Автор работы: Пользователь скрыл имя, 25 Октября 2010 в 12:16, Не определен
Рассказ
These conclusions were derived from Ausubel s cognitive perspective, which ran counter to prevailing behavioristic paradigms on which the Audiolingual Method was based. But Ausubel's criticism may have been ahead of its time, for in 1964 few teachers were ready to entertain doubts about the widely accepted method. (See the vignette at the end of this chapter for a further discussion of the Audiolingual Method.)
By the 1970s and 1980s, criticism of earlier direct analogies between first and second language acquisition had reached full steam. Stern (1970), Cook (1973, 1995), and Schachter (1988), among others, addressed the inconsistencies of such analogies, but at the same time recognized the legitimate similarities that, if viewed cautiously, allowed one to draw some constructive conclusions about second language learning.
The comparison of first and second language acquisition can easily be oversimplified. At the very least, one needs to approach the comparison by first considering the differences between children and adults. It is, in one sense, illogical to compare the first language acquisition of a child with the second language acquisition of an adult (see Schachter 1988; Scovel 1999). This involves trying to draw analogies not only between first and second language learning situations but also between children and adults. It is much more logical to compare first and second language learning in children or to compare second language learning in children and adults. Nevertheless, child first language acquisition and adult second language acquisition are common and important categories of acquisition to compare. It is reasonable, therefore, to view the latter type of comparison within a matrix of possible comparisons. Figure 3.1 represents four possible categories to compare, defined by age and type of acquisition. Note that the vertical shaded area between the child and the adult is purposely broad to account for varying definitions of adulthood. In general, however, an adult is considered to be one who has reached the age of puberty.
| CHILD | ADULT | L1 = First language
L2 = Second language С = Child A = Adult | ||
| L1 | C1 | A1 | ||
| L2 | C2 | A2 |
Figure 3.1. First and second language acquisition in adults and children
Cell A1 is clearly representative of an abnormal situation. There have been few recorded instances of an adult acquiring a first language. In one widely publicized instance, Curtiss (1977) wrote about Genie, a thirteen-year-old girl who had been socially isolated and abused all her life until she was discovered, and who was then faced with the task of acquiring a first language. Accounts of "wolf children" and instances of severe disability fall into this category. Since we need not deal with abnormal or pathological cases of language acquisition, we can ignore category Al. That leaves three possible comparisons:
In the C1-C2 comparison (holding age constant), one is manipulating the language variable. However, it is important to remember that a two-year-old and an eleven-year-old exhibit vast cognitive, affective, and physical differences, and that comparisons of all three types must be treated with caution when varying ages of children are being considered. In the C2-A2 comparison, one is holding language constant and manipulating the differences between children and adults. Such comparisons are, for obvious reasons, the most fruitful in yielding analogies for adult second language classroom instruction. The third comparison, C1-A2, unfortunately manipulates both variables. Many of the traditional comparisons were of this type; however, such comparisons must be made only with extreme caution because of the enormous cognitive, affective, and physical differences between children and adults.
Much of the focus of the rest of this chapter will be made on C2-A2 and C1-C2 comparisons. In both cases, comparisons will be embedded within a number of issues, controversies, and other topics that have attracted the attention of researchers interested in the relationship of age to acquisition.
Most discussions about age and acquisition center on the question of whether there is a critical period for language acquisition: a biologically determined period of life when language can be acquired more easily and beyond which time language is increasingly difficult to acquire. The Critical Period Hypothesis (CPH) claims that there is such a biological timetable. Initially the notion of a critical period was connected only to first language acquisition. Pathological studies of children who failed to acquire their first language, or aspects thereof, became fuel for arguments of biologically determined predispositions, timed for release, which would wane if the correct environmental stimuli were not present at the crucial stage. We have already seen, in the last chapter, that researchers like Lenneberg (1967) and Bickerton (1981) made strong statements in favor of a critical period before which and after which certain abilities do not develop. Second language researchers have outlined the possibilities of extrapolating the CPH to second language contexts (see Bialystok 1997; Singleton & Lengyel 1995;Scovel 1988,1999 for useful summaries).The "classic" argument is that a critical point for second language acquisition occurs around puberty, beyond which people seem to be relatively incapable of acquiring a second language. This has led some to assume, incorrectly, that by the age of twelve or thirteen you are "over the hill" when it comes to the possibility of successful second language learning. Such an assumption must be viewed in the light of what it really means to be "successful" in learning a second language, and particularly the role of accent as a component of success. To examine these issues, we will first look at neurological and phonological considerations, then examine cognitive, affective, and linguistic considerations.
One of the most promising areas of inquiry in age and acquisition research has been the study of the function of the brain in the process of acquisition (see Schumann 1998, Jacobs & Schumann 1992, and Scovel 1988 for synopses). How might neurological development affect second language success? Does the maturation of the brain at some stage spell the doom of language acquisition ability?
Some scholars have singled out the lateralization of the brain as the key to answering such a question. There is evidence in neurological research that as the human brain matures, certain functions are assigned, or "lateralized," to the left hemisphere of the brain, and certain other functions to the right hemisphere. Intellectual, logical, and analytic functions appear to be largely located in the left hemisphere, while the right hemisphere controls functions related to emotional and social needs. (See Chapter 5 for more discussion of left- and right-brain functioning.) Language functions appear to be controlled mainly in the left hemisphere, although there is a good deal of conflicting evidence. For example, patients who have had left hemi-spherectomies have been capable of comprehending and producing an amazing amount of language (see Zangwill 1971: 220). But in general, a stroke or accident victim who suffers a lesion in the left hemisphere will manifest some language impairment, which is less often the case with right hemisphere lesions.
While questions about how language is lateralized in the brain are interesting indeed, a more crucial question for second language researchers has centered on when lateralization takes place, and how that lateralization process affects language acquisition. Eric Lenneberg (1967) and others suggested that lateralization is a slow process that begins around the age of two and is completed around puberty. During this time the child is neurologically assigning functions little by little to one side of the brain or the other; included in these functions, of course, is language. And it has been found that children up to the age of puberty who suffer injury to the left hemisphere are able to relocalize linguistic functions to the right hemisphere, to "relearn" their first language with relatively little impairment. Thomas Scovel (1969) extended these findings to propose a relationship between lateralization and second language acquisition. He suggested that the plasticity of the brain prior to puberty enables children to acquire not only their first language but also a second language, and that possibly it is the very accomplishment of lateralization that makes it difficult for people to be able ever again to easily acquire fluent control of a second language, or at least to acquire it with what Alexander Guiora et al. (1972a) called "authentic" (nativelike) pronunciation.
While Scovel's (1969) suggestion had only marginal experimental basis, it prompted him (Scovel 1988) and other researchers (e.g., Singleton & Lengyel 1995) to take a careful look at neurological factors in first and second language acquisition. This research considered the possibility that there is a critical period not only for first language acquisition but also, by extension, for second language acquisition. Much of the neurological argument centers on the time of lateralization. While Lenneberg (1967) contended that lateralization is complete around puberty, Norman Geschwind (1970), among others, suggested a much earlier age. Stephen Krashen (1973) cited research to support the completion of lateralization around age five. Krashen's suggestion does not grossly conflict with research on first language acquisition if one considers "fluency" in the first language to be achieved by age five. Scovel (1984:1) cautioned against assuming, with Krashen, that lateralization is complete by age five. "One must be careful to distinguish between emergence' of lateralization (at birth, but quite evident at five) and 'completion' (only evident at about puberty)." If lateralization is not completed until puberty, then one can still construct arguments for a critical period based on lateralization.
One of the most compelling arguments for an accent-related critical period came from Thomas Scovel's (1988) fascinating multidisciplinary review of the evidence that has been amassed. Scovel cited evidence for a sociobiological critical period in various species of mammals and birds. (Others, such as Neapolitan et al. 1988, had drawn analogies between the acquisition of birdsong and human language acquisition.) Scovel's evidence pointed toward the development of a socially bonding accent at puberty, enabling species (a) to form an identity with their own community as they anticipate roles of parenting and leadership, and (b) to attract mates of "their own kind" in an instinctive drive to maintain their own species.
If the stabilization of an accepted, authentic accent is biologically preprogrammed for baboons and birds, why not for human beings? The socio-biological evidence that Scovel cited persuades us to conclude that native accents, and therefore "foreign" accents after puberty, may be a genetic leftover that, in our widespread human practice of mating across dialectal, linguistic, and racial barriers, is no longer necessary for the preservation of the human species. "In other words," explained Scovel (1988: 80), "an accent emerging after puberty is the price we pay for our preordained ability to be articulate apes."
Following
another line of research,Walsh and Diller (1981:18) concluded that different
aspects of a second language are learned optimally at different ages:
Lower-order
processes such as pronunciation are dependent on early maturing and
less adaptive macroneural circuits, which makes foreign accents difficult
to overcome after childhood. Higher-order language functions, such as
semantic relations, are more dependent on late maturing neural circuits,
which may explain why college students can learn many times the amount
of grammar and vocabulary that elementary school students can learn
in a given period of time.
This conclusion lends support for a neurologically based critical period, but principally for the acquisition of an authentic (nativelike) accent, and not very strongly for the acquisition of communicative fluency and other "higher-order" processes. We return to the latter issue in the next section.
Yet another branch of neurolinguistic research focused on the role of the right hemisphere in the acquisition of a second language. Obler (1981:58) noted that in second language learning, there is significant right hemisphere participation and that "this participation is particularly active during the early stages of learning the second language." But this "participation" to some extent consists of what we will later (Chapter 5) define as "strategies" of acquisition. Obler cited the strategy of guessing at meanings, and of using formulaic utterances, as examples of right hemisphere activity. Others (Genesee 1982; Seliger 1982) also found support for right hemisphere involvement in the form of complex language processing as opposed to early language acquisition.
Genesee (1982: 321) concluded that "there may be greater right hemisphere involvement in language processing in bilinguals who acquire their second language late relative to their first language and in bilinguals who learn it in informal contexts." While this conclusion may appear to contradict Obler's statement above, it does not. Obler found support for more right hemisphere activity during the early stages of second language acquisition, but her conclusions were drawn from a study of seventh-, ninth-, and eleventh-grade subjects—all postpubescent. Such studies seem to suggest that second language learners, particularly adult learners, might benefit from more encouragement of right-brain activity in the classroom context. But, as Scovel (1982: 324-325) noted, that sort of conclusion needs to be cautious, since the research provides a good deal of conflicting evidence, some of which has been grossly misinterpreted in "an unhappy marriage of single-minded neuropsychologists and double-minded educationalists. . . . Brain research ... will not provide a quick fix to our teaching problems."
Some adults have been known to acquire an authentic accent in a second language after the age of puberty, but such individuals are few and far between. Anthropologist Jane Hill (1970) provided an intriguing response to Scovel's (1969) study by citing anthropological research on non-Western societies that yielded evidence that adults can, in the normal course of their lives, acquire second languages perfectly. One unique instance of second language acquisition in adulthood was reported by Sorenson (1967), who studied the Tukano culture of South America. At least two dozen languages were spoken among these communities, and each tribal group, identified by the language it speaks, is an exogamous unit; that is, people must marry outside their group, and hence almost always marry someone who speaks another language. Sorenson reported that during adolescence, individuals actively and almost suddenly began to speak two or three other languages to which they had been exposed at some point. Moreover, "in adulthood [a person] may acquire more languages; as he approaches old age, field observation indicates, he will go on to perfect his knowledge of all the languages at his disposal" (Sorenson 1967: 678). In conclusion, Hill (1970: 247-248) suggested that
the language acquisition situation seen in adult language learners in the largely monolingual American English middle class speech communities ... may have been inappropriately taken to be a universal situation in proposing an innatist explanation for adult foreign accents. Multilingual speech communities of various types deserve careful study... We will have to explore the influence of social and cultural roles which language and phonation play, and the role which attitudes about language play, as an alternative or a supplement to the cerebral dominance theory as an explanation of adult foreign accents.
Hill's challenge was taken up in subsequent decades. Flege (1987) and Morris and Gerstman (1986), for example, cited motivation, affective variables, social factors, and the quality of input as important in explaining the apparent advantage of the child. However, both Long (1990b) and Patkowski (1990) disputed such conclusions and sided with Scovel in their relatively strong interpretation of an age-related critical period for first and second language acquisition.
Implicit in the comments of the preceding section is the assumption that the emergence of what we commonly call "foreign accent" is of some importance in our arguments about age and acquisition. We can appreciate the fact that given the existence of several hundred muscles (throat, larynx, mouth, lips, tongue, and others) that are used in the articulation of human speech, a tremendous degree of muscular control is required to achieve the fluency of a native speaker of a language. At birth the speech muscles are developed only to the extent that the larynx can control sustained cries. These speech muscles gradually develop, and control of some complex sounds in certain languages (in English the r and l are typical) is sometimes not achieved until after age five, although complete phonemic control is present in virtually all children before puberty.
Research on the acquisition of authentic control of the phonology of a foreign language supports the notion of a critical period. Most of the evidence indicates that persons beyond the age of puberty do not acquire what has come to be called authentic (native-speaker) pronunciation of the second language. Possible causes of such an age-based factor have already been discussed: neuromuscular plasticity, cerebral development, sociobiological programs, and the environment of sociocultural influences.
It is tempting immediately to cite exceptions to the rule ("My Aunt Mary learned French at twenty-five, and everyone in France said she sounded just like a native"). These exceptions, however, appear to be (a) isolated instances or (b) only anecdotally supported. True, there are special people who possess somewhere within their competence the ability to override neurobiological critical period effects and to achieve a virtually perfect nativelike pronunciation of a foreign language. But in terms of statistical probability (see Scovel 1988), it is clear that the chances of any one individual commencing a second language after puberty and achieving a scientifically verifiable authentic native accent are infinitesimal.
So, where do we go from here? First, some sample studies, spanning two decades, will serve as examples of the kind of research on adult phonological acquisition that appears to contradict Scovel's "strong" CPH.
Gerald Neufeld (1977, 1979, 1980) undertook a set of studies to determine to what extent adults could approximate native-speaker accents in a second language never before encountered. In his earliest experiment, twenty adult native English speakers were taught to imitate ten utterances, each from one to sixteen syllables in length, in Japanese and in Chinese. Native-speaking Japanese and Chinese judges listened to the taped imitations. The results indicated that eleven of the Japanese and nine of the Chinese imitations were judged to have been produced by "native speakers." While Neufeld recognized the limitations of his own studies, he suggested that "older students have neither lost their sensitivity to subtle differences in sounds, rhythm, and pitch nor the ability to reproduce these sounds and contours" (1979: 234). Nevertheless, Scovel (1988: 154-159) and Long (1990b: 266-268) later pointed out glaring experimental flaws in Neufeld's experiments, stemming from the methodology used to judge "native speaker" and from the information initially given to the judges.
In more recent years, Moyer (1999) and Bongaerts, Planken, and Schils (1995) have also challenged the strong version of the CPH. Moyer's study with native English-speaking graduate students of German upheld the strong CPH: subjects' performance was not judged to be comparable to native speakers of German. The Bongaerts et al. study reported on a group of adult Dutch speakers of English, all late learners, who recorded a monologue, a reading of a short text, and readings of isolated sentences and isolated words. Some of the non-native performances, for some of the trials, were judged to have come from native speakers. However, in a later review of this study, Scovel (1997:213) carefully noted that it was also the case that many native speakers of English in their study were judged to be nonnative! The earlier Neufeld experiments and these more recent studies have thus essentially left the strong CPH unchallenged.