Teaching Apology Formulas at the Discourse Level: Are Instructional Effects Maintained Over Time?

ELIA 16, 2016, pp. 13-48 13 Alicia Martínez-Flor DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Estudios de lingüística inglesa aplicada TEACHING APOLOGY FORMULAS AT THE DISCOURSE LEVEL: ARE INSTRUCTIONAL EFFECTS MAINTAINED OVER TIME? LA ENSEÑANZA DE FÓRMULAS DE DISCULPA A UN NIVEL DISCURSIVO: ¿SE MANTIENEN LOS EFECTOS DE LA INSTRUCCIÓN A LARGO PLAZO? Alicia Martínez-Flor Universitat Jaume I, Castellón, Spain [email protected] DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 This study explores the effects of teaching apologies at a discourse level on a group of Spanish learners of English as a foreign language both in the short and in the long-term. It adopts a one-group pre-, post- and delayed post-test design to particularly examine the effectiveness of a pedagogical model on learners’ appropriate use of apology formulas not only after immediately receiving instruction, but also five months later. An interactive discourse completion test was used to analyse learners’ performance when apologising in different contrasting scenarios. Results show that the types of apology formulas produced by learners in the pretest differ significantly from those produced in the post-test, as well as in ELIA 16, 2016, pp. 13-48 14 DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Teaching apology formulas at the discourse level the delayed post-test. After receiving instruction, learners produced more elaborated apologetic responses attending to the sociopragmatic aspects involved in the different situations, a performance that was maintained over time. These results seem to demonstrate the positive benefits of instruction to foster learners’ pragmatic knowledge in terms of both frequency and variety of apology formulas not only in the short-term but also in the long-term. Key words: teaching pragmatics, apologies, durability of pragmatic instruction, EFL context Este estudio explora el efecto de enseñar disculpas a un nivel discursivo a un grupo de estudiantes españoles que aprenden inglés como lengua extranjera tanto a corto como a largo plazo. Adopta pues un diseño de un grupo, prepost- y post-test a largo plazo para examinar en concreto la eficacia de incorporar un modelo de enseñanza en el uso apropiado de fórmulas de disculpa no solo inmediatamente después de recibir la instrucción, sino también cinco meses más tarde. Se utilizó un test interactivo para completar el discurso y analizar el comportamiento de los estudiantes cuando pedían disculpas en diferentes situaciones. Los resultados indican que los tipos de fórmulas de disculpa que utilizan los estudiantes en el pre-test difieren significativamente de las utilizadas tanto en el post-test como en test distribuido a largo plazo. Después de recibir instrucción, los estudiantes usaron respuestas de disculpa más elaboradas prestando atención a los aspectos sociopragmáticos de cada situación, un comportamiento que se mantuvo a largo plazo. Estos resultados parecen demostrar los beneficios positivos de la enseñanza para fomentar el conocimiento pragmático de los estudiantes tanto en la frecuencia como en la variedad de fórmulas de disculpa no solo a corto sino también a largo plazo. Palabras clave: enseñanza de la pragmática, disculpas, durabilidad de la instrucción, contexto de inglés como lengua extranjera ELIA 16, 2016, pp. 13-48 15 Alicia Martínez-Flor DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 1. Introduction The effect of instructional intervention in the development of learners’ pragmatic competence in both second language (L2) and foreign language (FL) contexts has aroused the interest of an increasing number of interlanguage pragmatics (ILP) researchers over the last decades (see Jeon and Kaya, 2006; Takahashi, 2010; Taguchi 2011, 2015 for a review of the research undertaken). Results from this investigation have demonstrated the positive role of engaging learners in an instructional period and the teachability of all pragmatic aspects being examined (i.e. various speech acts, pragmatic fluency, discourse markers, modal particles or hedging devices, among others). Among the different speech acts, apologies have been one of the target pragmatic features addressed in interventional ILP studies. However, most of the research focusing on teaching apologies has not investigated them from a discursive perspective and has not considered the retention of the knowledge acquired about this speech act over time. In order to address these issues, the present study aims to examine the effect of teaching English as a Foreign Language (EFL) learners’ appropriate use of apology formulas at the discourse level and whether the pragmatic gains are maintained not only after immediately receiving instruction, but also five months later. 2. Literature Review 2.1. Apologies Apologies fall into the category of expressive speech acts (Searle, 1969), whose goal is to provide a remedy for an offense and restore harmony between the speaker and the hearer (Leech, 1983). In Bergman and Kasper’s (1993: 82) terms, apologies can be defined as “compensatory action to an offense in the doing of which S (the speaker) was causally involved and which is costly to H (the hearer)”. Thus, the speech act of apologising can be expected to appear post-event in a negotiated sequence between two interlocutors, where an offensive action created by the speaker has damaged the hearer. In this sense, apologies are politeness devices used in ELIA 16, 2016, pp. 13-48 16 DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Teaching apology formulas at the discourse level a communicative situation where the apologiser needs to take into account the other interlocutor’s face as an attempt to repair or restore damage to face (Brown and Levinson, 1987). However, it is important to highlight that an apology involves different aspects of face depending on the perspective from which it is considered. For the hearer, an apology is a face-saving act because it provides support for the hearer’s negative face as “it is made clear that he/ she has been harmed by the speaker’s actions” (Sabaté-Dalmau and Curell- Gotor, 2007: 291). In contrast, for the speaker, “an apology is a facethreatening act (FTA) as it damages the speaker’s positive face” (Warga and Schölmberger, 2007: 223). Indeed, it implies the acceptance that something wrong has been done, whether on purpose or not. In this case, the apology can adopt a defensive orientation towards saving one’s own face by justifying or explaining the reason for his/her failure (Trosborg, 1995). Considering therefore the complexity involved in an apologetic exchange, the speaker needs to be aware of the existence of different formulas that may be used to appropriately communicate in a remedial exchange. The choice of these apology formulas, which range from more direct to indirect realisations used to mitigate the FTA, is based on three different parameters: i) the speaker’s perception of the severity of the offense involved in the communicative act (Olshtain and Weinbach, 1987), as well as ii) the degree of social distance and iii) social power between the two interlocutors (Brown and Levinson, 1987). The extent to which these factors may influence the speaker’s assumption of responsibility will result in his/her choice of expressing an apology totally explicitly or just emphasising his/her innocence by using an indirect explanation or by simply not feeling the need to apologise. In any case, the speaker needs to know how to appropriately perform apologies (if he/she chooses to do so) so that harmony can be restored and, consequently, avoid communication breakdowns between the two interlocutors. In fact, as Olshtain and Cohen (1983) point out, a more serious offense would require a more elaborated apologetic strategy, ELIA 16, 2016, pp. 13-48 17 Alicia Martínez-Flor DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 whereas a less severe offense might only need a less intensified apology. In order to do so, and particularly in FL settings where there are limited opportunities to encounter authentic apologetic sequences, pedagogical intervention seems to be necessary. 2.2. Interventional Studies on Apologies Apologies have been the focus of attention in a series of ILP studies that have highlighted the need for instructional intervention on this speech act. More specifically, all of them have targeted pedagogical instruction on Iranian EFL learners (Eslami-Rasekh et al., 2004; Eslami-Rasekh and Eslami-Rasekh, 2008; Eslami-Rasekh and Mardani, 2010; Birjandi and Derakhshan, 2013; Simin et al., 2014; Derakhshan and Eslami-Rasekh, 2015). Eslami-Rasekh et al. (2004) conducted a pre-test/post-test design with a control group to examine the effect of explicit metapragmatic instruction on learners’ comprehension of apologies. The instruction took about 30 minutes of each two-hour class period over twelve weeks and consisted of teacher-fronted discussions, cooperative grouping, role plays and other pragmatically oriented tasks. Results indicated that learners’ apology comprehension improved significantly, thus supporting the benefits of implementing explicit instruction in FL classrooms to facilitate interlanguage pragmatic development. Following the same design (pretest/post-test with a control group), Eslami-Rasekh and Eslami-Rasekh (2008) analysed the effect of instruction on learners’ awareness and production of apologies. The treatment lasted for fourteen weeks, and 30 minutes of each three-hour class time were devoted to pragmatic related activities. Those activities included a combination of teacher-fronted discussions, peer work, role plays, semi-structured interviews, small-group discussion, introspective feedback and metapragmatic assessment tasks. Findings showed that the interventional group recognised and performed apology speech act schemes significantly better than the control group. In another study, Eslami-Rasekh and Mardani (2010) focused on the ELIA 16, 2016, pp. 13-48 18 DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Teaching apology formulas at the discourse level effects of explicit instruction on the frequency of apology intensifying devices adopting a pre-test/post-test design with a control group. The instructional treatment involved a series of activities including small data cards showing apology situations, a model dialogue for role play activities, a questionnaire for diagnostic assessment, feedback and discussion. The results supported the positive effect of pragmatic instruction to enhance the appropriateness of the use of intensifiers when apologising since learners from the experimental group outperformed those from the control group. More recently, the use of audiovisual input or technology have been incorporated in the instructional treatments to ascertain the impact they may exert on learner’s apology development. Birjandi and Derakhshan’s (2013) study incorporated a pre-test/post-test design with a control group to investigate the effectiveness of consciousness-raising video-driven prompts on the comprehension of apologies. Learners were assigned to three instructional treatment groups (i.e. metapragmatic, formsearch and role play) and exposed to vignettes extracted from different episodes of Flash Forward and Stargate TV series, and the film Annie Hall for nine 60-minute sessions of instruction twice a week. Results revealed the significant impact of using audiovisual input on the development of apology comprehension, since the three instructional treatment groups benefited from the instruction received and outperformed the control group. Similar results were obtained in the research carried out by Derakhshan and Eslami-Rasekh (2015), which made use of the same consciousness-raising video-driven prompts to examine their effectiveness with thee different types of instruction (i.e. discussion, role-play and interactive translation). Learners were also exposed to video extracts extracted from different episodes of Flash Forward and Stargate TV series, and the film Annie Hall for nine 90-minute sessions of instruction twice a week. Results indicated that learners’ awareness of apologies benefited from all three types of instruction, ascertaining thus the potential of using audiovisual materials, particularly in FL contexts. Finally, Simin et al. (2014) investigated the role of instruction in developing learners’ pragmatic awareness on recognition and production of ELIA 16, 2016, pp. 13-48 19 Alicia Martínez-Flor DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 apologies with the help of e-communication through exchange of e-mails. The study followed a pre-test/post-test design with two experimental groups (one receiving explicit instruction and the other receiving implicit instruction). The treatment, which lasted one semester of four months, was based on the provision of different hypothetical apology situations in which the learners had to write e-mail exchanges to their instructor. The explicit instruction group received feedback and discussion on each situation, whereas the implicit instruction group did not get corrective feedback of the e-mails being sent. The analysis of the e-mail exchanges revealed that the learners receiving explicit instruction gained significantly in terms of pragmatic proficiency required for strategies of apology. The authors thus support the positive benefits of incorporating this type of technological tool when teaching pragmatics. The above review of the interventional pragmatic studies on apologies has overall demonstrated the positive effect of instruction. Nonetheless, all are set in the Iranian EFL context, so as Eslami-Rasekh and Eslami-Rasekh (2008) indicate, more research is needed to widen the range of the participants’ first language (L1). Apart from this aspect, two more issues remain to be dealt with. The first one refers to the fact that none of these studies has examined apologies at the discourse level. As Félix-Brasdefer (2006) and Kasper (2006) mention, speech acts need to be understood as part of an interactional exchange in which usually two or more interlocutors construct a communicative sequence over multiple turns. On this account, different studies have considered the interactive nature of conversation when teaching speech acts within the level of discourse. For instance, focusing on compliment responses, Huth (2006) investigated the effects of teaching American learners of German how to realise L2 complimentresponse sequences in talk-in-interaction. To that end, the author developed a teaching unit of five different phases based on authentic compliment sequences in German. Results from this study showed that providing learners with opportunities for exposure and practice of the sequential organisation of compliment response sequences were positive as they were ELIA 16, 2016, pp. 13-48 20 DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Teaching apology formulas at the discourse level able to appropriately engage in talk-in-interaction. Dealing with refusals, the studies by Alcón and Guzmán (2010) and Alcón (2012) examined the benefits of pragmatic instruction at the discourse level on learners’ awareness of this speech act. Alcón and Guzmán (2010) implemented an instructional treatment based on Félix-Brasdefer’s (2006) model to teach the negotiation of refusals by using conversational analysis (CA) tools. Results from their study showed that learners not only improved their awareness of the pragmalinguistic and sociopragmatic issues involved in performing appropriate refusals, but also paid attention to other conversational skills such as turn taking or negotiation strategies. As a follow-up study, Alcón (2012) examined the effect of bilingualism on learners’ attention and comprehension of refusals. Findings from this study also illustrated the positive effect of teaching refusals at the discourse level since learners’ awareness of this speech act increased. In the light of these positive findings, it seems that empirical studies that address apology-based instruction within sequential exchanges at the level of discourse are needed. In fact, recently Limberg (2015) proposes a series of principles that should be considered when teaching apologies in the EFL context, being the last of them that of practising speech acts within sequential structures. Specifically, the interactive structure of apology situations should be emphasised by making learners understand the complexity of an apology interaction, as it does not occur in isolation in speech. As the author (2015: 283) highlights, apologies “are often initiated by a perceived form of offense (being late), sometimes followed by a complaint (“I’ve been waiting for a while now”), then an apology, which can be a sequence consisting of multiple turns itself, and finally a response, again possibly realised through another multi-turn sequence”. The second issue that is worth mentioning is related to the fact that none of the previous interventional studies on apologies has made use of delayed post-tests to assess the long-term instructional effects. As Kasper and Rose (2002) highlight, the incorporation of measures that assess the outcomes of pragmatic instruction some time after the end of the treatment period has been considered as necessary in order to support the benefits of such instruction. However, using a delayed post-test, although desirable, ELIA 16, 2016, pp. 13-48 21 Alicia Martínez-Flor DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 is sometimes not possible because of the lack of access to the original instructional group after the completion of the main study. In spite of this drawback, a number of studies have successfully included a delayed posttest addressing a variety of targeted pragmatic features. The studies by Lyster (1994) and Liddicoat and Crozet (2001), for instance, focused on particular aspects related to the sociopragmatic competence in French, namely the distinction between the use of French tu/vous in different informal and formal contexts, and responding appropriately to a question about the weekend in French. Other studies have examined different speech acts, such as giving and responding to compliments (Ishihara, 2004), refusals (Félix-Brasdefer, 2008) or requests and request modifiers (Codina, 2008; Li, 2012; Martínez-Flor, 2012). Finally, research conducted by Nguyen et al. (2012) and Fordyce (2014) has dealt with other pragmatic aspects rarely being examined in interventional pragmatics research, namely the speech act set of constructive criticism and epistemic stance, respectively. Results from all these studies reveal that instruction on pragmatics tends to be durable, with the only exception of some of the formal aspects of the language in Liddicoat and Crozet’s (2001) and Fordyce’s (2014) studies that were just partially sustained over time. However, none of them has focused on apologies as the targeted pragmatic feature. Additionally, with the exception of the studies by Liddicoat and Crozet (2001) and Ishihara (2004), whose delayed post-test was one year later, and Martínez- Flor’s (2012) and Fordyce’s (2014) research, whose delayed post-test was five months later, the delayed post-test in the rest of the other studies took place between two and six weeks after the treatment. It could be argued then that this is not enough time to strictly confirm the durability of instructional effects. Therefore, research that sheds light on whether the effect of apology-based instruction is retained over a substantial period of time is needed. 2.3. Purpose of the Study Considering i) the lack of research on the instructional effects of English apologies on learners from less studied L1 backgrounds (Eslami-Rasekh ELIA 16, 2016, pp. 13-48 28 DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Teaching apology formulas at the discourse level 4) an offer of repair, 5) a promise of forbearance. These five options are presented on a scale that ranges from the most (expression 1) to the least direct way of apologising (expression 5). They can be used either by themselves or by combining them. Additionally, the first two formulas can be used across all apology situations, whereas the latter three formulas are situation-specific and vary depending on the damage caused (Kondo, 2010). Formulas Examples 1. Expression of apology a. regret b. offer of apology c. request for forgiveness Sorry I apologise Pardon me 2. Explanation or account The traffic was terrible 3. Acknowledgement of responsibility a. Accepting the blame b. Expressing self-deficiency c. Showing embarrassment d. Justifying the hearer e. Expressing lack of intent f. Refusing to acknowledge It’s my fault I didn’t see you I feel awful about it You are right I didn’t mean to It wasn’t my fault 4. Offer of repair I’ll pay for the broken vase 5. Promise of forbearance It won’t happen again Table 3. Apology formulas (adapted from Olshtain and Cohen, 1983; Blum-Kulka et al., 1989) The conventional expression of apology, which is the most direct way of apologising, involves the use of performative verbs that express regret for having violated a particular social norm. This expression involves three subtypes, which are i) an expression of regret (e.g. “I’m sorry”), ii) an offer of apology (e.g. “I apologise”); and iii) a request of forgiveness (e.g. “Excuse me”). These semantic formulas are not language specific so that each language may have different ways of expressing directness by ELIA 16, 2016, pp. 13-48 29 Alicia Martínez-Flor DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 either using a performative or a combination of them. When giving an explanationof the situation, which is seen as an indirect form of apology, the speaker justifies him/herself by explaining that the cause of the offense was beyond his/her control (e.g. “The bus was late”). Acknowledgement of responsibility involves that the speaker assumes his/her fault and provides an apology for the damage caused. The various subtypes that are included within this main expression are displayed by forming a continuum which ranges from explicitly assuming the responsibility of the complaint for the offense, whereby the speaker recognises fault in causing the offense, to refusing to acknowledge the guilt. This formula involves six sub-types, namely i) accepting the blame (e.g. “My mistake”); ii) expressing self-deficiency(e.g. “I was confused”); iii) showing embarrassment (e.g. “I feel awful about it”); iv) justifying the hearer(e.g. “You’re right to be angry”); v) expressing lack of intent(e.g. “I didn’t mean to upset you”); and vi) refusing toacknowledge (e.g. “It wasn’t my fault”). An offer of repairis used when the speaker tries to offer a repair because either a physical offense or damage is done (e.g. “I’ll help you to get up”). Finally, a promise of forbearanceis employed when the apologiser promises that the offense will not be repeated (e.g. “This won’t happen again”). After classifying each apology formula on this taxonomy, a statistical analysis of the data was conducted using a version 14.0 of the Statistical Package for the Social Science (SPSS). To discern whether the differences in the two measures of pre-test and post-test, as well as pre-test and delayed post-test were significant or not, a t-test for related samples was used. 4. Results and Discussion The purpose of this study was to examine the effect of teaching EFL learners’ use of apology formulas from a discursive perspective both in the short and in the long-term. The first research question focused on the instructional effects regards the amount and type of apology formulas ELIA 16, 2016, pp. 13-48 30 DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Teaching apology formulas at the discourse level produced by learners. In relation to the amount, the analysis of the 240 apology samples (10 participants x 8 situations x 3 tests) yielded a total of 1162 apology formulas. Of these, 298 were identified in the pre-test data and 448 in the post-test data. Results from applying the statistical procedure show that the mean of apology formulas used per learner in the pre-test was much lower (M=32.60) than in the post-test (M=48.60) (see Table 4), and the difference in both cases statistically significant (p=.05). N Mean Sig. Pre-test 10 29.80 .042 Post-test 10 44.80 Table 4. Differences as regards the overall use of apology formulas in the pre-test and post-test. Sig. at p<0.05 level Additionally, a difference was also observed regarding the type of general formulas (i.e. expression of apology, explanation or account, acknowledgment of responsibility, offer of repair and promise of forbearance) being used prior to instruction (pre-test) and immediately after it (post-test). As presented in Table 5, the results from applying the statistical procedure show that direct expressions of apology were more frequent in the pre-test (M=26.50) than in the post-test (M=22.20). In contrast, the other four types of apology formulas displayed a higher frequency in the post-test (M=5.20, M=6.70, M=4.90 and M=5.80 respectively) than in the pre-test (M=1.80, M=0.50, M=1.00 and 0 occurrences respectively), all differences being statistically significant (p=.000). Table 5. Differences as regards the type of formulas for apologies in the pre-test and post-test. Sig. at p<0.05 level * t cannot be calculated because the standard deviation is 0 ELIA 16, 2016, pp. 13-48 31 Alicia Martínez-Flor DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 From the above results we can claim that implementing pedagogical intervention in the EFL classroom had a positive effect on both the amount and type of apology formulas used by learners immediately after being engaged in the instructional treatment. These findings therefore confirm previous ILP research supporting the fact that instruction does make a difference in the pragmatic realm (see Jeon and Kaya 2006; Takahashi 2010; Taguchi 2011, 2015 for a review), and are in line with previous studies that have particularly examined the teachability of the speech act of apologies in FL settings (Eslami-Rasekh et al., 2004; Eslami-Rasekh and Eslami-Rasekh, 2008; Eslami-Rasekh and Mardani, 2010; Birjandi and Derakhshan, 2013; Simin et al., 2014; Derakhshan and Eslami-Rasekh, 2015). More specifically, the present study has showed that teaching apologies following a discursive approach helped learners’ pragmatic knowledge in terms of both frequency and variety of apology formulas being used. Thus, in line with previous research that has focused on teaching refusals at the discourse level (Alcón and Guzmán, 2010; Alcón, 2012), this study has also incorporated the use of audiovisual material to offer contextually rich input that presents apologies in interaction (i.e. as in a conversation in which they need to be co-constructed over a series of turns). Additionally, the study has also included awareness-raising activities to make learners attend to this input, as well as production tasks so that learners had an opportunity to try out the newly acquired apology formulas in contextualised communicative situations. Finally, they received corrective feedback from both their peers and the teacher to provide them with an opportunity for gap noticing regarding their output. In short, the study included the three necessary conditions (i.e. appropriate input presentation, meaningful output engagement and feedback provision) for pragmatic learning to take place (Martínez-Flor and Usó-Juan, 2010). To further explore the positive benefits of teaching apologies at a discourse level, the second research question examined whether the instructional effects were maintained in the long-term, more specifically five months after the instruction had finished. Regarding the amount of apology formulas, from the total of 1162 apology formulas found in the 240 ELIA 16, 2016, pp. 13-48 32 DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Teaching apology formulas at the discourse level apology samples, 416 were identified in the delayed post-test data. Results from applying the statistical procedure show that the mean of apology formulas used per learner in the pre-test was much lower (M=32.60) than in the delayed post-test (M=45.90) (see Table 6), the difference in both cases being statistically significant (p=.000). Table 6. Differences as regards the overall use of apology formulas in the pre-test and delayed post-test. Sig. at p<0.05 level The type of general apology formulas being used prior to instruction (pre-test) and five months later (delayed post-test) was also examined. As illustrated in Table 7, the results from applying the statistical procedure show that direct expressions of apology were more frequent in the pre-test (M=26.50) than in the delayed post-test (M=21.90). In contrast, the other four types of apology formulas displayed a higher frequency in the delayed post-test (M=4.80, M=5.30, M=4.40 and M=5.20 respectively) than in the pre-test (M=1.80, M=0.50, M=1.00 and 0 occurrences respectively), all differences being statistically significant (p=.000). Table 7. Differences as regards the type of formulas for apologies in the pre-test and delayed post-test. Sig. at p<0.05 level * t cannot be calculated because the standard deviation is 0 ELIA 16, 2016, pp. 13-48 33 Alicia Martínez-Flor DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 These results seem to indicate that the learning and use of a variety of apology formulas was retained in the longer term. Previous ILP research has found that the immediate gains on different targeted pragmatic features, such as sociolinguistic competence (Lyster, 1994), compliments (Ishihara, 2004), refusals (Félix-Brasdefer, 2008), requests (Codina, 2008; Li, 2012; Martínez-Flor, 2012) or the speech act set of constructive criticism (Nguyen et al., 2012) were maintained in the delayed post-test. The present study thus supports this previous research and adds new insights by addressing a different instructional aspect, that of apologies, that was taught at a discourse level. It does, however, partly contrast with the study by Fordyce (2014) which showed that some of the initial learning of forms dealing with epistemic stance were not retained over time. It should be mentioned that the focus of instruction was a morphosyntactic aspect, which is probably less amenable to instruction, and the instructional treatment did not include practice. Additionally, the instructional treatment only lasted three 45-minute sessions based on analysing a set of four written texts. The quality of input presented to learners as well as the duration of the instructional sessions could have also exerted an influence in the partial loss of gains. Therefore, it appears that the choice of contextualised enriched input and the inclusion of production activities integrated in a well-designed pedagogical model following a discursive approach that lasts several sessions was decisive in retaining learners’ knowledge of a variety of apology strategies five months after having participated in the treatment. In order to shed more light on these findings, a more detailed and qualitative analysis of the different subtypes of apology formulas employed within each general strategy was conducted. By examining Table 8, that shows the frequency (f) and percentage (%) of apology formulas used by learners on pre-test, post-test and delayed post-test, it appears that the distribution of use was different across the three measures. ELIA 16, 2016, pp. 13-48 34 DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Teaching apology formulas at the discourse level Formula Subtype Pre-test Post-test Delayed post-test f % f % f % Expression of apology regret 262 87.92 198 44.20 195 46.88 offer of apology 31.00 16 3.57 16 3.85 request for forgiveness 0 0.00 8 1.78 8 1.92 Sub-total 265 88.92 222 49.55 219 52.65 Explanation or account 18 6.04 52 11.61 48 11.53 Sub-total 18 6.04 52 11.61 48 11.53 Acknowledgement of responsibility accepting the blame 5 1.68 13 2.90 9 2.16 expressing self-deficiency 0 0.00 9 2.00 6 1.43 showing embarrassment 0 0.00 15 3.34 11 2.63 justifying the hearer 0 0.00 7 1.55 8 1.92 expressing lack of intent 0 0.00 14 3.11 12 2.88 refusing to acknowledge 0 0.00 9 2.00 7 1.68 Sub-total 5 1.68 67 14.96 53 12.74 Offer of repair 10 3.36 49 10.94 44 10.58 Sub-total 10 3.36 49 10.94 44 10.58 Promise of forbearance 0 0.00 58 12.94 52 12.50 Sub-total 0 0.00 58 12.94 52 12.50 TOTAL 298 100.0 448 100.0 416 100.0 Table 8. Frequency of apology formulas used by learners in the pre-test, post-test and delayed post-test (n=1162). ELIA 16, 2016, pp. 13-48 35 Alicia Martínez-Flor DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Starting with learners’ use of particular apology formulas in the pre-test, it can be observed in Table 8 that the variety of formulas employed by learners was very limited. The most frequent type of formula was the direct expression of apology amounting to 88.92%. Within this formula, learners widely used the subtype of regret (87.92%), whereas the sub-type of offer of apology was scarcely used (1.00%) and no instances of request for forgivenesswere found. The remaining types of apology formulas were utilised to a much a lesser extent (in the case of explanation or account (6.04%), acknowledgement of responsibility (1.68%) and offer of repair (3.36 %) or not used at all regarding the formula of promise of forbearance. It is also worth mentioning that among the different subtypes within the formula of acknowledgment of responsibility, only that of accepting the blamewas used with no instances of the rest of the five subtypes, namely those of expressing self-deficiency, showing embarrassment, justifying the hearer, expressing lack of intentand refusing to acknowledge. This seems to indicate that prior to instruction learners mainly resorted to the isolated use of the formula I’m sorry (direct expression of regret) when apologising in different situations, independently of the politeness and sociopragmatic aspects involved in each of them. Similar findings were found in the studies by Félix-Brasdefer (2008) and Martínez- Flor (2012) on refusals and request modifiers respectively, since learners’ performance when refusing or requesting in a variety of contrasting scenarios was highly direct. The authors indicated that since learners had not been made aware of the face-threatening nature involved in those speech acts, very few instances of downgraders or mitigators that would have served to soften their responses were used. Moving on to learners’ responses in the post-test, Table 8 clearly shows a substantial increase in learners’ use of other types apology formulas. In fact, they employed all subtypes of formulas relying, therefore, not only on the explicit expression of apology. Within their use of this formula, which in total amounted to 49.55%, learners also employed the three subtypes, namely those of regret (44.20%), offer of apology (3.57%) and request for forgiveness (1.78%). Similarly, within the formula of acknowledgment of responsibility, which in total amounted to 14.96%, ELIA 16, 2016, pp. 13-48 36 DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Teaching apology formulas at the discourse level the six different subtypes were used, namely those of accepting the blame (2.90%), expressing self-deficiency (2.00%), showing embarrassment (3.34%), justifying the hearer (1.55%), expressing lack of intent (3.11%) and refusing to acknowledge (2.00%). The remaining three main types of apology formulas were also all used in the following order of frequency: promise of forbearance (12.94%), explanation or account (11.61%) and offer of repair (10.94%). These results provide useful insights with regard to the immediate instructional effects on apology formulas, since the high levels of directness decreased and learners’ preference of more indirect and elaborated apologies increased. Thus, it appears that the instruction implemented following a discursive approach positively affected learners’ apology performance, as they had the opportunity to learn how sociopragmatic and politeness aspects influence the co-construction of an apology sequence. The study by Simin et al. (2014), which focused on the development of learners’ awareness and production of apologies with the help of e-communication through exchange of e-mails, also showed that after instruction, learners’ e-mails were more elaborated and involved a combination of strategies to apologise appropriately and effectively. Finally, it was also interesting to explore whether the instructional gains were maintained in the longer term. The results from the delayed post-test (see Table 8 above) show that learners kept using the full range of apology formulas being taught, since instances from all subtypes were found. Starting with the expression of apology, which in total amounted to 52.65%, learners maintained the use of the other three subtypes, namely those of regret (46.88%), offer of apology (3.85%) and request for forgiveness (1.92%). Similarly, within the formula of acknowledgment of responsibility, which in total amounted to 12.74%, the six different subtypes were also still used, namely those of accepting the blame (2.16%), expressing self-deficiency (1.43%), showing embarrassment (2.63%), justifying the hearer (1.92%), expressing lack of intent (2.88%) and refusing to acknowledge (1.68%). The remaining three main types of apology formulas were also maintained, since all of them were still used in the following order of frequency: promise of forbearance (12.50%), ELIA 16, 2016, pp. 13-48 37 Alicia Martínez-Flor DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 explanation or account (11.53%) and offer of repair (10.58%). These findings show therefore not only immediate effects of instruction on learners’ variety of strategies when apologising, but also delayed effects on their performance as measured five months after instruction. Previous research on compliments (Ishihara, 2004) and request modifiers (Codina, 2008) also indicated that learners maintained the use of a variety of downgraders and mitigators when giving and responding to compliments as well as requesting in different communicative situations by the time of a delayed post-test. In order to illustrate the results displayed in Table 8 above, Example 1 presents learners’ performance in scenario 8 from the IDCT (Usó-Juan and Martínez-Flor, 2014) on the three different occasions: Scenario 8 A. You are the chief executive of a Hotel Group. Your group is seeking for a general manager in a recently opened hotel in London. A prerequisite to get this job is to have a Master’s Degree in Business Administration (MBA). You found that an applicant for the job, who is currently working as a receptionist in one of your hotels, has lied about having an MBA. You don’t know this person but, as one of your workers, you want to talk to him/her about this fact. You explain: B. You work as a receptionist in a worldwide hotel chain. The hotel group is seeking for a general manager in a recently opened hotel in London. Candidates are required to have a Master’s Degree in Business Administration (MBA) and although you don’t have it, you decide to lie about having this Master’s degree. The interview committee has discovered you used false educational credentials and now the chief executive of the Hotel Group, who you don’t know, wants to talk to you about this fact. You listen and respond: ELIA 16, 2016, pp. 13-48 44 DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Teaching apology formulas at the discourse level References Alcón, E. (2012). Teachability and bilingualism effects on third language learners’ pragmatic knowledge. Intercultural Pragmatics, 9, 511-541. https://doi. org/10.1515/ip-2012-0028 Alcón, E., & Guzmán, J. (2010). The effect of instruction on learners’ pragmatic awareness: A focus on refusals. International Journal of English Studies, 10, 65-80. Bergman, M. L. & Kasper, G. (1993). Perception and performance in native and nonnative apology. In G. Kasper & S. Blum-Kulka (Eds.), Interlanguage Pragmatics (pp. 82-107). Oxford: Oxford University Press. Birjandi, P. & Derakhshan, A. (2013). Pragmatic comprehension of apology, request and refusal: An investigation on the effect of consciousnessraising video-driven prompts. Applied Research on English Language, 3, 67-86. Blum-Kulka, S., House, J. & Kasper, G. (1989). Cross-cultural pragmatics: Requests and apologies. Norwood NJ: Ablex. Brown, P. & Levinson, S. C. (1987). Politeness: Some universals in language usage. Cambridge: Cambridge University Press. Codina, V. (2008). The immediate vs. delayed effect of instruction on mitigators in relation to the learner’s language proficiency in English. In E. Alcón (Ed.), Learning how to Request in an Instructed Language Learning Context (pp. 227-256). Bern: Peter Lang. Derakhshan, A. & Eslami-Rasekh, Z. (2015). The effect of consciousness-raising instruction on the pragmatic development of apology and request. TESLEJ, 18,1-24. Eslami-Rasekh, A. & Mardani, M. (2010). Investigating the effects of teaching apology speech act, with a focus on intensifying strategies, on pragmatic development of EFL learners: The Iranian context. The International Journal of Language Society and Culture, 30, 96-103. Eslami-Rasekh, Z. & Eslami-Rasekh, A. (2008). Enhancing the pragmatic competence of non-native English-speaking teachers candidates (NNESTCs) in an EFL cotext. In E. Alcón & A. Martínez-Flor (Eds.), Investigating Pragmatics in Foreign Language Learning, Teaching and Testing (pp. 178-197). Clevedon: Multilingual Matters. ELIA 16, 2016, pp. 13-48 45 Alicia Martínez-Flor DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Eslami-Rasekh, Z., Eslami-Rasekh, A. & Fatahi, A. (2004). The effect of explicit metapragmaic instruction on the speech act awareness of advanced EFL students. TESL-EJ, 8, 1-12. Félix-Brasdefer, C. (2006). Teaching the negotiation of multi-turn speech acts: Using conversation-analytic tools to teach pragmatics in the FL classroom. In K. Bardovi-Harlig, C. Félix-Brasdefer & A. S. Omar (Eds.), Pragmatics and Language Learning, 11 (pp. 165-198). Honolulu, HI: National Foreign Language Resource Center, University of Hawai’i at Manoa. Félix-Brasdefer, C. (2008). Pedagogical intervention and the development of pragmatic competence in learning Spanish as a foreign language. Issues in Applied Linguistics, 16, 49-84. Fordyce, K. (2014). The differential effects of explicit and implicit instruction on EFL learners’ use of epistemic stance. Applied Linguistics, 35, 6-28. https://doi.org/10.1093/applin/ams076 Huth, T. (2006). Negotiating structure and culture: L2 learners’ realization of L2 compliment-response sequences in talk-in-interaction. Journal of Pragmatics, 38, 2025-2050. https://doi.org/10.1016/j.pragma.2006.04.010 Ishihara, N. (2004). Exploring the immediate and delayed effects of formal instruction: Teaching giving and responding to compliments. TESOL Journal, 21, 37-70. Jeon, E. H. & Kaya, T. (2006). Effects of L2 instruction on interlanguage pragmatic development. In J. Norris & L. Ortega (Eds.), Synthetizing research on language learning and teaching (pp. 165-211). Philadelphia, PA: John Benjamins. https://doi.org/10.1075/lllt.13.10jeo Kasper, G. (2006). Speech acts in interaction: Towards discursive pragmatics. In K. Bardovi-Harlig, C. Félix-Brasdefer & A. S. Omar (Eds.), Pragmatics and Language Learning, 11 (pp. 281-314). Honolulu, HI: National Foreign Language Resource Center, University of Hawai’i at Manoa. Kasper, G. & Rose, K. H. (2002). Pragmatic Development in a Second Language. Oxford: Blackwell. Kondo, S. (2010). Apologies. Raising learners’ cross-cultural awareness. In A. Martínez-Flor & E. Usó-Juan (Eds.), Speech act performance. Theoretical, empirical and methodological issues (pp. 145-162). Amsterdam: John Benjamins. https://doi.org/10.1075/lllt.26.09kon Leech, G. (1983). Principles of pragmatics. London: Longman. ELIA 16, 2016, pp. 13-48 46 DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Teaching apology formulas at the discourse level Li, S. (2012). The effects of input-based practice on pragmatic development of requests in L2 Chinese. Language Learning, 62, 403-438. https://doi. org/10.1111/j.1467-9922.2011.00629.x Liddicoat, A. J. & Crozet, C. (2001). Acquiring French interactional norms through instruction. In K. R. Rose & G. Kasper (Eds.), Pragmatics in Language Teaching (pp. 125-144). Cambridge: Cambridge University Press. https:// doi.org/10.1017/CBO9781139524797.012 Limberg, H. (2015). Principles for pragmatics teaching: Apologies in the EFL classroom. System, 69, 275-285. https://doi.org/10.1093/elt/ccv012 Lyster, R. (1994). The effect of functional-analytical teaching on aspects of French immersion students’ sociolinguistic competence. Applied Linguistics, 15, 263-287. https://doi.org/10.1093/applin/15.3.263 Martí Arnándiz, O. (2012). Are women more polite than men? The effect of gender on the use of request modifiers by Spanish EFL university students. In L. Ruiz de Zarobe & Y. Ruiz de Zarobe (Eds.), Speech acts and politeness across languages and cultures (pp. 307-336). Bern: Peter Lang. Martínez-Flor, A. (2012). Examining EFL learners’ long-term instructional effects when mitigating requests. In M. Economidou-Kogetsidis & H. Woodfield (Eds.), Interlanguage Request Modification (pp. 243-274). Amsterdam: John Benjamins. https://doi.org/10.1075/pbns.217.08mar Martínez-Flor, A. & Usó-Juan, E. (2010). Pragmatics and speech act performance. In A. Martínez-Flor & E. Usó-Juan (Eds.), Speech act performance. Theoretical, empirical and methodological issues (pp. 3-20). Amsterdam: John Benjamins. https://doi.org/10.1075/lllt.26 Nguyen, T. T. M., Hanh, P. T. & Pham, M. T. (2012). The relative effects of explicit and implicit form-focused instruction on the development of L2 pragmatic competence. Journal of Pragmatics, 44, 416-434. https://doi. org/10.1016/j.pragma.2012.01.003 Olshtain, E. & Cohen, A. D. (1983). Apology: A speech act set. In N. Wolfson & E. Judd (Eds.), Sociolinguistics and second language acquisition (pp.18- 35). Rowley MA: Newbury House. Olshtain, E. & Weinbach, L. (1987). Complaints: A study of speech act behaviour among native and nonnative speakers of Hebrew. In J. Verschueren & M. Bertuccelli-Papi (Eds.), The pragmatic perspective: Selected papers from the 1985 International Pragmatics Conference (pp. 195-208). Amsterdam: John Benjamins. ELIA 16, 2016, pp. 13-48 47 Alicia Martínez-Flor DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Quick Placement Test. (2001). Paper and Pen Test. Oxford: Oxford University Press. Sabaté i Dalmau, M. & Curell i Gotor, H. (2007). From “sorry very much” to “I’m ever so sorry:” Acquisitional patterns in L2 apologies by Catalan learners of English. Intercultural Pragmatics, 4, 287-315. Searle, J. R. (1969). Speech acts: An essay in the philosophy of language. Cambridge: Cambridge University Press. https://doi.org/10.1017/ CBO9781139173438 Simin, S., Eslami-Rasekh, Z., Eslami-Rasekh, A. & Ketabi, S. (2014). The effect of explicit teaching of apologies on Persian EFL learners’ performance: When e-communication helps. International Journal of Research Studies in Language Learning, 3, 71-84. https://doi.org/10.5861/ijrsll.2014.661 Taguchi, N. (2011). Teaching pragmatics: Trends and issues. Annual Review of Applied Linguistics, 31, 289-310. https://doi.org/10.1017/ S0267190511000018 Taguchi, N. (2015). Instructed pragmatics at a glance: Where instructional studies were, are, and should be going. Language Teaching, 48, 1-50. https://doi. org/10.1017/S0261444814000263 Takahashi, S. (2010). Assessing learnability in second language pragmatic”. In A. Trosborg (Ed.), Handbook of pragmatics: 7 (pp. 391-421). Berlin: Mouton de Gruyter. Trosborg, A. (1995). Interlanguage pragmatics: Requests, complaints and apologies. Berlin: Mouton de Gruyter. https://doi.org/10.1515/9783110885286 Usó-Juan, E. & Martínez-Flor, A. (2014). Reorienting the assessment of the conventional expressions of complaining and apologising: From singleresponse to interactive DCTs. Iranian Journal of Language Testing, 4, 113-136. Warga, M. & Schölmberger, U. (2007). The acquisition of French apologetic behaviour in a study abroad context. Intercultural Pragmatics, 4, 221- 251. https://doi.org/10.1515/IP.2007.012 Woodfield, H. (2012). Pragmatic variation in learner perception: The role of retrospective verbal report in L2 speech act research. In C. Félix- Brasdeer & D. A. Koike (Eds.), Pragmatic variation in first and second language contexts (pp. 209-237). Amsterdam: John Benjamins. https:// doi.org/10.1075/impact.31.08woo ELIA 16, 2016, pp. 13-48 48 DOI: http://dx.doi.org/10.12795/elia.2016.i16.02 Teaching apology formulas at the discourse level First version received: June 2016 Final version accepted: October 2016