- snjk
-
这个解释起来就麻烦了。而且好像没有Ravem这个人吧?
这个假设追根溯源的话要先从生物学家Haeckel 1868年的假设开始谈起,""ontogeny is a concise recapitulation of phylogeny"" 简言之,Haeckel假设说,从胚胎到一个成熟的个体所经历的几个阶段正好反映了整个进化过程的几个阶段。这一假设后来有所发展,科学家们开始对各个阶段进行了对比。
这个思想被Jacobson引入到语言学界。 简言之,人类从学习一门语言,到掌握一门语言(first language acquisition),再到因为大脑受伤而丢失语言(language loss),这三个过程也反映了语言的历史变化(language change)。Jacobson主要是对发音(phoneme)做的一些研究,发现有的phonemes小孩很快就学会了,而有些phonemes要很长时间才学的会。他认为这种先后顺序也放映了phonemes在人类语言中的发展过程,并且他认为这一发展过程也反映在语法等其他语言方面。
这一想法也被运用到了SLA领域,但我不知道是谁介绍近来的。下面是一段英文描述:
The supposed parallels between language history and first language acquisition (or second language acquisition, for that matter) are usually labelled as the recapitulation hypothesis; the supposed parallelism between language acquisition and language loss as the regression hypothesis.我估计这一假设在SLA的意思应该和语言迁移有点一致(但我不肯定):在学习一门外语的时候,和母语相一致的地方很容易习得,不一样的地方需要时间慢慢学习。而且你应该知道外语学习的U-shape规律,这和Regression Hypothesis也有一致的地方。国内估计这方面的文章比较少,你可以看一下Gass & Selinker (2008)写的SLA的书,很全面,我记得好像提到过这个假设。这本书国内可以弄到,有PDF版本。
- 再也不做稀饭了
-
再现假设 或重演假设
The theory of recapitulation, also called the biogenetic law or embryological parallelism—and often expressed as "ontogeny recapitulates phylogeny"—is a biological hypothesis that in developing from embryo to adult, animals go through stages resembling or representing successive stages in the evolution of their remote ancestors. With different formulations, such ideas have been applied and extended to several fields and areas, including the origin of language, biology, cognition and mental activities,anthropology, education theory and developmental psychology.While some examples of embryonic stages show that superficial features of ancestral organisms exist, the theory of recapitulation itself has been completely disproven within the field of biology. By contrast, there is no consensus against its validity outside of biology; recapitulation theory is still considered plausible and applied by some researchers in fields like Behavioral Development,[8] the study of the origin of language,[9] and others.
Does Second Language Learning Resemble First Language Learning in Children?
This question is relevant to the field of adult SLA from an historical perspective, as child language acquisition became a focus of researchers before that of SLA by adults. Perception precedes production in child L1 acquisition. Foster-Cohen (1999) points to infants" modulation of loudness and pitch of sounds produced as evidence of their awareness of the potential significance of their speech sounds. Child L1 acquisition research documents a systematic process of different patterns and stages of development. From babbling of consonant-vowel sequences to an infant"s first single words (9 mo-1.6 yrs.), and then onto a two-word phase (1.6-2 yrs.) and a spurt of vocabulary during the second year, Foster-Cohen"s research shows the relationship between the steady decrease in babbling as the increase in content words is observed. Perception of sounds that the child cannot yet produce is clearly demonstrated when a parent or adult imitates a child"s mispronunciation and the young child angrily refutes his/her mispronunciation (Foster-Cohen 1999; Smith 1973). Children perceive words before they can produce them, thus deleting syllables (nana for banana) or sounds (mo for more) or substitutions (hostible for hospital; flambilly for family) (Ingram 1986; my daughter"s utterances 1991-94).
A child"s L1 acquisition of syntax begins with content words in the correct word order, then proceeds to a “telegraphic” phase, and onto a multiword phase (see Lightbown and Spada 1999, pp. 7-8). Researchers have devised a method by which to measure children"s utterances: the mean length of utterance (MLU) which averages the number of morphemes over 100 utterances, and is thus a realistic, standardized measure of L1 development (as opposed to a child"s age). Brown (1973) first observed a predictable order of morpheme acquisition in children who were English L2 learners and found that the order of acquisition of the inflectional morphemes studied did not reflect the frequency of usage by the children"s parents. Brown (1973) explored the notions of stress and saliency, syllabicity, and lack of exceptions as possible reasons for a particular order of morpheme acquisition.
Child L1 acquisition research refuted behaviorist theory of language acquisition as imitation and “habits”. Cazden (1972, p. 92) showed evidence of a child applying the regular past tense (-ed) rule to all verbs used, even when adult repeatedly modeled the irregular past-tense verb. Any parent of young children can cite examples of their child"s constructed grammars, even if he/she does not understand linguistic theories of language acquisition. Child L2 acquisition brings into focus the theory of a “critical period” of language learning. Foster-Cohen (1999, p. 7-8) and McLaughlin (1978, p. 99) both define child SLA as sequential acquisition and not simultaneous acquisition. Researchers tend to regard the childhood years between five and nine as a “critical period” when the L1 grammar is in place. In studying child L2 learners, researchers found that the L2 learning environment made a crucial difference for 7-8 year-old L2 learners of French (Selinker, Swain, and Dumas 1975). The absence of L1 speakers in the child"s L2 environment put emphasis on the quality of input in relation to the L2 outcome. McLaughlin (1978b) posited that there is no L1 transfer for the child L2 learner unless the child is isolated from his/her peers of the TL, as in the immersion setting. The theory is that the child recapitulates the L2 grammar as if it were an L1, due to the greater social context the child is within.
There are two other hypotheses concerning child L2 and L1 acquisition that merit mentioning. There is the Regression hypothesis (McLaughlin 1978) which compares L1 acquisition skills with L2 data (see also Ervin-Tripp 1974), and the Recapitulation hypothesis which claims recapitulation of the L2 learning process of a native speaker of the TL. McLaughlin showed the studies of Milon (1974) and Ravem (1968, 1974) as supporting this hypothesis. Juxtaposed to this hypothesis, McLaughlin cites children"s occasional use of L1 structures to solve the “riddle” of L2 structures (Wode 1976). McLaughlin claims a unitary process that characterizes all language acquisition whereby both L1 and L2 learners use the same SLA strategies.
The L1-L2 Hypothesis of Dulay and Burt (1974, 1975) states that child SLA is similar to L1 acquisition. Dulay and Burt claim a mentalist perspective of “creative construction” in which children"s process of reconstructing L2 rules is guided by “universal innate mechanisms” (1974a, p. 37) until a difference between input and production is resolved. This is a universal grammar (UG) approach that emphasizes the mental processes of L2 learners" strategies, regardless of their L1. Morpheme order studies were conducted in response to these hypotheses. Dulay and Burt (1974a) were the first to apply Brown"s L1 findings to child SLA. Since only minimal differences were observed between L2 learners of different L1 backgrounds, production similarity was attributed to universal developmental factors. Lakshmanan (1995) posits that child SLA research provides more insight into biological factors and the constraints of UG principles and the possibility that some form of UG is available for at least a given period of time after L1 acquisition.
As Richards and Sampson (1974) noted, it is practically impossible to understand the relationship between child L1 and L2 acquisition due to various factors that interact in determining the learner"s approximate system. I feel that much more research is needed in the field of child SLA in a variety of environments. These results could shed light on cognitive processes in relation to language acquisition processes. This is a fascinating area of research that is only in its infancy.