This is an open access article distributed under the terms of the Creative Commons
Attribution License (Offer and acceptance of political apologies can be a profound form of human interactions (Lazare, 2004). Political apology per se constitutes a powerful tool in re-examining traumatic events and historical sufferings (Skevi & Samaras, 2017). Worldwide history abounds with examples of political apologies (MacLachlan, 2010). In recent years, there is a consistently increasing number of political apologies following historical wrongdoings and injustices in a historical, national and political level (Celermajer, 2009; Gibney, 2008; Zoodsma et al., 2021). To this end, politicians seem to face past wrongdoings and injustices by making morally-based amends for such wrongs (Howard-Hassmann & Gibney, 2008). It is well acknowledged that attempts of re-establishing trust through recognizing truths and making moral amends are of utmost importance in an effort to ensure victims’ dignity and recognition of their self-worth. However, a still empirical question remains as to whether political apologies have such a potential to achieve establishment of trust, support and recognition and, if they do so, what are the most effective apology forms for this purpose (Gkinopoulos et al., 2022). The focus of this article is on two key forms of political apology, a self-focused apology and a self-other focused apology (Berndsen et al., 2015; Slocum, 2006; Slocum et al., 2011). Through an experimental study, this paper evidences the effectiveness of self-other focused apology in a real-life example of a traumatic event and vulnerable population. In the next section, I outline key components of political apologies framing the overall problematic around the aforementioned self and self-other continuum.
Understanding the Content and Meaning of Apologies
Broadly defined, an apology constitutes a speech act aiming to reconcile two or more parties (Bavelas, 2004; Celermajer, 2009; Tavuchis, 1991; Thompson, 2008). Speech is the cornerstone of an action of apology (Bavelas, 2004), that is an obligation on the part of offenders to say that they apologise. The question of what makes an apology effective, beyond its wording, prompted a number of studies (Alexander et al., 2006; Berry & Worthington, 2001; Brown, 2004; Corlett, 2006; Lazare, 2004; Risen & Gilovich, 2007; Steiner, 2000), which concluded that in order to be effective and perceived as sincere, apologies should: acknowledge that a wrong has been committed; acknowledge that the transgressor accepts responsibility for this wrong act; involve a transgressor, who offers to make a sort of atonement for the wrong act that has been committed; provide reassurances that the wrong act will not be repeated.
Interaction Between Offenders and Victims as Part of Political Apologies
It is a well-established empirical observation that when transgressors offer apologies for wrongdoings and injustices, then victims are more likely to forgive them and support their efforts to restore damages (e.g. Brown et al., 2008; Fehr & Gelfand, 2010; Leonard et al., 2011; McCullough et al., 1997). However, there is an interplay between various psychosocial principles, which determine or mediate the extent of forgiveness and support for offenders (see, for reviews, Exline et al., 2003; Fehr et al., 2010) and explain the different and, sometimes, pessimistic signals that research on apologies has provided (e.g. Philpot & Hornsey, 2008; Wohl et al., 2012; Wohl et al., 2015).
Empirical research has confirmed the importance of perceived sincerity of apologies for promoting forgiveness (Okimoto et al., 2013; Philpot & Hornsey, 2011; Wohl et al., 2012). Further studies (Berndsen et al., 2015; Okimoto et al., 2015) have found that perceived sincerity of apology mediates the effect of apology on intergroup forgiveness and, most importantly, the effect of victim-focused apology on intergroup trust and forgiveness. Wohl et al. (2013) have showed that people forgive and support offenders, when they perceive their apology as sincere and remorseful in a context of immediate interaction, where offenders and victims can be interlocutors. Dovidio et al. (2002) and Nadler and Liviatan (2006) provide also evidence that an apology that gives voice to victims and is perceived as highly sincere and remorseful, increases trust towards offenders and elicits positive emotions that, in turn, lead to positive intergroup relations. Allan (2007) posits that not just any apology offered has the ability to promote trust. Instead, true remorsefulness constitutes a key component of apologies. Empirical studies have shown that victim-focused apologies and reparative actions increase intergroup trust (Blatz & Philpot, 2010; Blatz & Ross, 2012; Hornsey & Wohl, 2013) and that the perception of an apology as deserved and sincere is positively correlated with perceptions of apologisers and the content of their apology as trustworthy (Berndsen et al., 2015; Wohl et al., 2013).
Acts of apology has a particular meaning only in interplay with the affected victims, serving their moral and psychological needs, acknowledging their suffering and reassuring them of a safer future (Lazare, 2004). Indeed, victim-focused institutional apologies satisfy the needs of victims (Shnabel et al., 2015) and restore harmonic life (Tavuchis, 1991). This social harmony is reflected on positive perceptions of emotional climate by victims when they are involved in reparatory activities (Páez, Espinosa, & Bobowik, 2013), reporting positive social well-being (Rimé et al., 2011), positive perceived emotional climate and intentions to support the apologisers’ policies (Shnabel & Nadler, 2008). Subsequent studies have, indeed, confirmed that victim-inclusive forms of apology gain more political support than victim-exclusive forms, because of better psychosocial healing and more highly perceived repair of the wrongdoing (Blatz & Philpot, 2010; Blatz et al., 2009; Kitagawa & Chu, 2021). Likewise, Berndsen et al. (2015) suggested that victim-focused apologies yield more positive outcomes in terms of support, forgiveness and perceived sincerity of offenders than offender-focused apologies. Sincere and trustworthy apologies should, thus, necessarily communicate a shared understanding of victims’ suffering.
Self-Focused and Self-Other Focused Apologies: The Staircase Model of Intergroup Apologies and the Authentic Apology Model
Two key conceptual models provide valuable insights on the distinction between offender and victim-focused apologies and inform the scope of the current study. To begin with, the Staircase Model of Intergroup Relations (SMIR) (Wohl et al., 2011) constitutes a framework of identifying the context where apologies are effective. SMIR requires setting straight the records of history, allowing shared interpretation of events on the part of offenders and victims. The Model also enables victims to have a voice, to become understood and to be heard. In such an immediate interactive context, SMIR provides opportunities for contact between offenders and victims and a discussion of reparations on the basis of a shared understanding of what victims need. Exemplifying an incremental reconciliatory process, SMIR is supposed to positively effect on support and forgiveness of offenders (Nunney & Manstead, 2021).
In a similar framework of victim-inclusive apologies, Slocum (2006) developed the Authentic Apology Model (AAM), a two-dimensional model of apologetic behaviour that precisely focuses on the content of apologies in relation to victim inclusion. According to one of the dimensions of this model that is of interest to this paper, apology operates along a continuum from a self-focus to a self-other focus apology. The self-focus side involves the wrongdoer’s admission of liability for the wrong, their regret and offer of restitution for the wrong. Emphasis is given more on offenders’ needs than victims’ needs (Allan, 2007). The self-other focus involves acknowledgement of the harm caused, expression of sorrow for harming victims, as well as attempts of repairing the harm by addressing victims’ emotional and other needs in a context of interaction between offenders and victims. Later research (e.g. Attwood, 2008; Slocum et al., 2011) confirmed the higher perceived sincerity, trust and forgiveness of offenders in the self-other apology than in self-focus apology. However, emphasis was still given on the wording of apologies (Berndsen et al., 2015; Gkinopoulos et al., 2022), that is the way that political apologies are rhetorically framed and the word content of those apologies rather than the setting they took place or the interlocutors involved in them.
Aims and Hypotheses
Despite the undoubted importance of immediate interaction between victims and offenders and the voice that victims must have in the process of damage restoration, it still remains an empirical question of ongoing interest as to which form of apology is considered as more effective in contexts of traumatic events. To date, much attention has been paid to the content and wording of apologies (Allan, 2007; Berndsen et al., 2015). However, studies tend to neglect dramaturgical elements of an apology, for instance in terms of physical interactions with victims and how they influence sincerity or reinforcing intergroup trust (Allan, 2007; Cels, 2015; Schmitt et al., 2004). Specifically, in the context of a performance-based approach in the natural setting of occurrence of wrongdoings, it is not yet clear which components of an apology (e.g. those that include victims vs those that exclude victims) impact on perceived sincerity and, thus, on trust, as well as which components elicit positive emotions yielding higher political support towards offenders.
The current study goes beyond the utterance of political apology itself and its wording, to consider the different foci of a political apology on self/offenders and others/victims in the aftermath of a real natural disaster. To date, literature and empirical research gives very little consideration to how the aforementioned forms occur in real-life cases of major transgressions. There is already some sporadic experimental evidence on victim-focused apologies as more sincere and trustworthy forms of apology, which are more likely to lead to forgiveness and support of offenders that offender-focused apologies (e.g. Berndsen et al., 2015; Philpot & Hornsey, 2008). This evidence, however, often lacks ecological validity as it either relies on theoretical assumptions that still need to be tested, or on simulation of forms of apologies in experimental settings with imaginative scenarios and convenience samples (see Gkinopoulos et al., 2022). As Berndsen et al. (2015) suggest, in order to reach to robust conclusions regarding the effectiveness of victim-focused political apologies, dramatic examples of real-life transgressions need to be explored.
The present study aims to build on existing findings and advance the relationship between victim-focused apology and political support towards the government, filling the gap of previous research in two ways: (a) by deploying actual non-verbal content of a real political apology, utilizing ecologically valid, rather than manipulated, material of images that capture two key important dimensions of a political apology; the inclusion or exclusion of victims; (b) by recruiting participants from the disaster zone of eastern Attica as victims, who suffered from the disaster. In doing so, the current study also responds to the need for inclusion of underrepresented and vulnerable populations in psychological research to increase its credibility (Vazire et al., 2022).
Based on the previous research showing the effect of victim-focused apology on sincerity, trust and positive emotional climate, two hypotheses were outlined: First, based on previous research showing that victim-focused apologies yield greater support towards the apologiser (Berndsen et al., 2015), are perceived as more sincere, promote forgiveness (Okimoto et al., 2015) and increase trust towards offenders (Nadler & Liviatan, 2006), it is expected that victim-focused apology will be perceived as more sincere, trustworthy, leading to greater perceptions of positive emotional climate and higher political support (Hypothesis 1). While a victim-focused apology increases the likelihood of support or forgiveness towards the apologiser, it is assumed that this link could have indirect paths via trust and perceptions of apologies as sincere and remorseful, along with positive emotions about apologizers’ future intentions (Nadler & Liviatan, 2006; Shnabel et al., 2015). To this end, in addition to testing main effects of apology condition on perceived sincerity, trust, positive emotional climate and political support towards the government, we expected that victim-focused apology would predict political support towards the government via high perceived sincerity, trust and positive emotional climate as three mediators in the aforementioned relationship (Hypothesis 2).
Concluding, in order to test the aforementioned two hypotheses, the consequences of two types of apology in a natural real-life context of a disaster will be studied. The next section will provide contextual information that will help understand the disastrous event per se, as well as the then socio-political situation.
Context of the Present Studies
On Monday 23rd July 2018 a catastrophic bushfire broke out in eastern Attica. Weather conditions on that day were conducive enough for the bushfire to spread and destroy numerous properties, cars and thousands acres. Mati, a small area of eastern Attica, was mostly affected by the disaster. In Mati, 103 people died and hundreds were severely injured and hospitalized. People died in their effort to escape the fires, because of the increased wind intensity that helped fan the flames at a rapid pace, while swam in order to avoid the smoke and flames and were drown. Boats were mobilised in a rescue operation as the fires took hold on Monday night. On Wednesday 25th July 2018 Alexis Tsipras, the then Greek Prime Minister, announced a 3-day national mourning in the name of the victims of that disaster. Under the pressure of the widespread disappointment about the management of the crisis, on Friday 27th July 2018 and after three days of national mourning, Alexis Tsipras offered an apologetic speech, taking the responsibility for this ‘tragedy’ (Souliotis, 2018). The pressure against Alexis Tsipras was intense and accusations were strong. Specifically, the Greek government and Alexis Tsipras were accused of a series of omissions and mistakes, such as the violation of the then institutional framework, incomplete planning and inadequate preparation of the General Secretariat for Civil Protection, incorrect estimations of the pace of wildfires ignoring the weather and winds forecasts, the National Early Warning System that was legislated, but never worked properly, and several weaknesses and malfunctions in the Fire and Police departments that caused a chaos.
In the beginning of national mourning, numerous members of the cabinet and the majority of the members of parliament of the opposition insisted on the urgent need of a public apology and taking responsibility, along with actions for restoration of damages and compensation for damaged properties. Some members of the parliament even strongly recommended an immediate visit on the disaster zone, in order for the victims to be heard. Alexis Tsipras issued a public apology on 27th July in the Greek parliament, but only members of the cabinet and journalists could join. On Monday 30th July, following his public apology on Friday 27th July, Alexis Tsipras visited the disaster zone as a continuation of his prior verbal apology statement to members of the cabinet and journalists. During the visit at the disaster zone, Alexis Tsipras directly spoke to victims, heard their concerns and reassured them in terms of the restorative actions that the government would plan to take (Souliotis, 2019). The disaster obviously affected the image of SYRIZA in the public opinion with people expressing disappointment and anger against the government and Alexis Tsipras and the way that he and his government dealt with the crisis. At the same time, people’s trust in politics, in general, was shaken (Souliotis, 2018) However, amidst people’s complaints that seldom do politicians take precaution measures to prevent such disasters, it seems that Tsipras’s apology and visit in the disaster zone eased victim’s anger and disappointment, as they felt that they were given a voice (Souliotis, 2018). In subsequent public opinion polls, there was a decrease of popularity towards, Alexis Tsipras and his party, but he continued to precede New Democracy in measures of voting intention. It is worth noting that the decrease was smaller after Tsipras’s visit in Mati (Telloglou, 2019), might prove how important are immediate interactions with victims in contexts of apologizing for wrongdoings.
Reflecting back on the literature, Alexis Tsipras’s apology seems to accurately capture the two key dimensions of political apologies: (a) a self-focused; Alexis Tsipras issued a statement on 27th July, addressing the members of the parliament and journalists. He mainly focused on his party’s needs and on explanations of the wrongful behavior to other members of the parliament and the journalists, verbalizing the overall distress that the disaster itself had caused (Allan, 2007); (b) a self-other focused; Alexis Tsipras apologized for the disaster and took the responsibility in a context of immediate interaction with victims in his visit at the disaster zone, on 30th July. He then admitted the harm done to the victims, validated the consequences for the victims and verbalized the sorrow because of the suffering experienced by the victims (Slocum et al., 2011).
Method
Pilot Study 1
The aim of the pilot Study 1 was twofold: First, to make sure that the perceived action of all images concerns an apology and not another action. Second, to make sure that images differ on the victim inclusion/exclusion dimension and not on any other characteristic. All newspapers circulating during that period were collected and analysed in terms of their visual content of the bushfire description. Newspapers were collected via the Athens News Agency, https://amna.gr, (in Greek: Αθηναϊκό Πρακτορείο Ειδήσεων), which is a major news agency located in Greece. The Agency hosts all newspapers, which are readable online too. This Agency was selected as the most representative, in terms of the news content, to collect the visual material of images from the catastrophe. Table 1 below shows the names of newspapers circulating during that period, along with their front-page titles.
Table 1
Newspaper Titles and Front-Page Headings
| Newspaper Titles | Front-Page Headings |
|---|---|
| The Daily | Fiery Disaster in Attica |
| The News | Dante’s Inferno |
| The Nation | Attica in Ashes |
| The Newspaper of Redactors | Fiery War |
| The Dawn | Difficult Times for Attica and Greece |
| Free Press | Bushfire-Killer |
| The Radical | Crime With the Politics That Sacrifices the Fire Prevention as Guilty |
| The Liberal | At the Mercy of the Bushfire |
| Anti-News | Bushfire as an Organized Plan of Arsonists |
| Democracy | Days of 2007 |
The relative frequency of each image used in the occasion of the bushfire, was calculated. Images with the highest relative frequency were included in the subsequent analyses. Images were collected from newspapers between 31st July and 1st August, following Tsipras’s public apology statement on the 27th and his visit at the disaster zone on the 30th July. Image 4 (see Appendix 1) was used as control image. The following images were those with the highest relative frequency.
Image 1
Image 1 captures Alexis Tsipras’s apologetic statement, surrounded by the members of the cabinet.
Note. The photography was taken by journalist Theodoros Tosounidis on 27th July 2018.
Image 2
In Image 2, Alexis Tsipras applauds firemen for their massive efforts to tackle the bushfire.
Note. This photography was taken by journalist Nikolaos Leandros on 30th July 2018.
Image 3
Finally, Image 3 shows Alexis Tsipras in immediate talks with people affected by the bushfire.
Note. The photography was taken by journalist Nikolaos Lagios on 30th July 2018.
Once images with the highest relative frequency were identified, the research proceeded in the next phase. 90 participants, 44 (39.6%) females and 46 (41.4%) males, with ages ranging from 19 to 61 years old (M = 30.9, SD = 10.8) took part in pilot Study 1. All participants were residents in Municipality of Marathonas, the area that was affected by the bushfire. Recruitment was completed through the local municipalities and the support of the Athens branch of the Red Cross. Recruitment was launched on 5th August 2018, five days after the Greek Prime Minister’s visit of the disaster zone of Mati on 29th July 2018. Participants were first asked to provide demographic information (i.e. age, gender and area of residence). Participants were presented with the four images, the three images above and the control image. Participants were asked to write what each image described and then in which dimension they thought that images differ. Participants indicated that Images 1, 2 and 3 described the Prime Minister’s political apology (N = 70, N = 68, N = 75 responses, respectively). For the control image 33 participants responded that this image was related to the news broadly. Regarding the dimension on which images differ, most participants (N = 71) indicated inclusion of real victims of wildfires as the dimension that images differ between each other. For detailed presentation of responses in questions of this pilot study and frequencies, see Supplementary Materials. Having ensured that images describe an apology and that the key difference between images lies on the inclusion of victims in interaction with the Prime Minister, the second pilot study will test the typicality of the apology action of the images.
Pilot Study 2
This pilot study aimed to test the extent of typicality of an apology action for the collected images. 90 participants, 47 (42.3%) females and 43 (38.7%) males, with ages ranging from 19 to 65 years old (M = 30.04, SD = 10.7) took part in pilot Study 2. All participants were residents in Municipality of Marathonas, the area that was affected by wildfires. Recruitment was completed through the local municipalities and the support of the Athens branch of the Red Cross. Recruitment was launched on 13th August 2018, fifteen days after the Greek Prime Minister’s visit of the disaster zone of Mati on 29th July 2018. Participants were first asked to provide demographic information (i.e. age, gender and area of residence). Through an online survey participants were instructed to pay attention to each of the three images and then for each image separately to assess its typicality as an apology image, responding to the 5-point Likert scale question: (“To what extent do you think that this image is typical of the Prime Minister Alexis Tsipras’s apology and responsibility for the July 2018 wildfires in Attica?”), where 1 = not at all and 5 = very much. In order to test beliefs of typicality of apology action of images, an One Sample t-test was conducted. As Table 2 shows, Images 1 and 3 were the most typical of a political apology action, while Images 2 and 4 were not significantly typical of a political apology action.
Table 2
Levels of Typicality of Apology for Each of the Three Images
| Images | M | SD | One Sample t-test |
|---|---|---|---|
| Image 1 | 4.73 | .52 | t(29) = 18.2, p = .001 |
| Image 2 | 2.70 | .98 | t(29) = 1.66, p = .107 |
| Image 3 | 4.50 | .62 | t(29) = 13.04, p = .001 |
| Image 4 | 2.92 | .73 | t(29) = 1.01, p = .329 |
To sum up, the two pilot studies were conducted to ensure the robustness of experimental material, verify its interpretations, ensuring that images were, indeed, perceived as representative and typical of a political apology action. Next, the experimental study design and results will be presented.
Experimental Study
Participants and Sampling Strategy
222 participants, 128 (57.7%) females and 94 (42.3%) males, with ages ranging from 18 to 72 years old (M = 37.65, SD = 11.82) took part in the main study. All participants were residents in Municipality of Marathonas, the area that was affected by wildfires. Recruitment was completed through the local municipalities and the support of the Athens branch of the Red Cross. Data collection lasted about a month, between 19th August – 29th September 2018. Data was collected in areas of east Attica in Municipality of Marathonas– close to or within the disaster zone. A power analysis was conducted using G*Power. A small effect of f2 = .10 was assumed, with α = .05 and 95% of power, which resulted in a sample size of at least 188 participants.
Procedure
The Qualtrics online survey that was developed, was first created in English and then was translated to Greek by the first author and a research assistant at the University of Surrey, who are both bilingual. Prior to data collection, ethical approval was obtained from the ethics committee of the University of Surrey. Participants provided written informed consent in advance of taking part in the study.
Participants were first asked to provide demographic information (i.e. age, gender and area of residence. Then, three conditions were designed to compose the main study. In Condition A participants were presented with Image 1 (victim-exclusive apology). In Condition B participants were presented with Image 3 (victim-inclusive apology). In control condition participants were simply presented with an image of the building of the Greek parliament.
Participants in Condition A were given the following instruction: (“The image below illustrates an aspect of the Greek Prime Minister Alexis Tsipras’s apology and taking responsibility for the wildfires in Attica on July 2018. Alexis Tsipras takes the responsibility for the catastrophe vis-à-vis the ministers of the cabinet, who surround him. Please, pay attention to the image and then respond to the questions that follow”). Participants in Condition B were given the following instruction: (“The image below illustrates an aspect of the Greek Prime Minister Alexis Tsipras’s apology and taking responsibility for the wildfires in Attica on July 2018. Alexis Tsipras takes the responsibility for the catastrophe vis-à-vis the victims who were affected by wildfires and surround him. Please, pay attention to the image and then respond to the questions that follow”). Then, participants were asked to respond to a number of dependent measures.
Dependent Measures
Unless otherwise specified, participants responded to a 5-point scale (1 = completely disagree up to 5 = completely agree). Full list of items per scale is presented in Appendix 2.
Perceived Sincerity
Four items about the perceived sincerity of the prime minister’s apology were used in line with the four criteria that Hatcher (2011) uses to define an effective sincere apology, matched with the affirmation, affect and action components of Slocum’s (2006) model (see, for a review, Okimoto et al., 2015). Example item was: The prime minister ‘Acknowledges that what he did was wrong’; ‘Accepts responsibility for his action’; ‘Makes attempts to make up, or atone, for the wrong committed’; ‘Gives assurances that it would not happen again’ (α = .94).
Perceived Trust
Kim et al. (2004) have defined trust as that psychological state, which comprises “the intention to accept vulnerability based on positive expectations of the intentions or behaviour of another” (p. 105). In line with this definition, three items were used, with the following item as example: Do you believe that the prime minister and his government ‘Is reassuring that policies will be implemented to avoid wrongdoings in the future’ (α = .98).
Perceived Positive Emotional Climate
Four items were used, adjusted from the Climate Dimension Scale CD 24 (de Rivera & Páez, 2007), asking for participants to rate their perceptions of the socio-emotional climate in Greece at the time following the apology. Example item was: ‘People in the country feel secured that there is enough food, water, medicine, and shelter for themselves and their families, and that they will continue having these goods’ (α = .93).
Political Support Towards the Government
Four items, developed by Pirralha and Weber (2014), were used asking for participants to rate their satisfaction with, evaluation and support of the government’s performance in public safety and environmental protection domains. These two domains were the most frequently discussed during the occurrence of the natural disaster in east Attica and governmental policies were under question regarding their effectiveness in these domains. Participants were asked on a 7-point scale (1 = extremely dissatisfied up to 7 = extremely satisfied) to rate how satisfied they are, for example, ‘with the way that Greece’s government is doing its job in public policy domain’. Furthermore, participants were asked to rate on a 7-point scale (1= extremely bad up to 7 = extremely good) what they think overall, for example, ‘about the state of public safety in Greece nowadays’ (α = .97).
Demographic Information
Participants were asked to provide demographic information regarding their age, gender and political orientation. In all subsequent analyses demographics were used as covariates.
Manipulation Check
Two items were used to check the manipulation of apology focus (victim-exclusive and or victim-inclusive). Using a 4-point scale (1 = not at all up to 4 = to a great extent), participants indicated a) to what extent the image focused on the inclusion of victims affected by the wildfires and b) to what extent the image focused on the interaction with victims affected by the wildfires (α = .98).
Results
Manipulation Check
Analysis of Variance (ANOVA) on perceived victim inclusion focus of the images revealed a significant effect of apology condition, F(2, 219) = 908.3, p < .001, = .16. Simple effects revealed that perceptions of victim inclusion were significantly higher in victim inclusion condition (M = 3.99, SD = .22, p < .001) than in victim exclusion condition (M = 1.95, SD = .58, p < .001). No significant differences were found between victim-exclusive condition and control condition (M = 1.52, SD = .32, p = .17).
Effect of Apology Condition on Perceived Sincerity, Trust, Positive Emotional Climate and Political Support Toward the Government
To test Hypothesis 1 (higher levels of perceived sincerity, trust, positive emotional climate and political support are expected in the victim-inclusive condition, than in victim-exclusive condition), a One-Way ANCOVA was conducted to detect the effect of apology condition on perceived sincerity, trust, positive emotional climate and political support. In order to control for the effects of age, gender and political orientation, we run the analysis with and without age, gender and political orientation as covariates. Adding age, gender and political orientation as covariates did not change the pattern of results. For the adjusted means and standard deviations after including covariates, see Supplementary Materials. Results confirmed the hypothesis and are summarized in Table 3 below.
Table 3
Means, Standard Deviations, and One-Way Analyses of Variance in Perceived Sincerity, Trust, Positive Emotional Climate and Political Support Toward the Government per Apology Condition
| Measure | Victim Exclusive
|
Victim Inclusive
|
Control Group
|
F (2, 221) | ||||
|---|---|---|---|---|---|---|---|---|
| M | SD | M | SD | M | SD | |||
| PS | 2.34 | .98 | 4.26 | .82 | 2.19 | .61 | 145.52*** | .57 |
| PT | 3.08 | 1.20 | 4.16 | .87 | 3.97 | .23 | 32.39*** | .22 |
| PPEC | 2.46 | .58 | 3.29 | .91 | 2.72 | .35 | 30.60*** | .21 |
| PoS | 3.71 | .83 | 4.06 | .87 | 3.17 | .53 | 11.59*** | .09 |
Note. N = 222. PS = Perceived Sincerity; PT = Perceived Trust; PPEC = Perceived Positive Emotional Climate; PoS = Political Support.
As Table 3 shows, there was a significant main effect of apology condition across all the dependent measures, confirming the prediction of Hypothesis 1. Pairwise comparisons revealed significant differences between (a) victim-inclusive and victim-exclusive condition; (b) victim-inclusive and control condition, with participants scoring higher in victim-inclusive condition in all the dependent measures. No significant differences were found in comparisons between victim-exclusive and control condition.
Association Between Apology Condition and Political Support Toward the Government via Perceived Sincerity, Trust and Positive Emotional Climate
To explore Hypothesis 2 regarding the hypothesised indirect effect of apology condition on political support of governmental policies via perceived sincerity, trust and positive emotional climate, a bootstrapped parallel mediation analysis (5,000 resamples) was conducted using the PROCESS macro for SPSS, Model 4 (Hayes, 2013) including three mediators with 5,000 bootstrapped resamples and a 95% confidence interval with indicator coding. The mediation analysis was conducted to examine whether the effect of victim-inclusive apology on political support towards the government could be explained by increased perceived sincerity, trust and positive emotional climate. The victim-inclusive apology condition was coded as the reference group and was compared to the victim-exclusive apology condition (D1) and control (D2) condition separately. Results are presented in Table 4 below. For more information about descriptive statistics and zero-order correlations, see Supplementary Materials.
Table 4
Results of Mediation Analysis
| Variables | β | SE | t | p |
|---|---|---|---|---|
| IV to Mediator | ||||
| Apology D1 → PS | 1.91 | .13 | 14.16 | < .001 |
| Apology D2 → PS | -.15 | .13 | -1.14 | .25 |
| Apology D1 → PT | 1.88 | .14 | 7.54 | < .001 |
| Apology D2 → PT | .88 | .14 | 6.19 | < .001 |
| Apology D1 → PPEC | .83 | .10 | 7.66 | < .001 |
| Apology D2 → PPEC | .26 | .10 | 2.46 | .01 |
| Mediator to DV | ||||
| PS | .02 | .06 | .41 | .68 |
| PT | .32 | .06 | 4.95 | < .001 |
| PPEC | .34 | .09 | 3.77 | < .01 |
| Boot Indirect Effect | Boot SE | CI Lower End | CI Upper End | |
| Indirect Effect of IV on DV through proposed mediator | ||||
| Total DI | -.40 | .13 | .12 | .67 |
| Total D2 | .66 | .13 | .39 | .93 |
| Apology D1 → PS → PoS | .05 | .11 | -.16 | .28 |
| Apology D2 → PS → PoS | -.04 | .01 | -.03 | .02 |
| Apology D1 → PT → PoS | .34 | .08 | .19 | .52 |
| Apology D2 → PT → PoS | .28 | .07 | .15 | .44 |
| Apology D1 → PPEC → PoS | .29 | .09 | .12 | .48 |
| Apology D2 → PPEC → PoS | .10 | .03 | .02 | .17 |
Note. N = 222. IV = independent variable; DV = dependent variable; CI = confidence interval. Unstandardized regression coefficients are reported. Boot- strap sample size = 5,000. PS = Perceived Sincerity; PT = Perceived Trust; PPEC = Perceived Positive Emotional Climate; PoS = Political Support Towards Governmental Policies.
Results confirmed that the indirect effect was significant in both cases. In particular, when compared to both the victim-exclusive condition and the control condition, victim inclusion significantly increased perceived trust of the apologizer and perceived positive emotional climate. Regarding perceived sincerity, when compared victim-inclusion condition with control condition, no statistical significant indirect effect was found for perceived sincerity. However, perceived sincerity was significantly increased when comparing victim-inclusion with victim-exclusion condition. Then, victim-inclusion significantly significantly increased perceived sincerity.
General Discussion, Limitations and Future Directions
The aim of this paper was to investigate the action of apology, its self/other focus and equivalent effects on political support towards the government via perceived sincerity, trust and positive emotional climate. Findings confirmed the two hypotheses. First, victim-inclusive apology was more highly perceived as sincere and trustworthy, eliciting more positive emotions and political support towards the government than the victim-exclusive apology (see, for example, Berndsen et al., 2015; Smith, 2008; Wohl et al., 2012). Furthermore, our mediation hypothesis that apology condition effects political support towards the government via perceived sincerity (Slocum et al., 2011), trust and positive emotional climate (e. g. Nadler & Liviatan, 2006; Shnabel et al., 2015) was also confirmed. Consistent with hypotheses and predictions, victim-focused apology heightened support towards governmental policies via increased perceived sincerity, trust and positive emotional climate in case of immediate interactions between the Prime Minister and victims of Mati. These findings were the first to evidence that a political apology that communicates the victims’ feelings, sufferings and concerns in a real-life context of interaction and dramatic transgressions, can foster support towards the government via perceptions of such apologetic form as highly sincere, trustworthy and eliciting positive emotions.
Regardless, it would be useful to emphasize again that winning political support does not constitute the raison d’être of a political apology and is not necessarily a prerequisite for the effectiveness of an apology (Wohl et al., 2011). Ingredients such as sincerity of an apology, trust or positive feelings elicited through the apologetic action provide important paths themselves too. The experimental study expanded previous findings (e.g. Berndsen et al., 2015; Shnabel et al., 2015; Smith, 2008; Wohl et al., 2012) on bringing the self-other aspect of apologies to the foreground and exploring further which aspects of apology are those that increase perceived sincerity, trust and positive emotions, as well as which aspects of an apology might directly affect political support towards the government in real-life settings of a dramatic transgression that caused the death of 103 people. To the best of our knowledge, this study is the first attempt to analyse apologetic actions and to empirically test the effectiveness of specific components of such actions in real-life settings. This novelty reflects the dramaturgical approach of political apologies (Cels, 2015), which emphasizes the need for considering the real interactive context in evaluating the effectiveness of public apologies, the roles of interlocutors and the contextual factors that frame the public sense-making/reactions in the aftermath of major transgressions. It is, finally, worth pointing to the fact that real-life victims were reached as respondents, a population that is hardly targeted so far in social psychological research, as well as the research on political or intergroup apologies (cf. Bobowik et al., 2017). Recruiting this sample, the current study responded to calls for the need to further explore the evaluation and utility of victim-focused apologies in contexts that exemplify dramatic cases of transgression (Berndsen et al., 2015; Hornsey & Wohl, 2013). Previous experimental work (e.g. Berndsen et al., 2015; Leonard et al., 2011) used examples of real rivalries limited to relations between two universities or similar examples of rivalries perceived more as friendly and less as adversarial (Berndsen et al., 2015; Hornsey & Wohl, 2013; Wohl et al., 2011).
Considering immediate interactions between offenders and victims in the aftermath of dramatic transgressions can help explain how a political apology may meet emotional and psychological needs of victims. In particular, comparing offender-focused and victim-focused forms of political apologies enhances our understanding of how meaningful political practices contribute to restoration of damage and trust in a context of an effective crisis management. Emphasizing on a victim-focused apology, the equal and active role of both victims and apologisers becomes prominent, directing our attention to extra-linguistic ingredients of a political apology, in a real-life setting, where interactions unfold. Policy-makers may benefit from such empirical evidence, on the grounds that transgressions occur worldwide increasing, in turn, the need for effective and convincing modes of political communication, restoration of trust and reassurance of reparative actions. It is worth acknowledging that we live ‘in the age of apologies’ (Gkinopoulos et al., 2022) and the need for robust experimental evidence of perceptions and evaluations of political apology forms in real-life transgressions is more urgent and informative than ever. Such evidence helps social psychologists and other political scientists understand underlying mechanisms that determine the effectiveness of one or another political apology form.
Beyond the contribution of this paper, we should not fail to acknowledge some limitations of the current study, which can be addressed by future researchers. First, the current study focused on a particular population of the disaster zone. In order to establish robustness of the design that used visual material from a real-life transgression capturing different forms of a political apology, it would be useful to detect similar effects in general populations and longitudinally. Second, we know that it is more usual that the leader of the offending group or one of its representatives offers an apology (Wohl et al., 2011). This explains the emphasis of the current study on capturing different forms of Alexis Tsipras’s political apology. However, since 2018 political leaders every year remember the disaster and issue brief statements. To this end, a fruitful avenue for future research would be to expand beyond the offender’s apology by empirically demonstrating effectiveness of political apology forms in cases that, for example, a political representative of an opposing party apologises or interacts with victims. Finally, an integrative methodological framework may also provide fruitful insights. Cels (2015) suggests a dramaturgical approach to political apologies deploying visual material, capturing different roles of interlocutors in different interactive contexts. Such an approach can complement political apology research and experimental designs by showcasing how political apologies are constructed in visual political discourse.