry. The examples will furthermore
nicely display the tension between a normative consciousness and awareness of
language and its simultaneous undermining by hybrid language use.
In the below scene the three adolescents Mehmet, Ugur and Kamil, aged 15 and
16, hang around in a self-service shop in their neighbourhood. They buy doughnuts
and fool about. At one point Mehmet swallows a piece of his doughnut in the
wrong way and he starts coughing to which Kamil responds by slapping Mehmet's
back and ironically wishes him to enjoy the meal. This prelude continues down
to line 5.
(9), (10),(11),(12)
The episode which is of main interest here starts after the pause of two seconds,
which Ugur introduces with "Stirb langsam" (line 7). Ugur thus comments
on Mehmet's ongoing coughing. "Stirb langsam" reminds Mehmet of an
episode when Ugur pronounced the same title of the video film as "Sitirb
langsam" for which he gets a loud laughter in response (lines 8 to 13).
The strong reaction is probably due to the pronounciation of "sitirb"
for "stirb". This epenthesis is regarded as a typical Turkish accent
and is highly stigmatized(13). While still laughing about
Ugur's mispronounciation, Mehmet directs his friends' attention to a sign in
the shop in which they hang around. This sign originally reads "EINGANG"
(entrance) but the first "N" had dropped leaving "EI GANG"
(line 15). This leads the three adolescents to a brief, fast and effective word
play, which actually cannot be adequately represented by the transcript above
(or by any other transcript). The sequence from line 15 to line 31 or 32 respectively
is fully dedicated to the polyfunctionality and to the associations of the truncated
presyllable "Ei" which in German of course means "egg" and
which is homophonous with the Turkish noun "ay" (moon, month) or -
if extended by the Turkish vowel [w] - with "bear" (ayi).
If the adolescents alternate languages, they do it in a way which seems to
have no restrictions when adding Turkish suffixes to German words (not vice
versa, however). This means for the word play above, that all kinds of German-Turkish
combinations have to be taken into account. Thus a German "Eigang"
(egg walk) may also be thought of as a German-Turkish "ay Gang", a
combination of "moon" or "month" with "Gang" (walk,
corridor, course (of a meal)).
It is probaby due to this multiple understandability that Kamil asks "Nerede
bunun ayi/Ei-i" (line 18 and line 20), to which Mehmet reacts with a laughter
and which leads Kamil to the variant: "Ay/Ei Gäng" (line 22).
Kamil thus transforms "Gang" into "Gäng", but the German
orthography does not show that Kamil's pronounciation is indeed [aw gæõ],
bringing a third language, i.e. English or American, into play. At this point
Ugur, who has been teased before, enters the play as well (line 24), though
it's not clear whether his contribution is one of participation or one of checking.
Also Mehmet, who had started the play, offers another variant (line 25 and 26):
Mehmet pronounces the complete German EINGANG now with a strong draw as if the
word was made up of three instead of two syllables. This is also an interesting
parallel to Ugur's alleged epenthetic pronounciation before (cf. line 09), in
that Mehmet inserts an additional vowel between the semivowelized [y] and the
reinserted [n], thus pronouncing it in a very Turkish way as *{ay/Ei}-yin-gang*.
Another allusion to Ugur's use of the stigmatized epenthetic form? Mehmet continues
by returning to Kamil's Angloamerican variant, caricaturing the 'heavy accent'
of a German speaking American: [awn gewn svaw gewn] (line 26). The intonation
pattern is roughly like that: ¯° 2°. At the same time Mehmet's
voice goes one pitch up. This variant derivates the verb "gehen" from
"Gang". On the paradigmatic level "ein" is substituted by
"zwei". "Ein gehen" or "eingehen", on which "ein
geyn" is based, is a proper German verb (construction) with different meanings.
"Zwei gehen", on the other hand, makes only sense as "two (persons
etc.) go". It is however the parallelism that counts which Kamil reconverts
into the nominal forms "Eingang (+) Zweigang" (line 28). Ugur fully
comes in now with a new version, in that he brings the "bear" (ayi)
into play (line 27) using a full fledged Turkish genitive construction "the
bear its walk" (ayi-GEN Gang-POSS. -> ayi-n-in Gang-i). Ugur's bear-version,
however, is not elaborated upon. Rather, Ugur adapts to Kamil's "Eingang
(+) Zweigang" to which he adds "Weitergang" or "zweiter
Gang" respectively (line 31). And as if "Weitergang" should be
taken literally, Mehmet opens up a new subject (line 33).
Obviously, the language play has reached its end here. And indeed, the last
two contributions were fully in German, presenting real existing words, far
enough away from the initial word. In the table below, the word play which lasted
only a few seconds is represented in a kind of overview by enumerating its various
stages.
It is quite normal that children and adolescents play with language, that they
test it, turn words upside down. That Mehmet, Kamil and Ugur do this in two
languages, that they extract and exploit the language material and the ambiguities
to play with out of two languages, is certainly the privilege of bilinguals.
Mehmet, Kamil and Ugur go to German Hauptschule, that is striving for the minimum
degree in secondary education. Their educational achievements are not brilliant.
Pupils like them are only too often, as we know, regarded as semilingual or
as defective bilinguals. But we also find a high degree of language awareness
expressed for example through the episode of stigmatising Ugur's epenthetic
pronounciation or the caricaturing of the American accent and - not to forget
- the missing "n" of "Eingang", which served as the immediate
cause of the play. All this reflects their normative awareness of language.
Part of the play are word derivations, conversions, paradigmatic substitutions,
parallelisms and continuously ambiguities, which are borne out of the two languages'
in-betweenness, fully exploiting the potentials of bilingualism. Playing in
between the two (and sometimes more) languages (or codes) is not unfrequent.
Mehmet and Kamil do it quite often as well as other adolescent bilinguals. In
another situation we find a kind of extempore composition, triggered by talking
about a football coach named "Wolfgang". The summary of it is presented
below (without the context in which it is embedded):
(14)
Besides the alliterative play with the letter "o", it is also the
bilabial initial sound "m" (Wolf'un oglu molf) which is conspicuous
here. This "m"-alternation is a typical Turkish etcetera-form: cf.
Hasan Masan meaning Hasan and his friends - a pattern used here in an expressive-poetic
function.
To give another, less extensive example: At one point, talking about fights
between football fans, the adolescents use an imaginary gun and rhythmically
shoot around, supported by a Turkish counting-out rhyme which was furthermore
accompanied by clapping hands: "Bir sana bi hava / bir sana bi hava"
("One for you, one into the air / one for you, one into the air"),
which then is altered into "Hava Ana", to "Mother Eve",
to profanely end as a German "Havanna Zigarre" (Havana cigar).
This kind of language performance in two languages often present extempore
poetry. In doing so, the performers do not only make use of their bilingualism,
but also exploit its possibilities of boundary crossing by fusing and blending
words as well as other expressive mechanisms (like the bilabial etcetera-marker).
The bilingual language players thus display quite a high normative awareness
about language and its potentials, even about word order processes. In a context,
which would pay estimation and respect to bilingual language use and linguistic
creativity, a high level of linguistic reflection and consciousness would be
attested to such speakers.
Besides vernacular and dialectal elements stylized elements of Gastarbeiterdeutsch
('guestworker' German, immigrants' Pidgin-German) are also integrated into the
performances. Stylized speech has well been researched by Rampton as one of
the 'crossing' phenomena in language ('stylized Asian English', cf. Rampton
1995). Stylized German is the - mostly exaggerated - imitation of the first
generation migrants' accent in speaking German. Ugur's alleged pronounciation
of "Stirb" as "sitirb" in transcript (3) is a typical instance
of this accent.
4.5 Stylisations as mimicry
The use of Gastarbeiterdeutsch or marked elements of it (emblems) as part of
the mixed language repertoire plays an important role in itself. As a matter
of fact, such emblems function as intertextual quotes. On the one hand, this
stylized variety of Gastarbeiterdeutsch is a copy or quotation of this particular
language variety as it was and still is spoken by the parent generation of migrants;
on the other hand, it is also a quotation of majority society's ascriptions
vis-à-vis migrants in general (as is used e.g. in foreigner talk, in
caricature, comedy shows and elsewhere). Throughout the more than 40 year old
process of postwar immigration to Germany (at least since visibility of immigration
has asked for political measures), proper command of German has become the tertium
comparationis for integration as defined by majority society; and noncompliance
with this demand could at any time be made the rationale for distinction and
discrimination (see above). Within the use of mixed language, however, Gastarbeiterdeutsch
is merely a quotation and a stylistic ingredient to play with. At the same time,
and this is most important to emphasize, it is not the authentic language of
those speakers who make use of it but it represents re-appropriations by mimicry.
This multiple role is nicely exemplified in the below excerpt. It's Mehmet
(Me) again, this time sitting together with his friend in his room listening
to Techno-Music. Mehmet's little niece and nephew are playing on the ground.
In the background one can sometimes hear the voice of Mehmet's mother. When
she enters the room the following minor dialogue develops between mother (Mo)
and son:
As we can see, Mehmet's mother does not approach her son in Turkish, but in
a loud and an extremely exaggerated Gastarbeiterdeutsch way of speaking. Asking
"WIE GE::::ST?" is certainly not meant as a kind of welcoming remark
towards her son or his friend. Both have been there all afternoon and in frequent
contact with the woman. Mo's question is not only marked by the metathesis of
"TS" to "ST" (the Standard German form is "Wie geht's"),
but in particular on the prosodic level by its loudness, the vowel lengthening
and the high pitched voice. Mehmet responds in the same extreme and exaggerated
way. "NIX GU:AT" is an apt and adequate answer in tone and voice and
furthermore diphthongizes the German [u:] in "gut" (good) into "guat",
which is an exaggerated form of Bavarian dialect. Furthermore, it contains the
highly stigmatised negation particle "nix". "Nix" is the
passepartout negator in Gastarbeiterdeutsch and is also Bavarian. Phrases like
"nix verstehen" (no understand) and "nix deutsch sprechen"
(no speak Geman) are caricature classics for ridiculizing migrant speech. His
mother's checking question terminates the sequence. She picks up the child and
leaves the room.
The de-contextualisation, the deplaced topic and the conditional irrelevance
of this three-turn-sequence speak strongly in favour of a metaphorical intertextual
language play between mother and son, in which the stylised form of Gastarbeiterdeutsch
come to fruition. This variety is omnipresent (cf. Androutsopoulos,
Hinnenkamp 2001). Its figurative use even releases ritual clichées
from rituality and topics from their thematic boundedness. Its function is purely
phatic: A We that reassures itself of its own identity via exaggerated and caricaturing
use of voices which are not their own (anymore) but which become re-appropriated
and re-contextualized in play, this time, however, stripped off its threatening
loadenness.
5. Discussion
5.1 'Speaking mixed' as a blurred genre in its own right
Based on the self-characterizations of the migrant adolescent users of mixed language,
I want to describe in particular its roots within migratory history and immigration
society and then, to locate it within a wider societal context of the migratory
discourse.
In the interviews which I led with the adolescent code-switchers and mixers,
they call their way of talking gemischt sprechen or karisik konusmak (both meaning
"speaking mixed"); some describe it as "speaking half German
half Turkish" (proponents of semilingualism would certainly like that!),
my informants in Augsburg refer to it quite metaphorically as yarim yamalak
konusmak, which could be glossed as "speaking halfly mend" or "patchwork
speaking". But in what everway they label their way of talking, there are
two things that are crucial to it:
- They have given this specific way of speaking an autonomous name. They thus
distinguish between this particular and other varieties of language and even
other languages. An internal differentiation according to the degree of switching
and mixing does not exist.
- The characterization of this variety by its users always expresses an activity:
It's not named by a noun such as "mixed language" or "patchwork
language", but it is always combined with verba dicendi-formulations
such as German sprechen and reden or Turkish konusmak (all meaning "to
speak"). That is, in that they are speaking mixed, they are doing something
very actively(15).
The relevance of this kind of activistic self-reference becomes clear on the background
of how the language of their parents (and sometimes grandparents) was labeled.
They spoke Gastarbeiterdeutsch. This characterization of the language variety
spoken by immigrants has even entered Hadumod Bußmann's German "Lexikon
der Sprachwissenschaft" (Linguistic Dictionary), where we find the following
entry: "Gastarbeiterdeutsch is a pidgin variant which developed in Germany
since the 60es and 70es and is characterized by paratactic sentence patterns,
limited lexicon, little redundancy, deletion of article, preposition, conjunction
and verbal inflection. All these features are generally occurring, irrespective
of the speaker's native language" (Bußmann 1983,
157; 1990, 262 f.; my translation, V.H.). The name Gastarbeiterdeutsch has its
root not in the users themselves but is an 'other'-characterization (by wider
society, media, linguists). And as we learn from the above entry, it was not conspicuous
because of its genesis in emergency multilingual language situations or in its
supportive function in untutored processes of making oneself understood, rather
it is only its deficits which were focussed on here.
To reiterize: Labels such as Gastarbeiterdeutsch and the notorious 'semilingualism'
were denominations given to the migrants by majority society. Contrary to that,
the generation of "speaking mixed" has given this name to their language
(variety) itself, without being labelled as such by others. Majority society,
furthermore, is no more the direct addressee of this language, all it may provide
now is overhearers. In conversing with members of majority society, these adolescents
have German at their disposal, with Turkish speaking people they speak Turkish,
and among themselves they use the mixed variety. These are the general options.
Of course, reality does not submit to such clear-cut categorizations. It would
be naive to pretend that the options or choices selected would always be optimal
ones. We find imbalances in language dominance, have to deal with incompetence,
with emergency solutions in order to reach one's communicative goal. How bilingual
speakers deal with imbalances can nicely be seen when within a group, the more
versatile speaker adjusts vis-à-vis a speaker who is less fluent in switching
languages, e.g. by converging toward ones partner's preferred or stronger language.
Another recipient design in this respect is the doubling principle, repeating
or paraphrasing the utterance (or part of it) in the other language (cf. Hinnenkamp
2004).
5.2 Gemischt sprechen and identity
Speaking mixed is not just one linguistic option among others but is also an expression
of a particular identity within the migratory process. The adolescents who grow
up under such polycultural and multilingual conditions are confronted with contradictory
demands, those from society, from their own community, from their peers - each
setting different standards which kind of linguistic and cultural conduct is the
right one, the wanted one. If we would fall back upon another of those often cited
essentialisms within the migratory discourse, we might say, these adolescents
use a 'split language' similar to how they possess a 'split identity'. Hence,
their mixed language would just be seen as another expression of their confusion
between two languages, two cultures, two socialisations etc. This, of course,
is extremely simplified if not wrong. It portraits language, culture and identity
in a one-to-one relationship and operates on a basis of rigid essentialist concepts.
Identity is regarded as a fixed and ready-made entity, like a suit which either
fits or does not fit. The constitution of identity (of identities, I should rather
say) is a permanent process and communication plays a main part in it. One does
not have identity but one operates, interacts and struggles with different identities.
Its formation is constituted in a continuous debate with other people in conflict,
with different social and societal demands (cf. Antaki, Widdicombe
1998). These identities are also borne out of boundary marking against majority
society or against one's parent generation.
Le Page and Tabouret-Keller (1985) speak of identity
in terms of their transformation into "acts of identity", i.e. in
particular communicative acts, in which an inventory of categories is used,
how to deal with one's own and with others' typifications and ascriptions, how
to claim, to confirm or mark off membership and affiliations, how to include
oneself into and exclude oneself from groups and communities, etc. Acts of identity
make these and other categories of localizing oneself individually or socially
relevant. One's localisation is neither free of contradictions nor is it permanent
and stable. It implies a continuous struggle between chosen and ascribed identities.
Gemischt sprechen is not simply an expression of a transitional social identity.
It does not simply juxtapose elements of different languages, but blends them,
creates new compositions, hybrid forms and it fills up a semantic room in society,
which was hitherto unoccupied and undefined. It mirrors an autonomous approach
by way of language alternation, language mixing and appropriations from both
linguistic communities and both 'cultures' (if we allow this simplification
for a moment). In this sense gemischt sprechen represents an autonomous hybridolectal
We-code. That is, code alternations do not correspond to alternations of metaphorical
We versus They-affiliations along the lines of Ingroup - Outgroup / We-Code
- They-Code (cf. Gumperz 1982), but represent a We-code
in its own right (see Swigart 1992; Meeuwis, Blommaert
1998; Sebba,Wootton 1998; Hinnenkamp
2000a).
6. Conclusions
6.1 Mirroring the history and social conditions of the migration process
The adolescent actors who grow up in the polycultural and polylinguistic space
of urban migration centres, develop specific hybrid forms and creations out of
the pool of codes at their disposal. This bricolage-argument is well known (cf.
Clarke 1976). The bilingual and bilingually mixed conversations
of migrant adolescents do not just present instances of code-switching in the
sense of juxtaposable rules from two languages plus their respective local functions.
What we get is an autonomous blurred and hybrid code, oscillating between two
languages, representing both languages, and, at the same time constituting something
third, what they call gemischt sprechen, karisik konusmak, yarim yamalak konusmak,
and the like - a linguistic code in its own right.
This mixed language functions as a mirror of the historical, social, cultural
and linguistic conditions, in which these adolescents grow up. Historically
the code is the critical response to the the majority society's demands on integration.
The demand is to have a good command of German and at the same time to be allowed
to preserve one's Turkishness as an "ethnic identity". Sociolinguistically
the adolescents react with an ingroup-language, a We-Code, which implies both,
deficit and competence, but first and foremost, difference and autonomy. The
latter two lead to exclusion, of the parent generation on the one hand, and
of majority society on the other hand. Both, however, deficit and competence,
as well as difference and autonomy, become re-integrated into an autonomous
code which is made up by the 'donating languages' - to adapt a word from the
beginnings of creole studies - but which are also distorted, caricatured and
reinterpreted. Mercer has commented this dynamism with respect to "black
film practice". But it fits perfectly into this context:
"Across a whole range of cultural forms there is a 'syncretic' dynamic
which critically appropriates elements from the mastercodes of the dominant
culture and 'creolises' them, disarticulating given signs and re-articulating
their symbolic meaning otherwise. The subversive force of this hybridizing tendency
is most apparent at the level of language itself where creoles, patois and Black
English decentre, destabilise and carnivalise the linguistic domination of 'English'"
(Mercer 1988, 57).
In a more general sense this pertains to all "mastercodes", so that
English could be substituted by other dominant languages or hegemonial codes.
The blurred language as used by the Turkish adolescents thus represents a kind
of feedback effect to the "mastercode", it constitutes a re-appropriation
and re-contextualization of a discourse that so far had been defined solely
by others. In this respect the adolescents' mixed code with all its implications
is part and parcel of a new discourse that stands in opposition to the above
described discourse-labels such as 'semilingualism', 'split identity', 'cultural
diremption', 'Gastarbeiterdeutsch', and the like.
6.2 Semilingualism or not?
The few examples discussed in the two previous chapters show no evidence of semilingualism
or double semilingualism(16). Certainly, prescriptivists will
find inconsistencies in language use as well as in the choice of language; most
probably they will find grammatical flaws and incorrectnesses. But spontaneous
and everyday language use is always of this kind. The Turkish of these adolescents
is the product of their German-Turkish life-world, the Turkish under the influence
of diaspora and German environment. Such diaspora-processes are normal and are
not stopped through the use of Turkish satellite TV in Germany, unless they get
systematically steered into the opposite direction by institutional support of
Turkish as a Second Language (in fact, some successful attempts have been made
in this direction).
But let's take just a superficial look at some of Hansegård's "lack
of competence"-criteria: We can't say much about (a), the size of the repertoire.
But certainly it is a repertoire composed of different and intertwined varieties
of German and Turkish. As to lack (b): "linguistic correctness" is
on the one hand reflected by speaking according to rules on the other hand on
the meta-level of norm-mocking (as displayed and discussed in the examples of
transcripts (2) and (5)). The lack of (d), the ability to create or neologize
is, of course, most obviously proved wrong by the examples. Certainly, those
who chose this as a criterion were not thinking of neologism made up of two
languages. Prescriptivist creativity is probably restricted to monosystems.
As to (e): How many functions of language did we come across in the few examples?
At least we can add the poetic function, the expressive function, the metalinguistic
function(17). And the last of the criteria (f) refer to a
richness or poorness in individual meanings: Of course we can't really say anything
about that, but just the richness of "Eingang", "Eigang"
and "Ay-Gang", respectively, in transcript (3) gives us a clue.
What about Cummins' points, that he made in 1994? We remember his implication
that variation implied semilingualism. As we do not share (rather not understand)
his view of 'variation', we can't really tell. But he obviously thinks of variation
in the formal educational system, where just one particular formal norm is accepted
(an elaborated, academic form of language). What we find in terms of variation
is traits of dialect, of jargon, of ethnolectal stylizations; we also find a
strong consciousness as to the use of these different varieties and as to the
norms of language; in particular as to the awareness of norms of the majority
language. And we find a playing around with the normative and the legitimatory
claims of German. We thus find a lot of variation in each language as well as
across languages. If semilingualism would imply variation of this kind, then
it is here where we could find an extreme case of it - but only because it does
not comply with the school's demand of formal language proficiency. Cummins'
argument would be invalid as soon as school would enlarge its scope of legitimate
varieties to bilingual codes, at least on the level of awareness and recognition
of the existence of more than just one language, German - and we have to add,
more than just one variety of German. This would certainly give a positive feedback
to the pupils' expertise and creativity outside school. Motivation by recognition
is one of the basic requirements to success.
Beginning with a discussion of the concept of semilingualism and double semilingualism,
I have argued that the role of language and bilingualism as pursued by the majority
discourse is ignorant of the development of migrants' language use merely because
it does not follow the concepts of double monolingualism and legitimate language
use. Semilingualism does not only maintain that migrant children are particularly
prone to "subtractive bilingualism", but it also implies a whole ideological
universe of explanations for this (and at the same time tries to explain away
societal reasons), focussing on the migrant child as an individual language
learner and on his or her individual failure of mastering school's formal language
demands. Again, we find a prescriptivist idea of language behind that, one that
complies with the norms of school, and less if not at all with the standards
of multicultural and multilingual community life.
* * *
I began my paper by quoting Wandruszka in length:
"For humans there exists neither a complete
control of language nor a completely homogeneous speech community. Never and
nowhere will we find a perfect, homogeneous monosystem; always and everywhere
we will just find imperfect heterogeneous polysystems. The relationship of humans
towards their language is not one of perfect monolingualism, but just the opposite,
it is one of imperfect polylingualism and one of polylingual imperfection."
(Wandruszka 1979, 313; translation mine, V.H.)
If school and the majority discourse took heed this insight into language as
an open, imperfect polysystem, then certainly migrants' languages as well as
their polylectal varieties, their mixtures and creations with and in between
languages would not anymore be ignored or denunciated as semilingualism.
Appendix
Legend to transcriptions
Notes
(1) I am very grateful to the comments made by the editor
of this volume, Frank-Olaf Radtke, and by an anymous peer reviewer. The latter's
suggestions were very valuable and interesting. For reasons of space, however,
I could not follow all of her/his ideas of additional points to be discussed,
but I will certainly take them into account in another publication.
(2) I owe this notion to J. Normann Jørgensen from the
University of Copenhagen. - For a sociolinguistic discussion of the different
forms of and constraints of bilingualism cf. classics like Grosjean
1983 or Romaine 1995; for a comprehensive reader
on bilingualism cf. Wei 2000.
(3) In a reflection on the correlation of migrant children's
school achievements and knowledge of language as yielded by the results of PISA,
Lemper comes to the conclusion that deficits in German within the migrant family
is the strongest correlational factor. He thus pleads for an all-generational
inclusion of language promotion: "According to PISA the demand to intensify
the institutionalization of German language courses on all levels, takes indisputable
priority: On the level of kindergarten and school as well as on the level of
extracurricular organizations, and not only for migrant kids, also for their
parents"(Lemper 2002, 18; my translation, V.H.).
- Of course, as we see, Lemper speaks only of German as the legitimate language
to be promoted.
(4) A word about integration. It means in first line striving
for inconspicuousness in terms of a Tönniesian concept of Gesellschaft,
that is rejecting the establishment of ethnic 'parallel societies' from the
side of majority society. This inconspicuousness is positively supported by
facilitating equal access to the goods and resources of majority society. In
terms of Gemeinschaft, social integration does not contradict cultural, linguistic
and religious freedom, except where it gets into conflict with claims of the
Gesellschaft. The role of the German language, e.g., in this sense is certainly
part of Gesellschaft, whereas the mothertongues of immigrants have kept their
place in the realm of Gemeinschaft. A change of this would only be due in the
face of accepting multilingualism as a societal form of plurilingua, including
institutional support of its maintenance.
(5) In particular within the age group of primary and secondary
education up to 18 years there is an overwhelming increase of pupils with a
migratory background. With the strong concentration on big cities, many inner
city schools have disproportionally high enrollment rates of migrant children.
(6) At the beginning of the nineteen hundred and sixties Professor
Georg Picht published a momentous book titled "Die deutsche Bildungskatastrophe"
(1964) ["The German Educational Catastrophe"].
Picht argued, that it was not lower intelligence that was responsible for school
failure but that socioeconomic, regional, cultural and gender conditions were
responsible for the selection process. This, he concluded, was hardly in accordance
with the claim of democracy and socioeconomic effectivity.
(7) It's in a way quite paradoxical that on the one hand "interdependence"
is claimed and that a "common underlying proficiency" (Cummins) makes
possible the transfer of cognitive/academic or literacy-related skills across
languages, that on the other hand surface aspects of different languages (e.g.
pronunciation, fluency, etc.) have to be kept clearly separate.
(8) Special thanks to Tuna Döger and Ahmet Atasever for
their support in getting the data. - Most of the data has been gathered in informal
situations by one of the adolescents themselves, mostly during spare time activities.
Not all participants of a taped conversation were thus informed before. They
were then asked afterwards if they agreed in using their data. They all agreed.
And furthermore, they were very enthusiastic that interest was shown via à
vis their language. The participants were also asked for additional information
about some personal data.
(9) This is a genuine Turkish genitive construction: bu-GENITIVE
SUFFIX ay/Ei-POSSESSIVE SUFFIX.
(10) K's pronounciation here is different: The German Umlaut
"ä" signifies that he says it with an English pronounciation,
so the equivalent translation would be "gang".
(11) This construction is actually not translatable into English:
The German verb "eingehen" means primarily "to die, to decay"
and in another sense also "to enter" (in relation to "Eingang"
- "entrance"). Of course, the German prefix "ein-" is identical
to the number/indefinite article "ein" (one, a). Thus, "zwei
gehen" with "zwei" meaning "two" is a parallel construction
to "ein gehen". The meaning of this two verb construction is manifold:
e.g. in 'guestworker' pidgin "one walk two walk".
(12) This is the parallel noun construction to line 26. However,
the nouns do not correspond to the verbs. "Weitergang" translated
as "continuation" could also be written "weiter Gang" meaning
"wide corridor".
(13) Ugur's alleged realisation of "stirb" as "sitirb"
does not only allude to the highly stigmatised pronounciation of Turks' "Gastarbeiterdeutsch"
(guestworkers'/immigrants' German) of inserting vowels in between consonant
clusters as a transfer from Turkish which has many more restrictions on consonant
clusters than German. Secondly, the German [i] is furthermore relaxed into a
Turkish centralized [I]. Thirdly, German dialectal features go lost. It also
remains unclear, if alone Ugur's defective pronounciation is responsible for
the laughter or if this is also due to a particular role constellation within
the group.
(14) Wolfsburg is the German town of the Volkswagen-motor
works.
(15) In the critical and antiethnicist movement of migrants
other labels are used as well which partly have become popularized by the books
of Feridun Zaimoglus (cf. e.g. Zaimoglu 1995). "Kanak
Sprak" based on the xenophobe invective "Kanake" is one such
label which formally and semantically has a wider extension than gemischt sprechen
(cf. Zaimoglu's preface in "Kanak Sprak";
also cf. Pfaff 2002). "Kanak Sprak" reflects
and absorbs the negative ascriptions as much as it is an expression of new self-confidence
and identity. But as it consists mainly of stylized and jargonized forms (and
is not bilingual), it becomes easily majorized by nonethnic jargon and comedy
shows (cf. Auer 2000; Keim, Androutsopoulos 2000; Androutsopoulos
2001).
(16) For more evidence of the whole spectrum of variation
and virtuosity, in particular for the 'artful' combination of the diverging
grammars of German and Turkish, see the discussion of data in Hinnenkamp
2004.
(17) Depending on the model of language functions, we will
find other ones as well. My ones are based on Hymes 1962.
References
Androutsopoulos,
Jannis. 2001. "Ultra korregd Alder!" - Zur medialen Stilisierung
und Popularisierung von ‚Türkendeutsch'. In: LINSE
- Linguistik Server Essen
Androutsopoulos,
Jannis; Hinnenkamp,
Volker. 2001. Code-Switching in der bilingualen Chat-Kommunikation - Ein
explorativer Blick auf #hellas und #turks. In: Beißwenger,
Michael, ed. Chatkommunikation. Sprache, Interaktion, Sozialität und
Identität in synchroner computervermittelter Kommunikation. Stuttgart, 367-401.
Antaki,
Charles; Widdicombe,
Sue. 1998. Identity as an achievement and as a tool. In: Antaki, Charles;
Widdicombe, Sue, eds. Identities in Talk. London, 1-14.
Auer,
Peter. 1988. A Conversation Analytic Approach to Codeswitching and Transfer.
In: Heller, Monica, ed. Codeswitching - Anthropological and Sociolinguistic Perspectives.
Berlin, New York, Amsterdam, 187-213.
Auer,
Peter, ed. 1998. Code-Switching in Conversation. Language, interaction and
identity. London, New York.
Auer,
Peter. 2000. Türkenslang - Ein Ethnolekt des Deutschen und seine möglichen
linguistischen Auswirkungen. Ms., Colloquium "Sprache und Lebensalter".
Zürich.
Bernstein, Basil. 1971. Class, Codes and Control. Volume
1. Theoretical Studies toward a Sociology of Education. London.
Bloomfield, Leonard. 1927. Literate and Illiterate
Speech. In: American Speech, vol. 2, 432-439 (reprinted in: Hymes,
Dell, ed. 1964. Language in Culture and Society - A Reader in Linguistics
and Anthropology. New York).
Bommes,
Michael; Radtke,
Frank-Olaf. 1993. Institutionalisierte Diskriminierung von Migrantenkindern.
Zur Herstellung ethnischer Differenz in der Schule. In: Zeitschrift
für Pädagogik, vol. 39, 483-497.
Bußmann, Hadumod. 1983/1990. Lexikon der Sprachwissenschaft.
Stuttgart.
Clarke, John. 1976. Style. In: Hall, Stuart; Jefferson,
Tony, eds. Resistance through Rituals - Youth Subcultures in Post-War Britain.
London.
Crystal,
David. 1987. The Cambridge Encyclopedia of Language. Cambridge.
Cummins,
Jim. 1976. The Influence of Bilingualism on Cognitive Growth - A Synthesis
of Research Findings and Explanatory Hypotheses. In: Working Papers in Bilingualism,
vol. 9, 1-43.
Cummins,
Jim. Linguistic Interdependence and the Educational Development of Bilingual Children. In: Review of Educational Research, vol. 49, 221-251.
Cummins,
Jim. 1994. Semilingualism. In: Encyclopedia of Language and Linguistics (2nd Edition). Oxford.
Dittmar,
Norbert. 1976. A Critical Survey of Sociolinguistics - Theory and Applications. New York.
Durán,
Richard, ed. 1981. Latino Language and Communication Behaviour. Norwood, NJ.
Eastman, Carol, ed. 1992. Codeswitching. In: Journal
of Multilingual and Multicultural Development, vol. 13 (1/2), 1-17.
Gogolin, Ingrid. 2001. Sprachenvielfalt durch Zuwanderung - Ein verschenkter Reichtum in der (Arbeits-)Welt? Paper presented at the workshop "Förderung von MigrantInnen in der beruflichen Bildung durch sprachbezogene Angebote" at the GPC - Good Practice Center Bonn, 24./25. Sept. 2001. (Also available under http://www.ingrid-gogolin.de/files/1_Gogolin.pdf)
Gomolla, Mechthild; Radtke, Frank-Olaf. 2002. Institutionelle Diskriminierung - Die Herstellung ethnischer Differenz in der Schule. Opladen.
Grosjean,
Francois. 1983. Life with two languages. Cambridge, MA.
Gumperz,
John J. 1982. Discourse Strategies, Cambridge.
Hansegård, Nils Erik. 1968. Tvåspråkighet eller halvspråkighet? Stockholm.
Hinnenkamp,
Volker. 1990. "Gastarbeiterlinguistik" und die Ethnisierung der Gastarbeiter. In: Dittrich, Eckhard; Radtke,
Frank-Olaf, eds. Ethnizität - Wissenschaft und Minderheiten. Opladen, 277-298.
Hinnenkamp,
Volker. 2000. "Gemischt sprechen" von Migrantenjugendlichen als Ausdruck ihrer Identität. In: Der
Deutschunterricht, vol. 5, 96-107.
Hinnenkamp,
Volker. 2003. Mixed Language Varieties of Migrant Adolescents and the Discourse of Hybridity. In: Jørgensen, J. Normann, ed. Bilingualism and Social Relations - Turkish Speakers in North-west Europe (Journal
of Multicultural and Multilingual Development), 12-41.
Hinnenkamp,
Volker. 2004. "Zwei zu bir miydi?" - Mischsprachliche Varietäten von Migrantenjugendlichen im Hybriditätsdiskurs. In: Hinnenkamp,
Volker; Meng,
Katharina, eds. Sprachgrenzen überspringen - Sprachliche Hybridität und polykulturelles Selbstverständnis. Tübingen.
Hinnenkamp,
Volker; Meng,
Katharina, eds. 2004. Sprachgrenzen überspringen - Sprachliche Hybridität und polykulturelles Selbstverständnis. Tübingen.
Hymes, Dell H. 1962. The Ethnography of Speaking. In: Gladwin, Thomas; Sturtevant, William C., eds. Anthropology and Human Behavior. Washington, DC, 13-53.
Kallmeyer,
Werner; Tandogan-Weidenhammer, Deniz; Keim, Inken. 2000.
Deutsch-Türkisches - Sprache und kommunikativer Stil von Migranten. In:
Sprachreport,
vol. 3, 2-8.
Keim,
Inken; Androutsopoulos,
Jannis. 2000. Hey Lan, isch geb dir konkret Handy - Deutsch-türkische Mischsprache und Deutsch mit ausländischem Akzent. Wie Sprechweisen der Straße durch die Medien populär werden. In: Frankfurter Allgemeine Zeitung (FAZ), 26. Januar 2000.
Labov, William. 1972. Language in the Inner City - Studies in the Black English Veracular. Philadelphia.
Lemper, Lothar Theodor. 2002. PISA und Migration - Paper delivered at the "7. Forum Migration" of the Otto Benecke Foundation at the Wissenschaftszentrum Bonn, 26. Sept. 2002. Bonn. (also available under www.obs-ev.de/pdf/PISAundMigration.pdf)
Le Page, Robert B.; Tabouret-Keller, Andrée. 1985. Acts of Identity. Cambridge.
MacSwan, Jeff. 2000. The Threshold Hypothesis, Semilingualism, and Other Contributions to a Deficit View of Linguistic Minorities. In: Hispanic
Journal of Behavioral Sciences, vol. 22 (1), 3-45.
Mahlstedt, Susanne. 1996. Zweisprachigkeitserziehung in gemischtsprachlichen Familien. Bern.
Mercer, Kobena. 1988. Diaspora Cultures and the Dialogic Imagination - The Aesthetics of Black Independent Film in Britain. In: Cham, Mbye B.; Andrade-Watkins, Claire, eds. Blackframes - Critical Perspectives on Black Independent Cinema. Cambridge, Mass., 50-61.
Pfaff,
Carol W.. 2000. Development and Use of et- and yap- by Turkish/German Bilingual Children. In: Göksel, Asli; Kerslake, Celia, eds. Studies on Turkish and Turkic Languages. Wiesbaden, 365-373.
Pfaff,
Carol W. 2004. "Kanaken in Alemannistan" - Feridun Zaimoglu's Representation of Migrant Language. In: Hinnenkamp,
Volker; Meng,
Katharina, eds. Sprachgrenzen überspringen - Sprachliche Hybridität und polykulturelles Selbstverständnis. Tübingen.
Picht,
Georg. 1964. Die deutsche Bildungskatastrophe. Freiburg.
Rampton, Ben. 1995. Crossing - Language and Ethnicity among Adolescents. London, New York.
Romaine, Suzanne. 1995. Bilingualism. 2. ed. Oxford.
Sebba, Mark. 1993. London Jamaican. London, New York.
Sebba, Mark; Wootton,
Anthony. 1998. We, They and Identity - Sequential Versus Identity-Related Explanation in Code-Switching. In: Auer,
Peter, ed. Code-Switching in Conversation - Language, Interaction and Identity. London, New York, 262-289.
Skutnabb-Kangas, Tove. 1981. Bilingualism or not? The Education of Minorities. Clevedon, UK.
Swigart, Leigh. 1992. Two Codes or One? - The Insider's View and the Description of Codeswitching in Dakar. In: Eastman,
Carol, ed. 1992. Codeswitching (Journal
of Multilingual and Multicultural Development), vol. 13 (1/2), 83-102.
Terkessidis, Mark. 2000. Migranten. Hamburg.
Wandruszka, Mario. 1979. Die Mehrsprachigkeit des Menschen. München.
Wei, Li. ed. 2000. The Bilingualism Reader. London.
Zaimoglu, Feridun. 1995. Kanak Sprak - 24 Misstöne vom Rande der Gesellschaft. Hamburg.
Keywords: identity, migration, migrant, migrant children, migrant pupils, school failure, bilingualism, semilingualism, multilingualism, polyphonia, switching, mixing, normativity, stylisation, ethnography, Turkish migrants, migrant language use, legitimate language, non-legitimate language, foreigner, conversation, hybrid language, language play, language performance, Pidgin-German, Gastarbeiterdeutsch, majority discourse
