An analysis of request production by university and secondary school EFL students

ES 25 (2003-04) - pp. 167-182 AN ANALYSIS OF REQUEST PRODUCTION BY UNIVERSITY AND SECONDARY SCHOOL EFL STUDENTS1 ALICIA MARTÍNEZ FLOR Universitat Jaume I The development of the pragmatic competence has been regarded as one of the main components of the different constructs of communicative competence (Bachman, 1990; Celce-Murcia, Dörnyei and Thurrell, 1995). Among the different pragmatic aspects analysed within the field of interlanguage pragmatics, the theory of speech acts has received a great deal of attention. In this paper, we focus particularly on the speech act of requesting, as one of the most threatening-speech acts (Brown and Levinson, 1987). However, most of the studies dealing with requests have been centred on second language settings. In this respect, in this study we aim at analysing the use of not only requests but also its peripheral modification devices by two groups of English as a foreign language learners, namely those of University and Secondary School students. Results indicate that the first group produced a higher number of requests and mitigators in comparison to the second group of learners. However, a qualitative analysis showed a lack of variety in learners’ production. Thus, in line with Kasper (1997), we suggest that instruction may be necessary to foster learners’ pragmatic competence in the foreign language context. 1. INTRODUCTION Over the last years an increasing body of research has been conducted on the field known as interlanguage pragmatics, which as claimed by Kasper (1989: 42) “seeks to describe and explain learners’ development and use of pragmatic knowledge”. In this sense, several scholars have argued that becoming communicatively competent in the target language does not only imply the acquisition of grammatical and linguistic aspects, but also pragmatic and discourse abilities. For this reason, pragmatic competence has been regarded as part of learners’ communicative competence, thus integrating it in the different constructs of communication (Bachman, 1990; Celce-Murcia, Dörnyei and Thurrell, 1995). Within this field, an area that has been extensively dealt with refers to the Theory of Speech Acts developed by Austin (1962) and Searle (1969, 1976). This last author proposed a taxonomy of illocutionary acts that has been addressed by many researchers, which included five main categories, namely those of representatives, directives, expressives, commissives and declarations. In this paper, 1 This study is part of a research project funded by a grant from the Autonomous Government of the Valencian Community (GV-00-147-09) and a grant from Caixa Castelló-Bancaixa (P1.1A2000-14). ALICIA MARTÍNEZ FLOR 168 we are going to pay attention to the second category, i.e. that of directives, which have been defined as “attempts by the speaker to get the hearer to do something” (Searle, 1979: 13). To this respect, since the speaker invades the hearer’s territory, this group of speech acts has also been addressed in the Politeness Theory developed by Brown and Levinson (1987). A key notion within this theory refers to face, since politeness is regarded as an activity, which serves to enhance, maintain, or protect face. Additionally, face can be positive or negative. Positive face refers to the desire to be liked, approved of, respected and appreciated by others, whereas negative face involves the wish to maintain one’s territory unimpeded. This concept of face is closely linked to directive speech acts, since according to Brown and Levinson (1987: 60), some speech acts intrinsically threaten face and, thus, are called face-threatening acts (FTAs). Therefore, in an interaction participants must engage in some form of face-work, to which they may behave in two ways: either seeking to avoid the face-threatening act, or deciding to do the FTA. If the participants decide the last option, that is, to do the FTA, they can either go off record, in which case the participants’ communicative intent may imply more than one intention, or they can go on record expressing their intentions clearly and unambiguously. Among the group of directive speech acts, we shall focus particularly on the speech act of requesting, which has been regarded as one of the most threateningspeech acts. As claimed by Trosborg (1995: 187), a request is “an illocutionary act whereby a speaker (requester) conveys to a hearer (requestee) that he/she wants the requestee to perform an act which is for the benefit of the speaker”. In this sense, on the one hand the speaker threatens the hearer’s negative face, and on the other hand, the requester may also run the risk of losing face, as the requestee may not comply speaker’s goals. Moreover, if we pay attention to Haverkate’s (1984) distinction between impositive and non-impositive exhortative speech acts, requests belong to the former, since the speaker imposes and exerts his/her influence over the hearer in order to obtain his/her intentions. Thus, as assumed by Trosborg (1995), although it is difficult to distinguish requests from other impositive speech acts, there are two features which characterise this particular speech act, namely those of (i) the fact that the benefits are for the requester, and (ii) the imposition over the requestee involves a cost to him/her. Apart from the speech act itself, it has been claimed by Trosborg (1995) and Sifianou (1999) that the speech act of requesting is made up of two main parts: the head and its peripheral modification devices. These last ones refer to mitigation items that are divided into two main groups, namely those of internal modification and external modification. As claimed by Trosborg (1995), the use of these markers when requesting are very important, since they perform the function of softening the threatening nature of the request on the hearer. Requests have been the object of much investigation within the field of interlanguage pragmatics. In fact, there have been a considerable number of studies focusing on the speech act of requesting, both cross-sectional (Scarcella, 1979; AN ANALYSIS OF REQUEST PRODUCTION… 169 Blum-Kulka and Olshtain, 1986; Takahashi and DuFon, 1989; Svanes, 1992; Trosborg, 1995; Hill, 1997; Hassall, 1997; Rose, 2000; Safont, 2001a), and longitudinal (Schmidt, 1983; Ellis, 1992; Ohta, 1997; Achiba, 2000). Moreover, there have also been some studies devoted to analysing the effects of pragmatic transfer as far as the speech act of requesting is concerned (Blum-Kulka, 1982; House and Kasper, 1987; Faerch and Kasper, 1989; Takahashi, 1992, 1993, 1995, 1996). All these developmental studies have regarded participants from diverse nationalities and different levels of proficiency in order to examine those factors that affect learners’ acquisition of requests, although all of them suggest that further research is needed to shed more light on this pragmatic developmental pattern in foreign language settings. Moreover, regarding the analysis of request head peripheral modification devices, only a few studies have focused on its use on the part of EFL learners (Campoy and Safont, 2001; Safont, 2001b). In this sense, in order to expand the scope of investigation about the speech act of requesting in foreign language learning contexts, we present this study which deals with learners’ production of this particular speech act in two EFL educational contexts, namely those of University and Secondary School. Furthermore, we also aim at analysing to what extent EFL learners modify their requests by using mitigation devices. Taking into consideration the above assumptions, the following research questions were formulated: 1) Does learners’ educational context (whether they belong to the University or to Secondary School) influence their degree of pragmatic competence on request production? 2) Do learners employ modification items when requesting in the two different EFL settings? 2. METHODOLOGY 2.1 Subjects The subjects for the present study were 232 female students learning English as a foreign language in two different contexts, those of University and Secondary School. We did not administer any previous proficiency test in order to classify the students into different levels of proficiency, since our aim was to obtain two groups of students according to their educational setting. The first group consisted of 117 University students from Universitat Jaume I, Castellón, who studied English as a compulsory subject in their particular degree courses, since this is a distinctive feature of this university. Their ages ranged between 18 and 26 years old, the average age being 20.2 years. The second group involved 115 Secondary School students from four different Secondary Schools situated in the province of Castellón. They were selected from the two last levels of this instructional period, ALICIA MARTÍNEZ FLOR 170 namely those of 1st and 2nd Bachillerato. At both levels, English was a compulsory subject of their syllabi, and students attended English classes three hours per week. The age of this group of Secondary School students ranged between 15 and 18 years old, the average age being 16.3 years. 2.2 Instrument In order to examine our subjects’ production of requests, and the use of modification devices softening them, we employed a written production test, which has also been called Discourse Completion Test (DCT) in the literature. According to Sasaki (1998: 458), these production tests “present a situation where a certain kind of speech act is expected, and the respondents are asked to provide what they think would be appropriate in that situation”. Concerning the use of the DCT as a research method in the field of interlanguage pragmatics, several authors (Kasper and Dahl, 1991; Rose, 1994; Sasaki, 1998) have attributed certain limitations to this instrument, as it has been regarded as too artificial by representing a test-like situation more than a real-life one. However, according to Houck and Gass (1996) and Beebe and Cummings (1996), the DCT is a useful tool which allows for a large amount of data collection in a relatively short period of time. Thus, for the purposes of our study, which was conducted in six different degrees at Universitat Jaume I, and in four Secondary Schools, the DCT was the most suitable research method to obtain information from a large sample of respondents. We should also mention that the DCT employed in our study was created by the LAELA2 research group in order to conduct research in the field of interlanguage pragmatics. It consisted of 20 different situations that elicited learners’ use of three speech acts, namely those of requesting, suggesting and advising. However, for the purposes of the present study we only took into consideration those eight situations regarding the speech act of requesting (see Appendix A). 2.3 Procedure When the Discourse Completion Test was distributed among the different learners, the researcher of the present study explained to them that the English department of Universitat Jaume I was interested in obtaining some information about students’ performance in a particular test. They were also told what to do and how to do it, but they were not told at any moment that they were taking part in a research investigation. They had 50 minutes to answer the test and they could ask about any doubts they had regarding vocabulary. 2 LAELA stands for “Lingüística Aplicada a L’ensenyament de la Llengua Anglesa” (Linguistics Applied to English teaching). AN ANALYSIS OF REQUEST PRODUCTION… 171 By means of this DCT, we analysed participants’ production of requesting both quantitatively and qualitatively. Firstly, we examined and counted the amount of requests that were produced in an appropriate way to the situation in which this particular speech act was required. Secondly, we paid attention to the type of linguistic formulae the participants had employed to express the speech act of requesting. In order to perform this qualitative analysis, we focused on Trosborg’s (1995) taxonomy for the speech act of requesting, since it is built on previous research within the area of pragmatics (Austin, 1962; Searle, 1969, 1976), as well as on the basis of Brown and Levinson’s (1987) politeness model distributed into on-record and off-record strategies. This typology has also been reformulated following the studies carried out by House and Kasper (1981), Blum-Kulka and Olshtain (1984), and Safont (2001a). As can be observed in Table 1 below, Trosborg’s (1995) taxonomy is mainly divided into the strategies of indirect requests, conventionally indirect requests (either hearer-oriented or speaker-based), and direct requests. Moreover, for the purposes of our study, we have decided to include an extra group of other types of strategies in case these might occur. Table 1. Taxonomy of request realisation strategies (adapted from Trosborg, 1995: 205) TYPE STRATEGY EXAMPLE Indirect Hints Statement Ability Could you...? Conventionally Can you...? Indirect Willingness Would you...? (hearer-oriented) Permission May I...? Suggestory Formulae How about...? CONVENTIONALLY INDIRECT Wishes I would like... (speaker-based) Desires/needs I want/need you to... Obligation You must... You have to... DIRECT Performatives I ask you to... Imperatives Lend me your car. Elliptical phrase Your car. Other types of strategies Additionally, we also took into account whether learners employed peripheral modification devices softening their production of requesting. To this respect, we also considered Trosborg’s (1995) taxonomy of request modification items, which is distributed into internal and external modifications (see Table 2). The use of these mitigators is important, since they represent different ways of modifying a request in order to make it more polite and decrease its threatening nature. ALICIA MARTÍNEZ FLOR 172 Table 2. Taxonomy of request modification items (based on Trosborg, 1995: 209-219) TYPE SUBTYPE EXAMPLE Syntactic downgraders I wonder if you would be able to pass me the bread Internal modification Lexical/phrasal downgraders Could you pass me the bread, please? Upgraders I’d be very grateful if you’d pass me the bread. Preparators Are you busy right now? I need that you pass me the bread. External modification Disarmers I’m sorry to trouble you, but could you pass me the bread? Supportive reasons Could you pass me the bread? I cannot reach it. 3. RESULTS AND DISCUSSION As stated before, our first research question focused on the analysis of EFL learners’ production of requesting in two different learning contexts. Figure 1 below presents the comparison of both groups of students. Figure 1. University and Secondary School students’ production of requests As can be observed, University students produced more appropriate requests than Secondary School students. Thus, the amount of request realisation strategies performed by University students amounted to 53.19%, whereas Secondary School students’ percentage of request formulae was 46.81%. In this sense, we may Request production 53,19 46,81 University students Secondary School students AN ANALYSIS OF REQUEST PRODUCTION… 173 assume that learners’ educational context might have influenced learners’ degree of pragmatic request production. Additionally, these findings may also indicate, in line with previous studies which have focused on proficiency effects (Takahashi and DuFon, 1989; Trosborg, 1995; Hassall, 1997; Hill, 1997), that students with a high level of proficiency, that is, the group which belonged to the University, performed better than those students from a lower level, namely those from the Secondary School setting. Moreover, apart from examining the number of requests produced by each group of students in quantitative terms, we shall now go on to pay attention to the type of linguistic strategies employed by our participants in the two different EFL environments in qualitative terms. In order to provide a more detailed account of the most frequently types used by learners in the two EFL contexts, we considered the distribution into indirect, conventionally indirect and direct strategies as can be observed in Figure 2. 0 10 20 30 40 50 60 70 80 90 100 Indirect Convent. indirect (heareroriented) Convent. indirect (speakerbased) Direct Other 3,88 79,84 5,71 6,32 4,25 1,4 4,36 7,86 4,92 81,46 University students Secondary School students Figure 2. Comparison of University and Secondary School students’ request strategies Figure 2 offers the comparison between University and Secondary School students’ percentage of the specific request linguistic realisations used according to the different strategies employed: indirect, conventionally indirect (both heareroriented and speaker-based), direct, and the use of other types of strategies. As can be seen, no relevant differences may be observed, since all types of request formulae were employed in almost a similar percentage by both groups of EFL students. The type most frequently employed by all students involved the conventionally indirect strategies oriented to the hearer, which amounted to 79.84% in University students, and to 81.46% in the case of Secondary School students. ALICIA MARTÍNEZ FLOR 174 This result is in line with previous studies focusing on EFL learners’ use of the speech act of requesting, which also pointed out students’ tendency to use more conventionally indirect strategies (Trosborg, 1995; Safont, 2001b). Furthermore, Table 3 below shows a more detailed analysis of the different request formulae performed by both University and Secondary School students following Trosborg’s (1995) taxonomy of request realisation strategies. Table 3. Comparison of University and Secondary School students’ request strategy types3 University students Secondary School students TYPE STRATEGY STRUCTURE O P M O P M Indirect Hints Statement 32 3.88% 10 1.4%. 6 Ability Could you...? 301 36.58% 231 170 23.88% 105 Conventionally Can you...? 223 27.10% 166 366 51.4% 96 indirect Willingness Would you...? 133 16.16% 59 41 5.76% 7 (heareroriented) Permission May I...? 3 0.42% Suggestory Formulae How about...? Wishes I would like... 6 0.73% 1 Conventionally indirect Desires/Needs I want/need you to... 41 4.98% 31 4.36% (speaker-based) Obligation You must... 2 0.24% 5 0.70% You have to... 4 0.49% Direct Performatives I ask you to... Imperatives Lend me your car. 46 5.59% 21 51 7.16% 2 Elliptical phrase Your car. Other types of strategies Questions 35 4.25% 2 35 4.92% TOTAL 823 100% 480 712 100% 210 The specific request linguistic realisations most frequently employed by University students referred to ability and willingness strategies. Thus, ability strategies were divided into two structures depending on whether they involved the use of the modal verb could (36.58%) or the modal verb can (27.10%). Regarding willingness strategies, their use amounted to 16.16%, although it is worth mentioning that University students did not only employ the expression Would you ...?, but also more elaborate structures with the correct use of Would you mind + Ving ...? or Would you be so kind as to ...?: 3 This table illustrates the different request strategy types displaying the information in three columns. Thus, the first column marked with an “O” refers to the Occurrences found of this particular speech act. The second column represented by a “P” consists of the Percentage of that particular strategy used. Finally, the third column addresses the Mitigation employed when requesting, and is represented by an “M”. AN ANALYSIS OF REQUEST PRODUCTION… 175 Example (1) - Situation 4 (see Appendix A) Two women who do not know each other are sitting together on a train and it is a non-smoking area. One of the women starts smoking. The other woman says: Would you be so kind as to stop smoking, please? The other request realisation strategies involved the use of imperatives (5.59%), structures indicating desires/needs (4.98%) or hints (3.88%), the others being less than 1%. Moreover, we should also mention that no instances of permission, suggestory formulae, performatives and elliptical phrases were found. Finally, in the group we added to include other types of strategies, University students produced questions to express requests, amounting to 4.25%. Focusing on Secondary School students’ use of particular request linguistic formulae, we may observe in Table 3 above that a total of 75.28% of the strategies employed were related to ability modal verbs distributed into the use of can, which amounted to 51.4% (more than half of the overall strategy use), and the use of could, which accounted for 23.88%. Furthermore, it is also worth noticing that the willingness strategies, which amounted to a 5.76%, were only expressed by the structure Would you ...?. Thus, in contrast to University students, who attempted to use more elaborate willingness structures, the Secondary School students who tried to employ the expression Would you mind + V-ing ...? failed to used it correctly, as the verb was not in the gerund form, and none of the Secondary School students employed the expression Would you be so kind as to ...?. However, like University students, Secondary School students also produced questions, amounting to 4.92%. Example (2) - Situation 1 (see Appendix A) You arrive in Zaragoza and go to the hotel. You want to know what number your room is. You say to the receptionist: What is the number of my room, please? Although differences have been found between the two EFL settings (which might have been due to proficiency effects), our findings are in line with previous studies that analysed EFL learners’ production of requests (Safont, 2001b). According to Safont (2001b) and Safont and Alcón (in press), the most frequently strategies employed by EFL learners referred to ability and willingness strategies, that is, those belonging to the group of conventionally indirect strategies. This use might have been due to the type of input learners receive in the classroom, either by teachers’ output or by FL materials. Apart from studying the head, that is the request itself, we were also interested in ascertaining the peripheral modification items employed with them. Thus, the second research question pointed to whether learners employ modification devices when requesting. As can be observed in Table 3 before, the third column represented by an “M” (Mitigation) addresses the number of mitigators used with