Language processing in speakers who acquire a second language under submersion conditions: Novel perspectives for psycholinguistic research on language processing
Prof. Dr. Eva Belke
Funded by Volkswagen Foundation, Dilthey Fellowship
Most of the psycholinguistic theories of bilingual language processing are based
on data from speakers who have acquired a second language late in life. In most
cases, they have acquired their second language through dedicated foreign
language learning programmes in their home country. Typically, such schooling
programmes feature immersion conditions, that is, language teaching that adapts
to the learners' needs. Critically, research in bilingual language processing
largely overlooks the sizable proportion of the population in Europe and
elsewhere in the world, who acquires a second language earlier in life and who
does so under submersion conditions. Such conditions typically apply to children
who must unavoidably attend school in the majority language, which differs from
their families' language. In Germany, approximately 25% of the children acquire
German as a second language under such conditions. They typically have more
intensive contact with the German language for the first time when they join
Kindergarten. Starting primary school, their German proficiency is often less
advanced than that of their monolingual peers. Nevertheless, German is the
teaching language at school, if only for pragmatic reasons - the children with
German as a second language often have diverse family languages, establishing up
to ten or more different family languages in a single classroom alone.
Even though many schools invest considerable effort in promoting the German
proficiency of children with German as a second language these efforts are
typically restricted to tutorial programmes offered in addition to regular
classroom teaching. To date, these programmes have failed to counteract the
tendency that children from families with German as a second language often fare
worse in international schooling assessments, such as PISA, than their
monolingual peers. One possible reason for this is that existing classroom
concepts for teaching German at primary schools often do not take into account
the multilingual reality present in the classrooms. Instead, all children are
typically taught with materials developed for German children, which poses
particular linguistic challenges to multilingual children, often appropriately
referred to as "sink-or-swim" situations.
Until now, experimental psycholinguistic research has neglected these children
and their language biographies. In addition, there is little contribution of
experimental psycholinguistics to date to the development and evaluation of
programmes for teaching in a second language in submersion contexts.
The first objective of this research project is to establish the group of
speakers, who have acquired a second language under submersion conditions, as a
relevant speaker group to psycholinguists investigating bilingual language
processing. To this end, we have consulted existing theories of bilingual
language processing in speakers of a foreign language, acquired late in life,
and derived hypotheses about what the implications of these models are for
speakers of a second language acquired earlier in life and under submersion
conditions. We focus on lexical processing, which is the research area in
psycholinguistic research on bilingualism that has been investigated most
intensively. Our aim is to establish in what way the processing strategies the
children we test in our experiments differ from those employed by foreign
The second objective is to evaluate didactic concepts used for teaching German
in mixed-language and mixed-proficiency classrooms. In this line of research, we
have collected a large longitudinal corpus of written texts from children taught
with different didactic approaches, comparing, across the board, their
linguistic development over time (from part 2 of grade 2 to the end of grade 4).
The data are currently being analysed.
A third objective of the project is to use experimental psycholinguistic methods
to establish the efficiency of teaching methods for learners of German as a
second language. To this end, we focus in particular on the acquisition of
gender-like grammatical subclasses. German is a gender-marking language, and its
nouns are notorious for providing very little semantic or phonological cues as
to the gender class they belong to. They appear to be largely arbitrarily
assigned to gender subclasses. In this section of the research project, we
examine what mechanisms help support the acquisition of gender-like subclasses
in an artificial language learning paradigm. We are particularly interested in
finding out how the exposure to linguistic input can be optimized in such a way
that the assignment of nouns that are phonologically unmarked for gender can be
acquired more easily.
Bebout, J. & Belke, E. (in press) Language play facilitates language learning:
Optimizing the input for rapid gender-like category induction. Cognitive
research: Principles and Implications.
Frieg, H. & Bebout, J. (in press). Strukturierte Sprachvermittlung mit der
Generativen Textproduktion. Praxis Fördern.
Kauffeldt, L., Kirschke, C., Bebout, J., Frieg, H., Belke, E., Hoffmann, R. &
Belke, G. (2014). Dschungeltanz und Monsterboogie: Singen und Spielen mit
Sprache [Dschungeltanz and Monsterboogie: Singing and Playing with Language].
Baltmannsweiler: Schneider Hohengehren.
(Songbook, EUR 16,00)
Frieg, H., Hilbert, C., & Belke, E. (2013). Sprachförderung bei einem Jungen
mit Deutsch als Zweitsprache: Wie erfolgreich sind implizite Verfahren?
[Language training in German as a second language: How effective are implicit
methods?]. Praxis Sprache, 1, 7-17.
von Lehmden, F., Kauffeldt, J., Belke, E., & Rohlfing, K. (2013). Das Vorlesen
von Kinderbüchern als implizites Mittel zur Sprachförderung im Bereich
Grammatik [Reading to children promotes implicit grammar acquisition]. Praxis
Sprache, 1, 18-27.
Frieg, H., Hilbert, C., Belke, E., & Belke, G. (2012). Die generative
Textproduktion [Generative text production]. Sprachheilarbeit, 57, 155-161.
Frieg, H., Stielow, A., Kitzinger, M., & Belke, E. (2012). AltübAsD: Ein
Verfahren zur altersübergreifenden Analyse schriftsprachlicher Daten in der
Grundschule [AltübAsD: Assessment of written text production in primary school
children of all grades]. DaZ, 4/2012, 9-24.