L1 Influence on Collocation Processing

Collocation use is undoubtedly critical to more succinct and native-like L2 production. However, even advanced L2 learners sometimes use collocations in ways that deviate from native speakers’ expectations. Previous literature reviews and many

empirical studies (Langacker, 1991; Murao, 2004; Shehata, 2008) all pointed to cross-linguistic transfer, or L1 influence, as one of the main factors. The two terms were used interchangeably with ‘cross-linguistic influence’ earlier presented by Kellerman and Smith (1986) because of the influence of another linguistic system or L2-related aspects of language loss. L2 learners might assume there was a perfect match between an L2 target word and a counterpart in their first language. This often caused a quite pervasive phenomenon of ‘false friends’ (Laufer, 1991) in L2 learner’s collocation use.

As indicated by many cognitive linguists, meaning and conceptual realizations were dynamic and complicated. Words do not have static senses. Word meaning, from the psycholinguistic view, involves social, cultural, cognitive, or cross-linguistic associations. It is simply to see that an L1 concept is taken in such a sophisticated way and it influences the understanding of L2 concept. Nevertheless, it is fuzzy to determine to what extent L1 and L2 concepts are closely similar or significantly different. Even if L1 and L2 concepts are similar, it is still unsure what concepts the learner activates to construct L2 meanings. Wolter (2006) claimed that post-pubescent L2 learners had a considerably sophisticated L1 lexicon. This lexical structure was further bolstered by a similarly sophisticated understanding of conceptual and experiential knowledge that indicated to the learner how the concepts associated with the forms in the language could be connected to each other. There was no particular reason to believe that they acquired knowledge of the linguistic structure when they initially approached an L2. On the contrary, it seemed highly unlikely that they began structuring L2 lexical knowledge from scratch when presented with new L2 lexical items. Wolter (2006) further elaborated it by using the example of Japanese learners of English’s lexical concept. Japanese learners of English depicted the size of a room as “narrow” or “wide”, rather than “small” or “big”. They indicated a small room as a

narrow room. The Japanese word for “small” (chiisai) was not normally collocated with “room” (heya) as frequently as “narrow” (semai). The L1 conceptualization was practical to construct L2 lexical networks, but it sometimes generated erroneous information and was culture specificity concerned.

Kellerman and Smith (1986) believed that learners could make use of their native language in their target language production where they had deficient target language proficiency. How learners relied on their native language to produce their target language was explained in terms of L1 influence within the framework of learners’

perception of the distance between the native language and the target language.

Moreover, the degree of markedness of L1 collocates had decisive influence for linguistic transfer. If learners perceived the distance between the target language and the native language to be greater, the influence of transfer was lower, while if they perceived distance between them to be a little, then they transferred the relevant items or structured from their native language. The other framework, markedness, referred to such features as irregularity, infrequency, semantic opaqueness, and unmarkedness as regularity, frequency, productiveness, semantic transparency, and coreness in his classification. If the markedness level was too high, transfer could be impeded, and if the markedness level was not high, it could function as language distance. Thus, the interaction of the two factors—distance and markedness, determined whether transfer was activated or not.

To prove L1 influence on learners’ target language production, Kellerman and Smith (1986) conducted two experiments: one concerned the transferability of idiomatic expressions and the other concerned native speakers’ intuitions about semantic space. Experiment One recruited 72 Dutch freshmen, sophomores and juniors of EFL (English for Foreign Language) learners. They were given 20 idiomatic expressions and asked them to examine whether they were correct English

or not. The result showed that idioms were not subject to cross-linguistic transfer, probably because they were marked expressions for them. Experiment Two included 81 Dutch learners of English to inquire the effect of L1 in terms of coreness and unmarkedness of words with peripheral and marked meaning. They surmised core meaning and unmarked meaning could be transferred to L2. The participants were given 17 sentences with core and non-core senses of the word like “broken” (i.e.

break), “break his leg” and “break the record” and judged which combinations were correct in English. The result confirmed Kellerman’s postulation and he argued that there were some factors such as core and frequency affecting transferability from mother tongue to L2. Thus, he strongly suggested that transfer from the native language played an important role in acquisition of the target language.

Many empirical studies have explored cross-linguistic influence on L2 learners’

collocation development. It included a wide range of themes, such as collocation knowledge, miscollocation, influential factors, correlations between collocation competence, language distance and congruency, etc. Table 2.4.1 shows a synopsis of on collocation studies on cross-linguistic influence that may help illuminate the various issues and findings in notions, features, and development of L2 collocation.

Table 2.4.1 Synopsis of Collocation Studies on Cross-linguistic Influence

Research Purpose Method Finding

How learners relied on L1 to produce L2 was explained within because most of them were marked expressions.

Fayez-Hussein embedded in the translation task.

Lennon (1996) To examine lexical choice of polysemous their ideas of core meaning of polysemous verbs and might be derailed by translation equivalents.

They tended to transfer their L1 to the target production or

overgeneralized the use of some common verbs such as put, go, and take, so that their deficient much higher than those of other amplifiers. When L2 learners produced collocations in their writing, these word combinations mostly were congruent to their L1.

Caroli (1998) To examine the relations between learners at the early stage of vocabulary development tended to use L1 features to select L2 collocates and the Italian equivalent collocations were easily acquired.

Liu (1999) To identify and

As with VN miscollocations, it was attributed to three main reasons: (a) L1 interference, (b) false concepts hypothesized, and (c) lack of knowledge about collocational restrictions in semantic-related

lexemes such as synonyms, tended to apply L1 to select possible collocations for an L2 under the great influence of L1.

Teachers should focus on teaching the collocations occurring in higher frequency and those congruent in learners’ L1 and L2.

Wolter (2006) To explain how L2 learners draw on L1

L1 lexical knowledge could be both a help and hindrance when making L2 word combinations. Also, the

Chen (2007) To study the

VN collocation was found to be the most demanding lexical

Lin (2010) To study Taiwanese and Chinese learners’ verb and L1 transfer were the most frequent type and cause. Taiwanese

acquisition; it was more difficult to acquire incongruent collocations;

and L2 collocations, once stored in memory, were processed