Introduction

Apology is one of the most visible forms of remedial action in everyday interaction. It is a speech act through which speakers acknowledge, mitigate, explain, or repair an offence that has affected an interlocutor, a relationship, or a shared interactional order. Within pragmatics and sociolinguistics, apologies have therefore been examined not merely as formulaic expressions of regret, but as culturally situated practices through which speakers negotiate responsibility, face, solidarity, politeness, and interpersonal repair. In Algerian Arabic, apology expressions circulate in highly varied forms, ranging from direct formulae of regret to longer sequences that combine explanation, justification, compensation, and relational reassurance.

A direct apologetic formula may vary according to the gender of the addressee and may include a kinship term that indexes closeness, solidarity and mitigation.

Example 1. Direct apology formula addressed to a male interlocutor

/smaħ.li xuː.ja/
سمحلي خويا
‘I am sorry, brother.’

Example 2. Direct apology formula addressed to a female interlocutor

/smaħi.li xti/
سمحيلي أختي
‘I am sorry, sister.’

These examples show that apology in Algerian Arabic is not only a matter of expressing regret. It also involves relational positioning through address terms such as /xuː.ja/ ‘my brother’ and /xti/ ‘my sister’, which soften the offence and help re-establish interpersonal proximity. Such forms reveal the importance of address, solidarity and social repair in the pragmatics of apology.

The present study is situated at the intersection of speech act theory, socio-pragmatics, and the analysis of gendered linguistic behaviour. It investigates the apology strategies used by Algerian male and female postgraduate students in four situations that require an apology. More specifically, it asks whether the participants rely on the same broad apology strategies, whether their frequencies differ by gender, and how Algerian Arabic apology practices can be described in relation to Sugimoto’s (1997) taxonomy of apology strategies.

The study is motivated by a gap in the literature. While apology has been widely examined in cross-cultural pragmatics, the Algerian context remains comparatively underrepresented, particularly when gender is introduced as an analytical variable. Studies by Allili (2016), Derki (2023), and Dahmani and Al Khalaf (2024) have begun to document apology strategies in Algerian Arabic and Kabyle-speaking contexts. However, further empirical descriptions are still needed in order to refine the account of how Algerian speakers combine direct and indirect apologetic moves, and how these combinations may vary according to interactional situation, social relationship and participant profile.

This article pursues two main objectives. First, it identifies the apology strategies employed by Algerian students in the responses elicited through the Discourse Completion Test. Second, it compares the strategies used by male and female participants, not in order to essentialise gendered behaviour, but to describe the tendencies observable within the limited sample under study. The contribution of the article lies in its focused description of apology strategies in Algerian Arabic, its use of a balanced male/female sample, and its identification of a recurrent composite pattern that the present study labels the mixed strategy.

Two research questions guide the analysis:

• What apology strategies are used by Algerian students in the four situations included in the Discourse Completion Test?

• What similarities and differences can be observed between male and female participants in their use of apology strategies?

The principal contribution of the study is thus threefold. It documents apology strategies in an Algerian Arabic-speaking sample; it tests the descriptive usefulness of Sugimoto’s taxonomy in a context different from the one in which it was initially developed; and it proposes the mixed strategy as a useful coding label for responses that combine several apology moves within one utterance. This final point is particularly relevant because the most frequent responses in the data are not simple formulae, but composite sequences in which remorse, explanation, justification, compensation or relational reassurance may be combined.

The gender variable is approached descriptively. The article does not assume that men and women possess fixed communicative natures. Rather, it compares the distribution of strategy types within a small, balanced sample and interprets differences as tendencies internal to the data. This methodological caution is important because gender differences in pragmatic behaviour cannot be reduced to essential or universal traits. They must be examined in relation to corpus size, situation type, offence severity, social distance and the interactional expectations associated with a given speech community. The analysis therefore avoids broad claims about male or female politeness and focuses instead on frequencies, strategy combinations and orientations toward interpersonal repair in the corpus.

The Algerian Arabic dimension also deserves emphasis. Algerian Arabic is not treated here as a homogeneous or monolithic language variety, but as a socially situated repertoire in which apologies may draw on kinship terms, religious expressions, face-saving routines and contextual explanations. The examples reproduced in the analysis are therefore not merely illustrations; they are part of the linguistic evidence. They show how an apology may be organised around formulae such as /smaħ.li/ and /smaħi.li/, and around relational address terms such as /xuː.ja/ and /xti/, which index closeness, solidarity and mitigation.

The study deliberately remains focused on apology as a speech act rather than on politeness in general. This delimitation is important because apology occupies a specific pragmatic position: it normally appears after a perceived offence and presupposes that the speaker recognises, negotiates or mitigates some form of responsibility. In the four situations used in the Discourse Completion Test, the offence concerns damaged objects, a broken promise or failure to complete an academic task. These situations are sufficiently ordinary to elicit routine apologetic forms, but also sufficiently varied to reveal whether participants prefer direct regret, explanation, compensation, or combinations of these moves.

By examining these responses, the study contributes to a more precise understanding of apology practices in Algerian Arabic. It also shows that apology strategies cannot be reduced to isolated formulae. In many cases, apology is organised as a sequence of moves through which the speaker acknowledges the offence, explains its circumstances, proposes repair and seeks to preserve the relationship with the interlocutor. The analysis that follows therefore considers both the type of strategy selected and the way in which several strategies may be combined within a single apologetic utterance.

1. Review of Related Literature

1.1 Speech Act Theory and Definitions of Apology

Speech act theory provides the general theoretical background of the present study. Austin (1962) argued that utterances do not merely describe reality but may also perform actions such as ordering, promising, thanking or apologising. Searle (1969, 1979) further developed this view by classifying speech acts into major functional categories, including representatives, directives, commissives, expressives and declarations. Apology belongs to the expressive domain because it communicates the speaker’s attitude toward an offence, a failure or a socially problematic action.

In pragmatic terms, apology is closely linked to the restoration of social relations. Olshtain (1989) presents apology as a speech act that provides support to a hearer affected by a violation. Goffman’s account of remedial work is also central, because apology is part of the interactional procedures through which social order is repaired after an offence. Goffman defines apologies as follows: “Remedial interchanges used to re-establish social harmony after a real or virtual offence” (Goffman, 1971).

This definition is important because it shows that apology cannot be reduced to a fixed linguistic formula. It is a socially situated act through which speakers acknowledge trouble, repair face and attempt to restore an interactional relationship. Apology strategies may therefore vary across languages, cultures, social roles, degrees of offence and interactional settings. In this sense, apology is both a linguistic act and a relational practice: it expresses regret, but it also negotiates responsibility, mitigates conflict and reopens the possibility of social continuity after a disruption.

1.2 Sugimoto’s Taxonomy of Apology Strategies

Sugimoto’s (1997) taxonomy is used as the principal analytical framework of the study. As reported in studies such as Bataineh and Bataineh (2006) and Muzhir and Raheem (2012), Sugimoto distinguishes three broad categories of apology strategies: primary strategies, secondary strategies and seldom-used strategies. These categories make it possible to classify apologetic responses according to the type of remedial move performed by the speaker. The following table summarises the taxonomy used for coding the responses in the present study.

Table 1. Sugimoto’s taxonomy of apology strategies as used in the present study

Category	Sub-strategy	Definition	Illustrative example
Primary strategies	Statement of remorse	The speaker directly expresses regret or sorrow.	I am very sorry for what happened.
Primary strategies	Accounts	The speaker explains what happened.	It rained while the book was in my hand.
Primary strategies	Description of damage	The speaker describes the damage or consequence of the offence.	The book was slightly damaged.
Primary strategies	Reparation	The speaker offers to repair the damage.	God willing, I will repair your device.
Secondary strategies	Compensation	The speaker offers to replace the damaged object or pay for it.	I will buy you another device.
Secondary strategies	Promise not to repeat the offence	The speaker promises that the offence will not happen again.	This will not happen again.
Seldom-used strategies	Explicit assessment of responsibility	The speaker assesses or limits responsibility for the offence.	Of course, I did not do it on purpose.
Seldom-used strategies	Contextualisation	The speaker describes the wider context in which the offence occurred.	The shops were closed.
Seldom-used strategies	Self-castigation	The speaker blames himself or herself for the offence.	It is my fault.
Seldom-used strategies	Gratitude	The speaker thanks the offended person for listening or forgiving.	Thank you for accepting my apology.

This taxonomy is useful because it does not limit apology to direct expressions such as I am sorry. It also accounts for explanations, repair offers, compensation, promises and contextualising moves. Such a framework is particularly relevant for the analysis of Algerian Arabic data, where apologetic sequences may combine direct regret with explanation, justification, relational address terms and offers of repair. The taxonomy therefore provides a stable point of departure, while still allowing the analysis to observe how speakers organise several apology moves within one response.

1.3 Apology, Gender and Algerian Arabic

Apology has been investigated from several cross-cultural and socio-pragmatic perspectives. Blum-Kulka and Olshtain (1984), Sugimoto (1997), Hussein and Hammouri (1998), and Soliman (2003) demonstrate that apologies differ across linguistic and cultural communities. Their findings support the view that pragmatic competence involves knowing not only which linguistic formula to use, but also how to adjust an apology to the offence, the interlocutor, the social relationship and the wider cultural context.

Research on gender and apology has produced mixed results. Holmes (1989, 1995) argues that women tend to apologise more frequently in some contexts, whereas other studies report more limited or context-dependent differences. Bataineh and Bataineh (2008), for example, found gender-related differences in apology strategies among Jordanian speakers, while Ali-Harb (2016) reported no statistically significant differences in an Arabic-speaking context. These divergent findings show that gender should be treated as an empirical variable rather than as a predetermined explanation. In other words, the analysis of gender and apology should not begin from essentialised assumptions about male or female speech, but from observable patterns in the data.

In the Algerian context, apology remains less extensively documented than in other linguistic and cultural settings. Allili (2016) examined apology strategies among Algerian university students of English and highlighted the influence of social power, social distance and offence severity. Derki (2023) specifically addressed gender and apology in Algerian Arabic and found that men and women used similar broad types of strategies, although their frequencies differed. Dahmani and Al Khalaf (2024) studied apology strategies among Algerian Kabyle speakers and identified culture-specific forms such as requesting patience, asking the hearer not to be angry and offering religiously marked expressions. These studies show that apology in Algerian contexts cannot be reduced to a simple expression of regret. It is embedded in social norms, address practices, expectations of repair and culturally meaningful ways of mitigating interpersonal disturbance.

The present study contributes to this still limited body of research by describing apology strategies in a balanced male/female sample of Algerian postgraduate students. Its positioning is especially important in relation to gender. Earlier studies have sometimes framed women as more apologetic or men as less apologetic. The present article avoids this binary simplification. It asks instead how apology moves are distributed in the corpus and whether different combinations become visible when responses are coded by gender. The emphasis therefore falls on patterns of use, frequency distributions and strategy combinations, not on essentialised identities.

The adoption of Sugimoto’s taxonomy offers a useful classificatory basis, but the Algerian data also require analytical flexibility. The mixed strategy, which appears frequently in the responses, does not contradict Sugimoto’s categories. Rather, it shows that participants often combine them. A single apology may express remorse, explain the context and offer compensation within the same utterance. Such sequencing is pragmatically meaningful because it allows the speaker to address responsibility, causality and repair at the same time. The mixed strategy is therefore treated in this study as a descriptive coding label for composite apologetic responses.

Against this background, the present study positions itself as a focused contribution to the socio-pragmatic description of apology strategies in Algerian Arabic. It does not claim to offer a statistically generalisable account of gendered apology behaviour. Rather, it documents the strategies produced by a balanced sample of Algerian male and female postgraduate students in four apology-eliciting situations. Its specific contribution lies in combining Sugimoto’s (1997) taxonomy with a cautious descriptive comparison of gender-based frequencies and in identifying mixed strategies as a recurrent pattern in the data. In this respect, the study extends existing research on apology in the Algerian context while acknowledging the exploratory nature of its findings.

The literature reviewed above shows that apology is shaped by several interacting factors: language, culture, offence severity, social power, social distance and gender. This study does not attempt to isolate all these variables. Instead, it focuses on one manageable comparison between male and female participants while keeping the Algerian Arabic context at the centre of the analysis. The aim is to produce a fine-grained descriptive account that may later be tested in broader studies based on larger samples, naturally occurring interaction and more diversified sociolinguistic variables.

2. Research Methodology

2.1 Research Design, Participants and Data Collection Instrument

This study adopts a descriptive socio-pragmatic design aimed at identifying the apology strategies used by Algerian students and examining whether male and female participants display convergent or divergent patterns in their production of apologies. The study is exploratory in scope: it does not seek to establish statistically generalisable laws, but rather to describe recurrent pragmatic tendencies within a clearly delimited group of Algerian postgraduate students.

To achieve these objectives, a Discourse Completion Test (DCT) was used as the main instrument of data collection. The DCT, widely employed in interlanguage pragmatics and cross-cultural pragmatics, enables researchers to elicit comparable responses to controlled communicative situations. In the present study, participants were asked to respond in writing to four situations in which an apology would normally be expected. These situations involved ordinary interpersonal contexts, such as damaging a borrowed object, breaking a promise, failing to complete an academic task, or damaging an item belonging to a sibling. The use of the DCT was therefore appropriate for the purpose of the article, since it allowed the researcher to obtain comparable apology data from all participants and to examine the strategies selected by male and female respondents under similar conditions.

The participants consisted of thirty Algerian postgraduate students studying in Jordan. The sample included fifteen male and fifteen female participants, whose ages ranged from twenty-four to twenty-six years. The balanced gender composition of the group made it possible to compare male and female responses descriptively. However, the sample was selected through convenience sampling, due to limitations of time, access and availability. Participants were recruited from the researcher’s immediate academic environment because they were accessible and willing to take part in the study. Since the sample is small and non-probability-based, the findings cannot be generalised to all Algerian students or to all Algerian Arabic speakers. They should instead be read as exploratory evidence that sheds light on apology practices within this particular group of Algerian postgraduate students.

2.2 Coding Procedure and Analytical Framework

After the DCT responses had been collected, each answer was examined, coded and counted manually. The coding procedure was guided primarily by Sugimoto’s (1997) taxonomy of apology strategies, which distinguishes between three broad categories: primary strategies, secondary strategies and seldom-used strategies. Primary strategies include direct expressions of remorse, accounts, reparation and description of damage. Secondary strategies include compensation and promises not to repeat the offence. Seldom-used strategies include explicit assessment of responsibility, contextualisation, self-castigation and gratitude.

During the coding process, an additional recurrent pattern was observed in the participants’ responses. Several apologies did not rely on a single strategy only, but combined two or more strategies within the same utterance. For example, a participant could express remorse, explain the circumstances of the offence and offer compensation in one apology sequence. These composite responses were therefore coded as mixed strategies. The introduction of this category does not replace Sugimoto’s framework; rather, it extends its descriptive application to the present data, since the Algerian responses often display a cumulative structure in which apology, justification and repair are combined.

The analysis proceeds in two stages. First, the overall distribution of apology strategies is presented across the entire data set in order to identify the dominant types of strategies used by the participants. Second, the responses of male and female participants are compared in terms of frequencies and percentages. The aim of this comparison is not to prove statistically significant gender differences, but to describe observable tendencies in the data. For this reason, the analysis remains descriptive and interpretive. Frequencies and percentages are used to organise the results and facilitate comparison, but they should not be read as evidence of statistically established gender-based differences.

This methodological caution is important. The study identifies patterns of apology use, but it does not claim to establish general laws about gender and apology in Algerian Arabic. The data allow us to observe how this particular group of male and female Algerian students responded to the four DCT situations, but they do not allow broad conclusions about all Algerian speakers. The discussion of gender differences must therefore remain cautious, especially when the differences observed are differences of frequency rather than differences in the types of strategies used.

2.3 Ethical Considerations and Methodological Limitations

Basic ethical principles were observed throughout the investigation. Before completing the DCT, participants were informed about the purpose of the study and about the academic use of the data. They were also told that participation was voluntary and that they were free to decline participation. Confidentiality was ensured by keeping the participants’ names anonymous throughout the process of analysis and reporting. The data collected were used exclusively for research purposes.

Several methodological limitations should nevertheless be acknowledged. First, the DCT elicits written responses to imagined situations. It is useful for collecting comparable pragmatic data, but it cannot fully reproduce the interactional complexity of naturally occurring conversation. It does not capture prosody, hesitation, gesture, repair initiated by the hearer, negotiation between interlocutors, or the sequential development of real apology exchanges. Second, the sample is limited to thirty Algerian postgraduate students studying in Jordan. Although the gender balance of the sample is useful for descriptive comparison, its small size prevents statistical generalisation. Third, the data were coded manually by the researcher. Manual coding is acceptable in exploratory pragmatic research, but future studies should strengthen reliability by involving at least two independent coders and reporting inter-coder agreement. Fourth, the four DCT situations may differ in terms of offence severity, social distance, relational closeness and the nature of the damaged object or broken obligation. These factors may influence the choice of apology strategy and should be examined more systematically in future research.

These limitations do not invalidate the study. Rather, they define the exact scope within which the findings should be interpreted. The study should be understood as an exploratory contribution to the socio-pragmatic analysis of apology strategies among Algerian students. It provides useful descriptive evidence and opens the way for broader investigations based on larger samples, naturally occurring interactions, inferential statistical analysis and a more detailed examination of variables such as age, gender, social status, offence severity and interlocutor relationship.

3. Results and Discussion

3.1 Overall Distribution of Apology Strategies

The first research question asks what apology strategies are used by Algerian students. In order to answer this question, the responses produced by fifteen male and fifteen female participants across four situations were coded and counted. Since each participant responded to four situations, the data set contains 120 coded responses. Table 2 presents the overall distribution of the broad strategy types.

Table 2. Frequencies and percentages of apology strategies used by Algerian students

Strategy type	Male (F)	Female (F)	Total (F)	Percentage
Primary strategies	16	15	31	25.83%
Secondary strategies	6	19	25	20.83%
Seldom-used strategies	3	5	8	6.67%
Mixed strategies	35	21	56	46.67%
Total	60	60	120	100%

Two observations emerge from Table 2. First, the participants used all three strategy types included in Sugimoto’s (1997) taxonomy: primary, secondary and seldom-used strategies. Second, the data show a frequent use of mixed strategies, a category not listed as an independent type in Sugimoto’s taxonomy. In this article, a mixed strategy refers to a response in which the participant combines two or more apology moves within a single utterance, for example remorse plus contextualisation plus compensation.

The mixed strategy is the most frequent category, accounting for 46.67% of all coded responses. This suggests that the participants often preferred elaborated apologies rather than relying on a single apologetic move. Primary strategies constitute the second most frequent category, with 25.83% of the responses. Secondary strategies account for 20.83%, while seldom-used strategies remain the least frequent, representing 6.67% of the total. The low frequency of seldom-used strategies may indicate that participants preferred direct and repair-oriented apologetic moves over more indirect or contextualising ones.

3.2 Gender-Based Comparison of Strategy Types

The second research question asks whether similarities or differences can be observed between male and female participants. The same frequencies can be read comparatively in Table 3, which presents the gender-based distribution of the four broad strategy types.

Table 3. Gender-based comparison of apology strategy types

Strategy type	Male (F)	Female (F)	Total (F)	Percentage
Primary strategies	16	15	31	25.83%
Secondary strategies	6	19	25	20.83%
Seldom-used strategies	3	5	8	6.67%
Mixed strategies	35	21	56	46.67%
Total	60	60	120	100%

Both groups produced the same total number of responses: sixty for male participants and sixty for female participants. The broad choice of strategy types is therefore shared across the two groups. The main differences lie in frequency. Male participants produced more mixed strategies (35 occurrences) than female participants (21 occurrences). Female participants, by contrast, produced more secondary strategies (19 occurrences) than male participants (6 occurrences). Primary strategies were almost equally distributed, with 16 male occurrences and 15 female occurrences. Seldom-used strategies were rare in both groups.

These findings should be interpreted cautiously. They do not prove that one gender is more apologetic or more polite than the other. They indicate only that, in this small sample, male participants more often combined several apologetic moves in one response, whereas female participants more often used secondary strategies, especially compensation.

3.3 Primary Strategies

Primary strategies include statements of remorse, accounts, reparations and descriptions of damage. Table 4 presents the sub-types of primary strategies identified in the data.

Table 4. Sub-types of primary strategies used by male and female participants

Group	Statement of remorse	Accounts	Reparation	Description of damage	Total
Male	15 (48.39%)	1 (3.23%)	0 (0%)	0 (0%)	16 (51.61%)
Female	15 (48.39%)	0 (0%)	0 (0%)	0 (0%)	15 (48.39%)
Total	30 (96.78%)	1 (3.23%)	0 (0%)	0 (0%)	31 (100%)

The two groups relied overwhelmingly on statements of remorse. Both male and female participants produced fifteen occurrences of this sub-strategy. Accounts appeared only once, and only in a male response. Reparation and description of damage were not used by either group. This distribution indicates a strong preference for direct expressions of apology in the primary category.

Table 5. Detached examples of primary strategies

Sub-strategy	English translation	Algerian Arabic example	IPA / transcription
Statement of remorse	I am very sorry, brother.	سمحلي خويا	/smaħ.li xuː.ja/
Accounts	There was too much wind while the umbrella was open in my hand.	كان بزاف ريح كي كانت لبرابلي في يدي مفتوحة	/kaːn bəzˈzaːf riːħ kiː kaːnət l.pa.ʁa.pliː fiː jɛd.di mɛfˈtuːħa/

The single occurrence of an account among male participants should not be overinterpreted. It indicates that explanation is available as a resource, but that it is not the preferred primary move in this corpus. Explanations become more visible when they are incorporated into mixed strategies or classified under contextualisation. This confirms the usefulness of examining not only isolated categories but also the way categories combine in actual responses.

The absence of reparation and description of damage in the primary category is noteworthy. In the situations involving damaged objects, one might expect participants to describe the damage before proposing repair. Instead, most participants moved directly to remorse or, in other categories, to compensation. This suggests that the acknowledgement of the interpersonal offence may be more salient than the detailed description of the material damage. In other words, the participants appear to orient first toward the offended person and only secondarily toward the damaged object.

3.4 Secondary Strategies

Secondary strategies include compensation and promise not to repeat the offence. In the present corpus, only compensation was used.

Table 6. Sub-types of secondary strategies used by male and female participants

Group	Compensation	Promise not to repeat	Total
Male	6 (24%)	0 (0%)	6 (24%)
Female	19 (76%)	0 (0%)	19 (76%)
Total	25 (100%)	0 (0%)	25 (100%)

The compensation strategy was used more frequently by female participants than by male participants. Female responses account for 76% of all secondary strategies, whereas male responses account for 24%. This pattern may suggest that, in this sample, female participants more often oriented their apology toward material repair or replacement. However, the limited size of the sample requires caution, and the result should not be generalised beyond the data.

Table 7. Detached example of a secondary strategy

Sub-strategy	English translation	Algerian Arabic example	IPA / transcription
Compensation	Your umbrella was broken; I will buy you another one.	تكسرت لباربلوي ديالك نشريلك وحدة أخرى	/tɛk.sə.rɛt l.pa.ʁa.plwi dja.lək nɛʃ.ri.lək waħ.da ʔux.ʁa/

The absence of promises not to repeat the offence may be explained by the specific situations. Breaking an object or failing to keep a promise may call more directly for immediate repair than for a future-oriented commitment. However, this interpretation remains tentative. A different DCT including recurring offences, lateness or repeated neglect might elicit more promises not to repeat the offence.

The dominance of compensation within secondary strategies is consistent with the nature of several DCT situations. When an umbrella or a watch is broken, compensation provides a concrete means of restoring the relationship and addressing the material consequence of the offence. It is therefore both an interpersonal and practical strategy: it signals responsibility while offering a solution.

3.5 Seldom-Used Strategies

Seldom-used strategies include explicit assessment of responsibility, contextualisation, gratitude and self-castigation. Table 8 presents their distribution in the corpus.

Table 8. Sub-types of seldom-used strategies used by male and female participants

Group	Explicit assessment	Contextualisation	Gratitude	Self-castigation	Total
Male	1 (12.5%)	2 (25%)	0 (0%)	0 (0%)	3 (37.5%)
Female	1 (12.5%)	4 (50%)	0 (0%)	0 (0%)	5 (62.5%)
Total	2 (25%)	6 (75%)	0 (0%)	0 (0%)	8 (100%)

Only two seldom-used strategies appear in the data: explicit assessment of responsibility and contextualisation. Gratitude and self-castigation do not occur. Contextualisation is the most frequent sub-type within this category, with six occurrences out of eight. Female participants produced four contextualising responses, while male participants produced two. This may indicate that some female participants in this sample provided slightly more contextual information when constructing an apology, but the low number of cases prevents any strong conclusion.

Table 9. Detached examples of seldom-used strategies

Sub-strategy	English translation	Algerian Arabic example	IPA / transcription
Explicit assessment of responsibility	Of course, I did not break it on purpose.	مشي بلعاني كسرتها	/ˈmaʃi blaʕˈni ˈkæsartˤha/
Contextualisation	Brother, the wind was strong; that is why it broke.	خويا كان ريح بزاف كسرها	/ˈxuː.ja kaːn rˤˈriːħ bəzˈzæːf ˈkæsarha/

Contextualisation is nevertheless significant. When participants mention strong wind, closed shops or circumstances beyond their control, they do not necessarily deny responsibility. Rather, they make the offence intelligible and attempt to reduce its face-threatening force. Contextualisation therefore functions as a mitigating move: it frames the offence as accidental, constrained or situationally explainable.

The limited use of seldom-used strategies may reflect the relatively direct character of the apology situations. Participants did not often thank the hearer for accepting the apology, nor did they blame themselves explicitly. Instead, they tended to combine remorse and repair. This does not mean that gratitude or self-castigation are absent from Algerian Arabic apology practices in general; it means only that they were not triggered by the four written scenarios used in this study.

3.6 Mixed Strategies

Mixed strategies are responses that combine two or more apology moves within the same utterance. They are not treated here as a replacement for Sugimoto’s taxonomy, but as an additional coding label needed to describe the composite responses found in the data.

Table 10. Distribution of mixed strategies by gender

Group	Number using mixed strategies	Percentage within mixed strategies
Male	35	62.50%
Female	21	37.50%
Total	56	100%

Table 11. Detached example of a mixed strategy

Strategy	English translation	Algerian Arabic example	IPA / transcription
Mixed strategy	I am sorry, my friend. There was strong wind, and it broke. I will buy you another one.	سمحلي خويا، كان ريح قوي وتكسرت، نشري لصاحبي واحدة خرى	/sɛˈmaħli ˈxuː.ja | kaːn rˤiːħ qaˈwiː | wətˤkæsˤːərət | nəʃri lˈsˤaħbi ˈwaħda xra/

The example combines three apologetic moves: a statement of remorse, contextualisation and compensation. Mixed strategies are the most frequent strategy type in the data, and they are used more often by male participants than by female participants. This may suggest that male participants in this sample tended to produce more elaborated apologetic sequences. Nevertheless, the difference is descriptive rather than inferential. It should therefore be interpreted as a tendency within the corpus rather than as a general claim about male or female apology behaviour.

The mixed strategy is the most important analytical result. It shows that a taxonomy built on separate categories may need an additional layer capable of capturing combinations. In pragmatic terms, such combinations are not accidental. They constitute complete apologetic sequences, often organised as follows: remorse first, explanation second, and repair or compensation third. This sequence is interactionally coherent because it moves from affective recognition to causal framing and then to practical repair.

The gender comparison confirms convergence more than opposition. Both male and female participants used the same broad categories and relied heavily on direct remorse. The differences observed concern the distribution of composite and secondary strategies. Male participants produced a higher number of mixed strategies; female participants produced a higher number of compensation strategies. These differences are analytically useful, but they remain limited to the corpus and should be tested through larger samples and naturalistic data.

Taken together, the results indicate that apology in this corpus is best understood as a combination of directness and elaboration. Directness appears in the high frequency of statements of remorse. Elaboration appears in the frequent use of mixed strategies, where participants add contextualisation and compensation to the initial apology. This combination allows speakers to fulfil several pragmatic tasks simultaneously: acknowledge the offence, mitigate the cause, preserve the relationship and offer repair.

3.7 Synthesis of Findings

The results of the four elicitation situations make it possible to formulate a consolidated reading of apology behaviour among the Algerian postgraduate students who participated in the study. The first major finding is that apology is not realised as a single, isolated formula. Although explicit expressions of remorse remain central, the participants frequently combined several pragmatic moves in the same response. This confirms that apology, in Algerian Arabic as represented in the corpus, is better understood as a composite remedial act through which the speaker recognises the offence, repairs the interpersonal imbalance and seeks to preserve the relationship with the offended party.

The distribution of the main categories provides a clear answer to the first research question. The participants used the three broad categories proposed by Sugimoto (1997): primary, secondary and seldom-used strategies. Primary strategies accounted for 31 occurrences out of 120 coded responses, that is 25.83%. Secondary strategies accounted for 25 occurrences, or 20.83%. Seldom-used strategies were less frequent, with 8 occurrences, or 6.67%. The most salient result, however, is the high frequency of mixed strategies, which reached 56 occurrences, representing 46.67% of the total. This category is analytically important because it shows that the respondents did not merely choose between direct and indirect apology forms; rather, they often organised apology as a sequence of regret, explanation and repair.

At the level of sub-strategies, the statement of remorse appears as the dominant primary device. Expressions equivalent to 'I am sorry' or 'forgive me' function as the most immediate way of acknowledging the offence and opening the remedial exchange. Compensation is the dominant secondary strategy, especially in situations involving material damage such as the broken umbrella or the broken watch. Contextualisation is the most visible seldom-used strategy, since some participants attempted to explain the circumstances of the offence by referring, for instance, to strong wind or to an accidental situation. By contrast, some sub-strategies remained absent from the corpus: description of damage, reparation, promise not to repeat the offence, gratitude and self-castigation. This absence should not be interpreted as proof that such strategies are unavailable in Algerian Arabic; it simply indicates that they were not selected by this small group of participants in the four situations proposed by the DCT.

The second research question concerned the possible role of gender. The data point to similarity more than categorical difference. Male and female participants used the same general strategy types, and both groups relied heavily on explicit remorse. The differences are therefore mainly differences of frequency, not differences of pragmatic repertoire. Male participants produced more mixed strategies, whereas female participants produced more secondary strategies, particularly compensation. Female participants also used contextualising responses slightly more often within the seldom-used category. These tendencies must be interpreted cautiously: the sample is small, the participants were selected through convenience sampling, and the study did not test statistical significance. Consequently, the findings should be presented as exploratory tendencies observed in this dataset rather than as general statements about Algerian men and women.

A further synthesis concerns the cultural and interactional profile of the responses. The corpus suggests that apology in the examined Algerian Arabic data tends to combine directness and relational repair. Direct formulae of remorse signal acknowledgement of the offence, while compensation and contextualisation help the speaker reduce the interpersonal tension caused by the violation. The mixed strategy is therefore not a marginal or accidental phenomenon; it reflects the tendency to intensify apology by accumulating several remedial components. This accumulation may be read as a pragmatic means of showing sincerity, mitigating face-threat and restoring social equilibrium.

Overall, the findings support the usefulness of Sugimoto's taxonomy while also showing the need to adapt existing models when they are applied to Algerian Arabic data. The introduction of the mixed strategy does not invalidate the original framework; rather, it supplements it by accounting for responses in which several categories occur together. The study thus contributes to the socio-pragmatic description of apology in Algerian Arabic and opens the way for larger-scale research that could compare elicited DCT data with naturally occurring conversations, examine the effect of offence severity and social distance, and test whether the tendencies observed here remain stable across wider regional, age and educational groups.

Conclusion

This study examined apology strategies used by Algerian male and female postgraduate students in four situations elicited through a Discourse Completion Test. It pursued two main objectives: first, to identify the apology strategies used by the participants; and second, to compare the broad tendencies observable in male and female responses. The analysis was guided by Sugimoto's (1997) distinction between primary, secondary and seldom-used strategies, while also taking into account the mixed strategy as an empirically observed pattern in the data.

The findings show that the participants used all three categories included in Sugimoto's framework. They also relied extensively on mixed strategies, which constituted the most frequent category in the corpus. This result is significant because it suggests that apology in the examined Algerian Arabic responses is often organised as a multi-component remedial act. Speakers do not only say that they are sorry; they may also explain the circumstances, offer compensation and attempt to restore the interpersonal relationship damaged by the offence.

With regard to gender, the study does not support a sharp opposition between male and female apology behaviour. Both groups used the same broad strategy types and shared a strong reliance on direct expressions of remorse. The differences observed in the corpus relate mainly to frequency: male participants used more mixed strategies, whereas female participants used more secondary strategies, especially compensation. These tendencies are important, but they must remain cautiously formulated because the sample is limited, non-random and exploratory.

The contribution of the study lies in its documentation of apology strategies in an Algerian Arabic context that remains under-represented in pragmatic research. It also shows the value of combining a recognised taxonomy with close attention to local linguistic and cultural forms. At the same time, the study has clear limitations. The number of participants is small, the data were elicited rather than naturally occurring, and the analysis would benefit from coding reliability procedures and from a wider range of variables, including social distance, relative power, offence severity, age, region and degree of familiarity between interlocutors.

Future research should therefore expand the corpus, include naturally occurring interactions, and compare Algerian Arabic apology practices with other Arabic varieties and with other sociolinguistic contexts. Such work would make it possible to determine whether the prominence of mixed strategies observed in this study is a stable feature of Algerian apology behaviour or a tendency linked to the particular situations, participants and elicitation method used here.