The Effect of L1 Linguistic and Cultural Background on L2 Pragmatic Competence

Abstract

This study examined the pragmatic competence and awareness of L2 learners of English, with a focus on their request strategies in L2 English, levels of (in)directness, request perspective, internal and external modifications, and the impact of tasks. The participants comprised 80 Cypriot Greek (CG) undergraduate students. The pragmatic tests were designed based on Speech Act Theory and included discourse completion tasks (DCTs) and multiple-choice questionnaires (MCQs). Additionally, a role-play and an interview task were employed to assess L2 learners’ pragmatic competence in oral discourse. The tasks featured four conditions related to power and social distance: [+social distance, status equals], [−social distance, status equals], [+social distance, hearer dominance], and [−social distance, hearer dominance]. Overall, the results indicated that L2 learners of English tended to prefer conventionally indirect requests, suggesting that L1 linguistic and cultural backgrounds influence expressions of politeness. Face-threatening situations with a high degree of imposition elicited more conventionally indirect requests. Data analysis revealed contextual variation but not a clear task effect. Factors such as social distance, hearer dominance, power, familiarity, and imposition impacted the degree of (in)directness, request perspective, and both internal and external modifications.

1. Introduction

Request speech acts and mitigating strategies in L1 and L2 have been widely and thoroughly studied across languages and cultures, revealing cross-linguistic differences (Blum-Kulka & Olshtain, 1984; Economidou-Kogetsidis, 2008, 2009, 2012, 2013; Ruiz de Zarobe & Ruiz de Zarobe, 2012; Webman Shafran, 2019). The appropriate use of request forms in social contexts depends on factors such as L1, culture, age and gender of the interlocutors, social distance, power relations, and the degree of imposition. Native speakers acquire pragmatic competence subconsciously through exposure to their native language from birth, while L2 learners must make a conscious effort to learn the pragmatic rules of the L2 (Blum-Kulka, 1982; Bardovi-Harlig & Hartford, 1990; Trosborg, 1995; Nelson et al., 2002; Webman Shafran, 2019).

This study aims to investigate the (in)directness of L2 English request strategies used by L1 CG undergraduate students, as well as request perspective and internal and external modification, and whether these factors are affected by the type of task—oral versus written—along with power and social distance variables. The focus is on their pragmatic behavior and interactional practices in different social settings, as well as the possible transfer from L1 CG to L2 English (Taguchi & Roever, 2017; Ogiermann & Bella, 2020).

The paper is structured as follows: Section 2 provides an overview of the theoretical framework of request speech acts, interlanguage pragmatics, request perspective, and internal and external modifications. Section 3 presents the methodology. Results are described in Section 4 and discussed in Section 5. Section 6 concludes the paper.

2. Request Speech Acts

2.1. (In)Directness and Request Speech Acts

Interlanguage pragmatics and politeness research address the issues of speech act production and mitigation strategies employed by L2 learners in intercultural communication, highlighting how these strategies differ from those native speakers due to L1 interference (Woodfield, 2006; Economidou-Kogetsidis, 2008).

As suggested by Searle (1969, 1975) and P. Brown and Levinson (1987), requests are inherently face-threatening speech acts. A speaker asks the hearer to perform an action, which poses a threat to the hearer’s negative face—the right of a person for freedom and non-distraction (P. Brown & Levinson, 1987). According to Austin (1962), Searle (1975), and Blum-Kulka and Olshtain (1984), there are three major levels of directness: direct, conventionally indirect, and non-conventionally indirect. The meaning and function of a direct utterance are transparent (e.g., “Close the door!”); there is no clear match between form and function in a conventionally indirect request (e.g., “Could you please close the door?”), as it uses the form of a question for the request function. However, the intended meaning can still be inferred, as the form is conventionalized as a request. The non-conventionally indirect strategy is the least transparent and clear, and it is the most indirect (e.g., “It is noisy outside.”), as it uses a statement for an implicit request and requires the hearer to make an extra effort to infer the implied meaning (Webman Shafran, 2019).

Previous research has shown that conventionally indirect requests are the most preferred strategies universally, although there are some variations across languages and cultures (Blum-Kulka et al., 1989; Trosborg, 1995; Stavans & Webman Shafran, 2017; Webman Shafran, 2019).

According to Blum-Kulka (2005) and Economidou-Kogetsidis (2008), there are three important dimensions to consider when analyzing the request speech acts of L2 learners: request strategy and (in)directness, internal or lexical modification, and external modification. The degree of indirectness and the amount and type of modification depend on various social and situational factors, such as social distance, power, and the imposition of the speech act (P. Brown & Levinson, 1987; Economidou-Kogetsidis, 2008). Interlocutors are expected to implement indirectness and mitigation strategies in the face of threats due to hearer dominance, power, social distance, and the imposition involved (Holmes & Stubbe, 2003). In this study, we adopt R. Brown and Gilman’s (1972) definition of power, meaning the ability of one interlocutor to control the behavior of the other(s) in a non-reciprocal way due to various factors, such as their position in society, socio-economic status, age, or background (Economidou-Kogetsidis, 2008).

English and CG differ in terms of cultural norms and values, as well as in the level of (in)directness: CG is more direct than English, which uses more mitigation devices (Katriel, 1986; Cohen, 1987; Webman Shafran, 2019). In CG, interlocutors tend to choose a direct strategy when there is equal status and no social distance between them, while a conventional indirect strategy is implemented in non-congruent conditions, such as social distance and hearer dominance. Due to cultural values in English-speaking societies, such as individuality, freedom of action, and privacy, native speakers of English seem to be less affected by power relations and social context; thus, they use conventional indirect strategies in nearly all situations, except in contexts of low imposition and intimate relationships, where direct requests are employed (Leech, 1983; P. Brown & Levinson, 1987; Blum-Kulka et al., 1989; Reiter, 2000; Tawalbeh & Al-Oqaily, 2012; Webman Shafran, 2019).

2.2. Internal and External Modification in Requests

As suggested by Blum-Kulka (2005), mitigation devices can help save or protect face; internal modification is associated with positive politeness, whereas external modification is associated with negative politeness.

Internal modification can be distinguished into downgraders and upgraders, which modify the head act of the request utterance and have either a softening or an intensifying effect, respectively (Blum-Kulka, 1989; Economidou-Kogetsidis, 2008). Downgraders can be syntactic, lexical, or phrasal; the latter is the focus of this paper. The analysis of lexical and phrasal downgraders in this research is based on the Cross-Cultural Speech Act Realisation Project (CCSARP) (Blum-Kulka & Olshtain, 1984; Blum-Kulka et al., 1989; Economidou-Kogetsidis, 2008, p. 115). The markers include “please”; consultative devices/openers (e.g., “would you mind”, “do you think”, “would it be all right if”, “is it/would it be possible”, “do you think I could …”, “is it all right?”); downtoners (e.g., “possibly”, “perhaps”); understaters (e.g., “a bit”, “a little”); hedges (e.g., “sort of ”, “rather”, “quite”); subjectivizers (e.g., “I’m afraid”, “I wonder”, “I think/suppose”); cajolers (e.g., “You know”, “I’d really like …”); and appealers (e.g., “Clean the table dear, will you?”, “…, ok/right?”).

External modification, which is realized in the form of supportive moves before or after the head act, is related to the context of the request speech act and its illocutionary force (Economidou-Kogetsidis, 2008). Explanations and justifications, serving as external modifiers, can have either an intensifying or a mitigating effect; the latter is the focus of the current paper.

In this study, we have implemented the external modification categories and classifications proposed by Economidou-Kogetsidis (2008, p. 116), based on the CCSARP project (Blum-Kulka & Olshtain, 1984; Blum-Kulka et al., 1989) and subsequent research (Blum-Kulka, 1985; Blum-Kulka & Levenston, 1987; Van Mulken, 1996). The categories include:

(1)

Grounders (i.e., explanations or justifications for a request) (e.g., “Judith, I missed the class yesterday. Could I borrow your notes?”);

(2)

Disarmers (e.g., “I know you don’t like lending out your notes, but could …”);

(3)

Preparators (e.g., “I’d like to ask you something …”, “Don’t you live in the same street as I do …?”);

(4)

Getting a precommitment (e.g., “Could you do me a favor? …”);

(5)

Promise of reward/making a promise (e.g., “Could you give me a lift home? We’ll use my car tomorrow.”);

(6)

Imposition minimizers (e.g., “Would you give me a lift, but only if you’re going my way?”);

(7)

Apologies (e.g., “I’m sorry to bother you, but could I request a few days off?”).

2.3. Request Perspective

The request perspective is one of the key areas of interest in research on request speech acts (e.g., Blum-Kulka, 1989; Reiter, 2000; Ogiermann, 2009; Ogiermann & Bella, 2020). As specified by Blum-Kulka (1989, 1991), there are four types of requests: hearer-oriented (e.g., “Can you?”), speaker-oriented (e.g., “Can I?”), inclusive (e.g., “Can we?”), and impersonal (e.g., “Can one?”). The speaker-oriented request is considered to be marginally more polite than the hearer-oriented request (Leech, 1983; Ogiermann & Bella, 2020).

The choice of request strategies and perspectives is associated with “the conventionalization of perspectives within strategy types” (Blum-Kulka, 1989, p. 59), variations in the degree of coerciveness (Blum-Kulka, 1991), and social implications (Ogiermann & Bella, 2020). In the case of the speaker perspective, imposition is minimized because the hearer, who is expected to perform an action, is not named. Conversely, the inclusive or joint perspective suggests solidarity among the participants in the communication (Blum-Kulka & Levenston, 1987; Economidou-Kogetsidis, 2013).

Cross-cultural research so far has shown a tendency for the choice of the hearer perspective in requests (Ogiermann & Bella, 2020). Blum-Kulka (1989) found that speakers of Australian English, Argentinian Spanish, Canadian French, and Hebrew prefer hearer-oriented conventionally indirect requests in DCTs. Similar results were obtained in Márquez Reiter’s (2000) role-play study, which focused on British English and Uruguayan Spanish. Ogiermann’s (2009) DCT study confirmed these findings, highlighting the prominence of the hearer perspective among Polish and Russian speakers compared to English and German speakers. The influence of language proficiency on the choice and preference for the hearer perspective was demonstrated in Pinto’s (2005) DCT, based on data from Mexican Spanish and American English speakers, as well as in Félix-Brasdefer’s (2007) role-play study involving American learners of Spanish.

L2 learners of English exhibit a strong preference for the hearer perspective compared to L1 speakers (German and Greek learners of L2 English: Woodfield, 2008; Cypriot Greek learners of L2 English: Woodfield & Economidou-Kogetsidis, 2010; Economidou-Kogetsidis, 2012, 2013; Chinese learners of L2 English: Lin, 2009; email requests: Sell & Haggerty, 2019). According to Ogiermann and Bella (2020), when speakers select the level of directness, this choice influences the request perspective: direct requests (imperative) adopt the hearer perspective, while indirect requests can encompass speaker, hearer, inclusive, or impersonal perspectives.

This study aims to investigate the following research questions:

• Is there any effect of ±social distance, hearer dominance, power, familiarity, and imposition on the following:
  • The degree of (in)directness
  • Request perspective
  • Internal modification
  • External modification
  of the request speech acts produced by L1 CG learners of L2 English?
• Is there any task effect (written vs. oral) on the L2 pragmatic competence and performance of L2 English learners?

3. Methodology

3.1. Participants

The participants in the study were 80 Greek Cypriot (CG) first-year undergraduate students (ages 17–25; 44 male and 36 female, with normal speech and hearing) who were learners of L2 English at a private, English-speaking university in Cyprus. Their L2 English proficiency ranged from low intermediate to advanced, with IELTS scores between 5 and 9 (mean score: 6.5). The participants reported no visual, reading, or language difficulties. Although English was not their major, it was a compulsory subject, taught twice per week for a total of four hours. Most of the students had contact with English speakers, and 52 of them had visited English-speaking countries (see

Table 1. Participants: background information.

Variables	Mean	SD	Min.	Max.
Age	19.2	2.4	17	27
AoO	11.3	2.5	6	19
Years of learning L2 English	19.8	2.7	4	18
Gender	Male	Female	Proficiency
	44	36	IELTS 5.0–5.5	18
Visits to English-speaking countries	Yes	No	IELTS 6.0–6.5	22
	52	28	IELTS 7.0–7.5	19
Contacts with people by using English	Yes	No	IELTS 8.0–8.5	14
	80	0	IELTS 9.0–9.5	7
			Mean	6.5
University	Private		80

). A non-probability convenience sample was used, as participants were selected based on their possession of targeted characteristics and availability, which aimed to ensure the homogeneity of the population participating in the study. Student participation was voluntary and in line with confidentiality considerations.

3.2. Materials

The pragmatic tests, based on the Speech Act Theory (Austin, 1962) as a theoretical framework, were implemented in the research. Specifically, a discourse completion task (DCT) was used, adapted from K. R. Rose (1994). A multiple-choice questionnaire (MCQ) was also employed, again adapted from K. R. Rose (1994). Additionally, an oral role-play task (ORP) was used to assess L2 learners’ pragmatic competence in oral discourse, as described in Appendix A.

MCQs, DCTs, and role-plays are popular tasks for eliciting comparable data among researchers examining requests from a cross-cultural and interlanguage pragmatic perspective (Blum-Kulka & Levenston, 1987; Blum-Kulka, 1989; Pinto, 2005; Woodfield, 2008; Lin, 2009; Ogiermann, 2009; Woodfield & Economidou-Kogetsidis, 2010; Reiter, 2000; Félix-Brasdefer, 2007; Economidou-Kogetsidis, 2012, 2013). One of the disadvantages of the DCT methodology is that it differs from naturally occurring speech and lacks its characteristics (Golato, 2003; Ogiermann, 2018). Therefore, a mixed-methods approach—a combination of oral and written data—was chosen to increase the validity and reliability of the research.

The multiple-choice questionnaire consisted of eight situations, with two for each condition. For each situation, the students had to choose one from four: direct strategy, conventionally indirect strategy, non-conventionally indirect strategy, and hint or no act of request. The discourse completion task also included eight situations, two for each condition, where students were prompted to produce a request speech act. Additionally, the oral role-play task featured eight situations as well, two for each condition, to elicit oral request speech production from L2 English learners. Each of the tasks examined four conditions regarding power and distance variables: [+social distance, status equals], [−social distance, status equals], [+social distance, hearer dominance], and [−social distance, hearer dominance]. Linguistic (socio-economic) background questionnaires were also utilized.

3.3. Procedure and Data Analysis

First, the participants were asked to complete the MCQ, followed by the DCT and the ORP. There was at least a two-week gap between each task. Both written and oral data were analyzed in terms of pragmatic appropriateness and linguistic accuracy; the amount of information; the degree of politeness, directness, and formality (Tannen, 1993); the request perspective (Ogiermann & Bella, 2020); and internal and external modification (Economidou-Kogetsidis, 2008). Coding categories from the CCSARP (Blum-Kulka et al., 1989) were employed to analyze the request strategies of the participants: direct strategy, conventionally indirect strategy, non-conventionally indirect strategy, or no act of request. Each situation, and subsequently each condition per task, was analyzed independently and then compared within the tasks and between/among the tasks using non-parametric statistics, specifically Pearson’s chi-square tests.

4. Results

4.1. Request Strategies and (In)Directness: Three Tasks

The overall analysis of the data established that L2 learners tend to choose a conventionally indirect strategy in the three tasks—the ORP, the MCQ, and the DCT—with the highest scores observed in the latter (see Example 1):

(1)

May I borrow your mobile phone for a bit? I really need to call my parents.

This finding aligns with previous research by Ogiermann (2009) and Economidou-Kogetsidis (2013). The MCQ task exhibited the highest rate of both the direct strategy and the non-conventionally indirect strategy among the tasks, while the ORP showed a higher prevalence of the no act of request strategy. This observation is consistent with the findings of K. R. Rose (1994), K. Rose and Ono (1995), and Hinkel (1997). One possible explanation for these results could be the effect of face-to-face interaction in the oral task, as participants might avoid using direct requests (Economidou-Kogetsidis, 2013). See

Table 2. Degree of (in)directness: total scores in the three tasks.

Strategy Task	Direct	Conventionally Indirect	Non-Conventionally Indirect	No Act of Request
MCQ	85/640	446/640	77/640	32/640
	13.29%	69.64%	12.10%	4.97%
DCT	51/640	545/640	38/640	6/640
	7.96%	85.18%	5.93%	0.93%
ORP	52/640	452/640	61/640	75/640
	8.18%	70.68%	9.48%	11.66%
Chi-square DCT vs. ORP	χ2 = 12.293, df = 1, p = 0.714	χ2 = 39.337, df = 1, p = 0.589	χ2 = 5.533, df = 1, p = 0.938	χ2 = 2.734, df = 1, p = 0.950
Chi-square DCT vs. MCQ	χ2 = 15.721, df = 1, p = 0.473	χ2 = 45.236, df = 1, p = 0.037	χ2 = 22.858, df = 1, p = 0.118	χ2 = 4.288, df = 1, p = 0.368
Chi-square MCQ vs. ORP	χ2 = 13.104, df = 1, p = 0.665	χ2 = 38.715, df = 1, p = 0.306	χ2 = 11.884, df = 1, p = 0.455	χ2 = 9.139, df = 1, p = 0.331
MCQ = multiple- choice questionnaire; DCT = discourse completion task; ORP = Oral Role Play.

for further details.

The results indicated that social distance, hearer dominance, power, familiarity, and imposition have an effect on the degree of (in)directness, as there is a statistically significant difference between the conditions [±social distance, hearer dominance] regarding the direct strategy, the conventionally indirect strategy, the non-conventionally indirect strategy, and the no act of request (see Appendix B for the Pearson chi-square statistics). In the DCT and ORP tasks, participants exhibited more instances of the conventionally indirect strategy in the status-equals conditions (see Example 2), while in the MCQ, they demonstrated this strategy more frequently in the ±hearer dominance conditions.

(2)

Hey man, can I borrow your pen?

These findings are at odds with the recent findings by Webman Shafran (2019) and may be attributed to L1 transfer effects.

The DCT and the ORP had more instances of the direct strategy in the +hearer dominance condition (see Example 3), while the MCQ displayed this strategy more frequently in the ±hearer dominance conditions.

(3)

Give me a lift!

One explanation for this could be the effect of the setting and the situation, as previous research on Greek language and culture (Sifianou, 1992a, 1992b; Pavlidou, 1994; Antonopoulou, 2001; Economidou-Kogetsidis, 2008) suggests that some requests may not threaten the positive or negative face of an individual if they take place in contexts such as shops, among family, or over the phone. The non-conventionally indirect strategy was used in all conditions except for the −social distance/status equals in the DCT and ORP, as well as the −social distance/hearer dominance condition in the MCQ; see Example 4:

(4)

My phone is dead. Would someone be so kind and lend me their phone for a while?

Regarding the no act of request strategy, it was not employed at all in the DCT, while it was primarily used in the −social distance conditions in both the MCQ and the ORP (see

Table 3. Degree of (in)directness: four conditions: three tasks.

+Social Distance Status Equals	Direct	Conventionally Indirect	Non-Conventionally Indirect	No Act of Request
(1) +Social distance/status equals condition
MCQ	5/160	135/160	20/160	0/160
	3.17%	84.14%	12.69%	0%
DCT	9/160	143/160	8/160	0/160
	5.62%	89.38%	5%	0%
ORP	10/160	132/160	7/160	11/160
	6.03%	82.75%	4.31%	6.89%
Chi-square DCT vs. ORP	χ2 = 0.631, df = 1, p = 0.960	χ2 = 0.866, df = 1, p = 0.929	χ2 = 0.405, df = 1, p = 0.817	N/A
Chi-square DCT vs. MCQ	χ2 = 0.405, df = 1, p = 0.817	χ2 = 0.837, df = 1, p = 0.933	χ2 = 0.517, df = 1, p = 0.972	N/A
Chi-square MCQ vs. ORP	χ2 = 0.496, df = 1, p = 0.780	χ2 = 3.770, df = 1, p = 0.438	χ2 = 0.211, df = 1, p = 0.900	χ2 = 2.421, df = 1, p = 0.298
(2) −Social distance/status equals condition
MCQ	53/160	75/160	20/160	12/160
	33.33%	46.82%	12.69%	7.16%
DCT	11/160	147/160	2/160	0/160
	6.87%	91.87%	1.26%	0%
ORP	10/160	121/160	4/160	25/160
	6.03%	75.86%	2.58%	15.53%
Chi-square DCT vs. ORP	χ2 = 0.875, df = 1, p = 0.928	χ2 = 3.941, df = 1, p = 0.414	χ2 = 0.056, df = 1, p = 0.814	N/A
Chi-square DCT vs. MCQ	χ2 = 6.016, df = 1, p = 0.198	χ2 = 4.579, df = 1, p = 0.333	χ2 = 0.265, df = 1, p = 0.876	N/A
Chi-square MCQ vs. ORP	χ2 = 1.360, df = 1, p = 0.851	χ2 = 3.774, df = 1, p = 0.437	χ2 = 0.875, df = 1, p = 0.646	χ2 = 4.293, df = 1, p = 0.038
(3) −Social distance/hearer dominance condition
MCQ	0/160	146/160	6/160	8/160
	0%	91.28%	3.96%	4.76%
DCT	20/160	122/160	15/160	3/160
	12.50%	76.25%	9.38%	1.87%
ORP	11/160	105/160	18/160	26/160j
	6.89%	65.51%	11.20%	16.37%
Chi-square DCT vs. ORP	χ2 = 2.421, df = 1, p = 0.120	χ2 = 7.485, df = 1, p = 0.112	χ2 = 4.371, df = 1, p = 0.112	χ2 = 0.932, df = 1, p = 0.627
Chi-square DCT vs. MCQ	N/A	χ2 = 1.085, df = 1, p = 0.897	χ2 = 0.098, df = 1, p = 0.952	χ2 = 0.195, df = 1, p = 0.907
Chi-square MCQ vs. ORP	N/A	χ2 = 3.039, df = 1, p = 0.551	χ2 = 1.581, df = 1, p = 0.209	χ2 = 3.491, df = 1, p = 0.479
(4) +Social distance/hearer dominance condition
MCQ	27/160	91/160	28/160	14/160
	16.66%	57.15%	17.46%	8.73%
DCT	13/160	131/160	13/160	3/160
	8.12%	81.89%	8.12%	1.87%
ORP	25/160	92/160	29/160	14/160
	15.51%	57.75%	18.12%	8.62%
Chi-square DCT vs. ORP	χ2 = 0.770, df = 1, p = 0.942	χ2 = 2.817, df = 1, p = 0.589	χ2 = 1.653, df = 1, p = 0.438	χ2 = 0.187, df = 1, p = 0.911
Chi-square DCT vs. MCQ	χ2 = 6.060, df = 1, p = 0.195	χ2 = 906, df = 1, p = 0.924	χ2 = 381, df = 1, p = 0.827	χ2 = 5.540, df = 1, p = 0.019
Chi-square MCQ vs. ORP	χ2 = 2.852, df = 1, p = 0.583	χ2 = 1.411, df = 1, p = 0.842	χ2 = 1.740, df = 1, p = 0.783	χ2 = 1.957, df = 1, p = 0.376
MCQ = multiple-choice questionnaire; DCT = discourse completion task; ORP = Oral Role Play.

).

Overall, the analysis of the data did not reveal any significant task effect (oral vs. written) regarding the use of request strategies, with the exception of a few differences: between the DCT and MCQ tasks concerning the conventionally indirect strategy; between the MCQ and ORP regarding the no act of request in the −social distance/status equals; and between the DCT and the MCQ in terms of the no act of request in the +social distance/hearer dominance condition.

4.2. Request Perspective: Written vs. Oral Tasks

The analysis of the data showed that in both oral and written tasks, the CG participants tended to adopt a hearer perspective, with a slight preference for this perspective in the ORP compared to the DCT; see Example 5.

(5)

Excuse me sir, I forgot my wallet. Would you be so kind to lend me a few coins?

It should be noted that CG students implemented more speaker perspective in the written mode than in the oral mode; see Example 6.

(6)

Hello Ms. Do you know that my job is very hard. I would like to get more money per month, or I will resign.

Conversely, the use of the impersonal perspective was higher in the oral mode, which partially supports the findings of Economidou-Kogetsidis (2013) in her study on natural and DCT requests by L1 CG speakers in L2 English, particularly regarding hearer and speaker perspectives. However, in the current study, the frequency of using inclusive/joint and impersonal perspectives was higher than in her findings; see Example 7.

(7)

Is it possible to extend the deadline for the assignment submission please?

Overall, the data analysis indicated a task effect on request perspective, with the oral task eliciting more hearer perspective (nearly statistically significant) than the written task (see

Table 4. Request perspective: written vs. oral task.

Request Perspective	Hearer	Speaker	Inclusive	Impersonal
DCT	347/634	226/634	12/634	49/634
	54.73%	35.64%	1.89%	7.74%
ORP	345/565	121/565	5/565	94/565
	61%	21.33%	0.86%	16.81%
Chi-square	χ2 = 72.703, df = 1, p = 0.066	χ2 = 36.017, df = 1, p = 0.142	χ2 = 0.995, df = 1, p = 0.319	χ2 = 9.828, df = 1, p = 0.875
DCT = Discourse completion task; ORP = Oral Role Play.

). The prevalence of the hearer perspective in the L2 English data of the current study supports the general trend observed in previous research (e.g., Biesenbach-Lucas, 2007; Woodfield, 2008; Lin, 2009; Economidou-Kogetsidis, 2012; Ogiermann & Bella, 2020).

It appears that social distance, hearer dominance, power, familiarity, and imposition have an effect on request perspective, as there are statistically significant differences among the conditions of [±social distance, hearer dominance] in terms of hearer, speaker, and impersonal perspectives (see Appendix C for the Pearson chi-square test results). In the DCT, participants showed a higher score for the hearer perspective in conditions with +social distance (see Example 8), while in the ORP, a higher score was observed with –hearer dominance.

(8)

Excuse me, sir. Can you tell me please the directions to the city centre? I had never been in this city before. So, I do not know how to go there.

Regarding the speaker perspective, its rate was higher in the −social distance condition in the written mode (see Example 9) and in the +hearer dominance in the oral mode.

(9)

I think that I worked very hard the last months, and I deserve a higher salary.

Interestingly, the −social distance/+hearer dominance condition elicited more speaker perspective than hearer perspective in both modes; see Example 10.

(10)

Sir, I have to take part in the university competition. It is really important for me to focus on it, without being stressed for the assignment submission. So, I’d like to ask for an extension, if it is possible. Thanks.

This evidence corroborates the general trends observed in research in this field.

The impersonal perspective was used only in the −social distance/+hearer dominance condition in the DCT (see Example 11), whereas in the ORP, it was employed across all conditions (see

Table 5. Request perspective: four conditions: written vs. oral tasks.

Request Perspective	Hearer	Speaker	Inclusive	Impersonal
(1) +Social distance/status equals condition
DCT	87/160	71/160	0/160	2/160
	54.38%	44.37%	0%	1.25%
ORP	130/149	5/149	0/149	14/149
	87.06%	3.46%	0%	9.48%
Chi-square	χ2 = 8.781, df = 1, p = 0.067	χ2 = 2.863, df = 1, p = 0.239	N/A	N/A
(2) −Social distance/status equals condition
DCT	94/160	58/160	1/160	7/160
	58.75%	36.25%	0.63%	4.37%
ORP	105/135	6/135	0/135	24/135
	77.58%	4.32%	0%	18.10%
Chi-square	χ2 = 0.541, df = 1, p = 0.969	χ2 = 0.356, df = 1, p = 0.837	N/A	N/A
(3) −Social distance/hearer dominance condition
DCT	60/157	62/157	9/157	26/157
	38.22%	39.49%	5.73%	16.56%
ORP	46/134	61/134	0/134	27/134
	34.48%	45.68%	0%	19.84%
Chi-square	χ2 = 8.360, df = 1, p = 0.079	χ2 = 0.857, df = 1, p = 0.931	N/A	χ2 = 9.293, df = 1, p = 0.054
(4) +Social distance/hearer dominance condition
DCT	108/157	36/157	2/157	11/157
	68.78%	22.95%	1.27%	7%
ORP	67/146	47/146	5/146	27/146
	45.68%	31.89%	3.44%	18.96%
Chi-square	χ2 = 7.652, df = 1, p = 0.105	χ2 = 4.589, df = 1, p = 0.332	N/A	χ2 = 1.205, df = 1, p = 0.547
DCT = Discourse completion task; ORP = Oral Role Play.

).

(11)

Sir, considering my hard work and the production I provide at work, I think I deserve a salary rise. Thank you.

4.3. Internal Modification: Written vs. Oral Tasks

The analysis of the data revealed that there is no task effect, whether written or oral, in terms of lexical modification. In both tasks, the ORP and the DCT, the participants used more zero modification (see Examples 1–3) than the marker “please” (see Examples 7–8) and consultative devices (see Examples 4–5). The scores for other lexical and phrasal devices, such as downtoners, understaters, hedges, subjectivizers, cajolers, and appealers, were low, which provides further evidence supporting the findings of Economidou-Kogetsidis (2008, 2013). However, this finding contrasts with previous research by Faerch and Kasper (1989) and House (1989), which showed that their participants preferred to use clear and unambiguous means (such as the marker “please”) for the lexical modification of their requests and for explicitly indicating the illocutionary force (Economidou-Kogetsidis, 2008).

CG learners of L2 English may be influenced by their L1 and cultural values, as the use of this formulaic expression in Greek is not as frequent or conventionalized as it is in English. In Greek, this expression is often associated more with formality than politeness (Sifianou, 1992a; Economidou-Kogetsidis, 2008), which reflects the positive politeness orientation of Greek culture (Sifianou, 1992a, 2001; Pavlidou, 1994, 1997, 1998; Makri-Tsilipakou, 2001; Economidou-Kogetsidis, 2003, 2005). In contrast, British culture favors negative politeness (Scollon & Scollon, 1983; Sifianou, 1992a, 2001; Wierzbicka, 2003; Economidou-Kogetsidis, 2003, 2004, 2005). The underuse of consultative devices/openers can also be explained by L1 transfer, as they are not widely used in CG and are associated with negative politeness, expressing tentativeness and pessimism about the request outcome, especially in negative constructions and conditional clauses (Sifianou, 1992a); see

Table 6. Internal modification: written vs. oral task.

Internal Modification	ZM	MP	CD/O	D	U	H	S	C	A
DCT	346/634	203/634	59/634	12/634	5/634	0/634	5/634	2/634	2/634
	54.57%	32.01%	9.35%	1.89%	0.78%	0%	0.78%	0.31%	0.31%
ORP	332/565	188/565	29/565	3/565	7/565	4/565	1/565	1/565	0/565
	58.83%	33.18%	5.17%	0.45%	1.29%	0.64%	0.22%	0.22%	0%
Chi-square
χ2	41.741	44.169	13.391	0.195	0.631	N/A	N/A	N/A	N/A
df	1	1	1	1	1	N/A	N/A	N/A	N/A
p	0.759	0.669	0.572	0.907	0.427	N/A	N/A	N/A	N/A
ZM = zero marking; MP = marker ‘please’; CD/O = consultative devices/openers; D = downtoners; U = understaters; H = hedges; S = subjectivizers; C = cajolers; A = appealers; DCT = Discourse completion task; ORP = Oral Role Play.

.

The findings suggest that there is an effect of social distance, hearer dominance, power, familiarity, and imposition on internal modification, as there is a statistically significant difference among the conditions [±social distance, hearer dominance] regarding zero marking and the use of marker “please” and consultative devices, but not for the other lexical and phrasal means (see Appendix D for the Pearson chi-square statistics). In the DCT task, participants used zero marking more frequently in contexts with +social distance, while in the ORP, this was mainly observed in the +hearer dominance conditions. Regarding the marker “please”, students preferred to use it more in the status-equals conditions, predominantly in the medial sentence position and both in the ORP and the DCT, which contradicts the findings of Webman Shafran (2019). As for consultative devices, they were mostly used in the −social distance/hearer dominance condition in the DCT (see

Table 7. Internal modification: four conditions: written vs. oral tasks.

Internal Modification	ZM	MP	CD/O	D	U	H	S	C	A
(1) +Social distance/status equals condition
DCT	89/160	63/160	5/160	2/160	0/160	0/160	0/160	0/160	1/160
	55.62%	39.37%	3.12%	1.27%	0%	0%	0%	0%	0.62%
ORP	70/149	60/149	12/149	1/149	6/149	0/149	0/149	0/149	0/149
	46.55%	40.51%	7.75%	0.86%	4.33%	0%	0%	0%	0%
Chi-square
χ2	1.653	0.540	3.472	N/A	N/A	N/A	N/A	N/A	N/A
df	1	1	1	N/A	N/A	N/A	N/A	N/A	N/A
p	0.799	0.969	0.176	N/A	N/A	N/A	N/A	N/A	N/A
(2) −Social distance/status equals condition
DCT	58/160	57/160	13/160	1/160	2/160	0/160	0/160	1/160	1/160
	53.12%	35.62%	8.15%	0.62%	1.25%	0%	0%	0.62%	0.62%
ORP	72/135	55/135	6/135	2/135	0/135	0/135	0/135	0/135	0/135
	53.44%	40.51%	4.31%	1.72%	0%	0%	0%	0%	0%
Chi-square
χ2	1.812	1.865	0.798	N/A	N/A	N/A	N/A	N/A	N/A
df	1	1	1	N/A	N/A	N/A	N/A	N/A	N/A
p	0.770	0.760	0.940	N/A	N/A	N/A	N/A	N/A	N/A
(3) −Social distance/hearer dominance condition
DCT	74/157	35/157	38/157	3/157	2/157	0/157	4/157	1/157	0/157
	47.13%	22.29%	24.23%	1.91%	1.27%	0%	2.54%	0.63%	0%
ORP	89/134	33/134	6/134	2/134	0/134	2/134	1/134	1/134	0/134
	66.37%	24.13%	4.31%	1.72%	0%	1.72%	0.86%	0.86%	0%
Chi-square
χ2	4.212	0.866	0.530	N/A	N/A	N/A	N/A	N/A	N/A
df	1	1	1	N/A	N/A	N/A	N/A	N/A	N/A
p	0.378	0.929	0.767	N/A	N/A	N/A	N/A	N/A	N/A
(4) +Social distance/hearer dominance condition
DCT	92/157	51/157	10/157	2/157	1/157	0/157	1/157	0/157	0/157
	58.59%	32.48%	6.39%	1.28%	0.63%	0%	0.63%	0%	0%
ORP	96/146	40/146	6/146	0/146	1/146	3/146	0/146	0/146	0/146
	65.51%	27.58%	4.31%	0%	0.86%	1.72%	0%	0%	0%
Chi-square
χ2	6.024	6.616	0.590	N/A	N/A	N/A	N/A	N/A	N/A
df	1	1	1	N/A	N/A	N/A	N/A	N/A	N/A
p	0.197	0.158	0.745	N/A	N/A	N/A	N/A	N/A	N/A
ZM = zero marking; MP = marker ‘please’; CD/O = consultative devices/openers; D = downtoners; U = understaters; H = hedges; S = subjectivizers; C = cajolers; A = appealers; DCT = Discourse completion task; ORP = Oral Role Play.

).

4.4. External Modification: Oral vs. Written Tasks

In general, the analysis of the data indicates that there is a difference between the oral and written tasks, although it is not statistically significant, regarding external modification. Participants used more zero marking in the ORP than in the DCT (see Example 3), while grounders (see Examples 1, 4, 6, 8, 9, 10, 11) and apology markers (see Examples 5, 8) were more prevalent in the written task than in the oral one. The other means of external modification, such as disarmers, preparators, precommitment markers, and promise markers, were not used frequently. There is a nearly statistically significant difference between the DCT and the ORP with respect to apologies in the +social distance/hearer dominance condition (see

Table 8. External modification: written vs. oral tasks.

External Modification	ZM	G	D	P	G/P	PR/MP	I/M	A
DCT	176/634	289/634	2/634	30/634	7/634	3/634	10/634	117/634
	27.76%	45.58%	0.34%	4.73%	1.10%	0.47%	1.57%	18.45%
ORP	358/565	114/565	1/565	0/565	3/565	6/565	3/565	80/565
	63.36%	20.28%	0.21%	0%	0.43%	1.07%	0.43%	14.22%
Chi-square
χ2	41.565	45.986	0.018	N/A	0.153	0.195	0.284	27.458
df	1	1	1	N/A	1	1	1	1
p	0.925	0.556	0.894	N/A	0.695	0.658	0.867	0.493
ZM = zero marking; G = grounders; D = disarmers; P = preparators; G/P = getting a precommitment; PR/MP = promise of a reward/make a promise; I/M = imposition minimizer; A = apology; DCT = Discourse completion task; ORP = Oral Role Play.

).

There is an effect of social distance, hearer dominance, power, familiarity, and imposition on external modification, as there is a statistically significant difference among the conditions [±social distance, hearer dominance] regarding zero marking and the use of grounders and apologies (see Appendix E for the Pearson chi-square statistics). The highest score for zero marking in the written task occurred in the +social distance/status-equals condition, while in the oral task, it was observed in the −social distance/status-equals condition; conversely, the situation for the use of grounders is opposite. Apologies were used in the +social distance contexts in the DCT, while in the ORP, they were predominantly found in the +hearer dominance condition (see

Table 9. External modification: four conditions: written vs. oral tasks.

External Modification	ZM	G	D	P	G/P	PR/MP	I/M	A
(1) +Social distance/status equals condition
DCT	67/160	54/160	0/160	3/160	2/160	2/160	1/160	31/160
	41.87%	33.75%	0%	1.87%	1.26%	1.26%	0.62%	19.37%
ORP	77/149	41/149	0/149	0/149	0/149	1/149	0/149	30/149
	51.72%	27.58%	0%	0%	0%	0.86%	0%	19.82%
Chi-square
χ2	4.154	6.133	N/A	N/A	N/A	N/A	N/A	10.855
df	1	1	N/A	N/A	N/A	N/A	N/A	1
p	0.386	0.189	N/A	N/A	N/A	N/A	N/A	0.210
(2) −Social distance/status equals condition
DCT	30/160	112/160	0/160	1/160	1/160	0/160	5/160	11/160
	18.75%	70%	0%	0.63%	0.63%	0%	3.12%	6.87%
ORP	98/135	29/135	0/135	0/135	3/135	1/135	1/135	3/135
	73.27%	21.55%	0%	0%	1.72%	0.87%	0.87%	1.72%
Chi-square
χ2	4.115	7.663	N/A	N/A	N/A	N/A	N/A	0.284
df	1	1	N/A	N/A	N/A	N/A	N/A	1
p	0.391	0.105	N/A	N/A	N/A	N/A	N/A	0.867
(3) −Social distance/hearer dominance condition
DCT	35/157	75/157	3/157	17/157	3/157	1/157	2/157	21/157
	22.29%	47.77%	1.91%	10.85%	1.91%	0.63%	1.27%	13.37%
ORP	75/134	22/134	1/134	0/134	0/134	4/134	1/134	31/134
	56%	16.37%	0.86%	0%	0%	2.58%	0.86%	23.27%
Chi-square
χ2	0.922	4.933	N/A	N/A	N/A	N/A	N/A	0.815
df	1	1	N/A	N/A	N/A	N/A	N/A	1
p	0.921	0.288	N/A	N/A	N/A	N/A	N/A	0.665
(4) +Social distance/hearer dominance condition
DCT	45/157	46/157	1/157	9/157	1/157	0/157	2/157	53/157
	28.66%	29.29%	0.65%	5.73%	0.65%	0%	1.27%	33.75%
ORP	103/146	23/146	0/146	0/146	0/146	0/146	0/146	20/146
	70.68%	15.51%	0%	0%	0%	0%	0%	13.79%
Chi-square
χ2	3.199	4.680	N/A	N/A	N/A	N/A	N/A	9.181
df	1	1	N/A	N/A	N/A	N/A	N/A	1
p	0.525	0.322	N/A	N/A	N/A	N/A	N/A	0.057
ZM = zero marking; G = grounders; D = disarmers; P = preparators; G/P = getting a precommitment; PR/MP = promise of a reward/make a promise; I/M = imposition minimizer; A = apology; DCT = Discourse completion task; ORP = Oral Role Play.

).

The findings are consistent with those of Economidou-Kogetsidis (2008), as CG learners used zero marking, consultative devices/openers, and the marker “please” more frequently than other lexical and phrasal devices for internal modification. However, this study examined both symmetrical and asymmetrical power/social distance situations. Regarding external modification, CG students used zero marking and grounders, which support the previous findings of Economidou-Kogetsidis (2008). However, they also used apologies and did not employ any disarmers. Disarmers help create common ground and mutual understanding between interlocutors, thus embodying characteristics of positive politeness. Being exposed to an L2 English environment, CG students tend to approximate their behavior to that of native English speakers and refrain from using disarmers. Apologies are considered a strategy of negative politeness (P. Brown & Levinson, 1987), which is favored in British culture. Therefore, the use of apologies by CG students could be a result of their immersion in the English-speaking environment of a British university in Cyprus rather than a transfer from their first language (L1). It appears that internal modification is more influenced by L1 transfer from CG, while external modification is shaped by the L2 English learning setting and culture.

In general, all three tasks exhibit the same trend regarding the level of directness, request strategies, internal and external modification, and request perspective. This is consistent with previous research by Beebe and Cummings (1996) and Economidou-Kogetsidis (2013). These results suggest that the data obtained from the oral and written tasks are comparable.

5. Discussion

Our study investigated requests in L2 English as produced by L1 speakers of Cypriot Greek, focusing on the possible effects of task type (oral vs. written), social distance, and the hearer’s power on the modes of directness, hearer perspective, and internal and external modification. The results showed that participants tended to use conventionally indirect strategies more than other strategies across all three tasks (MCQ, DCT, ORP). The overall prevalence of the conventionally indirect request strategy in the three tasks supports the idea of the centrality of conventionalization in politeness. This is consistent with the frequency-based or habit-based accounts of politeness and conventionalization, which relate to the inherent evaluation of the expression, the context, and “meta-knowledge about not what expressions mean but how often they mean that.” (Terkourafi, 2015, p. 17). The speaker is perceived as more polite and concerned for the hearer when using indirect requests due to a decrease in illocutionary force and degree of optionality (Leech, 1983; Culpeper, 2011; Terkourafi, 2015).

(Im)politeness is related to conventionalization rather than indirectness and depends on expressions, contexts, and speakers, regardless of the degree of face-threat. The conventional indirect request strategy is more closely associated with politeness than the non-conventional indirect strategy or hints due to the extra inferential load placed on the listener, which can minimize the politeness effect. Therefore, greater indirectness does not necessarily equate to greater politeness; it is based on the concepts of coerciveness and clarity, and an imbalance between them (e.g., indirect vs. direct) can lead to impoliteness (Blum-Kulka, 1987; Terkourafi, 2015).

The participants in the study demonstrated a low rate of non-conventional indirect strategy use, as its illocutionary force is ambiguous (Holtgraves & Yang, 1990; Terkourafi, 2015). The speakers generally employed conventionally indirect approaches and used fixed expressions influenced by extralinguistic contextual factors—such as the interlocutors’ gender, age, social class, their relationship, and the setting—rather than relying solely on variables like power, social distance, and hierarchy to communicate politeness (Manes & Wolfson, 1981; Terkourafi, 2001, 2002, 2009, 2012).

Although the participants of the current study—learners of L2 English with an L1 CG background—showed an overall strong preference for the hearer perspective, likely due to a possible L1 transfer from CG in terms of pragmatic knowledge and preferred perspective, their tendency to use the speaker perspective in the hearer-dominance situations suggests that they may be influenced by their English-speaking environment (a private, English-speaking university in Cyprus). Native speakers of English typically apply the speaker perspective (Ogiermann & Bella, 2020). However, further research is needed to compare CG students at private (English-speaking) and public (Greek-speaking) universities in Cyprus.

The findings of our study provide further evidence supporting the general pattern of the underuse of lexical modifiers by L2 English learners in comparison to native English speakers (Faerch & Kasper, 1989; Trosborg, 1995; Economidou-Kogetsidis, 2008, 2009; Wigglesworth & Yates, 2011; Hassall, 2001, 2012; Göy et al., 2012) and by learners of other L2s (e.g., Félix-Brasdefer, 2012; Shively, 2011 for Spanish; Bella, 2012 for Greek; and Barron, 2003 for German).

The conventionalized politeness marker “please” is used more frequently than downtoners (Cunningham, 2017). It is interesting to note that CG students utilized the marker “please” more in status-equals conditions, predominantly in the medial position of sentences, both in oral and written tasks. This finding does not align with the study by Webman Shafran (2019).

According to Ogiermann (2009), the politeness marker “please” is a mitigating device that is used cross-culturally with request speech acts, particularly with imperatives and conventional indirect forms of legitimate requests that involve a low level of imposition, but not with non-conventionally indirect strategies or hints (House, 1989; Sifianou, 2001; Wichmann, 2004; Economidou-Kogetsidis, 2005). The strongest directive force of the marker “please” is associated with its initial position in the sentence, which is typically used in speaker-dominance conditions. When “please” appears in the final position, it reflects the social rule of communication that urges the hearer to comply with the request in order to gain societal approval; this usually occurs in formal settings. In contrast, the medial position of “please” introduces variability in function and the level of directness (Sato, 2008; Webman Shafran, 2019). Overall, requests are more direct and transparent when the marker “please” is used (Blum-Kulka, 1985, 1987).

The participants in our study implemented more external than internal modifications, which aligns with the general trend among L2 English learners to overuse external modifiers (Kasper, 1981; Blum-Kulka & Olshtain, 1986; Faerch & Kasper, 1989; Hassall, 2001, 2012). They may opt for the explicit realization of politeness (Faerch & Kasper, 1989) or find this type of modification easier than lexical alternatives (Hassall, 2001, 2012). Grounders were the most frequent type of external modification, consistent with previous research by Economidou-Kogetsidis (2008, 2009). Additionally, the use of apologies was notably high, which could be attributed to immersion in an English-speaking environment (Cunningham, 2017).

The differences between internal and external modifications in the current study can be attributed to L1 and L2 influence, respectively, as well as to cross-linguistic and cross-cultural interference. These differences reflect the polarity in the pragmatic behavior of vertical (Greek: solidarity is affected by power asymmetries, with an emphasis on group belonging) and horizontal (British: solidarity is not affected by power asymmetries) societies (R. Brown, 1965; Vassiliou et al., 1972; Hofstede, 1980, 1983; O’Driscoll, 1996; Koutsantoni, 2004; Economidou-Kogetsidis, 2008).

Power Distance is one of the dimensions of Hofstede’s model, a framework developed by Geert Hofstede to compare cultural values (de Mooij & Hofstede, 2010; Epaminonda, 2022). It reflects how much less powerful members of institutions or organizations accept and expect unequal distributions of power within a society. In high Power Distance cultures, hierarchical structures are accepted as natural, with individuals respecting their designated roles without questioning inequalities. Relations tend to be paternalistic and autocratic, with authority centralized. Conversely, in low Power Distance societies, there is a preference for more equal and democratic relationships, with people seeking justification for power differences and favoring consultative approaches. Overall, Power Distance indicates how societies manage and perceive social inequalities. According to Hofstede’s dimensions, Cyprus has a medium to high score in power distance (Epaminonda, 2022).

Methodologically, this study is innovative in that it compares three different tasks: a written multiple-choice questionnaire (MCQ) and a discourse completion task (DCT), both of which are controlled elicitation tasks focusing on speech acts and metapragmatic awareness in isolation from surrounding discourse, thereby lacking authenticity and social contextualization (Blum-Kulka & Olshtain, 1984, 1986; Faerch & Kasper, 1989; Golato, 2003; Economidou-Kogetsidis, 2008, 2009, 2010; Woodfield & Economidou-Kogetsidis, 2010; Cunningham, 2017); and an oral role-play task (ORP), which, while still an elicitation task, operates under more authentic spoken discourse conditions (Sasaki, 1998; Kasper, 2000; Yuan, 2001; Otcu & Zeyrek, 2008; Economidou-Kogetsidis, 2012; Göy et al., 2012; Hassall, 2001, 2012; Woodfield, 2012; Al-Gahtani & Roever, 2014).

It is also important to note that we investigated the L2 English immersion setting and its impact on the use of request strategies and mitigating devices under ±social distance and ±hearer dominance conditions, as data collection took place at an English-speaking private university in Cyprus. The use of different tasks that measure online and offline pragmatic knowledge, as well as socio-pragmatic competence and performance, in (non)-interactive formats (Felix-Brasdefer, 2010) with the same participants in identical social situations increases the validity and reliability of the research and the generalizability of the results. We did not find a significant task effect, as both oral and written tasks exhibited similar patterns regarding request strategies, request perspective, and mitigating devices. This finding may indicate the comparability of the tasks and the efficiency of the mixed-methods approach in providing insight into the issue of request speech acts in L2 English.

This study offers important pedagogical implications, as its findings can inform the instruction of pragmatics in ESL/EFL settings. Developing pragmatic competence is crucial for effective communication and relies on sociolinguistic rules that help prevent misunderstandings and inappropriate behavior in specific contexts. Various teaching approaches and strategies can support L2 English learners in enhancing their pragmatic awareness. It is essential that the teaching of L2 English pragmatics is adequately represented and integrated into instructional materials. When teaching and learning pragmatics in L2 and/or EFL contexts, educators should consider students’ proficiency levels, abilities, areas of difficulty, and challenges with applying pragmatic knowledge in both production and comprehension—challenges often stemming from limited competence or L1 transfer.

6. Conclusions

This study investigated the pragmatic competence and awareness of L2 learners of English with an L1 CG background. The focus of the research was on their request strategies in L2 English, including levels of (in)directness, request perspective, internal and external modification, and the task effect. Additionally, the study considered taking cross-linguistic and cross-cultural comparisons, social constraints, and situational and individual variation.

The significance of this study lies in its attempt to investigate request strategies and (in)directness, as well as request perspective, internal modification, and lexical modification across all possible situations/contexts: [±social distance, hearer dominance]. It also examined the task effect, comparing the written and oral modes, based on methodological triangulation. The major finding of this investigation is that the three tasks—a multiple-choice questionnaire, a discourse completion task, and an oral role-play task—are comparable and can elicit valid and reliable results. Statistically, it was confirmed that the use of request strategies, the level of (in)directness, and the request perspective, as well as internal and external modification, depend on social distance and hierarchy, power, familiarity, and imposition, as well as on the learning environment, specifically at an English-speaking private university in Cyprus.

The mixed-methods approach of this study provided a more comprehensive understanding of pragmatic phenomena and their variability. Multiple tools can be employed to assess L2 learners’ pragmatic competence and to increase awareness of the significance of pragmatics and politeness. It is crucial to select an appropriate task aligned with the research design and research questions to ensure the validity and reliability of the findings. When doing so, factors such as practicality, authenticity, and the comparability of tasks to natural speech should be considered, along with socio-pragmatic knowledge, as well as the structure and content of the task.

Overall, the findings of this study indicate that additional efforts are necessary to enhance pragmatic awareness and to integrate pragmatics and politeness more prominently into the EFL curriculum and teaching resources. Employing alternative, contextualized, and communication-focused teaching approaches can encourage students to become more engaged and motivated to develop pragmatic skills by considering language and cultural differences alongside linguistic and extralinguistic factors. These insights can inform the explicit instruction of L2 English pragmatics and intercultural communication.

Further research is required with a focus on both private and public university settings in Cyprus, as well as secondary and tertiary levels, involving CG and other L1 students’ backgrounds. This will help investigate the effects of context and L1 transfer on the development of pragmatic competence and performance among L2 English learners. It is important to collect and analyze naturally occurring oral and written data, such as conversations (Beebe & Cummings, 1996; Yuan, 2001; Golato, 2003; Félix-Brasdefer, 2007), emails (Hartford & Bardovi-Harlig, 1996; Reinhardt, 2010; Economidou-Kogetsidis, 2011), (a)synchronous computer-mediated communication, and oral and written corpora (Cunningham, 2017). This research should combine both cross-sectional and longitudinal methods, adding conditions such as ±speaker dominance and enhanced situational prompts. Additionally, attention should be paid to both lexical and syntactic modification, including downgraders and intensifiers, as well as phonetic and prosodic features. The phonology–pragmatics interface should be explored based on elicited versus natural comprehension and production, utilizing both online and offline data.