Conceptualizing Forgiveness , Granting Forgiveness , and Seeking Forgiveness : A Turkish-French Comparison

The study examined the validity of the Turkish versions of three forgiveness-related questionnaires: the Conceptualizations of Forgiveness Questionnaire, the Forgivingness Questionnaire, and the Disposition to Seek Forgiveness Questionnaire. These questionnaires were translated from English to Turkish. The study also compared Turkish and French participants’ scores on these scales. The sample comprised 252 students living in Turkey and 216 students living in France. The four-factor model of conceptualization of forgiveness, the three-factor model of forgivingness, and the three-factor model of disposition to seek forgiveness were shown to hold in a Turkish sample of students. The associations observed in previous studies between several subscales of these questionnaires were also found in the Turkish sample, thus providing additional evidence of the construct validity of the Turkish versions. Finally, it was shown that (a) the unconditional forgiveness and unconditional seeking of forgiveness mean scores were lower among the Turkish participants than among the French participants, and (b) the sensitivity to circumstances scores were higher among the Turkish participants than among the French participants.

Forgiveness has been defined as the "forswearing of negative affect and judgment by viewing the wrongdoer with compassion and love, in the face of a wrongdoer's considerable injustice" (Enright et al., 1991, p. 123).Since the end of the 1990s, forgiveness has increasingly attracted the attention of social and clinical psychologists (Worthington, 2005).This is largely due to the fact that forgiveness has been empirically proven to constitute an efficient way for durably transforming, and even ending, interpersonal conflicts (Worthington, 2006).In modern, multicultural societies, misunderstandings and conflicts may arise between persons from different cultural or religious backgrounds.Thus, it is important to explore the possibly diverse meanings forgiveness may have among different populations (Ho & Fung, 2011;Sandage & Williamson, 2005).
The present study had three objectives.Firstly, it was aimed at examining, on a sample of Turkish students, the validity of three forgiveness-related questionnaires: the Conceptualizations of Forgiveness Questionnaire (CFQ, Mullet, Girard, & Bakhshi, 2004), the Forgivingness Questionnaire (FQ, Suwartono, Prawasti, & Mullet, 2007), and the Disposition to Seek Forgiveness Questionnaire (DSFQ, Chiaramello, Munoz Sastre, & Mullet, 2008).Secondly, it was aimed at comparing the scores observed using the previous Turkish adaptation of Snyder's Heartland Forgiveness Scale (Bugay & Demir, 2010a;Bugay, Demir, & Delevi, 2012) with the scores that are to be observed using the translated version of the FQ.Thirdly, it was aimed at comparing the CFQ, FQ and DSFQ scores observed in the Turkish sample with corresponding scores observed in previous studies conducted in France.

Conceptualizations of Forgiveness
A set of cross-cultural studies has recently investigated the way people conceptualize forgiveness.Factorial studies (Mullet et al., 2004) have evidenced a four-factor model of conceptualizations: (a) Forgiveness as a moral behavior, an evaluative factor, (b) Forgiveness as a change of heart, a factor that refers to the nature of forgiveness, (c) Forgiveness encourages repentance, a factor that refers to the possible effects of forgiveness, and (d), Forgiveness as a broad process, a factor that refers to the scope of forgiveness.This four-factor structure, initially found on a sample of French adults aged 18-80, has been able to fit data from samples of: (a) Congolese adults aged 18-82, (b) Uruguayan adults aged 17-84, (c) Colombian students aged about 22, and (d) Indian students aged about 26 (for an overview, see Tripathi & Mullet, 2010).
In the five cultures examined so far, forgiveness was regarded as a moral behavior, and significantly more among the Latin Americans than the other groups.Also, all cultures considered forgiveness a broad process, but intergroup differences were strong.The Congolese and the Latin Americans adhered to this view significantly more than the other groups.Regarding the consequences of forgiveness, strong differences were also present: The Indians and the Congolese adhered to the view that forgiveness has positive consequences (e.g., encourages repentance) more than the other groups.Finally, regarding change of heart, differences among the four groups were also strong.The Congolese and the Indians adhered to this view much more than the other groups.

Dispositional Forgiveness
Another set of cross-cultural studies has examined dispositional forgiveness, which Roberts (1995) defined as "the disposition to abort one's anger (or altogether to miss getting angry) at persons one takes to have wronged one culpably, by seeing them in the benevolent terms provided by reasons characteristic of forgiving" (p.290).Roberts coined the term forgivingness: an overall disposition to forgive, a disposition that manifests itself in most circumstances in life, whereas forgiveness only applies to particular circumstances (particular offenses).Factorial studies have evidenced a three-factor model of forgivingness: (a) Lasting Resentment expresses the victim's tendency to hold negative emotions, negative cognitions, and exhibit avoidance behaviors toward the offender, even in the presence of positive circumstances, (b) Sensitivity to Circumstances before forgiving expresses the victim's ability to analyze the pro and cons of harmful situations, and to build on the many circumstances of these situations for deciding whether to forgive the offender, and (c) Unconditional Forgiveness expresses the victim's tendency to harbor positive attitudes toward the offender even in the absence of positive circumstances.Unconditional Forgiveness can be viewed as the product of a type of personal, spiritual growth that may be relatively independent of external influences.Scores on these scales have been shown to be significantly related with behavioral measurements.As example, lasting resentment and unconditional forgiveness scores have been shown to be respectively higher and lower among homicide offenders than among control adults (Menezes Fonseca, Neto, & Mullet, 2012).This ternary structure, initially found on a sample of French adults aged 18-80, has been able to fit data from samples of: (a) Italian adults aged 18-80 living in Sardinia; (b) Portuguese students and adults aged 18-40 living in the north of the country; (c) Indonesian students, aged about 21, living in Java; (d) Tutsi victims of the 1994 genocide in Rwanda, (e) Chinese adults aged 18-86 from the Buddhist community, from the Christian community, or non-religious, living in Macao; and (f) Indian students aged about 26, from the Hindu community (for an overview, see Tripathi & Mullet, 2010).
The way people conceptualize forgiveness has been shown to be associated with their level of forgivingness.Notably, a positive association was found in western Europeans between unconditional forgiveness and the beliefs that (a) forgiveness can decrease negative feelings and increase positive feelings toward the offender (change of heart) and (b) forgiveness is a broad process that is not limited to the victim-offender dyad.A positive association was also found between the view that forgiveness is immoral and lasting resentment.Interestingly, exactly the same pattern of associations was found among Indian students, supporting the view that conceptualizations and daily practice of forgiveness are related in the same way in both cultures (Tripathi & Mullet, 2010).

Disposition to Seek Forgiveness
A third set of cross-cultural studies has examined the disposition to seek forgiveness.Chiaramello et al. (2009), using a French sample, have found a three-factor structure that closely paralleled the three-factor structure that has been detailed above regarding forgivingness: (a) inability in seeking forgiveness, (b) sensitivity to circumstances before seeking forgiveness, and (c) unconditional seeking of forgiveness.
Strong associations have been found between scores on the FQ and scores on the DSFQ, namely between lasting resentment and incapacity at seeking forgiveness, between both sensitivity to circumstances factors, and between unconditional granting of forgiveness and unconditional seeking of forgiveness.As in the case of forgivingness, the ternary structure has been shown to fit data obtained in five countries, as diverse than Angola, Brazil, Indonesia, Mozambique, and Portugal (Neto, Pinto, Suwartono, Chiaramello, & Mullet, 2011).It seems fair to conclude that this structure has cross-cultural generality.Scores on these scales have also been shown to be significantly related with behavioral measurements (Menezes Fonseca, Neto, & Mullet, 2012).

Research on Forgiveness in Turkey
Even though forgiveness has been well established in the field of psychology over the last 20 years (e.g., Freedman & Enright, 1996;Girard & Mullet, 1997;McCullough, Worthington & Rachal, 1997), the concept has, until recently, been largely neglected in Turkey.In the absence of a Turkish version of forgiveness measurement, only a few studies (Alpay, 2009;Taysin, 2007) have investigated forgiveness of others in marriage by using a single-item measure of forgiveness (i.e., I forgive him/her for what he/she did to me).
Fortunately, a Turkish version of Heartland Forgiveness Scale (Thompson et al., 2005) --a disposition to forgive scale created in the US, has recently been published (Bugay & Demir, 2010a;Bugay, Demir & Delevi, 2012), and, as a result, research in the area of forgiveness has expanded in Turkey.Studies using this scale have investigated (a) aspects of transgression as predictors of forgiveness of others (Bugay & Demir, 2010b), (b) social-cognitive, emotional and behavioral variables as determinants of self-forgiveness (Bugay, 2010), and (c) the role of rumination as a mediator between life satisfaction and forgiveness (Bugay & Demir, 2011).The creation of new validated forgiveness-related questionnaires in Turkish was motivated by the fact that (a) forgivingness has been shown to be a disposition involving more than one factor, and (b) absolutely no validated measures of disposition to seek forgiveness and conceptualization of forgiveness has been adapted to the Turkish population.

Hypotheses
Our first hypothesis was that the four-factor model of conceptualization of forgiveness, the three-factor model of forgivingness, and the three-factor model of disposition to seek forgiveness that have been repeatedly evidenced in other cultures would be found in the Turkish sample.The second hypothesis was that between (a) the score of forgiveness of others of the Heartland Scale and (b) the lasting resentment and the unconditional granting of forgiveness scores of the FQ, a strong association would be observed.
The third hypothesis was that between the CFQ and the FQ, the strongest associations observed in previous studies (Ballester, Muñoz Sastre, & Mullet, 2009;Tripathi & Mullet, 2010), namely between lasting resentment and immoral behavior, between lasting resentment and broad process, between sensitivity to circumstances before forgiving and encourage moral behavior, and between unconditional granting of forgiveness and broad process would be observed.The fourth hypothesis was that between the FQ and the DSFQ, the strongest associations observed in previous studies (Chiaramello et al., 2008), namely between lasting resentment and incapacity at seeking forgiveness, between both sensitivity to circumstances factors, and between Unconditional granting of forgiveness and unconditional seeking of forgiveness would be observed.
The fifth hypothesis was that between the CFQ and the DSFQ, several strong associations would be observed that would parallel the ones that had been observed in previous studies between the CFQ and the FQ, namely between incapacity at seeking forgiveness and immoral behavior, between sensitivity to circumstances before seeking forgiveness and encourage moral behavior, and between unconditional seeking of forgiveness and broad process.
Finally, the sixth hypothesis was that several differences will be observed between Turkish and French scores.In particular it was expected that Turkish scores regarding the unconditional forgiveness factor would be lower than the French scores.This hypothesis was based on previous results found in Lebanon (Mullet & Azar, 2009).Owing to the strong relationship between unconditional forgiveness and unconditional seeking of forgiveness, this expected difference is likely to extent to scores on the unconditional seeking of forgiveness factor.

Participants
The sample consisted of 252 students (51% females) from a public university in Turkey.One hundred fifty two participants were undergraduate students and 102 were graduate students.Their mean age was 23.21 years (SD = 3.88).For the cross country comparison the data were taken from Ballester et al. (2004), and from Chiaramello et al. (2008).The mean age of the comparison samples was 35.55 (SD = 14.51), and the percentage of females was 66% (N = 216).

Material
Four questionnaires were used: the CFQ, the FQ, the DSFQ, and the Heartland Forgiveness Scale (HFS).The CFQ consisted of 20 items, five for each of the four subscales: change of heart, encourages repentance, immoral behavior and broad process.These items are shown in Table 1.The FQ consisted of 17 items distributed in three subscales: lasting resentment, sensitivity to circumstances before forgiving and unconditional granting of forgiveness.These items are shown in Table 2.The DSFQ consisted of 15 items, five items for each of the three subscales: inability in seeking forgiveness, sensitivity to circumstances, and unconditional seeking of forgiveness.These items are shown in Table 3.
The fourth questionnaire, the HFS was used as a validity criterion.It is composed of 18 items, six for each of the three scales: Forgiveness of self (e.g., "Although I feel bad at first when I mess up, over time I can give myself some slack."), forgiveness of others (e.g., "I continue to punish a person who has done something that I think is wrong."),and forgiveness of situations (e.g., "When things go wrong for reasons that can't be controlled, I get stuck in negative thoughts about it.").As indicated above, the validity of the Turkish version of the HFS has already been established by Bugay et al. (2012).

Translation Process
The following steps were implemented to ensure equivalence in meaning.The questionnaires were first translated from English to Turkish by five bilingual counselors with Ph.D. degrees, and several possible translations were elaborated for each item.Second, three faculty members at the Department of Psychological Counseling and Guidance were presented the translated items and the original items, and instructed to indicate the best fitting translation for each item.The Turkish version of the questionnaires was subsequently shown to two English language teachers with M.S. degrees for additional checking.Finally, the ultimate version of the questionnaires was reviewed by a Turkish language teacher who controlled the accuracy of the Turkish formulations used.

Conceptualizations of Forgiveness
A CFA was conducted on the 20 items of the CFQ.The model tested was the four-factor correlated model suggested by Mullet et al. (2004).The results are shown in Table 1.The observed RMSEA value was .06[.05-.07], and the observed Chi²/df value was 323/164 = 1.97; that is, lower than 5.The CFI and RMR values were .87 and .070,respectively.
A MANCOVA was conducted with Country (Turkey vs. France) as the independent variable, Change of Heart, Encourages Repentance, Immoral Behavior and Broad Process as the dependent variables and Age and Gender as the covariates.The Country effect was significant, R (4, 464) = 13.88,p < .001.
As a significant effect was observed overall, four separate ANCOVAs were conducted, one for each of the dependant variable.Regarding Change of Heart, the country effect was not significant.Participants from Turkey tended to agree with the idea that forgiveness involves a change of heart more or less to the same extent (M=4.99)than participants from France (M=4.73).Regarding Encourages Repentance, the country effect was significant.Participants from Turkey tended to agree with the idea that forgiveness encourages repentance (M=6.25)more than participants from France did (M=4.93),F(1, 467) = 42.40,p < .001.Regarding Immoral Behavior, the country effect was not significant.Participants from Turkey tended to agree with the idea that forgiveness is an immoral behavior more or less to the same extent (M=2.35)than participants from France (M=2.45).Finally, regarding Broad Process, the country effect was significant.Participants from Turkey tended to agree with the idea that forgiveness can be a broad process (M=6.89)more than participants from France did (M=6.02),F(1, 467) = 19.05,p < .001.

Dispositional Forgiveness
A CFA was conducted on the 15 items of the FQ.The model tested was the three-factor correlated model suggested by Suwartono et al. (2007).The results are shown in As a significant effect was observed overall, three separate ANCOVAs were conducted, one for each of the dependant variable.Regarding Lasting Resentment, the country effect was not significant.Participants from Turkey tended to report a level of lasting resentment (M=3.38) that was close to the one reported by participants from France (M=3.27).Regarding Sensitivity to Circumstances, the country effect was significant.Participants from Turkey tended to report a level of sensitivity to circumstances (M=6.88) that was slightly higher than the one reported by participants from France (M=6.18),F(1, 467) = 15.17,p < .001.Finally, regarding Unconditional Forgiveness, the country effect was also significant.Participants from Turkey tended to report a level of unconditional forgiveness (M=3.76) that was slightly lower than the one reported by participants from France (M=4.42),F (1, 467) = 7.42, p < .01.

Disposition to Seek Forgiveness
A CFA was conducted on the 15 items of the DSFQ.The model tested was the three-factor correlated model suggested by Chiaramello et al. (2008).The results are shown in As a significant effect was observed overall, three separate ANCOVAs were conducted, one for each of the dependant variable.Regarding Incapacity at Seeking Forgiveness, the country effect was significant.Participants from Turkey tended to report a level of incapacity at seeking forgiveness (M=2.96) that was slightly higher than the one reported by participants from France (M=2.47),F(1, 440) = 7.37, p < .01.Regarding Sensitivity to Circumstances, the country effect was significant.Participants from Turkey tended to report a level of sensitivity to circumstances (M=6.69) that was higher than the one reported by participants from France (M=6.04),F(1, 440) = 12.44, p < .001.Finally, regarding Unconditional Seeking of Forgiveness, the country effect was also significant.Participants from Turkey tended to report a level of unconditional seeking of forgiveness (M=5.10) that was slightly lower than the one reported by participants from France (M=5.752),F(1, 440) = 7.52, p < .01.

Correlational Analyses
Correlation coefficients were computed between age, gender and all the forgiveness-related variables, including the HFS.The results are shown in Table 4. Owing to the great number of coefficients, the significance threshold was set at .001.

Discussion
The study examined the validity of the Turkish version of three forgiveness-related questionnaires: the Conceptualizations of Forgiveness Questionnaire, the Forgivingness Questionnaire, and the Disposition to Seek Forgiveness Questionnaires.As hypothesized, the four-factor model of conceptualization of forgiveness, the three-factor model of forgivingness, and the three-factor model of disposition to seek forgiveness were found to hold in the Turkish sample.There is thus empirical evidence for the construct validity of the Turkish version of these questionnaires.The good fit of the three-factor model of forgivingness in the Turkish sample had, however, to be contrasted with the bad fit of this model in a Lebanese sample previously reported by Mullet and Azar (2009).In their study, a specific apologies factor was present in the ternary structure, in addition to the unconditional forgiveness factor and a modified lasting resentment factor.This difference in the nature of the factors was attributed to the fact that among Muslims there is a specific status attributed to apologies and repentance in the forgiveness process.The present study leads to think that this difference in structure may be specific to Lebanese Muslims.
As hypothesized, strong correlations were observed between the score of forgiveness of others of the HFS on the one hand, and the scores on the lasting resentment and unconditional granting of forgiveness scales of the FQ.
There is thus empirical evidence for the convergent validity of the Turkish version of these two questionnaires.
As hypothesized, the strongest associations observed in previous studies between (a) the CFQ and the FQ, namely between lasting resentment and immoral behavior, between lasting resentment and broad process, between sensitivity to circumstances before forgiving and encourage moral behavior, and between unconditional granting of forgiveness and broad process, and (b) between the FQ and the DSFQ, namely between lasting resentment and incapacity at seeking forgiveness, between both sensitivity to circumstances factors, and between unconditional granting of forgiveness and unconditional seeking of forgiveness were observed in the Turkish sample.These associations provide additional evidence for the construct validity of the Turkish version of these questionnaires.
As hypothesized, several associations paralleling the ones that had been observed in previous studies were observed between the CFQ and the DSFQ.Incapacity at seeking forgiveness and immoral behavior were shown to be significantly correlated as well as (a) sensitivity to circumstances before forgiving and encourage moral behavior and (b) unconditional seeking of forgiveness and broad process, thus providing still more evidence for the construct validity of the Turkish version of these questionnaires.
Finally, as hypothesized, the unconditional forgiveness and unconditional seeking of forgiveness mean scores were lower among the Turkish participants than among the French participants.These differences were consistent with findings by Azar and Mullet (2009).Other differences have been found.In particular, Turkish participants, more than the French tend to conceptualize forgiveness as a broad process and as a process that encourages the offender's repentance.The Turkish may thus be close in their views to the Indians and the Latin Americans, respectively, two cultures that are considered as less individualistic than the French one.Also, the Turkish, more than the French were sensitive to the circumstances of the offense before forgiving or before seeking forgiveness.This finding is consistent with the view developed earlier that among Muslims, a specific status in the forgiveness process is attributed to apologies and repentance, two undoubtedly important circumstances in this kind of situation.
In summary, valid forgiveness-related questionnaires are now at the disposal of Turkish researchers who are interested in examining different aspects of forgiveness among Turkish-speaking students and young adults.Future studies should, for example, explore the relationship between Turkish people's attitude towards forgiveness and the way (a) they behave in the family (e.g., family violence), (b) they behave at school (e.g., bullying), (c) they conceive of the functioning of institutions (e.g., the educational system, the police, the justice system), (d) they consider certain national events (e.g., mass violence) or certain major international events (e.g., terrorism).

Table 1 .
Results of the confirmatory analysis for the conceptualization items, means, standard deviations and Cronbach's alpha

Table 2 .
Results of the confirmatory factor analysis for forgivingness ıtems, means, standard deviation, and Chronbach's alphas Country as the independent variable, Lasting Resentment, Sensitivity to Circumstances and Unconditional Forgiveness as the dependent variables and Age and Gender as the covariates.The Country effect was significant, R (3, 465) = 8.54, p < .001.

Table 3 .
Results of the confirmatory factor analysis for the disposition to seek items.means, standard deviations and Cronbach's alphas Country as the independent variable, Incapacity at Seeking Forgiveness, Sensitivity to Circumstances and Unconditional Seeking of Forgiveness as the dependent variables and Age and Gender as the covariates.The Country effect was significant, R(3, 438) = 8.22, p < .001.

Table 4 .
Correlations between age, gender, and the forgiveness-related variables