Cross-cultural and situational variation in requesting behaviour:

Cross-cultural and situational variation in requesting behaviour: Perceptions of social situations and strategic usage of request patterns Maria Economidou-Kogetsidis * Department of Languages and Literature, School of Humanities, Social Sciences and Law, University of Nicosia, 46 Makedonitissas Avenue, 1700 Nicosia, Cyprus 1. Introduction One of the fundamental tenets of pragmatics is that utterances and verbal communication in general should be interpreted based on the sociocultural context (Blum-Kulka, 1997; Kasper and Rose, 2001) and that ‘‘context is crucial in understanding language’’ (Becker et al., 1989:521). A number of studies (Becker et al., 1989; Blum-Kulka and House, 1989; Trosborg, 1995; Le Pair, 1996; Fukushima, 1996, 2000; Ballesteros Martin, 2001, 2002; Kwong, 2004 to name but a few) have demonstrated how speakers’ speech acts and the degree of directness/indirectness they employ in a specific situation are indeed influenced by certain social/contextual variables. The most widely discussed and tested variables are the social variables of social distance (D), social power (P) and imposition (R) of the requested act, proposed by Brown and Levinson’s (1978, 1987) influential model of politeness. Brown and Levinson’s main argument (1978, 1987) is that the social variables of D, P and R are the most important factors influencing speakers’ linguistic choices. The authors maintain that these factors combine in an additive fashion, thus the bigger the hearer’s power, social distance and degree of imposition of the act, the greater the face threat will be and the greater the degree of indirectness to be employed by the speaker. In other words, according to Brown and Levinson (1978, 1987), there is a positive correlation between these variables and the degree of indirectness employed. Journal of Pragmatics 42 (2010) 2262–2281 A R T I C L E I N F O Article history: Received 19 May 2009 Received in revised form 28 January 2010 Accepted 1 February 2010 Keywords: Familiarity Power Imposition Greek Requests Directness A B S T R A C T Brown and Levinson (1978) and Brown and Levinson (1987) is that social distance, power and imposition are the most important factors influencing speakers’ linguistic choices and that there is a positive correlation between these social variables and the degree of indirectness employed. The present study investigates the relationship between social/ situational and cultural factors and native and non-native speakers’ requesting behaviour. It examines whether the requestive strategies of English native speakers and of Greek ESL learners follow a similar trend across different social situations. It additionally tests whether thesame social situations are perceived and rated similarly by the two groups and the extent to which the speakers’ directness is influenced by familiarity, power and imposition. Results have shown that there are high levels of cross-cultural agreement between the two groups for trends of situational variation but also some cross-cultural disagreement on the specific directness levels employed for given situations. Even though significant differences in the speakers’ assessment of social reality seems to explain the differences in their linguistic choices to some extent, this study argues thatpower, familiarity and imposition alone do not always decide speakers’ directness. A number of intervening situational and cultural factors need to be taken into account when interpreting speakers’ linguistic choices. 2010 Elsevier B.V. All rights reserved. * Tel.: +357 22461566x105; fax: +357 22357665. E-mail addresses: [email protected], [email protected]. Contents lists available at ScienceDirect Journal of Pragmatics journal homepage: www.elsevier.com/locate/pragma 0378-2166/$ – see front matter 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.pragma.2010.02.001 However, in the light of newempirical evidence and criticisms of their model since the time of their 1st edition (1978), the authorsmodify their claimthat these three factors ‘‘subsumeall others that have a principal effect on such assessment’’ (Brown and Levinson, 1978:85). The authors admit that ‘‘theremay be a residue of other factorswhich are not capturedwithin the P, D, and R dimensions’’ (Brown and Levinson, 1987:16) and refer to ‘‘liking’’ and the presence of an audience as two such factors. A large number of empirical studies, some of which will be discussed below, have provided support for the relationship existing between these social variables and the speakers’ choice of requesting strategies and/or modification, while a number of others have not. Some studies have focused more on other situational/contextual factors and argued that factors such as speakers and hearers’ rights and obligations (Blum-Kulka and House, 1989; Held, 1996), the estimated likelihood for compliance, the estimated difficulty in making the request (Blum-Kulka and House, 1989), the magnitude of face threat (Baxter, 1984), the urgency of request (Held, 1996), the speaker’s embarrassment (Held, 1996), the request goal, the setting and the medium (Blum-Kulka et al., 1985), the interlocutors’ age (Rintell, 1981; Blum-Kulka et al., 1985; Held, 1996) and sex (Blum-Kulka et al., 1985), and the formality/informality of the situation (Felix-Brasdefer, 2006), can be equally or more influential factors. Nevertheless, what is sometimes ignored by studies is that some of these factors can, in fact, be subsumed under Brown and Levinson’s (1978, 1987) P, D and R factors. Right and obligations, for instance, can be subsumed under P and possibly R (see sections 3.2 and 3.3), while the magnitude of face threat is, according to Brown and Levinson (1978, 1987), the sum total of P + D + R. To make things more complicated, the way people perceive and assess such social variables is not only situation dependent but also culturally specific (Brown and Levinson, 1987:248) as ‘‘members of different cultures might differ in their perceptions of social situations as well as in the relative importance attributed to any of the social parameters mentioned’’ (Blum-Kulka and House, 1989:137). Spencer-Oatey (1993) warns researchers that ‘‘different sociocultural groups may hold differing norms regarding the power and distance of a given role relationship’’ and argues that, ‘‘whenever possible, researchers should include assessment checks of people’s perceptions of these variables’’ (1993:28). The present study is concerned with ‘‘the relationship between situational and cultural factors in determining variable patterns of requestive behaviour’’ (Blum-Kulka and House, 1989:123). As such, it investigates the requesting behaviour and directness choices of Greek learners of English (henceforth GLs) and of British English native speakers (henceforth ENSs) across a number of socially different situations. Using data collected from a discourse completion test and interviews, it examines whether the requestive strategies of native and non-native speakers follow a similar trend across different social situations, and the extent to which cultural factors interact with social/situational factors. It additionally tests whether the same social situations are perceived and rated similarly by the two language groups, as far as power, familiarity and imposition are concerned. A further aim of this study is to trace the possible links between these emic perceptions of social situations and the observed strategic usage of request patterns. In order to do this, the contextual parameters of power, distance and imposition are correlated with the three types of request strategies: most direct, conventionally indirect and non-conventionally indirect. The research questions of this study are therefore as follows: (a) To what extent do GLs and ENSs agree on the variation of their requestive behaviour according to the social situation? (b) Do Greek learners (GLs) and British English native speakers (ENSs) perceive the same social situations differently in terms of (a) social power (b) familiarity and (c) imposition of the requested act? (c) To what extent is the GLs’ selection of requesting strategies determined by the social variables of familiarity, power and imposition?—Are there any possible links/correlations between their perceptions of social reality and the observed usage of request patterns? In the sections below, I firstly review some of the empirical findings in relation to the social variables of D, P and R and in relation to other situational/contextual variables, and then offer a short discussion of Greek speakers’ directness and politeness orientation. Then I explain how the variables of D, P and R are used and defined in this study. The results and a discussion of results are presented in the sections that follow. 2. Background 2.1. Relationship of variables and speech act realisation 2.1.1. Power (P) According to Brown and Levinson’s predictions (1978, 1987), the bigger the face threat (computed by the three variables D, P and R), the higher the number of the strategy to be employed. In Kasper’s words (1994:3209), there is ‘‘a positive correlation between the weight of contextual factors (social distance, power and imposition) and politeness investment’’. Brown and Levinson’s predictions have been both confirmed and rejected by various studies concerned either with requests or with other speech acts. Cherry (1988), for instance, examined a number of letters written by academics to the president of an American university and found that power was not influential on the subjects’ requesting strategies. Similarly, McLaughlin et al. (1983) found that relationship dominance had no influence on offender accounts. Unlike Cherry (1988) and McLaughlin et al. (1983), however, Baxter (1984) found that ‘‘persons with power used less politeness than less powerful persons’’ (1984:427). Her study additionally found that gender emerged as an also relatively powerful predictor of M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 2263 politeness strategies thus supporting the view that other variables can be more influential than distance, power and imposition, and that these three variables alone do not always determine speakers’ requesting strategies. Similar results regarding power were produced by Blum-Kulka et al.’s study (1985), which investigated requests in Israeli society. From all the variables considered, the most important predictors of variance were the relative power of the speaker and the request goal. Blum-Kulka and House (1989), examining the social parameters influencing the choice of directness level in Hebrew, German and Argentinean speakers, also found that one of the most important factors influencing a request’s level of directness was the speaker’s relative dominance. Yet, importantly, Blum-Kulka and House (1989) identified a number of additional influential factors. These were the hearer’s degree of obligation to carry out the act, his/her right to demand compliance and the estimated likelihood of compliance. The authors also distinguish between ‘‘context external’’ and ‘‘context internal’’ factors (Blum-Kulka and House, 1989:130–131, taken from Brown and Fraser, 1979) which explained their subjects’ linguistic choices in certain social situations. ‘‘Context external’’ factors include the categories of social distance, social power and participants’ rights and obligations. ‘‘Context internal’’ factors, on the other hand, refer to factors ‘‘which are directly related to its requestive nature, such as the type of request goal, the degree of imposition involved for the speaker relative to the specific goal, and the prerequisites needed for compliance’’ (Blum-Kulka and House, 1989:131). Scarcella and Brunak (1981) investigated the effect of status in adult first and second language performance and found that high-level L2speakers, like native speakers, usedmore negative politeness towards superiors. Similarly, BrownandGilman (1989)who examinedShakespeare’s tragedies alsofoundthe power variable to influence the speakers’ requestiveproduction.A similar finding was received by Holtgraves and Yang’s study (1990) on requests produced by American and Korean speakers, while Cansler and Stiles (1981), Holtgraves et al. (1989), Lustig and King (1980) also found support for the influence of power. Even more recent studies received support for Brown and Levinson’s (1978, 1987) claim. Aeginitou’s study (1994) on facework and politeness in the Greek EFL classroom, found power to be a determinant of politeness moves. Fukushima’s study (2000), on requests in English and Japanese, also revealed positive correlations between the variable of power and choice of requesting strategies in a number of socially different situations. Similarly, Trosborg (1995) who analysed aspects of interlanguage pragmatics in Danish learners of English found that native speakers of English used significantly more indirect strategies (hints) when interacting with authority figures (1995:276), and Kwong’s study (2004), investigating refusals produced by Korean native speakers and American English speakers, showed that Korean speakers tended to take a more mitigating approach in dealing with a higher status person, unlike English speakers who did not seem to be particularly sensitive to status differences. 2.1.2. Distance (D) The variable of distance (D) ‘‘is the one which has received the most contradictory results, as well as criticisms’’ (Fukushima, 2000:76). Some studies supported Brown and Levinson’s (1978, 1987) prediction regarding D, while a number of others did not. More specifically, Trosborg (1995), in the study mentioned above, found her English participants to employ fewer hints and more conventional indirectness with friends than with strangers. Similarly, Le Pair (1996), confirmed that the Spaniards of his study employed more face-saving strategies of the negative polite type in situations where there was a relatively high social distance between the interlocutors, while they exhibited a preference for direct strategies in those cases where there was no or little social distance between the interlocutors. More recent studies, on the other hand, mostly concernedwith Spanish speakers, showed howhigher requesting directness is expected between status-equal friends (-D). This was found in Felix-Brasdefer’s (2006) results on refusal strategies among male speakers ofMexican Spanishwhich showed that ‘‘social power and social distance are conditioning factors in the selection of linguistic strategies’’ (2006:2177) in the Mexican community examined. Brown and Levinson’s (1978, 1987) predictions were also confirmed in Felix-Brasdefer’s study (2005) concerning indirectness and politeness in Mexican Spanish and in Ma´rquez Reiter’s (2000, 2002) publication on linguistic politeness in Britain and Uruguay. Ma´rquez Reiter’s (2000, 2002) Spanish males and females framed their requests according to the social distance so that more directnesswas employedwhen social distance was smaller. Diaz Perez (1999) similarly found that the higher the degree of imposition and/or social distance, themore conventionally indirect strategies her English and Spanish subjects employed. Finally, BallesterosMartin (2001, 2002, cited in Ma´rquez Reiter and Placencia, 2005:178) showed how ‘‘mitigated requests in Spanish tended to be employed when there was social distance between the interlocutors and when the speakers considered that it was their right to request something fromthe addressee. By the same token, if the speaker has more social power than the hearer and more of a right to ‘demand’, he observed that he/she is less likely to make use of mitigation’’ (Ma´rquez Reiter and Placencia, 2005:178). Contrary to the above findings, however, Fukushima’s results (2000) in the study mentioned above, did not confirm the positive correlation between distance and directness. Her results revealed negative correlations between the two, indicating that the bigger the social distance between the speaker (S) and the hearer (H), themore direct the requesting strategies became. Such partial confirmation of Brown and Levinson’s prediction was similarly received in Holtgraves and Yang’s study (1990). Eventhoughthey founddistance tobe associatedwiththe requestsproducedbyAmericanandKorean speakers, they also found that bald-on-record requestswere perceived by bothAmericans and Koreans as indicating the greatest distance rather than the leastdistance. Similarly, Baxter (1984) in the same study of compliance-gaining aspolitenessmentionedabove, foundcloseness to be associated with politeness strategies but ‘‘in a manner opposite to that expected by Brown and Levinson’’ (Baxter, 1984:453): direct strategies were given higher likelihood ratings in distant relationships than in close relationships. Blum-Kulka and House (1989), on the other hand, in their study mentioned above, found that the relative social distance between the interlocutors was not a determinant factor for the subjects’ requestive directness. Similar results were obtained 2264 M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 in Blum-Kulka et al.’s study (1985) on the language of requesting in Israeli society. The authors found that neither sex nor the relative social distance had a strong predictive value. Brown and Gilman (1989), who analysed Shakespeare’s tragedies, also came to the conclusion that distance had no effect on politeness behaviour, while McLaughlin et al. (1983), contrary to Brown and Levinson’s thinking, also showed that intimacy did not predict that level of politeness in offender accounts. 2.1.3. Imposition (R) As far as the effect of imposition (R) on speakers’ linguistic strategies is concerned, Brown and Gilman (1989) found rating of imposition to be consistent with Brown and Levinson’s prediction of politeness. Similarly, McLaughlin et al. (1983) received support for the weighting of imposition in managing failure events by finding politeness to increase with the severity of the social offence. Trosborg (1995:277) found hints to be employed more in situations in which the requested service was felt to be special and carried a high degree of imposition. Fukushima (2000) similarly received positive correlations between the choice of requesting strategies and the variable of imposition in half of the situations studied. More recently, Schauer’s (2007) longitudinal study of German EFL university students indicated that her three participating groups (British English native speakers, study abroad German learners of English and ‘at home’ German learners of English) all used ‘‘a considerably larger variety of external modifiers in situations involving a high imposition request’’ (2007:211). However, in Baxter’s study (1984), imposition provided no predictive force in accounting for politeness while Hartford and Bardovi-Harlig (1996) also report mixed results in the email requests fromstudents to university lecturers. The authors found that lowand high levels of impositionwere found in both negative and positive affect requests, thus coming to the conclusion that imposition is not a factor that strongly affects the evaluation of the requests of the native speakers. However, thiswas not the case for non-native speakers. Their study showed that levels of imposition may indeed play a role in the evaluation of nonnative speakers’ requests. Finally, Rose’s (2000) cross-sectional study of the interlanguage development of young Cantonese learners of English revealed little evidence of sensitivity to situational variation related to imposition and social status. The exact relationship between the social variables and the degree of directness/indirectness employed by speakers is still very much under investigation. The same can also be said for the degree of influence that these variables have on the speakers’ linguistic choices. What is also important, however, is that the above variables, as well as other factors, often interact with each other in any given social situation. Finally, what the exact components of D, P and R precisely are, is also a subject for discussion. In section 3, I consider the variables of social power, familiarity (rather than social distance) and imposition from the viewpoint of their components and explain how these variables are defined and used throughout the present study. 2.2. Greek and English speakers: directness and politeness Numerous cross-cultural studies have indicated that there is variation in the speech-act performance between two different speech communities, especially in relation to the level of directness of their request realisation. Such differences have been observed between the English and the Greek interactional styles, with Greek speakers being found more direct in their speech behaviour than British English and American English speakers (Tannen, 1981; Sifianou, 1992a; Economidou- Kogetsidis, 2003, 2005). According to Sifianou (1992a), on the verbal level, requests, wishes, advice and suggestions among Greek members of the in-group ‘‘are expressed structurally more directly than in English because they are not perceived as impositions to the same extent’’ (1992a:42). Such high directness is particularly evident in certain requests that are not facethreatening in the Greek culture, especially in contexts such as in shops, on the phone or in the family (Sifianou, 1992a,b; Pavlidou, 1994; Antonopoulou, 2001; Economidou-Kogetsidis, 2005). In such contexts, the degree of familiarity is either very high (i.e. among family members), the participants’ roles (and therefore rights and obligations) are culturally and situationally fixed and specific (‘standard’ situations (House, 1989)), or the result benefits the addressee (i.e. in service encounters) (Antonopoulou, 2001; Economidou-Kogetsidis, 2005). According to Antonopoulou (2001:246), in such contexts, mitigating devices are ‘‘hardly necessary to mitigate impositions, as there are no impositions involved’’. It has been argued that this tendency for higher directness on the part of Greek speakers is consistent with the ethos of directness of the Greek society and its preference for informality, spontaneity, solidarity, in-group membership and overt expressions of feelings (Vassiliou et al., 1972; Triandis and Vassiliou, 1972). Following Brown and Levinson’s well-known distinction between positive and negative politeness (1978, 1987), the Greek culture has importantly been found to favour positive politeness (Sifianou, 1992a, 2001 [1989]; Pavlidou, 1994, 1997, 1998; Makri-Tsilipakou, 2001; Economidou- Kogetsidis, 2003, 2005), particularly in the personal sphere. Such positive politeness is often manifested through the Greek speakers’ extensive use of bald-on-record strategies (e.g. imperative or elliptical requests), their emphasis on external rather than internal request modification (in the case of Greek learners of English—cf. Economidou-Kogetsidis, 2008, 2009) and the wide use of other linguistic devices (i.e. in-group markers) that promote solidarity and intimacy. The British culture, on the other hand, has been characterized as a culture with a negative politeness orientation, favouring tact, individuality, avoiding impositions and redressing threats to face (Scollon and Scollon, 1983; Wierzbicka, 1991; Sifianou, 1992a, 2001 [1989]; Economidou-Kogetsidis, 2003, 2004, 2005). According to Sifianou (1992a) who compared politeness phenomena in England and Greece, ‘‘the English seem to place a higher value on privacy and individuality’’ (1992a:41) (i.e. the negative aspect of face) and English requests are more sparingly employed as they are perceived as impositions to a greater extent. As a result, English requests are preferably expressed more elaborately and indirectly and higher restrictions are placed on bald-on-record strategies. English native speakers tend to acknowledge the M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 2265 addressee’s autonomy by inviting them to comply with the request and minimize impositions typically through the use of conventional indirectness and internal mitigating devices (Blum-Kulka, 1989). Following the different cultural orientation and linguistic preferences above, it is expected that the GLs’ requesting production in the present study will also be characterised by a high degree of directness. 3. Defining the social variables The terminological inconsistency associated with the above variables makes it essential that we clarify the way that the above variables can be defined before proceeding with the discussion of their use. A number of authors (Wierzbicka, 1991; Spencer-Oatey, 1996; Fukushima, 2000) have emphasised the importance for precise use of the above terminology, as crosscultural pragmatic studies often ‘‘use the same terms with different meanings, or different terms with the same meaning, and that the parameters are rarely explicitly defined’’ (Spencer-Oatey, 1996:1). 3.1. Familiarity (F)—a definition A number of different labels are often used for this dimension, perhaps the most common ones being: ‘‘(social) distance’’ (Brown and Levinson, 1978, 1987; Blum-Kulka et al., 1985, 1989; Fukushima, 2000; Holtgraves and Yang, 1990; Thomas, 1995; Ma´rquez Reiter, 2002), ‘‘solidarity’’ (Brown and Gilman, 1972), ‘‘familiarity’’ (Olshtain, 1989; Fukushima, 1990), ‘‘closeness’’ (Baxter, 1984) (for an extended list and discussion on the different labels used see Fukushima, 2000; Spencer- Oatey, 1996). In many of the definitions used in the studies motioned above, ‘familiarity’ seems to be a component of social distance rather than an identical term to social distance. Spencer-Oatey (1996) summarises that ‘‘authors have variously interpreted distance as comprising one or more of the following (often overlapping) components: - Social similarity/difference (e.g. Brown and Gilman, 1972 [1960]. - Frequency of contact (e.g. Slugoski and Turnbull, 1988). - Length of acquaintance (e.g. Slugoski and Turnbull, 1988). - Familiarity, or how well people know each other (e.g. Holmes, 1990). - Sense of like-mindedness (e.g. Brown and Gilman, 1972 [1960]). - Positive/negative affect (e.g. Baxter, 1984)’’. (Spencer-Oatey, 1996:7) The present empirical study uses the term ‘‘familiarity’’ (F) instead of ‘‘social distance’’ as it is felt that the term ‘‘familiarity’’ was easier to understand by the participants of the study who were asked to fill in the questionnaire. The term ‘familiarity’ as used in the present study refers to closeness and intimacy. It follows part of Fukushima’s definition of ‘social distance’ as it is presented in her study (2000:87), with the ‘affect’ factor taken out. Fukushima’s (2000) explains that social distance can be determined by (a) whether people are similar or different, (b) how well they know each other, and (c) whether they like each other (affect) (Fukushima, 2000:87). Social variables such as age, social class, ethnicity, occupation, sex, beliefs, and value systems are some of the components that account for whether people are similar or different, while length of acquaintance, frequency of contact and the amount of self-disclosure are the components that account for how well people know each other (Fukushima, 2000:87). Not all of these components need to exist in a relationship. Familiarity might exist because of one or more of the above components and the degree of importance of each component may vary not only situationally but, more importantly, cross-culturally. I would agree with Fukushima’s definition above regarding the term ‘‘familiarity’’ but would argue that familiarity is not, in my view, determined by affect. Referring to the example of two cousins who grew up together, have known each other for years, live near each other (and see each other often), one could claim that even though both parties are similar in age, ethnicity, beliefs, social class, they might still not like each other very much because they are so different. The degree of familiarity between the two remains extremely high regardless of the fact that there is no affect. 3.2. Social power—a definition As with distance, social power also carries the problem of terminological inconsistency. A number of studies have again used different terms for this dimension, the most common ones being: ‘‘(social) power’’ (Brown and Levinson, 1978, 1987; Blum-Kulka et al., 1985; Craig et al., 1986; Olshtain, 1989; Holtgraves and Yang, 1990; Aeginitou, 1994; Fukushima, 2000; Ma´rquez Reiter, 2002), ‘‘status’’ (Scarcella and Brunak, 1981; Craig et al., 1986; Becker et al., 1989; Bardovi-Harlig and Hartford, 1990; Rose, 2000; Nelson et al., 2002; Kwong, 2004) and ‘‘dominance’’ (Blum-Kulka and House, 1989; Trosborg, 1995) (for an extended list and discussion on the different labels used see Fukushima, 2000; Spencer-Oatey, 1996). In their interpretations of the ‘vertical’ dimension of interlocutor relations, authors have emphasised one or more of the following aspects: - Power of control (e.g. Brown and Gilman, 1972 [1960]; Brown and Levinson, 1987 [1978]). - Social status or rank (e.g. Cansler and Stiles, 1981). 2266 M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 - Authority, or the legitimate right to exert influence (e.g. Leichty and Applegate, 1991). - A general notion of equality–inequality (e.g. Holtgraves, 1986). (Spencer-Oatey, 1996:11) The conception of ‘‘power’’ in the present study follows Brown and Gilman’s (1972 [1960]) definition below which emphasises control of another person’s behaviour, rather than social rank or social status. ‘‘One person may be said to have power over another in the degree that he is able to control the behaviour of the other. Power is a relationship between at least two persons, and it is nonreciprocal in the sense that both cannot have power in the same area of behaviour. There are many bases of power-physical strength, wealth, age, sex, institutionalised role in the church, the state, the army or within the family’’ (Brown and Gilman, 1972:255). The components of power include such factors as social status, social class, institutionalised role, age, sex, wealth, physical strength, regional or ethnic identity (Fukushima, 2000). As a consequence of these, someone has power over someone else, so the notion of equality-inequality arises. ‘‘If someone has power over someone else, s/he can control the other to some extent. S/he has authority or the legitimate right to exert influence’’ (Fukushima, 2000:85). What is therefore also added to this conceptualisation of power is the notion of legitimate right to exert influence, as proposed in Leichty and Applegate’s work (1991). ‘‘Legitimate power’’ (Leichty and Applegate, 1991:481) can be the result of social status, rank or institutionalised role depending on the specific situation. This agrees with what Thomas (1995:127) refers to as ‘‘legitimate power’’ which is ‘‘the type of powermost subject to cross-cultural variation’’ (Thomas, 1995:127). Equally importantly, social power is also closely associated with the rights and obligations associated with the role. According to Blum-Kulka and House (1989), rights and obligations are definable on the grounds of social relations (context external factors). In a customer-waiter relationship, for example,waiters are obliged to provide good service and, if they fail to do so, customers have the right to complain, so in this respect, it is the customer who has greater power. The above components are not only situation-specific but also culture-specific. ‘‘In an egalitarian culture, for example, social class may not be an important component, but in a culture which emphasises vertical relationships, social class or status may become important, as is evidently the case in Japan’’ (Fukushima, 2000:85). Similarly, customers in one culture might not have the same rights as customers in a different, more customer-care oriented culture. 3.3. Imposition—a definition Unlike power and familiarity, the meaning of imposition ‘‘does not vary much among researchers’’ (Fukushima, 2000:84). One area of inconsistency, however, is whether rights and obligations are components of imposition. Some authors seem to include them in their definition of imposition (Brown and Levinson, 1987; Fukushima, 2000), while others do not (Thomas, 1995). For Thomas (1995) the definition of imposition is ‘‘how great the request you are making is’’ (Thomas, 1995:130) and the answer depends on the value of what is being asked for. Thomas (1995) refers to Goffman’s (1967) notion of ‘‘free goods’’1 and ‘‘non-free goods’’ to explain the value of what is being asked. For Fukushima (2000) the size of imposition depends on factors such as time, effort, financial burden and psychological burden on the part of the addressee. Fukushima’s ‘‘standpoint towards imposition is to include rights and obligations, since imposition will also be influenced by whether the requester has a right to make a certain request and whether the requestee has an obligation to pursue the request’’ (Fukushima, 2000:88). For example, in certain cultures university lecturers might have the right to request the student to carry their books and notes back to their office at the end of the lecture, whereas in other cultures (especially in Western cultures) this might constitute a greater imposition, as such a task falls outside students’ obligations and teachers’ rights. The term ‘imposition’ in the present study is defined using Thomas’ definition (1995:130) (‘‘how great is the request you are making?’’) and Fukushima’s (2000) proposed factors which affect the size of the imposition (e.g. time, effort, financial burden, psychological burden, rights and obligations). Indeed, rights and obligations need to be included under the variable of imposition as ‘‘the degree of imposition of the requested act will be high if the requester does not have a right to ask a certain request, and the requestee does not have an obligation to pursue it’’ (Fukushima, 2000:88). This might be seen in the case of standard situations where there is high obligation for Hto comply and a strong right for S to pose the request. There is, therefore, low difficulty for both H to pose the request and for H to accept and carry out the requested act. Yet, as pointed out above, whether S has the right to pose the request or not is related to power. This shows how the variables are often interrelated as some of their components overlap. Similarly to familiarity and power, imposition may also vary individually, situationally and cross-culturally. It is important to note that the components presented in the definition of each variable are my interpretations, and the participants in this study might have different components of each variable in mind. However, an examination of what each subject considers to be the components of each variable falls outside the scope of the present study. 1 Goffman’s (1967) term ‘‘free goods’’ refers to those small things (i.e. coffee, cigarettes, and glass of water) which are considered as having no value as such and can therefore be obtained without someone making a specific request. Of course, what constitutes free and non-free goods often differs from culture to culture. M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 2267 4. Methods and procedures 4.1. Participants One hundred and ninety two participants took part in this study: 92 native speakers of British English and 100 native speakers of Greek, second language speakers of English. All the participants were full-time university students in the UK and they were either enrolled in an undergraduate or a postgraduate study. In order to ensure greater internal validity, only students aged 17–30 were included in the study. The mean age for the Greek participants was 22.4 and the mean age for the British participants was 21.7. For the Greek participants, there were 51 males and 49 females. For the British participants, there were 45 males and 47 females. The Greek participants had been in the U.K. for an average of 2.8 years at the time of the study. They all grew up in Greece and were native speakers of the Greek language. English was a foreign language for all of them and none of them had lived in the UK or in any other country for any period of time prior to the present study. As they were all enrolled for a Bachelor or a Masters degree, they all satisfied the English Language requirement for entry to an undergraduate or a postgraduate study having passed one of the formal English language tests (e.g. ESOL level 2, IELTS Grade 6.0, TOEFL 550 +, GCSE or IGCSE English Language: Minimum C, etc.). Their English language proficiency level can therefore be characterized as ‘advanced’. The British participants came from England, Wales, Scotland and Ireland. Completed questionnaires from respondents who reported to have lived in a foreign country for more than one year were rejected. Similarly, responses from participants who reported to be native speakers of an additional language other than English were also eliminated. As Britain is a multicultural society, no ethnic-minority participants were used in order to avoid influences from other cultures and languages as much as possible. Thus only students of white European ethnic origin were selected to participate (for a discussion of ‘‘British English’’ see Stewart, 2005:116). 4.2. Data collection procedures 4.2.1. The DCT A written discourse completion task (WDCT), previously pilot-tested, was the main elicitation instrument of the study but interview data were also collected. The discourse completion task was designed to elicit requests in writing in the English language. Through the use of the elicitation instrument, a total number of 1920 requests were collected, coded and analysed. The DCT consisted of two parts. The first part, which was a discourse-completion, open-ended questionnaire, comprised 10 social situations, all of which had been designed to elicit requests. The pilot study had already proved that they could effectively elicit requesting strategies and a number of these situations had already been successfully used in other studies (Blum-Kulka and Olshtain, 1984; Blum-Kulka et al., 1989; Takahashi, 1992). Participants were again given a short description of the situation, which specified the setting, the familiarity and the social power between the participants. They were then asked to complete the dialogue, responding as they thought the person in the situation would, thus providing a requesting strategy. The social situations represented by this questionnaire varied by design along the dimensions of social power and familiarity. The scenarios were as follows: 1. Tuition fees: The speaker gives private language tuition to the hearer. The hearer forgets to pay him/her so the speaker has to remind him/her about the money. 2. Job leave: The speaker is a new employee and has to ask his/her boss for a few days’ leave to see his/her dying grandmother. 3. Baby-sitting: An emergency comes up so the speaker has to ask his/her parents to baby-sit for the evening. 4. Restaurant: The speaker is in a restaurant with a group of friends and gives his/her order to the waiter. 5. Assignment extension: A student asks for an assignment extension from one of his/her lecturers who knows him/her best. 6. Booking ticket: The speaker rings an airline reservation centre to book a flight. 7. Bank loan: The speaker asks the bank manager for a loan of £2000. 8. Dinner: The speaker asks his/her best friend to pass him/her the salad during dinner. 9. Street directions: The speaker asks a pedestrian for some directions. 10. Police: The speaker is a traffic warden and has to ask a driver, who has parked on a yellow line and is blocking a garage entrance, to move his/her car immediately. Six role constellations were represented: (a) SP (social power): x > y/F (familiarity): high (situation 1), (b) SP (social power): x > y/F (familiarity): low (situation 10), (c) SP (social power): x < y/F (familiarity): high (situation 5), (d) SP (social power): x < y/F (familiarity): low (situation 2) (situation 7), (e) SP (social power): x = y/F (familiarity): high (situation 3) (situation 8), (f) SP (social power): x = y/F (familiarity): low (situation 4) (situation 6) (situation 9). These descriptions are of course my own perceptions and might not represent the participants’ views (e.g. SP might for some participants rest on the H rather than on the S in situation 1, and the interlocutors’ relationship might be seen as carrying low rather than high familiarity). For this reason, the participants’ perception of social reality is something which is examined in the second part of the DCT (the situation assessment questionnaire). 2268 M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 These selected situations were also chosen in such a way as to include both ‘‘standard’’ and ‘‘non-standard’’ (House, 1989) situations. Three ‘‘standard’’ situations such as Restaurant, Ticket and Police are situations, which present ‘‘a pre-fixed constellation of rights and obligations’’ (Blum-Kulka and House, 1989:142), where the listener has a high obligation to comply and the speaker has a strong right and a low degree of difficulty to pose the request. Such situations are standard situations in the sense that precisely those parameters which make requesting such potentially difficult work interactionally are relatively set or standardised. Results in previous studies (House, 1989) have shown that the notion of a ‘standard situation’ is a crucial determinant of learners’ choice of requestive strategies. Finally, the chosen situations intended to elicit both requests for action and requests for information. This was decided in order to achieve greater requesting variation and because requests for information ((‘‘Asks’’): Hassall, 1999) were analysed separately on their own scale of directness. TheWDCT,whichwas themain elicitation instrument used for themost extensive research project on speech acts today (the CCSARP; Blum-Kulka and Olshtain, 1984; Blum-Kulka et al., 1989), and numerous cross-cultural and interlanguage studies have been conducted since then using this instrument (Blum-Kulka and Olshtain, 1986; Fearch and Kasper, 1989; Fukushima, 1990; Takahashi, 1992; VanMulken, 1996; Lee, 2005;Woodfield, 2006; Economidou-Kogetsidis, 2008, 2009), still remains the most heavily criticised research instrument in linguistic studies for a number of justifiable reasons. Methodological concerns include its hypothetical nature of the situations and how representative the WDCT data are to authentic spoken language (Rintell and Mitchell, 1989; Hartford and Bardovi-Harlig, 1992; Sasaki, 1998).Woodfield (2005, 2008), for example, reports on negative comments from research participants regarding both the authenticity of the research tasks and the reliability of written responses. Golato (2003:91–92), similarly suggests that written DCTs are inappropriate for studying actual language as they require participants not to conversationally interact but to ‘‘articulate what they believe would be situationally appropriate responses within possible, yet imaginary, interactional settings’’. Bardovi-Harlig and Hartford (2005:11) also argue that among the instruments available to researchers for examining production in pragmatics studies, production questionnaires do not generate interactive language samples, while Kasper (2008:294) points out that ‘‘DCTs and other questionnaire formats elicit intuitional data rather than data on language use and behaviour’’. While acknowledging the extensive criticisms of WDCTs in the literature, the DCT was still employed because the focus of the present study was on such intuitional data. The study focuses on the nature of the pragmalinguistic and sociopragmatic resources employed by learners and native speakers rather than on the nature of speech acts in interaction (Kasper, 2008). Kasper and Rose (2002) agree that ‘‘when carefully designed, DCTs provide useful information about speakers’ pragmalinguistic knowledge of the strategies and linguistic forms by which communicative acts can be implemented and about their sociopragmatic knowledge of the context factors under which particular strategic and linguistic choices are appropriate’’ (2002:96). Acknowledging, therefore, the fact that the language collected is not authentic spoken language, the findings of the present study should be regarded as findings relating to what speakers tend to view as being pragmatically appropriate linguistic behaviour rather than findings deriving from actual discourse. The use of DCT additionally offered a number of important practical advantages to the purposes of the present study. Although DCTs do not offer interactive language samples, they can offer data of high comparability due to the controlled nature of the task (Bardovi-Harlig and Hartford, 2005:11). Even though naturally occurring data would have been the ideal data, such data for cross-cultural studies presents major practical limitations. Legal and ethical considerations aside, it would be practically impossible to collect large data of a given speech act in the same situational and interpersonal context. As Rintell and Mitchell point out (1989:250) ‘‘it is impossible to control the contextual variables so as to ensure that the same context will be repeated even once’’ (Rintell and Mitchell, 1989:250). For the present study, the DCT allowed great control of the contextual variables important to the study (Rintell and Mitchell, 1989:250) and permitted the gathering of large data ‘‘on a wide range of difficult-to-observe linguistic phenomena’’ (Billmyer and Varghese, 2000:518). At the same time, it allowed control over the demographic information related to the subjects used. 4.2.2. The situational assessment questionnaire The second part of the DCT was a situational assessment questionnaire which aimed to find out whether these situations are perceived and rated similarly by the two language groups. Such cross-cultural variation in the perception of social reality has been linked to differences in the speakers’ requestive behaviour. The Greek learners, for example, might be found to decrease their requestive directness with higher social power and increase it accordingly with higher familiarity. The situational assessment questionnaire (see Fig. 1) followed the same structure as the instrument used by Blum- Kulka and House (1989), with the main difference being that it concerned only three social parameters: (a) social power, (b) familiarity, and (c) imposition of the requested act. Participants were asked to rate each of the above social dimensions on a Likert scale of 1–3, where 1 was the lowest and 3 the highest. They were asked to report not only on the speaker’s social power but also on the listener’s social power. Four social dimensions for each situation were therefore examined. To keep evaluations as specific to the situations as possible, the questions contained explicit reference to the interlocutors’ relationship (e.g. parents, best friend, and driver) or name (e.g. Mike). To avoid any confusion in relation to the terminology of ‘‘familiarity’’, ‘‘power’’ and ‘‘imposition’’, short and simple explanations of these terms were provided at the beginning of the questionnaire. M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 2269 4.2.3. The interviews Semi-structured interviews were also conducted in relation to both native speakers and the learners’ production. These interviews aimed to gain some qualitative insights into the participants’ views, beliefs and opinions regarding requests and politeness. As interviewing was not the main research instrument of the study and was simply used as a complimentary tool, only four informants were interviewed (2 GLs and 2 ENSs). Among other things, the interviewees were asked to comment on the situational features of social power and familiarity between the two interactants, as well as on the imposition of the requested act. They were additionally asked to comment on how and whether their written response would have been any different if one of these social parameters had been any different. The interviewees were asked to take a few minutes to read their copy of completed questionnaire in order to familiarise themselves with the situations and their own responses again. The duration of each interview was approximately 30 min long. The total interview time collected was therefore 120 min. 5. Data analysis The requesting strategies yielded were analysed within a shared analytical framework. The analysis was based on an independent evaluation of each request head act according to its degree of directness. Directness is taken to mean the degree to which the speaker’s illocutionary intent is apparent from the locution (Blum-Kulka et al., 1989:278),while head act is ‘‘theminimal unitwhich can realize a request’’, in otherwords ‘‘the core of the request sequence’’ (Blum-Kulka et al., 1989:275). Based on the dimension of directness/indirectness, a coding scheme was therefore developed for the cross-linguistic analysis of both English and Greek requesting strategies. This coding scheme, which was in effect a directness/indirectness model, was based on the coding scheme of the Cross-Cultural Speech Act Realisation Project (CCSARP) (Blum-Kulka and Olshtain, 1984; Blum-Kulka et al., 1989), with a number of modifications and adaptations made to it to account for requests received not successfully covered by the taxonomy’s existing predicting value.2 In addition to the above, requests for action and requests for information were analysed on a separate scale of directness. As Hassall (1999) argues, requests for information (‘‘asks’’) need to be analysed separately from other requests because of problems presented by one strategy for asking for information, the ‘‘direct question’’. A direct question (e.g. ‘‘Where is the train station?’’) is the most direct way of all to ask for information, but as a means of asking anything other than information (e.g. for goods or a service) (e.g. mother to son before going out: ‘‘Where is your jacket?’’), a question is not direct (Hassall, 1999:594–595). Both requests for action and requests for information were analysed along the following main directness levels of (a) most direct strategies,3 (b) conventionally indirect strategies,4 and (c) non-conventionally indirect strategies.5 A forth option (d) of opting out of the act altogether (‘don’t do the face-threatening act (FTA))’ was also offered. Fig. 1. Sample from the situational assessment questionnaire. 2 The data of the present study, for example, yielded a new requesting sub-strategy which included ‘reminder requests’. Such utterances served to remind the hearer about an expected or a prohibited action. Such a request might be (a) an expectation request (e.g. ‘‘Mike, you haven’t paid me today.’’, ‘‘You owe me some money, don’t you think?’’), or (b) a prohibited-action request (e.g.. ‘‘You can’t park here!’’). Another new category developed included ‘pre-decided statements’. These are utterances in which the speaker chooses to skip the requestive illocution altogether and instead simply checks whether his/her decision about the course of action he/she will take is acceptable by the hearer. Such requests are normally employed in situations where S feels it would be more appropriate to help himself/herself rather than employ some kind of requestive construction (i.e. for free goods) (e.g. ‘‘I won’t be in tomorrow. I hope it’s OK.’’). Both reminder requests and pre-decided requests were classified in the most-direct strategy level of requests. 3 This strategy level which covers ‘‘explicit requests that are syntactically marked as such, for example, imperatives, or by other verbal means that name the act as a request, such as performatives (Austin, 1962) and hedged performatives (Fraser, 1975)’’ (Blum-Kulka, 1989:46) (e.g. ‘‘Please look after the kids for the evening.’’, ‘‘I’m here to ask for an extension’’). Obligation, want and need statements (e.g. ‘‘I need you to pay me’’) also fall into this level. 4 This strategy level covers strategies that realise the act by reference to contextual preconditions necessary for its performance, as conventionalised in a given language (Blum-Kulka, 1989:47). Conventioanlly indirect requests might typically include utterances which contain a suggestion for H to do X (Blum- Kulka et al., 1989:18) (e.g. ‘‘How about looking the kids tonight?’’), or utterances containing reference to preparatory conditions (e.g. ability, willingness, possibility) as conventionalised in any specific language (Blum-Kulka et al., 1989:18) (e.g. ‘‘Could you look after the kids tonight?’’, ‘‘Would it be possible to take a few days off work?’’). 5 Non-conventionally indirect strategy level covers the open-ended group of indirect strategies that realise the request either by partial reference to the object or element needed for the implementation of the act or by reliance on contextual clues (Blum-Kulka, 1989:47). This category includes strategies which are not conventionalised in the language and hence require more inferencing activity for the hearer to derive the speaker’s requestive intent. Strong and mild hints, as well as ‘‘non-explicit question directives’’ (Ervin-Tripp, 1976) (e.g. ‘‘Where’s your jacket?’’) belong to this most indirect strategy level. 2270 M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 6. Results 6.1. Situational and cross-cultural variation of main requesting strategies To what extent do GLs and ENSs agree on the variation of their requestive behaviour according to the social situation? The analysis of the distribution of the main request strategy types in the ten situations examined revealed that: (a) There are high levels of cross-cultural agreement between the Greek and the English speakers for trends of situational variation namely, relatively higher levels of directness are licensed insome situations thaninothers across bothcultures (see Figs. 2–5). (b) There is some cross-cultural variation in choices of directness level within some situations. Thus, the two groups were found to disagree on the specific directness levels appropriate for given situations (Table 1). Despite the general cross-cultural agreement between the ENSs and the GLs for trends of situation variation and, as can be observed from Table 1, different results were revealed for each DCT situation regarding the main requesting strategies employed by the two groups. In order to identify which of these percentage differences were statistically significant, chi-square tests of Independence were conducted. The statistical analysis results showed that there was a significant difference in the choice of the main requesting strategies between GLs and ENSs in the Tuition fees (situation 1) and the Assignment situation (situation 5) (chi-square was significant at the 0.01 level). Looking more closely at situation 1 (Tuition fees), the ENSs were significantly more direct than the GLs. More specifically, 50% of the strategies employed by the British Fig. 3. The use of conventionally indirect strategies in the ten request situations by ENSs and GLs. Fig. 2. The use of direct strategies in the ten request situations by ENSs and GLs. M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 2271 participants belonged to the most direct strategy level, whereas only 20.5% of GLs’ strategies belonged to this directness level. The results also indicated that GLs strongly favoured conventional indirectness in this situation, whereas only 25.6% of ENSs opted for conventional indirectness. Interestingly, the results of the Tuition fees situation revealed that more GLs opted out of this face-threatening act altogether by choosing to say nothing as compared to ENSs (compare 8.3% vs. 1.3%—see Table 1). Looking more closely at situation 5 (Assignment) (Table 1), it was the Greek participants who used significantly more direct strategies (50.6%) compared to ENSs (10.7%). The vast majority of ENSs’ strategies were conventionally indirect (85.7%), whereas only 49.4% of GLs’ strategies were conventionally indirect. While ENSs made some use of nonconventionally indirect strategies (3.6%), GLs made no use of these strategies at all (0%). Yet, despite the above disagreement in the participants’ choices of directness levels in these two social situations, the use of the three main strategy types (direct/ conventionally indirect/ non-conventionally indirect) follows a similar trend across the majority of the situations in both groups. The highest degree of cross-cultural agreement can be observed in the Babysitting, Restaurant, Dinner, and Directions situations (see Table 1). Higher levels of direct strategies are licensed by both groups in the Restaurant, Ticket and Bank loan situation. Similarly, higher levels of conventional indirectness are licensed by both groups in the Baby-sitting, Dinner and Directions situation (see Table 1). Finally, neither the ENSs nor the GLs favoured non-conventional indirectness in the situations tested. 6.2. Cross-cultural variation in the assessment of social variables Do GLs and ENSs perceive the same social situations differently in terms of (a) social power, (b) familiarity and (c) imposition of the requested act? People from different cultures often hold differing views regarding the social and situational factors involved and the relative importance they attribute to the parameters involved (Blum-Kulka and House, 1989; Spencer-Oatey, 1992, 1993, 1997; Fukushima, 2000). The argument here is that differences in the participants’ perception of social reality can lead to differences in the participants’ linguistic choices (i.e. requesting strategy selection and degree of directness). Fig. 4. The use of non-conventionally indirect strategies in the ten request situations by ENSs and GLs. Fig. 5. The use of ‘Don’t do the FTA’ in the ten request situations by ENSs and GLs. 2272 M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 In order to investigate this research question, mean scores of each language group (Greek/British) were calculated. These scores were calculated in relation to the above social variables and to each social situation. T-tests, a ‘‘procedure that tests the difference between two groups for normally distributed interval data’’ (Hatch and Lazaration, 1991:249), were conducted in order to establish whether the mean differences were statistically significant. The mean scores and the t-test results are all presented in Table 2. As can also be observed from Table 2, cross-cultural disagreement was perceived in the participants’ assessment of these social variables in the majority of the situations. More specifically, the two groups assessed the relative weight of these parameters differently in situations 1, 3, 5, 6, 7, 8 and 10 (i.e. in all situations except fromsituations 4 and 9). The resultswere as follows: (1) There was a significant difference in the assessment of familiarity in the Tuition fees situation. ENSs assessed the familiarity between S and H as greater as compared to GLs. (2) There was a significant difference in the assessment of social power that S has over H in the Tuition fees, Baby-sitting and Bank loan situation. ENSs assessed the social power that S has over H in the Tuition fees situation as greater as compared to GLs. GLs, on the other hand, assessed the social power that S has over his/her parents (Baby-sitting situation) and over his/her best friend (Dinner situation) as greater as compared to ENSs. Table 1 Percentage distribution and chi-square results of main request strategy types in ten social situations (nGreek = 100, nEnglish = 92). Situation Strategy type GLs % ENSs % Chi-square results Situation 1 (Tuition fees) 1. Most direct 20.5 50.0 x2 = 19.44 2. Conventionally indirect 50.7 25.6 df = 3 3. Non-conventionally indirect 20.5 23.1 p = 0.000** 4. Do not do FTA 8.3 1.3 Situation 2 (Job leave) 1. Most direct 51.9 38.4 x2 = 4.41 2. Conventionally indirect 46.9 61.6 df = 2 3. Non-conventionally indirect 1.2 0 p = 0.110, NS 4. Do not do FTA 0 0 Situation 3 (Baby-sitting) 1. Most direct 10.8 11.7 x2 = 0.76 2. Conventionally indirect 84.3 85.9 df = 2 3. Non-conventionally indirect 4.9 2.4 p = 0.684, NS 4. Do not do FTA 0 0 Situation 4 (Restaurant) 1. Most direct 67.5 57.0 x2 = 1.85 2. Conventionally indirect 32.5 43.0 df = 1 3. Non-conventionally indirect 0 0 p = 0.173, NS 4. Do not do FTA 0 0 Situation 5 (Assignment) 1. Most direct 50.6 10.7 x2 = 32.85 2. Conventionally indirect 49.4 85.7 df = 2 3. Non-conventionally indirect 0 3.6 p = 0.000** 4. Do not do FTA 0 0 Situation 6 (Ticket) 1. Most direct 78.0 65.1 x2 = 3.87 2. Conventionally indirect 18.3 31.4 df = 2 3. Non-conventionally indirect 3.7 3.5 p = 0.145, NS 4. Do not do FTA 0.0 0.0 Situation 7 (Bank loan) 1. Most direct 72.1 61.9 x2 = 2.64 2. Conventionally indirect 24.4 35.7 df = 2 3. Non-conventionally indirect 3.5 2.4 p = 0.267, NS 4. Do not do FTA 0 0 Situation 8 (Dinner) 1. Most direct 20.8 27.3 x2 = 0.89 2. Conventionally indirect 77.9 71.4 df = 2 3. Non-conventionally indirect 1.3 1.3 p = 0.640, NS 4. Do not do FTA 0 0 Situation 9 (Directions) 1. Most direct/bald-on record 1.2 1.2 x2 = 0.98 2. Conventionally indirect 98.8 97.6 df = 2 3. Non-conventionally indirect 0 1.2 p = 0.612, NS 4. Do not do FTA 0 0 Situation 10 (Police) 1. Most direct 56.2 44.4 x2 = 3.62 2. Conventionally indirect 43.8 52.8 df = 2 3. Non-conventionally indirect 0 2.8 p = 0.164, NS 4. Do not do FTA 0 0 ** p < 0.01, NS: not significant. M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 2273 (3) There was a significant difference in the assessment of social power thatHhas over S in the Bank loan situation. The social power that the bank manager (H) has over S was assessed as bigger by ENSs than by GLs. (4) There was a significant difference in the assessment of imposition of the requested act in the Assignment, Ticket, Dinner and Police situation. In the Assignment situation, the level of imposition was rated higher by the ENSs than by the GLs. In the Ticket, Dinner and Police situation, on the other hand, the level of imposition of the requested act was rated higher by the GLs than the ENSs. 6.3. Correlation analysis between assessment of social reality and levels of directness To what extent is the GLs’ selection of requesting strategies determined by the social variables of familiarly, power and imposition? In other words, are there any systematic links/correlations between their perceptions of social reality and the observed levels of directness? The participants were expected to vary their requesting strategies depending on their perception of the above social variables (i.e. the greater the familiarity between S and H, the more direct their requesting strategies are expected to become). Statistical analyses were conducted to test out the correlation between the learners’ assessment of social variables and their choice of requesting strategies. The Spearman Rank-Order Correlation (Rho) test was used for this analysis as ‘‘the Spearman correlation is appropriate for both rank-order data and interval data with the strength of ranks’’ (Hatch and Lazaration, 1991:451). The results of Spearman’s correlation, presented in Table 3, revealed that there are correlations between the learners’ choice of requesting strategies and their assessment of social variables only in the Baby-sitting (situation 3) and the Dinner (situation 8) situation. More specifically, results showed that: (1) There is no correlation between the learners’ degree of directness and their assessment of familiarity. (2) There is no correlation between the learners’ degree of directness and their assessment of H’ social power. (3) There is a negative correlation between the learners’ degree of directness and their assessment of the speaker’s social power in the Dinner situation (rs = 0.269, p = 0.019). This indicates that the greater the degree of P the speaker has over his/her best friend, the more direct the requesting strategies become. Table 2 T-test results and mean scores of social variables by ENSs and GLs. Familiarity Social power of S Social power of H Imposition Situation 1 (Tuition fees) G: 1.91 G: 2.18 G: 1.53 G: 1.84 B: 2.06 B: 2.48 B: 1.43 B: 1.87 t(162) = 2.018* t(163) = 3.041** t(161) = 0.985 t(161) = 0.312 Situation 2 (Job leave) G: 1.43 G: 1.30 G: 2.63 G: 2.21 B: 1.29 B: 1.30 B: 2.59 B: 2.25 t(161.5) = 1.700 t(162) = 0.055 t(163) = 0.431 t(159) = 0.333 Situation 3 (Baby-sitting) G: 2.87 G: 2.31 G: 2.30 G: 1.86 B: 2.95 B: 1.91 B: 2.25 B: 1.81 t(138.3) = 1.615 t(158.5) = 4.149** t(162) =0.524 t(144.7) = 0.497 Situation 4 (Restaurant) G: 1.21 G: 2.27 G: 1.36 G: 1.64 B: 1.28 B: 2.36 B: 1.50 B: 1.46 t(147.3) = 0.805 t(151) = 0.886 t(151) = 1.523 t(151) = 1.569 Situation 5 (Assignment) G: 1.95 G: 1.20 G: 2.61 G: 2.14 B: 2.11 B: 1.27 B: 2.69 B: 2.45 t(161) = 1.782 t(162) = 1.039 t(162) = 0.908 t(156.1) = 3.130** Situation 6 (Ticket) G: 1.22 G: 1.96 G: 1.62 G: 1.64 B: 1.20 B: 2.14 B: 1.74 B: 1.41 t(157.9) = 0.272 t(164) = 1.502 t(163) = 1.207 t(151.5) = 1.207* Situation 7 (Bank loan) G: 1.26 G: 1.46 G: 2.40 G: 2.00 B: 1.31 B: 1.38 B: 2.69 B: 1.99 t(163) = 0.673 t(163) = 0.803 t(159.6) = 3.071** t(162) = 0.109 Situation 8 (Dinner) G: 2.82 G: 2.26 G: 2.25 G: 1.71 B: 2.91 B: 1.97 B: 2.07 B: 1.49 t(143.01) = 1.295 t(148.05) = 2.496* t(149.9) = 1.607 t(144.7) = 1.929* Situation 9 (Directions) G: 1.16 G: 1.19 G: 1.38 G: 1.47 B: 1.10 B: 1.32 B: 1.48 B: 1.52 t(127.01) = 0.999 t(159.37) = 1.542 t(158.38) = 1.038 t(163) = 0.573 Situation 10 (Police) G: 1.31 G: 2.69 G: 1.26 G: 2.18 B: 1.21 B: 2.72 B: 1.22 B: 1.85 t(136.51) = 1.127 t(144) = 0.352 t(144) = 0.403 t(143) = 2.844* *p < 0.05, **p 0.01. G = Greek subjects/mean scores, B = British subjects/mean scores. 2274 M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 (4) There is a negative correlation between the learners’ degree of directness and their assessment of imposition in the Babysitting situation (rs = 0.228, p = 0.042). This indicates that the greater the imposition, the more direct the learners’ strategies become. 7. Discussion The findings in relation to the participants’ variation of requesting behaviour clearly show that P, F and R interact with cultural and other situational factors. While both the GLs and the ENSs varied their requests by situation, they differed in their specific choices within each situation. This finding agrees with Blum-Kulka and House’s (1989) finding whose study of Australian English, German, French, Hebrew and Argentinian Spanish speakers’ requests showed that ‘‘while the overall distribution along the scale of indirectness follows similar patterns in all languages, the specific proportions in the choices between the more direct and less direct strategies are culture-specific’’ (Blum-Kulka and House, 1989:133). The results showed that the GLs were significantly more indirect in the Tuition fees situation than the ENSs. That was contrary to the finding of previous studies (Tannen, 1981; Sifianou, 1992a; Economidou-Kogetsidis, 2005) which showed that Greek speakers are more direct in their speech behaviour than British and American English speakers. An explanation regarding these differences in the subjects’ directness could be found in differences in their conception of social relations. Results from the situational assessment analysis above indicated that GLs and ENSs hold differing norms regarding the familiarity and the social power involved in the Tuition fees situation. More specifically, the ENSs assessed both the familiarity between S and H and the social power that S has over H as higher. As a result, the ENSs were the ones who employed more direct requests to remind for the owed fee. Very direct requests such as ‘‘That’ll be £10 then Mike’’, ‘‘Can I have the money please?’’ or ‘‘Do you have your money ready?’’ were used extensively by the British participants who, also according to the interviews, saw this request as being well within their professional rights: ‘‘- Interviewer: Some people’s response was ‘‘I would never ask for the money. I feel embarrassed.’’ What do you think about this statement? - Interviewee: ‘‘. . .at the end of the day it is your job and if you went to work somewhere else and you didn’t get paid, you’d want them to pay you. So if he doesn’t pay you it’s more or less the same. You’ve taught him for that lesson, you’ve given him a lesson but he hasn’t paid you. . .So I wouldn’t really agree with the statement.’’ (Male interviewee—English participant) Greek participants, on the other hand, primarily opted for conventionally indirect strategies, while a number of them opted out of the face threatening act altogether, preferring not to request the fee at all. This was also supported by the Greek participants interviewed who emphasised that unless they were really desperate for the money, they would not ask until next time. Table 3 Spearman’s rank order correlation coefficients between the GLs’ choice of requesting strategies and the assessment of social variables. Familiarity Social power of S Social power of H Imposition Situation 1 (Tuition fees) rs = 0.143 rs = 0.132 rs = 0.107 rs = 0.117 p = 0.233, NS p = 0.270, NS p = 0.376, NS p = 0.335, NS Situation 2 (Job leave) rs = 0.094 rs = 0.030 rs = 0.023 rs = 0.121 p = 0.415, NS p = 0.797, NS p = 0.842, NS p = 0.293, NS Situation 3 (Baby-sitting) rs = 0.150 rs = 0.090 rs = 0.153 rs = 0.228 p = 0.178, NS p = 0.424, NS p = 0.175, NS p = 0.042* Situation 4 (Restaurant) rs = 0.126 rs = 0.038 rs = 0.161 rs = 0.143 p = 0.280, NS p = 0.745, NS p = 0.167, NS p = 0.222, NS Situation 5 (Assignment) rs = 0.139 rs = 0.120 rs = 0.082 rs = 0.068 p = 0.221, NS p = 0.292, NS p = 0.470, NS p = 0.549, NS Situation 6 (Ticket) rs = 0.029 rs = 0.088 rs = 0.005 rs = 0.109 p = 0.799, NS p = 0.434, NS p = 0.968, NS p = 0.336, NS Situation 7 (Bank loan) rs = 0.049 rs = 0.124 rs = 0.019 rs = 0.106 p = 0.668, NS p = 0.272, NS p = 0.866, NS p = 0.353, NS Situation 8 (Dinner) rs = 0.194 rs = 0.269 rs = 0.115 rs = 0.017 p = 0.093, NS p = 0.019* p = 0.326, NS p = 0.882, NS Situation 9 (Directions) rs = 0.037 rs = 0.045 rs = 0.079 rs = 0.090 p = 0.742, NS p = 0.691, NS p = 0.491, NS p = 0.427, NS Situation 10 (Police) rs = 0.136 rs = 0.089 rs = 0.069 rs = 0.082 p = 0.255, NS p = 0.459, NS p = 0.567, NS p = 0.492, NS * p < 0.05. M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 2275 ‘‘If I didn’t really need this money, then I’d wait until next time to see whether he remembers [my italics]. . .I wouldn’t want to make him feel that I’m pushing him to pay me. . ..’’ [translated from Greek] (Female interviewee—Greek participant) This finding can be seen as tentatively supporting Brown and Levinson’s main argument (1978, 1987) that the social variables of social distance, power and imposition are important factors influencing the choice of politeness strategies. Examining the interview comments more closely, it can additionally be argued that such cross-cultural disagreement on the degree of directness employed can more specifically be attributed to such ‘‘context external’’ (Blum-Kulka and House, 1989:131) factors such as rights and obligations, which, as mentioned above, are directly associated with social power. Blum-Kulka and House (1989:145) explain that: ‘‘in the sociolinguistic literature, estimates of right and obligations are considered to affect a request level’s of directness in a correlated relationship: The greater the right of the speaker to ask and the greater the obligation of the hearer to comply, the less the motivation for the use of indirectness (Labov and Fanshel, 1977)’’ (Blum-Kulka and House, 1989:145–146) How strong these rights and obligations are is culture-specific and in the Tuition fees situation the speaker’s right to remind for the owed fee and the hearer’s obligation to pay is probably seen as bigger by the British rather than by the Greek participants. This can be confirmed by the fact that the ENSs assessed the social power that S has over H as higher. This finding is also in line with the negative politeness orientation of the British society which values personal autonomy and independence. For the Greek speakers, cultural factors such as positive politeness considerations seem to play a more decisive role in the linguistic choices than P, F and R, something which could also explain the fact that no correlations between the learners’ directness and their assessment of F, P and R emerged. Such positive politeness considerations might include protecting H’s face from embarrassment, showing concern for H’s feelings by avoiding disapproval and criticism and/ or avoiding a taboo topic such a money request. It seems very likely that the request for the owed fee in this situation does not constitute a threat to the addressee’s negative face, as far as the Greek learners are concerned. Contrary to Brown and Levinson’s theory (1978, 1987) according to which requests are by nature face threatening acts that threaten the addressee’s (H) negative face want (Brown and Levinson, 1987:65), this request can be seen as constituting a threat to the addressee’s positive face instead, ‘‘by indicating that the speaker does not care about the addressee’s feelings’’ (Brown and Levinson, 1987:66). According to Brown and Levinson’s theory (1978, 1987), this happens if the addressee expresses disapproval, reprimands, accusations, insults, gives the addressee possible reasons to be embarrassed by him, mentions of taboo topics, etc. (Brown and Levinson, 1987:66–67). Reminding the addressee for the owed money is a taboo topic in the Greek society and can not only cause embarrassment, but also be interpreted as a reprimand, disapproval or even accusation towards H. As a result, it can also threaten the speaker’s positive face by threatening his/her ‘‘desire to be ratified, understood, approved of, liked or admired’’ (Brown and Levinson, 1987:62). The use of more indirect strategies on the part of the Greek learners can therefore be seen as a positive politeness strategy used to minimise the above risks, and it is in line with the general positive politeness orientation of the Greek society. The Greek participants’ positive politeness considerations were vividly revealed in their interview comments which made reference to embarrassment and loss of trust: ‘‘Well. . .If I wasn’t desperate for the money I wouldn’t want to look pushy...I wouldn’t want to make him feel that I’m pushing him to pay me. . .or that I don’t trust him enough. . .or that I think he might do a runner...’’ [my italics] [translated from Greek] (Male interviewee—Greek participant) ‘‘. . .I think by asking Mike for the money, your request could have a personal effect on him. . .he is a student. . .he might not have enough money and he might be planning to pay you next week. . .I might embarrass him. . .’’ [my italics] [translated from Greek] (Female interviewee—Greek participant) Yet, GLs were significantly more direct than ENSs in the Assignment situation (i.e. asking a lecturer for an assignment extension). While ENSs opted primarily for conventionally indirect strategies and more specifically for query preparatory (possibility) substrategies, GLs opted extensively for direct strategies and more specifically for want and need statements (e.g. ‘‘Hello. It’s great need. I hadserious personal problemsandI need to take anextensionfor the assignment.’’).Apossible explanation regarding the differences in the participants’ requesting choices in the Assignment situation could again be found in differences in their subjects’ perception of social reality and more specifically, in differences in their perception of the imposition of the requested act. The results presented above (Table 2) indicated that there were significant differences in the participants’ assessment of the imposition involved in this situation. More specifically, the GLs assessed the imposition of requesting an assignment extension as lower than the ENSs did, therefore employing higher requestive directness. A link between the subjects’ assessment of social reality and their choice of requesting strategies and degree of directness can therefore be suggested. 2276 M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 In this Assignment situation, interactants have a student–lecturer relationship. As pointed out by Fukushima (2000) and in the definition of imposition above, ‘‘the degree of imposition of the requested act is related to the requester’s right to make a certain request. How that right is perceived is culturally influenced’’ (2000:183). The fact that the British participants assessed the imposition as higher here may be accounted for by different perceptions of the speaker’s rights and of the size of factors such as time, effort and psychological burden on the part of the addressee. It might be the case that the Greek learners regarded an extension request as being within their student rights, or as part of the lecturer’s routine responsibilities, which involves minimal effort, time or psychological burden on his/her part. The interviews carried out with two Greek participants in relation to this situation showed that the only concerns that they had with this request were not concerns regarding the degree of imposition involved, but concerns regarding their own loss of face in case the lecturer doubted their sincerity and thought they were lying. This finding is also consistent with the ethos of directness and positive politeness orientation of the Greek society. According to Sifianou (1992a), ‘‘in positive politeness cultures where people depend on each other more and rely on their personal experience of those others, explicit expression of needs and feelings is expected. . .’’ (my italics) (1992a:42). Following this, it could therefore be argued that the use of higher directness (through an extensive use of need and want statements) by GLs in this situation is not only the result of their perception of the related imposition, but also of additional situational/ ‘‘context external’’ (Blum-Kulka and House, 1989:131) factors such as the requestive goal and the urgency of the request. This would agree with Blum-Kulka et al.’s (1985) finding which showed the request goal to be a particularly influential factor. The fact that the correlation analysis did not confirm a correlation between the learners’ choice of strategies and their assessment of the social variables tested makes it rather probable that other situational factors are at play. What also becomes apparent from the finding regarding the GLs’ assessment of social variables, is that the learners lacked L2-native-like sociopragmatic competence. The results have showed that the learners have ‘‘different perceptions of what constitutes appropriate linguistic behaviour’’ (Thomas, 1983:99) and it can therefore be claimed that their situational assessment revealed an instance of ‘‘sociopragmatic failure’’ (Thomas, 1983:99). Such failure occurs when learners assess the relevant situational factors on the basis of their native sociopragmatic norms rather than the norms of the target culture. According to Jaworski (1994), sociopragmatic failure, being social before linguistic, is often much more difficult to be corrected and overcome, as this would involve ‘‘making changes in the students’ own beliefs and value system’’ (Jaworski, 1994:51). Despite the disagreement in the participants’ choices of directness levels and assessment of social reality in the above two situations (Tuition fees and Assignment), the use of the three main strategy types follows a similar trend across the majority of the situations in both groups.More specifically, as stated above, a high degree of cross-cultural agreement regarding the use of direct strategieswas licensed by both groups in theRestaurant, Ticket and Bank loansituation. This cross-cultural agreementon the appropriateness of high directness can be attributed to such context external factors again such as rights and obligations, associated with the ‘standard’ nature of Restaurant, Ticket and Bank loan situation. Even though the Bank loan situation is not typically ‘standard’ in the sense that theH(the bankmanager)doesnot have a highobligation to comply (i.e.whether a loanwill be granted is at the discretion of the bank and the bankmanager), the speaker nevertheless has a strong right to pose the request as part of the nature of the exchange. Despite the fact that the GLs assessed H’s social power as higher, the decision of both groups seems to have been more powerfully influenced by their perception of the involved rights and obligations. Similarly, higher levels of conventional indirectness were licensed by both groups in the Baby-sitting, Dinner and Street directions situation despite differences in the way that both groups assessed the social variables P, F and R in the Dinner and Baby-sitting situation. This seems to suggest that the above variables were not particularity influential in the participants’ linguistic choices. This cross-cultural agreement on the appropriateness of conventional indirectness in these three situations can be attributed to such ‘‘context internal’’ (Blum-Kulka and House, 1989:131) factors ‘‘which are directly related to its requestive nature, such as the type of request goal, the degree of imposition involved for the speaker relative to the specific goal, and the prerequisites needed for compliance’’ (Blum-Kulka and House, 1989:131). Indeed, as Blum-Kulka and House (1989) explain, when receiving preference of conventionality in similar situations, situations such as the above ‘‘concern asking for a favour, and in all the ability and willingness of the addressee to comply are important prerequisites for compliance’’ (1989:131), rendering questions related to ability and willingness contextually highly relevant. According to Clark (1979), speakers can more easily back out of admitting a requestive intent and hearers avoid a requestive interpretation by making reference to the literal meaning of a conventional indirect utterance and by checking feasibility of compliance. Thus in the case of a request that entails imposing on someone’s time, the ‘‘speaker might first try and establish whether the relevant condition holds’’ (Blum-Kulka and House, 1989:131). One of the Greek interviewees also supports this point by emphasising reference to his parents’ ability to baby-sit: ‘‘I would use a question mark to make it more polite. I wouldn’t express it as a fact. My parents might have plans for the evening and might not be able to look after the children. I would ask first ‘Are you doing anything tonight?’’ (Male interviewee—Greek participant) Yet, looking at the Spearman correlation results (presented above), a negative correlation between the Greek learners’ degree of directness and their assessment of the imposition in the Baby-sitting situation can be observed. This indicated that the bigger the imposition, the more direct the learners’ strategies become. This partly confirms Brown and Levinson’s (1978, 1987) claim, in the sense that the variable of imposition is related to the choice of requesting strategies, but it does not M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 2277 support their claim that the bigger the imposition, the higher the number of strategy to be employed. Appeals here may again be made to cultural factors at play concerning the Greek culture and its members for preference for directness and honesty over tact and minimising impositions. Similarly, looking at the correlation results related to the Dinner situation, a negative correlation between the learners’ degree of directness and their assessment of the S’s social power emerged. This indicated that the bigger the degree of power the speaker has over his/her friend, the more direct the requests become. Unlike above, this finding is consistent with Brown and Levinson’s prediction which expects directness to increase with speaker’s power. 8. Limitations of the study Turning to the limitations of the present study, a number of points need to be raised in order to warrant attention for future research studies. The first point concerns the use of a written discourse completion test as the main methodological instrument. As acknowledged in section 4.2.1, the DCT has been extensively criticised in the interlanguage pragmatics literature primarily due to its non-interactive nature (Bardovi-Harlig and Hartford, 2005:11) and for being a tool which yields written language samples that might not be representative to authentic, spontaneous spoken language. It must therefore be acknowledged that the move in pragmatics research towards capturing more authentic, interactive data (Cohen, 2004, 2007; Kasper, 2008) is particularly important and similar future studies should aim at analysing spoken data gathered with a different methodological tool that allows more interaction (e.g. oral role-plays). A second limitation of the present study concerns the nature of some of the situations used in the DCT. More specifically, situations 10 (Police) and 3 (Baby-sitting) lack naturalness as they refer to scenarios that the student population might not easily associate with. In situation 10, the participants have to play the role of a ‘traffic warden’ while in situation 3 the participants have to play the role of a ‘parent’. The problem with situations that lack naturalness is a problem also noted by Fukushima (2000:135) and Bonikowska (1988:170) and future research studies should try and avoid such unrealistic situations. Thirdly, the participant profile of the present study was limited to undergraduate and postgraduate students and it is therefore important to be aware that these findings might not be generalisable to other social groups. Further research with other social groups and perhaps with a wider range of different social situations is necessary in order to draw more general conclusions. 9. Summary and conclusions The analysis of the distribution of the main request strategy types in the ten, socially different situations examined revealed that there are high levels of cross-cultural agreement between the Greek learners of English and the English native speakers for trends of situational variation, namely, relatively higher levels of directness are licensed in some situations than in others across both cultures. This cross-cultural agreement can be explained by the ‘standard’ nature of the situations where they used higher directness (Restaurant, Ticket, Bank loan situation), and the requestive nature of the Baby-sitting, Dinner and Directions situations which warranted for conventional indirectness. The study, however, also revealed some cross-cultural variation in choices of directness level within some situations, as the two groups were found to disagree on the specific directness levels employed for given situations. In explaining this variation, appeals were made to some significant differences in the speakers’ perception of social reality. Following the situational assessment test employed in the present study, a number of significant differences in the speakers’ assessment of familiarity, power and imposition were revealed, along with a degree of sociopragmatic failure on the part of the GLs. The significant differences in the speakers’ degree of requestive directness in the Tuition fees and Assignment situations were explained following the speakers’ assessment differences of social reality, therefore providing support for the relationship existing between the variables of P, F and R and the speakers’ choice of requesting strategies. Nevertheless, even though speakers’ assessment of P, F and R seems to explain the differences in their linguistic choices to some extent, following the results of the correlation analysis carried out, it could additionally be argued that none of these social variables was particularly influential on the speakers’ requestive production. This finding challenges Brown and Levinson’s (1978, 1987) claim that the variables of power, distance and imposition are the most important factors influencing speakers’ linguistic choices. Following the results of the study, a number of other situational ‘‘context external’’ and ‘‘context internal’’ (Blum-Kulka and House, 1989) factors were suggested as being more influential on the speakers’ strategy selection. More specifically, appeals were made to ‘context external’ factors such as rights and obligations concerning the differences which emerged in the Tuition fees and Assignment situation. Other ‘context internal’ factors such as ‘the request goal’ and the ‘urgency of the request’, along with cultural factors such as ‘positive politeness considerations’ were also found to successfully explain speakers’ linguistic choices. The results of this study clearly point towards a complex picture which relies on the culture-dependency of sociolinguistic variables. As Blum-Kulka and House (1989) point out, ‘‘the cultural factor operates in determining general levels of directness; it explains the differences between the directness means obtained between languages over and beyond situational variations’’ (1989:150). The picture which emerged from this study is a complex one and shows that a number of intervening social, contextual and cultural factors interact with one another in order to explain speakers’ linguistic choices. 2278 M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 The intricate ways with which these variables interact demand further research. What the exact role of factors such as ‘rights and obligations’, ‘urgency of the request’ and ‘politeness considerations’ is needs to be examined in greater depth and possibly with the employment of different methodological tools. Acknowledgements I would like to express my thanks to the participants of this study, the editor of the Journal of Pragmatics and the two anonymous reviewers for their detailed and insightful comments. References Aeginitou, Violeta, 1994. ‘‘Begin your sentences with a capital letter’’: facework and politeness in the Greek EFL Classroom. Unpublished PhD thesis. University of Lancaster, Lancaster. Antonopoulou, Eleni, 2001. Brief service encounters: gender and politeness. In: Bayraktaroglu, A., Sifianou, M. (Eds.), Linguistic Politeness Across Boundaries: The Case of Greek and Turkish. John Benjamins, Amsterdam, pp. 241–270. Austin, John L., 1962. How to do Things with Words. Clarendon Press, Oxford. Ballesteros Martin, Francisco Jose´ , 2001. La contesia espanola frente a la cortesia inglesa. Estudio pragmalinguistico de las exhortaciones impositivas. Estudios Ingleses de la Universidad Computense 9, 171–207. Ballesteros Martin, Francisco Jose´ , 2002. Mechanismos de atenuacion en espanole ingles. Implicaciones pragmaticas en la cortesia. Circulo de Linguisitica Aplicada a la Communicacion 11, 2–21. Bardovi-Harlig, Kathleen, Hartford, Beverly, 1990. Congruence in native and nonnative conversations: status balance in the academic advising session. Language Learning 40 (4), 467–501. Bardovi-Harlig, Kathleen, Hartford, Beverly, 2005. Institutional discourse and interlanguage pragmatic research. In: Bardovi-Harlig, K., Hartford, B. (Eds.), Interlanguage Pragmatics: Exploring Institutional Talk. Lawrence Erlbaum, London, pp. 7–36. Baxter, Leslie, 1984. An investigation of compliance gaining as politeness. Human Communication Research 10 (3), 427–456. Becker, Judith A., Kimmel, Herber, Bevill, Michael L., 1989. The Interactive effects of request form and speaker status on judgments of requests. Journal of Psycholinguistic Research 18 (5), 521–531. Billmyer, Kristine, Varghese, Manka, 2000. Investigating instrument-based pragmatic variability: effects of enhancing discourse completion tests. Applied Linguistics 21 (4), 517–552. Blum-Kulka, Shoshana, 1997. Discourse pragmatics. In: Van Dijk, T.A. (Ed.), Discourse as Social Interaction. Sage, London, pp. 38–63. Blum-Kulka, Shoshana, 1989. Playing it safe: the role of conventionality in indirectness. In: Blum-Kulka, S., House, J., Kasper, G. (Eds.), Cross-cultural Pragmatics: Requests and Apologies. Ablex, Norwood, NJ, pp. 37–70. Blum-Kulka, Shoshana, Danet, Brenda, Gherson, Rimona, 1985. The language of requesting in Israeli society. In: Forgas, J. (Ed.), Language and Social Situation. Springer Verlag, New York/Berlin, pp. 113–141. Blum-Kulka, Shoshana, House, Juliane, 1989. Cross-cultural and situational variation in requesting behavior. In: Blum-Kulka, S., House, J., Kasper, G. (Eds.), Cross-cultural Pragmatics: Requests and Apologies. Ablex, Norwood, NJ, pp. 123–154. Blum-Kulka, Shoshana, House, Juliane, Kasper, Gabriele (Eds.), 1989. Cross-Cultural Pragmatics: Requests and Apologies. Ablex, Norwood, NJ. Blum-Kulka, Shoshana, Olshtain, Elite, 1984. Requests and apologies: a cross-cultural study of speech act realization patterns. Applied Linguistics 5 (3), 196– 213. Blum-Kulka, Shoshana, Olshtain, Elite, 1986. Too many words: length of utterance and pragmatic failure. Studies in Second Language Acquisition 8 (2), 165– 179. Bonikowska, Malgorzata, 1988. The choice of opting out. Applied Linguistics 9 (2), 169–181. Brown, Penelope, Fraser, Colin, 1979. Speech as a marker of situation. In: Scherer, K., Gils, H. (Eds.), Social Markers in Speech. Cambridge University Press, Cambridge, pp. 33–63. Brown, Roger, Gilman, Albert, 1972. Pronouns of power and solidarity. In: Gigliogli, P. (Ed.), Language and Social Context. Penguin, Harmondsworth, pp. 252–282. Brown, Roger, Gilman, Albert, 1989. Politeness theory and Shakespeare’s four major tragedies. Language in Society 18 (2), 159–212. Brown, Penelope, Levinson, Stephen, 1978. Universals in language usage: politeness phenomena. In: Goody, E. (Ed.), Questions and Politeness. Cambridge University Press, Cambridge, pp. 56–289. Brown, Penelope, Levinson, Stephen, 1987. Politeness: Some Universals in Language Usage. Cambridge University Press, Cambridge. Cansler, David, Stiles, William, 1981. Relative status and interpersonal presumptuousness. Journal of Experimental Social Psychology 17 (5), 459–471. Cherry, Roger D., 1988. Politeness in written persuasion. Journal of Pragmatics 12 (1), 63–81. Craig, Robert T., Tracy, Karen, Spisak, Frances, 1986. The discourse of requests: assessment of a politeness approach. Human Communication Research 12 (4), 437–468. Clark, Herbert, 1979. Responding to indirect speech acts. Cognitive Psychology 11, 430–477. Cohen, Andrew, 2004. Assessing speech acts in a second language. In: Boxer, D., Cohen, A. (Eds.), Studying Speaking to Inform Second Language Learning. Multilingual Matters, Clevedon, UK, pp. 302–327. Cohen, Andrew, 2007. Teaching and assessing L2 pragmatics: What can we expect from learners? In: Paper Presented at the American Association for Applied Linguistics, CA April, 2007. Diaz Perez, Francisco Javier, 1999. Requesting in British English and Peninsular Spanish: a comparative analysis of politeness strategies. In: Alvarez Benito, G. (Ed.), Lenguas en Contacto. Mergablum, Seville, pp. 171–180. Economidou-Kogetsidis, Maria, 2003. Requesting strategies and cross-cultural pragmatics: Greek and English. Unpublished PhD thesis, University of Nottingham, Nottingham, UK. Economidou-Kogetsidis, Maria, 2004. Requestive directness in intercultural communication: Greek and English. In: Leigh, J., Loo, E. (Eds.), Outer Limits: A Reader in Communication and Behaviour Across Cultures. Language Australia, Melbourne, pp. 25–48. Economidou-Kogetsidis, Maria, 2005. Yes, tell me please, what time is the midday flight from Athens arriving?’’: telephone service encounters and politeness. Intercultural Pragmatics 2 (3), 253–273. Economidou-Kogetsidis, Maria, 2008. Internal and external mitigation in interlanguage request production: the case of Greek learners of English. Journal of Politeness Research: Language, Behaviour, Culture 4 (1), 111–138. Economidou-Kogetsidis, Maria, 2009. Interlanguage request modification: the use of lexical/phrasal downgraders and mitigating supportive moves. Multilingua 28 (1), 79–111. Ervin-Tripp, Susan, 1976. Is Sybil there? The structure of some American English directives. Language in Society 5 (1), 25–66. Fearch, Claus, Kasper, Gabriele, 1989. Internal and external modification in interlanguage request realization. In: Blum-Kulka, S., House, J., Kasper, G. (Eds.), Cross-cultural Pragmatics: Requests and Apologies. Ablex, Norwood, NJ, pp. 221–247. Felix-Brasdefer, Cesar J., 2006. Linguistic politeness in Mexico: refusal strategies among male speakers of Mexican Spanish. Journal of Pragmatics 38 (12), 2158–2187. M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 2279 Felix-Brasdefer, Cesar J., 2005. Indirectness and politeness in Mexican requests. In: Eddington, D. (Ed.), Selected Proceedings of the 7th Hispanic Linguistic Symposium. Cascadilla Press, Somerville, MA, pp. 66–78. Fraser, Bruce, 1975. Hedged performatives. In: Cole, P.,Morgan, J. (Eds.), Syntax and Semantics, Volume 3: Speech Acts. Academic Press, New York, pp. 187– 210. Fukushima, Saeko, 1990. Offers and requests: performance by Japanese learners of English. World Englishes 9 (3), 317–325. Fukushima, Saeko, 1996. Request strategies in British English and Japanese. Language Sciences 18 (3–4), 671–688. Fukushima, Saeko, 2000. Requests and Culture: Politeness in British English and Japanese. Peter Lang, Berlin. Goffman, Ervin, 1967. Interaction Ritual: Essays in Face-to-face Behavior. Anchor, New York. Golato, Andrea, 2003. Studying complement responses: a comparison of DCTs and recordings of naturally occurring talk. Applied Linguistics 24 (1), 90–121. Hartford, Beverly, Bardovi-Harlig, Kathleen, 1996. ‘‘At your earliest convenience’’: a study of written student requests to faculty. In: Bouton, L.F. (Ed.), Pragmatics and Language Learning, Monograph Series, Volume 7. Division of English as an International Language, University of Illinois, Urbana- Campaign, pp. 55–69. Hartford, Beverly, Bardovi-Harlig, Kathleen, 1992. Experimental and observational data in the study of interlanguage pragmatics. In: Bouton, L., Kachru, Y. (Eds.), Pragmatics and Language Learning Monograph series Volume 3. Division of English as an international language, University of Illinois, Urbana- Champaign, pp. 33–52. Hassall, Tim, 1999. Request strategies in Indonesian. Pragmatics 9 (4), 585–606. Hatch, Evelyn, Lazaraton, Anne, 1991. The Research Manual: Design and Statistics for Applied Linguistics. Newbury House, New York. Held, Gudrun, 1996. Two polite speech acts in contrastive view: aspects of the realization of requesting and thanking in French and Italian. In: Hellinger, M., Ammon, U. (Eds.), Contrastive Sociolinguistics. Mouton de Gruyter, Berlin, New York, pp. 363–384. Holmes, Janet, 1990. Apologies in New Zeland English. Language in Society 19, 155–199. Holtgraves, Thomas, 1986. Language structure in social interaction: perceptions of direct and indirect speech acts and interactants who use them. Journal of Personality and Social Psychology 51 (2), 305–314. Holtgraves, Thomas, Srull, Thomas K., Socall, Daniel, 1989. Conversation memory: the effects of speaker status on memory for the assertiveness of conversation remarks. Journal of Personality and Social Psychology 56 (2), 149–160. Holtgraves, Thomas, Yang, Joong-Nam, 1990. Politeness as universal: cross-cultural perceptions of request strategies and inferences based on their use. Journal of Personality and Social Psychology 59 (4), 719–729. House, Juliane, 1989. Politeness in English and German: the functions of ‘please’ and ‘bitte’. In: Blum-Kulka, S., House, J., Kasper, G. (Eds.), Cross-cultural Pragmatics: Requests and Apologies. Ablex, Norwood, NJ, pp. 96–119. Jaworski, Adam, 1994. Pragmatic failure in a second language: greeting responses in English by Polish students. International Journal of Applied Linguistics 32 (1), 41–55. Kasper, Gabriele, 1994. Politeness. In: Asher, R.E., Simpson, J.M.Y. (Eds.), The Encyclopaedia of Language and Linguistics. Pergamon Press, Oxford, pp. 3206– 3211. Kasper, Gabriele, 2008. Data collection in pragmatics research. In: Spencer-Oatey, H. (Ed.), Culturally Speaking: Culture, Communication and Politeness Theory. Continuum, London, pp. 316–341. Kasper, Gabriele, Rose, Kenneth, 2001. Pragmatics in language teaching. In: Rose, K., Kasper, G. (Eds.), Pragmatics in Language Teaching. Cambridge Applied Linguistics. Cambridge University Press, Cambridge, pp. 1–9. Kasper, Gabriele, Rose, Kenneth, 2002. Pragmatic Development in a Second Language. Blackwell, Oxford. Kwong, Jihyun, 2004. Expressing refusals in Korean and in American English. Multilingua 23 (4), 339–364. Labov, William, Fanshel, David, 1977. Therapeutic Discourse. Academic press, New York. Lee, Cynthia, 2005. A cross-linguistic study on the linguistic expressions of Cantonese and English requests. Pragmatics 15 (2–3), 395–422. Le Pair, Rob, 1996. Spanish requests strategies: a cross-cultural analysis from an intercultural perspective. Language Sciences 118 (3–4), 651–670. Leichty, Greg, Applegate, James, 1991. Social-cognitive and situational influences on the use of face-saving persuasive strategies. Human Communication Research 17 (3), 451–484. Lustig, Myron W., King, Stephen, 1980. The effect of communication apprehension and situation on communication strategy choices. Human Communication Research 7 (1), 74–82. Makri-Tsilipakou, Marianthi, 2001. Congratulations and bravo! In: Sifianou, M., Bayraktaroglu, A. (Eds.), Linguistic Politeness Across Boundaries: The Case of Greek and Turkish. John Benjamins, Amsterdam, pp. 137–176. Ma´rquez Reiter, Rosina, 2000. Linguistic Politeness in Britain and Uruguay: A Contrastive Study of Requests and Apologies. John Benjamins, Amsterdam/ Philadelphia. Ma´rquez Reiter, Rosina, 2002. A contrastive study of conventional indirectness in Spanish: evidence from Peninsular and Uruguayan Spanish. Pragmatics 12 (2), 135–151. Ma´rquez Reiter, Rosina, Placencia, Maria Elena, 2005. Spanish Pragmatics. Palgrave McMillan, New York. McLaughlin, Margaret, Cody, Michael, O’Hair, Henry D., 1983. The management of failure events: some contextual determinants of accounting behaviour. Human Communication Research 9 (3), 208–224. Nelson, Gayle L., Carson, Joan, Al Batal, Mahmoud, El Bakary, Waguida, 2002. Corss-cultural pragmatics: strategy use in Egyptian Arabic and American English refusals. Applied Linguistics 23 (2), 163–189. Olshtain, Elite, 1989. Apologies across languages. In: Blum-Kulka, S., Kasper, G., House, J. (Eds.), Cross-Cultural Pragmatics: Requests and Apologies. Ablex, Norwood, NJ, pp. 155–217. Pavlidou, Theodossia, 1994. Contrasting German–Greek politeness and the consequences. Journal of Pragmatics 21 (5), 487–511. Pavlidou, Theodossia, 1997. The last five turns: preliminary remarks on closings in Greek and German telephone calls. International Journal of the Sociology of Language 126, 145–162. Pavlidou, Theodossia, 1998. Greek and German telephone closings: patterns of confirmation and agreement. Pragmatics 8 (1), 79–94. Rintell, Ellen M., 1981. Sociolinguistic variation and pragmatic ability: a look at learners. International Journal of the Sociology of Language 27, 11–33. Rintell, Ellen, Mitchell, Candace J., 1989. Studying requests and apologies: an inquiry into method. In: Blum-Kulka, S., Kasper, G., House, J. (Eds.), Cross- Cultural Pragmatics: Requests and Apologies. Ablex, Norwood, NJ, pp. 248–272. Rose, Kenneth, 2000. An exploratory cross-sectional study of interlanguage pragmatic development. Studies in Second Language Acquisition 22, 27–67. Sasaki, Miyuki, 1998. Investigating EFL students’ production of speech acts: a comparison of production questionnaires and role plays. Journal of Pragmatics 30 (4), 457–484. Scarcella, Robin, Brunak, Joanna, 1981. On speaking politely in a second language. International Journal of Sociology of Language 27, 59–75. Schauer, Gila A., 2007. Finding the right words in the study abroad context: the development of German learners’ use of external modifiers in English. Intercultural Pragmatics 4 (2), 193–220. Scollon, Ron, Scollon, Suzanne, 1983. Face in interethnic communication. In: Richards, J., Schmidt, R. (Eds.), Language and Communication. Longman, London/New York, pp. 156–188. Sifianou, Maria, 1989. On the telephone again!: differences in telephone behaviour Language in Society 18 (4), 524–544. Sifianou, Maria, 1992a. Politeness Phenomena in England and Greece, A Cross-cultural Perspective. Clarendon Press, Oxford. Sifianou, Maria, 1992b. The use of diminutives in expressing politeness: modern Greek versus English. Journal of Pragmatics 17 (2), 155–173. Sifianou, Maria, 2001 [1989] On the telephone again! Differences in telephone behavior. In: Georgakopoulou, A., Spanaki, M. (Eds.), A Reader in Greek Sociolinguistics. Studies in Modern Greek Language, Culture and Communication Peter Lang, pp. 137–159. First published in 1989 in Language in Society 18 (4), 527–544. 2280 M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 Slugoski, Ben, Turnbull, William, 1988. Cruel to be kind and kind to be cruel: sarcasm, banter and social relations. Journal of Language and Social Psychology 7 (2), 101–121. Spencer-Oatey, Helen, 1992. Cross-Cultural politeness: British and Chinese conceptions of the tutor–student relationship. Unpublished PhD thesis. University of Lancaster, Lancaster. Spencer-Oatey, Helen, 1993. Conceptions of social relations and pragmatics research. Journal of Pragmatics 20 (1), 27–47. Spencer-Oatey, Helen, 1996. Reconsidering power and distance. Journal of Pragmatics 26 (1), 1–24. Spencer-Oatey, Helen, 1997. Unequal relationships in high and low power distance societies: a comparative study of tutor–student role relations in Britain and China. Journal of Cross-Cultural Psychology 28 (3), 284–302. Stewart, Miranda, 2005. Politeness in Britain. In: Hickey, L., Stewart, M. (Eds.), Politeness in Europe. Multilingua Matters, Clevedon, pp. 116–129. Takahashi, Satomi, 1992. Transferability of indirect request strategies. University of Hawaii Working Papers in ESL 11 (1), 69–124. Tannen, Deborah, 1981. Indirectness in discourse: ethnicity as conversational style. Discourse Processes 4 (3), 221–238. Thomas, Jenny, 1983. Cross-cultural pragmatic failure. Applied Linguistics 4 (2), 91–112. Thomas, Jenny, 1995. Meaning in Interaction—An Introduction to Pragmatics. Longman, London. Triandis, Harry, Vassiliou, Vasso, 1972. A comparative analysis of subjunctive culture. In: Triandis, H. (Ed.), The Analysis of Subjective Culture, Comparative Studies in Behavioral Science. Wiley, New York, pp. 299–335. Trosborg, Anna, 1995. Interlanguage Pragmatics. Mouton de Gruyter, Berlin. Van Mulken, Margot, 1996. Politeness markers in French and Dutch requests. Language Sciences 8 (3–4), 689–702. Vassiliou, Vasso, Triandis, Harry C., Vassiliou, George, McGuire, Howard, 1972. Interpersonal contact and stereotyping. In: Triandis, H.C. (Ed.), The Analysis of Subjective Culture, Comparative Studies in Behavioral Science. Wiley, New York, pp. 89–115. Wierzbicka, Anna, 1991. Trends in Linguistics, Studies and Monographs 53, Cross Cultural Pragmatics: The Semantics of Human Interaction. Mouton de Gruyter, Berlin, New York. Woodfield, Helen, 2005. Problematizing discourse completion tasks: voices from verbal report. In: Paper Presented at the 14th World Congress of Applied Linguistics (AILA), Madison, WI, USA. Woodfield, Helen, 2006. Requests in English: ESL learners’ responses to written discourse completion tasks. In: Paper Presented at the 31st International LAUD Symposium. Intercultural Pragmatics, Linguistics, Social and Cognitive Approaches, Landau/Pfalz, Germany. Woodfield, Helen, 2008. Problematizing discourse completion tasks: voices from verbal report. Evaluation and Research in Education 21, 43–69. Maria Economidou-Kogetsidis is currently an assistant professor at the University of Nicosia, Cyprus. She holds a doctorate in Cross-cultural pragmatics from the School of English Studies of the University of Nottingham, UK. Her research areas are cross-cultural communication, interlanguage and intercultural pragmatics, sociopragmatics, pragmalinguistics and politeness. Her publications have appeared in Intercultural Pragmatics, Journal of Politeness Research and Multilingua. Her current research focuses on the pragmatic performance of Greek Cypriot learners of English. M. Economidou-Kogetsidis / Journal of Pragmatics 42 (2010) 2262–2281 2281 according case 1 and 2 to choose a research topic, must talk about these two case. And research area should abut pratmatics. Structure: 1. introduction 2. a summary( 3. a critical review( 4. conclusion 2?please analysis by pragmatics 3?lecture 7 is about pragmatics,the ppt I have upload in additional file. 4?referenceHarvard style, and I need 13 reference.