Mitigation, Rapport, and Identity Construction in Workplace Requests

Abstract

This study investigates how Greek professionals formulate upward requests and simultaneously manage rapport and workplace identity within hierarchical exchanges. The data comprise 400 written requests elicited through a discourse–completion task from 100 participants, supplemented by follow-up interviews. Integrating pragmatic perspectives on request mitigation with Spencer-Oatey’s Rapport-Management model and a social constructionist perspective on identity, the analysis reveals a distinctive “direct-yet-mitigated” style: syntactically direct head acts (typically want- or need-statements) various mitigating devices. This mitigation enables speakers to preserve superiors’ face, assert entitlement, and invoke shared corporate goals in a single move. Crucially, rapport work is intertwined with identity construction. Strategic oscillation between deference and entitlement projects four recurrent professional personae: the deferential subordinate, the competent and deserving employee, the cooperative team-player, and the rights-aware negotiator. Speakers shift among these personae to calibrate relational distance, demonstrating that rapport management functions not merely as a politeness calculus but as a resource for dynamic identity performance. This study thus bridges micro-pragmatic choices and macro social meanings, showing how linguistic mitigation safeguards interpersonal harmony while scripting desirable workplace selves.

1. Introduction

In today’s increasingly globalised and multilingual workplaces, the ability to manage interactional delicacy through language is critical to professional success. Among the many speech activities that constitute workplace discourse, requests are particularly revealing: they simultaneously pursue transactional goals, negotiate social relations, and project professional identities. Building on sociopragmatic scholarship that has mapped the intricate interplay of power, politeness, and identity at work (e.g., Brown & Levinson, 1987; Vine, 2004; Holmes, 2005; Spencer-Oatey, 2008), the present study focuses on a setting that remains under-represented in this literature: Greek professional communication. By examining how native Greek speakers formulate requests to hierarchical superiors, we seek to broaden the empirical base of request studies and to show how language-specific resources are mobilised to perform rapport management work.

Although a ‘speech-act’ lens is adopted to identify request events, the analysis is resolutely discourse-oriented. That is, requests are treated not as de-contextualised head acts but as sequential bundles of moves whose pragmatic weight is distributed across internal and external modifiers, stance markers, and justificatory material. Blum-Kulka et al.’s (1989) taxonomy provides the descriptive scaffolding for classifying directness, mitigation, and perspective, while Spencer-Oatey’s (2000, 2008) Rapport-Management framework licences the interpretation of those linguistic choices as strategic calibrations of face, rights, and goals. In doing so, we answer recent calls (Marra & Dawson, 2021) to integrate identity construction more explicitly into workplace-pragmatics research by showing how Greek professionals craft particular personae through their requestive behaviour.

This study draws on written responses elicited with a Discourse Completion Test (DCT) administered to 100 employees across diverse Greek businesses. The written responses were supplemented with retrospective interviews that probed participants’ intentions and contextual assumptions, thereby gaining insight into the metapragmatic reasoning that underlies their linguistic choices. This mixed-method design enabled us to link quantitative patterns in the DCT corpus to the speaker-internal considerations that motivated them.

The analysis revealed a notable tension: although participants overwhelmingly relied on direct head acts, they embedded them within rather dense layers of modification, formal address forms, and self- or business-oriented justifications. It is argued that this configuration possibly reflects a local norm that balances institutional hierarchy with relational solidarity: directness indexes entitlement derived from contractual rights, whereas extensive mitigation preserves the superior’s face and affirms shared organisational goals. By mapping these patterns onto Spencer-Oatey’s (2000, 2008) interactional bases of face, sociality rights, and goals, this study demonstrates how rapport management and identity work unfold in tandem.

This paper proceeds as follows: After situating this study within the broader literature on workplace requests and outlining key concepts of Rapport Management and identity (Section 2), data collection and coding procedures are described in detail (Section 3). Quantitative findings on strategy distribution and qualitative insights from the verbal reports are presented (Section 4), before discussing how these choices enact rapport and professional identities (Section 5). The conclusion reflects on the implications of our findings and points to avenues for future cross-linguistic research on request practices in professional settings.

2. Theoretical Background

2.1. Workplace Requests

Research on requests in professional settings has moved steadily from a narrow focus on linguistic form and politeness to a multidimensional view that embeds request-making in workplace practice, organisational power, and identity work. Early applications of Brown and Levinson’s (1987) politeness theory cast the request as a paradigmatic face-threatening act: because it limits the addressee’s freedom of action and risks negative evaluation, speakers routinely deploy mitigation to soften the imposition. Subsequent studies have shown just how sensitive that mitigation is to institutional power. Vine’s (2004) conversation-analytic study of one-to-one policy-unit meetings found that subordinates rarely use bare imperatives; instead, they frame almost all their request tokens as epistemic-modal interrogatives (can/could/would you…), signalling the addressee’s right to refuse and marking the requester’s lower institutional footing. Likewise, Koester’s (2010) ABOT corpus demonstrates that requests cluster in procedural or favour-seeking genres and appear chiefly in mitigated formats (can prefaced questions or imperatives softened by just), whereas stronger deontic modals (have to, need to) are reserved for unequivocal directives.

Quantitative corpus work confirms the power-mitigation link at scale. Drawing on the 2-million-word Wellington Language in the Workplace (WLWP) corpus, Holmes and Stubbe (2015) report that managers favour bald imperatives in routine downward directives, while upward or extra-duty moves are cast as modal interrogatives that explicitly acknowledge superiors’ decision rights. Yet, as they note, speakers continually recalibrate directness to perform competence, collegiality, or expertise, hinting that relational concerns extend beyond politeness per se.

A growing body of work therefore shifts the analytical lens from mitigation alone to the identities that requests help construct. In WLWP data, Holmes (2006) shows managers alternating between blunt orders that index an authoritative, locally “masculine” stance and hedged, inclusive requests that evoke a collaborative, “feminine” persona, turning each request into a site of gender negotiation. Schnurr (2025) extends this insight across modalities: terse email imperatives to subordinates, rapport-building requests to clients, and the micro-choices of “can you…?” versus a blunt imperative in instant messaging all position leaders along a collaborative–hierarchical continuum and invite co-workers to ratify or resist those identities.

Cross-cultural and interlanguage research adds a further layer, showing that mitigation and identity cues do not travel smoothly across linguistic and cultural boundaries. Studies of Spanish, Korean, Greek, and BELF workplaces (e.g., Félix-Brasdefer, 2005; Economidou-Kogetsidis, 2011; Kankaanranta & Louhiala-Salminen, 2013) document pragmatic “misfires” when users transfer L1 request norms or negotiate new ones in lingua-franca teams. Here, metapragmatic awareness becomes pivotal: speakers overtly comment on style (“I didn’t mean to sound pushy…”) or attach softening cues (“just to clarify”) to align expectations (Holmes & Marra, 2007; Kádár & Haugh, 2013).

Taken together, these strands converge on two claims. First, request design is acutely sensitive to institutional power, cultural convention, and the speaker’s metapragmatic stance; mitigation is therefore best viewed as a dynamic resource rather than a fixed formula. Second, because requests publicly negotiate rights and obligations, they have exceptional capacity to project professional, gendered, and cultural identities. Recent discourse-oriented studies accordingly treat the request not merely as a politeness problem but as a primary site where workplace relationships and selves are made visible.

The present study deploys well-established instruments in the cross-cultural and interlanguage pragmatics literature, most notably Blum-Kulka et al.’s (1989) taxonomy of request strategies, while being situated within a discourse-oriented paradigm that privileges participants’ own construal of the interaction and the broader social meanings negotiated in talk. To that end, this study draws on Spencer-Oatey’s (2000) Rapport Management (RM) model, which was developed to remedy the face-based limitations of Brown and Levinson. In addition, following Marra and Dawson’s (2021, p. 482) exhortation that workplace-discourse research should be strongly concerned with identity construction, the present investigation examines how identities are enacted through requestive behaviour. The next section therefore (i) outlines the principal tenets of RM, (ii) clarifies this study’s conceptualisation of identity, and (iii) highlights the relation of RM with identity construction and workplace requests.

2.2. Rapport, Identity, and Requests

Spencer-Oatey’s (2000, 2005, 2008, 2015) rapport-management model conceives interaction as the ongoing construction, maintenance, enhancement, or challenge of social relationships. It is organised around three analytically distinct but interdependent rapport bases:

• Face sensitivities. Drawing on classic face theory, Spencer-Oatey distinguishes between quality face, i.e., the value speakers claim for their personal competence, abilities and character, and social-identity face, i.e., the value attached to their group memberships and institutional roles. A threat to either sub-component jeopardises the speaker’s public self-image and therefore the harmony of the encounter.
• Sociality rights and obligations. These are expectations about how people should be treated. They comprise two dimensions:

  (a)

  equity rights, which demand a fair distribution of costs and benefits, and

  (b)

  association rights, which regulate the appropriate degree of involvement or distance.

• Interactional goals. Participants enter an encounter with pragmatic ends they wish to accomplish. Goals may be complementary, neutral, or incompatible; misalignment is a frequent but often unnoticed source of relational friction.

These bases are negotiated within five interactional domains (illocutionary, discourse, participation, stylistic, and paralinguistic) and are oriented through four rapport stances: enhancement, maintenance, neglect, and challenge.

Adopting a social-constructionist view of identity (Benwell & Stokoe, 2006), the present study treats these rapport processes as the very mechanisms through which speakers’ multiple, mutable identities are assembled.

Identities are “social positioning of the self” (Bucholtz & Hall, 2005, pp. 585–586) and emerge through situated linguistic choices shaped by context (De Fina et al., 2006). Because discourse not only reflects but actively produces the social world (Hall & Bucholtz, 2013), every attempt to manage face, rights, or goals simultaneously projects a version of the professional self. Empirical research (see, e.g., Ho, 2010; Castro, 2018; Bella & Ogiermann, 2022) has shown that requesters craft their messages to highlight “desirable self-aspects” (Simon, 2004) or “desirable identities” (Schlenker, 1986). Schlenker (1986, p. 25) defines ‘desirable identities’ as those self-images and self-identifications that people endorse to serve their goals or purposes. He explains that “desirable identity images represent what people believe that they can be and should be in particular contexts” (emphasis in the original) that foster cooperation. In Spencer-Oatey’s (2000, 2008) model, such positive self-presentation aims at rapport enhancement (see also Spencer-Oatey, 2007). The model therefore offers a precise, layered matrix for tracing how professionals enhance rapport, engineer compliance, and simultaneously negotiate who they believe they can or should be in workplace contexts.

3. Method

3.1. Participants and Data Collection Procedures

The participants of this study were one hundred (100) Greek native speakers (34 male, 66 female) and their age ranged between 28 and 50 years. At the time of this study, they all worked in various jobs, but mostly in education, tourism, sales, advertising, publishing, banks, and various types of office work. The gender distribution reflects voluntary self-selection during recruitment; no exclusionary criteria were imposed.

The data was collected using a discourse completion test (DCT) designed to elicit speech acts in a workplace setting across sixteen situations. Four of the situations were designed to elicit requests, while eight (four apologies and four refusals) served as distractors. This analysis examines subjects’ responses in the four request scenarios: (1) requesting a leave of absence during a busy period, (2) asking for a new computer to replace an inefficient one, (3) requesting a salary increase, and (4) seeking an extension for a team project (see Appendix A for full scenarios). A version of the first situation was used in Wigglesworth and Yates (2007), while the rest were designed specifically for the purposes of the present study. Although the status of the participants as employees requesting something from an employer established hierarchy, the original DCT scenarios did not specify the social distance between the participants. This was a deliberate choice, since, among other things, this study aimed to investigate how the participants’ responses reflected their perceptions of this type of relationship in general.

Although natural data are ideal for interaction analysis (Kasper, 2000, p. 318), ethnographic studies of speech acts face two problems: contextual variables remain uncontrolled and target acts are unpredictable (Gass & Houck, 1999, p. 25; Ogiermann, 2009, pp. 67–70). As Kasper (2000, p. 320) notes, collecting enough authentic tokens of a specific pragmatic feature can be impractical. For precisely this reason, the present study adopted a discourse completion task: it offers a controlled yet flexible elicitation method that guarantees the target speech act will occur in identical contextual conditions for all participants, thereby making cross-speaker comparisons both feasible and reliable.

While DCTs “cannot capture the dynamics of social face-to-face interaction” (Félix-Brasdefer, 2010, p. 47), research shows they successfully elicit linguistic phenomena otherwise hard to observe (Billmyer & Varghese, 2000, p. 518; Economidou-Kogetsidis, 2008, p. 117), and their patterns generally align with natural data (Billmyer & Varghese, 2000, p. 518). Nevertheless, DCT output “can never be the same as authentic conversation” (Kasper, 2000, p. 318); hence, the findings approximate, but do not fully mirror, natural speech (Ogiermann, 2009, p. 68).

To address some limitations of the DCT and gain deeper insight into the rationale behind sociopragmatic choices (Cohen, 1996, p. 256), retrospective verbal reports were collected alongside the DCT data. Immediately after completing the DCT, participants took part in interviews. Due to practical constraints, interviews were conducted with 25 of the participants. The researcher focused on the target situations and primarily asked structured questions regarding the participants’ cognitions and intentions (see Ericsson & Simon, 1993). The key questions were

• What were you paying attention to when responding?
• What were you trying to achieve by saying what you said?

3.2. Data Analysis

Requests were classified following Blum-Kulka et al.’s (1989) original framework, and later expansions and additions by Trosborg (1995) and Yates (2010) adapted to fit the present study’s data. The classification accounted for the degree of directness of the main request (head act), as well as internal and external modification and request perspective.

Head acts were coded into three levels of directness as identified by Blum-Kulka et al. (1989): (1) direct, (2) conventionally indirect, and (3) non-conventionally direct. The non-conventionally direct category was represented solely by hints (e.g., αν αντικαταστήσουμε αυτόν τον παλιό υπολογιστή θα έχει όφελος η εργασία μας, (‘replacing this old computer will improve our work’). The conventionally indirect category was realised exclusively through query preparatories, such as θα μπορούσα μήπως να πάρω την άδεια που μου υπολείπεται; (‘could I maybe take the rest of my leave of absence?’).

In contrast, direct requests were expressed through three main strategies: (a) want statements (e.g., θέλω να πάρω μια βδομάδα άδεια ‘I want one week of leave’), (b) need statements (e.g., χρειάζεται να ζητήσουμε παράταση ‘we need to ask for an extension’), and (c) hedged performatives (e.g., θα ήθελα να ζητήσω μια παράταση ‘I would like to ask for an extension’).

Internal modification was examined through both syntactic and lexical means. The primary syntactic modifiers identified in the data included (a) past marking (e.g., ήθελα να ζητήσω μια παράταση ‘I wanted to ask for an extension’), (b) conditional forms (e.g., θα ήθελα να πάρω την άδεια που δικαιούμαι ‘I would like to get the leave I am entitled to’), and (c) conditional clauses (e.g., αν έχετε χρόνο, θα ήθελα να συζητήσουμε για τον μισθό μου ‘If you have the time, I would like us to discuss my salary’).

Lexical modification, on the other hand, was evident in the use of downtoners (ίσως ‘perhaps’, μήπως ‘maybe’), the politeness marker παρακαλώ ‘please’, empathetic markers (καταλαβαίνω ‘I understand’, ξέρω/γνωρίζω ‘I know’), interpersonal markers (ξέρετε ‘you know’, καταλαβαίνετε ‘you realize/understand’), and subjectivisers (νομίζω ‘I think’, πιστεύω ‘I believe’).

External modification in the data was achieved through (a) preparators (e.g., μπορώ να σας μιλήσω για λίγο; ‘can I talk to you for a bit?’), (b) reasons/explanations (e.g., θα χρειαστούμε περισσότερο χρόνο για να γίνει σωστή δουλειά ‘we will need more time to do a good job’), (c) disarmers (e.g., καταλαβαίνω ότι έχουμε πολλή δουλειά αυτή την περίοδο ‘I do understand how busy we are’), and (d) promises of reward (e.g., μπορώ να βοηθήσω να βρούμε έναν σχετικά φτηνό υπολογιστή ‘I can help find a relatively cheap computer’).

Request perspective was investigated based on Blum-Kulka et al.’s (1989) classification, which includes hearer, speaker, inclusive (hearer and speaker), and impersonal perspectives.

Finally, the analysis took into account the employment of the plural of formality (polite plural) as well as the use of formal address forms (e.g., κύριε/κυρία (Mr/Mrs) combined with a surname) and professional titles (e.g., κύριε Προϊστάμενε (Mr Manager) and κύριε Διευθυντά (Mr Director)).

For the statistical analyses, a strictly non-parametric approach was adopted because each speaker contributed small, paired frequency counts that violated normality assumptions. An omnibus Friedman test was first applied to detect within-participant differences across the four experimental situations; this test is the rank-based analogue of a repeated-measures ANOVA and is appropriate for more than two related samples. Whenever the Friedman test reached significance, pair-wise Wilcoxon signed-rank tests were carried out to pinpoint which situations differed, with Holm’s sequential Bonferroni procedure used to control the family-wise error rate. Effect sizes were reported with Kendall’s W for the omnibus test (indicating the strength of concordance among conditions).

The participants’ linguistic choices, together with insights gleaned from their verbal reports, were analysed to examine how they managed rapport and negotiated desirable identities through their contributions.

4. Results

A common feature observed across all situations under examination was the consistent use of the plural of formality. Not a single instance was recorded in which an employee addressed their employer using a singular form. Moreover, professional titles and formal forms of address appeared in 33% and 47% of the requests, respectively.

As shown in

Table 1. Distribution of head acts across the four situations.

	S1		S2		S3		S4
	n	%	n	%	n	%	n	%
Direct
Hedged performative	9	8.3	3	2.6	21	18.8	17	14.3
Need statement	19	17.6	80	73.4	6	5.4	68	57.1
Want statement	71	65.7	5	4.6	66	58.9	9	7.6
Total direct	99	91.6	88	80.6	93	83.1	94	79
Conventionally indirect
Query preparatory	6	5.6	7	6.5	2	1.8	9	7.6
Total CI	6	5.6	7	6.5	2	1.8	9	7.6
Non conventionally indirect
Hint	3	2.8	14	12.9	17	15.1	16	13.4
Total NCI	3	2.8	14	12.9	17	15.1	16	13.4
Total HAs	108	100	109	100	112	100	119	100

, another common feature was a marked preference for direct request strategies, with an almost complete absence of interrogative constructions and a predominant use of declaratives. Therefore, in their majority, the requests emerging in all situations are what Yates (2010) calls “apparently assertive”. However, an interesting difference emerged in the preference for want and need statements across the four situations. In Situations 1 and 3, there was a clear tendency toward want statements, which accounted for 65.7% and 58.9% of requests, respectively. In contrast, in Situations 2 and 4, need statements were the preferred format, occurring at rates of 73.4% and 57.1%, respectively. These differences were found to be statistically significant. The Friedman test showed a clear situational effect on the distribution of want statements, χ²(3) = 178.95, p < 0.0001, Kendall’s W = 0.60. Holm-corrected Wilcoxon signed-rank tests confirmed that Situations 1 and 3 contained significantly more want statements than Situations 2 and 4 (all p < 0.001). The complementary pattern held for need statements: The Friedman test yielded χ²(3) = 126.91, p < 0.0001, W = 0.42, and pair-wise Wilcoxon tests showed that Situations 2 and 4 featured significantly more need statements than Situations 1 and 3 (all p < 0.001).

The assertiveness of the head acts was softened through the use of syntactic (

Table 2. Distribution of syntactic modifiers across the four situations.

	S1		S2		S3		S4
	n	%	n	%	n	%	n	%
Past	12	10.9	11	20.8	15	13.6	21	33.9
Conditional form	84	75.7	23	43.4	77	70	32	51.6
Conditional clause	15	13.5	19	35.8	18	16.4	9	14.5
Total	111	100	53	100	110	100	62	100

) and lexical modifiers (

Table 3. Distribution of lexical modifiers across the four situations.

	S1		S2		S3		S4
	n	%	n	%	n	%	n	%
Subjectiviser	13	9.8	5	8.2	42	35.3	25	33.4
Downtoner	20	15	9	14.8	15	12.6	10	13.3
Empathetic marker	59	44.4	27	44.2	29	24.4	16	21.3
Interpersonal marker	37	27.8	20	32.8	26	21.8	24	32
Politeness marker	4	3	0	0	7	5.9	0	0
Total	133	100	61	100	119	100	75	100

), and most notably, the high frequency of external modifiers (

Table 4. Distribution of external modifiers across the four situations.

	S1		S2		S3		S4
	n	%	n	%	n	%	n	%
Preparator	28	8.6	31	10.8	36	13.7	22	8.7
Reason	182	56.2	160	55.6	194	73.8	171	67.9
Disarmer	99	30.6	78	27	30	11.4	59	23.4
Promise of reward	15	4.6	19	6.6	3	1.1	0	0
Total	324	100	288	100	263	100	252	100

).

As shown in Table 2, the use of syntactic modification was considerably more frequent in Situations 1 and 3 than in Situations 2 and 4. The difference was found to be statistically significant with the Friedman test indicating a clear situational effect, χ²(3) = 115.48, p = 7.3 × 10⁻²⁵, Kendall’s W = 0.38. Holm-corrected Wilcoxon signed-rank tests confirmed that Situations 1 and 3 contained significantly more syntactic modifiers than Situations 2 and 4 (all p < 0.001), while Situations 1 and 3 did not differ (p = 0.83) and the S2–S4 contrast reached only marginal significance (p = 0.008).

The conditional form was the most frequently used syntactic modifier across all situations, although its frequencies were considerably higher in Situations 1 and 3 (75.7% and 70%, respectively) than in 2 and 4 (43.4% and 51.6%, respectively). In Greek, the conditional is formed by combining the modal particle θα with the past form of a verb. For example, θα ήθελα (=θα + θέλω ‘want’ _{- PAST}) corresponds to the conventionalized I would like in English (Sifianou, 1992, p. 150). The mitigating function of the conditional stems from its ability to be replaced by an indicative form (e.g., θέλω ‘want’/ήθελα ‘wanted’). However, the conditional form is perceived as more formal (Sifianou, 1992, p. 150), creating greater social distance and thereby reducing the potential threat more effectively than its indicative counterparts.

As with syntactic modification, lexical modification in Situations 1 and 3 displays higher frequencies than in Situations 2 and 4 (see Table 3) with the Friedman test confirming a strong situational effect (χ²(3) = 120.22, p = 6.9 × 10⁻²⁶, Kendall’s W = 0.40). Holm-corrected Wilcoxon signed-rank tests showed that Situations 1 and 3 contained significantly more lexical modifiers than Situations 2 and 4 (all p < 0.001), while Situations 1 and 3 did not differ (p = 0.054). These findings parallel the syntactic-modification pattern, underscoring a consistent contextual boost for lexical elaboration in Situations 1 and 3.

Table 3 indicates that empathetic and interpersonal markers are the most frequently employed lexical modifiers in all situations.

External modification emerged as the most commonly used type of modification by participants across all situations. According to Table 4, the most frequently used external modification device in all situations was reasons/explanations for the requests (S1 = 56.2, S2 = 55.6, S3 = 78.5, S4 = 67.9). It is important to note that, in terms of content, two types of reasons seemed to dominate in the participants’ responses: (a) reasons emphasising the requester’s value and efficiency as an employee who merits having their request fulfilled, and (b) reasons stressing the potential benefit to the business if the requestee were to comply with the request. The first type was notably common in Situations 1 and 3, less frequent in Situation 4, and quite rare in Situation 2. The second type was characteristic of Situations 2 and 4. Examples 1 and 2 from Situations 3 and 4, respectively, serve as examples of this.

(1)

Ξέρω ότι είμαι σχετικά καινούργια στην εταιρεία αλλά πιστεύω έχω αποδείξει την αφοσίωση μου σε αυτήν αλλά και τις ικανότητες μου στην διεκπεραίωση των εργασιών που μου αναθέτετε, ως εκ τούτου θεωρώ ότι στο σημείο αυτό θα πρέπει να επανεξετάσουμε από κοινού το ύψος του μηνιαίου μου μισθού.

‘I know that I am relatively new to the company, but I believe I have proven my dedication to it as well as my abilities in carrying out the tasks you assign to me. Therefore, I think it is time for us to inclusively reconsider the amount of my monthly salary.’

(2)

Κύριε Διευθυντά, πιστεύω ότι πρέπει να ζητήσουμε παράταση στο deadline του project. Θεωρω ότι θα είναι καλύτερο για την εικόνα της εταιρείας να ζητήσουμε επιπλέον χρόνο και να είμαστε συνεπείς, παραδίδοντας μια ποιοτική δουλειά, παρά να έρθουμε αντιμέτωποι με καθυστερήσεις και ενδεχομένως μια όχι τόσο ποιοτική δουλειά.

‘Sir [lit. Mr. Manager], I believe we should request an extension to the project deadline. I think it would be better for the company’s image to ask for more time and ensure we deliver quality work on time, rather than facing delays and potentially delivering work of lower quality.’

In example 1, the explanation (‘I believe I have demonstrated my commitment to the company, as well as my ability to effectively complete the tasks assigned to me’) leading up to the main request serves to justify the request by highlighting the employee’s character and work performance as valid grounds for a salary increase. In contrast, in example 2, the explanation after the request (‘I believe it would be in the company’s best interest to request more time, ensuring we deliver high-quality work on schedule, rather than risk delays and potentially lower-quality output’) focuses on highlighting the advantages the company would gain if the request were granted.

Disarmers worked proactively to alleviate or minimise potential conflict in a conversation. Their primary goal was to address any concerns, doubts, or objections the interlocutor might have, making it less likely that they will oppose or challenge the speaker. They were used frequently across all situations, with higher frequencies in Situations 1 (30.6%) and 2 (27%) compared to Situations 3 (11.4%) and 4 (23.4%). In all cases, the main disarming strategy employed by participants was acknowledging the potential concerns of the requestee. In Situations 1 and 2, the content of the disarmers was largely shaped by the information provided in the respective DCT scenarios (e.g., ‘I know we are very busy at the moment’, or ‘I realize it’s not the best time for extra expenses’). This suggests that the higher frequency of disarmers in these situations may be influenced by the data collection instrument. However, despite the scenario in Situation 3 also providing information that could have supported the use of disarmers (the employee being relatively new to the job), their occurrence was relatively rare. Interestingly, even though the scenario in Situation 4 did not offer explicit cues to prompt the use of disarmers, this strategy was still employed quite frequently (e.g., ‘I understand that deadlines are binding’, or ‘I know we are in a hurry’).

One rather unexpected finding was the presence of preparators in the data. Given the written format of the data collection instrument and the inability of the interlocutor to respond to questions such as μπορώ να σας απασχολήσω λίγο; (‘Can I occupy you for a moment?’), one might have anticipated that preparators would be rare or entirely absent from the data. However, their appearance, though at relatively low frequencies (see Table 4), highlights the importance participants attach to preparing the ground for the upcoming request.

Finally, promises of reward were rather infrequent and highly situation-specific, occurring mainly in Situations 1 and 2 (see Table 4). They primarily involved offers from the employee to minimise any potential disruption to the business caused by fulfilling their request, such as assisting with shifts upon their return (S1) or helping to find a cost-effective replacement for the computer (S2).

Notably, the sequencing of head acts and external modifiers formed a consistent pattern in the data. In most cases, requests began with a disarmer, followed by reasons for the request, leading to the head act, which was often accompanied by additional reasons and explanations and less frequently by promises of reward. That is, the head act was typically surrounded by different types of external mitigation. Example 3 is indicative of this rather standard request structure.

(3)

Γνωρίζω ότι είναι δύσκολη εποχή για άδειες, αλλά έχω υποσχεθεί στα παιδιά μου να πάμε κάπου φέτος. Είμαστε όλοι πολύ κουρασμένοι και χρειαζόμαστε λίγο χρόνο μαζί. Θα ήθελα λοιπόν να λείψω για μια εβδομάδα τον επόμενο μήνα. Έχω ακόμα υπόλοιπο δύο εβδομάδες, οπότε καταλαβαίνετε ότι δε ζητάω κάτι πέρα από το κανονικό. Δεν έχω πρόβλημα να καλύψω κάποιες βάρδιες συναδέλφων όταν επιστρέψω.

‘I know it is a difficult time for a leave of absence, but I have promised my children to go somewhere this year. We are all very tired and we need some time together. So, I would like to be away for a week next month. I still have two weeks of leave left, so you realise I am not asking for something out of the ordinary. It wouldn’t be a problem for me to cover some colleagues’ shifts when I get back.’

Regarding request perspective, one of the most significant findings was the remarkably low frequency of the hearer perspective (see

Table 5. Distribution of request perspective across the four situations.

	S1		S2		S3		S4
	n	%	n	%	n	%	n	%
Hearer	2	1.9	7	7.4	10	10.5	9	8.7
Speaker	80	76.2	11	11.6	59	62.1	17	16.5
Inclusive	7	6.7	54	56.8	4	4.2	46	44.7
Impersonal	16	15.2	23	24.2	22	23.2	31	30.1
Total	105	100	95	100	95	100	103	100

), suggesting that participants consistently avoided framing the request as something the hearer was supposed to do for them. The speaker perspective appeared most frequently, primarily due to its systematic use in S1 and S3 (76.2% and 62.1%, respectively), whereas the inclusive perspective dominated in S2 and S4 (56.8 and 44.7, respectively). Additionally, as shown in Table 5, nearly one-third of the participants in each situation opted for an impersonal perspective, avoiding explicit mention of the agent performing the action.

The following section discusses the participants’ linguistic choices in light of their verbal reports (Section 5.1) and assesses the implications of these choices for rapport management and identity construction (Section 5.2).

5. Discussion

5.1. Participants’ Linguistic Choices

As stated in Section 4, participants consistently employed the plural of formality in all requests across all situations, signalling deference toward hierarchically superior interlocutors. Furthermore, the data revealed high frequencies of formal address forms. These findings suggest that the informants felt compelled to address their superiors with a level of deference and formality, thereby framing the relationship not only in terms of hierarchy but also in terms of social distance. This should not, however, be interpreted as evidence that employees in Greece invariably adopt such formal address. Rather, it indicates that a certain degree of formality is considered appropriate when making requests to superiors within professional contexts.

The participants’ use of request perspective was also crucial in this respect. As indicated by the results of this study, participants in all situations systematically avoided the hearer perspective. According to Blum-Kulka et al. (1989, p. 59), this is always a strategic choice, since “avoiding to name the hearer as actor can reduce the form’s level of coerciveness” (see also Bella, 2005; Ogiermann & Bella, 2020). Moreover, the high frequencies of joint perspective in all situations confirm Vine’s (2004, p. 199) contention that even where there is a power/status difference, indices of joint effort and cooperation may be striking.

Despite the deferential tone conveyed through the use of the plural of formality and formal address forms, the data revealed a clear predominance of direct, rather than indirect, request strategies. Notably, a marked preference for want-statements was observed in Situations 1 and 3, whereas need-statements predominated in Situations 2 and 4. Participants’ verbal reports offered insight into this variation, suggesting that their differing attitudes toward the two sets of situations influenced their choice of request formulation. Their comments are illustrative:1

P#3: They were a bit different [referring to the situations]. In 1 and 3 I was asking for something for myself, so it was tricky. I felt I needed to be more polite, tread carefully. It was my right to have what I was asking for, but still… In 2 and 4 not so much. I mean, what I was asking for in those had to do with the good of the business. It was in my employer’s best interest to comply.

P#68: I felt I was entitled to a holiday and to be paid fairly. I tried to show this clearly. Well, in this and this [pointing to Situations 2 and 4] what I was asking for was not only necessary for me. It was about the job being done well; it was something needed for the benefit of everyone involved.

These comments highlight a key contrast in the framing of requests. In Situations 1 and 3, participants tended to present their requests as grounded in entitlement and personal need, often accompanied by a stronger tone of self-enhancement and speaker perspective. In contrast, Situations 2 and 4 were more frequently framed using inclusive perspective and justifications that highlighted the benefit to the business or workplace function. This pattern suggests that participants were attuned not only to the content of their requests but also to the perceived legitimacy of the need, adjusting their strategies accordingly.

Moreover, all situations exhibited high frequencies of external modification. According to Blum-Kulka et al.’s (1989), external modification functions as a positive politeness strategy, aiming to reduce the force of the request through supportive moves. This was reflected in the present data, where participants employed high frequencies of such strategies in all situations to maintain rapport and express consideration.

The portrayal of the request’s fulfilment as beneficial to the business, rather than to the employee, appears to have licenced the lower frequencies of syntactic and lexical mitigation found in Situations 2 and 4 compared to 1 and 3. This pattern aligns with the types of reasons given in support of the requests. As noted in Section 4, justifications appealing to the benefit of the business were considerably more frequent in Situations 2 and 4, whereas those highlighting the employees’ professional value and efficiency were preferred in Situations 1 and 3. This preference was echoed in participants’ comments:

P#32 (on Situation 3): I know it may sound like I am bragging here, saying what a good employee I am and all, but I had to show her that I deserved this raise. I didn’t want her to think I’m just being greedy. It may sound like bragging, but I was trying to be convincing and polite.

P#81 (on Situation 2): It was important to show them that it was not just about my own convenience. Computers must work properly for the job to be done properly.

Ultimately, participants sought to balance politeness with assertiveness, demonstrate concern for the workplace’s effective functioning, and navigate the tension between personal and organisational needs with situations foregrounding personal benefit (1 and 3) being perceived as more sensitive or challenging than those aligned with collective goals or team success (2 and 4). Crucially, participants appeared to be navigating between what Spencer-Oatey (2008, p. 29) refers to as linguistic strategies of associative restraint and strategies of associative expressiveness. The former correspond broadly to Brown and Levinson’s (1987) negative politeness strategies, while the latter align with their positive politeness strategies.

Associative restraint was primarily manifested with the polite plural, formal modes of address, syntactic modifications, and speaker perspective aimed at minimising the imposition of the request. Associative expressiveness, on the other hand, sought to establish common ground by invoking shared attitudes, knowledge, and empathy, asserting reciprocity and in-group membership with the requestee. Its main manifestations in the data included disarmers, reasons portraying participants as interested in the benefit of the business, empathetic and interpersonal markers, and inclusive perspective.

5.2. Rapport and Identity Construction

The analysis of the pragma-linguistic realisations of the four scenarios demonstrates that participants’ choice of head-act, internal, and external modification, address forms and request perspective, systematically respond to the rapport-management concerns identified by Spencer-Oatey (2008). Three categories appear to dominate the data:

First, requesters work to enhance their own social identity face by supplying self-promotional justifications and by foregrounding professional competence when asking for leave and salary adjustments. These self-enhancements aim to secure a favourable personal evaluation while simultaneously legitimising the request. Although it is more prominent in Situations 1 and 3, the participants’ marked positioning as members of a group with common interests and goals also points to a worthy employee, albeit more indirectly.

Second, participants attend to association rights in at least two ways: (1) by calibrating politeness resources to institutional hierarchy, and (2) by foregrounding the business’s interests and by deploying disarmers that acknowledge the manager’s workload. The uniform use of polite plural address forms and the near-absence of second-person pronouns minimise imposition and protect the superior’s public self-image, thereby sustaining an unequal yet harmonious relationship.

Third, speakers invoke equity rights whenever they frame leave and pay adjustments as matters of fairness or contractual entitlement. The prevalence of direct want statements in these scenarios signals confidence in claiming rights without necessarily threatening rapport.

Together, these three orientations shape the linguistic design of requests and foreshadow the speaker identities that emerge from the data. That is, the same patterns that enable rapport maintenance and enhancement also function to project distinct yet interrelated professional personas. Specifically, the data supports the emergence of four complementary identity dimensions: (a) the deferential-hierarchical self, (b) the competent, deserving professional, (c) the team player, and (d) the rights-bearing, but considerate, negotiator. These identities manifest themselves as follows:

(a)

The deferential self

Consistent recourse to the polite plural and the systematic use of formal address forms index a speaker stance that foregrounds institutional hierarchy. These forms, combined with the notable absence of hearer-oriented perspective markers, serve to protect the superior’s social identity face while simultaneously calibrating distance within the dyad. In effect, speakers position themselves as respectful subordinates who recognise the legitimacy of the workplace’s power differential reflecting the broader institutional expectations of deference in hierarchical settings.

(b)

The competent, deserving professional

Self-promotional justifications, particularly salient in the leave and salary scenarios, activate the social identity face component of rapport management. Here, participants construct an identity as diligent and capable professionals whose performance justifies recognition and reward. As illustrated in Section 5.1, one requester explicitly references her “dedication and abilities” prior to submitting a pay-rise request, a move that functions both to elevate perceived competence and to legitimise the imposition. Retrospective reflections reinforce this identity: Participant #32 notes that she had to “show that I deserved this raise,” directly tying entitlement to demonstrable professional merit. Such self-enhancing strategies, while overtly evaluative, also reflect a culturally sanctioned narrative of worthiness grounded in achievement.

(c)

The team player

In the computer-upgrade and deadline-extension scenarios, participants frequently reframe individual requests as serving broader organisational interests. By deploying inclusive pronouns and emphasising collective outcomes (e.g., “computers must work properly for the job”), speakers satisfy the association rights dimension of rapport and project an identity aligned with institutional goals. This construction positions the requester not merely as a beneficiary of change but as a contributor to organisational efficacy and shared success. As discussed earlier, these strategies subtly recast potentially self-serving acts as expressions of loyalty and collaboration.2

(d)

The rights-bearing and considerate negotiator

Despite the inherently face-threatening nature of requests, participants predominantly employ direct head acts, signalling confident appeals to equity rights. However, this assertiveness is consistently tempered by dense mitigation, including external modifiers, disarmers, and reciprocal commitments, which attend to the superior’s position and workload. The resulting discursive identity is that of a knowledgeable and confident employee who claims entitlements while remaining sensitive to relational dynamics. This balanced stance—assertive yet deferential—reflects the dynamic calibration of rapport strategies observed across scenarios and exemplifies Spencer-Oatey’s (2008, p. 17) notion of “dynamic calibration of face wants”.

Those identities, which were found to be systematically co-existent and not mutually exclusive, point to what has been previously called an egalitarian—deferential interdependence orientation (Bella 2021): speakers assert their rights and competence through direct forms while simultaneously neutralising face threat with formality, rapport-enhancing modifiers, and explicit appeals to shared goals. These findings mirror closely Bella’s (2021) findings on Greek university students’ requests to faculty, suggesting a possibly culturally entrenched practice in Greek hierarchical settings

6. Conclusions

This study set out to investigate how Greek professionals craft requests to hierarchical superiors and, in so doing, negotiate rapport and workplace identities. Despite several limitations, including the use of a written DCT, the restricted set of scenarios addressed solely to superiors, a gender-skewed sample that may limit generalisability, and the absence of comparison with peer-to-peer or subordinate-directed requests that would reveal whether the observed “direct-yet-mitigated” pattern is hierarchy-specific, this study yields valuable preliminary insights into Greek workplace requesting.

The coexistence of entitlement-based directness with extensive mitigation and formal politeness supports the central argument that requesting in institutional settings is simultaneously a relational and identity-negotiating act. Interpreting participants’ pragmatic choices through the lens of rapport management and identity construction exposes a multilayered self-image (deferential, competent, cooperative, and rights-aware) that Greek employees project.

Beyond theoretical contribution, these findings carry practical implications: HR managers and intercultural trainers can prepare expatriate supervisors for Greek offices by stressing that directness does not imply brusqueness when paired with appropriate redressive moves, and by designing communication workshops that balance clarity of task orientation with face-sensitive politeness routines.

Future work should pair controlled instruments like those used here with naturally occurring workplace corpora and widen the comparative lens to additional cultural contexts; such research will refine our understanding of how directness, mitigation, and identity co-evolve in real-time professional discourse and test the generalisability of the relational style identified in this study.