THE IMPACT OF THE TRUTH AND RECONCILIATION COMMISSION ON PSYCHOLOGICAL DISTRESS AND FORGIVENESS IN SOUTH AFRICA

Abstract

BACKGROUND

Legislation to establish a Truth and Reconciliation Commission (TRC) was passed soon after election of South Africa’s first democratic government. Discourse around the TRC focused on the importance of bearing witness to the past, and on the healing powers of forgiveness. However, there was also a concern that individuals with TRC relevant experience would simply be re-traumatized by participation in the process. To date, there has been little empirical data for either hypothesis.

METHODS

A nationally representative survey of the South African population (n=4351) was undertaken 6 to 8 years after the TRC process began. Information about subjects’ exposure to and participation in the TRC was collected, and views about the testimony of survivors and perpetrators were assessed. To determine the predictors of distress, anger, and forgiveness, linear regressions were undertaken with inclusion of demographic variables, exposure to TRC variables, and attitudes to the TRC.

RESULTS

Distress was significantly associated with specific demographic factors (female gender, less education), with having a TRC related experience to share, and with negative perceptions of the TRC (a negative view of survivors’ testimony). Anger had similar associations but was also predicted by lower age. Forgiveness was associated with age and education, with being Coloured, and with having a positive view of perpetrator’s testimony, while it was inversely associated with having a TRC experience to share. Distress and anger correlated inversely with forgiveness. Perceptions of the TRC were moderately positive irrespective of many demographic variables (race, education, age).

CONCLUSION

In this cross-sectional study, causal relationships are difficult to ascertain. Nevertheless, relationships between increased distress/anger, having a TRC relevant experience to share, and negative perceptions of the TRC, support a view that bearing testimony is not necessarily helpful to survivors. However, in the population as a whole, moderately positive attitudes towards the TRC across sociodemographic variables support a view that the TRC helped provide knowledge and acknowledgment of the past.

INTRODUCTION

In recent decades, many newly democratic countries have grappled with the question of how best to respond to the gross human rights violations committed by past regimes (8;12). In South Africa, the first democratic government was elected in 1994, and it responded to the atrocities of apartheid by passing the Promotion of National Unity and Reconciliation Act, which in turn established a Truth and Reconciliation Commission (TRC). The Act provided the possibility of reparations for victims of human rights violations, and for amnesty from prosecution for perpetrators.

A key aspect of the Act was that it focused on the construct of restorative justice rather than on retributive justice (4;9). This was partly a political compromise, the result of a negotiated settlement between representatives of the old and the new regimes, which took into account the slim chances of successful prosecution of perpetrators. In addition, however, there was active discussion, during the planning and the execution of the TRC, on the importance of bearing witness to the past, and on the possibility of national reconciliation via a process of truth-telling and forgiveness.

Thus, the Act stated that the objectives of the TRC were “to promote national unity and reconciliation in a spirit of understanding which transcends the conflicts and divisions of the past”. This would be done by giving victims an opportunity to relate the violations they had suffered, and by giving amnesty to those perpetrators who gave full disclosure. The vice-chairperson of the TRC, Dr. Alex Boraine, stated in a lecture on the TRC process, "South Africa has decided to say no to amnesia and yes to remembrance; to say no to full-scale prosecutions and yes to forgiveness" (7).

In the South African mental health community there was some debate about the psychological value of the Truth and Reconciliation Commission (16;17). On the one hand, the emphasis on bearing witness and on forgiveness, was redolent of a psychological discourse on the importance of catharsis in healing, and of reports of the value of “testimony therapy” (1;11). On the other hand, the TRC was a quasi-legal procedure which did not necessarily provide victims of gross human rights violations with appropriate treatment, and which might even re-traumatize them. Even if the TRC were helpful for national reconciliation, the question of its value for particular individuals was therefore contentious.

To date, however, there have been few empirical data to support a relationship between exposure to or participation in the TRC, and levels of distress, anger, or forgiveness in South Africa. Kaminer et al (13), for example, found that in people who had experienced gross human rights violations in South Africa, there was a significant association between increased levels of psychopathology, and lower levels of forgiveness. A nationally representative survey of mental disorders in South Africa, the South Africa Stress and Health Study (SASH), undertaken after the conclusion of the TRC process (19), provided an opportunity to assess these relationships.

The data were used to address two hypotheses, which emerge from the debate about the TRC outlined earlier. According to the first hypothesis, bearing witness to the TRC (as assessed by media exposure) would be associated with a positive perception of the TRC and with greater forgiveness. On the second hypothesis, having a TRC relevant experience to share, or direct participation in the TRC, would be associated with greater distress and anger.

METHODS

The SASH was a national probability sample of 4351 adult South Africans living in both households and hostel quarters (19). Hostel quarters were included to maximize coverage of young working age males. The sample did not include individuals in institutions or in the military. The sample was selected using a three stage probability sample design. The first stage involved the selection of stratified primary sample areas based on the 2001 South African Census Enumeration Areas (EAs). The second stage involved the sampling of housing units from each EA and the third stage was the random selection of one adult respondent from each housing unit.

SASH interviewers were extensively trained in centralized group sessions lasting one week. The interviews were conducted face to face in seven different languages: English, Afrikaans, Zulu, Xhosa, Northern Sotho, Southern Sotho, and Tswana. Interviews lasted an average of three and a half hours, with some requiring more than one visit to complete. Data were collected between January 2002 and June 2004. Field interviewers made up to three attempts to contact each respondent. The overall response rate was 85.5%.

The SASH data set includes person-level analysis weights that incorporate sample selection, nonresponse and post-stratification factors (19). This weight was used in computing estimates of descriptive statistics for the survey population (e.g. estimates of population means and proportions) and for estimation of analytical statistics (e.g. regression coefficients) required in modeling relationships among variables in the survey population.

The SASH included several indicators of both participation in the TRC and perceptions of the TRC. Participation in the TRC was assessed by a series of four questions reflecting exposure to the TRC. Respondents were asked whether they had: 1) provided testimony or information to the TRC; 2) attended public meetings of the TRC; 3) watched or listened to the TRC hearings on television or radio (dichotomized as “a lot” or “some” versus “a little” or “not at all”). The fourth measure attempted to capture the presence of exposure to a human rights violation that was not shared with the TRC. Subjects who indicated that they had not provided testimony or information to the TRC were asked if they had ever had an experience that was the kind that the TRC should want to know about. All questions were coded dichotomously.

The first two measures of TRC perceptions asked respondents to rate the work of the TRC, on a scale from poor to excellent, in terms of both their expectations when the TRC started its work (Initial Perceptions) and their final summary judgment of the TRC’s performance (Final Perception). In addition, two scales assessed how respondents perceived both the victims and the perpetrators who testified before the TRC. The positive perception of victims scale summed three items that reflected feeling that victims were truthful, treated well, and should be compensated (alpha reliability = .62). The positive perception of perpetrators scale consists of 4-items that indicate that perpetrators were truthful, treated well, gave genuine apologies, and should be given amnesty if truly sorry (alpha=.66). Higher scores on both scales reflect more positive perceptions of the TRC.

Two scales were used to represent psychological distress. First, we utilized the 30-day non-specific distress (NSD) scale (14). This scale consists of the following ten items where respondents were asked “In the past 30 days, how often did you feel: (1) nervous; (2) that nothing could calm you down; (3) hopeless; (4) restless or fidgety; (5) so restless that you could not sit still; (6) depressed; (7) that everything was an effort; (8) so sad that nothing could cheer you up; (9) worthless; (10) tired out for no good reason?” Response categories were all, most, some, or a little. All items were reverse coded and summed so that a high score represents high NSD (alpha = .88).

The second scale measured anger/hostility. Respondents were asked to reflect on the past 30 days and note how often they (1) were you irritable or grumpy; (2) were you mad or angry; (3) were you so angry that you felt out of control; (4) did you have an urge to hit, push, or hurt someone; (5) did you have an urge to break or smash something?” Items were coded so that high values represent high anger/hostility (alpha = .83).

Forgiveness of others was assessed using a two item scale, adapted from previous community surveys of this construct in the United States (18). Respondents were asked to state their level of agreement or disagreement with the statements (1) I have forgiven those who have hurt me, and (2) When I have hurt someone I often ask the other person’s forgiveness (alpha=.55). This scale reflects an orientation of social forgiveness, with a higher score indicating greater levels of forgiveness.

Associations of distress, anger, and forgiveness with sociodemographic variables, exposure to TRC variables, and perceptions of the TRC were assessed using hierarchical ordinary least squares regression models. During apartheid in South Africa, people were classified as “Whites”, “Blacks”, “Coloureds”, or “Indians”, and although we are concerned not to further reify these distinctions, understanding subjects’ historical experiences and their consequent mental health patterns require recourse to using this racial terminology in sociodemographic analyses. On step 1 demographic variables were entered (Model 1), on step 2 exposure to TRC variables were entered (Model 2), and on step 3 perceptions of the TRC were entered (Model 3). The final model included all variables (Model 4). The associations between the three outcome variables, psychological distress, anger, and forgiveness were also examined.

A small group of respondents in the SASH sample (n=26) reported testifying to the TRC in some form. These individuals were asked questions concerning their interactions with the TRC, and a 4-item scale was created denoting perceptions of the experience. Respondents indicated whether they felt the TRC spent enough time with them, treated them with dignity, listened to them tell their story, and took them seriously. The median of the scale was used to demarcate positive experiences (n=16) from negative experiences (n=10) with the TRC. T-tests were conducted to examine differences between the groups on psychological distress, anger, and forgiveness.

RESULTS

Table 1 displays the sociodemographic variations in percentages of affirmative responses to the four TRC participation questions. As stated above, only a small number of respondents (0.7%) in the sample participated in providing testimony to the TRC. Those who were employed were more likely to report having testified to the TRC. Compared to those with a positive experience of testifying, individuals with a negative experience (i.e., the TRC did not spend enough time, listen to them, and treated them with less respect) scored significantly higher on distress and anger (p<.001) (data not shown).

Table 1.

Sociodemographic Variations in Participation with the TRC

	Provided Information %	Attended (% a lot/some /a little)	Media Exposure (%a lot/some)	Experience to share with %
Total	0.7	12.6	39.7	1.7
Sex
Female	0.4	9.8	35.2	0.9
Male	1.0*	15.7***	44.9***	2.6***
Marital Status
Married	0.8	12.4	40.5	1.6
Unmarried	0.6	12.7	38.9	1.8
Age
18–34	0.8	12.8	39.9	1.9
35–49	0.7	12.1	39.4	1.4
50–64	0.8	13.2	43.0	1.8
65+	0.0	11.1	30.6*	1.5
Education
None	0.3	10.4	30.0	1.3
Grade 1–7	0.3	10.9	29.2	1.1
Grade 8–11	0.6	14.0	38.7	2.3
Grade 12	1.0	12.6	46.2	1.5
Grade 13+	1.2	12.2	49.0***	1.7
Employment
Employed	1.2	13.5	45.2	2.3
Unemployed	0.5*	12.1	37.2***	1.5
Race
White	0.1	6.7	33.1	0.1
Black	0.8	13.9	40.3	2.0
Coloured	0.4	8.3	40.7	1.6
Indian	1.0	11.2***	43.7*	0.5*
Income
None	0.3	19.5	41.9	2.3
R 1–2,999	0.5	12.9	37.8	3.0
R 3,000–5,999	0.9	13.0	45.0	1.0
R 6,000–11,999	0.3	14.2	45.9	2.3
R 12,000+	0.9	10.2***	38.2**	1.0**
Area
Rural	0.7	12.6	35.1	1.2
Urban	0.7	12.6	42.7***	2.0*

Chi-square was used to test for significance against the reference group (first row in each section)

^*p<=.05;

^**p<.01;

^***p<.001

A small number (1.7%) of subjects who did not testify before the TRC reported having an experience that was the kind that the TRC should want to know about. Males were significantly more likely than females, Blacks and Coloureds were significantly more likely than Whites or Indians, and individuals from urban areas were significantly more likely than those in rural areas to report having a TRC relevant experience that was not shared.

Of the sample population, 12.6 percent reported having attended a TRC event. Males, blacks, and those with no income were significantly more likely to report having attended a public meeting of the TRC when compared to other sociodemographic groupings. Conversely, Whites and those with the highest levels of income were less likely to have attended a TRC function.

The TRC was heard on radio or seen on television by approximately 40 percent of the population, but there was wide sociodemographic variation in that exposure. Women, the elderly (65+), those with lower levels of education, the unemployed, Whites, individuals in rural areas, and those with lower levels of income were significantly less likely to report having watched or listened to the TRC proceedings.

Table 2 displays sociodemographic variations in perceptions of the TRC. Note the percentage above median is closer to 40% than to 50% because the “don’t know” and “refused” categories are high.

Table 2.

Sociodemographic Variations in Perceptions of the TRC

	Positive Perception of Victims (%>median)	Positive Perception of Perpetrators (%>median)	Initial Perception (%good/ excellent)	Final Perception (%good/ excellent)
Total	37.4	40.3	54.5	45.4
Sex
Female	36.3	39.3	52.7	43.0
Male	38.7	41.5	56.7**	48.1**
Marital Status
Married	36.1	39.0	52.4	44.0
Unmarried	38.7	41.6	56.8**	46.8
Age
18–34	36.3	41.4	56.6	48.2
35–49	40.4	39.6	53.4	42.5
50–64	37.5	41.2	54.4	44.4
65+	30.4*	30.2*	41.3***	36.4***
Education
None	34.5	33.7	45.9	37.2
Grade 1–7	35.5	32.3	48.7	42.3
Grade 8–11	38.8	39.0	54.6	45.9
Grade 12	39.4	46.3	60.8	49.7
Grade 13+	34.8	46.7***	55.8***	44.9**
Employment
Employed	37.9	42.5	56.7	46.0
Unemployed	37.2	39.3	53.6	45.1
Race
White	22.5	44.2	35.1	29.5
Black	39.0	39.0	57.4	47.6
Coloured	38.8	43.5	52.7	43.4
Indian	40.8***	48.4*	52.9***	48.4***
Income
None	32.3	37.2	57.0	43.9
R 1–2,999	39.1	40.6	56.9	47.1
R 3,000–5,999	38.9	42.5	56.2	48.3
R 6,000–11,999	42.9	45.3	63.2	56.7
R 12,000+	36.8	39.8	51.1***	42.7***
Area
Rural	37.2	39.4	55.0	47.3
Urban	37.5	40.9	54.3	44.2

Chi-square was used to test for significance against the reference group (first row in each section)

^*p<=.05;

^**p<.01;

^***p<.001

The overall perception of TRC victims was generally positive (mean item score of 9.8 out of 12). Older individuals had a less positive view of TRC victims than younger ones. Whites had a markedly less positive perception of the TRC victims than did other racial groups. The overall perception of TRC perpetrators (10.6 out of 16) was less positive than that of victims. The view of TRC perpetrators is roughly equivalent to a neutral outlook. Older and less educated individuals had less positive views of TRC perpetrators, while Blacks had less positive perceptions of TRC perpetrators than other groups.

Table 2 also shows the mean overall ratings for the two global views of TRC performance, Initial Perception and Final Perception. While there was a significant drop in the view of the job the TRC performed from the initial to final perception (paired t-test; t=13.9, p<.001) the general ratings of approximately 3 are representative of a “very good” job.

There were several differences by sociodemographics for the initial perceptions of the TRC. Individuals who were male, unmarried, younger, and more educated had a more positive view of the job the TRC would do, while Whites and those in the highest income bracket had a more negative view. The final perceptions of the TRC performance, while lower than initial views, show a similar demographic pattern. Individuals who were male, younger, and more educated had a more positive view, while Whites and both those with the lowest and highest levels of income had a more negative view.

Table 3 displays the results of four hierarchical OLS regression models estimating the relationship between demographic variables, TRC participation, and TRC perception with psychological distress. Model 1 displays the association of sociodemographic variables on distress and serves as a control for subsequent regressions estimating TRC variables. Higher distress is associated with female gender, lower education levels, unemployment, and being black. Model 2 shows that participation in the TRC significantly predicts distress after controlling for sociodemographics. In particular, attending the TRC and having an experience that the TRC might like to know about were associated with higher distress. In model 3 perceptions of the TRC add significant variance in predicting distress as well. Individuals with negative views of TRC victim’s testimony reported higher distress. In the final model (model 4) both TRC participation and perception of victims remain as significant predictors of distress.

Table 3.

OLS Regression Coefficients for the Association between TRC Variables and Psychological Distress (N=4050)

Model	1	2	3	4
Demographics
Male	−.083***	−.105***	−.089***	−.107***
Married	.024	.023	.021	.018
Age	−.002	−.002*	−.002*	−.002*
Education	−.024***	−.028***	−.026***	−.028***
Employed	−.097***	−.105***	−.099***	−.104***
Race (omitted = Black)
White	−.283***	−.238***	−.268***	−.240***
Coloured	−.149***	−.131**	−.143***	−.130**
Indian	−.193**	−.179**	−.178**	−.169*
Participation in TRC
Media Exposure		.067**		.060*
Attended		.198***		.186***
Provided info		.347*		.340*
Had experience		.265**		.238*
Perceptions of TRC
Initial perception (omitted = poor)
Good/excellent			.036	.012
Fair			.067	.063
DK/Refused			−.013	−.004
Final perception (omitted = poor)
Good/excellent			−.055	−.062
Fair			−.097	−.088
DK/Refused			−.012	.009
View of Victims (omitted = negative)
Positive			−.100***	−.092***
DK/Refused			−.181**	−.179**
View of Perpetrators (omitted = negative)
Positive			−.042	−.042
DK/Refused			−.063	−.046

^*p<=.05;

^**p<.01;

^***p<.001

A similar set of four regressions are presented in Table 4 estimating the effects of the TRC on anger. Model 1 shows that higher levels of anger are associated with female gender and younger age. Blacks have significantly higher anger levels than Whites but are not different from Coloureds and Indians. In model two the addition of TRC participation indicates that attending the TRC is associated with higher anger. The addition of TRC perception variables in model 3 adds to the prediction of anger through negative perceptions of the TRC victim’s testimony. In the overall Model 4 demographic, TRC participation, and TRC perceptions variables remain significant predictors of anger.

Table 4.

OLS Regression Coefficients for the Association between TRC Variables and Anger (N=4043)

Model	1	2	3	4
Demographics
Male	−.061**	−.075***	−.068**	−.078***
Married	.013	.013	.009	.008
Age	−.003***	−.004***	−.003***	−.004***
Education	−.005	−.007*	−.007*	−.008*
Employed	−.002	−.007	−.004	−.006
Race (omitted = Black)
White	−.137***	−.111**	−.131***	−.116**
Coloured	.033	.044	.039	.047
Indian	−.048	−.040	−.040	−.036
Participation in TRC
Media Exposure		.028		.014
Attended		.132***		.122***
Provided info		.226		.235
Had experience		.146		.112
Perceptions of TRC
Initial perception (omitted = poor)
Good/excellent			−.023	−.036
Fair			.040	.038
DK/Refused			−.112	−.109
Final perception (omitted = poor)
Good/excellent			−.103	−.105
Fair			−.133*	−.129*
DK/Refused			−.066	−.057
View of Victims (omitted = negative)
Positive			−.051*	−.045
DK/Refused			−.044	−.043
View of Perpetrtors (omitted = negative)
Positive			−.039	−.038
DK/Refused			−.069	−.061

^*p<=.05;

^**p<.01;

^***p<.001

Table 5 presents the regression estimates for TRC variables on forgiveness of others. The sociodemographic regression (model 1) reveals that greater forgiveness is associated with female gender, being married, older age, and higher education. In addition, coloureds also report higher levels of forgiveness of others than do blacks. When the TRC participation variables are added in Model 2 they add to the prediction of forgiveness. Individuals exposed to the TRC via the media report greater forgiveness of others while those who attended the TRC or provided information to the TRC report being less forgiving of others. Model 3 shows that TRC perception variables are positively associated with forgiveness of others. Higher forgiveness is reflected in more positive views of victim and perpetrator testimony alike. In the overall Model 4 demographic variables, TRC participation, and TRC perceptions remain significant predictors of forgiveness, although a relationship between having an experience to share with the TRC is now associated with increased forgiveness.

Table 5.

OLS Regression Coefficients for the Association between TRC Variables and Forgiveness of Others (N=3992)

Model	1	2	3	4
Demographics
Male	−.074**	−.070**	−.078**	−.070**
Married	.073**	.072**	.078**	.079**
Age	.005***	.005***	.005***	.005***
Education	.016***	.015***	.014***	.014***
Employed	−.034	−.036	−.036	−.036
Race (omitted = Black)
White	.041	.037	.061	.048
Coloured	.129**	.116**	.133***	.119**
Indian	.013	.008	−.002	−.005
Participation in TRC
Media Exposure		.108***		.054
Attended		−.248***		−.243***
Provided info		−.279*		−.288*
Had experience		.168		.198*
Perceptions of TRC
Initial perception (omitted = poor)
Good/excellent			−.048	−.037
Fair			−.058	−.061
DK/Refused			−.035	−.033
Final perception (omitted = poor)
Good/excellent			.111	.114
Fair			.000	.006
DK/Refused			−.100	−.095
View of Victims (omitted = negative)
Positive			.077**	.060*
DK/Refused			.105	.102
View of Perpetrators (omitted = negative)
Positive			.141***	.139***
DK/Refused			.143**	.141**

^*p<=.05;

^**p<.01;

^***p<.001

For all three outcome variables (distress, anger, and forgiveness) both TRC participation and perceptions are significant predictors after controlling for demographics. The TRC variables account for a relatively low proportion of the variance (approximately 4%) in these psychological dimensions. It is notable that the TRC variables do not appear to mediate relationships between race and psychological outcomes. Distress and anger were themselves highly correlated (r=.66, p<.0001). Forgiveness was inversely, but more modestly, associated with both distress (r=−.080, p<.0001) and anger (r=−.081, p<.0001).

DISCUSSION

Although only a small percentage of the South African population testified in front of the TRC, more than 1 in 10 attended hearings, and around 40% were exposed to the TRC via the media (albeit to a variable extent). Thus, the TRC certainly achieved its aim of engaging a substantial portion of the South African population. Nevertheless, it is notable that relatively few subjects (1.7%) felt that they had had an experience that would be of interest to the TRC, perhaps indicative of the frequent focus of the TRC on high-profile cases involving gross human rights violations. That women, the elderly, those with lower levels of education, the unemployed, individuals in rural areas, and those with lower levels of income were less likely to report having watched or listened to the TRC proceedings is consistent with the lack of resources in contemporary South Africa, and the consequent difficulties of achieving full participation of the citizenry in public debate.

The data here provide some evidence for both of the hypotheses presented earlier. The relatively widespread participation of the public in the TRC was accompanied by relatively positive global perceptions of the TRC. Although perceptions became more negative over time, although perceptions of the way in which the TRC treated victims were more positive than perceptions of the way in which perpetrators were treated, and although not all sectors of the population were equally positive about the TRC, the relatively positive global perceptions of the TRC support the argument that the South African TRC has been a transparent and effective social process, that helped bring knowledge and acknowledgment (3;17). The association of decreased distress (especially) and anger with a more positive view of victims, and of increased forgiveness with a more positive view of perpetrators (especially) and victims, is consistent with an argument that the TRC may in fact have contributed to a process of reconciliation.

At the same time, there was significant associations between having attended the TRC with increased distress / anger and decreased forgiveness, and having had an experience to share with the TRC and increased distress. This is consistent with a large literature emphasizing such points as the association between trauma exposure and subsequent distress (6), and the failure of trauma debriefing to help decrease levels of distress (15). Such data suggest that no matter how successful the TRC was as a national exercise, individuals exposed to gross human rights violations, and with subsequent psychological distress, may well require additional kinds of social support or psychological intervention. They are consistent with other data emphasizing that some of those who testified before the TRC found it a painful and disempowering experience (10). Similarly, in the absence of other measures, the TRC, on its own, fails to address many troubling issues in South Africa.

Indeed, in the South African context, not surprisingly, race and gender continue to matter. Blacks and males had generally higher levels of participation and more positive perceptions of the TRC than Whites and females. Previous data that males are more forgiving than females were not supported (2;13). On the other hand, it can be noted that race and gender accounted for a relatively small amount of the variance in distress, anger; and forgiveness. Despite markedly different histories during colonial and apartheid times, there are some similarities in the subjective experiences of South Africans of different racial and gender groups, and there are likely to be a number of shared psychological processes.

This analysis is necessarily limited by the relative inability of cross-sectional designs to address issues of causality. Associations of increased levels of distress / anger, and decreased forgiveness, with having attended the TRC, do not necessarily mean that the TRC had a negative effect, but instead may suggest that those with increased levels of distress and decreased forgiveness as a result of events at the core of the TRC (e.g. experience of human rights violations), may have been more likely to choose to attend TRC meetings. Furthermore the extent of the variance accounted for by the TRC variables is low.

Other limitations include the fact that the TRC process played out over several years, so that a single report can be misleading. For example, it was only after this survey was conducted that many of the reparations were formally announced, and such reparations were viewed as insufficient by many South Africans. Only a small number of participants here chose to participate in the TRC, so the analysis of this subgroup has low statistical power. Finally, it should be noted that psychological distress, anger, and forgiveness are broad constructs (and some of the scales had low standardized Cronbach alpha’s), and that individual differences in levels of specific psychopathology also need to be examined in relationship to TRC variables.

In summary, although the TRC may not have met the early expectations of South Africans, overall many people were exposed to the TRC, and participants in our survey had a moderately positive view of the TRC. It needs to be emphasized that the TRC was an imperfect process; in particular perpetrators have not often been brought to book, and victims have not often been adequately compensated. As discussed above, the effects of the TRC at an individual level may have been positive for some, but negative for others. At the same, there appears to be some data to support the view that the South African TRC has been a transparent and effective social process, that may serve as a useful model for similar commissions in other parts of the globe (5).