Excerpt
Table of contents
List of abbreviations
List of figures
1 Introduction
2 The origins of the notion of a sensitive period in second language acquisition
3 Selected studies on the Critical and Sensitive Period Hypothesis In second language acquisition
3.1 Delimitation of terms
3.1.1 Critical versus sensitive period
3.1.2 Initial learning rate versus ultimate attainment
3.2 The sensitive period in natural settings
3.2.1 Empirical results
3.2.2 Explanatory approaches
3.3 The sensitive period in instructional settings
4 The notion of a sensitive period in pedagogical concepts of bilingual kindergartens
4.1 Method
4.2 Results of the Analysis
5 Conclusion
Bibliography
Appendix
List of Abbreviations
illustration not visible in this excerpt
List of figures
Figure 1: The process of the critical period
Figure 2: Near-nativeness in the context of the concept of a critical period
1 Introduction
‘A tree must be bent while it is young.’ (Proverb)
Proverbs usually have an element of truth. The German speakers will certainly also know the German equivalent ‘Was Hänschen nicht lernt, lernt Hans nimmermehr.’ Do these proverbs also apply to the acquisition of a second language? Are young learners trees that ‘must be bent’ before they are too old to reach a certain language proficiency?
In first and second language research, the concept of a ‘critical period’ respectively of a ‘sensitive period’ was developed to explain probable advantages of children in the process of language acquisition. The aim of this term paper is to find out whether there is a sensitive period for second language learning and to explore possible explanatory approaches.
Within this research, the ‘Critical Period Hypothesis’ (CPH) represents the major focus. According to this hypothesis, “there is a period during which language acquisition is easy and complete (i.e. native-speaker ability is achieved) and beyond which it is difficult and typically incomplete” (ELLIS, 1997: 67). This means, it is assumed that language learners must begin at an early age to learn a language in order to reach native-like proficiency.
Generally, empirical data from course books, handbook articles and journal articles as well as commentaries by different authors are used in order to fol- low the research question up. First, a framework concerning the origins of the idea of a sensitive period from first language acquisition, and the clarification of particular terms will be created. Then, the distinction between possible ad- vantages of young learners in natural as well as instructional settings will be of interest.
The findings on a critical or sensitive period have also found their ways into the pedagogical practice. It is obvious that especially bilingual kindergartens, that promote an early start in language learning in general, will somehow re- fer to the idea of a sensitive period in second language acquisition. There- fore, the second part of this term paper will present an analysis of pedagogi- cal concepts of these kinds of day care institutions with respect to the inte- gration of this idea. Website information of different bilingual kindergartens will serve as sources.
2 The origins of the notion of a sensitive period in second language acquisition
The concept of a sensitive period for second language acquisition goes back to research and theoretical models regarding first language acquisition. The neuroscientists W. Penfield and L. Roberts made the idea of a critical period in language acquisition public in the end of the 1950s. Based on studies on brain damage and experiences with their own children, they claimed a spe- cialized capacity of the child’s brain for language learning (cf. HYLTENSTAM, ABRAHAMSSON, 2003: 539; MUNOZ, 2006: 1). In the 1960s, these findings received theoretical support by the linguist and neurologist E. Lenneberg and his ‘Critical Period Hypothesis’.
Generally, research on brain activity has shown that there is a lateralization, i.e. a specialization of the two halves of the brain, with regard to language processing. Although there are many areas active in these processes, the core linguistic processes are characteristically located in the left hemisphere. This specialization increases as the brain matures. At the same time, the other half of the brain loses gradually its ability to take over functions of the counterpart; in other words, it loses its plasticity (cf. SAVILLE-TROIKE, 2007: 68).
Lenneberg worked with studies of recovery of aphasia1 as well as studies of feral children, which means children who went through social isolation and had not learned a language before puberty. The case of Genie is one of the most popular ones amongst these studies. She was completely isolated until the age of thirteen and consequently did not acquire a first language (L1) as a child. After her discovery in 1970, she developed substantial communica- Is there a sensitive period for second language learning? 3 tive abilities, but lacked many grammatical rules (cf. ELLIS, 2007: 67f; SAVILLE-TROIKE, 2007: 82).
According to Lenneberg, the lateralization of the language functions in the brain is typically completed at puberty. He noted in his studies with children who suffered from brain damage at a certain age, namely puberty, that other parts of the brain could assume the functions of the injured parts due to the still active plasticity of the brain. Normal language acquisition was possible in these cases. However, if injuries of the language areas or social isolation took place after puberty, language acquisition was difficult or incomplete (cf. COLLIER, 1987; ELLIS, 1997: 67; HERSCHENSOHN, 2007: 16f; HYLTEN- STAM, ABRAHAMSSON, 2003: 539; SAVILLE-TROIKE, 2007: 68f).
Nonetheless, “the extant evidence of language learning by feral children is so scarce that it cannot be used to provide strong support for either this or op- posing views. (…) Likewise, the pathological kind of evidence provided by aphasic patients needs to be treated with caution.” (MUNOZ, 2006: 1)
On the basis of the findings in first language acquisition research, a model for a critical period in second language acquisition was developed. In accordance with this model, such a period consists of an onset, a peak and an offset or terminus (cf. figure 1) (cf. HERSCHENSOHN, 2007: 10; HYLTENSTAM, ABRAHAMSSON, 2003: 558ff; MEISEL, 2007: 2).
Figure 1: The process of the critical period
illustration not visible in this excerpt
Source: MEISEL, 2007: 103.
The onset is either defined by the beginning of primary language acquisition or by points when there is a characteristic acceleration in linguistic development, depending on the author cited. Then, the peak of this optimal period for second language acquisition is reached quite fast (cf. HYLTEN-STAM, ABRAHAMSSON, 2003: 558ff; MEISEL, 2007: 2).
The issue of the offset is the most debated one in the discussion about a critical period (cf. HYLTENSTAM, ABRAHAMSSON, 2003: 558). For in- stance, the question whether the terminus of such a period dies slowly down as indicated in figure 1 or whether it ends abruptly depends on the definition as either a critical or a sensitive period (cf. chapter 3.1.1, same page). Fur- thermore, it is assumed that there are different points of offset for different aspects of language, such as phonology or syntax (cf. MEISEL, 2007: 102f; SAVILLE-TROIKE, 2007: 83).
3 Selected studies on the Critical and Sensitive Period Hypothesis in second language acquisition
3.1 Delimitation of terms
3.1.1 Critical versus sensitive period
Whereas Lenneberg titled his hypothesis with the term ‘critical’, the idea of a ‘Sensitive Period Hypothesis’ (SPH) can be found in different sources by different authors, too. Although both terms are also often used synonymously in the literature, there are differences between them (cf. HYLTENSTAM, ABRAHAMSSON, 2003: 556).
Both refer to the concept of maturational constraints and a sensitivity for lan- guage input during an early stage of life. The CPH follows the idea of matura- tional constraints on the basis of irreversible changes in brain functions (cf. HERSCHENSOHN, 2007: 11). According to this hypothesis, “maturation is thought to take place and come to an end within an early phase of the life span, abruptly set off from the rest at a specific age (puberty or earlier)” (HYLTENSTAM, ABRAHAMSSON, 2003: 556). By contrast, the sensitive period is believed to be more gradual in its final point. Characteristically, the sensitivity for language input diminishes over a longer period of time. The point in time, ]when this happens, cannot be assigned definitely. It can range from later childhood or puberty to adolescence (cf. GASS, SELINKER, 2008: 405; HYLTENSTAM, ABRAHAMSSON, 2003: 556).
As the title of this research paper already suggests, the notion of a sensitive period is favoured within this context.
3.1.2 Initial learning rate versus ultimate attainment
When researchers examine language competences of younger and older language learners assigning the term ‘better’ to one particular group, it is necessary to distinguish between two aspects. On the one hand, the as- sessment of ‘better’ results can refer to the rate of learning. On the other hand, the ultimate attainment can be of interest with regard to its approxima- tion to native-like language proficiency (cf. DIMROTH, 2007: 115f).
Overall, different studies have shown that older learners, adults as well as older children, acquire a second language (L2) faster in the first phase of the acquisition process. Young children are slower first, but finally reach a higher level of proficiency than the older ones. However, these advantages of older respectively younger learners might apply to particular aspects of a language (cf. AGULLÓ, 2006: 367; HYLTENSTAM, ABRAHAMSSON, 2003: 545f; MUNOZ, 2006: 2f).
It is assumed that older learners use their capacities for explicit learning and analytical thinking which young children lack. Consequently, older learners achieve better results particularly in L2 syntax, morphology and other literacy-related skills in the first stage. In contrast, younger learners make use of implicit mechanisms developing more native-like intuition. Language aspects which are not concerned with cognitive dimensions and academic skills, such as oral fluency and accent, are better mastered by young children (cf. MUNOZ, 2006: 6, 10f; SAVILLE-TROIKE, 2007: 84).
In most of the studies on a sensitive period in second language acquisition, the focus is on the ultimate language proficiency of younger and older lan- guage learners. Nevertheless, the differentiation between initial learning rate and ultimate attainment should also have an impact on the evaluation of em- pirical studies. The length of residence of immigrants, whose language com- petences in the L2 are measured, should range from a minimum of 10 years and above. Otherwise, it could not be ensured that the results apply to the ultimate attainment and not to initial rate effects (cf. DEKEYSER, 2000, cited in MUNOZ, 2008: 582).
3.2 The sensitive period in natural settings
3.2.1 Empirical results
Already in 1982, Krashen examined about 25 studies which dealt with the effect of the age on initial learning rate as well as on final attainment in second language acquisition. Overall, the results commonly indicated that older language learners acquire a L2 faster than younger ones in the first stage and that the final attainment of young children is superior to ultimate language achievements of the older children and adults (cf. KRASHEN et.al., 1982, cited in PATKOWSKI, 1990: 80).
In 1989, Johnson and Newport conducted a study, which represents a ‘land- mark study’ for several researchers. 46 adult Chinese and Korean L2 learn- ers of English took part in this study which applied a grammaticality judge- ment test. The test persons’ ages of arrival in the United States ranged from 3 to 39. In sum, the group of persons with the lowest age of arrival, namely 3 to 7 years, showed results which corresponded to native-like proficiency. By contrast, the performance of the L2 speakers with an age of arrival above 7 declined. From the age of 17 as point of arrival, the linear decline of the test persons’ results disappeared (cf. JOHNSON, NEWPORT, 1989, cited in HYLTENSTAM, ABRAHAMSSON, 2003: 547f).
Nonetheless, the study has also been criticised due to methodological weaknesses. Amongst others, the minimum length of residence of the test persons, which was defined with five years, was questioned (cf. HYLTENSTAM, ABRAHAMSSON, 2003: 548).
DeKeyer replicated Johnson’s and Newport’s study in 2000 avoiding most of the weaknesses, although there were still some critics. In this study, the par- ticipants’ L1 was Hungarian. Their minimum length of stay was set with 10 years. Within this context, a correlation between the age of arrival and the scores in the grammaticality judgement test was found as it was the case in the preceding study (DEKEYER, 2000, cited in HYLTENSTAM, ABRA- HAMSSON, 2003: 548f).
Nonetheless, the idea of a sensible period in language learning, which was confirmed by the previous studies, have been challenged by some research- ers.
From time to time, there are reports on persons who reached nativespeakers abilities even though they started learning the particular L2 as adults. Consequently, it is inferred that not all language learners underlie a sensitive period (cf. DIMROTH, 2007: 117; ELLIS, 1997: 68). Some critics of the CPH or SPH even conclude from these findings and from the fact that there are significant similarities in the sequences of second language acquisition irrespective of the age that also adults somehow have access to inborn linguistic knowledge (cf. DIMROTH, 2007: 117).
By contrast, the argument of examples of native-like adult starters is opposed by other authors. According to them, it is questionable whether L2 learners ever can reach real native-like proficiency. In other words, bilingual or multi- lingual competences cannot be compared to the ones of native speakers (cf. GASS, SELINKER, 2008: 407; HYLTENSTAM, ABRAHAMSSON, 2003: 555f; MUNOZ, 2008: 580). Hyltenstam and Abrahamsson even suggest that “those studies that claim nativelike attainment in young learners generally do so on the basis of under-analyzed data.” (HYLTENSTAM, ABRAHAMSSON, 2003: 570) They state that data from L1 research implies maturational effects already from an age of approximately 12 months hypothesizing that these effects of maturational constraints are also distinct in the acquisition of a L2. Consequently, also L2 learners, who started from a very early age on, could not achieve native-like proficiency. However, the differences between the language competences and performances of very early starters on the one hand and native-speakers on the other hand are extremely difficult to distin-guish, which is meant by their term of “under-analyzed data” (cf. HYLTENSTAM, ABRAHAMSSON, 2003: 570ff).
[...]
1 ‘Aphasia’ as a cover term refers to a number of acquired language disorders due to brain damage which were caused by an accident, a tumor or vascular problems (cf. BUSSMANN, 1996: 27).