当前位置:首页 >> 文学研究 >>

2010 Lexical gap-filling mechanisms in foreign language writing


Available online at www.sciencedirect.com

System 38 (2010) 529e538

www.elsevier.com/locate/system

Lexical gap-?lling mechanisms in foreign language writing
? Ma. Pilar Agust?n Llach*
? ?o, Universidad de La Rioja, Department of Modern Philologies, C/ San Jose de Calasanz s/n, 26004 Logron La Rioja, Spain Received 29 March 2010; revised 19 July 2010; accepted 18 September 2010

Abstract The present paper intends to investigate the lexical gap-?lling behaviour of primary and secondary Spanish learners of English. When there is a mismatch between the learners lexical knowledge and their communicative needs, then a lexical gap arises. Learners resort to different mechanisms to compensate for that lack of lexical knowledge. Basically, we distinguish among several L1- and L2-based mechanisms. We analysed the lexical gap-?lling behaviour of 203 learners in their written compositions. We collected data when students were in 4th grade of Primary Education and four years later when they were in 8th grade. Results revealed that as learners progress in their learning of the foreign language they gradually abandon use of the L1 and substitute L1based mechanisms such as borrowings, coinages and calques by other L2-based ones such as semantic and formal approximations, circumlocutions, and use of general words. We observed that L2 pro?ciency interrelates with L1 and L2 in?uence in different ways depending on the speci?c mechanism used, so that some mechanisms which are frequent at 4th grade decrease their presence, basically borrowing, and some other tend to increase, basically L2-based mechanisms as linguistic pro?ciency, awareness and cognitive development increase. ? 2010 Elsevier Ltd. All rights reserved.
Keywords: Lexical gap; Lexical innovation; EFL Spanish learners

1. Introduction This paper examines the patterns of gap-?lling behaviour in vocabulary production across grades. When mismatches arise between learners’ communicative needs and the lexical tools available, they have to resort to different mechanisms to cover that lexical gap and compensate for the breakdown due to insuf?cient competence (cf. ? ? Dewaele, 1998; Gonzalez Alvarez, 2004; Poulisse, 1993; Rababah, 2002). These mechanisms have been frequently known as communication or compensatory strategies. These compensatory strategies have a lexical character, since lexis is crucial and essential for communication. A misinterpretation or absence of a lexical term generally leads to impairment or inhibition of communication. 2. Lexical gap-?lling Lexical gaps are the consequence of a lexical problem. Lexical problems arise when the learner has a message to communicate which contains some information he/she is unable to code in terms of lexical items. It may be that the
* Tel.: ?34 941299435; fax: ?34 941299419. E-mail address: Maria-del-pilar.agustin@unirioja.es. 0346-251X/$ - see front matter ? 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.system.2010.09.009

530

Ma.P. Agust??n Llach / System 38 (2010) 529e538

learner lacks the knowledge of the target lexical item, it does not exist in his/her lexicon because he/she has failed to learn the item, or because the concept is not lexicalised in the target language (TL), or he/she may be unable to retrieve that lexical item from memory since he/she may have temporarily lost access to that item (cf. Poulisse, 1993: 178). Contrastive linguistics has evidenced lexicosemantic divergences among languages, especially if they belong to different typological families, such as English (Germanic origin) and Spanish (Latin origin). These lexical divergences are prone to be the source of lexical gaps for language learners. From the taxonomies on compensatory strategies in the literature (Poulisse, 1993; Salazar Campillo, 2006) we can distinguish among reconceptualization strategies such as circumlocution and literal translation, and substitution strategies such as borrowing, foreignizing, and use of all purpose words. Learners can draw on their previous linguistic knowledge, e.g., L1, or on their existing TL knowledge. Most research on communication strategies considers these strategies as a conscious and voluntary behaviour (cf. Poulisse, 1993; Rababah, 2002). However, we believe that lexical gap-?lling mechanisms can also be unconscious. This means that learners are not aware of their lack of lexical knowledge and search in their minds for what they think is the correct target lexical item. In fact, in the examples of the present study these mechanisms do not always result from conscious planning or application. The difference between the two types of lexical gap-?lling mechanisms, voluntary and unconscious, lies in the fact that the former may imply a message adjustment with meaning reduction, i.e., circumlocution, whereas the latter may bring forth a change in meaning and even cause communication disruption, i.e., substitution of target lexical item. Whichever, in both cases learners who use mechanisms to cover lexical gaps do not abandon communication, but remain active in the communication process thanks to these means. One of these conscious vocabulary compensatory strategies used in solving a vocabulary problem or de?ciency is circumlocution. It consists in describing the salient features of the object, or action the learners cannot name in the foreign language (FL) (Salazar Campillo, 2006). Circumlocutions are classi?ed depending on the features selected for description. Thus, we distinguish among function, superordinate, description of physical characteristics of object, and location. Salazar Campillo (2006) concludes that circumlocution is a useful and broadly used compensatory strategy that facilitates message transmission and the bridging of lexical gaps. Nevertheless, her study presents a main shortcoming which concerns the elicitation procedure. She had learners place physical objects in different locations and had learners negotiate the meaning of unknown terms. This task design would have favoured the use of circumlocution. In this sense, Poulisse (1993: 164) claims that the use of compensatory strategy type is task speci?c, so a certain task will promote the use of some compensatory strategies or other. ? ? Other compensatory strategies are characterized by the use of lexical inventions. Gonzalez Alvarez (2004: 1) claims that lexical innovations are coined to ?ll gaps in the limited L2 lexicon in the learners’ attempt to overcome communication dif?culties, and in this sense, they frequently appear in the literature as compensatory strategies. When facing a lexical problem FL learners make a creative use of language. Depending on the products of lexical innovation and the underlying mechanisms and processes thereby evidenced, researchers distinguish between interlingual and intralingual lexical innovations. Intralingual sub-categories of lexical inventions include slips of the tongue, overgeneralizations (Blum and Levenston, 1977; Dewaele, 1998), circumlocutions (Rababah, 2002; Salazar Campillo, 2006), or use of all purpose words (Rababah, 2002). Interlingual categories of lexical inventions refer to calques generated from transfer from their L1 and other known FLs (Dewaele, 1998: 478e484; see also Liaw, 1996; Rababah, 2002) such as foreignizing (Liaw, 1996; Poulisse, 1993; Ringbom, 1986; Singleton, 1987), word coinage (Clark, 1980; Kocoglu, 1997; Liaw, 1996; Rababah, 2002), and transfer or codeswitching (Blum and Levenston, 1977; Poulisse, 1993; Rababah, 2002). Compensatory strategy use or lexical gap-?lling behaviour seems to be determined by TL pro?ciency (Celaya and ? ? Ruiz de Zarobe, 2008; Gonzalez Alvarez, 2004; Naves et al., 2005; Poulisse, 1993), the task to be accomplished (Poulisse, 1993), and the cognitive demands of each strategy (Poulisse, 1993). The more cognitively demanding the compensatory strategy is, the more effective it is in communicative terms (Poulisse, 1993). Communicative effectiveness seems to be a crucial element in research about compensatory strategies. In this sense, Poulisse (1993) points to reduction strategies such as circumlocution as the most cognitively demanding and therefore most effective ones, and to substitution strategies, such as borrowing from the L1, as the least cognitively demanding, hence the least effective. Investigating lexical gap-?lling mechanisms can be valuable in revealing the cognitive processes underling vocabulary acquisition and use. In this sense, attempts at identifying lexical compensatory strategies are relevant for research. In the present paper, we set out to study the ways in which learners deal with the dif?culties they encounter during the course of communication when their lexical resources are inadequate. A taxonomy of lexical gap-?lling

Ma.P. Agust??n Llach / System 38 (2010) 529e538

531

mechanisms or strategies is provided and young FL learners’ written compositions explored for the categories of this taxonomy. 3. Research questions This study does not intend to be exhaustive, but just exploratory in nature. The native language and the already existing TL lexical resources serve as sources for those lexical gap-?lling mechanisms. More speci?cally, our research questions are: 1. Are L1-based mechanisms more frequent than L2-based mechanisms to ?ll the lexical gaps of 4th graders when writing in the foreign language? 2. Are L1-based mechanisms more frequent than L2-based mechanisms to ?ll the lexical gaps of 8th graders when writing in the foreign language? 3. Are there differences across grades and pro?ciency levels concerning L1- and L2-based mechanisms to ?ll lexical gaps when writing in the foreign language?

4. Method This is a longitudinal study that stretches over several years with two moments of data collection. This intends to be basically a descriptive study of the lexical ?lling behaviour of learners at two different moments of their second language acquisition (SLA) process. 4.1. Participants The writings of 203 Spanish EFL learners are followed over a four-year period and scrutinized for L1 and L2-based mechanisms of lexical gap-?lling with two measuring moments along their language acquisition process. First, data were collected when learners were in 4th grade of primary education and had received around 419 h of instruction in English. At this time learners were between 9 and 10 years of age. Four years later, when the same learners were in grade 8 and after 839 h of instruction, we proceeded to collect our second data set. At this second time learners were between 13 and 14 years of age. We had students perform a pro?ciency test consisting in a cloze test and in a reading comprehension test. Results revealed that as learners pass grade they also gain in pro?ciency. We performed a test of means comparison between the cloze test scores of informants in 4th grade and later in 8th grade. The distribution of the sample did not meet normality assumptions, therefore, a non-parametric test of means comparison was carried out. Since the sample contained the same number of informants at both testing times, we performed a paired Wilcoxon signed rank test for two related samples. The test revealed that 8th graders show signi?cantly better results for both the cloze and the reading comprehension test (Z ? ?11.122, p < .001 for the cloze test and Z ? ?10.599, p < .001 for the reading test). Table 1 summarises information about informants’ characteristics concerning their age, L1, exposure to the TL, and pro?ciency tests results. 4.2. Instruments of data collection Participants were required to write a timed composition in English. They had 30 min to complete the task. The instructions for the task were given in Spanish so as to avoid any comprehension problems. Learners had to write a letter to a prospective English host family, where they introduced themselves and talked about their family, friends,
Table 1 Participants’ characteristics concerning their grade, age, exposure to the TL, L1 and cloze and reading test scores. N 203 203 Grade 4th 8th Age 9e10 13e14 Hours of instruction (aprox.) 419 839 L1 Spanish Spanish Cloze test 2.86 5.44 Reading test 1.62 4.00

532

Ma.P. Agust??n Llach / System 38 (2010) 529e538

school, hobbies and any other thing they considered interesting about themselves. Learners had no limitations apart from the time and the fact that they had to write the letter in English. 4.3. Procedures and analyses Compositions were written in the classroom during a regular English session. Learners were not allowed to use any books, notes, or dictionaries, nor to ask for any help. Once data were collected, compositions were converted into computer-readable ?les and scrutinized for mechanisms of lexical gap-?lling. We identi?ed three types of mechanisms based on the L1, namely borrowing, foreignizing, and calque, and four other types based on the L2: formal approximation, semantic approximation, use of all purpose words and circumlocution (cf. Rababah, 2002). Each category is explained in more detail in Table 2 below. Statistical analyses were performed in order to determine descriptive statistics for each lexical gap-?lling mechanism at each grade tested such as mean values, standard deviations, maximum and minimum values. Moreover, inferential statistics were also carried out to ?nd out signi?cant differences in lexical gap-?lling across the grades tested. The SPSS 15.0 was used to calculate descriptive and inferential statistics.
Table 2 Lexical gap-?lling mechanisms. Lexical gap-?lling mechanism Borrowing Explanation Insertion of an L1 word into the L2 syntax, without any adaptation (cf. Celaya and Torras, 2001) Adaptation of an L1 word to the L2 orthography or morphology in order to make it sound or look English (Ringbom, 1983; Celaya and Torras, 2001) Examples  My grandmother is coja (Eng. lame)  My father is big and lento (Eng. slow)  My rabbit is small, very divert (Sp. divertido, Eng. funny).  In mai house is famili: fatter, matter, tater and mai (Sp. tato, Eng. familiar for “brother”).  My table study is blue and big (literal translation from mesa de estudio, Eng. desk).  My fathers are Marta and Juan (literal translation from padres, Eng. parents).  My class is big (class for “classroom”).  I am tall and my hear is very long (hear for “hair”).  In the city there are very shops (very for “many”).  My bedroom is great (great for “huge” or “big”).  Thing  Make  Do  talking bird for “parrot”.

Foreignizing

Calque

A learner literally translates the word from the L1. This has to do with the transfer of semantic features from an L1 word to an L2 equivalent but with different contextual distribution (see e.g., Ringbom, 1987, 2001; Zimmermann, 1986a,b, 1987) A confusion of two L2 words which are formally similar either in spelling or in sound Mechanism that appears when two L2 words are confused because of semantic similarity, i.e., they have similar meanings, but they are functionally different learners lack knowledge of speci?c words and tend to use all purpose words (cf. Rababah, 2002). learners who do not know how to say a word in the foreign language opt for a description of that word (cf. Poulisse, 1993; Rababah, 2002; Salazar Campillo, 2006):

Formal approximation

Semantic approximation

All purpose words Circumlocution

Ma.P. Agust??n Llach / System 38 (2010) 529e538 Table 3 Descriptive statistics for learners’ essays in number of words. N Grade 4 Grade 8 203 203 Min. 13 21 Max. 277 419 Mean 93.15 165.63

533

S.D. 53,805 65,857

5. Results and discussion The Appendix offers by way of illustration examples of a grade 4 and a grade 8 essay. Results showed that learners’ essays increase in length as learners rise in grade. In particular, essays in 4th grade contain 93.15 words on average, while in grade 8, the mean composition length increases up to 165.63 words. Nevertheless, the ?gures of the standard deviations show that in grade 8 the sample is more heterogeneous with bigger differences among learners. In other words, the gap between the shortest and the longest composition in grade 4 amounts to 264 words, whereas in grade 8 the difference between the shortest and the longest essay rises to 398 words. There seems to be more instability among learners as far as their writing is concerned as they rise in grade and increase pro?ciency. Table 3 offers the ?gures of descriptive statistics for composition length. In order to ?nd out whether essay length increases signi?cantly, we performed a Wilcoxon signed rank test for two related samples, because the data did not meet the normality assumption. This test revealed that learners write signi?cantly longer compositions when they are in grade 8 than they had while in grade 4. Figures are the following Z ? 10.63, p < .001. Analysis of the compositions revealed from the ?rst moment that the categories of all purpose words and of circumlocution were very dif?cult to identify in our sample. Concerning the use of general words instead of speci?c terms, we realized that it was very dif?cult to discern whether the production of words such as “thing”, or “do” was the result of lack of knowledge or rather of preference for the general word to adapt to the informal letter learners were required to write. With all, we concentrated on the production of the general words when we considered they were used as a compensatory communication strategy: “thing”, “make”, and “do” and observed that the production of all purpose words was very much lower in 4th grade than four years later in grade 8. The word “thing” appeared 20 times in 8th grade, but not a single instance in 4th grade. In 4th grade the words “make” and “do” appear 1 and 5 times, respectively, whereas they are present 4 and 43 times four years later in 8th grade. Some examples of this strategy are: (11) (12) (13) (14) (15) (16) I do sausages. I do karaoke alone in the computer. I go with my friend Carla to buy some things. My friends are very special, we go to the cinema, to the park to many things. In my city usually make competitions of dance. I read, play football, play basketball, surf the Internet and make cook dinner.

From these observations, we can conclude that this L2-based lexical gap-?lling strategy is more frequent at higher levels of pro?ciency and at older ages. A possible explanation for this may be the higher linguistic awareness of learners at the second data collection moment. We can argue that 8th grade learners realize the high communicative potential of all purpose words: with little effort learners can bridge the lexical gap in an easy and ef?cient way. With regards to the production of circumlocution, we could ?nd very few instances of such strategy. For learners in 4th grade we found a total of 3 examples of circumlocution and 7 instances four years later in 8th grade. Table 4 presents all these instances. We believe a higher cognitive and lexical development as well as a higher linguistic awareness is necessary for learners to be able to resort to circumlocutions to cover their lexical gaps. This is in line with Poulisse’s (1993) argument that circumlocutions are the most demanding in cognitive terms of all compensatory strategies. If we consider this together with the young age of the participants it is no surprise that circumlocutions are sparse and that they get more frequent as learners get older and more pro?cient. It was in any case dif?cult to discern whether circumlocutions or paraphrases are the result of lack of lexical knowledge or a true desire to explain a fact.

534

Ma.P. Agust??n Llach / System 38 (2010) 529e538 Table 4 Examples of circumlocutions found in the data. 4th grade Dogs in the mountain for wolves Thirty days for month Run de bicicle for ride 8th grade Play phone for phoning Go to the streets for go out The volcano science for vulcanology The world of the photo for photography The most important city for the capital I play snow for I snowboard Double?oor for duplex apartment

Use of all purpose words and circumlocution, and especially the latter are conscious mechanisms of lexical gap?lling which are in themselves correct language, whereas the remaining four strategies result in a lexical error and may well be unconscious. Our ?rst research question asked about L1- and L2-based mechanisms to ?ll the lexical gaps of 4th graders when writing in the FL. Likewise, research question two also asked about prominence of L1 or L2-based mechanisms in 8th grade. Leaving out the mechanisms referred to, namely, use of all purpose words and circumlocution for the special reasons and features mentioned above, we can observe that L1-based mechanisms are more frequent in 4th grade than L2-based mechanisms. Similarly, four years later in grade 8, L1-based mechanisms also surpassed L2-based in the lexical gap-?lling behaviour of learners. However, in 8th grade the differences are much slighter between L1 and L2based strategies. Table 5 offers the descriptive results. In order to ascertain whether differences between L1- and L2-based lexical gap-?lling strategies were signi?cant and could be generalized, we decided to conduct several non-parametric tests for means comparisons, since normality assumptions could not be met. In particular, we performed Wilcoxon signed rank tests for related samples. The Wilcoxon signed rank tests performed show signi?cant differences in favour of L1-based mechanisms of lexical gap?lling for both grades analysed. Signi?cance values are higher for 4th grade, however (Z ? 7.14, p < .001, for 4th grade, for 8th grade Z ? 2.22, p < .05). From this result, we can conclude that L1-based mechanisms are always the most frequent ones. However, we can also observe a tendency for L1-based lexical gap-?lling strategies to decrease, whereas strategies derived from TL use tend to increase their presence as learners get older and more pro?cient (cf. Naves et al., 2005). Moreover, the total number of instances where learners resort to compensatory strategies to make do for a lexical gap reduces as they gain pro?ciency. We can argue that as learners become more experienced in the TL, they have a larger lexical repertoire and therefore, fewer lexical gaps to bridge. Similarly, they also dispose of a larger number of vocabulary resources to compensate for the lack of lexical knowledge and do not need to recur to ? ? the L1. Gonzalez Alvarez (2004: 104) made similar observations concerning this reduction and she interprets her results in the same key. Results on compensatory strategies can also be brought forward to account for our results. In this sense, Poulisse (1993: 164) observed that less pro?cient learners made more use of compensatory strategies and of L1-based strategies. The last research question sought to explore the differences across grades and pro?ciency levels concerning L1and L2-based mechanisms to ?ll lexical gaps when writing in the FL. As can be observed from Table 5, L1-based mechanisms are much more frequent in 4th grade than four years later. By contrast, only very slight differences can be appreciated in the production of L2-based mechanisms with a faint tendency to increase with grade. Wilcoxon signed rank tests of means comparison for two related samples (non-parametric) reveal that L1-based mechanisms decrease signi?cantly from 4th to 8th grade (Z ? 4.95, p < .001), whereas the increase in the use of L2-based mechanisms remains non-signi?cant (Z ? 0.47, p ? .63).

Table 5 Mean instances per composition of L1- and L2-based lexical gap-?lling mechanisms in learners’ essays. L1-based 4th grade 8th grade 2.9 1.4 L2-based 0.95 1

Ma.P. Agust??n Llach / System 38 (2010) 529e538 Table 6 Mean instances per composition of the different categories of lexical gap-?lling mechanisms in the learners’ essays. Borrowing 4th grade 8th grade 1.77 0.49 Foreignizing 0.58 0.64 Calque 0.55 0.28 Formal approximation 0.34 0.41

535

Semantic approximation 0.61 0.59

The signi?cant decrease of L1-based strategies is mainly the result of a dramatic decrease in the production of borrowings from 4th to 8th grade. The rest of the categories do not change notably and present very similar mean values for both grades examined.1 Table 6 offers the descriptive results of the different categories of lexical gap-?lling mechanisms. From the observation of the data, we noticed that learners in 4th grade tend to bridge their lexical gaps by using borrowings. They write very basic sentences of the type: Subject Verb Object (SVO) and they simply substitute the words they do not know by their Spanish L1 equivalents. Even four years later in grade 8, learners express very simple ideas and thoughts for which they know the words to use in the L2. Our results show that calque has a reduced frequency of application, and in any case lower than that of foreignizing. We believe that being both of limited communicative success, foreignizing is easier or less demanding to produce than calque, so FL learners opt for this strategy. Wilcoxon signed rank tests of means comparison for two related samples (non-parametric) showed that only for borrowings and calques is there a signi?cant change from 4th to 8th grade. The remaining gap-?lling mechanisms do not experience any signi?cant change. Our results con?rm the already claimed greater presence of L2 elements and mechanisms in the FL learners’ production as learning progresses. Table 7 presents these results. The results reveal a clear tendency towards the abandonment of the L1 when writing in the TL as learners’ pro?ciency increases. This is very interesting and can serve as support for the view of L1 use as a strategy to overcome communication problems. In addition, the gradual discard of the L1 evidences its initial role as a scaffolding to develop the TL system at the early stages. Moreover, this colligates with the idea purported by Zimmermann (1987) that lexical creation is a strategy for lexical search. FL learners’ search their minds for the targeted item even if it is not present in their mental lexicon or cannot be retrieved. In those cases, FL learners use synonyms in the L2 or L1 or paraphrases or other mechanisms. Another possible explanation of the results can be made in light of language processing theories. If the recourse to L1-based mechanisms of lexical gap-?lling re?ects the activation level of that language, then we can conclude that the L1 has a higher level of activation in less pro?cient learners and is the ? ? preferred source of information for lexical gap-?lling (Gonzalez Alvarez, 2004: 104; see also Dewaele, 1998). The communicative success of these mechanisms is un-even. Some lexical gap-?lling strategies are more successful than others in communicative terms. Thus, we agree with Djokic (1999: 126e129) in that the communicative success of borrowings and foreignizings depends on the reader’s knowledge of the learner’s L1. Calques transfer the message more or less correctly, although some instances of literal translation may cause momentary confusion or even misunderstanding. In this case, “the reader’s knowledge of the student’s L1 and the linguistic context play an important role in the transfer of the message” (Djokic, 1999: 128). L2 lexical gap-?lling mechanisms may cause some confusion, but most usually the linguistic context clari?es the message. The issue of communicative success is closely related to Poulisse’s (1993) linkage of communicative effectiveness and cognitive demands of lexical gap-?lling mechanisms exposed above. L2-based mechanisms are more demanding in cognitive terms, since they require better lexical knowledge and linguistic awareness on the part of the learners. In this sense, they are also more effective and successful in communicative terms, especially when learner and researcher (writer and reader) do not share the same native tongue. Our participants’ behaviour in creating new words to overcome lexical de?cits in communication may also re?ect the cognitive demands of the task. As we have commented above, learners in grade 4 recur more frequently to L1based strategies than four years later. This could be related to learners’ higher pro?ciency level in 8th grade and the fact that it was the ?fth time learners completed the writing task. Test effects may have played a role together with pro?ciency in determining the type of lexical gap-?lling mechanisms learners employ.

1

For examples of these mechanisms, please refer to the methodology, Section 4.

536

Ma.P. Agust??n Llach / System 38 (2010) 529e538

Table 7 Wilcoxon signed rank tests for means comparison for 4th and 8th grade lexical gap-?lling mechanisms. Borrowing Z P value *Signi?cant. 6.06 0.000* Foreignizing 1.02 0.305 Calque 3.55 0.000* Formal approximation 1 0.317 Semantic approximation 0.27 0.784

Our results resemble those of Liaw (1996) in which foreignizing and literal translation are the most frequent categories together with approximations. Circumlocutions and use of all purpose words are much less frequent. Contrary to our examples, no instances of codeswitching or borrowing were found in Liaw’s data (1996). The fact that Liaw’s readers are native English student peers and that learners were identi?ed as having a high level of pro?ciency in English may account for this fact. The strategy of codeswitching would make no sense in this case. In Djokic’s (1999) study L2-based strategies were most frequent. Literal translation, foreignizing, and borrowing had a much lower occurrence. The fact that Djokic’s (1999) learners are adult learners of high pro?ciency who know other FLs, and have a higher linguistic awareness may ? ? well explain the difference from our data. Gonzalez Alvarez (2004) analysed the production of lexical innovations of FL learners at three different pro?ciency levels and the results she obtained were that L1-based lexical innovations, especially borrowings or unmodi?ed L1 words, reduced as learners became more pro?cient, whereas L2-based lexical innovations increased with language level. Dewaele (1998) obtained results in the same key. 6. Conclusion The main concern of language learners is accuracy and ?uency, that is to get their meaning across in the most accurate way possible. In order to do this, learners resort to lexical gap-?lling mechanisms when linguistic resources are inadequate and insuf?cient for their communicative purposes. When low pro?cient learners have a lexical knowledge gap, then they most frequently resort to the L1 to compensate and solve that lack of knowledge. Basically borrowings, but also calques, are the most frequent mechanisms of lexical gap-?lling found in our data. Nevertheless, L1-based strategies, although most frequent at the two levels tested in the present study, tend to decrease. L2-based mechanisms by contrast tend to increase, though not signi?cantly. As a way of summary, it can be concluded from our data that learners at different grades follow different mechanisms to ?ll their lexical gaps in written production with a remarkable decline of general L1-based in?uence at the second measuring time and a slight increase of general L2-based resources. Secondly, L2 pro?ciency interrelates with L1 and L2 in?uence in different ways depending on the speci?c mechanism used, so that some mechanisms which are frequent at 4th grade decrease their presence, basically borrowing, and some other tend to increase, basically L2-based mechanisms. This result is in line with previous research pervasive all through the lexical ? ? acquisition process (Celaya, 1989; Celaya and Torras, 2001; Gonzalez Alvarez, 2004). Most FL learners beginning the SLA process start from the assumption that every word in the L1 has an equivalent in the L2. Perhaps thinking in the L1 or resorting to the L1 may be “the only way a leaner can ever begin to communicate in a second language” (Blum and Levenston, 1977: 16). Our results re?ect this argument and the idea that as learners progress in the L2 they gradually abandon L1 use. Recourse to the mother tongue may be the only communication strategy available to learners of low pro?ciency, and a good way to promote progress. As learners gain pro?ciency which leads to lexical knowledge development and get older which results in higher linguistic awareness, they start to incorporate more cognitively demanding lexical gap-?lling mechanisms, basically those based on TL use. Instructing students in the use of compensatory strategies can help learners cope with vocabulary problems. Indeed, we agree with Salazar Campillo (2006: 13) when she claims that the participants’ behaviour in choosing compensatory strategies can have important consequences in the educational setting, since we should instruct learners to manage problematic vocabulary or lexical gaps when confronted with them. Learners can become better strategy users through training (Salazar Campillo, 2006: 13). Current FL teaching programs advocate for instruction aimed at the development of communicative competence. Therefore, apart from helping FL learners acquire linguistic and sociocultural competences, teachers are also responsible for promoting the development of strategic competence, so that learners can react and overcome inhibitions and avoidance when facing unknown vocabulary items. Training could involve ? both strategy application activities and strategy awareness raising (cf. Mart?n Leralta, 2006; Salazar Campillo, 2006).

Ma.P. Agust??n Llach / System 38 (2010) 529e538

537

Teaching learners how to resort to voluntary L2-based lexical gap-?lling mechanisms can have positive effects on communication and on making learners aware of their behaviour when encountering a lexical gap. This may encourage learners to take risks when using the L2 and more importantly they may decide to remain in the communication process and get across their communicative goal (Liaw, 1996) instead of abandoning communication. Additionally and in relation to this, FL learners should explicitly be taught and learn some general, basic words such as “make”, “do”, “thing”, function words or the 1000 most frequent words because these are of general and extended use and can help learners overcome many vocabulary problems. Moreover, these can also be used for constructing paraphrases or circumlocutions (Salazar Campillo, 2006: 13). Future research should concentrate on examining lexical gap-?lling behaviour of older and more pro?cient learners to further observe the tendency of L2-based mechanisms to increase. Exploration of the relation between L2 pro?ciency and different types of L2-based mechanisms such as circumlocution, semantic approximation, use of all purpose words can also throw up intriguing results concerning learners’ preferences depending on linguistic mastery. Comparisons of written and oral performance would also be of great interest. We have provided some preliminary results for written compositions, but we believe that the more spontaneous oral speech would yield more instances of lexical gaps and in turn of lexical gap-?lling mechanisms. Still, an interesting design would be one that compared learners of different ages and the same pro?ciency level as well as learners of the same age but different pro?ciency level to see how lexical competence, linguistic awareness, and cognitive development relate and how relevant they are in determining lexical gap-?lling behaviour. Acknowledgements This study has been carried out under the auspices of a research project funded by the Spanish ‘Ministerio de ? Ciencia y Tecnolog?a’ and FEDER, Grant n HUM2006-09775-C02-02/FILO. I have to acknowledge my gratitude to the mathematician Montserrat San Martin for her help with the statistical analysis. I am also greatly indebted to Dr. ? ? Elsa Gonzalez Alvarez for her insightful comments on an earlier version of this paper. In any case, any remaining errors are my own. Appendix Example essay grade 4: Hello, Is XXX: in my street have got trees, shops, houses and schools. in my home have got two televisions and nine lamps. I am 9 years old. My Gcandmathes has got a dog. The my Sister’s bedroom is very small. The my bedroom is small. the my mather’s bedroom is very big. In my school is very big. my teacher in Inglish is very well studiant. I like very, very, very football. I like basketball. I like tennis. I like go to the cinema. I like go with my skates. I go to the cinema the Saturdays. my street is very contamination. the shops is very small. In the sportcenter, is very big and is orange. Example essay grade 8: Dear family Edwars, ? Hi! My name is XXX. I’m 13 years old and I live in Logrono, the capital city of La Rioja in Spain (La Rioja, I imagine you know is a place with a very good wine). I go to a middle school called XXX. I’m a very quietly girl and ? my hobbies are read, sing and skate. I love ice-skating and in Logrono there is an icekurt. Is there an icekurt in Oxford? I hope so, because is my favourite activity. Do your children like skating? I think we will be able to spend a very good time together. One question, in your house, will I share a room with Helen? It doesn’t matter if I have to do it, but I ? prefer to sleep alone in a room. In my house in Logrono I sleep in a very beautiful room with the wals painted orange. It’s big and from the window you can see St. Miguel’s parks and the mountains. My brother Hector sleeps in a room next to mine and he is very noisy Sometimes I can’t sleep well because of him! He is 12 and he is funny and kind, but sometimes I fall out with him. He enerves me! My mom is called XXX and she is very kind, but she gets angry easily. My dad is called XXX and he uis funny, but sometimes he is a little strict. In my home we all love reading and mi house is full of books, in the living room in the bedrooms, even in the kitchen! Hope to see you soon! XXX.

538

Ma.P. Agust??n Llach / System 38 (2010) 529e538

References
Blum, S., Levenston, E.A., 1977. Strategies of Communication Though Lexical Avoidance in the Speech and Writing of Second Language Teachers and Learners and in Translation. Available online. http://eric.ed.gov/ERICDocs/data/ericdocs2sql/content_storage_01/0000019b/80/ 37/9f/7b.pdf. Celaya, M.L., 1989. The role of the L1 in the acquisition and use of the L2. New Perspectives. BELLS: Barcelona English Language and Literature Studies 2, 73e80. Celaya, M.L., Torras, M.R., 2001. L1 in?uence and EFL vocabulary: do children rely more on L1 than adult learners? In: Proceedings of the 25th AEDEAN Meeting. December 13e15, University of Granada. pp. 1e14. Celaya, M.L., Y. Ruiz de Zarobe, 2008. CLIL, Age, and L1 in?uence. Paper presented at the XXXII AEDEAN Conference, Palma de Mallorca, 13the15th November. Clark, E., 1980. Lexical Innovations: How Children Learn to Create New Words. Available online. In: Papers and Reports on Child Language Development, 18. http://eric.ed.gov/ERICDocs/data/ericdocs2sql/content_storage_01/0000019b/80/34/58/44.pdf. Dewaele, J.M., 1998. Lexical inventions: French interlanguage as L2 versus L3. Applied Linguistics 19 (4), 471e490. Djokic, D., 1999. Lexical errors in l2 learning and communication. Rassegna italiana di linguistica applicata 31 (1), 123e135. ? ? Gonzalez Alvarez, E., 2004. Interlanguage Lexical Innovation. Lincom Europa, Munchen. ¨ Kocoglu, Z., 1997. The Role of Gender in Communication Strategy Use. Paper Presented at the 31st Annual Meeting of the Teachers of English to Speakers of Other Languages, Orlando, FL, March 11e15. Available online. http://www.eric.ed.gov/ERICDocs/data/ericdocs2sql/content_ storage_01/0000019b/80/16/c1/75.pdf. Liaw, M.-L., 1996. Communicative Devices Used by EFL Students in E-Mail Writing. Available online. http://eric.ed.gov/ERICDocs/data/ ericdocs2sql/content_storage_01/0000019b/80/14/cb/f3.pdf. ? ? ? ? Mart?n Leralta, S., 2006. La integracion de la competencia estrategica en el curr?culo de lengua extranjera. ELUA 20, 233e257. Naves, T., Miralpeix, I., Celaya, M.L., 2005. Who transfer more. And what? Cross-linguistic in?uence in relation to school grade and language dominance in EFL. International Journal of Multilingualism 2 (2), 113e134. Poulisse, N., 1993. A Theoretical account of lexical communication strategies. In: Schreuder, R., Weltens, B. (Eds.), The Bilingual Lexicon. John Benjamins, Amsterdam/ Philadelphia, pp. 157e189. Rababah, G., 2002. Second Language Communication Strategies: De?nitions, Taxonomies et al., Taxonomies, Data Elicitation Methodology and Teachability Issues. Available online. http://eric.ed.gov/ERICDocs/data/ericdocs2sql/content_storage_01/0000019b/80/1a/bc/3a.pdf. Ringbom, H., 1983. Borrowing and lexical transfer. Applied Linguistics 4, 207e212. Ringbom, H., 1986. Cross-linguistic in?uence and the foreign language learning process. In: Kellerman, E., Sharwood-Smith, M. (Eds.), Crosslinguistic In?uence in Second Language Acquisition. Pergamon Press, New York, pp. 150e162. Ringbom, H., 1987. The Role of the First Language in Foreign Language Learning. Multilingual Matters, Clevedon. Ringbom, H., 2001. Lexical transfer in L3 production. In: Cenoz, J., Hufeisen, B., Jessner, U. (Eds.), Cross-linguistic In?uence in Third Language Acquisition: Psycholinguistic Perspectives. Multilingual Matters, Clevedon, pp. 59e68. Salazar Campillo, P., 2006. The use of circumlocution in the foreign language Context. Porta Linguarum 5, 7e15. Singleton, D., 1987. Mother and other tongue in?uence on learner French. Studies in Second Language Acquisition 9, 327e346. Zimmermann, R., 1986a. Classi?cation and distribution of lexical errors in the written work of German learners of English. Papers and Studies in Contrastive Linguistics 21, 31e40. Zimmermann, R., 1986b. Semantics and lexical error analysis. Englisch- Amerikanische Studien 2, 294e305. Zimmermann, R., 1987. Form-oriented and content-oriented lexical errors in L2 learners. IRAL 25, 55e67.


相关文章:
gap filling
gap filling_英语学习_外语学习_教育专区。Summary of Gap Filling for CET 6 Words and Phrases and Expressions 1.固有的:Intrinsic For instance: CelSummary ...
Reading Gap Filling Dec 09
Spectral Sound Gap Filli... 4页 免费 2010-4 necklace gap fill... ... English Language Profici... 24页 免费如要投诉违规内容,请到百度文库投诉...
雅思听力解题指南:Gap Filling
智课网 IELTS 备考资料 雅思听力解题指南:Gap Filling 摘要: 为了帮助考生们更好地复习雅思考试,小马过河网雅思频道为各位考生整理了雅思听力解题指南: Gap Filling...
gap filling
Steven Pinker, an expert in children’s language acquisition, warns that children cannot develop language properly by watching television. They need to ...
simple-present-gap-fill_3296
simple-present-gap-fill_3296_少儿英语_幼儿教育_教育专区 暂无评价|0人阅读|0次下载simple-present-gap-fill_3296_少儿英语_幼儿教育_教育专区。SIMPLE PRESENT ...
雅思听力题型全解:填图题(GapFilling)
智课网 IELTS 备考资料 雅思听力题型全解:填图题(GapFilling) 题型介绍: You are asked to fill in the blank with limited words according to what you have...
5aUnit2 gapfill
2页 免费 5aUnit5 gapfill 2页 免费 5AUnit2测试 2页 2财富值如要投诉违规内容,请到百度文库投诉中心;如要提出功能问题或意见建议,请点击此处进行反馈。...
5aUnit5 gapfill
5aUnit4 gapfill 1页 免费 牛津小学英语5aunit5 30页 2财富值 Week 5 gapfill 暂无评价 2页 免费 5AUnit5 Review and check 暂无评价 24页 5财富值 5A...
雅思听力常见题型的解题指南:Gap Filling
智课网 IELTS 备考资料 雅思听力常见题型的解题指南:Gap Filling 摘要: 摘要: 小马过河雅思频道为各位考生整理了雅思听力常见题型的解题指南:Gap Filling,供考生们...
STAR Process Gap-Fill
STAR Process Gap-Fill_法律资料_人文社科_专业资料 暂无评价|0人阅读|0次下载|举报文档 STAR Process Gap-Fill_法律资料_人文社科_专业资料。Teacher’s Key ...
更多相关标签: