Abstract
This study investigates differential effects of etymological elaboration and rote memorization on idiom acquisition and retention in Chinese college EFL learners.The sample (N = 70) involved two intact groups of college students. Subjects in one group received instruction in the form of etymological familiarity while subjects in the comparison group were asked to memorize idioms on the basis of their given meaning. Subjects were then administered two post-tests: one to assess initial idiom acquisition, and another four weeks later to assess idiom retention. Statistical analyses produced sufficient support for both the first hypothesis (that the etymological familiarity condition would acquire more idioms than the rote memorization condition), and the second hypothesis--that the etymological familiarity would exhibit superior retention rates.The finding of this study suggests that for the purpose of encouraging long-term retention, elaborating on the original usage of idioms is preferable to asking students to learn idioms by rote.

Keywords:etymology, rote memorization, idiom acquisition and retention, Chinese college EFL learners

Introduction
Idiomaticity has recently become a respectable area of study in psycholinguistics, linguistics, developmental psychology, and neuropsychology (Cacciari and Tabossi, 1993). This interest in idiomaticity is well founded, given that American or British English, for example, contains many thousands of formulaic phrases and expressions that the ordinary speaker must somehow learn (as is evident in the many idiom and slang dictionaries currently available). Since language production relies heavily on an ability to string together prefabricated, memorized multi-word expressions, that is, whole phrases which are stored in memory, and retrieved from it, as unbroken-up chunks, people are not considered competent speakers of the target language until they master the various clichéd, idiomatic expressions that are ubiquitous in everyday discourse (Bobrow and Bell, 1973; Boers et al, 2004).

Since building and maintaining a large idiomatic expression stock is such an essential part of achieving proficiency in a language, especially for foreign and second language learners, many researchers in recent years have turned their attention to idiom studies, a domain that seems to have become something of a cottage industry within the field of SLA and TEFL. Associating an idiom with its etymology in particular has become an especially popular subject, as demonstrated by the recent appearance of many studies devoted to the topic (e.g., Gibbs, 1997; Boers, 2001, Boers et al, 2004; Grant, 2004; Grand et al, 2004; Kövecses and Szabó, 1996).Although an ample body of evidence now exists confirming the possibility and reality of etymological elaboration as an insightful technique for learning idiomatic expressions (Boers, 2001; Boers et al, 2004), the pertinent question for theoretical researchers to ask now is not whether through etymological elaboration, idiom acquisition occurs, but the extent to which it occurs and the variables that promote its occurrence.

Of greater relevance to language educators is the practical question of how methods designed to stimulate idiom acquisition fare in comparison to more direct approaches such as semantic mapping (Gibbs, 1997) and rote learning(Lewis, 1997). Although research over the past few years has validated etymological familiarity as an insightful method, little evidence has been adduced demonstrating its merits relative to other forms of instruction. Data of this nature should prove especially useful in illuminating the debate between those on one end of the spectrum who encourage a greater, sometimes almost exclusive, reliance on relatively holistic pathways to promote idiom acquisition (e.g., Boers, 2001; Boers et al, 2004) and others who call for more direct instruction of idiomatic expressions (e.g., Bobrow and Bell, 1973; Lewis, 1997).

The present study was undertaken to furnish data for this debate via a quasi-experimental design and to measure the differential effects of two methods on idiom acquisition: (1) etymological elaboration and (2) rote memorization. More specifically, the objective of this study was to determine the extent to which etymological elaboration, in contrast to the rote memorization, facilitates idiom acquisition and retention in Chinese College EFL Learners. This paper firstly presents (1) views on the nature of idioms on which the two different methods were based, (2) methodology and design of the study, and then the result and conclusion is given at the end of the article.

Rationale for different instruction of idioms
Idioms have been the subject of investigation by linguists and psychologists for a number of years. Both groups have been concerned largely with the representation and status of idioms in the lexicon, that is, with the question as to whether idioms are stored, accessed, and subject to grammatical rules in the same way as single lexical items (Bobrow and Bell, 1973; Lewis, 1997). Whereas the experiments of psychologists by and large have investigated how speakers process non-literal language, linguists generally have tried to determine the status of idioms in the lexicon by comparing their syntactic behavior with that of single-word lexical items. The central question is whether or not idioms are decomposable into individual chunks or constituents that are semantically non-vacuous(Gibbs, 1983, 1991, 1992; Boers, 2000).
Since one of the traditional criteria for classifying an expression as idiomatic has been its non-compositional nature (Fernando and Flavell, 1981), many linguists claim that the lexical make-up of idiomatic expressions is completely arbitrary (e.g., Hatch & Brown, 1995).If an expression is said to be non-compositional, it is believed that its meaning cannot be inferred by simply adding up the semantics of its constituents. As a result, idioms appear to be quite arbitrary.

This claim is equivalent to saying there is no possibility of learning idioms in any insightful way and that, instead, learners must persevere through a long and painstaking process of blind memorization (see Lewis, 1997; Boers et al. 2004).Because of this alleged arbitrary nature of idioms, it has long been taken for granted in second or foreign language teaching that the only way learners could learn such expressions was believed to be rote memorization because of the absence of reliable clues inside the expressions themselves (Cooper,1999).

Whereas traditional linguists view idioms as arbitrary units, researchers in cognitive sciences argue that idioms are partly compositional and analyzable(see Hatch & Brown, 1995; Gibbs, 1997). The reason idioms were labeled arbitrary might be that most traditional language scholars tended to examine only a few of these conventional phrases, such as by and large.Yet a closer look at idiomaticity, one that examines idioms more broadly and seeks greater generalizations across different idiomatic phrases, reveals that idioms are not frozen, arbitrary, but indeed partly motivated.This means that, while it is true that the idiomatic meaning is not fully predictable on the basis of a literal reading, the derivation from that literal sense can nonetheless be‘explained’. A general consensus among researchers now is that many idioms can be ‘motivated’along various lines (Kövecses and Szabó, 1995; Boers, 2001; Boerset al, 2004; Gibbs, 1997; Lakoff, 1987, 1993).Many idioms appear to be motivated by conceptual metaphors or metonymies,for example, hit the ceiling,get hot under the collar,flip your lid motivated by the anger is heated fluid in a container metaphor, lend a (helping) hand by a metonymy where the hand stands for the action.

Of special interest to linguistic researchers, however, is a class of figurative idioms that are derived from associations in rather specific experiential domains, and that can be motivated by their literal usage in those original contexts(e.g., Boers, 2001; Boers et al, 2004). The idioms,for example, a shot in the arm, keep one’s finger on the pulse of something instantiate metaphors or metonymies, but they are obviously motivated by their original usage in a concrete scene, that is, a health and medicine scene (i.e. doctor giving a patient an injection of drug or medicine to stimulate her/him, doctor putting his fingers on the pulse of a patient to diagnose). Although the idiom, a race against time,reflects the time is a moving object metaphor, it is also motivated by the imagery of a concrete scene(i.e. a racing contest).A safe pair of hands illustrates the hand for the action metonymy, but adds sporting imagery (at least for those who are aware that this expression is derived from ball games, especially cricket).
The central question for language educators to ask now is not whether idiom acquisition can occur through etymological familiarity, but the extent to which it occurs and the variables that promote its occurrence. In addition, more researches should be conducted in various settings with different samples to provide evidence in support of its merits relative to other forms of instruction.In accordance with prior research findings, the following hypotheses were posed:

Hypothesis 1: Subjects in the experimental group will acquire more idiomatic expressions through receiving instruction in the form of etymological elaboration than subjects in the comparison group who intentionally attempt to learn the same idiomatic expressions via rote memorization.

Hypothesis 2: Subjects who acquire idiomatic expressions through etymological elaboration will exhibit higher retention rates than subjects who learn the same material through rote memorization.

   The inspiration for this study and the fundamentals of its design are rooted in the study conducted by Boers, Demecheleer and Eyckmans (2004). Their study involved having one group who was told the literal, original usage of the target idioms while an equivalent group from the same population served as a comparison one who was NOT told the etymological information of the same material. The experimental group was informed, for example, that the idiomatic usage of The Home Secretary is on the ropes is derived from its literal usage in the context of boxing matches. A pretest and post-test were administered to measure the effects of the treatment. Scores on a between-groups revealed that the subjects who had been familiar with the etymological information acquired significantly more idioms than the comparative subjects.

Though building on the basic design of the Boerset al. (2004) experiment, the present study differs from it in several principal ways. First, the independent variable investigated in Boerset al. was etymological familiarity alone and the value of other methods was not measured at all. Skeptics may argue that the same results would be achieved through other direct methods. In this study, however,subjects did implement two strategies: etymological elaboration and rote memorization.
Perhaps the most salient way in which this study differs from its predecessor is with regard to the dependent variable involved. In Boerset al. the dependent variable was the number of the target idioms the subjects acquired after the treatment. The present study, however, was conducted to determine not only idiom acquisition [2]defined as the median number of idioms gained by a group or an individual subject from pretest to post-test 1, but also idiom retention defined operationally as the median number of idioms gained from pretest to post-test 2.

Method

Design
Since it was impossible for the subjects to be randomly selected, this study consists of a two-group, pre/post-test quasi-experimental design, involving two intact classes chosen to serve as the experimental and comparison groups.The duration of instruction in this study was four weeks. Each week includes three class sessions. Each class session consists of two periods (40 minutes each). To minimize the chances of inter-subject communication between groups, Tuesday/Thursday classes were chosen to function as the experimental group (n = 35), and Monday/Wednesday sections were selected to function as the comparison(n = 35).Instruction in both groups was incorporated as part of the regular course curriculum for the respective sections.Only data from those subjects who had taken all the three idiom tests were used in this study. The final n-sizes of both groups were effectuated by excluding all dropped-outs.

Subjects
The subjects were 70 newly enrolled college level Chinese EFL learners studying in the English Department of a university located in Shandong province, China. Of the students who participated, 47 were male and 23 were female, with an average age of 19.5. In terms of homogeneity, all the subjects have the same cultural and language background, with Chinese as their first language, on the one hand;they had learned the target language from the same standard textbooks approved by the State Education Commission for about six years, taken the same entrance examination, and recruited into the university with scores above the required level, on the other hand. The similar background and English proficiency that the subjects possess makes the two groups as equivalent as possible.

Table 1
The difference between the two groups on the pretest in idiom proficiency

Group	N	Mean	SD	df	F	Sig.
Exp	35	28.77	7.60451	1	.631	.430
Con	35	30.11	6.49783	1

This kind of similarity, however, in no way rule out the possibility of heterogeneity in idiom proficiency between the two groups. Therefore, a pretest was administered to both groups to serve as a baseline and to control for inter-group differences in idiom proficiency prior to administration of the treatment. A one-way analysis of variance of means failed to reveal a significant difference (P > .05) (see Table 1 above). Because the participants were not randomly sampled, no claim is here made that the two intact classes chosen constituted a representative sample of all college foreign language learners.

Materials
The material used to prepare subjects in the two groups for instruction was 50idiomspreselected by the researcher and his one colleague from the Oxford Dictionary of Idioms (Speake, 1999), the Dictionary of Idioms and Their Origins(Flavell and Flavell, 2000), and the textbook New Horizon: College English (Zheng, 2001) designated by the school authority for the students. Only idioms that the researcher and his colleague believed would be unfamiliar to most subjects were chosen.The process of idiom selection was subjective in that the researcher relied on previous experience teaching the target population rather than a frequency list to determine the appropriateness of target idioms. Although a few of the idioms selected were anachronistic, the majority of the idioms were relatively contemporary.As a source of etymological information for the selected idioms the researcher used the Dictionary edited by Speake (1999), which actually gives explanations as to the origins of the expressions.

The Dictionary edited by Flavell and Flavell (2000) served as an additional source of etymological information.

The test consisted of two components: a 10-item gap-fill section (worth 60 points) and a 10-item multiple-choice section (worth 40 points). The design of these components was to produce a well-rounded and versatile instrument that would accommodate subjects’ different cognitive styles. The first section is a gap-fill task where the learner was presented with a meaningful context from which the keyword of the idiom was missing.This task (essentially a modified cloze test) was designed specifically to elicit the key part of the idiom with the purpose of testing the learner’s recollection. In the multiple-choice section, there were 10 sentences, in each of which an idiom was embedded. One correct answer (i.e. the correct paraphrase of the target idiom) and four distracters were supplied for each sentence. The students were required to tick the correct literal paraphrase of the target idiom from the five possible choices, which were labeled were A through E, with choice E consistently reading “none of the above.”A random order of the answer choices was computer-generated to avoid possible confounding effect that might occur due to an organized distribution pattern.

The 50 idiom selected for the two groups (see Appendix) contained all 34 items that appeared on the two post-tests (30% of idioms tested in the first test was again used in the second test), plus an additional 14 idioms that the researcher and his colleague felt would benefit the subjects in their academic careers.The assignment of the total score for the two sections was determined on the different difficulty of answers rather than theoretical grounds. According to the previous teaching experience, the difficulty degree of gap-fill task appears to be higher than that of the multiple-choice. Before the implementation of testing, the tests constructed by the researcher were submitted to an experienced professor for his suggestions on which some modifications were made to enhance the reliability of the tests.

Procedures of instruction
After completing the pretest,subjects were instructed to begin their designated assignments. For the comparison group whose instructor was the researcher’s colleague, the assignment was to memorize the 50 target idioms, each of which the idiomatic meaning was given in Chinese, along with a sentence made with it in English. Subjects were not informed of the original usage of the target idioms. Upon distribution of the idioms, subjects were informed that they would be tested on some of the idioms at the end of the semester. Subjects were not instructed to use any particular technique (e.g., the keyword method, semantic mapping) but were merely told to commit the idioms to memory.

Concomitantly, the experimental group whose instructor was the researcher began to receive instruction about the literal, original usage of the same idioms given to the comparison group. For example, with respect to the original usage of the idiom, somebody is on the rack, the students would first be told that “The ‘rack’ was an instrument of torture, used in the past for punishing and hurting people, that is, their arms and legs were tied to a wooden frame and then pulled in opposite directions, stretching the body.” And then they would be asked to interpret the idiom by combining its etymology and a meaningful context such as “After three weeks had passed and she had still not heard from her daughter, Joan was on the rack.”In order to enhance the learners’ awareness of the literal usage of idioms, sometimes they would be required to hypothesize the original usage of the target idioms. Subjects in this group were instructed to focus on the etymological information and informed they would be tested over the original usage of the idioms at the end of the semester, and at no time were subjects told to expect a test on the idioms themselves.

Data collection and analysis
After the instruction for about four weeks, the first post-test was administered to both groups simultaneously to assess initial acquisition in their own classrooms respectively. The time limit was 50 minutes, the time of one period of a class session.To avoid biasing subjects, information concerning the nature and purpose of the study was not supplied. Subjects in both groups were told merely that they would receive the results of the test within several weeks. Approximately four weeks from the initial post-test, both groups were tested again to assess idiom retention rates over an extended period.

The time interval between the two post-tests was determined on practical rather than theoretical grounds. After subjects completed the second post-test, subjects in both groups were notified as to the purpose of the study and the function of the tests. Participants were assured that their identity would remain anonymous.

After subjects completed the second post-test, all data pertaining to the experiment were collected, scored, and analyzed. In the process of scoring, the testing sheets for the two groups were mixed up and evaluated by the researcher and his colleague independently at first, and then, if the two scores were given by the two raters without greater discrepancies, the participant’s final mark would be the mean of the two marks; if, however, the two scores were very different, the two raters would determine the final score in consultation. The correlation between the two raters in scoring the subjects was measured by the correlation coefficients. The internal consistency of the raters was quite high, since the Pearson Correlation coefficient reached .972 point with significant correlations (p< 0.01).

After data collecting, statistical analyses were performed using SPSS (Statistical Package for Social Sciences) for Windows, version 13.0 at the Computer Center of the university.To test Hypothesis 1 and 2, descriptive statistics, the mean, standard deviation, range and median, were computed for each group across all three administrations of the idiom test to examine the variability of the scores. A one-way analysis of variance test (ANOVA) [3] was then run, both within and between groups, to determine (a) the extent to which group performance changed over time and (b) the degree to which either group differed relative to the other at each administration of the test.

Results and discussions
Descriptive statistics across all three administrations of the idiom test for the experimental group (EG) and the comparison group (CG)were computed and presented in Table 2.

Table 2: Descriptive Statistics

		Administration			Gains & Retention
Group		Pretest	Test 1	Test 2	Pretest to Test 1	Test 1 to Test 2	Pretest to Test 2
Exp.	Mean	28.771	82.943	80.886	54.172	-2.057	52.115
	SD	7.605	4.964	5.027	5.903	6.058	8.280
	Median	29	84	80	54	-4	53
	Range	28	20	22	21	22	38
Con.	Mean	30.114	72.886	66.000	42.772	-6.886	35.886
	SD	6.498	7.071	8.42	7.013	6.876	9.308
	Median	30	73	66	41	-5	35
	Range	21	24	34	32	31	46

Table 2 reveals a marked difference in performance on both idiom acquisition and retention post-tests between the two groups.

With regard to the idiom acquisition from pretest to post-test 1, the mean of the EGis54.17, and that of the CG is 42.77. The median of the EG is 54, while that of the CG is 41. Both the mean and median demonstrate that the subjects of the EG acquired more idioms than those of the CG due to the treatment given to the former group. Moreover, the range of the scores for the EG is 21, whereas that of the CG is 32.The standard deviation for the EG (5.90 points) is quite smaller than that of the CG group (7.01 points). The comparison of the range and standard deviation of the two groups indicates that there was more variation among the subjects’ scores of the CG from pretest to post-test 1 as compared with the EG group.
In addition, through a one-way ANOVA(see Table 3 below), we can see that the difference between the means of scores of the two groups on idiom acquisition is statistically significant (p< .05, df = 1, F = 56.88). The results from the idiom acquisition tests given to the two groups provided substantial evidence in support of the first hypothesis of this present study.

Table 3
The difference between groups and retention within groups

see pdf

Concerning the difference between the two groups on the test for idiom retention from pretest to post-test 2, the data in Table 2 show that the mean of the EG is 52.115 and that of the CG is 35.886, with the median of the former group 53 and that of the latter one 35. The range of the EG is 38, whereas that of the CG is 46. The standard deviation for the EG (8.28) is smaller than that of the CG group (9.31). Considering the results from the descriptive statistics, one can claim that the subjects of the EG group acquired more idioms eight weeks later from pretest to post-test 2 and performed more homogeneously than those of the CG on the test for idiom retention. Based on the statistical data from one-way ANOVA (see Table 3 above), it is clear that the difference between the two groups on idiom retention test is statistically significant (p < .05, df = 1, F= 57.323).

From the comparison within groups, we can see that the mean difference between idiom acquisition (the mean from pretest to post-test 1) and idiom retention(the mean from pretest to post-test 2) was vastly greater for the CG group, which actually experienced a 6.9%decline in mean performance. One-way ANOVA reveals the decline is significant at p = .001 (df = 1, F = 12.219). By contrast, the EG group experienced only a 2.057% decline without significant difference (p> .05, df = 1, F = 2.339). In support of Hypothesis 2, the subjects of the EG exhibited significantly higher idiom retention rates than those of the CG group.

Putting all obtained data together, the results of this study reveal that the performance of students in the experimental group was significantly better than that of students in the comparison group on both idiom acquisition and retention tests. This superior performance is very apparent in Figure 1 presenting the group trends. The line for idiom acquisition rises rapidly for the experimental group, but slowly for the comparison one, on the one hand; the idiom retention falls slowly for the experimental group, but rapidly for the comparison group, on the other hand. Although only further research can ascertain the outcome of a third post-test, it is not unreasonable to assume--given the known behavior of the two groups--that the condition of etymological familiarity would surpass the rote memorization condition

Figure 1
Grand Means

Although it is not clear what exactly might be going on in the learner’s brain for the knowledge of original usage of idioms to have this attested mnemonic effect, several theories may help to explicate the behavior of the subjects in this study. One that offers a possible rationale for the experimental group’ superior performance is the Dual Coding theory (Clark and Paivio, 1991; Paivio,1986), which holds that the mental system consists of two separate subsystems---the imagery one whose functions include the analysis of scenes and generation of mental images, and verbal one which is specialized for dealing with verbal language. These systems are separate but interconnected, so that they can function in parallel, or in an integrated manner.

   Paivio's dual coding theory supports the idea that people learn by connecting verbal information and mental images, that is, when the learning materials involving related verbal and imagery information rather than verbal materials or imagery materials alone. It also appears that information presented through the imagery channel is more salient and better remembered than information presented through the verbal channel.As a global theory, dual coding theory is invoked to explain a variety of cognitive phenomena. As applied specifically to idiom acquisition and retention, the theory predicts that etymological information of an idiom can create or activate a mental image or a picture of a concrete scene in the students’ mind, and this image can then be stored in memory alongside the verbal form,and which can subsequently provide an extra pathway for recall when encountering the idiom again (Gibbs, 1997; Boers et al, 2004). This theory, together with the “levels of processing” view, can offer an even more compelling explanation for the experimental group’s performance.

   Relying heavily on trace theory, the level-of-processing model (Cermak and Craik, 1979; Craik and Lockhart, 1972) postulates separate stages for sensory, working, and long-term memory. Craik and Lockhart (1972) suggest that learners process information or stimuli at multiple levels simultaneously and that the durability of a given memory trace is a positive function of the depth to which the encoding information is processed. In other words, the deeper the processing, the more information will be retained in the long-term memory. It remains unclear that the extent to which cognitive effort would be consumed for rote memorization, but it is explicit that challenging students to process an idiom through its literal, original usage would indeed engage them in problem solving and would thus require cognitive effort at a ‘deeper’ level, which can enhance retention.What is most important to note here is that, according to this model, a “deep” processing is achieved not by any quantitative operation, such as mental rehearsal and repetition, but by a qualitative one, which involves focusing on the “motivated” meaning of idioms by their literal, original usage or other lines such as metaphorical processing.

Despite substantial evidence in support of the hypotheses that etymological elaboration facilitates both idiom acquisition and retention, some skeptics may argue that it functions only if the idiomatic meaning is derived from its original usage in a fairly straightforward way. Perhaps the results of this experiment that are highly in favor of the experimental group could be attributed to the presence of ‘etymologically transparent’ idioms in the post-tests. With respect to “etymological transparency”, however, Boers et al. (2004) contend that it is inevitably a subjective experience and should be conceived as a gradable concept with idioms occupying a position on a continuum between extremely transparent (e.g.,by and large) and extremely opaque (e.g., give someone the cold shoulder).In a previous study (Boers et al, 2004), however, the mnemonic effect of adding a touch of etymology to idioms was shown to be equally profound for opaque idioms as for transparent ones.

All in all,the encouraging results of this experiment indicate that the subjects of the experimental group exhibits superior performance on the post-tests for idiom acquisition and retention fare in comparison to the subjects who learn the material through rote memorization and this superior performance seems to be the result of etymological elaboration given to them.

Conclusion, implications and limitations
In this article, we have explored the mnemonic effect produced by etymological elaboration (i.e. elaborating on the original usage of idioms). While previous studies had already provided ample evidence of the potential benefits of this pedagogical technique (e.g., Boers et al, 2004), the present study was intended to investigate the impact of this technique versus rote memorization on idiom acquisition and retention.Two hypotheses were measured in this study: etymological elaboration can help learners (1) acquire more idioms and (2) exhibit higher retention rates than rote memorization. The experimental data reported in the study are quite encouraging with regard to both two hypotheses. The finding of this study suggests that for the purpose of encouraging long-term retention, raising learners’ awareness of the original usage of idioms is preferable to asking students to memorize idioms on the basis of their given meaning.

   Though this research has not been exhaustive, an attempt has been made to find an insightful use for etymology in teaching foreign language, especially teaching idioms. Idiomatic expressions are often a part-and-parcel of authentic material and many thousands of them exist in British or American English the learners must somehow learn. For most foreign or second language learners, it is a hard nut to crack to deal with such expressions in class because of their property of semantic eccentricity. In order to adequately grasp such expressions and keep them in long-term memory the learner requires a feasible strategy that follows a reliable approach (Krishna, 2006). As a teacher of English as a foreign language,he/she also needs to provide reliable strategies for students to learn idioms ubiquitous in authentic language. The results of the study show that there possibly exist such strategies, one of which is elaborating on the literal, original usage of idioms when encountered in class. In addition, it would make sense in classroom contexts to challenge students to hypothesize the original usage of idioms because this task can engage them in problem solving and would require cognitive effort at a ‘deeper’ cognitive level.

The results of the study also suggest that, because of the mnemonic effect etymology produces on idiom acquisition and retention,etymology courses for language students of all grade levels should be in order. Unfortunately, such courses seem to be rare in foreign language classrooms in higher education of China. As a result, students often have little knowledge about the etymology of most figurative vocabularies and multi-word expressions. Given the efficacy of etymological elaboration in idiom acquisition, it does not make good pedagogical sense to subordinate etymology courses to other language courses. An etymology course, no doubt, should remain quite indispensable from foreign or second language classrooms in higher education.

   This study, nonetheless, suffer from a few shortcomings. First, the time interval between the first post-test and second post-test was very short, only four weeks. Second, in the post-tests, the target idioms tested were embedded in a meaningful context, but not presented in isolation. There would probably be an unclear cut between the function of contextual clues and etymological familiarity in idiom tests. Finally, the subjects were not randomly selected and the implementation of the study was within one university. Studies can be conducted with samples randomly selected in different universities and see the results.

Notes
[1] Although the term “acquisition” is conceptually different from the concept “learning” in the field of SLA (e.g., Krashen, 1982), the two terms are used interchangeably in this present study.
[2] Although the term “acquisition” is sometimes used as a blanket term to encompass the notion of retention, for purposes of analysis and operationalization, the two constructs acquisition and retention are considered separate in this study and are measured independently.
[3] According to Geoffrey et al. (1992, p. 128-9), the analysis of variance can be used to analyze experiments with any number of conditions, including, of course, two-group experiment because the t test is a special case of the F test. To be more specific, if you were to conduct a t test and an F test on the data from the same two-group experiment, you would reach exactly the same conclusions. The reason the conclusions would be identical is that the two statistical tests are algebraically equivalent, that is,

                               

References
Bobrow, S. and S. Bell. (1973). On catching on to idiomatic expressions. Memory and Cognition, 1, 343-346.

Boers, F. (2000). Metaphor awareness and vocabulary retention. Applied Linguistics, 21(4), 553-571.

Boers, F. (2001). Remembering figurative idioms by hypothesizing about their origins. Prospect, 16, 35-43.

Boers, F., Demecheleer, M. and Eyckmans, J. (2004). Etymological elaboration as a strategy for learning idioms. In P. Bogaards and and B. Laufer (Eds.), Vocabulary in a second language: Selection, acquisition and testing(pp. 53-78), Amsterdam and Philadelphia: John Benjamins.

Cacciari, C. (1993). The place of idioms in a literal and metaphorical world.In C. Cacciari and P. Tabossi (Eds.), Idioms, processing, structure and interpretation (pp.76-87). Hillsdale, New Jersey: Lawrence Erlbaum Associates.

Cermak, L. and Craik, F. (1979). Levels of processing in human memory. Hillsdale,New Jerse: Lawrence Erlbaum.

Clark, J.M. and Paivio, A. (1991). Dual coding theory and education. Educational Psychology Review, 3, 233-262.

Cooper, T. (1999).Processing of idioms by L2 learners of English. TESOL Quarterly, 33, 233-262.

Craik, F.I.M., & Lockhart, R.S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behaviour, 11, 671-684.

Fernando, C. and Flavell, R. (1981). On idiom: Critical views and perspectives. Exeter Linguistic Studies,5. Exeter: University of Exeter.

Flavell, L. and Flavell, R. (2000).Dictionary of idioms and their origins. London:Kyle Cathie.

Grant, L. 2004: A corpus-based investigation of idiomatic multi word units. PhD dissertation, Victoria University of Wellington.

Grant, L. and Bauer, L. 2004: Criteria for re-defining idioms: Are we barking up the wrong tree? Applied Linguistics, 25, 38–61.

Geoffrey, K. et al. 1992. Introduction to design and analysis: A student's handbook. New York: W. H. Freeman and Company.

Gibbs, R. W.(1997). Metaphor in idiom comprehension. Journal of Memory and Language, 37, 141-154.

Gibbs, R. W. and J. E. O'Brien. (1990). Idioms and mental imagery: The metaphorical motivation for idiomatic meaning. Cognition,36, 35-68.

Gibbs, R. W., & Nayak, N. (1991). Why idioms mean what they do. Journal of Experimental Psychology:General, 120, 93-95.

Gibbs, R. (1983). Do people always process the literal meaning of indirect requests? Journal of Experimental Psychology: Learning, Memory, and Cognition, 3, 524-533.

Gibbs, R. W. (1992). What do idioms really mean? Journal of Memory and Language,31, 485-506.

Hatch, E. and C. Brown. (1995). Vocabulary, semantics, and language education. Cambridge: Cambridge University Press.

Kövecses, Z. and Szabó, P. (1995). Idioms: A view from cognitive semantics. Applied Linguistics, 17, 326-355.

Krishna, K. (2006). Metaphorical systems and their implications to teaching English as a foreign language.
Asian EFL Journal, 6(1), 2006. Retrieved April 20, 2006 from
http://www.asian-efl-journal.com/04_kk.php.

Lakoff, G. (1993). The contemporary theory of metaphor. In A. Ortony (Eds.), Metaphor and thought (pp. 203-225). Cambridge: Cambridge University Press.

Lakoff, G. (1987).Women, fire, and dangerous things. Chicago: The University of Chicago Press

Larsen-Freeman, D. & Long, M. H. (1991). An introduction to second language acquisition research. Beijing: Foreign Language Teaching and Research Press.

Lewis, M. (1997). Implementing the lexical approach: Putting theory into practice. Hove: LTP.

Li Xiuping. (2005). An analysis of Chinese EFL learners' beliefs about the role of rote learning in vocabulary learning strategies.
Asian EFL Journal, 7(4), 2005. Retrieved March 16, 2006 from
http://www.asian-efl-journal.com/December_05_xl.php

Paivio, A. (1986). Mental representations. Oxford: Oxford University Press.

Speake, J., editor.(1999). Oxford dictionary of idioms. Oxford: Oxford University Press.

Zheng, Shutang. (2001). New horizon: College English. Beijing: Foreign Language Teaching and Research Press.

Appendix
The 50target idioms selected for the subjects
Idioms tested in the first post-test
The gap-fill section:
A shot in the arm; Show your true colors; Be off one’s trolley; Take it on the chin; Burn your bridges; A one-man band; Put a damper on something; Be left high and dry; Pull something out of the hat; Keep one’s finger on the pulse of something.

The multiple-choice section:
Go with the flow; Bite the hand that feeds you; Jump in at the deep end; Play second fiddle to someone; At the drop of a hat; Live from hand to mouth; A flash in the pan; Run the gauntlet; Rest on your laurels; Keep your cards close to the chest.

Idioms tested in the second post-test
The gap-fill section:
Be worth one’s salt; Keep a tight rein on someone; In the nick of time; Ride roughshod over someone; Come home to roost; Keep one’s finger on the pulse of something; A chip off the old block; Below the belt; Take the bait; Keep your cards close to the chest.

The gap-fill section:
A ball and chain; Beat about the bush; Let the cat out of the bag; Make a beeline for something;Lose your shirt; Kill the goose that lays the golden egg; Play to the gallery; Set off on the wrong foot; Pull something out of the hat; A flash in the pan.

The 14 idioms NOT tested in the post-tests
The gloves are off; The dice are loaded against someone; Have a finger in the pie; A feather in your cap; in the lap of the gods; Ride herd on someone;Cry for the moon; Cut the Gordian knot; Shoot from the hip; Do a hatchet job; Get your second wind; Have bigger fish to fry; Meet somebody halfway; One for the road.