Comparing the effects of explicit and implicit teaching using literary and nonliterary materials on learner’s pragmatic comprehension and production

The present paper explored how plays could contribute to pragmatic development when employed as a medium of implicit or explicit instruction. 80 undergraduate English-major students were divided into four experimental groups two of which were literary and two nonliterary. Implicit Literary, as one of the literary groups was exposed to typographically enhanced plays containing the speech acts of, refusal, apology, and request. Explicit Literary, the other literary group received the same speech acts and metapragmatic instructions on the acts as well. For the nonliterary groups, dialogs which contained the given functions served as the medium of instruction for either Implicit Nonliterary containing enhanced input or Explicit Nonliterary having input plus metapragmatic information. All participants took a MultipleChoice Discourse Completion Test (MDCT) and a Written Discourse Completion Test (WDCT) both before and after the instruction. Analysis of the participants’ performance on the pretest and posttest WDCT did not reveal any significant difference between the literary and nonliterary groups. Concerning the mode of instruction, explicit groups outperformed the implicit ones. Pre and posttest analyses of the participants’ performance on the MDCT revealed no significant differences among the groups in terms of the four teaching conditions implying that EFL learners’ knowledge of speech acts can be boosted by pragmatic instruction irrespective of the medium or mode of instruction. Subjects: Bilingualism / ESL; Language Teaching & Learning; Literature


ABOUT THE AUTHORS
Mohammad Baqerzadeh Hosseini is Assistant Professor in Applied Linguistics. He is faculty member at Islamic Azad University, Naragh branch, Iran. His primary research interests include pragmatics, critical discourse analysis, and vocabulary.
Hossein Pourghasemian is Assistant Professor in Applied Linguistics. He is faculty member at Qom University of Technology, Qom, Iran. His primary research interests include literary criticism, second language acquisition, teaching writing skill, and discourse analysis.

PUBLIC INTEREST STATEMENT
In order to communicate effectively in a foreign language, one needs to be able to both understand one's interlocutor's meaning and make one's meaning understood. This ability is called pragmatic competence. We as teachers must equip language learners with such a knowledge. In this study we checked whether we could better impart such knowledge using literature (play or drama in this case). The comparison of the results of student performance indicated no edge for literary materials over nonliterary ones. Kasper (1996) defines interlanguage pragmatics as the study of how nonnative speakers comprehend, acquire and employ L2 pragmatic knowledge. This area of inquiry has usurped a great deal of attention in the past two decades leading to greater attention to the teaching of pragmatic competence. Since 1990s a host of empirical studies have been conducted especially attempting to address three main issues concerning pragmatic development such as the teachability of pragmatic features, its instructional efficiency compared with mere exposure and the effectiveness of various approaches to teaching pragmatics (Rose, 2005). Pragmatic development has been traced and studied through several empirical studies dealing with instructional methods and learning opportunities for pragmatic instruction (Taguchi, 2011). According to Rose (2005), teaching pragmatic features is possible and teaching pragmatics can be more beneficial than mere exposure. As Taguchi (2015) asserts, however, it is early to decide which approaches or methods are more effective in teaching pragmatics.

Introduction
The explicit-implicit dichotomy of teaching pragmatics has dominated the literature addressing the differential effects of various instructional methods on pragmatic development. The explicit method of pragmatics instruction has proved to be more effective than implicit teaching, yet more research is needed to make any categorical claim about the superiority of one mode over the other (Takahashi, 2010). It might be that the two modes can be realized in many ways leading to the difficulty of coming at a definite conclusion.
Research on teaching methods and approaches to teaching pragmatics necessitates the generation of a variety of teaching materials to be employed in the classroom (Taguchi, 2015). Although inherently significant, the effects of teaching materials have neither independently nor in combination with a method of instruction come under study. Literature can be employed as a material for both explicit and implicit teaching. Literature has long been used in teaching English as a resource providing both motivating and authentic content. In addition to boosting grammatical and lexical knowledge, literature can familiarize EFL learners with the target culture and its social practices and norms (Hall, 2005;Kim, 2004). As it is rich in pragmalinguistic and sociopragmatic aspects of pragmatic competence, literature can serve as an appropriate means for facilitating pragmatic instruction.
According to Goodwin (2001), communicative competence including pragmatic awareness can be taught effectively through drama or play. A wide array of pragmatic abilities can be developed and practiced through mini-dialogs, drama and role plays (Olshtain & Cohen, 1991). However, the linguistic side of communicative competence has attracted the attention of researchers, while the pragmatic side has gone almost unnoticed (Iida, 2012;Paesani, 2005). The present study intends to explore the potential of plays, being predominantly dialogic in nature, in boosting pragmatic development through teaching certain speech acts either explicitly or implicitly.

Pragmatics instruction studies
Pragmatic development has been studied through interventional studies which can be classified into instruction versus no instruction or mere exposure studies, teachability studies and studies dealing with various teaching approaches. In a review of the literature on pragmatics instruction, Rose (2005) found that in general pragmatic features are teachable and that instruction can lead to pragmatic development. Nevertheless, it is still not fully known which approach to teaching pragmatics yields better results.
Experimental studies generally suggest that explicit teaching of pragmatics which involves metapragmatic information including social status, psychological and social distance, and the degree of imposition has proven to be more effective than implicit instruction (Takahashi, 2010). In explicit teaching the learners' attention is drawn towards specific linguistic features and the discovery of how these features are played out in contextual factors. Explicit teaching is in line with the noticing hypothesis requiring conscious attention to information related to pragmatics in the L2 class, rather than learners' mere exposure to pragmatics-rich input. Pragmatic development may not ensue from mere exposure to pragmatic input or the learning pace might be very slow (Alcon, 2005;Rose, 2005). Only through activities involving attention to form-function-context mappings and focal pragmatic forms, implicit teaching can be effective in pragmatic development (Taguchi, 2015). Consciousness-raising activities and typographically enriched instructional materials can be employed in implicit teaching to this end. Upon reviewing empirical studies comparing the effects of explicit and implicit instructions to pragmatic development, Taguchi (2015) identified three categories which were difference (between explicit and implicit teaching), no difference (between explicit and implicit teaching) and mixed results (difference in explicit and implicit teaching in one respect but no difference in another).
In the first category, the difference category, there was a significant difference between explicit and implicit methods and in most of these studies the explicit group fared better. Nguyen, Pham, and Pham (2012), for instance, investigated the relative effectiveness of explicit teaching (metapragmatic explanation and correction of errors of forms and meanings) and implicit form-focused instruction (pragmalinguistic input enhancement and recast activities) on the development of English pragmatic competence. They found that the explicit group performed significantly better than their implicit counterpart.
A number of studies, however, displayed no significant difference between explicit and implicit pragmatic instructional methods. For example, Takimoto's (2006) evaluation of the relative effectiveness of structured input instruction (implicit method) and structured input instruction plus feedback (explicit method) for teaching English polite request forms revealed that the performance of the two treatment groups in the study was significantly better than the control groups, though the two experimental groups did not have significant differences.
The third category included studies which came up with mixed results when comparing the explicit-implicit modes of instruction. For instance, in a study of the influence of implicit versus explicit instruction on the EFL learners' ability to employ request strategies, Alcon (2005) came up with mixed findings. The results indicated that the learners benefitted from both modes as far as the awareness of requests was concerned. Although the explicit group did better on the production task, the two groups were not significantly different on the recognition task.

Literature teaching studies
For a long time language teaching was equal to the teaching of literature as the appreciation of literature was the main purpose of learning a foreign language. However, with the advent of new methods and approaches particularly the communicative approach, teaching literature has not been so welcome in second language context as it is in first language situation (Iida, 2013). The reasons may range from the lexical and syntactic complexity of literary texts which renders them difficult to understand (Chen, 2006;Lazar, 1994), to the alleged time-consuming nature of literary reading (Paran, 2006). In spite of all these difficulties, literature can solve one of the perennial problems of language teaching, i.e., "the search for engaging and authentic content" (Bibby, 2012, p. 5). Iida (2012) believes that the new interest in literature lies in the fact that applied linguistics has shifted from mere theoretical concerns to practical ones. Now, literature is not only seen as the provider of motivating content but also as the necessary context for language teaching (Bibby, 2012).
Although there have been a good number of studies on the use of literature in educational contexts, especially in the language classrooms, these studies, according to Paran (2008), have mostly dealt with the impact of teaching literature on the development of L2 skills and components, and the pragmatic aspects of communicative competence have not received due attention. For example, in an attempt to use poetry to extend learners' understanding of the uses and meanings of linguistic structures, Hanauer (2001) focused on the potential of the poetry reading task to direct learners' attention to formal aspects of the language ("focus on form").
The effect of literary discussions on the learners' interactions was studied by Kim (2004) who attempted to investigate the effects of students' interactions with literary texts (short stories and novels) and with their peers on their language development. His study revealed that literary discussions led to greater enjoyment of reading literature, and the resultant social interactions boosted the learners ability to express themselves more conveniently in English.
In another empirical study, Wang (2009) dealt with the effect of reading novels on the learners' skills and their problem-analysis abilities. The study revealed that literature instruction enhanced the learners' grammatical and lexical knowledge and their ability in the four language skills. It also increased learners' overall general English proficiency and improved their L2 cultural knowledge.
The empirical studies conducted on literature use in classroom in general back the rationale of employing literature in L2 instruction. Regardless of genre differences, diversities in methodology, approaches and contexts, literature has commended itself as a positive tool in L2 learning. The afore-cited studies suggest that literature can lead to the improvement of the learners' cultural and linguistic competence, and hence develop their communicative competence.

Purpose of the study
The major bulk of studies concentrating on aspects of teaching literature have mostly dealt with the contribution of literature to increasing the overall English proficiency or the facilitated learning of language skills and few studies have addressed the effects of literature teaching on the learners' pragmatic competence. The studies addressing interlanguage pragmatics have mainly considered social and cognitive aspects of teaching pragmatics and have mostly ignored techniques or practices which can contribute to pragmatic development. Moreover, few studies have directly dealt with the potential of literary genres such as plays in promoting pragmatic development of EFL learners.
Considering the fact that literature can contribute to both linguistic (Hanauer, 2001;Paesani, 2005) and cultural competence (Allington & Swann, 2009;Kim, 2004) and also the fact that plays on account of their dialogic structure can manifest speech acts, the present study intends to investigate the possible contribution of instruction through plays on learners' pragmatic development. In other words, this study intends to see whether instruction, either explicit or implicit, through literary materials (plays) makes a difference. More specifically it seeks to answer the following questions: (1) Do the four teaching conditions (explicit literary, implicit literary, explicit nonliterary and implicit nonliterary) contribute differently to the recognition and production of the given speech acts?
(2) Do the medium (literary vs. nonliterary) and mode (explicit vs. implicit) of instruction contribute differently to the recognition and production of the given speech acts?

Participants
The study comprised 80 EFL students majoring in English language and literature and translation studies at universities and institutes of higher education in Qom, Iran. The participants were freshmen (48 females and 32 males) who were between 18 to 23 years old. since there was not a large student population to choose the participants from, we had to be content with our four intact groups labeled as Explicit Literary (EL), Implicit Literary (IL) and Explicit Nonliterary (EN), Implicit Nonliterary (IN), depending on the treatment conditions of the study.
For the purposes of the study, authenticating the DCTs and deciding the answer key for the MDCT, we asked 20 native English speakers (11 North Americans, 6 British, and 3 other) to participate in this study. They were 10 males and 10 females of whom 12 were over 50, 5 were 40-50, 2 were between 30-40 years old and one was under 20.

Instruments
The instruments of the present study were as follows: A test of General English Proficiency used to homogenize the groups both within and without consisting of three sections: reading comprehension, vocabulary and structure (the test was an adaptation of the English proficiency test developed by one of the universities in Iran). The test piloted on one of the groups while being compared with a TOEFL test (2005) administered on the same group, and the scores showed a high correlation with the TOEFL scores (r = 0.86). Since the test was a truncated form of a longer test, Cronbach's Alpha analysis was run, and the results revealed a high reliability (α = 0.78) for the test.
The two other tests, the MDCT and the WDCT both including 12 items, were meant to test the participants' recognition and production of the speech acts of apology, request, and refusal (4 items for each speech act). In every task, a scenario was included which supplied the necessary context based on the status and distance of the participants who were concerned in a given situation. As for the MDCT, the test takers were to choose the most appropriate option given the relevant parameters of the situation (Appendix A). In the case of the WDCT, the test takers were asked to write down what they would normally say in every one of the given conditions or situations (Appendix B). Valipour and Jadidi (2015) provided the refusal items for the tests and Khatib and Ahmadi-Safa (2011) provided request and apology sections of the two tests. As both the MDCT and the WDCT were posted online as a google form on Linked-in, and native speakers were requested to answer the tests and pass their views on the items. In the case of the MDCT, the option most frequently chosen by the native speakers was considered as the key for each item on the test.

Materials
In order to provide material for the pragmatic teaching in this study, a few one-act plays, and a number of dialogs were used, which were accompanied with some metapragmatic information for the explicit groups. The following plays were used in this study: The Boor by Anton Chekhov, Roulette by Douglas Hill, St Martin's Summer by Cosmo Hamilton, Her Tongue by Henry Arthur Jones, and Bloody Mary by Greg Vovos. To ensure easy access and to increase input enhancement, the plays were thoroughly scanned and all acts and their related adjuncts were underlined. The dialogs for the nonliterary groups were mostly adopted from Functions (Matreyek, 1990) and to a lesser extent from Four Corners series (Richards & Bohlke, 2012). For each specific speech act, the corresponding dialogs were carefully chosen and care was taken to provide the necessary structures and the contexts for the fulfillment of the speech acts. In all, for every instructional session six dialogs were used.
For their first lesson before the dialogs and the plays, the participants in the explicit groups (Explicit Literary and Explicit Nonliterary) received the metapragmatic information which consisted of Olshtain and Cohen (1983)'s definition of the apology speech act and a classification of apology strategies. In their second lesson, the two explicit groups were given an explanation of the request speech act and the levels of directness related to this speech act based on Blum-Kulka and Olshtain (1984). Politeness was also explained to them in terms of imposition, position, distance and degree of directness. The external modifiers, or the adjuncts to request were explained in the third lesson. The speech act of refusal including the related adjuncts of this speech act and its categorization based on Salazar, Safont, and Codina (2009) was explained in the fourth session. In the fifth lesson, the participants were instructed how to refuse invitations, suggestions and requests made by people of different status and of degrees of familiarity.

Data collection
In the first place, for the sake of homogenizing the participants in terms of their general proficiency both inside the groups and among the four groups, the proficiency test was administered to them.
Then, to determine the participants' level of pragmatic competence prior to the inception of instructional sessions and to ascertain that the level of pragmatic competence was not very significantly different among and inside the groups, the pretests (the MDCT and WDCT) were given to the participants. Following these steps, the five-week pragmatic instructions began and continued.
In the case of the explicit groups, the procedure was follows: at first the metapragmatic information was reviewed by the teacher and the corresponding speech act was explained thoroughly to the participants with regard to the related situations and the people involved. Then the students were required to read the dialogs or the play to identify the main act, its related adjuncts and the kinds of the relevant strategies employed to actualize the speech act. The students were also asked to decide if the speech act had been properly actualized with the actors participating in the dialogs or the play. Each lesson lasted between 25 to 30 minutes.
The implicit groups did not have any exposure to the metapragmatic information, the participants in these groups only had to read the dialogs or the play and to concentrate on the underlined parts and regard the actualizations of the speech acts and see if the actors involved had been successful in materializing the speech acts properly. The completion of each lesson took 15 to 20 minutes. At last, the same WDCT and MDCT tests which were given as the pretest were administered to the students as the posttest one week after the final instructional session.

Data analysis
The present study involved one independent variable (teaching condition) with four levels (explicitliterary, implicit-literary, explicit-nonliterary, and implicit-nonliterary), and one dependent variable (pragmatic performance) with two levels (pragmatic comprehension and production). The mode of instruction (either explicit or implicit instruction) concerned the teaching approach and the medium of instruction concerned the instructional materials, either literary or non-literary. Pragmatic comprehension and pragmatic production which involved recognition and production of the given speech acts were indexed based on the participants' scores on the MDCT and WDCT respectively.
The WDCT was rated by the two authors of the study and another researcher who was also a nonnative speaker but a professional (who made research in interlanguage pragmatics). Rating was done by employing Taguchi's (2006) rating scale of appropriateness (Appendix C).Based on the accurate and appropriate production of the relevant speech acts in the given situations, the scale had a 6-point rating scale in which no performance equaled 0 and high appropriate and accurate performance was equal to 5, showing high excellence. It must also be noted that the inter-rater reliability among the raters was shown to be high (the Pearson correlation coefficient was r = .90). The MDCT score was scored using the answer key developed from the responses provided by the native participants.
To compare the conditions of the groups before starting the instruction, the participants' scores on the proficiency test and the WDCT and MDCT pretests were analyzed through ANOVA. Since a statistically significant difference was found among the groups, the authors used ANCOVA to compare the performance of the groups on the MDCT and WDCT posttests.

Results
To ascertain the homogeneity of the four groups regarding the general English proficiency of the participants inside the groups, the one-way ANOVA was employed to compare the means of the groups. As Table 1 displays, the P value was calculated to be (Sig = .00) which is manifestly smaller than the critical value (.05) confirming the significant differences among the groups.
The one-way ANOVA of the participants' MDCT pretest scores ( Table 2) which was carried out to compare the groups' performance on the MDCT pretest, revealed that there was not significant differences among the four groups (Sig = .57), suggesting that the four groups were equal in terms of recognizing the speech acts before starting the instruction.
However, applying one-way ANOVA to the scores of the four groups on the WDCT pretests displayed the P value to be (Sig. = .00) which was manifestly less than the critical .05 value, implying that the four groups were significantly different in their production of the speech acts before the onset of instruction (Table 3). That is, in terms of their pragmatic production ability, the four groups of the study were different and were not homogeneous.
The first research question asked whether the performance of the four groups changed due to the four instructional conditions of pragmatics. Since the groups displayed significant differences in their proficiency level, to analyze their performance on the posttests ANCOVA was employed. The one-way ANCOVA of the participants' scores on MDCT revealed that the groups were not significantly different as far as the recognition of speech acts under the four different instructional methods was concerned regarding the P value (Sig = .08) which was greater than .05 signifying that the difference was not very considerable (Table 4).
In the case of the participants' scores on the WDCT posttest, the one-way ANCOVA analysis displayed a significant difference (Sig = .00) among the four groups under the four instructional conditions of teaching pragmatics. As the results on Table 5 reveal, the four groups were different and each teaching condition led to differences in the production of speech acts.
The Scheffe post hoc test was employed to pin down the degree of the significant differences among the four groups. As the findings on Table 6 reveal there were significant differences between IL and EN (Sig = .00) and between EL and IL (Sig = .01); the other differences were not significant. Put simply, the performance of EL and EN was significantly better than IL on the WDCT  posttest and the performances of IN, EL, and EN were similar although some insignificant differences were observed among them.
The effect of the medium and mode of instruction on the learners' pragmatic performance was the concern of the second research question. It intended to probe into the possible differences between the nonliterary and literary groups, and between the explicit and implicit groups on the recognition and production of the speech acts.
As Tables 7 and 8 indicate the medium (play or dialog) of presenting pragmatic lessons caused no significant difference in the learners' recognition of the most fitting speech act for the intended situations. Neither did the medium (play/dialog) of instruction affect significantly the learners' production of the speech acts in the given situations. The one-way ANCOVA analysis of the participants' scores in the nonliterary (implicit nonliterary and explicit nonliterary) groups and the literary (implicit literary and explicit literary) groups on the MDCT posttest as displayed in (Table 7) showed no significant difference between the two groups since (Sig = .08) was above the critical .05 value. Moreover, As indicated by Table 8, the one-way ANCOVA analysis, of the performance of the groups on the WDCT posttest revealed no significant difference between the literary (EL and IL together) and nonliterary (EN and IN together) groups since the P value (Sig = .10) for the medium of instruction was manifestly higher than.05.   As for the mode of instruction, Table 9 indicates the mode of presenting pragmatic lessons whether implicitly or explicitly did not significantly affect the learners' recognition of the most suitable speech act for the situation. The one-way ANCOVA analysis of the participants' performance on the MDCT posttest revealed that there existed no significant differences (Sig = .058) between the explicit (explicit literary and explicit nonliterary) and implicit (implicit literary and implicit nonliterary) groups.
Nevertheless, the mode of presenting pragmatic lessons had a significant effect on the learners' performance on the WDCT posttest. As indicated by Table 10, there existed a significant difference between the implicit and explicit groups since (Sig = .001) is considerably lower than the critical value of .05.

Discussion
The present study intended to probe into the potential of plays,due to their dialogic nature, as means of facilitating the instruction of pragmatic functions through both implicit and explicit ways of teaching. Nevertheless, the findings of the study revealed no particular advantage for the plays as a medium of pragmatic instruction. The findings of the study are reviewed regarding the relevant theory and previous studies.
The main concern of the present study was to discover if the four groups involved in this study differed on account of differing instructional conditions. The results obtained were mixed regarding the productive and receptive aspects of their pragmatic competence. In terms of pragmatic comprehension, i.e. recognizing the most suitable instance of speech acts in the given situations, the four groups displayed no significant differences as analyses of the statistical data revealed. Since the plays and the dialogs, as the medium of instruction, differed only in their length, this finding should not be very surprising. However, regarding the mode of instruction, our findings are not congruous with a major bulk of the studies in the field which mostly point to the superiority of explicit instruction over implicit teaching (Rose, 2005;Taguchi, 2015).
As far as pragmatic production was concerned, there were significant differences among the groups, in the production of the speech acts: EN (Explicit Nonliterary) displayed the best performance, next stood EL (Explicit Literary), IN (Implicit Nonliterary) ranked third and the fourth was IL (Implicit literary). This implies that teaching pragmatics explicitly has an advantage over its implicit instruction. This finding is congruous with a host of past studies (e.g. Koike & Pearson, 2005;Nguyen et al., 2012) advocating explicit mode of instruction over implicit teaching. At the same time, it implies that explicit instruction of pragmatics through nonliterary materials such as dialogs might be more effective than using long literary materials like a play. As for the medium of instruction, comparing nonliterary (Explicit Nonliterary and Implicit Nonliterary) groups with the literary (Explicit Literary and Implicit Literary) groups displayed no significant difference between them as far as both pragmatic comprehension and production were concerned. This implies that instructional material did not differentially affect the participants' performance. Due to the dearth of precedence in the interlanguage pragmatics literature on the use of plays, and also due to the fact that the studies in the field have only paid a secondary attention to the medium of offering pragmatic teaching, we cannot make any substantial claims over the fitness or unfitness of our finding. In terms of pragmatic production, the difference between pretest and posttest scores of the non-literary groups was better than that of their literary counterparts. This can be accounted for by the fact that the nonliterary groups were exposed to several dialogs each of which providing enough context on the situation, and as such the learners were provided with easy access to instances of the acts within a short space. At the same time, the dialogs had been selected from textbooks specially written for instructional purposes. Whereas, the literary groups had to follow one relatively long extended dialog, i.e. play, which did not follow any pedagogical aim, and these learners had to process the sporadic cases of the given speech act within a wide space. This fact might be theoretically explained with reference to cognitive psychology.
In one respect, Ellis (2013) asserts that the frequency of exposure leads to better learning and the more times one experiences something, the stronger his memory is activated for it and the more fluently that thing will be accessed. In this regard, the learners in the nonliterary groups had greater exposure to more instances of the intended speech acts than their literary counterparts since the dialogs provided for the nonliterary groups contained at least one instantiation of the speech act, while the play provided for the literary groups did not possess as many cases of that particular function. Moreover, according to Van Patten (2004) meaning and form compete to capture the reader's attention, and the learners will attend to form only when they can understand input easily. Based on this line of reasoning, while reading a play, learners could have been so much absorbed with meaning, or more specifically with the story, that they have not paid attention to the rather dispersed and sporadic instances of a particular speech act in the play.
Regarding the mode of instruction, mixed results were obtained. In terms of pragmatic comprehension, our findings indicated no significant differences between the implicit (Implicit Literary and Implicit Nonliterary) and the explicit (Explicit Literary and Explicit Nonliterary) groups. In other words, the performance of the groups in recognizing pragmatics was not influenced by the mode of instruction whether it was explicit or implicit. In this respect, our study is in congruity with nodifference studies (e.g. Martinez-Flor, 2006;Takimoto, 2006). These studies assert no superiority for explicit instruction and imply that implicit teaching can be equally effective.
As for pragmatic production, our study revealed better performance for the explicit group, a finding in line with a number of studies conducted in the field (Rose, 2005;Taguchi, 2015;Takahashi, 2010). Upon reviewing 58 experimental studies in interlanguage pragmatics, Taguchi (2015) found an edge for explicit form-focused instruction involving metapragmatic information over implicit instruction. Therefore, even when boosted by enhancement techniques, input exposure per se could not result in as much effective learning as obtained from the explicit instruction. The reason might be sought in the greater degree of amenability of pragmatic features to explicit, rather than implicit, instruction (Rose, 2005). When the learners are exposed to explicit metapragmatic instruction on the sociopragmatic and pragmalinguistic features of the speech acts, accompanied with explicit awareness-raising tasks, probably their metapragmatic awareness is raised. Taguchi (2015) asserts that if the implicit condition aspires to be as effective as the explicit condition, instruction should be geared to involve the effective processing of the form-functioncontext mappings by the learners.

Conclusion
The study intended to probe into the possibility of boosting pragmatic competence through play as a literary genre which served as the medium of pragmatic comprehension and production, it also attempted to take into account the dichotomy of explicit versus implicit approach to pragmatics instruction. In terms of pragmatic comprehension, the groups under the four medium-mode configurations of teaching conditions did not display any significant differences; however, they revealed significant differences in their pragmatic production. In addition, the results indicated that it did not make much difference whether pragmatics was taught through plays or it was taught through employing specially designed dialogs. Moreover, pragmatic comprehension showed not to be critically affected by the mode of explicit or implicit instruction but the mode of instruction yielded differential pragmatic production. Therefore, what matters is the act of teaching pragmatics and not leaving it untaught. In other words, teaching the pragmatic features of a language, regardless of the mode, materials, or conditions, is effective in the development of the learners' pragmatic competence, and the different instructional conditions can facilitate the pragmatic awareness necessary for the recognition and production of speech acts. In a nutshell, the findings underscore the significant role of pragmatic input and point to the advantages of teaching pragmatics in EFL situations. It is recommended that material developers and teachers incorporate authentic material rich in pragmatic features with the necessary metapragmatic information to foster learners' pragmatic development. Finally, as there is a dearth of research in the area of using literature for pragmatics instruction, it is suggested that the future studies address use of different literary genres in teaching speech acts and other pragmatic features.