安 徽 大 学
硕 士 学 位 论 文
题目 Error Analysis and Mother Tongue Transfer
专 业 英 语 语 言 学
研究方向 语 言 学
申请人 汪 春 娣
届 别 二 零 零 三
导 师 朱 跃 教 授
年 月 日
Acknowledgement
For the two years of my study in The School of Foreign Studies, I would like to express my appreciative thanks to Professor Hua Quankun, Professor Hong Zengliu, Professor Zhou Fangzhu, Professor Zhang Ming, Professor He Gongjie, Professor Xiao Shuhui, Professor Zhang Zuwu, Professor Li Yongfang and many others for their delicate instructions. I am especially grateful to Professor Zhu Yue whose critical comments and sound advice in the writing of this paper have been invaluable and much appreciated.
Finally, I would like to acknowledge that this paper would not have been possible without the help of many of my colleagues.
1.Introduction--------------------------------------------------------------------1
2.The literature review----------------------------------------------------------2
2.1 Error analysis and its psychological basis 2
2.2 The procedures of EA 3
2.3 Related research of EA from abroad and the limitations of EA 5
2.4 Related research in China 9
3. Research design--------------------------------------------------------------10
3.1 Quantitative design 11
3.1.1 Subject 11
3.1.2 Instrument 11
3.1.2.1 Error sample 11
3.1.2.2 Questionnaire 11
3.2 Qualitative design 12
3.2.1 Descriptive analysis 12
3.2.2 Interview 12
4. Data collection and analysis-----------------------------------------------12
4.1 Collection of errors 12
4.2 Descriptive statistics of error analysis 14
4.3 Descriptive statistics of the sources of errors 15
5. Discussion---------------------------------------------------------------------18
5.1 Interlingual errors or interferences 18
5.2 Intralingual errors or developmental errors 27
5.3 The other errors 31
5.4 The result of 21 percent interference 34
5.5 Insight into English teaching 36
6. Conclusion--------------------------------------------------------------------39
Reference 41
Mother tongue transfer has been a controversial issue in foreign language learning. Much empirical work on EA has been focused on determining what proportions of errors in a corpus are due to inter-lingual transfer as opposed to intra-lingual transfer since 1970s. One of the studies of this kind worth mentioning is carried out by Dulay and Burt (1974). They have selected 179 Spanish children and studied the errors made in their English learning. After the analysis of the errors, they claim that the proportion of mother tongue interference is only 3 per cent. Other researches show that the mean percentage is about 30 percent (Ellis, 2001). In China, however, the researches in this aspect are not often seen.
Based on the collection of the errors from 355 compositions of the second year college students, the paper identifies 15 types of common errors and their frequency distributions. The quantitative analysis of the errors and the study of their frequency distributions show that the error frequency distributions are not distinctive. Through the quantitative analysis of the data from the questionnaires, the paper finds that, in China, the errors due to mother tongue interference amount to 21 per cent. Through the qualitative error analysis, the paper discovers that the students’ uses of verb tenses, plural forms of nouns, sentence structures, articles, prepositions and cognate words are liable to be affected by their mother tongue. After a careful analysis of the students’ errors, the paper puts forward suggestions that feasible measures be taken to deal with the problems in English teaching so as to improve students’ proficiency in English.
Key words：error analysis mother tongue transfer
inter-lingual error intra-lingual error
Chinese interference developmental error
论文摘要
母语迁移一直是二语习得领域引起争议的问题。自二十世纪七十年代初以来，国外错误分析的焦点一直是母语迁移在错误语料库中所占的比例；其中最引人瞩目的人是Dulay and Burt (1974)，他们通过对179名母语为西班牙语的一组儿童学习英语所犯错误分析发现只有3%的错误可归为母语干扰；另一些错误分析统计也发现由母语干扰导致的错误一般只占30%左右（Ellis, 2001）；而国内错误分析的相关研究少有涉及母语干扰的具体比例。
本文通过对355名大学二年级学生书面作文错误的收集，归纳出大学生写作中常见的15种错误及其分布比例。通过对错误的定量分析，得出各种错误在各变量之间的比例没有显著差异；通过对问卷的定量分析，得出在汉语环境下有21%的错误来自母语干扰；对具体的错误定性分析，还显示动词时态、名词复数、句子结构、冠词和介词等错误易受母语干扰。最后，本文对上述调查结果进行分析，指出在教学中应注意来自母语干扰的错误；提出应加强语法、词汇、阅读和写作等方面的教学，并强调教学应以读写为本。
关键词：错误分析 母语迁移 汉语干扰
语际错误 语内错误 学习过程中错误
1.Introduction
The study of second language acquisition (SLA) as a new discipline has developed quickly at the turning of the new century. There have been extensive researches done in this field. Among them, error analysis (EA) and language transfer are common and controversial issues which many linguists and foreign teachers have concerned themselves about for years. The study of error analysis rose at the end of 1960s and developed in 1970s, providing us an important method for SLA research. Together with the study of language transfer, the focus, which previously was on the learners’ external environment, now has been switched to the internal one—the learners themselves.
The underlying assumption of contrastive analysis (CA) was that error occurred primarily as a result of interference when the learners transferred native language habits to the use of second language (L2). Interference was believed to take place whenever the habits of the native language differed from those of the target language. CA gave way to EA as this assumption came to be challenged. Dulay and Burt (1974) claimed that the proportion of interference was only 3 percent. Wilkins (1972) even suggested abandoning the idea of “transfer”, since it could not account for all the difficulties and errors that learners had in their learning process. However, some recent findings show that the greater the difference between the two language systems, the larger the transfer will be (Koda, 1999).
Chinese interference in English learning has been recognized (Tran-Chi-Chau, 1974, Wang Zongyan, 1980, Sun Mianzhi, 2001). James (2001) points out that no responsible Contrastive Analyst ever claims that all foreign language errors are due to first language (L1) interference, or go to the opposite extreme of claiming that virtually no errors are caused by L1 transfer.
Under the circumstances of the above contradictory and confusing findings, the research attempts to explore the possible proportion of Chinese interference in Chinese context. The research questions to be addressed in this study are as follows:
(1) What are the most common errors that students commit in their writing?
(2) What proportion of the total errors in a corpus is due to transfer as apposed to intra-lingual errors in Chinese context?
1. The literature review
2.1 Error Analysis and its psychological basis
“Error” refers generally to the learner’s misuse or misunderstanding of the target language, may it be grammatical or pragmatic. An error is a noticeable deviation from the adult grammar of a native speaker, reflecting the limit of the learners’ competence in using the target language. It reveals the learners’ knowledge of the target language. Error Analysis is a type of bilingual comparison, a comparison between learners’ interlanguage and the target language. It is a methodology of describing second language learners’ language system. There are two justifications for the study of learners’ errors: The first is pedagogical, which claims that a good understanding of the nature of errors is necessary before a systematic means of eradicating them can be found. The second is theoretical, which claims that a study of learners’ errors is part of the systematic study of the learners’ language. By studying error analysis, it will be possible to identify the learning process that the learners are going through, thus finding out how learners’ second language proficiency develops (Hu Zhuangling, 2001).
With the development of the applied linguistics, two kinds of attitudes towards learners’ errors have turned up. According to the behaviourists accounts, errors are viewed as the result of the negative transfer of L1 habits. In other words, the learners fail to change their old habits so as to acquire new habits of the second language. According to the mentalists accounts, errors are predicted to be similar to those found in L1 acquisition because learners actively construct the grammar of a L2 as they progress. The errors they make are actually their incorrect hypotheses about the new language.
Another linguistic system, interlanguage (IL), is based on the observable output which results from a learner’s attempted production of a target language form. Interlanguage (Selinker, 1972) is formed when the learners attempt to learn a new language, and it has features of both the first language and the second language. Other terms that refer to the same basic ideas are ‘approximative system’ (Nemser, 1971) and ‘transitional competence’ (Corder, 1967).
According to Selinker, a theory of second language learning should be concerned with the surface of interlanguage so as to predict the learners’ behavioural events and make clear the psychological structure of adult language learners. They regard errors as evidence of learning process. By making hypotheses about the target language, the learners arrive at a particular interlanguage. Then they modify their hypotheses and go toward the target language. Selinker (1972) also notes that most L2 learners fail to reach target language competence, that is, they stop learning while the contained rules in their internalized rule system are different from those of the target system. This is referred to as ‘fossilization’, a special phenomenon in interlanguage.
Corder (1967) points out that errors are evidence of the learners’ in-built syllabus. Error analysis aims at exposing the relations between Universal Grammar and SLA, so as to find what strategies the language learners use in the process of language learning and what the causes of errors are. It is closely related with Chomsky’s points of Inner theory, Language Acquisition Device and Universal Grammar. Its psychological basis is cognitive theory.
1.2 Procedures of EA
Corder (1974) suggests that the procedures of EA are sample collection, identification, description, explanation and evaluation of errors.
The first step is the sample collection of errors. This starting point in EA is deciding what samples of learner language to use for the analysis and how to collect these samples. Two issues need to be made clear here: one is whether the learner language reflects natural, spontaneous language use, or is elicited in some way; the other issue is whether the samples of learner language are collected cross-sectionally (at a single point in time) or longitudinally (successive points over a period of time).
The second step is the identification of errors. Once a corpus of learner language has been collected, the errors in the corpus have to be identified. It is necessary to decide, therefore, what constitutes an error and to establish a procedure for recognizing one.
An error can be defined as a deviation from the norms of the target language. A second question concerns the distinction between errors and mistakes (Corder, 1967). An error takes place when the deviation arises as a result of lack of knowledge. It also represents lack of competence. A mistake occurs when learners fail to perform their competence. Corder argues that EA should be restricted to the study of errors and that mistakes should be eliminated from the analysis. A third question concerns whether the error is overt or covert (Corder, 1971). A fourth question concerns whether the analysis should examine only deviations in correctness or also deviations in appropriateness.
The third step is the description of errors. There are a number of classificatory systems that have been used in EA. Dulay and Burt (1973,1974) identify four types of error according to their psychological origins:
(a) interference-like errors
(b) first language developmental errors
(c) ambiguous errors
(d) unique errors
So far a comparatively complete classification has been provided by Dulay, Burt and Krashen (quoted in Liushaolong, 1998). They suggest that there are four kinds of error taxonomies:
(a) linguistic category classification: phonology, graphology, grammar, lexis, text or discourse;
(b) the surface structure taxonomy: omission, addition, misformation, misordering;
(c) comparative taxonomy: developmental, interlingual, ambiguous and unique;
(d) communicative effect taxonomy: global and local.
The fourth step is the explanation of errors. Explanation is concerned with establishing the source of the errors. This stage is the most important for SLA research as it involves an attempt to establish the processes responsible for L2 acquisition.
Abbott (1980) suggests that the aim of any EA is to provide a psychological explanation. Taylor (1986) points out that the error source may be psychological, sociolinguistic, epistemic, or may reside in the discourse structure. Different error classifications result in different research findings and different explanations of errors. Flick (1979) notes that error classification is an arbitrary matter subject to the individual biases and point of view of the researchers.
However, there is general agreement over the main diagnosis-based categories of error (James, 2001). There are four major categories:
(a) interlingual errors
(b) intralingual errors
(c) communication strategy-based errors
(d) induced errors
Interlingual errors refer to the errors which are caused by mother tongue interference; intralingual errors are errors in learner language that reflect learners’ transitional competence and which are the results of such learning processes as overgeneralization; communication strategy-based errors refer to errors resulting from the lack of communication strategy; and induced errors arise in learner language when learners are led to make errors that otherwise they would not make by the nature of the formal instruction they receive.
The final step is the error evaluation. Error evaluation studies proliferated in the late 1970s and in the 1980s, motivated quite explicitly by a desire to improve language pedagogy. Where the purpose of the error analysis is to help learners learn a L2, there is a need to evaluate errors. Some errors can be considered more serious than others because they are more likely to interfere with the intelligibility of what learners say or write. Teachers will want to focus their attention on these. For example, some errors, known as global errors, violate the overall structure of a sentence and for this reason may make it difficult to process. Other errors, known as local errors, affect only a single constituent in the sentence and are, perhaps, less likely to create any processing problems.
1.3 Related researches of EA from abroad and the limitations of EA
Corder’s “The Significance of Learners’ Errors” marked the beginning of error analysis. EA constituted the first serious attempt to investigate learner language in order to discover how learners acquire a L2. Its heyday was in the 1960s and 1970s. Since Corder, a lot of studies of learner errors have been carried out. Among them, there are case studies on children’s mother tongue acquisition (Bowerman, 1974, 1982, 1983, quoted in Bowerman, 1998), and researches on both children and adults’ SLA (Ravem, 1968; Milon, 1974; Dulay and Burt, 1971, 1974; Corder, 1967; Jain, 1974; Richards, 2001). Case studies on children’s language acquisition are longitudinal. They reflect children’s learning processes at different stages of development. Bowerman followed her two English-speaking daughters’ development, recorded their everyday spontaneous utterances, and found that the errors they had committed in the process of their mother tongue acquisition were systematic. At the same time, she also made comparison with the errors she had collected from other children. The errors include:
(a) overgeneralization of dative alternation, i.e. I said herno; Don’t say me that.
(b) causative formation, i.e. Do you want to see us disappear our heads?
(c) passivization, i.e. Why is the laundry stayed open all night?
(d) local alternation, i.e. spray paint on the wall.
(e) un-prefixation, I.e. I’m never going to unhate you.
Bowerman claimed that her utterance data had been proved by experimental data. Milon (1974) examined a seven-year old Japanese boy’s learning process of English negation and its erroneous forms indicated that it was similar to the process of his mother tongue acquisition.
However, the majority of EA researches have been cross-sectional. The bulk of the empirical work in EA has focused on determining what proportion of the total errors in a corpus is due to transfer as opposed to intralingual errors. A good example of the ‘proportion study’ that investigated this behaviourist/mentalist question is Dulay and Burt (1974). They made a survey on 179 Spanish children ranging in age from 5 to 8, about their English learning. After analysis of their 513 syntactic errors, they examined morphological features like past tense inflection. Based on this EA they claimed that 85 percent were developmental, 12 percent were unique, and only 3 percent interference.
Dulay and Burt’s research constituted a powerful attack on the Contrastive Analysis Hypothesis and was proved by some other scholars (Wilkins, 1972, Felix, 1980). However, other researches do not bear out their finding and there has been little agreement as to exactly what proportion of errors can be put down to transfer. The following table 1 (Ellis, 2000: 29) lists the percentage of interference errors reported in different studies since the early 1970s.
Study % of interference error Type of learner |
Grauberg 36 First language German-adult, (1971) advanced George 33 Mixed First language-adult,(1972) graduateDulay and Burt 3 First language Spanish-children, (1973) mixed level tran-Chi-Chau 51 First language Chinese-adult, (1974) mixed level Mukattash 23 First language Arabic-adult (1977) Flick 31 First language Spanish-adult, (1980) mixed level Lott 50 First language Italian-adult, (1983) university |
From table 1, we see that the mean percentage is more than 33 percent, Dulay and Burt’s 3 percent is conspicuously below the number. This is likely to be the result of a number of variables.
After that, EA is more likely to serve as a means for investigating a specific research question rather than for providing a comprehensive account of learners’ idiosyncratic forms. For example:
Felix (1981) and Pavesi (1986) use error analysis to compare the language produced by instructed and natural learners. Bardovi-Harlig and Bofman (1989) wished to investigate the differences between a group of learners who successfully passed the Indiana University Placement Exam and a group who failed to do so. They examined the nature of the errors which the two groups produced in one part of the examination--written composition--and found, unremarkably, that the pass group made fewer overall errors than the non-pass group and, more interestingly, that the major differences were in the number of lexical and morphological rather than syntactical errors. Santos (1987) carried out an error evaluation in order to investigate linguistic claims regarding markedness (Ellis, 2001: 70).
In recent years, studies on second language reading and writing in relation to mother tongue transfer and its psychological cognitive process have become the focus of SLA research (Francis, 1999, Koda, 1999, Sasaki, 1996). The latest findings show that the greater the difference, the greater the interference (Koda, 1999); the writing ability of mother tongue has a distinctive effect on second language level (Kim, 1997, Sasaki, 1996); the constructive process of mother tongue writing is similar to the second language writing and the writing strategy of mother tongue influences the second language writing (Kubota, 1998). James (2001) introduced ongoing work on Computerized Corpora of Error and professor Gui Shichun in China has also set up this kind of corpus as CLEC (Chinese Learner English Corpus).
A frequently mentioned limitation is that EA fails to provide a complete picture of learner language. Most of the EA studies are cross-sectional in nature, affording only a very static view of L2 acquisition. In many cases little care has been taken to separate out the errors made by learners at different stages of development. Therefore, EA has not proved very effective in helping us understand how learners develop knowledge of a L2 over time. However, this weakness is not a necessary one. EA can be used in longitudinal studies. A study of how learners’ errors change from one stage to another can shed light on the process of L2 acquisition (Ellis, 2001).
EA is limited in a second way with regard to the procedures of EA. Initially, the identification of errors is not easy work. There are sorts of errors which cannot be described as mere failures to memorize a segment of language or as occasional lapes in performance due to memory limitations, fatigue, and the like (Richards, 2001). Moreover, many error studies and classifications lacked reliability and had limited explanatory power, because of the lack of precise criteria for classifications, overlappings of some of the categories, and the subjectivity of their interpretations of errors.
The third problem is the insufficiency of error analysis. Hammarberg points out ‘insufficiency of error analysis’, which for him lies in its one-side practice of ‘analyzing out the errors and neglecting the careful description of the non-errors’ (Hammarberg, 1974: 185). Schachter (1974) also discovers what she has seen as a fundamental flaw in EA-- a failure to recognize that learners have a tendency to avoid target language items they are not sure about, and so not to commit errors which they would be expected to commit (James, 2001). For example, Schachter (1974) conducted an analysis of the relative clause errors produced by two sets of learners (one Arabic and Iranian, and the other Chinese and Japanese). She found that the first group of learners made more errors than the second group, despite the fact that relative clause structures existed in their L1s and did not exist in Japanese and Chinese. However, she also discovered that the Arabic and Iranian learners made many more attempts to use relative clauses than did the Japanese and Chinese learners. She concluded that learners resorted to avoidance if they found a structure difficult. EA, which focuses exclusively on what learners do, has no way of investigating avoidance and is, therefore, seriously limited.
However, EA has made a substantial contribution to SLA research. As Corder (1967) points out in “The significance of learners’ errors” that errors are significant in three respects: they tell the teacher what needs to be taught; they tell the researcher how learning proceeds; and they are a means whereby learners test their hypotheses about the L2. This is patently a very positive assessment of EA, announcing a program that might well take several decades and not just a heyday to complete.
2.4 Related research in China
In China, scholars have begun to introduce EA theory and its basis of IL (interlanguage) since 1980s. Professor Gui Shichu (1985) points out that EA is a common teaching method, through which an EA teacher can find the individuality and generality of the language errors of students. In his book “Psychological Linguistics (1985)”, he classifies errors into four categories and has discussed the two different attitudes toward them. Other professors set a special chapter in their books to introduce EA and have made comments about it. They conclude that second language learning is a creative construction (Wang Chuming, 1990, zhang Jianzhong, and Yu Hongzhen, 1998).
Scholars Dai Weidong and Shu Dingfang (1994) have made introductions and comments on the current situation of CA, EA and IL theory; and the other scholars are also involved in this kind of research (Tang Chenxian, 1997, Yang Lianrui, 1996). They have discussed research procedures, significance and limitations of EA and IL. Moreover, fossilization as a special phenomenon in IL has also been introduced (Wang Zongyan, 1988, Dai Weidong, 1994, Yang Lianrui, 1996).
In addition to the above theoretical survey studies of EA and its related theory IL, many teachers and scholars have begun to conduct experimental studies in order to inquire the linguistic and pragmatic errors in the use of TL (Huang Chidong, 1984, and He Ziran, 1986, quoted in Liu Shaozhong, 1997, Liu Shaolong, 2000, Xiao Xuyue, 2001, Cai Jinting, 2000, Sun Mianzhi, 2001). However, researches conducted on what proportion of Chinese interference in SLA are rare. The research being presented here is to give a further test whether 3 percent or 33 percent of mother tongue interference is valid in Chinese context.
3. Research design
This research employs both quantitative design and qualitative design. Quantitative research is empirical research where the data are in the form of numbers, and qualitative research is empirical research where the data are not in the form of numbers. Both of them are primary researches and are used to achieve different objectives. For the former, its goal is to examine variables identified and test hypotheses; whereas for the latter, the aim is to identify variables and generate hypotheses. In other words, quantitative research is more powerful when the research is to validate the findings in previous studies; qualitative research is more useful when the research is exploratory in nature (Wen Qiufang, 2002).
3.1 Quantitative design
3.1.1 Subject
The subjects of the research are 355 college students. 127 of them are four-year students from Anhui Normal University, and the other 228 are three-year students from Wuhu Teachers College. The following table 2 and table 3 are the detailed descriptions of the subjects.
Table 2
Number male female |
Four-year student 127 87 40 Three-year student 228 158 70 Total 355 245 110 |
Table 3
Number major non-major |
Four-year student 127 50 77 Three-year student 228 133 95 Total 355 183 172 |
3.1.2 Instrument
3.1.2.1 Samples collection
Students are required to write an essay entitled “My Favorite Person” in about 100-150 words within 30 minutes. The whole writing process is observed by their comprehensive English teachers in their regular class. Therefore, the samples can reflect natural, spontaneous language use, which are generally preferred.
3.1.2.2 Questionnaire
The questionnaire is to survey the sources of errors and the students’ psychological process of committing errors. Designed on the types of errors found in the students’ written products, it consists of 15 questions. Each has 3 objective choices A, B, C and one subjective choice D. The choices for each question are the same: choice A, interference (interlingual); choice B, developmental (intralingual); Choice C (neither A nor B) and choice D are open . The choices provided here are based on Dulay and Burt’s classification (1973,1974) with slight modification. Students are asked to answer the questionnaire, either by selecting an objective choice or by providing their own subjective one. The questionnaire is designed in Chinese, in order to avoid students’ misunderstandings of the questions. The work is to be finished in regular class with teachers’ guidance and observation.
3.2 Qualitative design
3.2.1 Descriptive analysis
Descriptive analysis is based on the data collected from students’ written products, such as analysis of frequency distribution of different categories of errors, the source of errors and the proportion of Chinese interference.
3.2.2 Interview
In order to know the students’ psychological process of committing errors further, we have individual interviews with 20 students. The interview is to be conducted by asking them questions. The whole process of interview is to be recorded.
4. Data collection and analysis
4.1 Collection of errors
Collection of errors consists of collection of students’ written products and identification of errors
The first stage is to collect students’ written products. Teachers responsible for different classes collect the students’ products, and bring them together. Among 400 essays, 355 are considered qualified except some copies, very short essays and empty papers. The following table 4 shows the factors of samples of learner language.
Table 4
Factors Description
A Language
Medium Learner written production
Genre An essay
Content My Favorite Person
B Learner
Level Intermediate
Mother tongue Chinese
English learning experience 8---10 classroom learning experiences
The second stage is the identification of errors. The criteria for identification of errors consist of 3 points:
(1) An error takes place when the deviation arises as a result of lack of knowledge;
(2) An error is overt deviation;
(3) An error is deviation in correctness.
So here errors refer to grammatical mistakes. The procedure of error identification has been conducted in three steps. First, students’ essays are corrected by two classes of the junior English majors from Anhui Normal University. Before doing this, they have read books on error analysis and finished an essay “Error analysis and mother tongue transfer”, so they are armed with the knowledge of EA. They correct the essays in their regular class according to the teachers’ guidance. Then, based on the first correction, a group of 6 competent and careful female students are selected to modify the first correction, in accordance with a common standard set by the teachers. At last, teachers make the final modification. After a hard, careful and painstaking process, 4798 errors have been identified. However, we have to point out that students’ productions are still open to modification.
4.2 Descriptive statistics of error analysis
Based on the classifications mentioned in the literature review and with reference to the most common errors in the college students’ writing, we first classified the 4798 errors into 3 categories: lexis, grammar and Chinglish sentence, and then subdivided them into 15 subcategories. The following table is the descriptive statistics of the 15 types of errors according to their frequencies and percentages that appear in students’ productions.
Descriptive analysis of error classification and the frequency distribution
Table 5
Type Brief description Frequency Percentage |
1 Errors in tense, aspect and voice of verbs 1216 25 2 Chinglish sentences 694 14 3 Redundant and improper words and 656 14 phrases 4 Spelling errors 589 12 5 Misuse of cognate words 502 10 6 Omission and misuse of propositions 361 8 7 Omission and misuse of articles 171 3 8 Errors in the plural form of nouns 170 3 9 Omission of copula “be” 91 2 10 Omission and misuse of possessive adjectives and pronouns 91 2 11 Misuse of the words with similar morphological forms 90 2 12 Omission and misuse of conjunctions 80 2 13 Misuse of infinitive, gender, -ing form, and -ed form of the verbs 58 1 14 “There be structure” errors 16 under1 15 Negation errors 13 under1 Total 4798 100 |
From table 5 we can see that verb errors in tense, aspect and voice run the top rank. Chinglish sentences take up the second place, and redundant and improper words and phases cover the third. First, the tense, voice and aspect of verbs are the key point of grammar, and also one of the most difficult parts for English learners. Moreover, Chinese does not share this feature. That might be the reason why 25 percent of the students’ errors occur here. Then, Chinglish sentences are the result of Chinese way of thinking and Chinese syntactic structure. Finally, from the concrete errors we know that redundant and improper words and phrases are mainly affected by the correlation of various factors.
Among 355 students, 6 variables exist: male, female, English major, non-English major, four-year student and three-year student, which belong to three groups respectively. The following table 6 shows the comparisons and correlations between each group.
Descriptive statistics of comparisons and correlations between each group of variables
Table 6
Variable Number of students Total error Average Correlation |
Male 110 1559 14.2 .0026 Female 245 3239 13.2 .0026 English major 183 2250 12.3 .0070 Non-English major 172 2548 14.8 .0070 Four year student 127 1582 12.5 .0042 Three year student 228 3216 14.1 .0042 |
The table indicates that there is slight difference among the averages, and the correlations between each group are not distinctive (r=.0026; -.0070; -.0042). In general, the performance of the whole students in intermediate stage is not unusual.
4.3 Descriptive statistics of the sources of errors
Based on quantitative analysis of questionnaire, the research determines proportions of the sources of errors. The following table is about descriptive statistics of frequency distribution of the sources of errors.
Table 7
Choices Frequency Percentage |
A(interference) 1102 21 B(developmental) 2678 50 C(neither A nor B) 632 12 D(the other) 907 17 |
Through the table we can see that the developmental errors cover half of the total. They are the main source of errors. To some extent, they have proved the Mentalist’s interpretation of SLA, the natural and developmental route and the L2=L1 acquisition hypothesis. Mother tongue interference takes up 21 percent, which is in agreement with the estimates varying from 3 percent to 25 percent of all errors (James, 2001). Furthermore, we know that the sources of errors are various, not only developmental errors, interference and the correlation of the two, but also many other factors.
Table 8 is the descriptive statistics of frequency distribution of the sources of errors among variables.
Table 8
Choice Variable |
interference developmental neither A nor B the other |
||||
Male |
320(19%) |
869(53%) |
204(12%) |
248(16%) |
|
Female |
782(21%) |
1809(49%) |
428(12%) |
650(18%) |
|
English major |
560(20%) |
1364(50%) |
350(13%) |
471(17%) |
|
Non-English major |
542(21%) |
1314(51%) |
282(11%) |
436(17%) |
|
Four year student |
379(20%) |
948(50%) |
181(10%) |
380(20%) |
|
Three year student |
723(21%) |
1730(51%) |
451(13%) |
527(15%) |
|
Table 8 shows that frequency distribution of each source of errors is more or less around the average. In addition to this, the proportion of both interference and developmental errors between English majors and four-year students are the same: 20 percent and 50 percent, with interference 1 percent lower than the average. Similarly, proportion of the two error sources between non-English majors and three-year students are also the same: 21 percent and 51 percent with developmental error, 1 percent higher than the average. To sum up, there is not distinctive difference between 6 variables, thus it indicates that students share the same attitude toward the source of errors.
The following table is the descriptive statistics of frequency distribution and percentage of the sources of the 15 types of errors.
Table 9
Choice Variable |
interference developmental neither A nor B ambiguous |
|||
1 |
64(18%) |
181(51%) |
39(11%) |
71(20%) |
2 |
291(82%) |
30(8%) |
9(2%) |
25(8%) |
3 |
69(19%) |
203(57%) |
23(7%) |
60(17%) |
4 |
0(0%) |
189(53%) |
34(10%) |
132(37%) |
5 |
57(16%) |
188(53%) |
36(10%) |
74(21%) |
6 |
51(14%) |
217(61%) |
26(8%) |
60(17%) |
7 |
86(24%) |
182(51%) |
38(11%) |
50(14%) |
8 |
52(15%) |
192(54%) |
46(13%) |
63(18%) |
9 |
70(20%) |
154(43%) |
61(17%) |
69(19%) |
10 |
77(22%) |
190(54%) |
56(16%) |
32(9%) |
11 |
17(5%) |
212(60%) |
31(9%) |
95(26%) |
12 |
92(26%) |
178(50%) |
54(15%) |
33(9%) |
13 |
30(8%) |
235(66%) |
40(11%) |
55(15%) |
14 |
76(21%) |
175(49%) |
61(17%) |
42(12%) |
15 |
86(24%) |
156(44%) |
67(19%) |
45(13%) |
From this table in combination with table 5, we can note 3 distinctive features of errors: first, the first 5 types of errors take up 75 percent of the total; secondly, 82 percent of students attribute Chinglish sentences to Chinese interference; thirdly, 0 percent (none) of the students considers that spelling errors are affected by Chinese interference. What is the source of spelling errors? We will discuss it in the following paragraphs.
5 Discussion
To learn a second language, learners unavoidably make errors due to the great difference between the native language and the foreign language. Through the research, we have known what the most common mistakes made by college students are in their writing. There are errors in phonology or orthography, lexicon, grammar and discourse. Most errors can be classified into two large categories. Now we will discuss the possible causes of the errors found in the students’ writing.
5.1 Interlingual error or interference
Interlingual error is also called mother tongue transfer. In the case of transfer, learners use their previous mother tongue experience as a means of organizing the second language data. The early stage of learning a second language is characterized by a good deal of interlingual transfer from native language, or interference of mother tongue. In the process of language learning, before the system and structure of the second language are established, the native language system and structure are the only linguistic system and structure in the previous experience upon which the learners can draw. For instance, the rules, phrases or items of mother tongue are mechanically applied or negatively transferred to the second language contexts. Interlingual transfer explains many errors made in the process of second language learning.
In studying the errors committed by students in their products, we know that, except spelling errors, the other 14 types of errors are more or less affected by the interference of Chinese. Conspicuously, 82 percent of the students think that “Chinglish” is the result of interference. Based on the concrete errors in students’ products, we find the following criteria may categorize interlingnal errors.
(a) Errors in tense, aspect and voice of verbs
Chinese is one of the many languages that do not use inflectional means to mark tense. It is obvious that the whole Chinese system of tense, modality, and aspect differs greatly from that of English. English verb tense is divided into three main respects: the present tense, the past tense and the future tense. There are altogether sixteen tenses in English. No wonder it is a hard task to use these complicated tenses properly. These errors cover 25 percent of the total. See the following examples:
“我昨天看电影。”
“我明天看电影。”
We can find they are exactly alike except two adverbial modifiers of time “昨天” and “明天”. But it is wrong to translate them into English according to the same rule. They should be :
“I went to see the film yesterday.”
“I’m going to see the film tomorrow.”
At the early stage of English learning, learners make some simple mistakes for the interference of Chinese. However, the strange thing is that they cannot escape errors like these even if they have mastered all kinds of tenses. The following are typical examples from students’ work.
(1)×A teacher was an important resource.
√A teacher is an important resource. (Misuse of present tense for past tense.)
(2)×If you want to be successful in the future, you must be paid attention to your study.
√If you want to be successful in the future, you must pay attention to your study. (Error of passive voice)
(3)×It rains outside.
√It is raining outside. (Misuse of present tense for present progressive.)
(4)×Several days later I’m recovered.
√Several days later I recovered. (Misuse of present tense and passive voice)
(5)×I wish you can realize your dream.
√I wish you could realize your dream.(The learner did not know that clause after “wish” should be in subjunctive mood.
The same example like:
(6)×If her friend did not encourage and help her, she couldn’t achieve success.
√If her friend had not encouraged and helped her, she couldn’t have achieved success.
(b) Chinglish sentence
Among Chinese learners, Chinglish is frequent not only in grammar but also in word. Chinglish is caused by the Chinese way of thinking and Chinese syntax, namely, Chinese sentence structures. Native English speakers make various mistakes, however, they do not write or speak Chinglish for they think the way the English people do. We all know a typical joke about Chinglish: “Good good study, day day up!”(好好学习，天天向上). As luck would have it, the research finds another variation: “Work work hard, and day day up!” Such mistakes are very ridiculous and reflect the serious problem existing among Chinese English learners. According to the data from questionnaire, 82 percent of the students admit Chinglish is the direct victim of Chinese interference. We analyze it in the following two respects:
(b1) Chinglish affected by the Chinese way of thinking
(1)×How much is the price of this pen?
√How much is this pen?
(2)×Mother used to go out to buy things on Friday.
√Mother used to go shopping on Friday.
(3)×Time passed in my ignorance.
√How time flies!
(4)×I suggest that we go out to play.
√I suggest that we go out for fun.
The above sentences are all word-for-word translations. In English the word “play” has more meanings than the Chinese word “玩”.
(5)×Because he was ill, so he didn’t go to school.
√Because he was ill, he didn’t go to school.
(6)×Although he likes English, but he can’t learn it well.
√Although he likes English, he can’t learner it well.
In the two sentences the habit of using Chinese conjunction has been completely transferred to the English diction.
(7)×The computer makes us to be able to learn more knowledge.
√The computer enables us to learn more knowledge.
(8)×The English teacher let us to do more exercises after class.
√The English teacher asked us to do more exercises after class.
Affected by the Chinese verb “让” and “使”, Chinese English learners produce many sentences like “let somebody to do something” and “make somebody to do something”.
(9) ×A: You are beautiful, Xiao wang!
B: You flatter me, I feel ashamed.
√A: You are beautiful, Xiao wang!
B: Thank you.
(10) ×A: Thank you for helping me.
B: It’s my duty to do so.
√A: Thank you for helping me.
B: It’s my pleasure.
Here Chinese culture has affected the way of thinking.
(b2) Chinglish affected by Chinese sentence structure
Co-occurrence of several verbs is frequently seen in Chinese syntax, and such structures are an indication of pidginization in language acquisition. For example:
(1) ×See book write.
√Write by looking at the book and copying from it.
(2) ×There are 5 students take part in the sports meeting this time.
√There are 5 students to take part in the sports meeting this time.
(3) ×I find, find, find, don’t have!
√I looked high and low for it but I could not find it.
Example (3) is a feature of using repetition to signify duration of action or state, and an obvious feature of the Chinese sentence pattern. However, too many repetitions in Chinese should not be directly transferred to English.
(4) ×You remember new words more and more, then your understanding becomes more and more easy.
√The more words you remember, the easier your reading becomes.
From the sentence we can see the interference of Chinese comparative degree of adjectives.
(5) ×I think he will not come.
√I don’t think he will come. (Influenced by Chinese word order )
(6) ×A: Are you hungry?
B: Hungry.
√A: Are you hungry?
B: Yes, I am.
(7) ×A: Has she recovered from her bad cold?
B: Recovered.
√A: Has she recovered from her bad cold?
B: Yes, she has.
Habitually, Chinese students often use a single verb instead of the SV or SVC structure.
(8) ×I don’t happy.
√I’m not happy.
(9) ×My mother tired because of hard work.
√My mother was tired because of hard work.
In Chinese, adjectives can be used as predicate alone, while in English the verb “be” is indispensable.
We have discussed various errors of Chinglish. How to get over them? The following is an error analysis in lexical interference.
Comparing Chinese and English, we find there are complete correspondence, partial correspondence and non-correspondence. Learners should pay more attention to the last two because they are related to a nation’s cultural background and ways of thinking.
Besides, learners need to read a great number of simplified English materials, such as, simplified original English books, magazines and newspapers, so as to acquire the English way of thinking and language awareness. When coming across a good sentence, learners can ask themselves whether they can use this beautiful sentence. If not, take it down, translate it into Chinese and retranslate it into English after some time and compare it with the original to see if there is Chinglish in it. The way of reading—translating—comparing is the process of avoiding “Chinglish”.
(c) Errors in the plural form of nouns
In Chinese, nouns are not marked for plural and lack of plural marking does occur frequently.
(1)×She bought some book for me.
√She bought some books for me.
(2)×We went our home respectively.
√We went our homes respectively.
It is obvious that the two examples are wrong in grammar because “book” and “home” are plural in meaning. We need to add a suffix “s” to the end of the words. The above examples are just one aspect of the singular and plural form. There are also many irregular plural forms of nouns, such as, knife, child, fish and deer. For example,
(3)×Two thiefs ran away.
√Two thieves ran away.
Learners know that the plural form of “child” is “children”, however, they still commit errors here. The following are different expressions found in the students’ work.
(4)×There are 2 childs in my family.
√There are 2 children in my family.
(5)×She loved us like her own childrens.
√She loved us like her own children.
If learners pay little attention to these words, they are likely to make more errors like this, even if the words are very simple. Learners may come across some words whose singular form and plural form are the same. “Sheep” can be “ a sheep” or “ flocks of sheep”. To make things more complicated, there are countable nouns and uncountable nouns in English. Learners are often confused about these words and fail to distinguish them. For example, the word “hair”, when used as a countable noun, is “hairs”, whereas when used as an uncountable noun, is “hair”. The following examples show the difference.
(6)×My mother has a few gray hair.
√My mother has a few gray hairs.
(7)×He has gray hairs.
√He has gray hair.
The following are more examples of the errors in countable and uncountable nouns.
(8)×I ate several breads for breakfast.
√I ate pieces of bread for breakfast.
(9)×There are some advanced equipments in our language lab.
√There is some advanced equipment in our language lab.
(10) ×Father told me a news.
√Father told me a piece of news.
“Fish”, used as a countable noun has two plural forms: fish and fishes. “Fish” is the more usual form used when referring to a mass of them in the water to be caught or seen, for example:
“My father caught a lot of fish in the river.”
And “fishes” is used to refer to different species of fish, for instance:
“I’m good at studying the fishes of the Atlantic Ocean.”
Learners should be alert to these words.
(d) Misuse or omission of articles
To some extent, misuse or omission of articles relate to the former point—the singular form and plural form of the noun. In Chinese, there are no definite articles or indefinite articles. Let us see the following phrase found in the students’writing:
“boy playing football”
It has two explanations. Learners may either forget the plural form of “boys playing football”, or commit an error of omission of articles. If learners indeed mean a singular form, an article must be put before “boy”. Either “a boy” or “the boy” is suitable here. See the following examples:
(1) ×Sun is red.
√The sun is red.
(2) ×I want to be teacher in the future.
√I want to be a teacher in the future.
(3) ×After class there will be discussion about the English corner.
√After class there will be a discussion about the English corner.
In the first example, definite article “the” should be used before unique nouns. In example (2) and (3), indefinite article “a” should be put before countable nouns. Learners do not make such errors because of carelessness, but because of their mother tongue interference. The Chinese have not the habit of emphasizing anything with a word. So the sentence “书在桌子上” is right in Chinese while the translated English sentence “Book is on table” is wrong. The correct form is: “The book is on the table”. The definite article “the” is widely used in English to make the following nouns refer to a specific person, thing, event or group. However, there are still some exceptions. Here is a sentence like this:
“Computer plays an important role in modern society.”
At the first sight, it is almost perfect, and it may also appear in the newspapers as the headline of news. In fact, “the computer” is the right expression. Learners have overlooked the article because they have got some words mixed up. There is no definite article before some nouns such as “society”, and “nature”. However, some other words are different. Here are four examples:
“Life is wonderful.”
“The life of Clinton”
“I’m interested in literature.”
“I’m interested in the literature of the contemporary time.”
Such nouns used in a general sense, there is no article before the noun; otherwise the article is needed. The indefinite article “a” has a different form before the vowels, and sometimes it is difficult to distinguish them. For example:
(4) ×I’ m an university student now.
√I’ m a university student now.
(5) ×Our math teacher was a honest and patient person.
√Our math teacher was an honest and patient person.
Although there are only three articles in English, learners commit many errors of them. The cause may be either their carelessness or their mother tongue interference.
(e) Misuse or omission of prepositions
First, this error is closely related to the collocation of words. For instance:
(1) × The teacher is strict to us.
√The teacher is strict with us.
(2) × My mother is busy on working at home.
√My mother is busy in working at home.
(3) ×Our teacher set a good example for us.
√Our teacher set a good example to us.
(4) ×I agree your opinion.
√I agree with you.
One feature of English vocabulary is that adjectives, nouns and verbs can collocate with many prepositions, and result in many different meanings. In addition to this, verbs are divided into transitive verbs and intransitive verbs and especially some active verbs can collocate with many prepositions. In Chinese, this phenomenon does not exist. Therefore, the above examples are influenced by Chinese directly or indirectly.
Another difference is that, in Chinese, prepositions can be used as predicate, like a verb. Sometimes this habit is directly transferred to the use of English. See the following sentences:
(6) ×Theory cannot instead of practice.
√Theory cannot take the place of practice.
Here a preposition replaces verb, because in Chinese they have the same lexical meaning.
Frankly speaking, this kind of errors can be easily avoided if learners are careful enough to look them up in the dictionary.
(f) Errors in cognate words
Errors occur in cognate words, because the items share morphological or semantic features with an item in the native language.
(1) ×His life was dangerous.
√His life was in danger.
(2) ×I’m interesting in it.
√I’m interested in it.
(3) ×He wants to success in the future.
√He wants to succeed in the future.
√He wants to be successful in the future.
(4) ×We may have been poor, but we were always respectfully
dressed.
√We may have been poor, but we were always respectably dressed.
(5) ×We all know that fail is the mother of succeed.
√We all know that failure is the mother of success.
The above errors are all misuses of different forms of an underlying word respectively because those different forms are morphologically related with each other.
(6) ×He opened lights. He closed lights.
√He turned on lights. He turned off lights.
Chinese influence may be seen in the meanings of the lexical items. “Open” and “close” are different from “turn on” and “turn off”, so the influence from Chinese is clear.
(7) ×Eat already?
√Have you eaten?
(8) ×I don’t like it already.
√I don’t like it anymore.
Many students use “already” as an aspect marker. Apparently, it is the result of influence from Chinese. In Chinese, we use “已经” to express something we have already finished, which is not suitable in English.
To sum up, instances of mother tongue interference in students’ products can be found at the level of morphology, syntax, vocabulary and meaning. In the process of English teaching, teachers should attach great importance to the Chinese interference so that learners do not make so many mistakes. Here, we should point out: because of the limitations of the research the paper cannot analyze Chinese interference in pronunciation.
5.2 Intralingual errors or developmental errors
In addition to errors due to transferring rules from mother tongue (interlingual error), learners also make many errors showing that they are processing the second language in their own terms. Errors of this kind are called Intralingual errors, which refer to overgeneralization of the rules of the target language. Intralingual transfer occurs when learners have learnt some rules of the target language, but have not learnt the restrictions on their use. Generalization is, of course, a fundamental learning strategy in language. In order to make sense of our world, learners allocate items to categories, and construct “rules” which predict how the different items will behave. Sometimes, their predictions are wrong, thus causing overgeneralization. According to Richards (2001), “overgeneralization” means learners often construct some new structures that do not exist in the target language because they take it for granted that these structures are analogues. Two reasons can explain the phenomenon of overgeneralization. 1) Sometimes, the rule does not apply to a particular item, even though learners have allocated the item to the appropriate category. They must therefore learn an exception to the general rule. 2) The item belongs to a different category, which is covered by another rule. Learners must therefore either allocate the item to a different category they know, or they must construct a new category and rule.
Overgeneralization errors:
(1) ×He brushes toothes every day.
√He brushes teeth every day.
The reason is that learners of English have learnt a rule of forming plurals. This makes them predict that a noun can be made plural by adding “s” or “es”.
(2) ×He comed to the library.
√He came to the library.
“Come” and “go” lie outside the scope of the general rule of forming the past tense. The past tense can be formed by adding suffix “d” or “ed” to the end of original verb. However, “come” and “go” are exceptions.
(3) ×We are knowing each other.
√We know each other.
It over-generalizes the use of the rule of forming progressive.
(4) ×This shows that how capable he is.
√This shows how capable he is.
It over-generalizes use of “that” for introducing a noun clause.
(5) ×We must does it.
√We must do it.
It’s because of the overgeneralization of the third person singular.
(6) ×Don’t make noise in the classroom.
√Don’t make noises in the classroom.
It over-generalizes “noise” as an uncountable noun. In fact, “noise” is either uncountable or countable, but here it is used as a countable noun.
(7) ×Who did speak English just now?
√Who spoke English just now?
It is because of the overgeneralization of the rule of using auxiliary verb “do” forming interrogatives.
(8) ×She didn’t saw her in the classroom yesterday morning.
√She didn’t see her in the classroom yesterday morning.
It is the overgeneralization of the rule for using past tense in the past time.
(9) ×What you are doing?
√What are you doing?
It over-generalizes the rule of word order, or sentence structure, which makes learners unable to cope with the complexity of word order. Especially, in interrogative sentences learners are required to reverse the rules they have learnt previously.
The following typical intralingual or developmental errors Richards (2001) lists are the most common errors in students’ products. The pattern of the two tables is from Richards, and most of the examples are from students’ work.
Errors in the production of verb group
Table 10
(1) be+verb stem for verb stem
We are live in the town.
I am not speak English well.
She is teach us Chinese.
(2) be+verb stem+for verb stem+ed
He was died several years ago.
One day it was happened.
Mother is came home from work.
(3) wrong form after do
We do found him in the classroom.
I don’t speaks English often.
She didn’t agreed with me that time.
(4) wrong form after modal verb
We must worked hard to improve our English.
I can saw it in her eyes.
We shouldn’t forgot our duty.
(5) “be” omitted before verb+stem+ed (participle)
He convinced of his innocence.
This time father satisfied with my achievement.
I born in a farmer’s family.
(6) “ed” omitted after be+partieiple verb stem
He was often punish by the teacher.
My homework was cover with words she corrected.
Some of us were go out.
(7) “be” omitted before verb+ing
Boys playing basketball in the playground.
My cousin growing fast.
My mother doing things with great patience.
(8) verb stem for stem+s
She ask us who is your favourite person.
He like to help the other.
Our neighbour say that she is a good mother.
Errors in the distribution of verb group
Table 11
(1) be+verb+ing for stem
I’m interesting in that.
She was satisfying with my answer.
She was finding by the police.
(2) be+verb+ing for verb stem
I am coming from Nanjing.
I’m knowing him.
I’m missing you very much.
(3) be+ not+verb +ing for do+not+verb
I am not liking him.
I am not thinking he is right.
She is not working every day.
(4)be+verb+ing for verb +ed in narrative
--- in the afternoon we were going back. On Saturday we were going down-town, and we were seeing a film and after we were meeting my brother.
(5) have+verb+ed for verb+ed
They had arrived just now.
He had left yesterday.
I have written a letter to my parents last Sunday.
(6) have+be+verb+ed for be+verb+ed
He has been married long ago.
He had been killed in 1956.
I had been punished by my parents yesterday.
(7) verb+ed for have+verb+ed
We kept in touch with each other up to now.
I visited my teacher twice since graduation.
So far I wrote two letters to her.
(8) be+verb+ed for berb+stem
Tomorrow is belonged to us.
If you can’t use it properly, error is occurred.
Many products in American are comed from China.
In short, intralingual errors, namely, developmental errors take up half of the total. To some extent, it proves the mentalist accounts and interlanguage theory that errors are predicted to be similar to those found in L1 acquisition and errors are evidence of learning process. So in English language teaching, teachers should pay more attention to different developmental stages of language learning.
5.3 The other errors
Based on questionnaire and interview, the research shows that students attribute the other 29 percent of errors to various factors, such as unsystematic knowledge of grammar, lack of training at usual time, limited vocabulary, complexity of English structure, poor memory, and nervousness. According to these factors, we divide the other errors into three subcategories: errors due to communication strategy, performance errors and spelling errors.
(a) Errors due to communication strategy
When learners have to convey something beyond their current knowledge of English, they have to turn to communication strategies for help.
According to James (2001), communication strategy refers to the learners’ assumption that if learners can say X in the L2, then they must be able to say Y. Lacking the required form, it must be all right to use another near-equivalent L2 item that they have learnt. The most general term for this is “approximation”. It takes on a number of forms, the first of which is to use a near synonym. An example is the substitute of “his success in exam” for “his good performance in exam”. Alternatively, one can use a subordinate term, for example: “fruits” for “grape”. A third option is to use an antonym or opposite: “not happy” for “sad”. A fourth option is to coin words by using ready-made or rote-learnt chunks and fragments of phrases or sentences from memory. Besides, some learners paraphrase how to convey what they do not know indirectly. For example: “should have + past participle structure”, it can indicate the situation or result that is opposite to the past. Learners who have not learned such a sentence structure have to write or speak long sentences instead: I should have finished this work last night, or I had the intention of doing the work last night, but I was too busy-----. Communication strategy can generate redundant, and improper words, phrases or sentences easily, sometimes, even inaccurate words or sentences. Therefore, redundant and improper words or phrases and some Chinglish sentences in students’ products find full interpretations here.
(b) Performance errors
According to Richards (2001), learners make performance errors when they are speaking or writing long sentences. These may be caused by carelessness, poor memory or complexity of a sentence.
(1) ×The number of students recorded by the school authorities are greater than last year.
√The number of students recorded by the school authorities is greater than last year.
The subject of the sentence is “number” not “authorities.
(2) ×That’s the problem I don’t know how to solve it.
√That’s the problem I don’t know how to solve
It is easy to make performance mistakes, because learners pay much attention to thoughts and to the development of the content, overlooking the form of it.
(c) Spelling errors
Finally, we want to discuss the possible causes of spelling errors. From table 5 we know that spelling errors are the fourth common errors that students commit in their products and table 9 shows that no one considers them to be the results of interference. Performance errors can explain part of them, such as, anxiety, bad memory, and carelessness. What are the main possible causes of the spelling errors? Let us make analysis of it.
James (2001) analyzed the spelling errors of German learners of English and found that the L2 spellers had at their disposal four ‘routes’ to spelling an L2 word: the routes via L1 phonology, via L2 phonology, and via L1 and L2 graphology. Moreover, these different resources may influence L2 spelling, either separately or in combination. L1 transfer was found to be operative in spelling, but not in proofreading.
Xiao xuyue (2001) divided various spelling errors into three categories: sound deviation, graphemic substitution and coinage.
Sound deviation refers to the errors that derive from incorrect pronunciation. It appears by means of operations as omission, addition and misordering of the sound; graphemic substitution refers to words which have the same pronunciation, but wrong form with target words, because of addition, omission of the silent sound, or double letters transfer; coinage refers to words coined by language learners by imitating some morphological features of other words. The following tables are examples of sound deviation, graphemic substitution and coinage in students’ products respectively.
Type |
Example |
|
Erroneous word |
Target words |
|
addition omission misordering |
condiction behavor sovled |
condition behavior solved |
Graphemic substitution
Table 13
Erroneous word Target word Error description |
playwriter playwright different form with same pronunciation arguement argument addition of silent sound frend friend omission of silent sound arrording according double letters transfer |
Coinage
Table 14
Error type Erroneous word Target word |
coinage requite require inflectional teached taught morphological virturous virtuous |
In addition to the above categories, there are also many slips. In general, all the errors above are what we have found from students’ products, and some of the errors are what learners always commit in the process of English learning. Both interlingual errors and intralingual errors, which are caused by negative transfer should be taken into account in the preparation of language teaching programmes. Teachers need to ascertain what are the particular problems for Chinese-speaking learners of English. Only by exploring the roots of errors can learners correct them and assure themselves of a bright future in English learning.
5.4 The result of 21 percent interference
The proportion of 21 percent of interference is in line with James’ estimate varying from 3 to 25 percent (James, 2001). However, it differs from some Chinese scholars’ estimates and findings. Professor Wang Zongyan (1980) points out that it is not easy to determine to what extent mother tongue interferes with Chinese learners in English learning. He agrees with the estimate that a third of the Chinese learners’ errors come from mother tongue interference. In addition, Professor Sun Mianzhi (2001) notes that 95 percent of the students admit while they translate or write in a foreign language, their mother tongue will interfere. The result is based on questionnaires answered by 200 second-year students, non-English majors. In combination with his many years of teaching experience, he concludes that students’ “output” errors affected by mother tongue interference are no less than two thirds. Moreover, Tran-chi-chan (1974) claims the result of 51 percent of Chinese interference by doing research on the first language Chinese adults with mixed levels. Then, is the result of 21 percent interference found in this research valid? The answer is definite. Apart from the limitation of error analysis: its lack of a precise criterion for error classification, an overlapping of some of the categories and the possibility of multiple explanations, there are still reasons from both learners and teachers. Many factors are involved in the above situation. The reasons are as follows:
The first problem comes from the students’ side. From the questionnaire and interview we know that most of the students attribute the other 29 percent of errors to their unsystematic knowledge of English grammar, lack of training and limited vocabulary. Let us analyze them respectively. First, errors coming from unsystematic knowledge of grammar indicate most of students have not laid a solid foundation in their elementary stage of English learning. However, because of the large enrollment in recent years, they have obtained places in the college. Nevertheless, their academic quality has been greatly reduced. That is why the president of Beijing University announced to society: in order to guarantee the quality of students, they will not enlarge their enrollment anymore. Then comes the lack of training and limited vocabulary. Because the majority of students in our college have taken part in self-taught examinations, systematic learning of English has been ignored. The others, because of access to the Internet, spend most of their spare time chatting and playing games on it. They are not willing to do more reading and writing. In short, measures have to be taken to guarantee the quality of college students.
The second problem comes from the teachers’ side. In recent years, because of the large enrollment of students, college is in great need of English teachers. So a teacher with 18 to 20 periods of class each week is a common phenomenon. Every day teachers are busy with their teaching. They have not enough time to prepare their lessons and to correct students’ work, let alone to improve themselves. Consequently, the teaching result can be imagined. In addition, this kind of teaching situation has been proved by the studies of some researchers. Liao Lifang, and Qin Aosong (2000) pointed out that 57 percent of English majors are not satisfied with current English teaching. And Wang Qimin (2002) shows that 88 percent of college students are not satisfied with the current English teaching.
In the English department of our college, because of the lack of teachers, some basic courses, like grammar, have been cancelled. In the course of teaching, students often ask teachers to teach them grammar. Most of the teachers take it for granted that the students have grasped basic grammar in the middle school and have attached no importance to the problem. From the research we know how serious the problem is! Errors of the verb group listed by Richards (2001) are tremendous in students’ essays.
To sum up, the problems reflected from the cause of the other errors indicate that our present situation of college English teaching is not satisfying. It needs attention from all levels of society.
5.5 Insight into English teaching
With the development of economy and China’s entry into WTO, English learning now enjoys widespread appeal. The stress on communicative competence highlights English teaching in colleges. It has been written in the syllabus and become the goal of college English teaching. The ultimate aim of English teaching is to cultivate student’s communicative competence. However, we cannot overlook one thing: communicative competence needs linguistic competence to pave the way. If learners have not mastered linguistic competence, how can they reach communicative competence? According to the current situation revealed by research, we know it is high time for us to take actions to deal with the problem.
So based on the research, firstly, we should attach much importance to grammar. English majors in our college need to restore the course of grammar in the basic stage of college English teaching. For non-English majors, attention should be given to grammar teaching in college English course.
Secondly, top priority should be given to vocabulary in comprehensive English teaching. Vocabulary is the most active and basic part in English language. From the research we know pure spelling errors of vocabulary are the fourth common errors students commit, and the other errors, such as errors in prepositions, articles, conjunctions, redundancy and improper words or phrases, misuse of similar morphological forms of an underlying word and omission of possessive adjectives and pronouns, and Chinglish are all more or less related with English vocabulary. The following is the points we should pay attention to in teaching vocabulary.
(a) Pronunciation
To know words, first learners learn how to pronounce them correctly. This is not difficult, and of course is not the point we will discuss. What we want to stress here is to make comparison or contrast between words, which share similar features either phonetically or phonologically. For example:
(1) words which are similar in phonetics:
be see peel weather
bee sea peal whether
(2) words which are similar in phonology:
tear latter decree chin
dear ladder degree thin
By comparison and contrast, spelling errors of sound deviation and graphemic substitution can be avoided at this level.
(b) Collocation
Collocation of words is closely related to the usage of preposition. In English, many words need to collocate with prepositions and result in various phrases in vocabulary. Take word “consist” for instance, “consist” is an intransitive verb which cannot be followed by an object directly, and it needs to collocate with a preposition first. Generally, “consist” can collocate with three prepositions and result in different meanings. For example:
(1) consist of: be made up of
The class consists of 30 students.
(2) consist with: in accordance with
Theory should consist with practice.
(3) consist in: have as the chief or only element
Some young people think that happiness consists in eating and enjoying.
In the course of vocabulary teaching, through a certain word, we elicit its collocation of a preposition or several prepositions, to analyze its different meanings and usages and then learn how to use them appropriately. As a result, learners may have a general knowledge about the usage of this word and errors related with preposition collocations can be avoided.
(c) Derivation
Derivation is one of the main processes of word-formation. In teaching vocabulary, we need to tell students knowledge about word-formation, and morphological morphemes in particular. A certain word, by preceding or following some morphemes, can derive a series of derivatives. When teaching a new word, we should elicit its derivatives. Through this way, learners can enlarge their vocabulary. On the other hand, they can reduce errors of words with similar morphological forms.
(d) Semantic meaning
Mode of thinking and moral standards of various nations is different from each other, as they are deeply rooted in the soil of national culture. The same thing to different peoples shares different implications. Take “west wind” for example. In western culture, “west wind” refers to warm and pleasant wind from the Atlantic, while in China it stands for cold and disagreeable wind from the north. The meaning of “West wind” differs from one culture to another semantically, because of different geographical locations. In the process of English teaching, we should attach importance to the semantic meanings of these culture-bound words, thus cultivating students’ cross-cultural awareness.
Thirdly, emphasis should be given to reading and writing. Reading is the main source of the input of a language. Even ordinary people need to read in order to keep informed. Errors in the students’ writing also reflect another weakness, that is, they have not done much reading and writing. We say that Chinglish is the direct victim of Chinese interference. However, if learners read much and write often, it can be avoided to a great extent. In Chinese context, the best way for learners to increase the input of English language is reading. Through reading, learners can not only acquire the English way of thinking and language awareness but also enlarge their vocabulary. The input in classroom teaching is far from enough, and learners need to do more reading after class. In addition, writing is an indispensable language activity in language learning. The physical act of writing helps them sort out ideas, strengthen the knowledge they have obtained from reading and listening, and internalize what they have learnt. This point has been proved by experiences of many senior English experts from home and abroad, for example：
Scholar Wu Gu-hua from Qinghua University says: “extensive reading is the basic way to enlarge vocabulary----extensive reading of good, native English books and reading material is the basic way to improve oral English.”(English Learning, 2000,10)
Professor He Qi-shen from Beijing Foreign Language Study University points out: “English way of thinking is an ideal state many English learners want to attain. From my own English learning experience, I realize that wide reading is one of the most effective ways to achieve the goal.”(English Learning, 2000,4)
6 Conclusion
The research has obtained two main results after analyzing the data collected. First, based on the collection of the errors from 355 essays of the second- year students, the paper has identified 15 types of common errors and studied their frequency distributions. These 15 types of errors are mainly related to the uses of verb tenses, vocabulary, etc.. The quantitative analysis of the errors and the study of their frequency distributions show that the error frequency distributions are not distinctive. Therefore, students’ performances either as a whole or as different groups are not unusual. Second, through quantitative analysis of questionnaires, the paper has found that, in China, errors due to mother tongue interference amount to 21 per cent, developmental errors 50 per cent and other errors 29 per cent.
The research reveals that the students’ uses of verb tenses, plural forms of nouns, sentence structures, articles, prepositions and cognate words are liable to be affected by their mother tongue. The paper, therefore, puts forward suggestions that feasible measures be taken to deal with the problems in English teaching so as to improve students’ proficiency in English.
Finally, the limitations of this research should be pointed out. As a part of a series of researches, this research is only cross-sectional, students involved in this research are at the same stage of English learning, and the samples only display a partial picture of all the errors which students have made in their compositions. In addition, there is still a lack of a precise criterion for error classifications, which may result in the overlapping of some error categories. Therefore, the future research will try to solve these problems and attempt to analyze errors not only in students’ written work but also in their oral work in different stages of second language development.
Abbott, G. 1980. ‘Toward a more rigorous analysis of foreign language errors’. International Review of Applied Linguistics 18.
Bardovi-Harling, K. and T. Bofman. 1989. ‘Attainment of syntactic and morphological accuracy by advanced language learners’. Studies in Second Language Acquisition 11:17-34.
Bowerman, M. 1988. The ‘No negative evidence’ problem: how do children avoid constructing an overly general grammar? In Explaining Language Universals, ed. J. Hawkins.
Corder, S.P. 1967. ‘The significance of learners’ errors’. IRAL, VOL V/4.
Corder, S.P.1971. ‘Describing the language learner’s language’. CILT Reports and Papers, No. 6. CILT.
Corder, S.P. 1974. ‘Error analysis’ in Allen and Corder (eds.) 1974.
Dulay, H. and M. Burt. 1973. ‘Should we teach children syntax?’ Language Learning 23.
Dulay, H. and M. Burt. 1974. ‘You can’t learn without goofing’ in Richards(ed.) 1974.
Dulay, H., M. Burt and S. Krashen. 1982. Language Two. Oxford: Oxford University Press.
Ellis, R. 2000. Second Language Acquisition. Shanghai: Shanghai Foreign Language Educational Press
Ellis, R.2000. Understanding Second Language Acquisition. Shanghai: Shanghai Foreign Language Educational Press.
Ellis, R. 2001. The Study of Second Language Acquisition. Shanghai: Shanghai Foreign Language Educational Press.
Felix, S. 1980. ‘Interference, interlanguage and related issues’ in Felix (ed.) 1980a.
Felix, S. 1981. ‘The effect of formal instruction on second language acquisition’. Language Learning 31:87-112.e
Fllick, W. 1979. ‘A multiple component approach to research in second language acquisition’ in Andersen (ed.) 1979a.
Francis, N. 1999. ‘Bilingualism, writing, and metalinguistic awareness: oral-literate interactions between first and second language’. Applied Psycholinguistics 20: 533-561.
Hammarberg, B. 1974. ‘The insufficiency of error analysis’ in Svartvik (ed.) 1973a.
Jain, M.P. 1974. ‘Error Analysis: Source, cause and significance’ in Richards (ed.) 1974.
James, C. 2001. Errors in language learning and use: Exploring Error Analysis. Beijing: Foreign Language Teaching and Research Press.
Kim, M. 1997. ‘L1 and L2 educational levels, intelligence, and the second language writing performance of Korean ESL students’. Unpublished M.A. dissertation. Department of ESL, University of Hawaii at Manoa, Honolulu,USA.
Koda, K. 1999. ‘Development of L2 intraword orthographic sensitivity and decoding skills’. Modern Language Journal 83/1: 51-64.
Kubota, R. 1998. ‘An investigation of L1-L2 transfer in writing among Japanese university students: Implications for contrastive rhetoric’. Journal of Second Language Writing 7/1.
Milon, J.1974. ‘The development of negation in English by a second language learner’. TESOL Quarterly 8/2.
Nemser, W. 1971. ‘Approximative systems of foreign language learners’. International Review of Applied Linguistics 9:115-23
Pavesi, M. 1986. ‘Markedness, discoursal modes and relative clause formation in a formal and informal context’. Studies in Second Language Acquisition 8:38-55.
Ravem.R. 1968. ‘Language acquisition in a second Language environment’. IRAL, VOL. VI/2.
Richards, J. 2001. ‘A non-contrastive approach to error analysis’. In the context of Language Teaching. Beijing: Foreign Language Teaching and Researching Press.
Richards, J. 2001. ‘Error analysis, interlanguage, and second language acquistion: a review’. In the context of language teaching. Beijing: Foreign Language Teaching and Research Press.
Santos, T. 1987. ‘Markedness theory and error evaluation: an experimental study’. Applied Linguistics 8:207-18.
Sasaki, M. & K. Hirose. 1996. ‘Explanatory variables for EFL students’ expository writing’. Language Learning 46:137-174.
Scharhter, J. 1974. ‘An error in error analysis’. Language Learning 27:205-14
Selinker, L. 1972. ‘Interlanguage’. International Review of Applied Linguistics 10,3
Taylor, D. 1986. ‘Ethnicity and language: a social psychological perspective’ in Giles et al. (eds.) 1980
Wilkins, D.A. 1972. Linguistics in Language Teaching. Edward Arnold.
蔡金亭（2000），中国学生过度语中的作格动词—一项实证研究，《外语教学与研究》，第7期。
戴炜栋，束定芳（1994），对比分析，错误分析和中介语研究中的若干问题，《外国语》，第5期。
桂诗春（1985），《心理语言学》，上海外语教育出版社，上海。
胡壮麟（2001），《语言学教程》，北京大学出版社，北京。
廖莉芳，秦傲松（2000），专业英语教学现状调查报告，《外语界》，第3期。
刘绍龙（1998），英语中介语句法范畴变异的动态研究—一项对儿童和大学生的跟踪调查，《现代外语》，第2期。
刘绍忠（1997），国外语际语用学研究现状与我国语际语用学研究的思考，《现代外语》，第3期。
孙勉志（2001），《汉语环境与英语学习》，上海外语教育出版社，上海。
唐承贤（1997），差错分析综述，《外语教学与研究》，第2期。
王初明（1990），《应用心理语言学》，湖南教育出版社，长沙。
王立非（2002），外国语习得研究新进展，《国外外语教学》，第2期。
王奇民（2002），制约大学英语教学效果的因素及对策，《外语界》，第4期。
王宗炎（1980），分光镜下的汉语干扰英语实例，《外国语》，第6期。
王宗炎（1988），《英汉教学语言学词典》，湖南教育出版社，长沙。
文秋芳（2002），《应用语言学研究方法与论文写作》，外语教学与研究出版社，北京。
肖旭月（2001），语音表征在取词过程中的作用—中国学生英语拼写错误的心理语言学分析，《外语教学与研究》，第1期。
杨连瑞（1996），试论中介语理论与外语教学，《外语教学》，第4期。
章兼中，俞红珍（1998），《英语教育心理学》，警官教育出版社，北京。
周方珠（1997），《英汉翻译原理》，安徽大学出版社，合肥。
朱 跃 （1999），《英语与社会》，安徽大学出版社，合肥。