Age related factors and L 2 learning . Implications for selecting an appropriate L 2 programme

Th is paper has been inspired by Dimroth and Stephany’s longitudinal study (Dimroth, 2008) on the acquisition of German by two Russian girls, members of diff erent age groups (8 and 14). Both girls were subjected to the same type of second language program called ‘submersion and withdrawal’, but showed diff erent progress. Whilst the 8-year old became fl uent in German, the 14-year old had diffi culty expressing herself in her second language. Th is paper considers various theoretical approaches and research fi ndings in order to account for the diff erences between the two learners. By all accounts there may be no single factor responsible for their individual success rates.


Introduction
In a study by Dimroth and Stephany (Dimroth, 2008) the acquisition of German by two Russian girls (sisters) age 8 and 14, was analyzed.Th e two girls arrived simultaneously in Germany with their parents and started to learn German in a nonguided way.Both girls attended the same school which had a policy of 'submersion and withdrawal' for the acquisition of German as an L2.Four years later, the parents moved to Vienna.At this point, the older child decided to move back to Russia, the younger stayed with her parents in Vienna, but would have liked to return to her friends in Germany.Th e results show that the younger child speaks German fl uently, while the older has many problems 1 cmarijana_2000@yahoo.com expressing herself in German.Th is paper discusses the possible reasons for the diff erent success rates of the two girls.

Monolingual system
When thinking about the diff erent success rates of the two L2 learner one of the possible reasons that comes to mind is the nature of the L2 system they were exposed to.First of all, the system of 'submersion and withdrawal' is a type of monolingual education for language minorities.Th ere is a generally accepted view nowadays that bilingual language learning has an advantage over the monolingual one, having in mind that monolingual education can result in certain negative side eff ects such as loss of minority language and sometimes language death of minority groups.
A case study by Saville-Troike (1982) illustrates how a monolingual policy frequently supports language maintenance and goes against assimilation, while a bilingual policy greatly supports assimilation of L2 speakers into the prevailing language group.Saville-Troike (1982) gives an example of two Pakistani groups: Pashto and Baluchi who share a common culture, but speak two diff erent languages and have diff erent attitudes to social organization.Th e policy of the fi rst tribe was that they required Pashto for full political participation, which means that monolingualism was supported, while the structure of Baluchi tribes allowed bilingual participation and more easily assimilated non-Balluchi speakers.Th e two diff erent policies fi nally contributed to the spread of Baluchi at the expense of the Pashto in the region.

Th e nature of 'submersion and withdrawal' L2 classes
According to Garcia (1997: 411) 'submersion and withdrawal' is a kind of L2 program which is easiest to plan, and therefore is widespread.Language minority students attend mainstream classes where no provisions are made for them.Th us, they are submersed in the majority language for all content, but they are 'withdrawn' or 'pulled out' for second language instruction with a language teacher.As soon as students become bilingual, the L2 instruction ceases.
Table 1: Baker (1993: 153)  Th is program is characterized by a fast conversion to the majority language and this according to Baker (1993: 199) 'stands chance of doing more harm than good' .

Implications for the Russian girls' case
Taking into account the general characteristics of 'submersion and withdrawal' program, it may be expected that the fi nal outcome with both of the L2 learners would be fast transition to German, and possibly fi nal loss of their L1.However, while the system worked in the case of the younger girl, but not in the case of the older one, it is obvious that their diff erent performances in their L2 German are not caused exclusively by the type of the system applied.Instead, other factors as well as possible interaction between several factors may have caused the diff erences in the two L2 learners.

Age factor
Diff erent performances of the two girls may be age related, having in mind that the younger girl is eight years old, while the older one is fourteen.According to Eckert (1997) the two girls fall within two diff erent age cohorts, the fi rst being in the period of childhood, the second one falling within the scope of adolescence.Th ese two age groups show diff erent properties.Children are more socioeconomically mobile: adults are not their primary linguistic models, but interaction with siblings, neighbours and friends infl uences them very much.Adolescents, on the other hand are less socioeconomically mobile and they feel that they belong to a separate age cohort.Th is is another possible reason why the younger learner was more successful than the older one.

Identity as a factor in language learning
Th ere is an increasing feeling of identity inherent to the period of adolescence, and since identity and language according to Fishman (1997: 329) always go together, it is reasonable that feeling of belonging to a certain group means also sharing the language of the group.What follows from this is that the fourteen-year old girl may have had more pronounced feelings of identity and may have seen herself as belonging to the Russian ethnic group and moreover to the exclusively Russian speaking group.If we also have in mind that the basic aims of the system 'submersion and withdrawal' are assimilation and monolingualism, then the causes of failure of the 14-year-old girl are more evident.She may have seen the system as a kind of suppression of her identity and therefore, she might have been liable to confl ict with native speakers, who are at the same time members of a diff erent culture.
L2 acquisition was quite diff erent for the younger girl who being in the period of childhood, and not having yet developed a strong feeling of belongingness to a certain group may not have seen the above described system of schooling as a kind of suppression and therefore was more willing to learn her L2.

Lerner-specifi c characteristics
Of course, we must not exclude the fact that a bilingual situation is always specifi c.According to Tabouret-Keller (1997:320) a bilingual speaker may gain diff erent feelings from a contact of two languages: sometimes this contact gives rise to feelings of inferiority, discrimination or exclusion from the dominant group or conversely feelings of familiarity and recognition amongst those who share the contact situation.Even the initial attitudes and expectations about a new language and new culture may be diff erent: some people reject their own group and wish to change and belong to some other group, while others value their own group membership and do not wish to acculturate.On some other occasions, however, people wish to be members of more than one group and be bicultural as well as bilingual.Saville-Troike (1982:198) reports problems of bilingual parents in the USA (who speak other languages at home) once their children begin school.Some children want to speak only English at home and they also force their parents to do so.On the other hand, some children attending bilingual school programmes willingly speak both of their languages.

Critical Period
Another possible reason, which is more linguistically grounded and which is also connected to the learner's age is the so called notion of Critical Period, which can as well be responsible for diff erent L2 outcomes in the two girls.
Biologist Lenneberg (1967) formulated his CPH (Critical Period Hypothesis) in respect to second/foreign language learning.According to him, the possibility of reaching native-like levels in L2 is age restricted.Critical period appears, according to Lenneberg round puberty, which is around the age of 12 or 13, and is caused by biological factors.Th e brain loses the ability of adaptation, the so-called neural plasticity, because language functions are supposed to be established by this age.Aft er this, age related 'window of opportunity' is closed, automatic learning of a second language by a mere exposure to it seems to disappear and from that point on foreign languages have to be taught and learned through a conscious and laboured eff ort.
If Lenneberg's hypothesis is true, it well explains the diff erence in the progress of the two girls.According to these criteria 'the window of opportunity' has already closed in case of the older girl, and that is why her performance in German is inferior when compared to her sister, and it is less likely that she would reach a native-like fl uency.On the other hand, the younger girl is well before the Criti-cal Period (according to Lenneberg at least), which explains her success in acquiring her L2 German.However, some other theorists (Pinker, 1994 etc.) place the boundaries of CP around the age of six.According to this hypothesis, the younger girl had too reached the CP when she began learning her L2, which then does not off er any explanations for the diff erences between the two sisters.However, if we take a middle solution proposed by Birdsong andMolis (1998[reported in Hyltenstam &Abrahamson, 2000]) who suggest the age eff ects over the whole life span, i.e. the younger one starts learning L2, the better the outcomes, this can still account for the diff erences in the L2 performance between the two girls.

Conclusion
Th is paper seeks to account for the diff erent learning outcomes of the two learners belonging to two diff erent age groups (age 8 and 14) and being exposed to the same 'submersion and withdrawal' learning system.Th e fi rst thing to be noted is that there is not a single factor which can account for the diff erence in the success rates of the two girls.One cannot criticize the system of 'submer-sion and withdrawal' and blame it for the failure of the older learner because it was obviously successful in the case of the younger learner.It could be suggested that this system may work only for younger L2 learners, while some other system may be bettersuited for the older ones.From the above data it also follows that the period of adolescence plays a special role in the formation of both learners.Obviously, it is in this period that individual's identity and feeling of belonging to a certain group is created, so this period of learner's life may be responsible for his/her attitude towards acquiring an L2.Although we cannot make any strong statements about the role of the Critical Period, having in mind that this is still an unresolved issue, there is still the fact that the rate of L2 learning was higher for the younger learner and this may imply that age eff ects increase over the life span.It follows that age is a very important factor in language learning and in order to produce the best results possible, the language learning system has to be designed as to suit the age of the learner.Th e older the learner, the more eff ort should be invested into L2 teaching techniques.Finally, one should not forget that bilingual situation is always specifi c and that there are individual learner characteristics that should be taken into account.Овај рад је инспирисан лонгитудиналном студијом К. Димрот и С. Хаберцетл (Димрот, 2008) о усвајању немачког језика од стране две младе Рускиње, које су припадале различитим старосним групама (осам и четрнаест година).Обе девојчице су биле изложене истом програму учења страног језика "Submersion and Withdrawal", али су показивале различит напредак.Док је осмогодишњакиња постала флуентна у немачком, четрнаестогодишњакиња је имала потешкоћа да се изрази на страном језику.У овом раду се разматрају различити теоријски приступи и резултати истраживања ради објашњења разлика између два ученика.По свему судећи, не постоји само један фактор који је одговоран за индивидуално постигнуће ученица.
illustrates the 'submersion and withdrawal' L2 programme by the following table: