14.4 Is there a critical age for second-language learning?
14.4 Is there a critical age for second-language learning?
In Chapter 4, on wild and isolated children, the concept of a critical age for first-language learning was raised. Recall, if you will, that although solid evidence was lacking, some theorists hypothesized that there was an age (puberty, for example) beyond which it would be impossible to acquire a first language. Brain changes were suggested as a possible explanation for such a psychological barrier.
Adults can learn a second language
It is reasonable to ask the same question about the acquisition of a second language. Is there any barrier to the learning of a second language and, if so, at what age does this barrier become operational? As far as adult second-language learning is concerned, we have the common observation that a very great number of adults do, in fact, learn the syntax of other languages perfectly. There are those who speak second languages so well that, on the basis of the grammar alone (not the pronunciation, which we shall deal with shortly), they would be judged native speakers.
No demonstrated critical age for learning syntax
There are, however, studies which demonstrate a differential effect for the age at which acquisition of syntax began. Patkowski (1980) had native speakers of English rate the syntax of transcripts of spontaneous speech from immigrants to the USA who had entered before or after the age of 15. Transcripts were used to remove any possible influence of accent on the raters. The scores showed two very distinguishable groups: those who arrived before 15 years of age scored very high, while those arriving after
scored lower. Johnson and Newport (1989) found the same effect when they had native speakers of Korean and Chinese rate the grammaticality of English sentences. The earlier the age of arrival, the better the subjects were at deter- mining the ungrammaticality of English sentences. A similar result was reported by Mayberry and Lock (2003): early exposure to language, no matter wheter spoken or signed, contributed the most to later success in learning a second language.
On the other hand, research exists providing data to refute the claim of a critical period for the acquisition of grammar. The same Mayberry and Lock (2003) study still shows that adult learners can achieve a near-native com- petence of second-language grammar. In another study using a grammatical- ity judgement test in which native speakers of French were compared with high-level learners of French, no differences were detected between the two groups on their test scores or in the process of how they judged the syntax (Birdsong, 1992). Further research (van Wuijtswinkel, 1994; White and Genesee, 1996, Birdsong and Molis, 2001) shows that even learners who begin to acquire a language after puberty can reach native-speaker levels. It is safe to affirm the view that there is no critical age in terms of acquiring the syntax of a second language.
Critical age for pronunciation
This brings us to pronunciation. Is it possible to learn a second language so well that one truly sounds like a native speaker? Mack (1986) and Perani et al. (2003) argue that even if highly competent second-language speakers seem to perform on a native level, experimental tasks will reveal the difference both in grammar and in pronunciation. Sebastián-Gallés and Bosch (2001) suppose that even pre-school age is already too late for aquiring near-native phonology. One psycholinguist, Thomas Scovel, has claimed that no adult can ever be successful in that regard. ‘The critical period for accentless speech simply means that adults will never learn to pass themselves off as native speakers phonologically . . . ’ (Scovel, 1988, p. 65). He describes this as the ‘Joseph Conrad phenomenon’, after the famous novelist and master of English prose, who, a native speaker of Polish, did not even begin to study English until he was 20. Scovel has in mind a certain category of adult second-language speakers: those who have mastered the grammatical and communicative complexities of another language but still speak with an accent.
We could add to this class two European-born US Secretaries of State,
Henry Kissinger (under President Nixon) and Zbigniew Brzezinski (under President Carter). Both of these men speak English that is heavily accented with their native languages, German and Polish, respectively. In other respects, they excel in the English language. Kissinger came to the USA when he was 15 and Brzezinski came when he was 10 years old. While Kissinger was beyond our posited motor-skill critical age of 13 years, discussed
earlier in this chapter, Brzezinski was not, but still he spoke with an accent. A more recent example of a foreign-born United States official is Arnold Schwarzenegger, who became governor of California in 2003. Born in Austria, he moved to the United States when he was 21. At present he appears to speak English with near-native speaker syntax, although his strong German accent may often obscure that fact!
However, there is a growing body of research that challenges any strong criterion of a critical period for accentless acquisition of a language such as the one that Scovel proposes. Some adults do appear to pick up accentless speech. Neufeld (1978) trained adult learners in the pronunciation of Chinese and Japanese. Judged by native speakers, half of them were able to pass as native. Furthermore, in two recent studies on this question, Bongaerts and associates (Bongaerts et al., 1995, 1997) argue that there were flaws in studies by researchers who claimed that late learners could not acquire correct pronunciation. The flaw was that in subject selection advanced learners were not included. Correcting for this error by including highly successful learners, Bongaerts and associates found that they could indeed pass for native speakers on a number of criteria. They suggest that ‘a very high motivation . . . continued access to target language input . . . [and] intensive instruction in the perception and in the production of the speech sounds’ (pp. 462–3) were instrumental for these late language learners’ acquisition of native-like pronunciation.
Matsui (2000) refers to a number of most recent studies proving that
some adults can pick up near-native foreign language pronunciation even without explicit instruction. Current neurological research of blind people suggests that the human brain retains its flexibility throughout life, and changes to adapt language challenges even for older people. Special brain exercises are even reported to help people regain language functions despite various deficits (Ariniello, 2000). Perception training, for example, resulted in long-term improvements in Japanese adults’ perception and production of English words with ‘r’ and ‘l’ sounds (Akahane-Yamada et al., 1996), and in Korean adults’ perception and production of English vowels (Yeon, 2003). Thus, while we would agree with Scovel that in second-language acquisi- tion there is no critical age for syntax, we cannot agree that there is an absolute critical age for pronunciation. Native-speaker pronunciation may
well be achieved by some adults.