Language Transfer Evidence in the Acquisition of Double Object Construction by Chinese EFL Learners ()
1. Introduction
The construction the present study concerned is recognized as “double object construction”. It has long been a heated topic in linguistics, generally defined as a three-place argument pattern with a verb followed by two noun phrases.
There have been a number of previous studies on double object construction from different perspectives. Under the framework of structuralism, many domestic linguists and linguists abroad have long devoted their endeavors to the classification of the types of double object construction on a particular language [1] [2] [3], but their research seldom probed into the relation between Chinese and English double object constructions. To the divisions on a particular language, specific as they were, as Ma himself put it, the research only stopped at the descriptive phase, which was not beneficial to teaching and learning. What’s more, as Shi (2004) pointed out, their results did not coincide with each other [4]. Under the framework of generative grammar, many linguists examined the transformational issues between dative construction and double object construction, which had no verdict and consistency, either. However, construction grammar sheds more light on the explanation of double object construction. It is also avoidable to be plunged into the controversy haunted between double object construction and dative construction. In recent years, with more application in contrastive and empirical studies, construction grammar has proved itself as a promising field in linguistics.
According to the notion of lexically governed rules proposed by Lakoff [5], many alternations of construction seem to be sensitive to lexical items, particularly verbs. That is to say, the verbs which were involved in double object construction can largely determine whether a given alternation applies or not. Many linguists did classification on double object verbs or/and on the verbs in double object construction, but most of which were concerning single language. In 2004, Shi put forward his division based on the comparison of Chinese and English double object construction. In his paper, the verbs were divided into three parts, namely, left-direction, right-direction and left-right-direction. However, as He (2008) pointed out that some verbs do not express the meaning of direction [6].
As to empirical studies, however, both at home and abroad, rarely did we see the systematic empirical studies on double object construction between Chinese and English, with only a couple of studies being touched upon it. In 2000, Hunston and Francis classified English double object construction via corpus study [7]. In 2007, on the basis of corpus, Hu did contrastive research on double object construction [8]. And in the same year, Xu conducted an empirical study among college students on the verb-classification by Shi. Nonetheless, Hu’s study was not comprehensive as it only centered on the core meaning of the structure. Xu’s study which was based on the classification by Shi, as mentioned above, was not without limitation. In view of this, more theoretical and empirical research is needed to deeply probe into this construction [9].
2. Literature Review
2.1. Theory of Construction Grammar
Construction grammar originally derived from Case Grammar proposed by Fillmore [10] in the 1960s. Another theoretical background of construction grammar can be dated back to the 1970s as Gestalt Grammar (or generative semantics as it is called) proposed by linguist Lakoff in 1977. Goldberg has further developed the construction grammar and has probed most deeply into this field. Most of the later research is based on her study.
Lakoff proposed the notion of lexically governed rules and pointed out that many alternations of construction seem to be sensitive to lexical items, particularly verbs [5]. In other words, the verb involved largely determines whether a given alternation applies or not.
Goldberg (1995 & 2001) found that of all the words in a sentence, verbs are the ones that carry the most information about the syntax and the semantics of the sentence, however, the predictive value of verbs with respect to the overall meaning of the sentence may not be as strong as assumed by traditional linguistic theories [5] [11].
Meanwhile, Goldberg (1995) proposed that constructions are typically associated with a family of closely related senses rather than a single, fixed abstract sense [5]. In other words, now that constructions are treated as the basic data type as morphemes, they should have polysemous senses like morphemes. Therefore, since verbs largely determines whether a construction is acceptable or not, the conclusion can be drawn as: from the semantic meaning of verbs, we can readily discuss the polysemy of a particular construction.
2.2. Theory of Language Transfer
Language transfer has been one of the major topics in the study of second language acquisition (SLA). The concept of Language Transfer was first proposed by Lado in one of his works Linguistics across Cultures in 1957 [12]. He held the view that, under the environment of second language acquisition, learners broadly depend on what they have mastered and often put the language forms and meanings to their second language learning. Weinreich (1953) used the term ‘interference’ for ‘instances of language of deviation from the norms of either language which in the speech of bilinguals as a result of their familiarity with more than one language’ [13].
Ellis described language transfer in such a way: in the case of similarities between the L1 and L2, it functioned positively, while in the case of differences, it functioned negatively [14]. And transfer will be positive when the first and second language habits are the same. In this case no errors will occur. Thus differences between the first and second language create learning difficulty which results in errors, while the similarities between the first and second language facilitate rapid and easy learning. Odlin defined linguistic transfer as: “transfer is the influence resulting from similarities and differences between the target language and any other language that has been previously (and perhaps imperfectly) acquired [15] ”. This is a widely accepted definition of “language transfer”. As a result, the definition of “transfer” has been considerably broadened in the field of contrastive linguistic studies and a neutral term—cross linguistic influence is adopted.
Inthegreat book Language Transfer, Odlinoffered “a classification of outcomes”: positive transfer, negative transfer (underproduction, overproduction, production errors, and misinterpretation) [15].
2.3. Classification of Double Object Verbs
Many domestic linguists, as well as linguists abroad, have made great efforts to confine the range of verbs, which could enter into double object construction or classify them.
Li divided the verbs from a semantic scope into eleven categories [1]. According to Pinker [16], English double object verbs can be divided into ten categories from semantic perspective, comprising verbs that inherently signify acts of “giving”, of “instantaneous causation of ballistic motion”, of “sending”, of “continuous causation in a deictically specified direction”, of “future having”, of “communication”, of “instrument of communication”, of “creation”, of “refusal” and of “obtaining”, etc. Lu (2002) made a specific investigation on those verbs without the meaning of ‘giving’ in Chinese double object construction and found that all together 104 verbs existed in Chinese double object construction [17]. Meanwhile, he did an empirical study from the syntactic perspective and attested that Chinese double object construction had the broad meaning of “gaining”. Shi (2004) put Chinese double object construction into three kinds, those verbs with the direction of left, those with the direction of right and those with the direction of left-and-right [4]. Based on this classification, he compared the Chinese double object construction with English and concluded that English double object construction only had the verbs with the direction of right. Ma divided double object construction into fourteen types: “the ‘giving’ type, the ‘getting’ type, the ‘quasi getting’ type, the ‘naming’ type, the ‘resulting’ type, the ‘cause’ type, the ‘occasion’ type, the ‘exchanging’ type, the ‘causative’ type, the ‘place’ type, the ‘measurement’ type, the ‘momentum’ type, the ‘time quantum’ type and the ‘IO’ being as the empty object [2] ”. Zhu (2007) put the verbs in DOO structure into two types: the type of “gaining” and the type of “fading” [18]. However, these classifications merely function as references to check whether the verb can add two objects but with no good for the study of double object construction since some double object verbs could not enter into this construction.
Goldberg classified the verbs which can enter into English double object construction into the following categories: “verbs of ‘giving’, verbs of ‘creation’, verbs of ‘obtaining’, verbs of ‘obligation’, verbs of ‘future having’, verbs of ‘permission’, verbs of ‘refusal’, etc [5] ”. Zhang & Wen put those verbs which can enter into English double object construction into five categories from syntactic perspective [19]. Hunston and Francis made verb-classification of English double object construction into six groups which were based on corpus study, namely, verbs concerned with giving someone something, or refusing to do so; verbs concerned with doing something for someone; verbs concerned with talking, writing or otherwise communicating something to someone; verbs concerned with giving someone a benefit or a disadvantage; verbs concerned with feeling and attitudes and verbs of other meanings [7]. Xu (2001) re-classified Goldberg’s classification mentioned above into three types [20]. They were those verbs with the meaning of “giving”, those verbs with no salient but latent meaning of “giving” and those verbs without the meaning of “giving”. Besides, he referred that the action of “giving” sometimes may favor receiver and sometimes may not; hence, he further divided those verbs with the meaning of “giving” into (positive) giving, (zero) giving and (negative) giving. In 2007, he further divided modern English double object construction into two genres. One was those sentences with the realization of giving and the other was those without the apparent action of it.
From the foregoing review, it is clear that many linguists did classify the verbs which can enter into Chinese and/or English double object construction; however, these classifications either lean to one type of language or seem to be too broad and scattered which are not beneficial for the teaching and for learners to master. Based on current literature reading, only Xu in 2007 did an empirical study on English and Chinese double object construction. In his paper, he adopted the classification of verbs by Shi (2004) [21]. Whereas, according to He (2008), this classification was not without shortages, since some verbs do not connote the meaning of direction and successful transfer [6].
2.4. Cross-Linguistic Similarities and Differences on Double Object Construction
As mentioned before, the verbs that were involved can largely determine whether a given alternation applies or not. From the philosophical perspective, the objective world is full of motional substances. The relation between motion and substance is of great importance in the objective world. Everything must undergo three phases: creating, developing and fading.
In respect to this, this paper adopts comparative analysis, attempting to analyze the double object construction from the three foregoing aspects, namely, “creation” (Substances come into being through creating), “giving” (In the long human history, with the realization of exchanging of goods, substances get development. And giving goods among human beings ensures the process of exchanging, so this aspect can also be called the type of “developing”) and “fading” (Substances end in disappearance). This classification is far from complete and by no means exhaustive, but it presents us a clear-cut analysis of double object construction between English and Chinese and is believed to be of great help in enhancing understanding. Here, the present study categorized the three aspects, as shown in Table 1.
Goldberg noted that the core meaning of double object construction is “giving” and from Table 1, it could be clearly seen that this core type exists in the Chinese and English double object construction, whereas the peripheral types in these two languages are in stark difference. In Chinese, verbs denoting the meaning of ‘creating’ could not enter into double object construction while English could; verbs denoting the meaning of fading could enter into the double object construction while English could not. For instance, we cannot say “我烤了一个面包给他” in Chinese, but we can say “I baked him a cake”. We cannot say‘I drank him a cup of tea’ in English but we can say “我喝了他一杯茶”.
![]()
Table 1. Similarities and differences on double object construction.
○ refers existence; × refers non-existence.
2.5. Objectives of Study
The present study examines the similarities and differences between English and Chinese double object construction via a more systematic and clear-cut analysis. In order to explore how Chinese EFL learners go about acquiring this construction and examine the implications this comparative analysis may have on Chinese EFL learners in their acquisition of English double object construction, an empirical study is implemented thereafter, aiming to answer the following research questions: 1) Are there any properties of double object construction which pose no problem for Chinese EFL learners, i.e., where learners never or seldom make errors and, if so, to what extent can those be predicted by language transfer? 2) In which type do Chinese EFL learners have the highest error rate?
3. Experimental Design
The detailed experimental design in attempting to answer the foregoing proposed questions is as follows.
3.1. Selecting Qualified Participants
Altogether 120 students were recruited as participants at random, comprising 40second-year university students as L2 learners of high proficiency, 40 2nd graders from Senior Middle School as L2 learners of intermediate proficiency and 40 2nd graders from Junior Middle School as L2 learners of low proficiency. To ensure the three groups of Chinese EFL learners representing different levels of English proficiency, the means and the standard deviation of the proficiency test for each group were first calculated. At the same time, one-way ANOVA was further performed to find out whether the observed differences in group means were statistically significant. The result was SS = 43115.304, MS = 21557.652, F = 316.662, P < 0.001. It could be seen that the main effect was significant, suggesting the three groups were indeed different in English proficiency.
3.2. Design of Empirical Study
The primary instrument of the present study was a closed-book exam consisting of 30 questions, with 15 Chinese and 15 English sentences, among which, three subclasses of the construction mentioned above were equally distributed in the test paper, five sentences for the “giving” type, the “creation” type and the ‘fading’ type respectively. In each subclass, five verbs were selected.
Considering that participants at different levels have the different command of English vocabulary and in view that the vocabulary limitation may, to some extent, affect the performance of participants, the questions presented in the test were under the pattern of a pair of co-appearance of Chinese sentence and its counterpart, as the Chinese sentence “玛丽教比尔法语” and the English sentence ‘Mary taught Bill French’. Under this pattern, any encountered with unfamiliar words could use the corresponding Chinese sentences as reference. See more details of the sample paper in Appendix 1.
The closed-book proofreading test papers were administered in three different groups, with a time limit of 30 minutes. The participants were required to judge the acceptability of these sentences by putting a tick/cross by the questions. To guarantee that participants knew exactly and understand readily what they are supposed to do in the test, the test instruction was given in the form of Chinese, be it literally or verbally. Each sentence a participant marked received point if and only if it was correct and no score would be given to a sentence in such situations as misjudging or no judging. The responses of participants were scored on a 2-point scale: 1 and 0. All of the data were processed by the software for statistics—SPSS 16.0.
4. Data Analyses and Discussion
As mentioned by Abu-Rabia and Wattad, analysis of Variance is an effective approach to prove the occurrence of the language transfer phenomenon, and the mean and standard deviations are the most widely utilized statistical tools used to determine the central tendency and variability of interval data [22]. Thus, the present study presents the means, together with standard deviation, obtained by the low, intermediate and advanced groups for judging the grammaticality of the sentences containing the type of “giving”, “creation” and “fading” in Table 2 and Table 3 .
It can be seen that the participants of different proficiency levels of English scored differently on three subclasses of double object construction and the performance on the core subclass was the best at any level. To further find out whether the observed differences in means were statistically significant, 3 one-way ANOVAs (repeated measures) were conducted respectively on three different levels.
The results of the participants in English and Chinese double object construction were as follows: for those at lower English proficiency: F = 46.977 (P < 0.001) in English and F = 15.994 (P < 0.001) in Chinese; for those at intermediate English proficiency: F = 84.469 (P < 0.001) in English and F = 16.431 (P < 0.001) in Chinese; and for those at advanced English proficiency: F = 36.749 (P < 0.001) in English and F = 12.265 (P < 0.001) in Chinese. These show that the means of different levels of proficiency with the core subclass of verbs were higher than and significantly different from those with the verbs of “creation” and the verbs of “fading”. And the means of advanced group with the core subclass of verbs were higher than and significantly different from those by low and intermediate groups. The findings mentioned above fairly well established that at different levels of English proficiency, Chinese EFL learners acquire the core
![]()
Table 2. Means and standard deviations of English double object construction of three subclasses by low, intermediate and advanced groups.
Note: SD = Standard deviation.
![]()
Table 3. Means and standard deviations of Chinese double object construction of three subclasses by low, intermediate and advanced groups.
Note: SD = Standard deviation.
subclass of English and Chinese double object construction better than peripheral ones and with the development of English proficiency levels, the advanced groupout performed the low and intermediate counterparts in the core subclass of English double object construction.
Also, three Paired-sample t-tests were applied as can be seen in Table 4 and Table 5. The results in Table 4 and Table 5 show that the mean scores for the type of “giving” and the other two types were significantly different, not only for L2 learner at low and intermediate levels, but also for those advanced learners, suggesting that when acquiring English double object construction, Chinese EFL learners, especially those at low and intermediate proficiency levels, experienced greater difficulty with the types of “creation” and “fading”.
5. Findings of Present Study
The findings shown in Table 2 and Table 3 were in consistency with the conclusion made by Hu on the basis of corpus [8], which suggested that most of the English double object construction produced by Chinese EFL learners were mainly restricted to several typical verbs; and also, they gave a powerful support to positive transfer by L1, which held the belief that similarities between the first and second language could facilitate rapid and easy learning; in other words, the facilitative effect of the L1 could be adduced by these findings [14] [23]. It is believed that in the second language environment, adequate second language input and the improvement of linguistic levels have definite and evident advantages on sentence comprehension and learning. From the data we collected and results we verified, it was clear to us that the high frequency of exposure to second
![]()
Table 4. Results of paired-samples t-tests for types of “giving” and “creating”.
![]()
Table 5. Results of paired-samples t-tests for types of “giving” and “fading”.
language and improvements of command of English resulted in the better performance of the advanced group. Hereby, it proved that language acquisition is a creative process in which learners are interacting with their environment to produce an internalized representation of the regularities they discover in the linguistic data to which they are exposed [24].
Statistical results shown in Table 4 and Table 5 confirm that these L2 learners did not syntactically differentiate English and Chinese double object construction on these two types. They used their previous mother-tongue experience as a means of organizing the second language data. And since learners of low levels of proficiency had less previous second language knowledge to draw on in making hypotheses about grammatical rules, and might therefore be expected to make more use of their first language knowledge.
Moreover, these findings gave a powerful support to the Theory of Transfer—negative transfer, which proposes that differences between first and second language created learning difficulty which resulted in errors [14]. It could be argued here that the errors can mainly be “overgeneralization and simplification” or “overuse and avoidance”. Needless to say, overgeneralization and simplification were at work. “He ate his little sister three mangos”, for example, could be a case of overgeneralization or overuse from the acceptable pattern of the first language in “他吃了他妹妹三个芒果”. And the wrong judgment on the sentence “Joe baked her boyfriend a cake”, for example, could be a case of simplification or avoidance since in modern Chinese double object construction, the sentence with the verbs of ‘creation’ is unacceptable.
6. Limitations
The present study tries to explore the acquisition of English double object construction by Chinese EFL learners. However, this study is subject to a number of limitations.
Due to limit time and the availability of resources, the present study is far from being a complete and perfect account for double object construction. It only focuses its attention on verbs; however, since the arguments in double object construction also play significant roles in its formation, the present research lacks further study in investigating and explaining the specific requirements and restrictions of its arguments. Also, the present study adopts a cross-sectional approach for investigating the acquisition of English double object construction. It is suggested that further research should adopt the longitudinal approach instead in pursuit of more reliable and valid findings.
In a nutshell, great efforts are needed to do more related work both theoretical and empirical.
7. Conclusion
This study is a tentative probe into the acquisition of English double object construction. The results of present study confirm that Chinese EFL learners at different levels of proficiency acquire the core subclass of double object construction (both Chinese and English) better than peripheral ones, which can be a powerful support to the proposal of Positive Transfer. The results also manifest Chinese EFL learners, with low and intermediate English proficiency in particular, face greater difficulty in acquiring the “creation” and “fading” type of double object construction. Still, the present study leaves much to be desired. It is hoped that this paper can devote its pygmy effort to and provide some suggestive implications and reference significance for Chinese English teaching and learning.
Acknowledgement
First, I would like to express hearty thanks to the editor of Open Journal of Applied Sciences who has given me enlightening suggestions and insightful comments. My special gratitude also goes to those kind-hearted students for their sincere help in the process of experimentation.
Appendix 1
Research Instruments
There are 30 English and Chinese sentences in this closed-book questionnaire. Some of them can be expressed with double object construction, while others cannot but with dative construction.
Note the following differences.
For example:
Double object construction:
姐姐给我一个本子。(√)
My elder sister gave me a notebook. (√)
Dative construction:
姐姐把一个本子给我。(√)
My elder sister gave a notebook to me. (√)
Read carefully and answer the following questions by putting “√” after the correct sentence and “×” after the wrong one. The completion time is 30 minutes.
1) 丽丽喝了我一瓶橘子汁。( )
2) Lily drank me a bottle of orange juice. ( )
3) 我写了爸爸一张纸条。( )
4) I wrote daddy a piece of message. ( )
5) 妈妈送了我一份生日礼物。( )
6) Mom sent me a birthday present. ( )
7) 我不小心打碎了大卫两个个杯子。( )
8) I accidentally broke David two cups. ( )
9) 多利煮了安娜一顿丰盛的晚餐。( )
10) Dolly cooked Anna a rich dinner. ( )
11) 布什烧了亚当两份文件。( )
12) Bush burnt Adam two files. ( )
13) 我问了老师一个很愚蠢的问题。( )
14) I asked the teacher a silly question. ( )
15) 小王建造了他自己一间小屋。( )
16) Mr. Wang built himself a hut. ( )
17) 奶奶告诉我们一个远古的童话。( )
18) Grandma told us an ancient fairy tale. ( )
19) 乔烤了她男朋友一个蛋糕。( )
20) Joe baked her boyfriend a cake. ( )
21) 他吃了他妹妹三个芒果。( )
22) He ate his little sister three mangos. ( )
23) 她妈妈织了她一双美丽的手套。( )
24) Her mother knitted her a pair of beautiful gloves. ( )
25) 那淘气的小家伙杀了他邻居家一只鸡。( )
26) The naughty kid killed his neighbor a chicken. ( )
27) 卡尔递给了乔一条项链。( )
28) Carl handed Joe a necklace. ( )
29) 比尔教我英文。( )
30) Bill taught me English. ( )