Introduction

Forgiveness represents a human virtue that reflects lack of resentment and involves benevolence, compassion and love towards a transgressor (Thompson et al., 2005; Worthington, 2001). The researchers have reported forgiveness to be facilitative for the development and restoration of self and interpersonal relationships (McCullough, 2008), mental health and well-being (Griffin et al., 2020; J. R. Webb & Toussaint, 2020; E. L. Worthington & Scherer, 2004) and physical health (Lawler et al., 2003; Toussaint et al., 2020). It has also been found to be an effective intervention method for interpersonal and emotional problems and promotes well-being (Baskin & Enright, 2004). The impacts and benefits of forgiveness are extended from intrapersonal, interpersonal and collective to organizational levels (Aquino et al., 2006; Noor et al., 2008). Many antecedents and correlates have been identified to unearth the nature and dynamics of forgiveness. Many intrapersonal and interpersonal factors such as perceived severity of transgression (Fincham et al., 2005); absence of apology (McCullough et al., 1998), the nature of relationship (McNulty, 2011), emotions (McCullough et al., 1998), cognitions (McCullough et al., 2007) and personality attributes (Berry et al., 2005) have been found to be closely linked with forgiveness.
            There are various forms of forgiveness that involve oneself, others, deity, society, and situations (Toussaint & Webb, 2005). Although a good number of studies have been conducted on interpersonal forgiveness, there has been a dearth of scientific studies on self-forgiveness. With its effectiveness in clinical interventions and the development of psychometric measures, self-forgiveness has recently attracted the attention of the researchers (Scherer et al., 2011; Wohl et al., 2008). Self-forgiveness refers to the desire to give up anger and grudges directed towards self as a result of perceived wrongdoings of one’s own on the one hand and to cultivate compassion, self-kindness and care toward oneself on the other (Enright, 1996).  Thus, self-forgiveness initiates a set of motivational changes that inbreed preventive and promotive strengths enabling a person to enhance positive cognition, behaviour and affect for self and affected others (Hall & Fincham, 2005).
            Based on the interpersonal model of forgiveness, Hall and Fincham (2005) have proposed the first model that suggested self-forgiveness to originate from shame and guilt that are caused by attributions and the severity of the transgressions. They further argued that guilt and conciliatory behaviours toward a victim and perceived forgiveness may facilitate or inhibit self-forgiveness. Empirical validation of the model showed that attributions, empathy, and shame were unrelated to self-forgiveness, and only transgression severity, guilt, conciliatory behaviours, and perceived forgiveness influenced the development of self-forgiveness beyond the effect of time (Hall & Fincham, 2008).
            Rangganadhan and Todorov (2010) reported Hall and Fincham’s model of self-forgiveness(Hall & Fincham, 2005) inadequate and put forth their own model in which guilt positively affects self-forgiveness via prompting both other-oriented empathy and conciliatory behaviour, while shame negatively affects self-forgiveness, both individually and by increasing personal distress empathy, or the experience of self-oriented distress at the recognition of another’s discomfort. McConnell et al. (2012) tested the original model proposed by Hall and Fincham for the third time and observed that the alternative model, which included severity of the offence, guilt, conciliatory behaviour, and perceived forgiveness from the victim, best accounted for the variance in self-forgiveness. McGaffin et al. (2013) integrated the findings of all the three previous models of self-forgiveness and suggested that guilt and shame lead to two distinct consequences. For example, guilt promotes other-oriented empathy, conciliatory behaviours and acceptance while shame may lead to acceptance and personal distress empathy. It provided for acceptance as a predictor of self-forgiveness and distinguished between other-oriented empathy and personal distress empathy.
            All these models of self-forgiveness have much in common and they all provide for the importance of transgression severity, shame, guilt, conciliatory behaviours, and perceived forgiveness from the victims. Most recently, a model of self-forgiveness has been presented by Mudgal and Tiwari (2017) who employed exploratory research design, a variant of mixed methods design, on Indian population and came up with three components of self-forgiveness namely, realization & reparation, guilt and attribution. The uniqueness of this model lies in the empirical validation of these components of self-forgiveness which suggested it to be significant predictors of general health and quality of life of the adults (Mudgal & Tiwari, 2017b). Self-forgiveness has been reported to be positively related to the positive health outcomes and decreased depression and anxiety (Maltby et al., 2001), and self-esteem, well-being, close-relationships, self-activity and social engagement in female victims (Coates, 1996), less suicidal behaviour(Hirsch et al., 2011), physical and mental health (Jon R. Webb et al., 2013), greater correctional involvement (Biron, 2007), lower levels of neuroticism and greater agreeableness (Biron, 2007; Butzen, 2009).

Self-Esteem and Self-Forgiveness

            It has been found that the individuals who experience an attack on their self-esteem are less able to forgive. It is the self-esteem that ascertains the level of threat due to any transgressions, it may be argued to be important for the study of forgiveness (Strelan & Zdaniuk, 2015). Researchers have also suggested that there is a need to test the process driving the relationship between self-esteem and self-forgiveness (Strelan & Zdaniuk, 2015).
The lower level of self-esteem encourages people to enter self-protection and develop avoidance that may affect self-forgiveness negatively. It has been reported that excess self-protection caused by poor self-esteem may decrease the accessibility of psychological resources needed for forgiveness to come out (Strelan & Zdaniuk, 2015). The severity of transgression has been reported to impede forgiveness in many studies (Fincham et al., 2005; Griffin et al., 2020) that necessitates the assessment of the roles of the positive self resources in regulating forgiveness (Strelan & Zdaniuk, 2015).
            A perusal of previous studies evinced that little is known about the extent to which an individual’s self-esteem affects forgiveness (Strelan & Zdaniuk, 2015). Some studies have been conducted employing trait self-esteem with mixed findings. A meta-analytic review showed that there was a very weak relationship between trait self-esteem and situation-specific forgiveness (Fehr et al., 2010). It may be argued that the transgression situation in which the individuals commit wrongdoing may also affect their self-esteem and may later impact their forgiveness. Some researchers have suggested that people use self-esteem as a resource (Steele et al., 1993) that acts as a standard or expectancy that is guided by their values and situational cues. It has been argued that although people try to maintain specific positive self-images, the primary motivation behind this image is to maintain global self-integrity, a general perception of their goodness, virtue, and efficacy. Thus, having high self-esteem facilitates individuals to affirm their resources and lesser use of defensive mechanisms or distort reality. Through their self-worth, people can have positive attitudes towards self and others. Moreover, individuals with high self-esteem are also more likely to acknowledge their own personal responsibility in the face of adversity, pain and wrongdoings. It has been argued that self-esteem originates from social comparison and denotes one’s ranking. It reflects one’s sense of self-worth and evaluations(Harter, 1999). It has also been argued that having high self-esteem is not always beneficial for individuals. Some ill-consequences of having higher self-esteem have also been identified by some researchers. For example, having higher self-esteem may involve puff the self up in putting others down, lead to a grandiose view of oneself and self-esteem may result from narcissism(Neff, 2011). Thus, high self-esteem originating from narcissism may lead to negative outcomes (anger, retaliation) while healthy self-esteem may culminate in positive ones (MacKinnon, 2015).

Self-Compassion and Self-Forgiveness

            Borrowed from Buddhist philosophy, self-compassion entered recently in the literature of Psychology. Self-compassion refers to a set of positive self-attitudes and behaviours that enhances the positive relationship with the self and self-worth that originates in the face of adversity, pain and failures. Self-compassion has been suggested to comprise of three interdependent and interrelated bipolar dimensions: Self-kindness vs. self-judgement, common humanity vs. Isolation and mindfulness vs. Over-identification. The first dimension is characterized by self-care and positive attitudes in adversity while the second facilitates the understanding that pain or failure is common experiences to all. The third aspect represents to become aware of one’s negative experiences and refrain from denying it (Neff, 2003a, 2003b). In essence, self-compassion denotes self-attitude, feelings and behaviours in the face of adversity, pain or failure of life. It is one of the significant self-resources that has attracted the attention of researchers in recent past and has been reported to be closely linked with positive health outcomes, academic success, well-being and performance (Rai & Tiwari, 2019; Verma & Tiwari, 2017).
             It has been reported that self-compassion and self-forgiveness are closely linked but their nature of association has not been understood properly (Conway-Williams, 2015). Self-forgiveness and self-compassion are similar in their definitions, components, processes and contributions to various life outcomes (Conway-Williams, 2015). An analysis of the definitions of these constructs showed that fostering compassion, generosity, humanity and love toward oneself involve in the constructs of self-forgiveness and self-compassion (Enright, 1996; Worthington et al., 2005). These two also have similarities in regulating a broad range of psychological functioning and are assumed to carry adaptive role in self-regulation, pro-social and positive virtues (Berry et al., 2005), health functioning (Christensen et al., 2011) and well-being (Pandey et al., 2019, 2020).
            These arguments evinced that self-compassion and self-forgiveness are linked closely but the empirical validation of their association is lacking (Conway-Williams, 2015). The search on PsycINFO in 2014 identified only three studies that reported their relationships with limited aspects of human behaviours (Conway-Williams, 2015). (Neff and Pommier (2013)examined the relationship between self-compassion and forgiveness of others in undergraduate students, older adults, and active mediators and reported it to be correlated with forgiveness of others in all three samples. There is a clear theoretical rationale for a direct relationship between self-compassion and self-forgiveness (Conway-Williams, 2015). Exploring the nature of the relationship between self-compassion and self-forgiveness may be important for the integration of these two areas of research as well as for furthering understanding of each of these constructs independently. It may be argued that self-compassion may impact one’s inclination to forgive oneself. Moreover, self-compassion is more frequent and relevant variable and comparatively easy to experience than self-forgiveness. In addition, cognitive, emotional, and experiential responses towards committing an offence may be linked with self-compassion that, in turn, may affect the likelihood of self-forgiveness. Self-compassion may lead one to experience less rumination, shame, and experiential avoidance that may facilitate self-forgiveness.
            Self-compassion and self-forgiveness occur at the time when an individual is facing adversity, pain and guilt. Both the processes are closely associated with self-knowledge and self-attitude. Both the processes have positive impacts on the individuals. Thus, they must be linked with each other. In spite of the pivotal role of these two psychological constructs, there is a dearth of studies relating these two constructs. Self-compassion has been suggested to be positively related to well-being and negatively linked with psychopathological symptoms (MacBeth & Gumley, 2012; Muris et al., 2018). Very recently, some researchers have shown their disagreements with the empirical validity of components of self-compassion model proposed by Neff (2003a) and questioned the suitability and usefulness of incorporating self-judgement, isolation and over-identification to her model (Muris et al., 2016). Muris and Petrocchi (2017) argued that self-kindness, common humanity and mindfulness (positive aspects of self-compassion) exhibit negative associations with mental health problems whereas self-judgement, isolation and over-identification (negative dimensions of self-compassion) show with psychopathological symptoms. Similar observations were also made by some other researchers who reported comparable findings regarding positive and negative self-compassion (López et al., 2015; Pfattheicher et al., 2017).
            Neff and associated researchers of self-compassion conducted the further statistical analysis and came to their previous conclusions that bifactor model reflects the most suitable model of self-compassion (Cleare et al., 2018; Neff et al., 2017). These efforts did not satisfy the opponents who questioned the inclusion of positive and negative dimensions in the proposed model of self-compassion. The opponents argued that self-judgment, isolation and over-identification represent negative attributes and carry a close connection with self-criticism (Zuroff et al., 1990), social withdrawal and loneliness (Rubin & Coplan, 2004), and self-focused rumination (Lyubomirsky & Nolen-Hoeksema, 1995). Following this line of arguments, Muris et al. (2018) have recently testified the face and empirical validity of positive and negative dimensions of self-compassion proposed by (Neff, 2003b) and argued positive self-compassion to have a close association with adaptive coping and healthy functioning and negative self-compassion to be linked with maladaptive behaviours (Muris et al., 2018). Similar to the findings of Muris et al. (2018), other researchers have also observed a positive correlation between positive self-compassion and social support seeking, and reassuring thoughts and negative correlations with negative self-compassion (Allen & Leary, 2010). Besides, negative self-compassion also showed its close association with passive reacting, avoidance and emotion suppression (Allen & Leary, 2010). Similarly, positive self-compassion showed its positive association with adaptive coping strategies such as active coping, planning and positive reframing and negative relationships with maladaptive strategies such as behavioural disengagement and self-blame (Sirois et al., 2015).
            Examining the association between self-compassion and self-forgiveness within more diverse populations has also been suggested (Conway-Williams, 2015). It is also needed to investigate the role of self-compassion in the self-forgiveness process (Wenzel et al., 2012). Majority of the study has employed three components model of self-compassion while ignoring the positive and negative dimensions which have very recently attracted the researchers. The findings suggest that conceptualization of self-compassion in terms of positive and negative dimensions are more efficacious to underscore the role of self-compassion in structuring the various life outcomes (Muris et al., 2018; Pandey et al., 2019). In the backdrop of these arguments, the present study aimed to explicate the mediating role of positive and negative dimensions of self-compassion between self-esteem and self-forgiveness with its three components, namely realization & reparation, guilt and attribution proposed by Mudgal and Tiwari (2017).

Objectives

          (1)     To understand the nature and magnitude of correlations among Self-Esteem, Positive and Negative Self-Compassion and various components of Self-Forgiveness,
          (2)     To estimate the variance accounted for by Self-Esteem, and Positive and Negative Self-Compassion in the scores of various components of Self-Forgiveness, and
          (3)     To explore the mediating role of Positive and Negative Self-Compassion between the relationship of Self-Esteem and the three components of Self-Forgiveness (Realization & Reparation, Guilt and Attribution).

Hypotheses

            On the basis of the understanding of the available literature, the following hypotheses have been formulated:
          (1)     There will be positive correlations among Self-Esteem, Positive Self-Compassion, Realization & Reparation, Attribution and Overall Self-Forgiveness while Negative correlation will be observed among the predictors and Guilt of the participants.
          (2)     There will be negative correlations among Negative Self-Compassion, Realization & Reparation, Attribution and Overall Self-Forgiveness whereas a positive correlation will be recorded between Negative Self-Compassion and Guilt of the participants.
          (3)     The Self-Esteem and Positive Self-Compassion will account for positive variability in Realization & Reparation, Attribution and Overall Self-Forgiveness while they will negatively contribute to the Guilt of the participants. On the hand, the Negative Self-Compassion will account for negative variability for Realization & Reparation, Attribution and Overall Self-Forgiveness and strengthening variance for Guilt of the participants.
          (4)     Positive and Negative Self-Compassion will mediate differently the relationship between Self-Esteem and Realization & Reparation, Guilt and Attribution.

Methods and Procedure

Participants

            The present study employed a convenient sampling method to choose the participants of the study. Initially, 351 participants pursuing their graduation and post-graduation programmes at schools of Arts, Commerce, Science and Law at Doctor Harisingh Gour Vishwavidyalaya (University), Sagar, Madhya Pradesh, India gave their consent with to participate. Out of these, only 300 participants with the age ranging from 18 years to 30 years (M= 20.04, SD = 1.77) ascertained their final participation.  The scores of these participants were screened for outliers through SPSS V25 as per the procedure outlined by Donald (2016). After the screening of the outliers, only data of 144 males (Age range = 18 years to 25 years, M = 22.10, SD = 1.66) and 124 females (Age range = 19 years to 30 years, M = 21.98, SD = 1.90) were left for final analyses.

Psychometric Tools

The following psychometric tools were employed to collect the data of the study:

Self-Esteem Scale

            Rosenberg Self-Esteem Scale is a unidimensional scale with 10-items that measures global self-worth by measuring both positive and negative feelings about the self (Rosenberg, 1965). The items seek an answer on a 4-point Likert scale format ranging from strongly agree to strongly disagree. It has been reported that people high in self-esteem show high attractiveness, better relationships and make better impressions on others. High self-esteem has also been shown to have a strong relation with happiness and lowered depression (Baumeister et al., 2003; Pandey et al., 2020).

Self-Compassion Scale

            Self-compassion of the participants was measured with the help of the Self-Compassion Scale (Neff, 2003a). The items of the self-compassion measure carry the thoughts, emotions and behaviours representing the various components of self-compassion. The scale consists of 26 items that comprises Self-Kindness (items-5, 12, 19, 23, 26), Self-Judgment (items-1, 8, 11, 16, 21), Common Humanity (items-3, 7, 10, 15), Isolation (items-4, 13, 18, 25), Mindfulness (items-9, 14, 17, 22) and Over-Identification (items-2, 6, 20, 24) with a five-point scale from almost never to almost always. The internal reliability of the measure has been reported to be satisfactory in many studies carried out across many populations (Allen et al., 2012; Neff & Pommier, 2013; Werner et al., 2012). The translated versions of this measure have also observed a six-factor structure of the scale across cultures (Arimitsu, 2014; Azizi et al., 2013; Castilho et al., 2015; Garcia-Campayo et al., 2014).

Self-Forgiveness Scale

            Standardized on Indian population with exploratory mixed methods design, the Self-Forgiveness Scale developed by Mudgal and Tiwari (2017) was used to assess the self-forgiveness of the participants. The self-forgiveness measure comprises of 30 items with a seven-point scale (1-very strongly disagree, 2-strongly disagree, 3-somewhat disagree, 4- neutral, 5-somewhat agree, 6- strongly agree, 7-very strongly agree). The scale represents three dimensions of Self -Forgiveness: Realization & Reparation (19 items), Guilt (6 items) and Attribution (5 items). The reliability coefficients were estimated by computing Cronbach’s Alpha for the three subscales and the overall scale of self-forgiveness were 0.90, 0.70, 0.62 and 0.77, respectively. Likewise, the validity was estimated by computing correlation between the scores of the overall self-forgiveness scale and overall quality of life (World Health Organization, 1996) that came to be 0.261.

Procedure

            The proposal of the study was submitted to the ethics committee of [BLINDED] India of approval. After the receipt of ethical approval, the scales were procured and the participants were consulted. Written consent was collected from each participant. Then, they were debriefed about the objectives of the study. The data was collected in the small groups of 5 to 10 participants at a time for accuracy and ease. Accordingly, the scales were supplied to each participant and asked to read the printed instructions on each scale. The scoring was completed as per the guidelines provided depicted in the manuals of the scales. No compensation was given to the participants for their participation in the study.

Data Analysis Plan

            The collected data were arranged according to the need of the statistical analyses and the same were treated with the help of SPSS v25. The data analyses were carried out in two phases: preliminary analysis, and mediation & path analyses. In the first phase, Coefficients of Correlation among self-esteem, self-compassion and self-forgiveness were carried out. Besides, hierarchical analyses assuming Self-Esteem and Self-Compassion (positive and negative self-compassion) as predictor variables and Self-Forgiveness with its three components (Realization & Reparation, Guilt and Attribution) as the criterion variables were also computed. The mediation analysis and path analysis were carried out in the second phase using SPSS V25.

Results

The results have been presented in two sub-sections: preliminary analyses, and mediation and path analyses.

Preliminary Analyses

At the stage of preliminary analyses, zero-order correlations were conducted to understand the nature and extent of association among self-esteem, components of self-forgiveness (realization & reparation, guilt &attribution) as well as positive and negative self-compassion (Table 1).
Table 1. Correlation matrix among self-esteem, components of self-forgiveness and positive/negative self-compassion
  Self-esteem Positive self-compassion Negative self-compassion Realization & Reparation   Guilt Attribution Overall Self-forgiveness
Self-esteem 1 .523** .269** .289** .002 -.148* .280**
Positive self-compassion .523** 1 -.140* .446** -.151* .150* .442**
Negative self-compassion .269** -.140* 1 -.180** .227** -.310** -.165**
Realization & Reparation .289** .446** -.180** 1 -.394** .218** .955**
Guilt .002 -.151* .227** -.394** 1 -.256** -.138*
Attribution -.148* .150* -.310** .218** -.256** 1 .301**
Overall Self-forgiveness .280** .442** -.165** .955** -.138* .301** 1