Language Attrition

Chapter (PDF Available) · March 2018with 4,745 Reads 
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
Cite this publication
The TESOL Encyclopedia of English Language Teaching, First Edition.
Edited by John I. Liontas (Project Editor: Margo DelliCarpini).
© 2018 John Wiley & Sons, Inc. Published 2018 by John Wiley & Sons, Inc.
DOI: 10.1002/9781118784235.eelt0843
Language Attrition
The Intellectual andSocial Context ofLanguage Attrition
Language attrition refers to a gradual reduction or loss of linguistic knowledge and
skills in an individual. It is defined as “the non-pathological decrease in proficiency
in a language that had previously been acquired by an individual” (Kopke &
Schmid, 2004, p. 3). This definition accurately captures the meaning of the term as
it is currently used in the field, and the use of the word “non-pathological” illus-
trates that the decline in proficiency is caused not by illness nor by deterioration of
or damage to the brain, but rather by a change in one’s contact with the language(s)
in question. This definition also makes it clear that the attrition phenomenon takes
place in an individual, rather than in groups or speech communities.
One way to understand language attrition is in terms of language acquisition,
which can be described as the process by which one’s proficiency in a first language
(L1) or second language (L2) increases. In the case of language attrition, lack of
contact with a language leads to reduced levels of proficiency in the attriting
language. The person undergoing attrition, the attriter, is typically a bilingual or a
multilingual individual whose L1 is being replaced by an L2, or whose L2 is being
replaced by the L1. It is of course possible for monolingual speakers to lose their
first and only language in an L1 environment; this can happen as part of normal
aging processes, or as part of an abnormal or pathological case of language dete-
rioration, such as aphasia or agnosia. However, language attrition involving loss
of the L1 in an L1 environment will not be included in this entry.
Aside from cases of L1 being lost in an L1 environment, there are other circum-
stances or conditions in which language attrition naturally occurs. These conditions
involve a change in linguistic environment that necessitates contact with another
language. A typical example would be the case of a family with a young child moving
to a foreign country, where the child quickly starts to acquire the new language spo-
ken in the environment. Given time, the child’s native language will slowly be
replaced by the ambient language input, which eventually becomes his or her domi-
nant language. In this case, the resulting outcome is L1 attrition in an L2 environ-
ment, where the attriting language is the child’s mother tongue. This situation
v1_lbp-L.indd 1 10/26/2017 8:11:17 PM
Language Attrition
includes language attrition in migrants exposed to a foreign language in a new host
country. While learning and using their new language, these individuals have limited
or no exposure to their L1, which results in the decline of their L1 knowledge and
skills. This decline can manifest itself as lexical access difficulty, disfluency, and
increased optionality in grammatical judgments, among other aspects. Second
language attrition, on the other hand, concerns L2 learners’ loss or maintenance of
their L2 after they have returned to their homeland, or after losing contact with the
L2. The resulting outcome is L2 attrition in an L1 environment. Thus, language attri-
tion can affect the mother tongue as well as any additional languages learned, and
the attrition literature often makes a distinction between L1 and L2 attrition. Whether
the attriting language is the L1 or L2, inevitable similarities exist between the two
types, and the majority of issues discussed here are relevant to both L1 and L2 attri-
tion. Given the nature of the current volume, however, more emphasis will be given
to L2 attrition. Any differences between the two types will be noted as appropriate.
Changes intheTopic andTreatment ofLanguage Attrition
As previously defined, language attrition refers to non-pathological loss of lin-
guistic knowledge and skills in an individual. Currently, language attrition is
widely regarded as an independent field of research subsumed under bilingual-
ism and second language acquisition (de Bot & Weltens, 1995; Schmid & Kopke,
2011). Traditionally, however, language attrition was regarded as part of the long-
established sociolinguistic practice of research on language contact, language shift,
and language death. Language shift involves a gradual decline in the use of and
competence in a language over generations, which in the most extreme cases can
result in language death within a community. The broader traditional scope of lan-
guage attrition in earlier literature is reflected by the fact that collections on the
topic of language attrition published before the 1980s included studies pertaining
primarily to intergenerational change (change between generations) in minority
languages, along with studies of aphasia.
The scope of language attrition began to shift with the holding of the first con-
ference on the subject at the University of Pennsylvania in 1980, which was specifi-
cally devoted to the topic of “Loss of Language Skills.” The papers presented at the
conference (published as conference proceedings in Lambert & Freed, 1982)
marked the beginning of attrition research as a recognized subfield of second
language acquisition that emphasized language attrition as non-pathological,
intragenerational loss of language in individuals over time. The 1980 conference
also provided a launching point for abundant research in the field of language
attrition. These changes in paradigm and pertinent research findings (see, for
example, Bahrick, 1984) began a process whereby language attrition came to be
seen as an individual, psycholinguistic phenomenon. Following Bahrick (1984),
scholars in Europe, namely Kees de Bot, Bert Weltens, and colleagues, advanced
the field with subsequent research, mostly focusing on the attrition of French as an
L2. Currently, both language attrition and language shift are deemed hyponyms of
v1_lbp-L.indd 2 10/26/2017 8:11:17 PM
Language Attrition 3
the umbrella term language loss, which refers to the decline of linguistic skills in
both communities and individuals (de Bot & Weltens, 1995; Hansen, 2001). Along
with language shift, a number of other terms have been introduced to refer to
concepts related to decreasing proficiency in one’s language, including language
regression, language death, code-switching, and code-mixing, to name a few.
While all of these terms refer to the gradual decline of language knowledge in
individuals or groups, each term represents a different focus.
Major Dimensions ofLanguage Attrition
Language attrition is a highly complex phenomenon, and several models and
theories have been proposed to account for its processes. These models and
hypotheses were originally proposed for L1 attrition and subsequently applied to
L2 attrition research.
Models andHypotheses
One of the earliest models introduced in language attrition research is the regres-
sion hypothesis, which concerns the order or sequence of attrition. Also known as
the “last in, first out” model, this hypothesis predicts that language will be lost in
the reverse order in which it was acquired. This means that the aspects of language
acquired by learners early on will be most resistant to loss, whereas those aspects
of language that were acquired last will be most susceptible to loss. Originally
developed in connection with research on aphasia, the regression hypothesis has
been investigated in both pathological and non-pathological contexts, yielding
conflicting findings; it seems to describe the course of attrition for some features,
but not others. This is understandable in that when an L1 attrites under the influ-
ence of an L2 or vice versa, many additional factors come into play, including the
attriter’s age, proficiency level, and literacy level, and typological features of the
L1 and L2, among others. Additionally, any study testing the regression hypothesis
should establish the acquisition sequence of the language in question to use as a
comparison point for the attrition sequence, which is not always easy to demon-
strate (Bardovi-Harlig & Stringer, 2010). Thus far, there is little empirical support
for this hypothesis.
A related proposal that has received considerable attention in both L1 and L2
attrition research is the critical threshold hypothesis. This hypothesis proposes
that the language knowledge least vulnerable to loss is not what is learned first,
but what is learned best (i.e., “best learned, last out”), and that when a certain
threshold of use is achieved, knowledge may be less susceptible to attrition. This
hypothesis is based on Neisser’s (1984) idea of a critical threshold, a point at which
a learner has a stable enough mental representation of some linguistic elements or
structures that they become resilient to loss. Bahrick’s (1984) research on language
retention in L2 learners, whose knowledge of L2-Spanish remained intact after 25
years of non-use, is often cited as evidence supporting the hypothesis. Another
v1_lbp-L.indd 3 10/26/2017 8:11:17 PM
Language Attrition
variant of the hypothesis predicts that once learners have achieved a certain
level of proficiency, knowledge of language is more easily retained. Here, an
individual’s pre-attrition attainment is regarded as the deciding factor for
attrition and maintenance, in that a certain level of proficiency may critically
determine the loss or non-loss of a language.
The next hypothesis of note concerns crosslinguistic influences, deemed to be
some of the most significant influences in both L2 acquisition and L1/L2 attrition.
Also referred to by some researchers as the interference hypothesis or the interlan-
guage hypothesis, this model predicts that attrition is caused by the newly dominant
(i.e., replacing) language. Findings have shown that structural properties in which
the attriting language differ quite markedly from the replacing language showed the
greatest attrition (e.g., Berman & Olshtain, 1983) and that typological universals
affect interference from the replacing language in the use of the attriting language
(Olshtain, 1989). The existence of crosslinguistic effects in both L1 and L2 attrition is
undeniable, and these effects seem to interact selectively with the attrition process.
In this respect, this hypothesis has often been called upon to interpret the results of
attrition data, as opposed to the hypothesis itself being tested empirically.
The final model introduced here is the dormant language hypothesis, which is
concerns with the general psycholinguistic issues of processing and memory. This
model differs from other proposed hypotheses in that it concerns the issue of what
happens to the attriter’s knowledge of language at the end of the attrition process.
More specifically, it questions whether the endpoint of attrition is the complete loss
of particular kinds of linguistic knowledge, or whether traces continue to remain in
the attriter’s mind. In the latter case, the problem thus involves difficulty of access
rather than a complete erasure of knowledge from the attriter’s mind. This hypoth-
esis is informed by the concept of “savings,” which has its origins in the widely
held psychological view of forgetting: information from memory merely becomes
inaccessible due to different factors, and can be retrieved again in the presence of
the right cues. This concept has been applied to L2 studies on relearning based on
the observation that relearning a language takes less time than learning it for the
first time, which implies access to unconscious linguistic knowledge rather than
complete loss. Proponents of the dormant language hypothesis maintain that
knowledge seemingly lost can be reactivated in a situation of relearning under
appropriate conditions and with the proper triggers. This hypothesis has received
empirical support from studies on relearning of forgotten words (e.g., Hansen,
2011). In addition, findings from studies using a relatively unconventional method
of hypnosis (e.g., Footnick, 2007) have shown that knowledge of a forgotten child-
hood language can be recalled using age-regression hypnosis. The question of
whether forgotten knowledge is permanently inaccessible or just difficult to access
has become a major point of contention—an issue that will be revisited later.
Content ofAttrition: What Gets Lost
One of the main topics in language attrition research pertains to the content of
attrition—that is, what is actually lost, or which components of the language
v1_lbp-L.indd 4 10/26/2017 8:11:18 PM
Language Attrition 5
system are most prone to attrition. Many studies concerned with the content of
language attrition have examined language skills rather than linguistic knowl-
edge. Findings have generally shown that receptive skills (i.e., reading and listen-
ing) are better retained than productive skills (i.e., speaking and writing), which
tend to be more vulnerable to attrition. In particular, oral fluency is known to be
highly susceptible to attrition, and it is also generally viewed as a precursor to the
onset of language attrition. Mirroring these findings, similar results have been
reported for receptive vocabulary and grammar, both of which tend to be better
retained than productive vocabulary and grammar.
Retention of the lexicon has been investigated to a greater extent than any other
area. In both L1 and L2 attrition research, the lexicon has been found to be most
prone to attrition, attriting quickly and more severely than other areas, such as
grammar or phonetics. The most common manifestations of lexical attrition
include lexical access difficulties and forgetting of unused vocabulary. Findings
have shown that words which are less frequently occurring or longer in length
tend to be more vulnerable to attrition. On the other hand, certain types of lexical
entries, such as formulas, conventional expressions, idioms, and social fillers, have
been found to be better retained than others (see Berman & Olshtain, 1983). Limited
research has also been conducted on morphology and syntax, but further research
is clearly needed, especially concerning the attrition of phonological, pragmatic,
and discourse skills.
Variables Affecting Language Attrition
Given the complex and multifaceted nature of language attrition, different fac-
tors should be taken into account in understanding the attrition phenomenon.
An important distinction was made early on between criterion variables and pre-
dictor variables affecting language attrition (Lambert & Freed, 1982), which have
been reconceptualized as linguistic and extralinguistic variables in the more recent
literature. Linguistic variables pertain mainly to the content and process of lan-
guage attrition. For example, many attempts have been made to describe exactly
which linguistic features are most vulnerable to loss, and how linguistic influ-
ence from the dominant or replacing language plays a role in the attriting lan-
guage, among other issues. Not surprisingly, these factors coincide with the
models and hypotheses proposed to explain the process of attrition, as well as
the linguistic aspects most prone to attrition, both of which have been addressed
previously. Extralinguistic variables, on the other hand, pertain to other factors
that influence the rate and quality of attrition, including the attriter’s age at the
onset of attrition, their attitudes and motivation, and their literacy in the attriting
language, to name a few. Some of the key factors most often discussed in the
literature are examined below.
Age at the onset of attrition is generally recognized as one of the most significant
factors in determining an individual’s vulnerability to language attrition. General
v1_lbp-L.indd 5 10/26/2017 8:11:18 PM
Language Attrition
tendency strongly suggests that the younger the attriter, the quicker and the more
severe the extent of language loss. This can be easily observed in L1 attrition con-
texts: the L1 remains fairly stable for migrants who leave their L1 environment
after puberty, whereas individuals who migrate before puberty can suffer signifi-
cant L1 attrition, both in the rate and extent of language attrition. Pallier’s (2007)
study, for example, reported on young Korean adoptees (ages 3–10) in France, who
lost their L1 completely after being removed from the L1-speaking environment
during childhood. Other studies that have investigated the link between age and
attrition have also demonstrated that the younger the child, the faster the rate
of attrition. However, while age seems to be a significant factor in L1 attrition,
investigating L1 attrition in young children can be quite tricky. In this regard,
attrition researchers have proposed that L1 attrition in children should be strictly
distinguished from L1 attrition in adolescents and adults, suggesting that the
former case could best be explained in terms of incomplete acquisition rather than
attrition. These researchers suggest that incomplete acquisition implies that some
grammatical aspects of the language are still in the process of being acquired (i.e.,
have yet to reach a stable endpoint); attrition, on the other hand, should be strictly
reserved for individuals whose L1 acquisition has been completed (i.e., reached a
stable endpoint) prior to the onset of language attrition. Thus, incomplete acquisi-
tion and attrition are currently viewed as two distinct phenomena that warrant
separate investigations.
Contrary to the rapid attrition of unused language skills reported in the case of
young children, findings have shown that adults are generally able to retain their
language knowledge and skills despite lack of exposure and use. In the case of L2
attrition, even if an adult learner has acquired less L2 than a child learner, the adult
is more likely to retain most of the L2 after returning to the L1 environment. This
means that adult learners, who are typically late starters, may not enjoy the same
ease of acquisition as their younger counterparts, but they are far better at retain-
ing what they have learned. This is well illustrated in Hansen’s (1980) study, which
tracked young participants who were considered to be fluent in L2 Hindi in pre-
school in India. These young participants had lost all traces of their L2 years later,
while their mother, who was less proficient in the L2 than the children when they
attended preschool, showed no decline in her L2 skills. In general, accumulated
findings show that adults who attain high levels of L2 proficiency tend to exhibit
a plateau of several years during which their language skills remain relatively
It should be noted that age inevitably interacts with another variable—the
learner’s level of literacy in the attriting language. Interestingly, ages 8–10 (often
referred to as the critical period of attrition that signals significant decrease in
proficiency) tend to coincide with the point at which children master literacy.
Thus, it may be hypothesized that the younger the learner and the more limited
the learner’s level of literacy, the stronger attrition may be. Whether stabilization
of the linguistic system is caused by brain maturation processes occurring at
that age, or whether it is facilitated in part by literacy skills, is an issue worth
examining further (Hansen, 2001; Kopke & Schmid, 2004).
v1_lbp-L.indd 6 10/26/2017 8:11:18 PM
Language Attrition 7
Pre-attrition attainment
This variable pertains to the peak proficiency level reached by the attriter before the
onset of attrition. Inevitably, the attriter’s pre-attrition attainment level plays a
much greater role in the study of L2 attrition than of L1 attrition, wherein the
process of acquisition is typically completed prior to the onset of attrition. In other
words, the pre-attrition attainment level in L1 attrition remains complete and
uniform, compared to L2 attrition where it serves as an important contributing
variable. Generally speaking, there exists an inverse relationship between the level
of pre-attrition attainment and the extent of attrition: the higher the learners’ L2
attainment, the lower the degree of attrition (see also the critical threshold
hypothesis discussed previously). Indeed, studies have repeatedly shown that the
higher the learner’s proficiency, the better the retention. This tendency has been
found to be most prominent in receptive skills, such as vocabulary recognition and
reading comprehension, which yielded high correlation between attainment and
retention. Hence, many researchers have identified pre-attrition L2 attainment as
the most significant predictor of language loss or retention, and higher attainment
in the L2 is generally deemed a good safeguard against attrition. However, it should
be noted that the pre-attrition attainment variable necessarily correlates with the
age variable, and can also correlate with other extralinguistic variables, including
the length of study or length of residence in the target language environment.
Attitudes andmotivation
Psychological constructs such as learner attitudes and motivation have been proven
to be some of the most important factors for success in L2 acquisition. These factors
also play an important role in language attrition, as they are closely related to issues
of language maintenance and use. Studies have demonstrated the importance of
attitudes and motivation not only in their effect on levels of peak attainment, but
also in their potential to influence learners’ participation in L2 practices or efforts to
seek out opportunities to use the L2, even after limited or no access to the input. It
is not difficult to see how motivation might influence the degree of attrition since
positive or negative emotions toward a language or culture are likely to increase or
decrease the learner’s engagement in that particular language. As with acquisition
studies, however, a number of methodological issues are involved in assessing
these constructs, which are subject to frequent changes. One such issue concerns
how to gauge the attriter’s motivation at certain points in time—for instance, at the
time of migration, or at the onset of attrition. This is further complicated by the fact
that it is difficult to precisely identify the onset of attrition (i.e., when the attrition
process actually starts). As noted, research findings strongly point to the level of
attainment as the best predictor of retention, with attitudes and motivation jointly
contributing as important indirect influences which assist learners in achieving
peak levels of attainment, and subsequently in maintaining the language.
Literacy is a variable that has begun to attract the attention of attrition researchers
in recent years. Well-developed literacy skills are thought to be important in
v1_lbp-L.indd 7 10/26/2017 8:11:18 PM
Language Attrition
helping to attain the critical threshold beyond which language is resistant to attri-
tion. Some research findings show that the level of writing skills attained is a
strong predictor of the learners’ subsequent retention of the language. Findings
have also suggested that older participants who retained their L2 better were also
better equipped with literacy skills in the attriting language (e.g., Olshtain, 1989).
Literacy skills also allow opportunities for further contact with the attriting lan-
guage in the form of reading and writing (e.g., writing letters to friends), which
can be especially helpful in the case of migration. Additionally, and perhaps more
importantly, written input can supply information that may not be as salient in
oral input, and seeing a word in print may help some learners to secure it in mem-
ory, thereby providing an additional mode of interaction with the language.
However, as can be expected, it is often difficult to separate this variable from age,
which necessarily correlates with the level of literacy.
Typological proximity between languages
Language attrition is also influenced by the typological proximity between the two
contact languages. In terms of vocabulary retention, findings have suggested that
L2 cognates were better retained than non-cognates, which were more vulnerable
to attrition. Similarly, some studies have reported a significant correlation between
the rate of lexical learning and loss where typological proximity between the two
interacting languages facilitated the attriters’ retention of vocabulary (e.g., Hansen,
2011). On the other hand, similarity between two languages has often been
assumed to influence erosion of grammatical and phonological skills. Thus far,
investigations have mostly focused on receptive skills in which the impacts of
similarity between the L1 and L2 have been conclusively demonstrated. Research
on other areas and skills should provide a more concrete picture of the effects of
L1–L2 similarity/difference in language attrition.
Manner of instruction
This variable pertains to the presence or absence of instruction during language
acquisition; and in the case of instructed learning contexts, the potential superiority
of certain teaching methods over others in terms of long-term retention. Findings
thus far suggest that teaching methods promoting explicit knowledge are more
likely to result in long-term retention than those promoting implicit methods, such
as immersion. However, because most studies reporting rapid loss of L2 compe-
tence under natural circumstances pertain to young children, age inevitably serves
as a confounding variable. Additionally, the duration of instruction and skills
emphasized in the L2 classroom have been identified as related variables.
Some of the most widely discussed variables affecting language attrition have been
introduced in this section. It should be noted that a number of other extralinguistic
factors have been identified in the attrition literature, which cannot be elaborated
upon due to space constraints. Some of these factors worth mentioning include
“level of education” (in the case of L1 attrition), “length of residence,” “time elapsed
since the onset of attrition,” and “amount of contact with the attriting language.”
v1_lbp-L.indd 8 10/26/2017 8:11:18 PM
Language Attrition 9
Current Emphases
It is perhaps not surprising that the traditional view of language attrition has been
somewhat negative, as losing a language (especially one’s native language) can
invoke a sense of inadequacy in an individual. However, it is becoming widely
recognized that language attrition, whether in a first or a second language, is a
normal process in language contact situations. Thus, further research on the pro-
cess of language attrition is becoming increasingly important, as it can provide
“another window on the dynamism of language, another testing ground for extant
theories of language development” (Hansen, 2001, p. 61). From a more practical
perspective, language attrition research is becoming increasingly important for its
potential to provide language maintenance strategies and curriculum strategies
for migrants returning to or leaving their home countries. Given such a prospect,
it comes as no surprise that the aspects of language attrition receiving attention in
recent literature concern topics such as maintenance and relearning, which have
the potential to inform both researchers and practitioners alike.
In view of the valuable insights that attrition research can offer, a deeper
understanding of the process of language attrition is becoming especially impor-
tant, as it is necessarily related to the maintenance of the target language, as well
as the maintenance of one’s native tongue. This is of special concern for scholars
and practitioners engaged in teaching English as an international language
(TEIL), given the diverse roles and status of English in different contexts around
the world. For example, English is taught as a foreign language in some coun-
tries, whereas it is taught and used as a medium of instruction or a lingua franca
in some others. In the case of countries such as Singapore, Hong Kong, and
Kenya, people use English as an official language in addition to their respective
native languages. In such cases, there may be less risk of L2 attrition since people
speak English on a regular basis at work or school. On the other hand, it is pos-
sible that the dominant use of EIL could potentially threaten the stability of one’s
L1, and along with it one’s national and cultural identities. This is especially
pertinent to situations in which English is used as an international language or a
lingua franca, or where it has established itself as a nativized variety, as in the
case of countries like India. In these contexts, learners’ L2 (EIL) may become the
more dominant language over their L1, which may subsequently decline to a less
dominant language. Given the diverse role of EIL in different countries around
the world, issues related to language maintenance are becoming especially per-
tinent in preserving both the learners’ L1 and L2, or dominant and non-dominant
Another topic related to language maintenance which is receiving attention for
both its theoretical and practical implications, is language relearning, wherein
findings suggest that learning a language a second time can be dramatically more
rapid than the first. Cognitive psychologists generally agree that information once
learned is not completely lost, and that although it may become inaccessible with
discontinued use, it may also be retrievable given the right cues. This means that
residual knowledge exists and can be reactivated, even for a language the attriter
v1_lbp-L.indd 9 10/26/2017 8:11:18 PM
Language Attrition
can no longer recall or recognize (see the dormant language hypothesis discussed
previously). However, some research findings have suggested otherwise. Recall
the findings from Pallier’s (2007) study of Korean children adopted by families in
France, who experienced sudden and complete isolation from their L1 after adop-
tion. The adoptees’ performance in various recognition tasks showed no traces of
their L1, such that these participants treated input from their attrited language in
the same way as input from a novel language. Based on these results, it may be
possible to draw the conclusion that the L1 can be completely lost in the absence
of sustained language input. However, proponents of the dormant language
hypothesis maintain that there may still be traces of the L1 that could potentially
be accessed in the presence of the right trigger. The existence of such traces might
be tested by means of a (re)learning paradigm by examining whether participants
like those in Pallier (2007), who evidently experienced complete loss of their
L1-Korean, would be able to acquire Korean faster and more efficiently than others
if they were to start (re)learning it. This empirical question remains to be tested in
future research, which could yield important implications for those who want to
relearn their lost language.
Whether language attrition reported in some studies is a case of actual loss or
simply a matter of diminished accessibility remains a prevalent topic of conten-
tion. Research findings that show attriters’ receptive skills to be relatively unim-
paired are often regarded as providing support for lack of access rather than lack of
knowledge. Given the ongoing debate, more research is needed to shed light on the
topic; however, several methodological complications are inevitably associated
with attrition research, as discussed below.
Future Directions inResearch andMethodology
While most aspects of language attrition affect both L1 and L2 attrition alike, the
two types differ in many respects when it comes to research design. One of the
major differences between L1 and L2 attrition research stems from the fact that it is
relatively easy to establish a baseline for skills for L1 attrition, but not for L2 attri-
tion. In other words, it can be safely assumed that the adult attriter’s attained L1
proficiency by definition reflects that of a “complete” native speaker. In the case of
L2 attrition, however, such assumptions cannot be made about the attriter’s attained
levels of proficiency, as each learner’s ultimate attainment is variable. Hence, one of
the most important methodological concerns in L2 (including EIL) attrition research
pertains to how to determine each individual’s pre-attrition attainment, especially
when dealing with language loss over the course of many years.
Another related issue concerns the fact that the participants’ pre-attrition data
may not be available in the same form as post-attrition data, making it difficult to
compare pre- and post-assessments. Once again, note that this is an issue for L2
attrition, but not for L1 attrition where the pre-attrition attainment reflects native
speaker proficiency. There are also cautions about selecting a reference group and
equating their data with the pre-attrition proficiency of the participants. This stems
v1_lbp-L.indd 10 10/26/2017 8:11:18 PM
Language Attrition 11
from the possibility that over many years, the particular language variety used by
the group selected as a reference point may have undergone change owing to social,
cultural, or immigration-related reasons. A similar concern for attrition researchers
is that the pre-attrition attainment variable is often inseparable from the partici-
pant’s age, the general tendency being that older children and learners with higher
proficiency are able to retain the language better. Since the two variables often go in
tandem, they should be carefully controlled in any attrition research.
Recently, “a model of assessment of L2 attrition” has been proposed by Bardovi-
Harlig and Stringer (2010) who identified one important set of variables especially
pertinent to L2 attrition research, which they termed “population variables.” In their
review of L2 attrition research to date, they identified three main populations of L2
attriters: (a) children returning to their home countries after residing abroad, (b)
returning missionaries, and (c) students in college or high school. Given that each
population exhibits different characteristics and is subject to different experiences,
Bardovi-Harlig and Stringer suggest that each group be best defined in terms of sets
of variables. These variables comprise those that have already been discussed in this
entry, including age, level of attained proficiency, attitudes and motivation, and lit-
eracy. Because these variables are bound to change over time, they should ideally be
tracked over time, during which reliable measures should be employed to assess
learners’ linguistic abilities both before and after attrition. As can be expected, the
proposed model entails a longitudinal study design with periodic assessment of
contributing factors, as well as repeated measures of the same participants to allow
for tracking of individual learners’ loss patterns. The practical constraints and diffi-
culties often associated with longitudinal studies notwithstanding, it is clear that
issues specific to attrition research can be best addressed by means of a carefully
designed longitudinal study. Longitudinal designs are especially desirable given
that studies of attrition typically cover a longer span of time than studies of acquisi-
tion. Such studies which track learners’ experiences, stretching from sustained expo-
sure to lack of exposure, can further deepen our understanding of the factors that
facilitate or impede language attention; this in turn can shed new insights on lan-
guage maintenance strategies, relearning, and curriculum development.
Bahrick, H. (1984). Fifty years of second language attrition: Implications for programmatic
research. Modern Language Journal, 68, 105–18.
Bardovi-Harlig, K., & Stringer, D. (2010). Variables in second language attrition. Studies in
Second Language Acquisition, 32, 1–45.
Berman, R., & Olshtain, E. (1983). Features of first language transfer in second language
attrition. Applied Linguistics, 4, 222–34.
de Bot, K., & Weltens, B. (1995). Foreign language attrition. Annual Review of Applied
Linguistics, 15, 151–64.
Footnick, R. (2007). A hidden language: Recovery of a lost language is triggered by hypnosis.
In B. Kopke, M. S. Schmid, M. Keijzer, & S. Dostert (Eds.), Language attrition: Theoretical
perspectives (pp. 169–87). Amsterdam, Netherlands: John Benjamins.
v1_lbp-L.indd 11 10/26/2017 8:11:18 PM
Language Attrition
Hansen, L. (1980). Learning and forgetting a second language: The acquisition, loss and reacquisition
of Hindi-Urdu negative structures by English-speaking children (Unpublished doctoral
dissertation). University of California, Berkeley.
Hansen L. (2001). Language attrition: The fate of the start. Annual Review of Applied
Linguistics, 21, 60–73.
Hansen, L. (2011). The acquisition, attrition and relearning of mission vocabulary. In M. S.
Schmid & W. Lowie (Eds.), Modelling bilingualism: From structure to chaos (pp. 115–34).
Amsterdam, Netherlands: John Benjamins.
Kopke, B., & Schmid, M. S. (2004). First language attrition: The next phase. In M. S. Schmid,
B. Kopke, M. Keijzer, & L. Weilemar (Eds.), First language attrition: Interdisciplinary
perspectives on methodological issues (pp. 1–43). Amsterdam, Netherlands: John Benjamins.
Lambert, R. D., & Freed, B. F. (1982). The loss of language skills. Rowley, MA: Newbury House.
Neisser, U. (1984). Interpreting Harry Bahrick’s discovery: What confers immunity against
forgetting? Journal of Experimental Psychology: General, 113, 32–5.
Olshtain, E. (1989). Is second language attrition the reversal of second language acquisition?
Studies in Second Language Acquisition, 11, 151–164.
Pallier, C. (2007). Critical periods in language acquisition and language attrition. In
B. Kopke, M. S. Schmid, M. Keijzer, & S. Dostert (Eds.), Language attrition: Theoretical
perspectives (pp. 155–68). Amsterdam, Netherlands: Benjamins.
Schmid, M. & Kopke, B. (2011). Second language acquisition and attrition. Language
Interaction Acquisition, 2(2), 185–96.
Suggested Readings
Hansen, L. (Ed.). (1999). Second language attrition in Japanese contexts. Cary, NC: Oxford
University Press.
Kopke, B., Schmid, M. S., & Keijzer, M., & Dostert, S. (Eds.). (2007). Language attrition:
Theoretical perspectives. Amsterdam, Netherlands: John Benjamins.
Murtagh, L. (2011). Second language attrition: Theory, research and challenges. In M.Schmid
& W. Lowie (Eds.), Modeling bilingualism: From structure to chaos (pp. 135–154). Amsterdam,
Netherlands: John Benjamins.
Schmid, M. (2011). Language attrition. New York, NY: Cambridge University Press.
v1_lbp-L.indd 12 10/26/2017 8:11:18 PM
  • ... While these migrants are busy learning and using their new language, they experience a gradual decline in their L1 knowledge and skills because they have limited or no exposure to their L1. This declining area includes lexical access difficulty, disfluency, and increased optionality in grammatical judgments (Park, 2018). ...
    ... It is generally suggested that the younger the attriter, the quicker and the more severe the extent of language loss. "L1 remains fairly stable for migrants who leave their L1 environment after puberty, whereas individuals who migrate before puberty can suffer significant L1 attrition, both in the rate and extent of language attrition" (Park, 2018). For example, Pallier's (2007) study showed that young Korean adoptees (ages 3-10) lost their L1 completely after being removed from the L1 speaking environment during childhood. ...
    ... Attrition researchers have proposed that L1 attrition in children should be strictly distinguished from L1 attrition in adolescents and adults, as they suggest that the former case could best be explained in terms of incomplete acquisition rather than attrition. These researchers suggest that incomplete acquisition implies that some grammatical aspects of the language are still in the process of being acquired; attrition, in contrast, should be strictly reserved for individuals whose L1 acquisition has been completed prior to the onset of language attrition (Park, 2018). ...
    Full-text available
    The paper attempts to throw light on the languages used by Bangladeshi expatriate students in Saudi Arabia as well as their feelings and thoughts towards their mother tongue. To meet this end, 175 students of Bangladesh International School in Jeddah (both Bengali and English medium) were asked to fill up a questionnaire and later some of them were interviewed randomly to have a better understanding of their approaches towards different languages. Since no research has yet been conducted in this field, this paper brings to light a number of exciting findings including a multilingual young Bangladeshi community abroad along with a visible demarcation regarding different sociolinguistic aspects such as the medium of instruction of education and their gender.
  • ... It should be noted that de Leeuw's observation is important by reason of other areas of research on language attrition, such as L1 attrition in young children who migrate to an L2dominant environment, for example. In this case, some authors would characterize the changes in the L1 as a result of incomplete L1 development (PARK, 2018). ...
    Full-text available
    In this study, we investigate the occurrence of L1 attrition among Argentinean learners of English (L2) living in their home country, which constitutes an L1-dominant environment. We analyzed the production of Voice Onset Time in word-initial plosives in L1 Spanish by monolingual and bilingual participants. We carried out both an inferential analysis and an individual verification of each participant’s production, which proved to be complementary. Our results suggest that not only is the L2 affected by the L1, but also the L1 can be modified in view of the contact with additional languages.
  • Chapter
    Full-text available
    The study of L1 attrition is currently entering its third decade. However, after twenty years of diligent investigation resulting in numerous theoretical and empirical papers the questions on this topic still greatly outweigh the answers. Findings from individual studies seem to indicate that it cannot even be said with any certainty whether an L1 in which a certain level of proficiency has been reached can ever undergo significant attrition, let alone how or why it might. This chapter will attempt to identify and clarify these issues, in order to establish the starting ground from which the individual papers will proceed. The first section will give a chronological overview of how the field of language attrition has developed over the past two decades, try to identify the concerns and forces that have shaped it, and point out developments and changes. The second section introduces those extralinguistic factors that have been shown to play important roles in attrition. The third section presents an overview of theoretical frameworks within which language attrition studies have been conducted, while the fourth section focuses on questions of research design. The final section is an outline of the structure and contents of this volume.
  • Article
    Our case study confirms two previous studies demonstrating that a 'lost' lan - guage can be recovered by hypnosis. These results point to the existence of an - other type of attrition. Here a distinction is made between this type of attrition of a 'hidden language' (HL), that is, one that has not been forgotten but rather has become inaccessible to the speaker, and the traditionally studied attrition of a 'forgotten language' (FL) or one that is being forgotten. A conflict hypoth - esis is proposed to explain HL attrition. An attempt is made to show how these two types of attrition could be distinguished and to explain why brain imagery analyses are important for this distinction.
  • Article
  • Article
    The study of language attrition, whether it is concerned with first or second languages, focuses on the effects resulting from an individual's reduced use of the attrited language. Such reduction in use can be due to a change in the linguistic environment or to the termination of an instructional program. In either case, some other language (or languages) is or becomes the dominant one.The present article reports on a series of studies, all focusing on individual attrition of English as a second language (ESL) in an environment where Hebrew is the dominant language. The predictor variables discussed are age, sociolinguistic features, input variables, and linguistic variables. The attrition process affecting English as a second language in a Hebrew dominant context seems to exhibit two major trends of change in language use: (a) a greater variability in the application of peripheral and highly marked structural rules, and (b) lower accessibility of specific lexical items. In each of these trends one can identify a limited reversal of the acquisition process, particularly with young children (5–8-year-olds) as well as a typological transfer process from the dominant language.