Second-language attrition occurs whenever the learner uses the second language to an “insufficient degree (De Bot & Weltens 1991:43) or due to environmental changes the language use is limited and another language is becoming the dominant one (Olshtain 1989: 151).
Contents |
“Almost everybody who has learned a foreign language shares the experience of forgetting the acquired language skills once the period of formal instruction is over” (Schöpper-Grabe 1998: 231) Second Language Attrition can be defined as the “non-pathological decrease in a language that had previously been acquired by an individual” (Köpke & Schmid 2004: 5) and described as the “[…] loss of skills in the individual over time (de Bot & Weltens, 1995).
Beginning in the 70's until today a new and especially young field in the area of Second Language Acquisition has developed. It is connected cross-sectional throughout different research areas. Language attrition in general is concerned with what is lost (linguistic focus), how it is lost (psycholinguistic and neurolinguistic focus) and why it is lost (sociolinguistic, sociologistic and anthropologistic focus) (Hansen 1999). Since over 25 years research has concentrated on Second Language Attrition. First studies, dealing with the topic of language loss or language attrition, were published in the late 70's (de Bot & Weltens 1989: 127). In 1980, as the University of Pennsylvania hosted the conference “The Loss of Language Skills”, finally it was recognized as a field in the research of Second Language Acquisition. Since then various research papers – mainly within America – have been published. Later several studies in Europe – especially the Netherlands – followed. In other countries, however, language attrition research was paid hardly any attention (de Bot & Weltens 1995). Compared to the field of Second Language Acquisition, Language attrition is quite young and therefore so much is still unknown. Because there exists yet so much uncertainty about acquisition of second languages in general, the question raises why it is necessary to engage with their loss.
The purpose of Language Attrition in general is to discover how, why and what is lost. The aim in foreign or second language attrition research is to find out why after an active learning process the language competence changes or even stops (Gleason 1982). Further, results from research in this area could, as Van Els and Weltens (1989) counter, contribute inter alia for the understanding of relations between acquisition and attrition (van Els 1989). L2/FL attrition research is mainly important because it provides interesting results for foreign language instruction. De Bot and Weltens (1995: 152) state, that “[r]esearch on language attrition can also have a considerable impact on curriculum planning or foreign language teaching”. The theoretical grounding of the language attrition research derives primarily from cognitive and psychological theories. Research in the area of language attrition concentrates generally on the loss of the L1 and L2. The first distinction that can be made is between pathological and natural language attrition. The former concentrates on language loss caused by a damage of the brain, injury, age or illness. However, this topic will not be investigated any further, because the language attrition in these cases is not caused by natural circumstances. Weltens (1987: 24) states another possible distinction, inter and intra generational language attrition. Inter generational language attrition is concerned with attrition within individuals, whereas intra generational language attrition concentrates on the attrition across different generations. Van Els (1986) distinguishes types of attrition in terms of which language is lost and in which environment it is being lost. Therefore he classifies:
To provide an answer how second language attrition happens, it is necessary to have a glance at the findings of the research of memory. Since its establishment by Ebbinghaus in the late 19th century the empirical research about learning still plays an important role in the modern research of memory.
Hermann Ebbinghaus contributed a lot to the research of the memory. He made the first empirical study concerning the function of the memory as to the storage and forgetting of information. His major finding was that the amount of learned knowledge depends on the amount of time investigated. Further, the more time is passing by, the more repetitions are necessary. Resulting, from the findings of Ebbinghaus, the first theory of forgetting was established, the decay theory. It says that if something new is learned a “memory trace” is formed. This trace is going to be decayed, if it is not used in the course of time, and by decaying of this trace forgetting takes place (Weltens 1987).
The Interference theory can be seen as one of the most important theories of forgetting. It indicates that prior, posterior or new learning information compete with already existing ones and therefore forgetting occurs. This inhibition can be divided into two kinds: the retroactive inhibition, where information acquired at a later point in time block the information that was acquired earlier. Proactive inhibition, means that information acquired in the past can infer with new information. Hence, a blocking can occur which inhibits the acquiring of the new target item. (Ecke 2004: 325).
Today, it is another most approved theory concerning the function of the memory (Schöpper-Grabe 1998:237). It says that the storage of information happens on different levels. Therefore, information or memory is not deleted, rather the access to the current level is blocked and hence, the information is not available. Hansen (1999: 10) quotes Loftus & Loftus (1976) in order to describe forgetting: “[…] much like being unable to find something that we have misplaced somewhere.” Cohen 1986 states, evidence for knowing that a learner is not able to “find” something, is the use of the so-called progressive retrieval. Thereby the learner is unable to express something that is in his mind and therefore uses an incorrect form and eventually remembers the correct one (Cohen 1986; Olshtain 1989). Time is considered as the decisive factor to measure how far the attrition has proceeded already (de Bot & Weltens, 1995).
It is necessary for a complete understanding of language attrition to have a look at the various hypotheses which try to explain how language changes over time.
The regression hypothesis can be named as the first established theory in language loss. Its tradition goes far back, further than any other theory. The first one, who designed it was Ribot in 1880, later on Freud took Ribot's idea up again and brought it in connection with aphasia. (Weltens & Schmid 2004: 211). In 1940 Roman Jakobson embedded it into a linguistic framework and claimed that language attrition is the mirror image of language acquisition (Weltens & Cohen 1989: 130). Even though only a few studies have tested this hypothesis it is quite attractive to many researchers. Because as Weltens and Schmid (2004: 212) state, children acquire the language in stages and therefore it was suggested that language competence in general appears in different layers and therefore attrition, as the mirror image of acquisition will also happen from the top layer to the bottom.
According to the regression hypothesis two similar approaches developed. Cohen (1975) started to conduct several studies on his own to determine “whether the last things learned are, in fact, the first things to be forgotten, and whether forgetting entails unlearning in reverse order from the original learning process.” (Cohen 1975: 128). He observed the attrition of Spanish, as the second language, among school children during the summer vacation. Cohen result’s supported the regression hypothesis and his last-learned-first-forgotten thesis, that some things, which are learned the last are the first to be forgotten, when the learner has no input of the target language anymore.
Another variation of the regression hypothesis is the best learned-last-forgotten hypothesis, which emphasises the intensity and quality of the acquired knowledge not the order. Therefore, the better something is learned, the longer it will remain, because the language component that is repeated again and again is automated and thus with a high probability sustain in the memory (Schöpper-Grabe 1998: 241).
It was introduced in the field by Andersen (Andersen 1982). He claims that second languages or foreign languages which share more differences with the respective mother tongue than similarities are more endangered to be forgotten than those similar to the L1. Another point is the attrition of components, which are less “functional”, “marked” or “frequent” compared to other elements (Weltens & Cohen 1989: 130). This hypothesis is more differentiated and complex as the regression hypothesis, because it considers aspects from first and second language acquisition research, language contact and aphasia research and the survey of pidgin and creole languages (Müller 1995). By means of this hypothesis research it tries to detect the aspects of language that are first to be forgotten.
Defining the process of language attrition it has to be considered that there exist several theories as well. Almost every researcher claims different stages how language attrition is happened.
Gardner (1982: 519520) believes that the process of second language attrition is divided into three points in time:
The time from time 1 to time 2 he describes as the Acquisition Period and between time line 2 and 3 as the Incubation Period (1982: 520). Further he states, that it is not enough to consider only the time that has passed between 2 and 3 to make statements about the attrition, rather the duration, relative success and nature of the acquisition period and the duration and content of the incubation phase has to be looked at as well (Gardner 1982: 520). The Acquisition Period is the time where language learning or language experience takes place, mainly from the first to the last lesson. During the Incubation Period no language training or language usage takes place and the forgetting may begin. He says that now, that language learning is not active anymore, a study about language attrition can be conducted (Gardner 1982a: 2).
The forgetting curve orientates on the typical forgetting curve by Ebbinghaus. He said, that already after a very short amount of time a forgetting process sets in immediately, stabilises and then levels off. Bahrick conducted a study, where he tested 773 persons with Spanish as their L2. His probates had varying acquisition and incubation periods, up to 50 years of non-active learning. He discovered a heavy attrition within the first 5 years, then stabilised for the next 20 years (Weltens & Cohen 1989: 130). According to Bahrick the knowledge that remained after 5 years is stored in the permastore. Neisser (1984) uses a different term, he prefers critical threshold, a level, which has to be reached, and beyond that threshold knowledge will resist decay. This theory has been approved in the literature a lot, because the process of language attrition is slower, if a certain level of competence in the L2 is achieved (Feuerhake 2004: 5). Contrary to these finding Weltens & Cohen (1989: 130) are reporting from studies, where different results were found. According to these findings the forgetting curve begins with an initial plateau, a period where the language competence is not affected at all. This follows then with the onset of attrition. Weltens explains these results, by the high proficiency of the probates (bilinguals and immersion students). However, unknown remains if the curve that follows this plateau might be exactly like the “normal” forgetting curve of language learners with a lower proficiency level (Weltens & Cohen 1989: 130).
Another phenomenon is relearning. Some studies show that despite the end of learning and no language input a residual learning can happen. Weltens (1989), who studied foreign language learners, identifies an increase in reading and listening comprehension. He says that it happens because a process of maturation in general happens. Schöpper-Grabe determines that contact and the intensity with the target language cannot be the only variable causing language attrition.
In the literature several factors are named for explaining why language competence is decreasing. Many researcher, however, regard the level of competence of the learner as essential for the following attrition. It is said, that the higher the level of competence the less attrition will take place. Thus, a reference to the theory of the critical threshold can be drawn. Similar to this theory it is claimed, that according to conducted studies, the higher the level of competence of the learner at the end of the incubation period, the fewer will be lost. Therefore duration, success and intensity of the language instruction or language input in general is vitally important. Weltens (1989) divides factors influencing language attrition into two categories, the first one are psychological factors, which are dependent upon the individual learner and divided into biological and cognitive aspects. The second category are sociopsychological factors, as the attitude towards the target language and culture and aligned with the motivation for acquiring the language. Further, factors, which are settled in the language environment, should be considered as well, e.g. the status and prestige of the language are meaningful, too. Another frequent occurring factor is age. A variable that seems to be quite important, especially observing language attrition in children. Even though children are regarded as the better foreign language learner, their cognitive development is less progressed compared to adults. Further, usually they haven’t learned to write or read in any language, and usually particularly not in the second language at all. Therefore their literacy skill in the L2 is very limited if not even there yet. Cohen (1989) conducted a study observing young children. He found out, that the attrition in an 8-year old boy was stronger than the one in his 12-year old sister. Tomiyama, suggested on the basis of his findings, that these children might not lose their knowledge of the L2 completely, moreover the access to such information is inaccessible and may vanish with time passing by (Feuerhake 2004). At the beginning of the 80's another, so far unnoticed factor, was introduced into the research field. Socioaffective factors as attitude, orientation and motivation are now accounted. On account of that, he established a socio-educational model of language acquisition. Thereby motivation and attitude influence the workload of the individual to keep their language competence. Further, individuals, who have positive attitudes towards the target language, seek possibilities and opportunities during the incubation period to retain their language competence (Gardner 1987: 521). However, the factor motivation is hardly considered examining language attrition. Especially during the last 10–15 years it became more and more acknowledged in the field of language acquisition rather than attrition. Only Gardner considered motivation as a possible factor influencing attrition. Even until today it is hardly recognised as an influencing factor and therefore exist only a few studies about motivation and its effects.
Feuerhake (2004: 7) reports that, looking at released studies, that have been conducted, it can be seen that all four competence areas2 are affected. Though some of these seem more likely to be affected than others, e.g. grammar and lexical knowledge are more likely to suffer a high attrition process. Showing loss in speaking competence, the first evidence is that the speech tempo decreases. Longer and more frequently occurring speech pauses, under which the fluency is suffering, are observable as well (Gardner 1987). Olshtain (1986) observed […] reduced accessibility in vocabulary retrieval in all situations of attrition where there is a reduction of language loss over longer periods of time.” (1986: 163). Further, gaps concerning grammatical knowledge, especially tenses and conjunction of verbs occur quite frequently. Nevertheless, it can be said that productive skills are more affected than receptive ones, which mainly remain constantly stable (Cohen 1989) and if the learner shows already signs of language attrition it is more likely that transfer from L1 will happen (Berman & Olshtain 1983). Cohen examined in his studies several strategies, the learner applies to compensate the lack of adequate speaking skills, e.g. one strategy is code-switching, to uphold the communication. Another phenomenon observable is a kind of “mixed-language”. Müller (1995) states that on many levels of speaking the learner falls back on a mixture between different languages. Still, it is important to mention, that, as with almost every study that has been conducted in the different sub-fields of Second Language Acquisition research, several problems arise. There are longitudinal vs. cross-sectional studies, different variables, which have been used, and mainly terms and conditions of acquisition and incubation period are not standardised, particularly the length of the incubation period (Feuerhake 2004: 8). That means, some studies only observe language attrition after language programs, other look at the attrition in between breaks of language programs and studies, which examine the attrition after change of environment, regarding language and living conditions (Cohen 1975, Olshtain 1989). Finally, studies reviewed in this paper show that attrition follows a certain order, e.g. productive skills are more affected than receptive skills. Mainly due to difficulties in lexical retrieval a loss in fluency seems to be the first signs of language attrition, followed by attrition in morphology and syntax. Further observations in language attrition are necessary, to give a better understanding of how the human mind deals with language (Hansen 1999: 78).
The following chapter is trying to explain motivation and its influence on language attrition. Until 1990 the sociopsychological model of Gardner dominated the research about motivation. Gardner and Lambert emphasise thereby the importance of attitude towards the language, the target country and language community (Feuerhake 2004).
According to Gardner and Lambert (1972) a learner is instrumental orientated if learning a foreign language has a function, e.g. for success in career terms. Thereby, the language becomes an instrument to achieve the higher purpose and the foreign language learning is concentrated on fulfilling the aim of the learner (Feuerhake 2004: 9).The integrative orientation follows the aim of acculturating with the target language and country as well as the integration into the target language community.
The instrumental and integrative orientation is not enough to cover all aspects of the term motivation, the term intrinsic and extrinsic motivation was added to the model. The term intrinsic is connected with behaviour, which results from the reward of the activity itself. The learner acts, because he is enjoying the activity or it is satisfying his curiosity. Mainly it is self-determined and the learner is eager to learn a foreign language because he wants to achieve a certain level of competence. The learner enjoys learning and the acquisition of a foreign language is challenging. Extrinsic motivated learners are orientated on external stimuli, e.g. positive feedback or expectations from others. In general four different types of extrinsic motivation can be distinguished (Bahar 2005):
Bahar (2005: 66) quotes Pintrich & Schunk (1996), who state that “[…] motivation involves various mental processes that lead to the initiation and maintenance of action […]”. Hence, motivation is a dynamic process that changes over time and the motivation of a learner as well might change during the learning process. Therefore, it cannot be seen as an isolated factor. Moreover, several other factors, which are settled within the learner, as well as in the environment, influence motivation and are responsible for its intensity and variability.