EDITORS: Köpke, Barbara; Schmid, Monika S.; Keijzer, Merel; Doster, Susan TITLE:
Language Attrition SUBTITLE: Theoretical Perspectives SERIES: Studies in
Bilingualism 33 PUBLISHER: John Benjamins YEAR: 2007 Julia Deák, Graduate
School of Education, University of Pennsylvania SUMMARY
This book follows up on the 2004 volume on first language (L1) attrition edited
by three of the same authors (Schmid, Köpke, Keijzer & Weilemar 2004). The focus
of this volume is on theory, and the 14 chapters cover a wide range of
perspectives and methodologies. As the introductory chapters make clear, there
are cognitive and neuropsychological perspectives to explore as well as social,
emotional, and life circumstance perspectives that look to patterns of language
use and language attitudes to explain attrition. The first three chapters, by
Köpke, Sharwood Smith, and de Bot, are the most overarchingly theoretical.
Köpke's chapter argues for a multicomponential view of attrition in which no one
factor can be expected to explain the incidence of attrition. She first
considers four brain mechanisms. The first is plasticity associated with age
effects, which explains the higher incidence of attrition among young immigrants
versus older ones. Second, she mentions activation thresholds (also discussed in
Paradis' chapter) which may render some existing information in the brain
irretrievable after long periods of disuse. Next she discusses the inhibitory
effects of neurons which might explain temporary L1 retrieval difficulties in
people actively trying to learn L2 (second language) as well as difficulty
inhibiting a very active L2 when trying to access a long-dormant L1. She also
discusses emotions as a brain mechanism, suggesting that in cases of early
trauma associated with an L1, or a high emotional investment in L2, the activity
of subcortical structures that are associated with emotion could cause attrition
effects. Köpke's main point is that these many factors interact, and they should
be studied together whenever possible. Studying attrition as an individual
neurological or cognitive phenomenon ignores important variation determined by
social or circumstantial factors. Although Köpke identifies some clusters of
factors, such as those related to age, or the linguistic and cultural
environment, she concludes that no cluster or individual factor is dominant, and
in fact the weight of influence of each may depend on the context, for example
in cases of low or high language proficiency. Aligning herself with Cook's
(1992) multicompetence model, she concludes that attrition should not be studied
as if it were an unusual condition with specific symptoms and one clear cause,
but rather as another dimension of variation in multilingual individuals,
stemming from multiple causes, and often ''remaining within the range of
perceived native-like proficiency'' (p. 31). Sharwood Smith's chapter
provides an overview of the Modular Growth and Use of Language (MOGUL) theory,
making the point that development includes both acquisition and attrition. MOGUL
is a way to conceptualize both acquisition and attrition by processing. It is
based on a theory of language in the brain which posits that linguistic forms in
the mind come into existence through the repeated processing of linguistic input
by phonological, syntactic, and conceptual-semantic processors (Jackendoff
2002). Input is not directly translated into forms as in a pure connectionist
model, since the architecture of the processors, perhaps influenced by Universal
Grammar (UG), constrain how structures are built or activated. The theory thus
includes concepts from both connectionist and UG approaches. Competition is also
invoked, as several existing structures compete for selection during both
comprehension and production. MOGUL adds to this theory by positing that growth
or decline in language comprehension and production can be understood as changes
in the long-term memory store of forms, or as changes in the accessibility and
use of these forms. De Bot's chapter argues that research on the ''how'' and
''why'' of language attrition could benefit from concepts from lifespan
developmental psychology and dynamic systems theory. In developmental
psychology, there is more and more work on the later stages of life, beyond
those most often studied by language researchers. Also, psychologists consider
the importance of major life events in development, and de Bot encourages
attrition researchers to look at the effects of ''language-related major life
events,'' such as entry into bilingual schooling, contact with speakers of other
languages, study abroad, migration, etc. Regarding Dynamic Systems Theory (DST),
De Bot emphasizes that language should be seen as an ever-changing system in
which growth and decline are two outcomes of the same process, which depends on
input and internal reorganization. The next three chapters draw the most from
theoretical linguistics. Myers-Scotton studies language shift in Xhosa-English
bilinguals, and tries to show that the shift to English grammar is abrupt and
not gradual. She does this by exploring whether there is a stage in which the
speaker uses ''critical grammatical morphemes'' from English in Xhosa phrases or
clauses. The hypothesis is that the speakers move from inserting English lexical
items directly into Xhosa to producing full English clauses. Myers-Scotton uses
cluster analysis to divide her sample into three groups which represent
different stages of shift; however, the groups overlap in terms of how many
English clauses they produced. She also finds support for her Abruptness
Hypothesis because critical grammatical morphemes from Xhosa are retained,
except in ''embedded English phrases or EL islands'' (p. 81). Tsimpli explores
which areas of L1 grammar are vulnerable to attrition under a theoretical model
based on the Minimalist Program (Chomsky 1995). Hers is an empirical study based
on offline data from one group (N=19), which learned an L2 to near-native levels
after childhood, and online data from another group (N=4), consisting of
immigrants' children who learned two languages during childhood. Her first
experiment shows evidence that a completely learned L1 can be influenced by a
strong L2 learned after childhood; the Greek-English bilinguals showed a ''change
in the direction of preference for postverbal subjects'' but still produced
postverbal subjects, showing that their L1 settings were not completely changed
by their L2 (p. 90). The second experiment showed that the bilingual young
adults had trouble judging the grammaticality of Greek determiners. Both
experiments were used to argue that attrition affects interface properties but
not pure syntactic L1 options. Gürel attempts to show transfer from L2 Turkish
to L1 English in terms of the binding of the overt pronoun in a set-theoretic
transfer model. Her 2002 paper concerned the reverse case of transfer from L2
English to L1 Turkish. In that case, the L1 binding options for the Turkish
pronoun o were loosened in accordance with the L2 English settings for him/her.
For the present experiments, Gürel expected the English L1 settings for possible
antecedents for himself/herself to be loosened towards the less restrictive
Turkish L2 settings for kendisi. The experiments reported in this chapter
yielded no significant findings, which prompted Gürel to invoke an explanation
based on frequency of use of the English L1 of the current group as contrasted
with the Turkish L1 of the prior group. Paradis' chapter on neurolinguistic
theories related to language attrition stands alone as a theoretical piece that
is actually cited by many of the other chapters. Paradis uses his
neurolinguistic theory of bilingualism (2004) to predict how frequency of use,
declarative vs. procedural storage of knowledge, and motivation impact the
selective attrition of some parts of language before others. The next
chapter, by Schmid, investigates the relationship between language use and
attrition in German-born immigrants in Canada and the Netherlands (N=106) who
had immigrated at least 10 years earlier (most immigrated about 35 years
earlier). She finds evidence of some attrition, but there is great individual
variation and no straightforward relationship between use and attrition.
Pallier discusses the critical period hypothesis and examines data from a group
of Korean adoptees in France, who were adopted at ages 3-10 and learned French
to nativelike proficiency. The adoptees performed similarly to French
monolinguals on recognition tests of Korean sentences, words, speech segments,
and phonemes, showing that one's first language is not permanently imprinted in
infancy, and a new first language can be learned up to age 10 in some subjects
if contact with the original language is completely severed. Footnick's
chapter investigates the first language knowledge of a French subject whose
contact with his family's African language was not completely severed, though he
claims to have undergone almost complete attrition. Footnick was able to help
her subject CK recover his ability to understand and speak Mina, a local
Togolese variant of Ewe, through hypnosis. Footnick argues that hypnosis reduces
activity in the anterior cingulated cortex, an area associated with pain and
conflict, which is why it can reverse the ''active forgetting'' that may be the
cause of attrition for a language associated with painful childhood memories,
though it is unclear whether her subject's experience was actually painful. The
final three chapters seem to address the question of how attrition actually
affects people. Prescher investigates the transcultural identity of immigrants
and their perceptions of attrition. She interviewed 20 German-born individuals
who immigranted to the Netherlands at least 10 years prior. All informants
reported experiencing some L1 attrition, and Prescher finds support for
Yoshizawa Meaders' (1997) model of transcultural identity development, which
includes an initial period of assimilation followed by a struggle to define the
transcultural identity and a return to identification with the original culture.
Ben-Rafael and Schmid also look at the connection between attitudes,
motivation and emotion on the one hand and language attrition on the other. The
authors compare two groups of immigrants who vary in their motivations for
immigrating and learning Hebrew: Francophone immigrants who came to Israel for
ideological reasons and Russian speakers who left home more for pragmatic
reasons. Unfortunately, the groups also differ in length of residence in Israel:
45 years on average for the Francophones and 14 years on average for the
Russians. The Francophones were found to code-switch more and to use borrowed
grammatical items as well as individual words in their L1 speech, indicating a
possible link between motivation and attrition of L1. In the final chapter of
the book, Jiménez writes about the method of using stimulated recall to probe
heritage language speakers' use of compensatory communicative strategies. He
gave heritage speakers of Spanish a narrative task, videotaped their monologues,
quickly transcribed their speech, and then played the video back for them,
asking them to explain what they were thinking when they paused, used
circumlocutions, or made various errors. His results give insight into the types
of production problems that attriters or incomplete acquirers of a language
face, as well as showing compensatory strategies for these problems in action.
EVALUATION
This book successfully presents many of the theories and methodologies currently
used to study language attrition, but it also claims to encourage an integration
of the many methods and viewpoints. While it makes clear what the various
dimensions of the phenomenon are likely to be, it does not really set a standard
for future research or argue convincingly for a comprehensive theoretical
framework that future researchers can use. While the book contains an
interesting collection of research papers, the empirical studies appear to
ignore the theoretical directions argued for in the early chapters. Köpke's plea
for an integrated, comprehensive approach is not answered by any of the
empirical papers. For example, Sharwood Smith's MOGUL theory is introduced in
his chapter, but then does not appear in any of the subsequent chapters. Most
chapters, perhaps because of their limited length, explore only one aspect of
language acquisition, either linguistic, psychological, neurological,
sociolinguistic or emotional. The fact that a few of the chapters presented
inconclusive results underscores the importance of a more comprehensive model
that includes all possible influencing factors at once. Nevertheless, the
collection of diverse works in this volume gives a wide snapshot of ideas about
language attrition today. For one, the book includes data on what appear to be
different types of attrition, from slight variation or code-switching patterns
still within the category of native or near-native competence, to transfer from
one known language to another, to active forgetting of a language associated
with trauma. Moving away from a view that attrition is only a pathological
occurrence, language attrition is reconceptualized as a natural and universal
part of multilingualism. The processing-based accounts of language development
in the book can easily accommodate predictions of attrition as well, connecting
attrition with a wide body of research on acquisition and multilingualism. It is
also notable that the connection with language attitudes, long-standing in the
language shift literature, is here brought to bear on studies of individual
speakers as well. In short, the volume is a broad introduction to theory and
research in various language-related fields, and it sets the stage for a broader
discussion of language attrition in years to come. REFERENCES
Cook, V.J. (1992). Evidence for multi-competence. _Language Learning_, 42, 4,
557-591. Jackendoff, R. (2002). _Foundations of language: Brain, meaning,
grammar, evolution_. New York: Oxford UP. Paradis, M. (2004). _A
Neurolinguistic Theory of Bilingualism_. Amsterdam: John Benjamins. Schmid,
M.S., Köpke, B., Keijzer, M. & Weilemar, L. (Eds.) (2004). _First Language
Attrition: Interdisciplinary Perspectives on Methodological Issues_. Amsterdam:
John Benjamins. Yoshizawa Meaders, N. (1997). The transcultural self. In P.H.
Elovitz & C. Kahn (Eds.), _Immigrant experiences: Personal narrative and
psychological analysis_ (pp. 47-59). Madison, WI: Fairleigh Dickinson University
Press. ABOUT THE REVIEWER
Julia Deák is a PhD Candidate in Educational Linguistics at the University of
Pennsylvania, Graduate School of Education. She studies second language
development using corpus methods and is interested in language attrition and
re-learning among college-aged heritage language learners.
|
