Introduction

There is no doubt that justice is important to people, and that it is motivating (see, for example, Lerner, 1980). Justice is concerned with fairness, and in the context of wrongdoing justice is typically understood in retributive terms: an offender must be punished or made to pay some form of compensation for their wrongdoing (Darley & Pittman, 2003). The desire for retribution is normative in Westernised societies (Ellsworth & Gross, 1994) and a central plank of Westernised legal systems (Darley, 2001), and there is physiological evidence that it is instinctual (Rusbult, Hannon, Stocker, & Finkel, 2005). Yet, there is also increasing evidence that wronged individuals embrace forgiveness as an appropriate response to transgressions (see, for example, McCullough, 2008). In this article, we ask, if justice is so important to people, who is more likely to forgive in the face of injustice?

The field of personality has, in one way, addressed this question. For example, it is well-established that the people most likely to forgive are those who score high on Agreeableness and empathy, whereas those least likely to forgive score high on Neuroticism and narcissism (see, for example, Berry, Worthington, O’Connor, Parrott, & Wade, 2005; McCullough & Hoyt, 2002; Strelan, 2007a). However, in this article, we take a different approach. Our focus is on the nexus between justice and forgiveness, particularly the motivations that underlie justice. If justice is important, motivating, instinctual and normative, what is the relation between the goals of justice—specifically, retributive justice—and forgiveness? Put another way, who is more likely to forgive, given that individuals also have a need to see justice done? Furthermore, we will address relations between justice goals and forgiveness taking into account the influence of motivational values. Although researchers have yet to explicate the relation between values and forgiveness, forgiveness is usually included as an exemplar of prosocial values (Schwartz, 1992), and individuals’ ideas about justice are closely linked to the values they hold (Feather, 1994).

We begin our analysis by noting that, historically, justice and forgiveness have been considered competitive or at least parallel constructs (see, for example, Exline & Baumeister, 2000; Reed & Aquino, 2003). Whereas justice is a fair-minded response, forgiveness is a compassionate response. When people forgive they forego the opportunity to respond retributively (Enright & Fitzgibbons, 2000); instead, they (at least) overcome negative cognitions, affect, and behaviours (Yamhure Thompson et al., 2005) and (at most) respond positively towards an offender (McCullough, Worthington, & Rachal, 1997). Forgiveness has also been equated with cancelling a debt (Exline & Baumeister, 2000) or not holding a grudge (Yamhure Thompson et al., 2005).

Nonetheless, recent studies suggest that justice and forgiveness may not be so incompatible, depending on how justice is conceptualised. Certainly, retributive justice is opposite to forgiveness (Strelan, Feather, & McKee, 2008). However, when the inclusive qualities of justice are emphasised—for example, social justice (Karremans & Van Lange, 2005); just world beliefs for the self (Strelan, 2007b; Strelan & Sutton, 2011); restorative justice (Strelan et al., 2008); and procedural and distributive justice for the self (Lucas, Young, Zhdanova, & Alexander, 2010)—individuals have been found to also endorse forgiveness. These studies suggest that when the goals of punishment are inclusive as opposed to retributive, differential relations with forgiveness may emerge.

Punishment Goals and Forgiveness

When people talk about a just response to wrongdoing, they typically refer to notions of retributive justice. Retributive justice, by definition, is concerned with how people should be punished for wrongdoing. As such it is a multi-faceted construct, insofar as it encapsulates many different goals—that is, people differ in their ideas about what punishment should achieve (Carlsmith, Darley, & Robinson, 2002).

In the criminal justice context, punishment goals are derived from philosophical penal theories. According to the absolute penal (or retributive) perspective (e.g. Kant, 1952), the violation of societal rules or norms requires a response that balances the scales of justice, and this may be achieved by punishing offenders. The punishment should be proportionate to the harm caused; in the process, offenders should get their just deserts. In this view, one goal of punishment is retribution; punishment is a moral response to something that has been done, and is an end in itself.

Conversely, relative penal (or utilitarian) theories (e.g. Bentham, 1962) propose that the goal of justice should be less a moral consideration where offenders get their just deserts and more an instrumental response. According to this perspective, the point of punishment is to minimise future offending, or control future behaviour. The social psychological and criminological literature has tended to posit three main punishment goals within the utilitarian framework: deterrence (of both offenders’ behaviour in future, and others in general), protection of society (e.g. by incapacitating the offender through incarceration), and offender rehabilitation (Carlsmith et al., 2002; Orth, 2003; Oswald, Hupfied, Klug, & Gabriel, 2002).

In recent years, a third perspective on wrongdoing has been embraced: restorative justice. Restorative justice seeks neither to punish nor protect but rather to repair. Specifically, it is concerned with healing relationships, whether literally or symbolically, between victims, offenders and communities (Braithwaite, 1989). The restorative justice model conceptualises transgressions as ‘social conflicts’ (de Keijser, van der Leeden, & Jackson, 2002). According to this view, justice is better served by the victim, offender and community engaging in a dialogue about the transgression, and making a shared decision about how the transgression will be dealt with (Braithwaite, 1989). Restorative justice may not be a punishment goal per se, but it certainly qualifies as a goal of justice. Indeed, experimental (Carlsmith et al., 2002; Ristovski & Wertheim, 2005), quasi-experimental (Strang, Sherman, & Angel, 2006), field (Sherman et al., 2005) and survey (Roberts & Stalans, 2004) data indicate that restorative justice concepts are important to people.

How might each of the different punishment goals relate to forgiveness? First, the retributive motivation for justice—that is, where punishment is an end in itself—is conceptually distinct from forgiveness, and has been empirically demonstrated as such (Strelan et al., 2008). Deterrence and societal protection goals might also be distinct from forgiveness. Although neither is concerned with punishment for its own sake, the focus of each is on punishment for the sake of the victim and society; ostensibly, neither is concerned with the welfare of the offender.

Conversely, the goal of rehabilitation may be conceptually compatible with forgiveness. Although rehabilitation may serve the function of protecting victims and society through reducing recidivism, it can also be understood as a more inclusive response to wrongdoing since an active effort is made to improve the offender’s welfare and outcomes.

Restorative justice is also an inclusive response to wrongdoing and therefore also has principles in common with forgiveness. Experimental (Ristovski & Wertheim, 2005) and field (Sherman et al., 2005) studies indicate that forgiveness can often play a role in restorative justice processes themselves. Furthermore, in one experiment where justice cognitions were manipulated, participants were more likely to associate restorative justice but not retributive justice with forgiveness (Strelan et al., 2008).

Values, Justice and Forgiveness

Following Rokeach (1973), Feather (1975, 1994, 1999) proposed that values can be conceived as abstract structures that involve the beliefs people hold about desirable ways of behaving or about desirable end states. Values have a normative quality about them and they transcend situations. They vary in their importance for self and they influence a person’s attitudes, beliefs, plans and decisions, and the way a person defines or interprets situations. They are hierarchically organised in terms of their importance for self and their activation depends on their importance for self and on the specific context. For example, a work of art would be more likely to trigger stimulation values and a transgression would be more likely to elicit benevolence or power values, but the activation of these respective values would also depend on how important they were for the individual.

Schwartz (1992) identified 10 motivationally distinct value types arranged in a circumplex model. These value types can be conceptualised along two bipolar dimensions: [a] self-enhancement values (power, achievement), which promote self-interest, are opposed to self-transcendent values (benevolence, universalism), which emphasise concern for others’ welfare; and [b] openness values (self-direction, stimulation), which reflect openness to change and new ideas, are opposed to conservation values (security, tradition and conformity), which emphasise maintenance of the status quo and threat avoidance. One value type, hedonism (pleasure and sensuous gratification for oneself), shares elements of self-enhancement and openness.

Of these higher-order value dimensions, openness values with their concern for self-direction and stimulation are less likely to be salient to justice and forgiveness. Due to their inclusive nature, we would expect that the self-transcending values of benevolence (preservation and enhancement of the welfare of significant others) and universalism (understanding, appreciation, tolerance and protection of all people) (Schwartz, 1992) would be associated with forgiveness and social justice and therefore, presumably, also restorative justice and rehabilitation goals, i.e. given that forgiveness and social and restorative justice and rehabilitation goals are conceptually compatible.

The self-enhancing values of power (status and prestige; control and dominance over others) and achievement (personal success through demonstrating competence according to social standards) reflect a concern for serving one’s own interests at the expense of others but also an awareness of the importance of maintaining authority and the salience of social standards (Schwartz, 1992). McKee and Feather (2008) found that self-enhancing values were positively related to retributive justice motives and punitive behaviour, including support for the death penalty. Thus, power in particular was expected to be positively related to the punishment goals of retribution, deterrence and societal protection.

Finally, conservation values, involving security, tradition and conformity, have been associated with right-wing authoritarianism and punitive responding (e.g. Feather, 2005). We expect that security values would be especially likely to be associated with retributive, deterrence and societal protection goals.

Aims and Hypotheses

The aim of the present study is to examine the extent to which attitudes towards different goals of justice are associated with tendencies towards forgiveness. In so doing, we make two initial inquiries. First, we address the inter-relations between specific punishment goals, restorative justice and forgiveness. Second, we relate values to justice goals and forgiveness. There is a paucity of research on the relation between values and justice-related variables, and this study contributes by investigating relations between values and forgiveness.

We propose three hypotheses. First, the extent to which justice and forgiveness are related depends on the goals of justice. Specifically, retributive, deterrence and societal protection goals will be negatively related to forgiveness, whereas restorative and rehabilitation goals will be positively associated with forgiveness.

Second, different values activate different justice goals. Thus, the self-transcending values of benevolence and universalism values will be positively associated with restorative justice, rehabilitation and forgiveness, whereas the self-enhancing values of power and achievement, and also security and hedonism, will be positively associated with retribution, deterrence and societal protection, and negatively related to forgiveness.

Our third hypothesis is based on the idea that values are by definition distal (see Feather, 1975, 1994, 1999; Rokeach, 1973; Schwartz, 1992). That is, they underlie attitudes, and therefore in this case, justice goals and forgiveness tendencies. Thus, while self-enhancing, self-transcending, and hedonism and security values will be significantly associated with forgiveness, the role of justice goals will be such that they will exert an influence on forgiveness over and above the contribution made by the values.

Method

Participants

There were 178 participants, all undergraduate psychology students at a large Australian university taking part for partial course credit (127 females and 51 males; M age = 20; SD = 4.41).

Materials and Procedure

Participants responded online to questionnaires measuring justice goals (retribution, deterrence, societal protection, rehabilitation and restorative), values, attitudes to forgiveness and dispositional forgiveness. Cronbach’s alpha (α), means and standard deviations for all measures are reported in Table 1.

Table 1 Means, standard deviations and Cronbach’s alpha for justice goals, values and forgivenessa
Full size table

Retribution, deterrence, societal protection and rehabilitation goals were measured using McKee’s (2005) Sentencing Goals Scale. This scale consists of 20 items measured on seven-point scales (1 = disagree; 7 = agree), with five items each for retribution (e.g. ‘Justice is not done if the offender is not punished in some way’), deterrence (e.g. ‘Crime rates would decrease if sentences were appropriately severe and publicized’), societal protection (e.g. ‘The purpose of court sentences should be to protect society from the offender’) and rehabilitation (e.g. ‘With the right approach most offenders can be rehabilitated back into society’). Total scores for each subscale range from 5 to 35 with higher scores indicating a more positive attitude towards each construct. The subscales have shown adequate internal reliability in previous studies (e.g. Feather & Souter, 2002).

Restorative justice goals were measured using a five-item scale developed for the present study. Three of the items are: ‘Justice should be about trying to heal relations between a victim, offender, and the community’; ‘Justice can be served if the victim, the offender, and the community meet to discuss, and find ways to repair, the harm that was done’; and ‘The ideal way to deal with crime is for a victim, the community, and the offender to meet to discuss how the crime should be dealt with’. The other two items were taken from de Keijser et al.’s (2002) measure of penal attitudes (‘The criminal justice system should accommodate the process of negotiation between offender and victim’; ‘Justice should be more about resolving conflicts, and less about punishing offenders’). All items were measured on seven-point scales (1 = disagree; 7 = agree). Total scores range from 5 to 35, with higher scores indicating a more positive attitude towards restorative justice.

Values were measured using the Schwartz Value Survey (Schwartz, 1992). This instrument consists of 57 values, representing 10 value types each accompanied by a brief descriptive phrase. Participants rated on a nine-point scale how important each value was as ‘a guiding principle in your life’ (−1 = opposed to my values; 7 = of supreme importance). To control for individual differences in scale use, following Schwartz and Rubel (2005) values were then ipsatized by centering each individual’s ratings for each value around his or her mean rating derived from all 57 values. Mean ratings were then calculated for each value type. As the value type benevolence as formulated by Schwartz (1992) contains the individual value of forgiving, the potential arose for an inflated estimation of the relation between the benevolence value type and other forgiveness variables in the analyses due to the overlap of constructs. To avoid this, a new measure of the value type benevolence was constructed which omitted the forgiving value, and this new variable was used in all reported analyses.

The value types, with their definitions in terms of their associated motivational goals, and the specific values associated with each type (in parentheses) for the current study were as follows (names and definitions for the value types are from Schwartz, 1992).

Universalism: understanding, appreciation, tolerance and protection for the welfare of all (broad-minded, wisdom, a world of beauty, equality, unity with nature, a world at peace, social justice and protecting the environment);

Benevolence: preservation and enhancement of the welfare of people with whom one is in frequent personal contact (honest, loyal, helpful and responsible);

Power: Social status and prestige, control, or dominance over people and resources (social power, wealth, authority and preservation of public image);

Achievement: personal success through demonstrating competence according to social standards (successful, ambitious, capable and influential);

Hedonism: pleasure and sensuous gratification (pleasure, enjoying life and self-indulgent);

Stimulation: excitement, novelty and challenge in life (a varied life, daring and an exciting life);

Self-direction: independent thought and action, choosing, creating, exploring (creativity, freedom, independent, curious and choosing own goals);

Tradition: respect, commitment, and acceptance of the customs and ideas provided by traditional culture or religion (respect for tradition, humble, accepting one’s portion in life, devout and moderate);

Conformity: restraint in actions, inclinations, and impulses likely to upset or harm others and violate social expectations or norms (self-discipline, obedient, politeness, honouring of parents and elders);

Security: safety, harmony, and stability of society, of relationships, and of self (family security, national security, reciprocation of favours, social order and clean).

Forgiveness attitudes were measured using Brown’s (2003) Attitudes Towards Forgiveness (ATF) scale. The ATF consists of six items (e.g. ‘It is admirable to be a forgiving person’) measured on seven-point scales (1 = strongly disagree; 7 = strongly agree). Total scores range from 6 to 42, with higher scores reflecting a stronger disposition to forgive.

Dispositional forgiveness was measured using the Forgiveness of Others subscale of the Heartland Forgiveness Scale (Yamhure Thompson et al., 2005). The subscale consists of six items (e.g. ‘When someone disappoints me, I can eventually move past it’), measured on seven-point scales (1 = almost always false of me; 7 = almost always true of me). Total scores range from 6 to 42, with higher scores reflecting a stronger disposition to forgive.

To minimise the number of analyses, and because the forgiveness attitudes and dispositional measures correlated quite well (r = .55), we combined these two measures into a single forgiveness score which we labelled ‘forgiveness’.

Results

Sample Characteristics

As Table 1 indicates, participants possessed fairly positive and equivalent attitudes towards each of the goals of justice. They tended to endorse self-transcendent values (universalism and benevolence), achievement, and elements of openness (hedonism and self-direction) and tended to place less emphasis on conservation values (tradition, conformity and security), power and stimulation. They were also fairly positively inclined towards forgiveness.

Relations Amongst Justice, Forgiveness and Values

Table 2 shows the correlations between the justice, values and forgiveness variables. As expected, retributive, deterrence and societal protection goals correlated positively and relatively strongly with each other, negatively with rehabilitation and restorative justice, and negatively or negligibly with universalism and benevolence values. Retributive and societal protection goals were positively related to security and power values. Retribution, deterrence and societal protection were unrelated to achievement and openness values but each was negatively related to forgiveness.

Table 2 Pearson product moment correlations between justice goals, values and forgivenessa
Full size table

Rehabilitation and restorative justice correlated positively with each other, with universalism, and with forgiveness, and negatively with power values. In addition, restorative justice was negatively related to achievement values. Apart from the negative relations between rehabilitation and security and hedonism values, rehabilitation and restorative justice were unrelated to any of the conservation and openness values.

In short, relations between each of the justice goals, values and forgiveness were generally as expected, with universalism and power values consistently related to the justice goals and forgiveness, and the orthogonal dimensions of conservation and openness values tending not to be related to justice goals and forgiveness. The one notable exception was benevolence, which was unrelated to any of the justice goals.

Values and Justice Goals Predicting Forgiveness

We hypothesised that while self-enhancing, self-transcending, and hedonism and security values would be significantly associated with forgiveness, justice goals would exert an influence over and above the contribution made by the values. Although achievement was unrelated to forgiveness, it was still included in analyses as our concern was with the combined influence of the six theoretically relevant value types.

Given their moderately strong correlations (rs > .60), the retributive, deterrence and societal protection goals were collapsed into an index labelled ‘punishment goals’, and the rehabilitation and restorative justice goals were collapsed into an index labelled ‘inclusive goals’. We then conducted a hierarchical multiple regression analysis in which the values were entered at step 1, and the punishment and inclusive goal indexes at step 2. The independent influence of the justice goals on forgiveness is demonstrated by the significance of the increase in the variance accounted for when the justice goals are included in the equation, as well as by the standardised β’s of the justice goals themselves. It may be seen from Table 3 that at step 1 the six value types predicted 26% of the variance on forgiveness, F(6, 160) = 9.40, p = .001. Entering the justice goals at step 2 resulted in a significant increment in variance to 42%, F(2, 158) = 20.89, p = .001, with punishment goals (4%), inclusive goals (5%), and benevolence (6%), and power (3%) values each accounting for significant amounts of unique variance (percentage in brackets calculated by squaring the semipartial correlations). Individuals endorsing punishment goals and power values tended to be less forgiving, while those for whom inclusive justice goals and benevolence values were important scored higher on forgiveness.

Table 3 Summary of hierarchical multiple regression analysis with values and justice goals predicting forgiveness
Full size table

Discussion

Justice and Forgiveness

Our aim in this study has been to identify who is most likely to forgive an injustice given that justice is important and motivating for people. One previous study (Strelan et al., 2008) demonstrated that manipulating restorative and retributive justice cognitions had differential effects on forgiveness. The present study extends this earlier work by treating the concept of retributive justice not in gross terms, but on the basis of the specific punishment goals that are associated with retributive justice. It seems that the extent to which justice and forgiveness are associated may depend on what people think justice is meant to achieve. As predicted, people are less likely to report forgiving tendencies when the goals of justice are operationalized as punishment oriented. Punishing transgressions because the offender should get his or her just deserts; punishing to deter offenders and others from repeating the transgression; or punishing an offender in order to protect society (e.g. by incapacitating them)—all are goals which appear to be antithetical to the notion of forgiveness. However, the reverse is true when justice is conceptualised in inclusive terms. People who think that the goals of justice should serve a more inclusive function, that is, by rehabilitating an offender or by putting in place processes that encourage the actual or symbolic restoration of relationships, are more likely to endorse forgiveness as a response to transgressions.

Values and Forgiveness

This is one of the first reported studies in which the relation between values and forgiveness has been explored. The self-transcending values of benevolence and universalism are positively related to forgiveness, whereas the self-enhancing values of power and achievement are negatively or negligibly related to forgiveness. These findings are as expected. Forgiveness and self-transcendent values are both inclusive in nature, each reflecting a concern for others’ welfare at both intimate and global levels. Indeed, forgiveness is usually listed as an exemplar of benevolence (e.g. Schwartz, 1992). Conversely, self-enhancing values tend to reflect the inter-related concerns of maintaining authority and social standards, and serving one’s own interests at the expense of others. As the label suggests, self-enhancing values are less concerned with the welfare of others.

Also as predicted, two conservation values (conformity and tradition) and two openness values (stimulation and self-direction) are unrelated to forgiveness. Conservation values emphasise maintenance of the status quo and threat avoidance and openness values reflect openness to change and new ideas and therefore each emphasises values fairly distinct from notions of forgiveness. That said, two values from the conservation-openness dimension, security and hedonism, are related (negatively) to forgivingness. People who value security are likely to have negative ATF. In Schwartz’s (1992) circumplex model, the conservation values of tradition and conformity lie on a dimension that is conceptually and empirically distinct from self-transcendence and self-enhancement (e.g. Schwartz & Rubel, 2005)—each of which is relevant to forgiveness—whereas security is located next to power and is therefore somewhat opposed to universalism. In other words, people who prefer security values may also be inclined towards values which emphasise the interests of the self before others and, in the interests of self-protection, the control and restraint of others if necessary. Thus security values, but not tradition and conformity values, emerge meaningfully when they are associated with forgiveness.

Hedonism values share elements with both openness and self-enhancement values, because hedonistic values emphasise the self-serving pursuit and gratification of new pleasurable and sensual experiences. In this study, these values are positively related to the openness values of stimulation and self-direction and the self-enhancing value of power, and negatively related to conservation and self-transcending values (see Table 2). Understandably, perhaps, the self-oriented aspect of hedonism discourages the endorsement of such other-oriented attitudes as forgiveness.

Justice and Values

As hypothesised, people who endorse retributive responses to transgressions—particularly retribution and societal protection goals—are more likely to endorse security and power values and less likely to embrace universalism values. The relations are reversed for people who endorse restorative and rehabilitative justice goals. Although there is a paucity of research on the relation between values and justice, these findings are consistent with some related work (McKee & Feather, 2008) which found that support for capital punishment was positively related to power and security values and endorsement of rehabilitation was positively related to benevolence and universalism values and negatively related to power and hedonism values.

There are two exceptions to our predictions which require comment. First, benevolence was unrelated to any of the justice goals. Individuals who value benevolence are concerned with the wellbeing of other individuals. However, in the present study, participants were asked to consider justice goals as broad concepts. It may be that if participants had been asked to consider an offending individual in a specific situation, benevolence values would have become more salient. Parenthetically, the nature of our enquiry may also help explain why Universalism values, which are concerned with others in general and which were significantly associated with four out of five justice goals, dropped out in the regression model whereas Benevolence retained a unique influence on forgiveness.

The second comment concerns the performance of the deterrence goal, which reflects a belief that the aim of punishment should be to send a message to the offender, and others, which discourages them from transgressing again. Although deterrence was positively and relatively strongly associated with retribution and societal protection, and negatively with rehabilitation and restorative justice and forgiveness, all as expected, it did not correlate significantly with any of the values. These null findings are especially curious in relation to power, security, and universalism values, each of which was consistently associated with the other justice goals. It is unclear why deterrence and values are apparently unrelated to each other. One possible explanation may lie in the way participants interpreted the deterrence items (e.g. ‘Crime rates would decrease if sentences were appropriately severe and publicized’). Deterrence is suggestive of controlling other people’s actions and therefore may trigger the authoritarian aspects of power and security values; conversely, deterrence might also suggest a concern for what is good for both an offender and society and therefore could also trigger benevolence and universalism values. In other words, deterrence may have resonated with participants regardless of the extent to which they endorsed self-transcending and self-enhancing values. Consequently, the correlations between the values and deterrence may have been negligible due to the fact that participants who scored high on power and participants who scored low on power (for example) may respond with similar ratings to items concerned with deterrence, but for different reasons.

Relations Amongst Values, Justice and Forgiveness

Together, self-transcending, self-enhancing, hedonism and security values have a moderate influence on forgiveness, accounting for 26% of the variance. The values most relevant to forgiveness are benevolence and power. Each consistently retains a unique and direct influence on forgiveness, regardless of the presence of justice goals. People who value the welfare of close others are more likely to forgive; those who value domination over others are less likely to forgive. Forgiveness appears to be more salient to such people when compared with those for whom universalism, achievement, security and hedonism values are important. Clearly, these additional self- or other-oriented values are relevant but they are not as influential when they are considered relative to power and benevolence.

The role of universalism requires comment. At the correlational level, universalism has a moderate association with forgiveness, yet when the other predictors are entered into the equation, the influence of universalism disappears and benevolence becomes the sole predictor from the self-transcending dimension. The regression analysis indicate that this is primarily a result of the variance shared between universalism and power in their relations with forgiveness. As predicted by the circumplex model, opposing value types such as universalism and power are negatively correlated with each other (see Table 2) and, as a consequence, tend to correlate in opposite directions with a third variable. In the case of the relations amongst universalism, power and forgiveness, this results in a reduction in the unique variance contributed by the two values, with the influence of power values being markedly reduced and the contribution of universalism values all but disappearing.

Finally, justice goals play a significant role over and above that of the values, predicting an additional amount of variance on forgiveness (16%). Thus, while values salient in the context of justice and forgiveness may be related to justice and forgiveness, the importance of the justice goals is such that they retain a unique influence on forgiveness after values are taken into account. People who endorse retributive, deterrence and societal protection goals are less likely to forgive; people who endorse restorative and rehabilitative goals are more likely to forgive.

In summary then, one may gain a better understanding of who is likely to be positively inclined towards forgiveness by attending to motivational values, specifically benevolence and power, and acknowledging people’s attitudes towards punishment goals. Differences between people in their endorsement of justice goals and their beliefs about forgiveness are related to their value systems, with benevolence and power values playing the major roles in predicting forgiveness; and, after controlling for the influence of values, justice goals also uniquely predict forgiveness.

Limitations

Some caveats should be noted. Our measures of justice were based on criminal justice terminology. While this may be advantageous in terms of face and ecological validity, it is possible that when participants responded they did so while placing themselves in the context of the criminal justice system. In other words, we have an indication of how people think about justice in one particular context; we cannot be sure that the results generalise to other situations, such as intimate relationships or the workplace. Future researchers should consider replicating these findings in alternative contexts, and/or developing measures that utilise language more appropriate to interpersonal or workplace environments.

In addition, the correlational nature of the study means that we cannot say anything about the direction of the relations we have hypothesised and observed. Indeed, given that the same sets of values theoretically are associated with both justice and forgiveness (Schwartz, 1992), it is reasonable to presume that the relation between justice and forgiveness may be bidirectional. For this study, we have proceeded on the basis that people’s attitudes towards justice influence the extent to which they are forgiving, but the relation could work the other way. That is, people’s attitudes and dispositions towards forgiving could influence their attitudes towards justice. At this early stage, our primary concern was with showing that, at a fundamental level, justice and forgiveness are related. Future researchers might focus on delineating the extent to which one is more likely to be contingent on the other (see, for example, Wenzel, Okimoto, Feather, & Platow (2008), for the theoretical position that forgiveness is one predictor of restorative justice).

Finally, while the present study demonstrates the relation amongst motivational values, justice goals and forgiveness, it is important to examine the role that all of these variables play within a more complex model describing the path from transgression to forgiveness, and using alternative methodologies. Do different types of transgressions prime different motivational values, for example? Are justice goals relevant in other situations such as those involving interpersonal transgressions or the work environment? Furthermore, what is the interplay with other relevant individual difference variables, such as justice sensitivity and justice orientation (see Schmitt, Gollwitzer, Maes, & Arbach, 2005)?

Implications

In this study, we have shown that people differ in the extent to which they associate retributive and inclusive justice attitudes with forgiveness, and that it is important to consider the role of values in considering the relations between justice and forgiveness. These results have important theoretical and practical implications for the justice–forgiveness relation. In theoretical terms, here is further evidence that justice and forgiveness may be more compatible than is often thought. People do associate forgiveness with justice, but it depends on the type of justice, or more specifically, on what individuals believe the goals of justice should be.

In practical terms, the findings provide a pointer towards predicting who is more likely to respond to inclusive and retributive processes following a transgression, not only within the criminal justice system, but perhaps also the workplace, and in intimate relationships (for example, in the contexts of relationship counselling or separation mediation). They show that values, particularly benevolence and power values, are relevant to both beliefs about justice goals and the willingness to forgive. People who believe that it is important to consider others’ welfare and wellbeing, and who embrace self-transcending values and who are less inclined to endorse self-enhancing values, are less likely to embrace retributive responses to transgressions and more likely to endorse inclusive justice goals and forgiveness.