|
Karen's Linguistics Issues, September 2008 | Previous Months |
A recent class survey on language learning I conducted with two of my classes of adult English learners brought some very interesting points to light. One prominent point, articulated on several occasions, regarded the use of verbs and collocation, which were sources of considerable frustration and anxiety. As anxiety can result in reduced achievement (MacIntyre and Gardner 1991), I thought it useful to carry out some research into semantic errors caused by transfer (‘interference’) in regard to verbs and their collocations.
Errors can provide ‘important evidence for the strength and weakness of a particular native language influence’ (Odlin 1989:23). They contain valuable information on strategies that learners use to acquire language and can play an important role in the study of foreign languages (Richards 1974, Dulay and Burt 1974).
Written work is a particular weakness for Japanese learners (Educational Testing Service (ETS) 1996). The ETS report that the Japanese ranked 115 out of 117 countries on the Test of Written English (TWE), which is part of the Test Of English as a Foreign Language (TOEFL). One significant cause of this underachievement may possibly be the lack of composition writing in English at Japanese junior and senior high schools with the exception of some sentence level Japanese to English translation (Hirose 2003).
It is in light of this that I intend, in this article, to study transfer errors made in my students’ written work, investigating to what extent learner’s level of English has on the lexical misselection of verbs.
This study, it is hoped, will provide the reader with an insight which will not only aid in the identification and sourcing of errors during class time, but will also provide a basis which may aid in the preparation of lesson materials, and in post-lesson reflective analysis.
Many attempts to categorise deviances concentrate mainly on Chomsky’s (1965) differentiation between ‘competence’ and ‘performance’. ‘Mistakes’ are performance-related deviances which can occur as a result of factors external to the competence of the speaker, such as tiredness or lack of concentration (Corder 1967, 1971, Jain 1974). They are due to ‘a failure to utilize a known system correctly’ (Brown 1994:205), chance-circumstances (Hammerly 1982) and have varying degrees of seriousness (Bell, 1981). James (1998) regards such mistakes as ‘slips’ and adds that they are quickly and easily self-correctable. Edge (1989:9) adds that a ‘slip’ has occurred ‘if the teacher thinks that a student could self-correct’.
Errors, on the other hand, are deviances caused by a lack of competence, are systematic (Corder 1967), and are a ‘noticeable deviation from the adult grammar of a native speaker’ (Brown 1994:205). Errors are not self-correctable but the teacher may think that the learner is familiar with the form (Edge 1989). Further learning must take place before they are correctable (James 1998).
Ellis (1994:51) argues that there is not a simple binary distinction between ‘error’ and ‘mistake’ for deviances, but ‘it is possible that the learner’s knowledge of the target form is only partial; the learner may not have learnt all the contexts in which the form…can be used’. It could be also that a certain form could have been ‘known’ or partially known at one point in time, but not known at a later point due to memory limitations, for example.
Edge (1989:10) adds another category of deviance; ‘attempts’, which are those deviances made, when the teacher knows that ‘the students have not yet learned the language necessary to express what the student wants to say’.
James (1998: 83) provides the most practical classification and is based on the ability of students themselves to self-correct their work, thus reducing the researcher’s subjective opinion. I will however include Edge’s (1989) category ‘attempts’ to separate highly erroneous sentences, where the pinpointing of individual errors and meaning is particularly challenging. Deviances will be classified as follows:
1) Slips. Errors in performance which are easily and quickly self-corrected by the producer without having them pointed out.
2) Mistakes. These can be self-corrected if pointed out. There are two classifications of mistake:
i) First-order mistakes (M1), which can be self-corrected if an indication is given that there is a deviance, say for example, in a sentence.
ii) Second-order mistakes (M2), which can be corrected if the exact location (the word or the phrase) is made clear to the producer of the sentence.
3) Errors. These deviances are not self-correctable and further learning must take place before correction can take place.
4) Attempts. As described above.
Overt and Covert Errors
Consider the following utterances made recently by a student:
1) I was seen by Tom Cruise.
2) The dog hurted me.
At first sight, it can be argued that (1) is grammatical, as it is a passivisation of the verb ‘see’, followed by an agent ‘Tom Cruise’. In its active sense, ‘Tom Cruise saw me’ would also be acceptable, even more so, some may say, than the passive form used in (1). One may ascertain though that the learner has not meant what was intended, due to the slight ‘awkwardness’ of the verb form. When asked to explain in greater depth, it became apparent that this was not the intended meaning of the utterance. The learner had actually intended to say that he had seen Tom Cruise on a recent visit to Tokyo. While it may be possible that ‘he was seen by Tom Cruise’, had the learner caught Mr Cruise’s eye, this was not the intended meaning of the sentence. Corder (1971) describes these deviances as covert as they are grammatical, but not a representation of the context being expressed.
Utterance (2) on the other hand, is ungrammatical in comparison to Standard English as the learner possibly over-generalised a rule, having failed to realise that hurt is an irregular verb form having ‘hurt’ as its past tense form. Corder (1971) coined the term overtly idiosyncratic to describe these deviances.
The ‘errors’ learners make are significant in a number of respects. According to Corder (1967), errors (a) tell the teacher how the learner is progressing towards his or her language goal, (b) provide the learner with an opportunity to test their L2 hypotheses and (c) provide the researcher with evidence of strategies or procedures the learner employs in learning language. Corder argues that ‘we are able to reconstruct his knowledge of the language to date, i.e. his transitional competence.’ Brown (1994:205) agrees, saying that errors reflect ‘the interlanguage competence of the learner.’
Far from being unwanted forms, Edge (1989) points out that errors children make when learning their first language are signs of positive development, and could indicate acquisition of language rules. This can also be related to adults learning a second language. If a learner makes the a errors such as ‘I goed to the movies last night.’, it can be an indication of:
a) What and how much the learner has acquired; in this case the learner knows that to form a past tense, he/she must add ‘ed’ to the verb.
b) What and how much is still to learn; that some verbs have irregular past forms.
For the teacher, such evidence of a student’s progress is essential. During the execution of lessons and the subsequent post-lesson analysis, teachers can assess acquired and yet-to-be acquired areas of their students’ competence and thus tailor the lesson material and processes in planning subsequent lessons.
Overall, teachers can gain a valuable insight into their learners’ interlanguage, which we will discuss in more detail.
When investigating learner language, it is important and useful to recognise that the learner has what James (1998:2) describes as three ‘codes’ or languages. Firstly, there is the learner’s first language, native language or mother tongue (L1) which is described by Richards and Schmidt (2002:350) as ‘(usually) the language which a person acquires in early childhood because it is spoken in the family and/or it is the language of the country where he or she is living’. Secondly, there is the foreign or target language (TL), which is described as ‘any language learned after one has learnt one’s native language’ (Richards and Schmidt 2002:472).
The third of these is a learner’s interlanguage (IL). This is a ‘separate linguistic system’ which a learner possesses which is identical to neither to the L1 of the learner, nor the norms of the TL the learner is in the process of learning. (Selinker 1972:35).
Nesmer (1971:55) refers to IL as an ‘approximative system’ which is the ‘deviant linguistic system actually employed by the learner attempting to utilize the target language’. Corder (1971:161) emphasises that it should be regarded as a ‘idiosyncratic dialect’ similar to those of infants and poets and as such, is grammatical in it own right. This association with L1 learners is strengthened by Corder’s (1977) ‘initial hypothesis’; stating that an L1 learner and an L2 learner have the same starting point. However, is this strictly the case?
Many adult Japanese learners of English do actually bring knowledge of the TL (in this case English), whether it is actual TL forms or TL culture. This is usually manifest in loans words. An adult Japanese learner of English, for examples, would bring to his/ her first learning experience knowledge of lexical forms such as ‘herushi’ (the adjective ‘healthy’) and ‘poketto’ (the noun ‘pocket’) which are however, subject to the phonological patterns of Japanese
Rutherford (1987:7-8) argues that a language learner will also bring two kinds of knowledge: ‘knowledge how’; which will enable learners to adapt ‘the new language into forms that will…serve the initial desire for rudimentary communication’ and ‘knowledge that’; an ‘unconscious ‘foreknowledge’ or innate ‘inkling’ of what shapes the organisation of the target language’. Klein and Perdue (1992) argue that on embarkation of a foreign language a learner knows about language and communication from past language learning experiences.
One could assume therefore, that such knowledge of loan words and strategies for communication would, at least in the initial stages of learning, have some influence on the learner and how he/she proceeds to communicate. However, can one expect that influence to continue through the various stages of learning? For an insight, it is necessary to look at transfer in more detail.
Figure 1: James’ representation of interlanguage (James 1998:3)
One may assume that as the learner’s interlanguage develops their representation of the TL would include more features of the TL than of his/her L1 thus reducing transfer. .
Causes of Transfer Errors
We will look at some of the possible causes of transfer errors Lott (1983) suggests. These can be seen in the figure 2 below:
Figure 2. Categorisation of deviances (Adapted from Ellis 1994, Odlin 1989, Lott 1983, James 1998)
Lott (1983) suggests three causes of error which are particularly useful to this study as they are very much related to transfer errors of lexical items:
1) Overextension of analogy
This occurs when there is a similarity between a form in the learner’s L1 and the target language, however the two forms have different meanings. An example of this could be the use of ‘claim’ by Japanese speaker rather than ‘complain’, as ‘kureim’ is a Japanese term borrowed from English and has a meaning similar to that of ‘complain’ in English.
2) Transfer of Structure
This is the preference of the learner to transfer aspects of the L1 rather than use those of the TL. This can be seen in utterances such as ‘*Recently I didn’t see any movies’ where the past tense is used as a substitute for the present perfect as it is the past tense which is used in Japanese, as in:
saikin eiga o mi - na - katta
recently movie (object marker) see not (simple past tense)
(I haven’t seen any movies recently)
and utterances such as ‘*yesterday we did a party’
kino pa-ti o shi mashita
yesterday party (object marker) do (simple past tense form)
(We had a party yesterday)
3) Interlingual/ Intralingual errors
These occur when two or more forms in the TL correspond to only one form in the L1. English speaking learners of Japanese may make errors in the forms iru and aru which are similar to the form there is in English except they are use for living things and inanimate objects respectively.
The third cause of transfer above, relates closely to the theory of markedness, which categorises forms depending on how specialised a certain language form tends to be. Marked forms, for example, are ‘‘special’ in relation to unmarked forms, which are more basic’ (Ellis 1994:319). In the example given above, it can be ascertained from this that, in Japanese, ‘iru’ and ‘aru’ are marked compared to the more-general English equivalent ‘there is’, as the two Japanese forms are more specialised than their English counterpart in that they relate to different kinds of object; animate and inanimate. It is interesting to note that ‘aru’, on the other hand is less marked than ‘there is’ as it can be used to mean both ‘there is’ and ‘I have’.
Eckman’s (1977) ‘Markedness Differential Hypothesis’ predicts that a TL form will be more difficult to learn if the TL form is more marked while its L1 equivalent is less marked. One could see that in terms of inanimate and animate objects, the Japanese forms would be more difficult for English speakers, but in terms of expressing what one has and what there is, the Japanese form ‘aru’ would be easier for English speakers. James (1998:183) points out that by ‘difficult’, Eckman meant that ‘transfer will be unsuccessful and will result in an error.
Therefore, from our discussion we can conclude that:
a) ‘Errors’ are a result of a lack of competence.
b) Japanese learners will have some knowledge of English and language learning strategies when learning commences.
c) Interlanguage may be formed by both L1 and TL.
d) As learning continues, the reliance on L1 will reduce.
e) Transfer errors may be a result of three possible transfer processes.
f) Markedness of certain TL lexical forms may make transfer unsuccessful.
Is L1 transfer more prominent in lower-level learners’ written work than it is in that of higher-level learners in regard to verb misselection errors?
As the independent variable for the analysis (that which influences (Nunan, 1992)) is L1 and the dependent variable (that which is influenced (Nunan, 1992)) are the errors in a particular written assignment, it is important to ensure that other variables are kept as constant as possible.
1) have been taught by the same teachers for five years.
2) have studied English in class for one hour and twenty minutes a week on a Monday morning in the same classroom
3) comprise wholly of students who are females who are retired, home workers or both.
4) comprise of learners between the ages of thirty five and sixty five.
5) comprise of learners who have studied from between two and a half years and five years at the Culture Centre.
6) have used texts books in class for the duration of their study at the Culture Centre. Since March 2002, these textbooks have been from the same series; New Interchange 2 (Richards 1997a) and New Interchange 3 (Richards 1997b).
7) comprise of students who use their native language (Japanese) in their everyday lives, with the exception of two participants who use English in some situations outside the classroom.
In addition many of the subject’s primary reason for learning English is travelling abroad or ‘communication with foreigners’ (See Appendix 1)
Both groups were asked to complete a composition titled ‘Describe a good day or a bad day in your life’. The same concept questions were used to ensure understanding of the task for both groups and examples of topics for the composition were elicited from students. The students were then given sufficient time to complete the composition (Ellis 1997). The use of dictionaries was not allowed, however the use of spell-checking machines was.
It should be noted that all research for this study was carried out at a time outside the regular class hours of the learners.
After the allocated time of one hour, the compositions were collected. They were copied and the copy was given back to the students.
‘Errors are often analysed by researchers on the assumption that they know what their subjects really wanted to say or written but rarely do they ask them’ (Berkoff 1982:9). As a result, the students were asked to present their written work in a ten-minute oral presentation, with a question and answer session one day after completing the written work. By employing this method, the exact contexts the students were writing were ascertained and hence the overt deviances were identified.
Upon identification of the overt deviances, the compositions were entered into a word processor and three separate copies, detailed below, of the essays were produced:
1) The ‘Slip’ Identification Task was a typed copy of the original composition.
2) The ‘First-Order Mistake’ Identification Task was a typed copy of the original composition with an indication to the learner of which sentences contained deviances. In cases where the sentence structure made it impossible to locate individual errors (‘attempts’) the entire sentence was underlined.
3) The ‘Second-Order Mistake’ Identification Task was a typed copy of the original composition with the phrases containing deviances underlined.
A sample of the tasks can be found in Appendix 2.
One week after the composition was written, the students were asked to check their own work. The Slip Identification Task was distributed, collected and slips were identified. A similar procedure was adopted with The First-Order Mistake Identification Task and The Second-Order Mistake Identification Task to identify ‘first-order mistakes’ and ‘second-order mistakes’ respectively. From this ‘errors’ in the students’ work were therefore, identified. The ‘errors’ were categorised further into the focus of this research; Synform verb errors.
He wanted to cancel [conceal] his guilt.
Here we can see that although the verb ‘cancel’ is correct in form and is of the required part of speech (in this case, a verb), it is not correct in meaning. The corrected form (in square parenthesis) was not selected by the learner and was replaced by an alternative taken from the learner’s interlanguage. I will use the term ‘synform verbs’ to describe these forms.
Therefore, we can ascertain that:
will be included in the sample as during the interview I ascertained that the word ‘start’ is incorrect as she had meant to write:
We first met each other at the airport.
On the other hand, a construction such as:
It has been taken for eleven hours.
will not be included as the learner was referring to duration of a train trip, but had erred in her choice of tenses.
There are some borderline cases such as:
About five years ago, I was change my job
In this case it was assumed, after further clarification, that the learner had made a grammatical error by failing to establish that ‘change’ in this context should be used as a verb.
A panel of five judges (three Japanese intermediate-English speakers and two British intermediate-Japanese speakers) was employed to determine to what extent they believed the ‘errors’ to be errors related to the transfer of the learners’ native language. They were given the ‘error questionnaires’ (see appendix 4. The compositions were available to the judges for reference) and asked to grade the errors on a scale of one to five. A ‘one’ would indicate that in his or her opinion no transfer had taken place. A ‘five’ would indicate that transfer had taken place in regard to the semantic errors of verbs in the sentences. A short meeting was then held so the judges could discuss the ‘errors’ and explain the provenance in regard to the Japanese language.
In the analysis of the ‘error questionnaires’, the extent of transfer on the verbs was calculated by totalling the results as follows:
If a verb error was graded by the five judges as (1 for low, 5 for high) 4,3,3,5,4 then the total for that would be 4+3+3+5+4= 19.
If a verb error was graded 1,2,2,1,1 then the total for that verb would be 7.
This would tend to suggest that transfer on the first example was stronger than on the second example as the total is higher.
The accumulated grade totals are on a scale of 5 (no transfer) to 25 (full transfer).
The rational behind this technique for measuring transfer is two-fold. Firstly, it allows a more objective view of errors than a single person apportioning the extent of transfer does, and secondly, it prevents a purely binary differentiation between transfer and non-transfer and adopts a representation of transfer on a scale.
|
|
Pre-Intermediate |
|||||||
|
Composition |
1 |
2 |
3 |
4 |
5 |
Total |
% |
Mean |
|
Slips |
3 |
0 |
0 |
0 |
0 |
3 |
3.70 |
0.6 |
|
Mistakes (First-order) |
0 |
1 |
4 |
4 |
1 |
10 |
12.35 |
2 |
|
Mistakes (Second-order) |
5 |
5 |
1 |
0 |
0 |
11 |
13.58 |
2.2 |
|
Errors |
15 |
9 |
13 |
12 |
6 |
55 |
67.90 |
11 |
|
Attempts |
0 |
0 |
2 |
0 |
0 |
2 |
2.47 |
0.4 |
|
Total Deviances |
24 |
17 |
23 |
20 |
12 |
81 |
100.00 |
|
|
Words written |
180 |
130 |
113 |
135 |
87 |
645 |
|
129 |
Figure 3: The type and amount of deviances for pre-intermediate group.
|
|
Intermediate |
|||||||
|
Composition |
6 |
7 |
8 |
9 |
10 |
Total |
% |
Mean |
|
Slips |
2 |
1 |
3 |
2 |
2 |
10 |
17.54 |
2 |
|
Mistakes (First-order) |
1 |
2 |
3 |
2 |
2 |
10 |
17.54 |
2 |
|
Mistakes(Second-order) |
0 |
0 |
2 |
2 |
0 |
4 |
7.02 |
0.8 |
|
Errors |
1 |
9 |
12 |
2 |
8 |
32 |
56.14 |
6.4 |
|
Attempts |
0 |
0 |
0 |
0 |
1 |
1 |
1.75 |
0.2 |
|
Total Deviances |
10 |
19 |
28 |
17 |
23 |
57 |
100.00 |
|
|
Words written |
125 |
183 |
173 |
137 |
127 |
745 |
|
149 |
Figure 4: The type and amount of deviances for intermediate group.
|
|
Grand Total |
Pre-intermediate group’s % of whole |
Intermediate group’s % of whole |
|
Slips |
13 |
23.08 |
76.92 |
|
Mistakes (First-order) |
20 |
50.00 |
50.00 |
|
Mistakes (Second-order) |
15 |
73.33 |
26.67 |
|
Errors |
87 |
63.22 |
36.78 |
|
Attempts |
3 |
66.67 |
33.33 |
|
Total Deviances |
138 |
58.70 |
41.30 |
|
Words written |
1390 |
46.40 |
53.60 |
Figure 5: A comparison of type and amounts of deviance produced by both groups.
As can be seen in figure 3 and 4 above, ‘errors’ were the most common form of deviance for both groups, comprising of 67.9% of deviances for the pre-intermediate group and 56.14% for that of the intermediate group. The mean average deviance rate for each composition in the PG was 11 while for the IG was 6.4. One could expect a higher number of competence ‘errors’ in the group of lower students. In general, the higher group were able to correct their deviances more readily than the lower group.
Analysis of Synform Verb Errors
Out of the 87 errors produced from the sample, 17 of these were, as has been discussed, synform verbs misselections. The PG produced 12 of these errors and the IG produced 5. The sentences containing these verb errors and the compilation of rating given by the five judges can be seen below in figure 6.
Pre-Intermediate Synform Verb Errors |
1 |
2 |
3 |
4 |
5 |
Total |
|
0 |
0 |
0 |
1 |
4 |
24 |
|
0 |
0 |
0 |
2 |
3 |
23 |
|
0 |
0 |
2 |
2 |
1 |
19 |
|
0 |
1 |
1 |
3 |
0 |
17 |
|
1 |
0 |
2 |
1 |
1 |
16 |
|
0 |
1 |
2 |
2 |
0 |
16 |
|
1 |
0 |
2 |
2 |
0 |
15 |
|
2 |
3 |
0 |
0 |
0 |
8 |
|
3 |
2 |
0 |
0 |
0 |
7 |
|
4 |
1 |
0 |
0 |
0 |
6 |
|
4 |
1 |
0 |
0 |
0 |
6 |
|
5 |
0 |
0 |
0 |
0 |
5 |
|
Mean |
12.1 |
|||||
|
Intermediate Synform Verb Errors |
1 |
2 |
3 |
4 |
5 |
Total |
|
0 |
0 |
1 |
2 |
2 |
21 |
|
0 |
1 |
0 |
3 |
1 |
19 |
|
0 |
1 |
1 |
1 |
2 |
19 |
|
1 |
1 |
2 |
1 |
0 |
13 |
|
3 |
1 |
1 |
0 |
0 |
8 |
|
Mean |
16 |
|||||
Figure 6: Sentences containing ‘Synform Verb’ errors and the ranking given by judges.
Whilst a larger corpus of errors would have enabled more accurate analysis, it was judged from these errors that although the lower group had a greater proportion of verb misselections, the extent to which transfer influenced selection was greater in the IG that in the PG.
Among the sentences that were judged to have been caused by transfer errors were the verbs used in sentences (1), (2), (13) and (14). These were judged to have been caused by a ‘transfer of structure’.
The error in sentence (1), as we have previously seen, was the misselection of the verb in the collocation ‘have a party’. The author had clearly not been aware of this collocation after having it brought to her attention, although the lexical item ‘have’ is known to be contained in her interlanguage. The judges agreed that in all likelihood, the error in ‘do a party’ was caused by a similar collocation in Japanese ‘pa-ti o suru’ (do a party):
Sentence (2) showed a similar possible transfer error. The score of 23 was reflected by comments that the verb ‘move’ in English has a similar meaning to ‘ido suru’ (literally ‘a move do) in Japanese. An interesting point to note is that although the lexical item was subject to possible transfer the grammatical aspect was not:
*I moved by bus
(Bus by move do (past marker))
I travelled by bus
Sentence (13), with a score of 21, was also considered the product of transfer:
*I looked for flowers that our garden didn’t have.
(our (possessive) garden in no flower (object marker) look for (past))
I looked for flowers that weren’t in our garden
It was argued however, that ‘nai’ could be used to negate most verbs, but in the context in which it is used in the Japanese above indicated that ‘nai’ represented a negation of the verb ‘have’, thus suggesting that transfer has occurred.
The intermediate learner’s error (in sentence (14)) was also thought to show signs of ‘transfer of structure’ (Lott 1983):
*It (that day) was very precious memories for me.
sonohi wa watashi ni tote taisetsu omoide de shita
(that day (subject marker) I to very important memory was (past))
That day holds very precious memories for me.
In this case, it is possible that the correct lexical item had not been known and she had relied on Japanese verb ‘desu’ (the base form of ‘de shita’) to compensate for a lack of competence.
Lott’s ‘overextension of analogy’ was thought to influence verb use in (15):
*He had troubled about his future
kare wa shourai o shimpai shite imashita
(he (subject marker) future (object marker) worry do (past))
He had been worried about his future.
Although ‘worry’ has a Japanese equivalent (‘shimpai suru’), Japanese also makes use of the loan word ‘toraburu’ (from the English ‘trouble’) to express a ‘problem’ or a ‘worry’. One possibility is that the author could have transferred the loan word from Japanese, wrongly equating that ‘toraburu’ and ‘trouble’ are semantically the same.
A number of examples may indicate that markedness causes difficulty in verb selection. One example is in sentence (15):
I have been rearing genpeiboku for seven years
(
I (subject seven year for ‘a kind (object grow
(tense
marker) of plant’ marker) rear marker))
raise
bring up
I have been growing ‘genpeiboku’ for seven years
As ‘genpeiboku’ is a kind of plant, from the four (of more) possibilities in English, which have similar meanings to ‘sodateru’, only one is correct; ‘grow’.
Another example is seen in
*We went to look at the sunrise
watashitachi wa hinode o mi ni iki mashita
we (subject sunrise (subject look to go tense marker (past))
marker) marker) watch
see
We went to watch the sunrise.
Again a number of English equivalents arise from the Japanese ‘miru’ (the base verb of ‘miai’ which corresponds to several English verbs; most commonly ‘see’, ‘watch’, and ‘look (at)’.
English verbs such as ‘grow’ or ‘watch’ have a high degree of markedness compared to the Japanese counterparts ‘sodateru’ and ‘miru’ as they have a more specialised or specific use. While this may cause more difficulty for the learner to acquire these verbs in English, it is not clear that transfer has occurred. In the case of the pre-intermediate learner who produced sentence (5), I know, as her teacher, that her interlanguage contains lexical items such as ‘watch’, ‘see’ and ‘look at’. This could indicate that the misselection was caused merely by chance, or by a lack of competence in differentiating between the verbs, rather than transfer, making what Lott (1983) describes as an interlingual/ intralingual error.
Conclusions and Teaching Implications
From the research, the case that L1 transfer is the source of certain errors seems difficult to dispute. One could argue however, that the corpus in the research was not ample enough to provide conclusive evidence whether errors in higher levels are influenced more by transfer than those in lower levels. A wider ranging corpus in terms of instances and differing levels was thought to be out of the range of a small-sized research report such as this.
There may be several points to note from this research. Firstly, if a teacher is to ascertain the nature of at least some his/her students’ errors, it is imperative to learn, or at least be aware of, the students’ native language(s). This may prove difficult in multilingual classes, however. Secondly, as the differing level of markedness between L1 and TL lexes can result in difficulty and possibly error, an investigation into the relationship between the two should be carried out in the planning stages of a lesson. In doing this, the teacher may find it beneficial to consider all the lexical possibilities between relatively unmarked L1 forms and TL forms that are more marked. As a result the teacher could then present the various TL forms not only what they collocate with, but also with what they do not collocate, therefore giving the learner a more complete understanding of the lexical forms. Thirdly, during the lesson and subsequent post-lesson analysis, teachers should be monitoring and assessing all errors to predict as far as possible, what aspects of the students’ interlanguages may need work.
As one may argue that the analysis of errors focuses only on the negative aspects of a learner language, there are certain patterns or tendencies that are evident. With research, and/or observation, the teacher can exploit these negative aspects to the advantage of their learners.
Bell, Roger T. (1981) An Introduction to Applied Linguistics. London: B.T. Batsford Ltd.
Berkoff, N.A. (1981) Error analysis Revisited. In Nickel, G. & Nehls, D. (1982) 7-17.
Brown, D.B. (1994) Principles of Language Learning and Teaching. 3rd Ed., New Jersey: Prentice Hall Regents.
Chomsky, N. (1965) Asoects of the Theory of Syntax. Cambridge, MA.:MIT Press.
Corder, S.P. (1967) ‘The Significance of Learners’ Errors’. In Richards, J.C. (1984), 19-27.
Corder, S.P. (1971) ‘Idiosyncratic Dialects and Error Analysis’ in Richards, J.C. (1984), 158-171.
Corder, S.P. (1977) ‘“Simple Codes” and the source of the learner’s heuristic hypothesis’. Studies in Second Language Acquisition 1, 1-10.
Dulay, H. & M. Burt, (1974). You can't learn without goofing. In Richards, J. C. (1984) 93-123.
Edge, J. (1989) Mistakes and Correction. London: Longman.
Educational Testing Service (1996). TOEFL: Test of Written English Guide (4th Ed.). Princeton, NJ: Educational Testing Services (ETS).
Eckman, F. (1977) ‘Markedness and the Contrastive Analysis Hypothesis’. Language Learning 27, 315-330.
Ellis, R. (1994) The Study of Second Language Acquisition. Oxford: Oxford University Press.
Ellis, R. (1997) Second Language Acquisition. Oxford: Oxford University Press.
Gass, S. (1996) ‘Second Language Acquisition and Linguistic Theory: The Role of Language Transfer’ in Richie and Bhatia (1996), 317-345.
Hammerly, H. (1982) Synthesis in Second Language Teaching: An Introduction to Languistics. Blain, Wash.: Second Language Publications.
Hirose, K. (2003) ‘Compairng L1 and L2 Organizational Patterns in the Argumentative Writing of Japanese EFL Students’ Journal of Second Language Writing 12, 181-209.
Jain, M.P. (1974) ‘Error Analysis: Source, Cause and Significance. In Richards, J.C. (1984) 189-215.
James, C. (1998) Errors in Language Learning and Use: Exploring Error Analysis. London: Longaman
Laufer, B. (1992) ‘Native Language effect on confusion of similar lexical forms’, in Marcus, M. and C. Mair, eds. New Direction in Contrastive Linguistics. Innsbruck: Innsbrucker Beitraege zur Kulturwissenschaft. vol.2 pp.199-209.
Klein, W. and C. Perdue (1992) Utterance Structure: Developing grammars again. Amsterdam: Benjamins.
Lott, D. (1983) ‘Analysing and Counteracting Interference Errors’. English Language Teaching Journal 37, 256-61.
MacIntyre, P. and R. Gardner (1991) ‘Methods and results in the study of foreign language anxiety: a review of the literature’. Language learning 41, 25-57.
Nesmer, W. (1971) ‘Approximative systems of Foreign Language Learners’ in Richards, J.C. (1984), 55-63.
Nickel, G. & D. Nehls, eds (1982 ) Error analysis, contrastive linguistics and second language learning. Papers from the 6th Interntional Congress of Applied Linguistics, Lund 1981. Heidelberg : Groos.
Nunan, D. (1992) Research Methods in Language Learning, Cambridge: Cambridge University Press.
Odlin, T. (1989) Language Transfer. Cambridge: Cambridge University Press.
Richards, J. (1974). A Non-Contrastive Approach to Error Analysis. In Richards, J.C. (1984) 172-188. Essex: Longman.
Richards, J.C. ed. (1984) Error Analysis: Perspectives on Second Language Acquisition. London: Longman.
Richards, J.C. (1997a) New Interchange 2. Cambridge: Cambridge University Press.
Richards, J.C. (1997b) New Interchange 3. Cambridge: Cambridge University Press.
Richards, J.C. & R. Schmidt (2002) Dictionary of Language Teaching & Applied Linguistics. 3rd Ed., London: Longman..
Ritchie, W.C. and T.K. Bhatia, eds. (1996) Handbook of second language acquisition. San Diego: Academic Press.
Rutherford, W. (1987) Second language grammar: Learning and Teaching. New York: Longman.
Schwartz, B. & R. Sprouse. (1996) ‘L2 cognitive states and the full transfer/full access model’. Second Language Research 12, 40-77.
Selinker, L. (1972) ‘Interlanguage’, in Richards, J.C. (1984), 31-54.
Appendix 1: Results of the Learning Questionnaire
1) How long have you been studying English this Culture Centre?
|
Student 1 |
Since 1999 |
|
Student 2 |
For 2 1/2 years |
|
Student 3 |
For 3 years |
|
Student 4 |
Since March 2001 |
|
Student 5 |
Five years |
Intermediate Group
|
Student 6 |
3 Years 3 months |
|
Student 7 |
For 3 Years |
|
Student 8 |
For 3 Years |
|
Student 9 |
Since 2000 |
|
Student 10 |
For 5 years |
2) What is your reason/ motivation for studying English?
|
Student 1 |
Travelling |
|
Student 2 |
I want to learn about foreign culture |
|
Student 3 |
Travelling |
|
Student 4 |
To speak when I travel abroad |
|
Student 5 |
To keep my mind active |
|
Student 6 |
Travelling |
|
Student 7 |
For a hobby |
|
Student 8 |
I want to communicate with foreigers |
|
Student 9 |
Travel |
|
Student 10 |
Hobby |
3) How often do you use English outside class?
|
Student 1 |
Never except homework |
|
Student 2 |
Never. Sometimes using internet. |
|
Student 3 |
Every week. I teach the Japanese class to foreigners |
|
Student 4 |
Only abroad |
|
Student 5 |
On trips, homework |
|
Student 6 |
When travelling |
|
Student 7 |
Homework |
|
Student 8 |
Travelling |
|
Student 9 |
Never. Sometimes when abroad. |
|
Student 10 |
Sometimes to my wife. |
Appendix 2: Identification of Error Tasks
Stage 1- Identification of Slips
Students were handed back the printout of their composition and asked to look over it for 5 minutes and correct any errors they saw.
I went to Akui hot springs near Sendai city and stayed there from 29th last month to New Year’s eve.
At 29th and New Year’s eve I moved by bus. At 30th I went to Kaleidoscopes Art museum by taxi for thirty minutes.
This museum is only one in the world.
At the ground floor, there were many kinds of kaleidoscopes were displayed and film of their history was shown.
At first floor, there were many kinds of pictures of kaleidoscopes
I felt very beautiful and mysterious.
By the way it didn’t snow at Sendai
But the top of the mountains near Sendai was covered with snow.
I had very happy time.
I hope I would like to go to Akui next summer and will swim in a spa pool.
Stage 2. Identification of First-Order Mistakes.
A second copy was given to the student with an ‘X’ next to the sentences which contain a deviance. The number indicates the number of deviances. An ‘O’ signifies a correct construction.
X I went to Akui hot springs near Sendai city and stayed there from 29th last month to New Year’s eve.
5X At 29th and New Year’s eve I moved by bus. At 30th I went to Kaleidoscopes Art museum by taxi for thirty minutes.
1X This museum is only one in the world.
3X At the ground floor, there were many kinds of kaleidoscopes were displayed and film of their history was shown.
1X At first floor, there were many kinds of pictures of kaleidoscopes
1X I felt very beautiful and mysterious.
1X By the way it didn’t snow at Sendai
1X But the top of the mountains near Sendai was covered with snow.
1X I had very happy time.
1X I hope I would like to go to Akui next summer and will swim in a spa pool.
Stage 3- Identification of Second-Order Mistakes
A third copy was give with the phrases, which contained the deviances underlined.
I went to Akui hot spring near Sendai city and stayed there from 29th last month to New Year’s eve.
At 29th and New year’s eve I moved by bus. At 30th I went to Kaleidoscopes Art museum by taxi for thirty minutes.
This museum is only one in the world.
At the ground floor, there were many kinds of kaleidoscopes were displayed and film of their history was shown.
At first floor, there were many kinds of pictures of kaleidoscopes
I felt very beautiful and mysterious.
By the way it didn’t snow at Sendai
But the top of the mountains near Sendai was covered with snow.
I had very happy time.
I hope I would like to go to Akui next summer and will swim in a spa pool.
Bio-data
Nicholas Delleman has been teaching English since 1994 years in England, Slovakia, and Turkey, and is currently working at various universities and companies in the Tokyo area, Japan.
© Nick Dellman 2008. All rights reserved.