ArticlePDF Available

Forgiveness and Collective Guilt Assignment to Historical Perpetrator Groups Depend on Level of Social Category Inclusiveness

Authors:

Abstract and Figures

The authors examined how categorization influences victimized group members' responses to contemporary members of a historical perpetrator group. Specifically, the authors tested whether increasing category inclusiveness--from the intergroup level to the maximally inclusive human level--leads to greater forgiveness of a historical perpetrator group and decreased collective guilt assignment for its harmdoing. Among Jewish North Americans (Experiments 1, 2, and 4) and Native Canadians (Experiment 3) human-level categorization resulted in more positive responses toward Germans and White Canadians, respectively, by decreasing the uniqueness of their past harmful actions toward the in-group. Increasing the inclusiveness of categorization led to greater forgiveness and lessened expectations that former out-group members should experience collective guilt compared with when categorization was at the intergroup level. Discussion focuses on obstacles that are likely to be encountered on the road to reconciliation between groups that have a history of conflictual relations.
Content may be subject to copyright.
Forgiveness and Collective Guilt Assignment to Historical Perpetrator
Groups Depend on Level of Social Category Inclusiveness
Michael J. A. Wohl
Carleton University
Nyla R. Branscombe
University of Kansas
The authors examined how categorization influences victimized group members’ responses to contem-
porary members of a historical perpetrator group. Specifically, the authors tested whether increasing
category inclusiveness—from the intergroup level to the maximally inclusive human level—leads to
greater forgiveness of a historical perpetrator group and decreased collective guilt assignment for its
harmdoing. Among Jewish North Americans (Experiments 1, 2, and 4) and Native Canadians (Experi-
ment 3) human-level categorization resulted in more positive responses toward Germans and White
Canadians, respectively, by decreasing the uniqueness of their past harmful actions toward the in-group.
Increasing the inclusiveness of categorization led to greater forgiveness and lessened expectations that
former out-group members should experience collective guilt compared with when categorization was at
the intergroup level. Discussion focuses on obstacles that are likely to be encountered on the road to
reconciliation between groups that have a history of conflictual relations.
Ideally, when an intergroup conflict comes to an end, the neg-
ative emotions directed at members of the out-group would also
cease. History makes clear, however, that the negative feelings
resulting from intergroup conflict linger on long after the violence
itself has terminated. Although peace may be restored, the psy-
chological consequences of the harm done during the conflict can
be felt for generations (Barkan, 2000; Wohl & Branscombe, 2004;
Yehuda et al., 2000). Transgressions committed against members
of the in-group in the past can evoke emotional responses that are
as intense as those for harmdoing committed more recently. Ob-
servers of interethnic conflict report that people speak of atrocities
committed against their group without necessarily differentiating
between events that occurred yesterday, a decade ago, or hundreds
of years ago (Ignatieff, 1997). Emotional responses based on
category membership can traverse generations, with the ancestors
who committed the wrong and contemporary members of the
perpetrator group being linked by a common category member-
ship. When people categorize others on the basis of their group
identity, guilt by association may be assigned to them, without
those individuals bearing any personal responsibility for the wrong
doing (Branscombe, Slugoski, & Kappen, 2004; Heider, 1958;
Hoffman, 1987).
Although more than a half century has passed since the libera-
tion of the Nazi concentration camps, the Holocaust continues to
be of considerable relevance for the identities of both Germans and
Jews. Reminders of the Nazi atrocities affect contemporary Ger-
mans’ political responses (Liu & Hilton, in press; Rensmann,
2004), as well as Jewish perceptions of present-day Germans
(Barkan, 2000; Goldhagen, 1996). Indeed, some believe that vig-
ilance is needed lest contemporary Germans display the same
behavioral tendencies as that of Germans from the Nazi period. For
example, in 1998, the American Jewish Committee opened a
permanent office in Berlin, Germany to “observe German behavior
so that the demons will not again come out of their closet” (Cowell,
1998). Thus, contemporary Germans are likely to be perceived by
Jewish people through a group-based lens that places contempo-
rary Germans and those of the Nazi period within the same frame
and in contrast to Jews.
We present four experiments that examine collective guilt as-
signment and willingness to forgive contemporary members of a
social category as a function of how the historical event and its
participants are categorized. We consider the consequences of
victimized group members perceiving the historical perpetrator
group and their own victimized group as distinct social groups
(e.g., Germans and Jews) or as part of a single, inclusive super-
ordinate category (e.g., humans). Depending on the level of cate-
gorization, differential responses toward contemporary Germans
are likely. In the intergroup context, Germans should be seen as
members of an out-group and be responded to relatively nega-
tively. When, however, categorization is at the more inclusive
human level, Germans should be responded to more positively
because they are included in the same category as the in-group.
Prior research has shown that as a consequence of such shifts in
categorization, affective responses to the same target can dramat-
ically differ (Powell, Branscombe, & Schmitt, in press; Schmitt,
Michael J. A. Wohl, Department of Psychology, Carleton University,
Ottawa, Ontario, Canada; Nyla R. Branscombe, Department of Psychol-
ogy, University of Kansas.
This research was facilitated by a Social Sciences and Humanities
Research Council of Canada Doctoral Fellowship to Michael J. A. Wohl.
Portions of this research were presented at the 2003 meeting of the Society
of Personality and Social Psychology, Los Angeles, CA. We are grateful to
the Jewish Students Associations at the University of Alberta, University of
Kansas, and University of Manitoba, as well as the Department of Native
Studies at the University of Alberta.
Correspondence concerning this article should be addressed to Michael
J. A. Wohl, Department of Psychology, Carleton University, 1125 Colonel
By Drive, B550 Loeb Building, Ottawa, Ontario, Canada K1S 5B6. E-mail:
michael_wohl@carleton.ca or nyla@ku.edu
Journal of Personality and Social Psychology Copyright 2005 by the American Psychological Association
2005, Vol. 88, No. 2, 288 –303 0022-3514/05/$12.00 DOI: 10.1037/0022-3514.88.2.288
288
Silvia, & Branscombe, 2000). We therefore expected that when the
categorization level used was at the group level, Germans would
be perceived as an out-group and historically victimized group
members would assign greater collective guilt and be less willing
to forgive contemporary Germans for the past compared with
when both groups are categorized as belonging to a single inclu-
sive category.
Impact of Social Categorization
Categorization is one of the most fundamental processes under-
lying human judgment. We categorize objects to understand them,
and similarly, we categorize people, including ourselves, to give
meaning to the social environment (Reynolds & Oakes, 2000).
Social categories such as Jew, German, White, and Native are used
because they provide normative information about how category
members are likely to behave and how they should be treated.
Categorizing the self as an in-group member generally results in
in-group favoritism (Brewer, 1979). However, because categori-
zation is inherently variable and tied to changes in context (Turner,
Oakes, Haslam, & McGarty, 1994), whether others are categorized
as alike to or different from the self can differ depending on the
context. According to self-categorization theory (Turner, Hogg,
Oakes, Reicher, & Wetherell, 1987), categorization can occur at
three distinct levels that vary in inclusiveness (i.e., personal, social,
and human). Categorization at the personal level results in the self
being perceived as different from other in-group members. When
categorization occurs at the social identity level, the self is seen as
like other in-group members but as different from out-group mem-
bers. The most inclusive superordinate level reflects the perception
of the self as like other human beings and, potentially, as different
from nonhumans. Thus, more socially shared similarities between
the self and others (who at lower levels of inclusiveness would be
perceived as out-group members) can be expected as the level of
inclusiveness increases (Haslam, Turner, Oakes, McGarty, &
Reynolds, 1998; Turner & Onorato, 1999).
When a specific social identity is salient, conduct should be
stereotypic of in-group norms (Reynolds, Turner, & Haslam,
2000). That is, when people self-categorize as members of a
particular social group, they perceive greater similarity among
those included within the category and greater differences from
those not included in the category. When the in-group has a history
of having been victimized by a particular out-group, negative
perceptions of that perpetrator group can be expected when people
categorize at the social identity level. Accordingly, we propose
that negative perceptions of an out-group with which the in-group
has an antagonistic history should be strongest when that out-
group is differentiated from the in-group. When categorization
occurs at a more inclusive level, in which both the former out-
group and the in-group are included (i.e., the human level), then
negative perceptions of that out-group category should be reduced.
Considerable research testing the common in-group identity
model has demonstrated that recategorizing former out-group
members as in-group members in a more inclusive superordinate
category can reduce intergroup antipathy (Gaertner, Dovidio,
Anastasio, Bachman, & Rust, 1993). For example, in a laboratory
experiment, intergroup bias was reduced when former competing
groups were recategorized as one larger group by means of spa-
tially integrated seating (Gaertner, Mann, Murrell, & Dovidio,
1989). Intergroup bias has also been reduced when competing
groups find complementary areas of expertise (Dovidio, Gaertner,
& Validzic, 1998). Thus, recategorization can reduce intergroup
bias by increasing perceived similarity between former out-group
and in-group members (Gaertner, Mann, Dovidio, Murrell, &
Pomare, 1990; Gaertner, Rust, Dovidio, Bachman, & Anastasio,
1994).
Although there is substantial evidence showing that recategori-
zation can reduce intergroup bias, the support thus far has come
largely from experiments using minimal groups in which partici-
pants evaluated specific in-group and out-group members follow-
ing personal contact with those individuals (e.g., Dovidio et al.,
1998; Gaertner et al., 1989; Nier et al., 2001). In the case of
historical victimization, however, contemporary members of the
victimized group often have had little or no contact with either the
perpetrators or their descendents. Evaluations of the perpetrator
group as a whole, rather than of particular individuals that partic-
ipants have interacted with, have not been investigated. The cur-
rent experiments were thus designed to experimentally assess
whether increasingly inclusive categorization can influence emo-
tional reactions toward contemporary members of the perpetrator
group as a whole among members of a historically victimized
group. Moreover, we assessed the consequences of including con-
temporary Jews and Germans in a single superordinate category
that encompasses all possible subgroups (i.e., humans). We hy-
pothesized that such inclusive categorization should be critical for
understanding when forgiveness of a former perpetrator group as a
whole can be expected and when collective guilt assignment will
be reduced.
Collective Guilt and Forgiveness
People feel personally guilty when the self is perceived as
having responsibility for violating a moral standard (Baumeister,
Stillwell, & Heatherton, 1994; Tangney & Fischer, 1995). Moving
beyond the individual, Smith (1993, 1999) proposed that people
are able to experience emotions on behalf of their group. Guilt at
the collective identity level can be experienced when people cat-
egorize themselves in terms of a group that they perceive has
committed wrongs against another group (Branscombe & Miron,
2004). Experiencing collective guilt for one’s group’s harmful
actions for which the personal self played no role depends on
perceiving a categorical association between the self and the
in-group that committed those actions (Branscombe, Doosje, &
McGarty, 2002). How much identity distress and guilt is felt
depends on how much the in-group is seen as deviating from
accepted moral standards (Bizman, Yinon, & Krotman, 2001).
When people perceive another group to have violated moral
standards, collective guilt may be assigned to members of that
social group. That is, people may expect members of a group to
feel remorse for their group’s past actions. Collective guilt assign-
ment therefore involves a prescriptive desire for all members of the
perpetrator group to feel remorse for their group’s actions. We
hypothesize that collective guilt assignment is most likely when
members of the out-group are seen in terms of the category
membership responsible for the immoral behavior. Historically
victimized group members may be especially likely to perceive
contemporary and past members of their perpetrator group as a
single category. As a result, all members of the perpetrator group
289
FORGIVENESS AND COLLECTIVE GUILT
may be expected to experience collective guilt when categorized in
intergroup terms. Although Germans born after World War II
clearly cannot be assigned any personal guilt for the Holocaust, as
they could not have played any role in that event, collective guilt
may still be assigned to Germans as a whole—to the extent that
Germans are categorized as a perpetrator group that is differenti-
ated from the in-group. Thus, Jewish people may assign collective
guilt to Germans as a whole when they categorize this former
perpetrator group as separate from their in-group.
Expecting a group to feel collective guilt implies that members
of that category have not been forgiven for their ancestors’ past
immoral actions. At the personal level, forgiveness involves relin-
quishing negative feelings toward the transgressor (Boon & Sul-
sky, 1997; Enright & North, 1998; Exline & Baumeister, 2000;
McCullough, Pargament, & Thoresen, 2000; Sandage, Worthing-
ton, Everett, & Hight, 2000). Such forgiveness permits the rela-
tionship between the conflicting parties to move forward following
the occurrence of a transgression (Minow, 1998; Zechmeister &
Romero, 2002). Forgiving a transgressor has repeatedly been
shown to result in less turmoil in interpersonal relationships and
more positive emotional reactions compared with not forgiving a
transgressor (McCullough, Bellah, Kilpatrick, & Johnson, 2001;
McCullough & Hoyt, 2002). Conversely, hostility, vengefulness,
and increased rumination tend to be greater in those who fail to
forgive their transgressor (McCullough et al., 2001). At the inter-
group level, forgiveness precludes harboring negative feelings
toward the perpetrator category as a whole (Tutu, 1999). By
forgiving the perpetrator category, the negative feelings associated
with a category of people (e.g., Germans as a whole) should be
lessened.
On the basis of self-categorization theory and the common
in-group identity model (Gaertner, Dovidio, Nier, Ward, &
Banker, 1999; Turner et al., 1987), we hypothesized that making a
more inclusive human identity salient among members of a vic-
timized group would decrease the degree of collective guilt as-
signed to contemporary members of the perpetrator group. Will-
ingness to forgive the perpetrator category for the past actions of
their group should also depend on level of categorization. Cate-
gorization at the more inclusive human identity level should result
in participants being more willing to forgive compared with cat-
egorization at the group level.
Our research considers the route by which shifts in level of
categorization among victimized group members might affect the
assignment of collective guilt and willingness to forgive perpetra-
tor group members. We propose that at the human level of cate-
gorization, people perceive genocide as less unique to Germans;
that is, genocide is seen as a more pervasive phenomenon among
the inclusive category— humans. To the extent that genocide is
seen as not something done uniquely by Germans against Jews, the
specific category of Germans should be seen as more deserving of
forgiveness and less deserving of the burden of guilt compared
with when genocide is an outcome that Germans alone have
perpetrated. To the extent that many human groups are perceived
to have committed similar harmful acts, one specific group (e.g.,
Germans) should not be singled out as particularly immoral. For
this reason, we predicted that when categorization shifts from the
intergroup level to a more inclusive level, genocide would be seen
as a more pervasive occurrence throughout human history. As
genocide is perceived to be more pervasive, there should be a
subsequent increase in willingness to forgive and a decrease in the
collective guilt assigned. Thus, we propose that perceived perva-
siveness of genocide mediates the effect of categorization level on
both collective guilt assignment and willingness to forgive con-
temporary members of a group with a history of having harmed the
in-group.
The model we tested is consistent with both self-categorization
theory and the common in-group identity model, in which shifts in
social categorization from the intergroup level to a more inclusive
or superordinate level lead to less negative perceptions of former
out-group members. In contrast to previous research on recatego-
rization, however, we studied responses among contemporary
members of a historically victimized group to members of a social
category who were not personally responsible for the injustice but
whose ancestors were responsible for one of the most heinous
crimes in human history. We provide empirical tests of our pro-
posed mediating mechanism (i.e., pervasiveness of genocide) by
which such shifts in categorization could reduce negative reactions
toward the perpetrator category as a whole. Moreover, we assess
the effect of such shifts in level of inclusiveness of categorization
for variables that are central to conflict resolution (e.g., willingness
to forgive and collective guilt assignment).
Overview
We conducted four experiments to test our hypotheses. In the
first two studies, which had Jewish participants, the Holocaust was
framed as either a German-Jewish intergroup event or as an event
reflecting something more general about human behavior. We
considered the consequences of shifting levels of inclusiveness—
from the social to the human level—for both forgiveness of con-
temporary Germans and collective guilt assignment to them. Using
two different methods, we manipulated how the Holocaust was
categorized: as reflecting what humans have done to other humans
or as reflecting what Germans did to Jews. We expected that Jews
would be more willing to forgive Germans for the past when they
categorized at the human identity level and that the guilt assigned
to contemporary Germans would be lower in the human identity
condition compared with the social identity condition. Further, we
hypothesized that the effect of inclusive categorization on willing-
ness to forgive and collective guilt assignment would be mediated
by perceived pervasiveness of the harmful actions. The more
pervasive such harmful acts are perceived to be, the less harshly
the specific salient perpetrator category should be judged.
In the third experiment, we sought to generalize this inclusive-
ness effect to a victimized group with a different history and
present day conditions. For centuries, White Canadians have mis-
treated Native Canadians. Indeed, the injustices experienced by
Native Canadians remain grim realities. As Russell (1998) re-
ported, the average Native Canadian’s income is two thirds of the
national average, fewer than 50% of Native Canadian homes have
sewer and water connections, almost 50% of Native Canadian
families live in overcrowded housing, and among Native Canadi-
ans living on reservations, 60% live below the poverty line. Thus,
Native Canadians are a population with a history of victimization
by White Canadians that continues in the present. In Experiment 3,
we assessed whether increasingly inclusive categorization can
increase forgiveness and reduce collective guilt assignment among
people whose victimization is ongoing.
290
WOHL AND BRANSCOMBE
In Experiment 4, we once again examined Jewish people’s
reactions to increasing the inclusiveness of categorization to test
two alternative mediating processes. In this experiment, we as-
sessed whether either differentiation between groups (contempo-
rary Jews and contemporary Germans) or differentiation within the
perpetrator group (contemporary Germans and Germans of the
Nazi era) can explain the effect of the categorization manipulation
on collective guilt assignment and forgiveness. In addition, Exper-
iment 4 provided an opportunity to assess additional outcome
variables that may flow from the intergroup emotional responses
of forgiveness and collective guilt. In particular, we sought to
determine whether the categorization manipulation would influ-
ence attempts to maintain social distance from individual members
of the historical perpetrator group (e.g., willingness to be friends
with a member of the perpetrator group) and social distance from
symbols of the perpetrator group as a whole (e.g., willingness to
buy products produced by the perpetrator group).
Experiment 1
Method
Participants
Jewish participants were contacted via the Internet through the Hillel
(i.e., Jewish Students Association) e-mail list at the University of Kansas.
Hillel members were asked to access a Website to fill out a short ques-
tionnaire concerning Jews’ perceptions of the Holocaust. Forty-seven par-
ticipants (26 men; 21 women) responded by accessing the Website and
completing the questionnaire. Participants ranged in age from 18 to 47
years (M 21.67, SD 4.56).
Procedure and Design
A link to an online questionnaire was provided to potential Jewish
participants in an e-mail sent to the Hillel members. This link directed the
participants to an online consent form and then randomly assigned each to
either the social (n 22) or human (n 25) identity salience condition.
In all conditions, participants were first asked to reflect on their views
concerning the Holocaust. For those participants assigned to the social
identity condition, the Holocaust was described as an event in which
Germans behaved aggressively toward Jews. In contrast, for those partic-
ipants assigned to the human identity condition, the Holocaust was de-
scribed as an event in which humans behaved aggressively toward other
humans.
Participants were then asked to indicate their agreement with a series of
Likert-type statements by clicking on their selected response option using
a1(definitely disagree)to8(definitely agree) scale. Four items assessed
willingness to forgive contemporary Germans for the Holocaust. These
items were “Germans today should be forgiven for what their group did to
Jews during World War II,” “Jews should move past their negative feelings
toward today’s Germans for the harm their group inflicted during World
War II,” “Today’s Germans should be forgiven for what their ancestors did
to Jews during World War II,” and “It is possible for me to forgive today’s
Germans for the Holocaust.” Four items, based on the construct validated
by Branscombe et al. (2004), assessed the degree to which Jews believe
collective guilt for the Holocaust should be assigned to contemporary
Germans. These items were “Germans of today should feel guilty about the
bad things that happened to Jews during World War II,” “Today’s Germans
should feel guilty about the awful things their ancestors did to Jews in
World War II,” “Germans today should feel regret for what their group did
to Jews during World War II,” and “All Germans should feel guilty about
the harm done to Jews during World War II.” In addition, three items
assessed perceived pervasiveness of genocide: “Harmful actions such as
those during the Holocaust have happened throughout human history,”
“There have been many similar instances of mass killing throughout human
history,” and “The Germans targeting Jews in World War II is similar to
many other examples of mass killings of people that have happened
throughout human history.” Last, to assess the effectiveness of the cate-
gorization manipulation, participants were asked to indicate the extent to
which “Germans and Jews share basic similarities” anchored at 1 (strongly
disagree)and8(strongly agree).
Results
Preliminary Analyses
Because there were neither participant gender effects on any of
the dependent measures nor interactions between participant gen-
der and condition, we collapsed across this variable. Table 1
provides a summary of the means and standard deviations for all of
the measured variables by condition. Table 2 presents the corre-
lations between the measured variables and the categorization
manipulation.
Experimental Effects
Manipulation check. The inclusiveness manipulation was as-
sessed in terms of perceived similarity between Jews and Germans.
In the social identity condition, we expected relatively few simi-
Table 1
Means and Standard Deviations for Measured Variables, Experiments 1– 4
Variable
Experiment 1 Experiment 2 Experiment 3 Experiment 4
Human Social Human Social Human Social Human Social
Pervasiveness of genocide
M 6.33 3.55 5.63 3.67 6.95 4.84 5.51 4.05
SD 1.36 1.84 1.82 1.58 1.37 1.21 2.10 1.45
Collective guilt assignment
M 5.47 6.75 3.70 4.97 5.08 6.28 4.23 5.49
SD 2.06 0.74 1.83 1.56 1.79 1.70 1.72 1.42
Willingness to forgive
M 5.84 4.52 5.67 4.63 5.55 4.05 5.47 4.56
SD 1.25 0.92 1.45 1.53 1.30 1.63 1.85 1.33
Note. Within a given experiment, all pairs of means are significantly different between conditions, ps .05.
291
FORGIVENESS AND COLLECTIVE GUILT
larities between the two groups to be perceived, and in the human
identity condition, we expected participants to perceive their in-
group and the Germans as relatively similar. As expected, in the
intergroup categorization condition, participants perceived Ger-
mans and Jews as somewhat dissimilar groups (M 3.59, SD
2.20), whereas they perceived the two groups as relatively similar
in the human categorization condition (M 6.36, SD 1.85), F(1,
45) 22.05, p .01, d 1.36. Thus, we were rather successful
in varying the perceived similarity between the groups as a func-
tion of the inclusiveness of categorization manipulation.
Pervasiveness of genocide. An overall pervasiveness of geno-
cide score was calculated by averaging the three items assessing
this perception (
.95). One-way analysis of variance (ANOVA)
yielded a significant effect of condition, F(1, 45) 35.59, p .01,
d 1.71. In the social identity condition, genocide was seen as
relatively unique to Germans (M 3.55, SD 1.84), whereas in
the human identity case, it was seen as pervasive and not unique to
Germans (M 6.33, SD 1.36).
Assignment of collective guilt. A mean of the four collective
guilt items was calculated to create an overall collective guilt
assignment score (
.95). Participants assigned significantly less
collective guilt to Germans when the more inclusive human-level
categorization was salient (M 5.47, SD 2.06) than they did
when categorization was at the social identity level (M 6.75,
SD 0.74), F(1, 45) 7.62, p .01, d 0.83.
Willingness to forgive. An overall willingness to forgive con-
temporary Germans score was created by averaging the four for-
giveness items (
.76). Participants were more willing to forgive
Germans when the human level of identity was salient (M 5.84,
SD 1.25) than they were when categorization was at the social
identity level (M 4.52, SD 0.92), F(1, 45) 16.55, p .01,
d 1.20.
Mediational Analysis for Collective Guilt
To determine whether the effect of our manipulation on collec-
tive guilt could be explained by perceived pervasiveness of geno-
cide, we used Baron and Kenny’s (1986) regression procedure for
testing mediation. Because we know from the ANOVAs that the
categorization manipulation reliably affected perceived pervasive-
ness of genocide and collective guilt acceptance, we proceeded to
test the full model. Thus, the categorization manipulation variable
(dummy coded as 1 in the social identity condition and 2 in the
human identity condition) and perceived pervasiveness of geno-
cide were included in a regression equation with collective guilt
assignment as the dependent variable, R
2
.23, F(2, 44) 6.42,
p .01. The coefficient associated with perceived pervasiveness
of genocide significantly predicted collective guilt assignment,
⫽⫺.38, t(45) ⫽⫺2.15, p .04, whereas the categorization
manipulation was no longer a significant predictor,
⫽⫺.13,
t(45) ⫽⫺.72, p .47. We then conducted a Sobel test to
determine whether the indirect effect of the categorization manip-
ulation on collective guilt assignment, by perceived pervasiveness
of genocide, was significantly different than zero. This test was
significant (z 2.03, p .05).
Mediational Analysis for Willingness to Forgive
We repeated the mediational analysis described for collective
guilt with willingness to forgive Germans as the dependent vari-
able. Because we know from the ANOVAs that the categorization
variable reliably affected perceived pervasiveness of genocide and
forgiveness, we proceeded to test the full model. Thus, the cate-
gorization manipulation variable and perceived pervasiveness of
genocide were entered into a regression model with willingness to
forgive as the dependent variable, R
2
.39, F(2, 44) 13.79, p
.01. The coefficient associated with perceived pervasiveness of
genocide was significant,
.46, t(45) 2.89, p .01, and the
categorization manipulation no longer significantly predicted will-
ingness to forgive,
.22, t(45) 1.36, p .18. We then
conducted a Sobel test to determine whether the indirect effect of
the categorization manipulation on willingness to forgive, by per-
ceived pervasiveness of genocide, was significantly different than
zero. This test was significant (z 2.60, p .01).
Discussion
Experiment 1 assessed the effect of level of inclusiveness on the
perceived pervasiveness of genocide, forgiveness of Germans, and
assignment of collective guilt to contemporary Germans. We pre-
dicted that the two levels of inclusiveness— humans or distinct
social groups—would have differential consequences for the as-
signment of collective guilt and willingness to forgive Germans for
the past by affecting the perceived pervasiveness of genocide. We
demonstrated that when the human level of identity was salient,
Jewish participants were less inclined to assign collective guilt and
more willing to forgive Germans for the Holocaust than they were
when social identity was salient. This effect was, as expected, fully
mediated by how widespread the harmful actions were seen to be.
When perceived as relatively unique to Germans, the actions were
not forgiven and considerable guilt was assigned, but when geno-
cide was seen as pervasive across human societies, Jewish partic-
ipants were more willing to forgive and less inclined to assign guilt
to contemporary Germans.
For intergroup forgiveness to take place, South Africa’s Arch-
bishop Tutu (1999) has argued that it is necessary to categorize the
Table 2
Correlations for Measured Variables for Experiments 1–3
Variable 1 2 3 4
Experiment 1
1. Categorization
2. Pervasiveness of genocide .67**
3. Forgiveness .52** .60**
4. Collective guilt assignment .38** .47** .50**
Experiment 2
1. Categorization
2. Pervasiveness of genocide .51**
3. Forgiveness .34* .63**
4. Collective guilt assignment .36* .52** .53**
Experiment 3
1. Categorization
2. Pervasiveness of harm .64**
3. Forgiveness .46** .63**
4. Collective guilt assignment .33* .50** .60**
Note. Categorization was coded 1 social identity condition and 2
human identity condition. All measured variables ranged on a scale from 1
to 8.
* p .05. ** p .01, two-tailed.
292
WOHL AND BRANSCOMBE
perpetrator and victim groups within a single inclusive national
group. During the Truth and Reconciliation Commission hearings,
which were aimed at examining the past as a basis for creating a
new future for South Africa, he reported perceiving that both
Blacks and Whites “belong in the one family, God’s family, the
human family” (p. 265). The results of the present experiment
confirm the causal effect of such inclusive categorization on for-
giveness and also demonstrate that this effect depends on shifts in
the perceived pervasiveness of the harmdoing. The effect of a more
inclusive categorization that includes Germans and Jews in the
same superordinate category (i.e., humans) on willingness to for-
give and on the desire for Germans to experience collective guilt
depended on whether genocide was perceived as something
uniquely done by Germans. That is, when people perceived the
Holocaust as due to the nature of Germans rather than human
nature, they were less likely to forgive Germans and they were
more likely to want contemporary Germans to experience collec-
tive guilt.
Although these results provide support for our major hypothe-
ses, the manipulation we used is open to an alternative interpreta-
tion. In Experiment 1, we manipulated level of categorization by
framing the Holocaust as either a human tragedy or one perpe-
trated by Germans. As a result, our manipulation may have implied
to participants that they should assign more guilt and be less
forgiving of Germans in the social identity condition than in the
human condition. By stating that the Holocaust was an event in
which Germans behaved aggressively toward Jews, it might have
linguistically suggested that Germans are particularly blamewor-
thy, whereas in the human identity condition, in which no refer-
ence was made to the specific perpetrator group, German blame
was not implied. Because the effect of this manipulation might be
attributable to differential linguistic reference to Germans, in Ex-
periment 2 we tested the same hypotheses using a manipulation of
level of categorization in which this linguistic possibility was
absent.
Experiment 2
Method
Participants
Jewish participants were contacted via the Internet through the Jewish
Students Association e-mail list at the University of Alberta, Edmonton,
Alberta, Canada. As in Experiment 1, participants were asked to fill out a
short questionnaire concerning Jews’ perceptions of the Holocaust. Thirty-
seven participants (19 men, 18 women) responded by accessing the Web-
site and completing the questionnaire. Participants ranged in age from 17
to 53 years (M 22.06, SD 5.96).
Design and Procedure
As in Experiment 1, participants were randomly assigned to one of two
categorization conditions (social vs. human identity) when they accessed
the Website. Again, all participants were told that they would be asked
questions concerning their perceptions of the Holocaust. The human iden-
tity condition (n 19) was identical to that used in Experiment 1. In this
condition, the Holocaust was described as an event that demonstrated how
humans had behaved aggressively toward other humans. In the new social
identity condition (n 18), participants also received this human descrip-
tion of the Holocaust but, to induce social-level categorization, participants
in the social identity condition were then asked to indicate whether they
were Jewish or non-Jewish and whether they were German or non-German
in origin. Thus, in Experiment 2, blame was no longer linguistically
implied in the social identity condition, although the two social catego-
ries—Germans versus Jews—were made salient. All participants indicated
they were Jewish and non-German.
Participants were then asked to indicate their agreement with a series of
Likert-type statements by clicking on their selected response option using
a1(definitely disagree)to8(definitely agree) scale. The dependent
measures for Experiment 2 were identical to those used in Experiment 1.
Four items assessed the degree to which Jews assigned collective guilt to
contemporary Germans for the Holocaust, four items assessed willingness
to forgive Germans, and three items assessed perceived pervasiveness of
genocide. The effectiveness of the categorization manipulation was again
assessed in terms of perceived similarity between Germans and Jews.
Results
Preliminary Analyses
Because there were neither participant gender effects on any of
the dependent measures nor significant interactions between par-
ticipant gender and condition, we collapsed across this variable.
Table 1 provides a summary of the means and standard deviations
for all of the measured variables by condition. Table 2 presents the
correlations between the measured variables and the categorization
manipulation.
Experimental Effects
Manipulation check. As was the case for Experiment 1, the
success of the categorization manipulation was assessed in terms
of the extent to which Germans were seen as sharing basic simi-
larities with Jews. In the human categorization condition, Jewish
participants perceived Germans and Jews as sharing more similar-
ities (M 5.84, SD 1.17) than did participants in the social
identity condition (M 4.67, SD 1.97), F(1, 45) 4.94, p
.03, d 0.72. Thus, this version of the categorization manipulation
was also successful at varying perceived similarity between the
groups.
Perceived pervasiveness of genocide. We predicted that peo-
ple would be more willing to see the Holocaust as one of many
instances of mass violence throughout human history when their
human identity was salient compared with when their social iden-
tity was salient. As was the case for Experiment 1, an overall mean
rating was calculated for the three pervasiveness of genocide items
(
.91). Level of categorization significantly affected perceived
pervasiveness of genocide, F(1, 35) 12.22, p .01, d 1.14.
As expected, when the human level of categorization was salient,
participants perceived genocide to be more pervasive (M 5.63,
SD 1.82) than they did when their social identity was salient
(M 3.67, SD 1.58).
Assignment of collective guilt. It was expected that Jewish
participants would assign more collective guilt to Germans in the
intergroup categorization condition compared with when the
human-level categorization was salient. To test this hypothesis, we
calculated an overall assignment of collective guilt score by aver-
aging the four assignment of collective guilt items (
.93). As
expected, when the intergroup level of categorization was salient,
Jews assigned more collective guilt to Germans (M 4.97, SD
293
FORGIVENESS AND COLLECTIVE GUILT
1.56) than they did when the human level was salient (M 3.70,
SD 1.83), F(1, 35) 5.19, p .03, d 0.75.
Willingness to forgive. An overall forgiveness score was cal-
culated by averaging the four forgiveness items (
.89). As
predicted, when the human level of categorization was used, Jews
were more willing to forgive contemporary Germans (M 5.67,
SD 1.45) than they were when the social level of categorization
was salient (M 4.63, SD 1.53), F(1, 35) 4.57, p .04, d
0.72.
Mediational Analysis for Collective Guilt
As in Experiment 1, we used Baron and Kenny’s regression
procedure for testing mediation. As the ANOVA indicated, the
categorization manipulation (coded as 1 social identity, 2
human identity) significantly predicted collective guilt acceptance
and perceived pervasiveness of genocide. When both the catego-
rization manipulation and perceived pervasiveness of genocide
variable were entered, the regression equation accounted for sub-
stantial variance in collective guilt assignment, R
2
.28, F(2,
34) 6.71, p .01. Perceived pervasiveness of genocide signif-
icantly predicted collective guilt,
⫽⫺.46, t(35) ⫽⫺2.70, p
.02, and the categorization manipulation was no longer significant,
⫽⫺.13, t(35) ⫽⫺.76, p .45. We then conducted a Sobel test
to determine whether the indirect effect of the categorization
manipulation on collective guilt assignment, by perceived perva-
siveness of genocide, was significantly different than zero. This
test was significant (z 2.14, p .04).
Mediational Analysis for Willingness to Forgive
The ANOVAs demonstrated that the categorization variable
reliably affected perceived pervasiveness of genocide and forgive-
ness. The mediational model was tested where the categorization
manipulation and perceived pervasiveness of genocide were both
entered into a regression model with willingness to forgive as the
dependent variable, R
2
.40, F(2, 34) 11.18, p .01. The
coefficient associated with perceived pervasiveness of genocide
was a significant predictor of willingness to forgive,
.61,
t(35) 3.94, p .01. The categorization manipulation variable,
however, was no longer a significant predictor of willingness to
forgive,
.03, t(35) 0.17, p .86. We then conducted a
Sobel test to determine whether the indirect effect of the catego-
rization manipulation on willingness to forgive, by perceived per-
vasiveness of genocide, was significantly different than zero. This
test was significant (z 2.63, p .01).
Discussion
The results of Experiment 1 were replicated with a different
manipulation of level of categorization. The effects of categoriza-
tion level on collective guilt assignment and willingness to forgive
Germans depended on shifts in the pervasiveness of genocide.
Encouraging Jews to categorize in human terms (i.e., we are all
members of the same group, humans) promoted forgiveness of
Germans and lowered collective guilt assignment by means of
changes in the perceived pervasiveness of genocide. These results
suggest that placing the harm committed in the past in a broader
historical context can indeed lead the harmed group to be more
willing to forgive and assign less collective guilt to contemporary
members of the perpetrator group.
Within the context of the Holocaust, the critical element for
Jewish perceivers appears to be whether Germans as a whole are
considered an antagonistic out-group in relation to the in-group or
are considered part of a shared superordinate category. In the
maximally inclusive human group case, Germans and Jews were
perceived as sharing substantial similarities, which dictates greater
forgiveness for the transgressions of the past compared with seeing
the groups as distinct and therefore deserving of differential
treatment.
These first two studies provide support for the hypothesis that
members of a victimized group can forgive contemporary mem-
bers of their former perpetrator group when a more inclusive level
of categorization is used. The research thus far, however, has
considered a single victimized group (Jews) and perpetrator group
(Germans) and responses to a particularly heinous event (the
Holocaust). An important question remains as to whether the
categorization effects we observed are specific to this particular
group-based trauma that ended more than a half century ago.
Specifically, can level of inclusiveness affect the responses of
other victimized groups, particularly those for whom the harmdo-
ing is still ongoing? Many social groups continue to be discrimi-
nated against by their historical perpetrator group. We address
whether shifts in level of categorization influence willingness to
forgive and assignment of collective guilt in a group that continues
to be discriminated against by their historical perpetrator group.
Experiment 3
In Experiment 3, we tested our hypotheses with a different
historically victimized group—Native Canadians. Canadian his-
tory is replete with accounts of Natives being murdered, tortured,
forcibly sterilized, and other crimes against humanity (Downey,
1999; Frideres, 1998; George, Kuhn, & Sweetman, 1996;
Jankowski & Moazzami, 1994). As in the United States, Native
lands were confiscated by White Canadians during the 19th cen-
tury through a series of one-sided treaties (Churchill, 1997; Fride-
res, 1998). Thus, much of Canadian history illustrates how White
people have victimized Native peoples.
Although Canadian social policy over the past half century has
removed official legal barriers against Native Canadians, they
remain a highly disadvantaged group. Native Canadians suffer
from lower levels of employment and are less likely to hold
managerial and professional jobs than are nonindigenous Canadi-
ans (Lock Kunz, Milan, & Schetagne, 2000). Thus, discrimination
against Native Canadians is ongoing, with much harm continuing
in the present. To test whether shifts in categorization influence
willingness to forgive and the assignment of collective guilt to
members of a current perpetrator group, we assessed its effect
among Native Canadians.
Method
Participants
Native Canadian students (14 men; 24 women) were asked to complete
a questionnaire about the treatment of Natives in Canada. A non-Native
experimenter solicited participants in three Department of Native Studies
third-year courses at the University of Alberta. These Native Canadian
294
WOHL AND BRANSCOMBE
participants had lived their lives in highly integrated settings (mean per-
centage of neighborhood White 65.59, SD 36.34; mean percentage of
Whites in school 85.67, SD 16.67; mean percentage of White
friends 45.11, SD 35.70). Participants ranged in age from 20 to 36
years (M 26.24, SD 4.24).
Procedure
Participants in each class were randomly assigned to one of two condi-
tions (social identity, n 19; human identity, n 19). In both conditions,
participants were first asked to reflect on their views concerning the
treatment of Natives in Canadian history. The treatment of Native people
was described to all participants as demonstrating how humans have
behaved aggressively toward other humans. In the social identity condition,
we then requested participants to simply check a box indicating whether
they were Native or non-Native and whether they were White or non-White
in origin. All participants indicated they were Native and non-White.
Participants then indicated their agreement with a series of items using
a1(definitely disagree)to8(definitely agree) scale. The four items that
assessed willingness to forgive (
.93) and the four items that assessed
collective guilt assignment (
.95) were identical to those used in the
previous two studies, although the target groups were changed to reflect
Native-White relations in Canada. In addition, we altered the perceived
pervasiveness of genocide items to reflect perceived pervasiveness of
intergroup harm (
.90). Last, as a check on the categorization manip-
ulation, participants were asked to indicate the extent to which they
perceived White and Native Canadians as sharing basic similarities.
Results
Preliminary Analyses
Because there were neither participant gender effects on any of
the dependent measures nor significant interactions between par-
ticipant gender and condition, we collapsed across this variable.
Table 1 provides a summary of the means and standard deviations
for all of the measured variables by condition. Table 2 presents the
correlations between the measured variables and the categorization
manipulation.
Experimental Effects
Manipulation check. The manipulation was once again suc-
cessful. Native Canadians in the human categorization condition
perceived Whites and Natives as sharing more basic similarities
(M 6.57, SD 1.02) than did participants in the intergroup
categorization condition (M 5.42, SD 0.96), F(1, 36) 13.00,
p .01, d 1.16.
Perceived pervasiveness of intergroup harm. The categoriza-
tion manipulation had a significant effect on perceived pervasive-
ness of intergroup harm, F(1, 35) 25.20, p .01, d 1.63. As
expected, Native Canadians perceived intergroup harm to be more
pervasive when categorization was at the human level (M 6.95,
SD 1.37) than when social categorization was induced (M
4.84, SD 1.21). As predicted, Native Canadian participants were
more willing to see the harm done by White Canadians as consis-
tent with the harm that many groups have committed against each
other when their shared identity with the perpetrator group was
salient.
Assignment of collective guilt. When the intergroup level of
categorization was salient, Native Canadians assigned more col-
lective guilt to White Canadians (M 6.28, SD 1.70) than they
did when the human level was salient (M 5.08, SD 1.79), F(1,
36) 4.49, p .05, d 0.69. Thus, when members of a
victimized group include members of a perpetrator group within a
more inclusive category that the in-group also shares, collective
guilt assignment decreases.
Willingness to forgive. Native Canadians were more willing to
forgive contemporary members of the perpetrator group when
human identity (M 5.55, SD 1.30) rather then social identity
was salient (M 4.05, SD 1.63), F(1, 36) 9.82, p .01, d
1.02. Thus, we were able to replicate our forgiveness results
concerning the impact of categorization among members of a
group that continues to experience considerable discrimination.
Categorization at the more inclusive human level led the victim-
ized group to be more forgiving of the perpetrator group.
Mediational Analysis for Collective Guilt
Baron and Kenny’s regression procedure was once again used to
test for mediation. Because we know from the ANOVAs that the
categorization variable reliably affected perceived pervasiveness
of intergroup harm and collective guilt assignment, we proceeded
to test the mediational model. Thus, the categorization manipula-
tion variable (dummy coded as 1 in the social identity condition
and 2 in the human identity condition) and perceived pervasive-
ness of intergroup harm were both entered as predictors of collec-
tive guilt assignment, R
2
.25, F(2, 35) 5.78, p .01.
Perceived pervasiveness of intergroup harm significantly predicted
collective guilt assignment,
⫽⫺.48, t(37) ⫽⫺2.53, p .02.
The categorization manipulation variable, however, no longer pre-
dicted collective guilt assignment,
⫽⫺.02, t(37) ⫽⫺.12, p
.90. We then conducted a Sobel test to determine whether the
indirect effect of the categorization manipulation on collective
guilt assignment, by perceived pervasiveness of intergroup harm,
was significantly different than zero. This test was significant (z
2.26, p .03).
Mediational Analysis for Willingness to Forgive
The categorization manipulation variable and perceived perva-
siveness of genocide were entered into a regression model with
willingness to forgive as the dependent variable, R
2
.40, F(2,
35) 11.57, p .01. The coefficient associated with perceived
pervasiveness of intergroup harm remained a significant predictor
of willingness to forgive,
.56, t(36) 3.27, p .01. The
categorization manipulation variable, however, was not a signifi-
cant predictor of willingness to forgive,
.10, t(36) 0.61, p
.54. We then conducted a Sobel test to determine whether the
indirect effect of the categorization manipulation on willingness to
forgive, by perceived pervasiveness of intergroup harm, was sig-
nificantly different than zero. This test was significant (z 2.74,
p .01).
Discussion
Experiment 3 replicated the results of the previous two experi-
ments with a different historically victimized group. The effect of
categorization level on collective guilt assignment and willingness
to forgive White Canadians depended on shifts in the pervasive-
ness of intergroup harmdoing. When Native Canadians categorized
295
FORGIVENESS AND COLLECTIVE GUILT
at the human level of identity, forgiveness of White Canadians was
increased and collective guilt assignment decreased by increasing
perceptions of the pervasiveness of group-based harmful actions.
These results suggest that shifts in the level of categorization can
influence willingness to forgive and assignment of collective guilt
in a group that continues to be mistreated by their historical
perpetrator group. Placement of the current harm in a broad his-
torical context led the harmed group to be more willing to forgive
and assign less collective guilt to contemporary members of the
perpetrator group.
Experiment 4
The purpose of Experiment 4 was to test alternative possible
mediators for the effect of categorization on willingness to forgive
and collective guilt assignment. First, we considered that because
of the common group membership inherent in human level cate-
gorization, Jews might be less likely to differentiate Germans from
Jewish people. When differences between groups are diminished,
by superordinate categorization, intergroup hostility is reduced
(Gaertner et al., 1993). Thus, to the extent that Germans as a whole
are not seen as different from Jews, the people that comprise the
German category might be forgiven more readily and assigned less
guilt. We also considered the possibility that human-level catego-
rization could result in differentiation between contemporary Ger-
mans and Germans of the Nazi era. According to social identity
theory (Tajfel & Turner, 1986), people are motivated to maintain
a positive view of the group to which they belong. When human-
level categorization is salient, people may attempt to maintain a
positive view of humanity by distancing the subset of people
deemed to have committed inhuman acts. Such a desire to maintain
a positive view of humanity might result in differentiation between
Germans of today and Germans of yesteryear. If so, such perceived
differentiation might mediate the effect of categorization level on
both collective guilt assignment and willingness to forgive con-
temporary members of a historical perpetrator group.
Thus, two potential alternative mediators were examined: (a)
differentiation between members of the historically victimized
in-group and the perpetrator group as a whole and (b) differenti-
ation between contemporary members of the perpetrator group and
perpetrator group members from the Nazi era. To test the viability
of these two potential mediators, we once again examined Jewish
people’s emotional reactions to contemporary Germans in condi-
tions that vary in level of inclusiveness.
Experiment 4 also allowed us to examine whether shifts in
categorization influence behavioral tendencies toward the perpe-
trator category. Smith (1993, 1999) has suggested that intergroup
emotions can play a substantial role in shaping intergroup behav-
ior. Thus, we assessed whether shifts in categorization would
influence contemporary Jews’ behavior toward contemporary Ger-
mans. We also considered the route by which shifts in level of
categorization among victimized group members might affect
behavioral tendencies toward the perpetrator group. Recently,
Yzerbyt, Dumont, Wigboldus, and Gordijn (2003) demonstrated
that behavioral willingness to intervene on behalf of a victim was
greatest when the victims were perceived to be members of the
same group as the perceivers. The intergroup emotional reactions
evoked fully mediated the effect of categorization on willingness
to intervene. In the spirit of Smith’s intergroup emotion model
(1993, 1999), and building on Yzerbyt et al.’s (2003) findings, we
posit that shifts in categorization influence behavioral tendencies
toward a historical perpetrator group and that changes in behav-
ioral tendencies are mediated by emotional reactions such as
collective guilt assignment and forgiveness.
Two types of behavioral tendencies were examined: (a) social
distance between members of the victimized group and individual
members of the perpetrator group (e.g., willingness to be friends
with a member of the perpetrator group) and (b) social distance
between members of the victimized group and symbols of the
perpetrator group as a whole (e.g., willingness to buy products
produced by the perpetrator group).
Method
Participants
Jewish participants were contacted via the Internet using the Jewish
Students Association e-mail list at the University of Manitoba, Winnipeg,
Manitoba, Canada. As in Experiment 1 and 2, participants were asked to
fill out a short questionnaire concerning Jews’ perceptions of the Holo-
caust. Fifty-six participants (28 men; 28 women) responded by accessing
the Website and completing the questionnaire. Participants ranged in age
from 18 to 45 years (M 22.15, SD 4.21).
Design and Procedure
As in the previous experiments with Jewish participants, when the
Website was accessed all participants were told that they would be asked
questions concerning their perceptions of the Holocaust. Participants were
then randomly assigned to one of two categorization conditions (social
identity, n 28; human identity, n 28). In the human identity condition,
the Holocaust was described as an event that demonstrated how humans
had behaved aggressively toward other humans. In the social identity
condition, participants also received this human description of the Holo-
caust, but to induce social-level categorization, participants in the social
identity condition were then asked to indicate whether they were Jewish or
non-Jewish and whether they were German or non-German in origin. All
participants indicated they were Jewish and non-German.
Participants were then asked to indicate their agreement with a series of
Likert-type statements by clicking on their selected response option using
a1(definitely disagree)to8(definitely agree) scale. The collective guilt
assignment, willingness to forgive, and perceived pervasiveness of geno-
cide measures were identical to those used in Experiments 1 and 2.
Participants were, however, also asked to indicate their agreement with an
additional series of Likert-type statements by clicking on their selected
response option usinga1(definitely disagree)to8(definitely agree) scale.
Three items assessed differentiation between German people and Jewish
people. These items were “I think that Germans and Jews have a lot in
common,” “There is little difference between Jewish and German people,”
and “The differences between Jewish and German people are small.” These
items were all reverse scored so that higher numbers indicate greater
perceived differences between the groups. Three items were also used to
assess the degree to which participants differentiated between contempo-
rary Germans and Germans of the Nazi era. These items were “Contem-
porary Germans are fundamentally different than Germans who lived
during World War II,” “Germans of today and Germans of the Nazi era are
basically the same” (reverse scored), and “I make a distinction between
today’s Germans and Germans of the Nazi era.” Higher scores on this
296
WOHL AND BRANSCOMBE
index indicate greater perceived differences between Germans in the
present compared with those of the past.
We also included items that assessed potential social distance responses
to shifting categorization. Four items assessed social distance from indi-
vidual contemporary members of the historical perpetrator group. These
items were “I would have no problem marrying a German person,” “I
would feel comfortable having a conversation with a German person,” “I
could be a close friend with a person from Germany,” and “I would feel
anxious with a German as my next door neighbor.” All items, except the
last, were reverse scored so that higher scores indicate greater social
distance from individual members of the perpetrator group. Four items
were also used to assess social distance from symbols of the perpetrator
category as a whole. These items were “I would like to travel to Germany
on a vacation,” “I would buy a German-made car,” “I would enjoy
attending Oktoberfest in Germany,” and “I would like to learn the German
language.” These items were all reverse scored so that higher numbers
indicate greater social distance from the category as a whole.
Again, to assess the effectiveness of the categorization manipulation, we
asked participants to indicate the extent to which “Germans and Jews share
basic similarities” anchored at 1 (strongly disagree)and8(strongly agree).
Results
Preliminary Analyses
There were neither participant gender effects on any of the
dependent measures nor significant interactions between partici-
pant gender and condition. Consequently, all results are reported
collapsed across gender. Table 1 provides a summary of the means
and standard deviations for all of the measured variables by
condition. Table 3 presents the correlations between the categori-
zation manipulation, the mediating variables, and the outcome
variables.
Experimental Effects
Manipulation check. Again, we successfully manipulated
level of categorization. Jewish participants in the human identity
condition perceived Germans and Jews as sharing more similari-
ties (M 5.00, SD 1.25) than did participants in the social
identity condition (M 4.11, SD 1.93), F(1, 54) 4.22, p
.05, d 0.55.
Perceived pervasiveness of genocide. An overall index was
calculated by averaging the three pervasiveness of genocide items
(
.90). Level of categorization significantly affected perceived
pervasiveness of genocide, F(1, 54) 9.27, p .01, d 0.81. As
expected, when the human level of categorization was salient,
Jews perceived genocide to be more pervasive (M 5.51, SD
2.10) than they did when their social identity was salient (M
4.05, SD 1.45).
Differentiation between groups. To assess whether the cate-
gorization manipulation affected perceived differences between
Jewish and German people, we created an overall differentiation
between groups score (
.90). Results indicated a marginally
significant effect of level of categorization, F(1, 54) 3.82, p
.055, d 0.52. That is, participants perceived greater differences
between the in-group and the out-group in the social identity
condition (M 5.26, SD 1.65) than they did in the human
identity condition (M 4.38, SD 1.72).
Differentiation within the perpetrator category. We also as-
sessed whether the categorization manipulation influenced the
extent to which contemporary Germans and Germans of the Nazi
era are differentiated. We created an overall differentiation within
the category Germans score (
.84). Level of categorization
marginally influenced differentiation within the perpetrator cate-
gory, F(1, 54) 3.48, p .06. Jewish participants tended to
perceive a greater difference between contemporary Germans and
Germans alive during the Holocaust in the human identity condi-
tion (M 3.77, SD 1.91) than did those in the social identity
condition (M 2.89, SD 1.60).
Assignment of collective guilt. It was expected that Jewish
participants would assign more collective guilt to Germans in the
intergroup categorization condition compared with that assigned
when the human-level categorization was salient. An overall as-
signment of collective guilt score was calculated by averaging the
four assignment of collective guilt items (
.85). As expected,
when the intergroup level of categorization was salient, Jews
Table 3
Correlations for Measured Variables, Experiment 4
Variable 12345678
1. Categorization
2. Pervasiveness .38**
3. Forgiveness .28* .42**
4. Collective guilt assignment .38** .43** .45**
5. Differentiation between
groups .26† .36** .08 .07
6. Differentiation within
perpetrator category .25† .11 .15 .09 .15
7. Social distance from
members of perpetrator
category .27* .28* .34* .42** .01 .04
8. Social distance from symbols
of perpetrator category .34* .07 .36** .35** .30* .03 .61**
Note. Categorization was coded 1 social identity condition and 2 human identity condition. All measured
variables ranged on a scale from 1 to 8.
p .07. * p .05. ** p .01, two-tailed.
297
FORGIVENESS AND COLLECTIVE GUILT
assigned more collective guilt to Germans (M 5.49, SD 1.42)
than when the human level was salient (M 4.23, SD 1.72),
F(1, 54) 8.89, p .01, d 0.80.
Willingness to forgive. An overall forgiveness score was cal-
culated by averaging the four forgiveness items (
.88). As
predicted, when the human level of categorization was used, Jews
were more willing to forgive contemporary Germans (M 5.47,
SD 1.85) than they were when the social level of categorization
was salient (M 4.56, SD 1.33), F(1, 54) 4.47, p .04, d
0.56.
Mediational Analysis for Collective Guilt
As in the previous experiments, Baron and Kenny’s regression
procedure was used to test for mediation. As the ANOVA indi-
cated, the categorization manipulation reliably affected perceived
pervasiveness of genocide and collective guilt assignment. There-
fore, both the categorization manipulation variable (coded as 1
social identity, 2 human identity) and perceived pervasiveness
of genocide were included in a regression model with collective
guilt assignment as the dependent variable, R
2
.24, F(2, 53)
8.36, p .01. The coefficient associated with perceived perva-
siveness of genocide significantly predicted collective guilt assign-
ment,
⫽⫺.34, t(54) ⫽⫺2.62, p .02, whereas the categori-
zation manipulation became a marginally significant predictor of
collective guilt,
⫽⫺.25, t(54) ⫽⫺1.90, p .06. We then
conducted a Sobel test to determine whether the indirect effect of
the categorization manipulation on collective guilt assignment, by
perceived pervasiveness of genocide, was significantly different
than zero. This test was significant (z 1.99, p .05).
We also examined the possibility that either perceived differ-
ences between groups (i.e., between Jews and Germans) or per-
ceived differences within the perpetrator category (i.e., between
contemporary Germans and Germans of the Nazi era) mediated the
categorization effect on collective guilt assignment. Neither of
these potential mediators met Baron and Kenny’s requirements for
mediation. Specifically, neither perceived differences between
groups,
⫽⫺.03, t(54) ⫽⫺.24, p .80, nor perceived differ-
ences within the group,
.01, t(54) 0.05, p .96, predicted
collective guilt assignment when either of these potential media-
tors was included in the equation with the categorization manip-
ulation variable.
Mediational Analysis for Willingness to Forgive
The categorization manipulation variable and perceived perva-
siveness of genocide were entered into a regression model with
willingness to forgive as the dependent variable, R
2
.19, F(2,
53) 6.36, p .01. Although the coefficient associated with
perceived pervasiveness of genocide was a significant predictor of
willingness to forgive,
.37, t(54) 2.77, p .01, the
categorization manipulation variable was not,
.14, t(54)
1.01, p .31. We then conducted a Sobel test to determine
whether the indirect effect of the categorization manipulation on
willingness to forgive, by perceived pervasiveness of genocide,
was significantly different than zero. This test was significant (z
2.05, p .05).
We also examined the possibility that either perceived differ-
ences between groups (i.e., between Jews and Germans) or per-
ceived differences within the perpetrator group (i.e., between con-
temporary Germans and Germans of the Nazi era) mediated the
categorization effect on willingness to forgive. As was the case for
collective guilt assignment, neither of these potential mediators
met Baron and Kenny’s requirements for mediation. Specifically,
neither perceived differences between groups,
.16, t(54)
1.19, p .23, nor perceived differences within the group,
.24, t(54) ⫽⫺1.78, p .08, predicted willingness to forgive
when either potential mediator was included in the equation with
the categorization manipulation variable.
Behavior Toward Contemporary Members of the
Historical Perpetrator Group
Social distance from individual members of the perpetrator
group. To assess whether the categorization manipulation af-
fected the extent to which Jewish participants were willing to
associate with members of the historical perpetrator group, we
created an overall social distance from contemporary members of
the perpetrator group score (
.87). Level of categorization
significantly influenced this index of social distance from individ-
ual category members, F(1, 54) 4.28, p .05, d 0.56. Jewish
participants distanced themselves more from members of the per-
petrator group in the social identity condition (M 3.19, SD
1.99) than they did in the human identity condition (M 2.21,
SD 1.49). We then sought to test whether this effect was
mediated by the emotional reactions of the participants. Because
we had already established the mediational role of perceived
pervasiveness of genocide on the intergroup emotional reaction
variables, for this test of mediation we examined the relations
among categorization, emotional reactions (collective guilt and
forgiveness), and social distance. Therefore, we regressed the
categorization manipulation and assignment of collective guilt on
social distance from individual members of the perpetrator group,
R
2
.19, F(2, 53) 6.25, p .01. Assignment of collective guilt
was a significant predictor of social distance from members of the
perpetrator group,
.37, t(54) 2.77, p .01, whereas the
categorization manipulation was not,
⫽⫺.13, t(54) ⫽⫺0.99,
p .32. Willingness to forgive was also examined as a possible
mediator of the effect of categorization on social distance. The
categorization manipulation and willingness to forgive were en-
tered into a regression model with social distance from individual
members of the perpetrator group as the dependent variable, R
2
.15, F(2, 53) 4.67, p .02. Although the coefficient associated
with willingness to forgive was a significant predictor of social
distance from contemporary members of the perpetrator group,
⫽⫺.29, t(54) 2.18, p .04, the categorization manipulation
was not,
⫽⫺.19, t(54) ⫽⫺1.45, p .15.
Social distance from symbols of the perpetrator category. We
also assessed whether our categorization manipulation influenced
Jewish people’s desired distance from symbols of the perpetrator
category as a whole (e.g., willingness to buy a German car). Thus,
we created an overall index of social distance from symbols of the
perpetrator category (
.84). A significant effect of categoriza-
tion was found such that Jewish people distanced themselves from
German symbols more in the social identity condition (M 4.96,
SD 1.90) than in the human identity condition (M 3.36, SD
1.91), F(1, 54) 6.83, p .02, d 0.70. As with the social
distance from members of the perpetrator group, we tested whether
298
WOHL AND BRANSCOMBE
the effect of categorization on social distance from symbols of the
perpetrator group was mediated by the participants’ emotional
reactions. We first focused on collective guilt assignment as a
possible mediator of social distance from symbols of the perpe-
trator group. With both categorization and assignment of collective
guilt entered as predictors, R
2
.17, F(2, 53) 5.53, p .01,
assignment of collective guilt remained a significant predictor of
social distance from symbols of the historical perpetrator group,
.27, t(54) 1.97, p .05, but categorization did not attain
significance,
⫽⫺.24, t(54) 1.74, p .08. We then tested
willingness to forgive as a mediator, with the categorization ma-
nipulation and willingness to forgive being regressed simulta-
neously on the social distance from symbols index, R
2
.19, F(2,
53) 6.18, p .01. Both willingness to forgive,
⫽⫺.29,
t(54) 2.24, p .03, and the categorization manipulation,
.26, t(54) ⫽⫺1.98, p .05, were significant predictors in this
equation. Because the effect of the categorization manipulation
was reduced when both categorization and willingness to forgive
and when the categorization manipulation and collective guilt
assignment were used as predictors this satisfied the requirements
for partial mediation (Baron & Kenny, 1986).
Discussion
Experiment 4 replicated the results of our previous experiments.
Categorizing at the human level promoted forgiveness of Germans
and lowered collective guilt assignment by changes in the per-
ceived pervasiveness of genocide. Other possible mediators of the
categorization effect were tested, but neither differentiation be-
tween contemporary Germans and contemporary Jews nor differ-
entiation between contemporary Germans and Germans of the
Nazi era satisfied the requirements for mediation. Shifting catego-
rization to a more inclusive level increased willingness to close the
social distance between contemporary Germans and contemporary
Jews. Specifically, when shared human category membership was
salient, Jewish people were more willing to be associated with
members of the perpetrator group as well as with the symbols that
reflect the perpetrator group as a whole. These results suggest that
inclusive categorization can indeed lead the harmed group to not
only increasingly forgive and assign less collective guilt to con-
temporary members of that group but also to alter their behavior
toward the perpetrator category as a whole.
General Discussion
Undeniably, the past century can be seen as “a catalog of our
capacity to wreak considerable harm on one another and our gross
inhumanity to our fellow humans” (Tutu, 1999, p. 124). This
“catalog” of human harmdoing is more likely to be perceived as
pervasive when categorization occurs at the most inclusive human
level. Shifting to an increasingly inclusive categorization altered
the perceived pervasiveness of genocide (Experiments 1, 2, and 4)
and the perceived pervasiveness of intergroup harm (Experiment
3). Thinking that groups have transgressed against each other
throughout human history affected willingness to forgive a specific
out-group for its harmful past and reduced collective guilt assign-
ment. When the harm done by the perpetrator group is seen as but
one example of the harm humans have perpetrated against their
fellow humans, it encourages forgiveness and discourages assign-
ment of collective guilt to that specific perpetrator group.
In Experiments 1, 2, and 4, human-level categorization altered
judgments of contemporary members of their perpetrator group.
When Jewish participants categorized at the human level rather
than the social level, they were more willing to forgive contem-
porary Germans for their ancestors’ transgressions and assigned
them less collective guilt. In Experiment 3, we replicated and
extended our results by demonstrating that shifts in categorization
can influence responses by a historically victimized group to their
current victimizers. When Native Canadians categorized at the
human level they were more willing to forgive White Canadians
and assigned them less collective guilt than when they categorized
at the social identity level.
We also examined the route by which shifts in level of catego-
rization among victimized group members affect the assignment of
collective guilt and willingness to forgive perpetrator group mem-
bers. In Experiments 1, 2, and 4, we found that the perceived
pervasiveness of genocide mediated the effect of categorization on
forgiveness and collective guilt assignment. When the human level
of categorization was salient, people perceived genocide as less
unique to Germans. To the extent that genocide was seen as not
something done uniquely by Germans against Jews, contemporary
Germans were forgiven more and assigned less collective guilt for
the acts of their ancestors (i.e., contemporary Germans were not
required to carry the burden of guilt). In Experiment 3, we exam-
ined perceived pervasiveness of intergroup harm as a mediator of
such affective responses to the perpetrator group. In this case we
addressed whether shifts in level of categorization influence will-
ingness to forgive and assignment of collective guilt in a group
(Native Canadians) that continues to be mistreated by their histor-
ical perpetrator group (White Canadians). As was the case with the
other experiments, when the human level of categorization was
salient, Native Canadians perceived that other human groups com-
mitted similar intergroup harm, which increased willingness to
forgive contemporary members of the historical perpetrator group
and decreased the desire for contemporary members of the perpe-
trator group to carry the burden of guilt for actions committed by
their ancestors. Thus, more inclusive levels of categorization may
be an integral part of the process needed for reconciliation between
groups. Making salient their shared human identity can result in
improved intergroup affective responses.
Our data also showed (Experiment 4) that shifting the level of
categorization influences behavioral tendencies toward contempo-
rary members of the historical perpetrator group. When categoriz-
ing at the human level, Jewish people desired less social distance
both from individual members of the historical perpetrator group
and from symbols of the perpetrator category as a whole. Consis-
tent with Smith (1993, 1999) and Yzerbyt et al. (2003), changes in
behavioral tendencies through saliency of a shared group mem-
bership can mediate group-level emotional reactions. Specifically,
to the extent that Jewish people were more willing to forgive and
assign less collective guilt, they desired less distance from con-
temporary Germans. Thus, our results do not merely represent an
attitudinal change, but they also hold potential for change in the
structural relations between members of the historical perpetrator
and members of the victimized group. Instead of creating distance
from members of the perpetrator group, as is most often the
outcome of group victimization (Tutu, 1999), Jewish people re-
299
FORGIVENESS AND COLLECTIVE GUILT
ported a greater willingness to engage in activities that would bring
the victimized and perpetrator group together when inclusive cat-
egorization was used.
Implications of Shifts in Categorization
Some have suggested that perceiving the out-group as part of an
in-group category as occurred in the human identity condition may
be a prerequisite for feeling empathy toward that out-group
(Turner et al., 1987). Groups in conflict may be more likely to take
a sympathetic stance toward the plight of out-group members if
those out-group members— by way of shifts in categorization—
become, in effect, members of a common in-group (Dovidio et al.,
1998; Gaertner et al., 1989, 1990, 1994). That is, the potential
divisiveness of group memberships may be moderated when an
inclusive identity is made salient, uniting members of opposing
groups in a common category (Gaertner et al., 1993). Gaertner et
al. (1994) found that the intergroup bias in multicultural high
schools decreased when students considered the student body as
one superordinate group rather than as several subgroups. In the
current study, we demonstrated that when victimized group mem-
bers’ human identity is made salient, the in-group can include
members of the perpetrator group. Consistent with the common
in-group identity model (Gaertner et al., 1993), biases were re-
duced such that members of a historical victimized group assigned
less collective guilt and were more willing to forgive contempo-
rary members of their perpetrator group.
The current data considerably extend existing recategorization
findings. First, our research does not simply involve recategorizing
former out-group members as now members of one’s own social
group (e.g., former greens are now blues like us), in which the two
groups are substantially at the same level of inclusiveness (e.g., the
social identity level for greens and blues). Rather, in all four
experiments both distinct ethnic and/or national groups are in-
cluded in a single inclusive category that encompasses all possible
subgroups. Indeed, the level of categorization used is one not
previously investigated (e.g., the maximally inclusive level of
humans). Second, our research demonstrates that such inclusive
categorization effects apply to members of the social category as a
whole and not just to specific individuals with whom participants
had engaged in personal interaction, as has been the case in
existing common in-group identity studies (Gaertner et al., 1989;
Nier et al., 2001). Third, our studies demonstrate that recategori-
zation effects at the human level extend beyond contemporary
circumstances. That is, the effect of human-level categorization
affects responses to past actions of the out-group (assignment of
collective guilt to Germans or White Canadians for the past) and
future-oriented responses toward the former out-group (willing-
ness to forgive Germans or White Canadians). We see this as akin
to Yehuda et al.’s (2000) demonstration that posttraumatic stress
reactions can traverse generations, to affect second-generation
Holocaust survivors. Such effects are not limited to only those with
personal experience with the harmdoing out-group. Although his-
torical victimization might be experienced directly, categorization
as a member of that victimized group can influence both emotional
(Studies 1, 2, 3, and 4) and behavioral (Study 4) responses to
contemporary members of the perpetrator group. Last, to date,
recategorization studies have focused almost exclusively on min-
imal groups or on real groups that have experienced relatively
minor injustices within a laboratory setting (Nier et al., 2001). Our
participants are, however, contemporary members of historically
victimized groups who are responding to and subsequently forgiv-
ing members of a social category who personally were not respon-
sible for the injustice but whose ancestors were responsible for
tremendous harm. As such, our findings have far-reaching impli-
cations for understanding real world intergroup conflict by iden-
tifying at least one process by which such historical conflicts may
be resolved.
Caveats
Some limitations of the current research should be noted. First,
the Jewish people who participated in our studies were not a
random sample of Jewish North Americans. Instead, we used a
self-selected sample of Jews who joined the Jewish Student’s
Association on their university campus. Nonetheless, our catego-
rization manipulation proved successful in both Canadian and
American Jewish samples. Second, because our participants are
members of a Jewish organization, one might expect their attitudes
regarding Germans to be rather solidified prior to participating in
our study. As a result, changing emotional reactions to their
historical perpetrator group should be rather difficult. Yet, despite
participants’ likely strong preexisting attitudes and familiarity with
Holocaust history, our manipulation was highly effective. Note
also, that all of the studies involving Jewish people were conducted
via the Internet, which lends itself to a potentially high-identified
sample. The strength of this method, however, is that for sensitive
matters such as emotional responses, there is evidence that less
socially desirable and more honest responses may be obtained via
the Internet (Evans, Garcia, Garcia, & Baron, 2003). Last, the
Native Canadian sample in Experiment 3 was raised and currently
resides in a large urban center. As such, our sample is highly
assimilated, as evidenced by their high levels of contact with
White Canadians. Postmes and Branscombe (2002) showed that
highly assimilated members of an ethnic minority group tend to be
fairly low in group identification. Nonetheless, our manipulation
was equally effective with Native Canadians as with Jewish
participants.
Potential Consequences of Recategorization
Although we were able to alter the perceived distinctiveness of
Germans and Jews, as well as of White and Native Canadians, the
sustainability of our inclusiveness manipulation is uncertain. Ac-
cording to social identity theory (Tajfel & Turner, 1986), because
people are motivated to maintain a positive view of their own
group, when the perpetrator group is included within a shared
superordinate category, perpetrator group members should be eval-
uated more favorably than they are when they are categorized as an
out-group. However, weakening boundaries between groups may
be perceived as threatening (Jetten, Spears, Hogg, & Manstead,
2000). Groups are motivated to achieve positive differentiation
when group distinctiveness is low and they are threatened by
similarity to a relevant out-group (Dovidio et al., 1998; Spears,
Doosje, & Ellemers, 1997). Perceived threat to in-group identity
may be exacerbated following recategorization when the now
similar out-group members are associated with the historical per-
petrator group. If victimized group members perceive themselves
300
WOHL AND BRANSCOMBE
to be similar to members of the perpetrator group, then derogation
of the out-group would, in effect, be derogation of the in-group as
well. Therefore, to maintain a positive social identity, victimized
group members cannot judge members of the perpetrator group
harshly in the human identity condition, for doing so would lower
the value of their in-group. By including the perpetrator group in
a common in-group, the new superordinate category might be
perceived as taking on some immoral characteristics. Because
people are motivated to think of the in-group in a positive light,
having immoral people in the group should be rather threatening to
the maintenance of a positive social identity (Branscombe, Wann,
Noel, & Coleman, 1993; see Marques & Paez, 1994, for a review).
As a result, victimized group members may resist being catego-
rized in the same group as members of the perpetrator group. Such
identity concerns may therefore inhibit the maintenance of human-
level categorization, and this should be particularly likely among
highly identified group members.
Shifts in categorization may be also opposed by historically
victimized group members because use of a more inclusive level of
categorization may make the process of seeking reparations for
past wrongs more difficult. In the current series of experiments,
perceived pervasiveness of intergroup harm mediated the effect of
categorization on willingness to forgive and collective guilt as-
signment. To the extent that people become cognizant of the
pervasiveness of intergroup harm, situational accounts may be-
come more apt than group-based dispositional accounts. That is,
dispositional accounts (and vilification of the out-group) can be
undermined by increasingly inclusive categorization. Indeed, when
situational explanations take precedence over dispositional expla-
nations, then attributions of responsibility to the perpetrator group
are lowered (Miller, Buddie, & Kretschmar, 2002; Miller, Gordon,
& Buddie, 1999). Because responsibility is the adhesive that
connects the actor (or in this case the group) to the event (Schlen-
ker, Britt, Pennington, Murphy & Doherty, 1994), the victimized
group may then have to forgo seeking retribution and, perhaps,
reparations.
Social psychologists have long been interested in understanding
the social contextual factors that facilitate genocide (e.g., Newman
& Erber, 2002; Staub, 1989). From Milgram’s obedience studies
(Milgram, 1963) to Tajfel and Turner’s (1986) theorizing concern-
ing the conditions that encourage the expression of intergroup
competition, the antecedents of group-based harmdoing have been
of central interest to social scientists. The purpose of our research
was to examine how contemporary members of historically vic-
timized groups can be prompted to respond in a more prosocial
direction toward contemporary members of the historical perpe-
trator group. Previous research (Branscombe, McGarty, & Doosje,
2003) has demonstrated that when the Holocaust is perceived as a
human tragedy, non-Jewish participants make more external attri-
butions for Germans’ actions. Some might argue that an increase
in external attributions is, in effect, a means of condoning the
crimes of Nazi Germany (e.g., Baumeister, 1997; Zillmer, Har-
rower, Ritzler, & Archer, 1995). Thus, it might be suggested that
by shifting categorization from the social to the human level we
are, de facto, promoting forgiveness of Nazis. We do not think this
is a valid conclusion to draw from our data. Examination of the
process underlying improved responses toward contemporary
members of a perpetrator category is not making allowances for
the Nazi genocide or any other instance of great harmdoing. Our
research represents a successful attempt at changing historically
victimized group members’ emotional reactions to people who did
not commit the crimes but are nonetheless associated with the
crimes because of their category membership. Thus, our partici-
pants exhibited the most positive emotional reactions to contem-
porary Germans in the inclusive condition, and this was mediated
by seeing similarities between the Nazi atrocities and other in-
stances of genocide. Although our participants did differentiate
between contemporary Germans and Germans of the Nazi era
(Experiment 4), this did not mediate the effects of categorization
on forgiveness and collective guilt assignment that we obtained.
Conclusions
In all four experiments, shifts in participants’ level of categori-
zation—from social to human— exerted powerful effects on emo-
tional responses toward contemporary members of the perpetrator
category. Maintaining human-level categorization, on the other
hand, might well be rather difficult. Many contextual factors are
likely to encourage categorization at the social identity level. In
fact, groups are likely to have powerful punishments for in-group
members who suggest that perpetrators and victims share anything
at all— especially the same social category. The results of our four
experiments clearly indicate that negative group-based feelings
toward the perpetrator category can be reduced with more inclu-
sive levels of categorization. We have demonstrated that willing-
ness to forgive and collective guilt assignment can be affected by
shifts in categorization. Our results are not simply about percep-
tions of an out-group, they also illustrate a greater willingness on
the part of victimized group members to let go of past transgres-
sions as a result of human-level categorization.
The resolution of social conflict involves more than changing
negative beliefs. As Archbishop Tutu’s Truth and Reconciliation
Commission in South Africa has shown us, what seems to be
needed is an ability to see out-group members as human like the
self (Tutu, 1999). Such shifts to a more inclusive level of catego-
rization can lead to improved evaluations of, and approach behav-
iors toward, contemporary members of a perpetrator group. Such
inclusive categorization in which all members of a society, regard-
less of subgroup membership, are perceived as part of the “human
family” may well be crucial for intergroup forgiveness and the
creation of a more peaceful future.
References
Barkan, E. (2000). The guilt of nations: Restitution and negotiating his-
torical injustices. New York: Norton.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable
distinction in social psychology research: Conceptual, strategic, and
statistical considerations. Journal of Personality and Social Psychology,
51, 1173–1182.
Baumeister, R. F. (1997). Evil: Inside human violence and cruelty. New
York: Freeman.
Baumeister, R. F., Stillwell, A. M., & Heatherton, T. F. (1994). Guilt: An
interpersonal approach. Psychological Bulletin, 115, 243–267.
Bizman, A., Yinon, Y., & Krotman, S. (2001). Group-based emotional
distress: An extension of self-discrepancy theory. Personality and Social
Psychology Bulletin, 27, 1291–1300.
Boon, S. D., & Sulsky, L. M. (1997). Attributions of blame and forgiveness
in romantic relationships: A policy-capturing study. Journal of Social
Behavior and Personality, 12, 19 44.
301
FORGIVENESS AND COLLECTIVE GUILT
Branscombe, N. R., Doosje, B., & McGarty, C. (2002). Antecedents and
consequences of collective guilt. In D. M. Mackie & E. R. Smith (Eds.),
From prejudice to intergroup emotions: Differentiated reactions to
social groups (pp. 44 66). Philadelphia: Psychology Press.
Branscombe, N. R., McGarty, C., & Doosje, B. (2003). How people
explain the Holocaust: Assignment of group-based guilt as a function of
level of categorization. Manuscript submitted for publication.
Branscombe, N. R., & Miron, A. M. (2004). Interpreting the ingroup’s
negative actions toward another group: Emotional reactions to appraised
harm. In L. Z. Tiedens & C. W. Leach (Eds.), The social life of emotions
(pp. 314 –335). New York: Cambridge University Press.
Branscombe, N. R., Slugoski, B., & Kappen, D. M. (2004). The measure-
ment of collective guilt: What it is and what it is not. In N. R.
Branscombe & B. Doosje (Eds.), Collective guilt: International perspec-
tives (pp. 16 –34). New York: Cambridge University Press.
Branscombe, N. R., Wann, D. L., Noel, J. G., & Coleman, J. (1993).
In-group or out-group extremity: Importance of the threatened identity.
Personality and Social Psychology Bulletin, 19, 381–388.
Brewer, M. B. (1979). In-group bias in the minimal intergroup situation: A
cognitive-motivational analysis. Psychological Bulletin, 86, 307–324.
Churchill, W. (1997). A little matter of genocide. San Francisco: City
Lights Books.
Cowell, A. (1998, February 10). U.S. Jewish group sets up office in Berlin
to watch for ‘demons.’ New York Times. Retrieved April 30, 2003, from
http://www.codoh.com/newsdesk/980210.html.
Dovidio, J. F., Gaertner, S. L., & Validzic, A. (1998). Intergroup bias:
Status, differentiation, and a common in-group identity. Journal of
Personality and Social Psychology, 75, 109 –120.
Downey, M. (1999, April). Canada’s genocide. Maclean’s, 112, 56 –58.
Enright, R., & North, J. (1998). Exploring forgiveness. Madison: Univer-
sity of Wisconsin Press.
Evans, D. C., Garcia, D. J., Garcia, D. M., & Baron, R. S. (2003). In the
privacy of their own homes: Using the Internet to assess racial bias.
Personality and Social Psychology Bulletin, 29, 273–284.
Exline, J. J., & Baumeister, R. F. (2000). Expressing forgiveness and
repentance: Benefits and barriers. In M. E. McCullough, K. I. Parga-
ment, & C. E. Thoresen (Eds.) Forgiveness: Theory, research and
practice (pp. 133–155). New York: Guilford Press.
Frideres, J. M. (1998). Aboriginal peoples in Canada: Contemporary
conflicts. Scarborough, Ontario, Canada: Prentice Hall.
Gaertner, S. L., Dovidio, J. F., Anastasio, P. A., Bachman, B. A., & Rust,
M. C. (1993). The common ingroup identity model: Recategorization
and the reduction of intergroup bias. In W. Stroebe & M. Hewstone
(Eds.), European review of social psychology (Vol. 4, pp. 1–26). New
York: Wiley.
Gaertner, S. L., Dovidio, J. F., Nier, J. A., Ward, C. M., & Banker, B. S.
(1999). Across cultural divides: The value of a superordinate identity. In
D. A. Prentice & D. T. Miller (Eds.), Cultural divides: Understanding
and overcoming group conflict (pp. 173–212). New York: Russell Sage
Foundation.
Gaertner, S. L., Mann, J. A., Dovidio, J. F., Murrell, A. J., & Pomare, M.
(1990). How does cooperation reduce intergroup bias? Journal of Per-
sonality and Social Psychology, 59, 692–704.
Gaertner, S. L., Mann, J. A., Murrell, A. J., & Dovidio, J. F. (1989).
Reducing intergroup bias: The benefits of recategorization. Journal of
Personality and Social Psychology, 57, 239 –249.
Gaertner, S. L., Rust, M. C., Dovidio, J. F., Bachman, B. A., & Anastasio,
P. A. (1994). The contact hypothesis: The role of common ingroup
identity on reducing intergroup bias. Small Group Research, 25, 224
249.
George, P., Kuhn, P., & Sweetman, A. (1996). Patterns of employment,
unemployment, and poverty: A comparative analysis of several aspects
of the employment experience of Aboriginal and non-Aboriginal Cana-
dians using 1991 PUMP. Royal Commission on Aboriginal People:
People to People, Nation to Nation. Ottawa, Ontario, Canada: Minister
of Supply and Services Canada.
Goldhagen, D. J. (1996). Hitler’s willing executioners: Ordinary Germans
and the Holocaust. New York: Knopf.
Haslam, S. A., Turner, J. C., Oakes, P. J., McGarty, C., & Reynolds, K. J.
(1998). The group as a basis for emergent stereotype consensus. In W.
Stroebe & M. Hewstone (Eds.), European review of social psychology
(Vol. 8, pp. 203–239). New York: Wiley.
Heider, F. (1958). The psychology of interpersonal relations. New York:
Wiley.
Hoffman, M. L. (1987). The contribution of empathy to justice and moral
judgment. In N. Eisenberg (Ed.), Empathy and its development (pp.
47– 80). Cambridge, United Kingdom: Cambridge University Press.
Ignatieff, M. (1997, March). The elusive goal of war trials. Harpers, 294,
15–18.
Jankowski, W. B., & Moazzami, B. (1994). Size distribution of income and
income inequity among the Native population of northwest Ontario.
Canadian Journal of Native Studies, 14, 47– 60.
Jetten, J., Spears, R., Hogg, M. A., & Manstead, A. S. R. (2000). Discrim-
ination constrained and justified: Variable effects of group variability
and in-group identification. Journal of Experimental Social Psychology,
36, 329 –356.
Liu, J. H., & Hilton, D. J. (in press). How the past weighs on the present:
Social representations of history and their role in identity politics. British
Journal of Social Psychology.
Lock Kunz, J., Milan, A., & Schetagne, S. (2000). Unequal access: A
Canadian profile of racial differences in education, employment and
income. Toronto, Ontario, Canada: Canadian Race Relations Founda-
tion.
Marques, J. M., & Paez, D. (1994). The “black sheep effect”: Social
categorization, rejection of ingroup deviates, and perception of group
variability. In W. Stroebe & M. Hewstone (Eds.), European review of
social psychology (Vol. 5, pp. 37– 68). New York: Wiley.
McCullough, M. E., Bellah, C. G., Kilpatrick, S. D., & Johnson, J. L.
(2001). Vengefulness: Relationships with forgiveness, rumination, well-
being and the Big Five. Personality and Social Psychology Bulletin, 27,
601– 610.
McCullough, M. E., & Hoyt, W. T. (2002). Transgression-related motiva-
tional dispositions: Personality substrates of forgiveness and their links
to the Big Five. Personality and Social Psychology Bulletin, 28, 1556
1573.
McCullough, M. E., Pargament, K. I., & Thoresen, C. E. (2000). The
psychology of forgiveness: History, conceptual issues, and overview. In
M. E. McCullough, K. I. Pargament, & C. E. Thoresen (Eds.), Forgive-
ness: Theory, research and practice (pp. 1–14). New York: Guilford
Press.
Milgram, S. (1963). Behavioral study of obedience. Journal of Abnormal
and Social Psychology, 67, 371–378.
Miller, A. G., Buddie, A. M., & Kretschmar, J. (2002). Explaining the
Holocaust: Does social psychology exonerate the perpetrators? In L.
Newman & R. Erber (Eds.), Understanding genocide: The social psy-
chology of the Holocaust (pp. 301–324). New York: Oxford University
Press.
Miller, A. G., Gordon, A. K., & Buddie, A. M. (1999). Accounting for evil
and cruelty: Is to explain to condone? Personality and Social Psychology
Review, 3, 254 –268.
Minow, M. (1998). Between vengeance and forgiveness. Boston: Beacon
Press.
Newman, L. S., & Erber, R. (2002). Epilogue: Social psychologists con-
front the Holocaust. In L. S. Newman & R. Erber (Eds.), Understanding
genocide: The social psychology of the Holocaust (pp. 325–345). New
York: Oxford University Press.
Nier, J. A., Gaertner, S. L., Dovidio, J. F., Banker, G. S., Ward, C. M., &
Rust, M. C. (2001). Changing interracial evaluations and behavior: The
302
WOHL AND BRANSCOMBE
effect of common group identity. Group Processes and Intergroup
Relations, 4, 229 –316.
Postmes, T., & Branscombe, N. R. (2002). Influence of long-term racial
environmental composition on subjective well-being in African Ameri-
cans. Journal of Personality and Social Psychology, 83, 735–751.
Powell, A. A., Branscombe, N. R., & Schmitt, M. T. (in press). Inequality
as “ingroup privilege” or “outgroup disadvantage”: The impact of group
focus on collective guilt and interracial attitudes. Personality and Social
Psychology Bulletin.
Rensmann, L. (2004). Collective guilt, national identity, and political
processes in contemporary Germany. In N. R. Branscombe & B. Doosje
(Eds.), Collective guilt: International perspectives (pp. 169–190). New
York: Cambridge University Press.
Reynolds, K. J., & Oakes, P. J. (2000). Variability in impression formation:
Investigating the role of motivation, capacity, and the categorization
process. Personality and Social Psychology Bulletin, 26, 355–373.
Reynolds, K. J., Turner, J. C., & Haslam, S. A. (2000). When are we better
than them and they worse than us? A closer look at social discrimination
in positive and negative domains. Journal of Personality and Social
Psychology, 78, 64 80.
Russell, B. (1998). Records of the federal Department of Indian Affairs at
the National Archives of Canada: A source for genealogical research.
Toronto, Ontario, Canada: Ontario Genealogical Society.
Sandage, S. J., Worthington, E. L., Everett, L., & Hight, T. L. (2000).
Seeking forgiveness: Theoretical context and an initial empirical study.
Journal of Psychology and Theology, 28, 21–35.
Schlenker, B. R., Britt, T. W., Pennington, J. W., Murphy, R., & Doherty,
K. J. (1994). The triangle model of responsibility. Psychological Review,
101, 632– 652.
Schmitt, M. T., Silvia, P. J., & Branscombe, N. R. (2000). The intersection
of self-evaluation maintenance and social identity theories: Intragroup
judgment in interpersonal and intergroup contexts. Personality and
Social Psychology Bulletin, 26, 1598 –1606.
Smith, E. R. (1993). Social identity and social emotions toward new
conceptualizations of prejudice. In D. M. Mackie & D. L. Hamilton
(Eds.), Affect, cognition, and stereotyping: Interactive processes in
group perception (pp. 297–315). San Diego, CA: Academic Press.
Smith, E. R. (1999). Affective and cognitive implications of group mem-
bership becoming part of the self: New models of prejudice and of the
self-concept. In D. Abrams & M. Hogg (Eds.), Social identity and social
cognition (pp. 183–196). Oxford, England: Blackwell.
Spears, R., Doosje, B., & Ellemers, N. (1997). Self-stereotyping in the face
of threats to group status and distinctiveness: The role of group identi-
fication. Personality and Social Psychology Bulletin, 23, 538 –553.
Staub, E. (1989). The roots of evil: The origins of genocide and other group
violence. New York: Cambridge University Press.
Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup
behavior. In S. Worchel & W. G. Austin (Eds.), Psychology of inter-
group relations (pp. 7–24). Chicago: Nelson-Hall.
Tangney, J. P., & Fischer, K. W. (Eds.). (1995). Self-conscious emotions:
The psychology of shame, guilt, embarrassment, and pride. New York:
Guilford Press.
Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S.
(1987). Rediscovering the social group: A self-categorization theory.
Oxford, England: Blackwell.
Turner, J. C., Oakes, P. J., Haslam, S. A., & McGarty, C. (1994). Self and
collective: Cognition and social context. Personality and Social Psy-
chology Bulletin, 20, 454 463.
Turner, J. C., & Onorato, R. S. (1999). Social identity, personality, and the
self-concept: A self-categorization perspective. In T. R. Tyler, R. M.
Kramer, & O. P. John (Eds.), The psychology of the social self (pp.
11– 46). Mahwah, NJ: Erlbaum.
Tutu, D. M. (1999). No future without forgiveness. New York: Doubleday.
Wohl, M. J. A., & Branscombe, N. R. (2004). Importance of social
categorization for forgiveness and collective guilt assignment for the
Holocaust. In N. R. Branscombe & B. Doosje (Eds.), Collective guilt:
International perspectives (pp. 284–305). New York: Cambridge Uni-
versity Press.
Yehuda, R., Bierer, L. M., Schmeidler, J., Aferiat, D. H., Breslau, I., &
Dolan, S. (2000). Low cortisol and risk for PTSD in adult offspring of
Holocaust survivors. American Journal of Psychiatry, 157, 1252–1259.
Yzerbyt, V., Dumont, M., Wigboldus, D., & Gordijn, E. (2003). I feel for
us: The impact of categorization and identification on emotions and
action tendencies. British Journal of Social Psychology, 42, 533–549.
Zechmeister, J. S., & Romero, C. (2002). Victim and offender accounts of
interpersonal conflict: Autobiographical narratives of forgiveness and
unforgiveness. Journal of Personality and Social Psychology, 82, 675–
686.
Zillmer, E. A., Harrower, M., Ritzler, B. A., & Archer, R. P. (1995). The
quest for the Nazi personality: A psychological investigation of Nazi war
criminals. Hillsdale, NJ: Erlbaum.
Received May 30, 2003
Revision received July 5, 2004
Accepted July 22, 2004
303
FORGIVENESS AND COLLECTIVE GUILT
... In the context of intergroup conflict, forgiveness has been considered as an essential element for restoring intergroup relations and promoting reconciliation (e.g., Hewstone et al., 2006;Tavuchis, 1991;Van Tongeren et al., 2014). Extensive research showed that forgiveness is linked to a positive change in attitudes towards a rival group (Wohl & Branscombe, 2005) as well as peaceful coexistence McLernon et al., 2004;Noor et al., 2008). Moreover, forgiveness was associated with less feelings of social distance towards the former adversary (Čehajić et al., 2008), less feelings of injustice (Wenzel & Okimoto, 2015), and greater prosocial behaviour towards the former adversary (Wohl & Branscombe, 2005). ...
... Extensive research showed that forgiveness is linked to a positive change in attitudes towards a rival group (Wohl & Branscombe, 2005) as well as peaceful coexistence McLernon et al., 2004;Noor et al., 2008). Moreover, forgiveness was associated with less feelings of social distance towards the former adversary (Čehajić et al., 2008), less feelings of injustice (Wenzel & Okimoto, 2015), and greater prosocial behaviour towards the former adversary (Wohl & Branscombe, 2005). ...
... Forgiveness has been defined in terms of leaving transgressions behind and moving forward (Hewstone et al., 2006;Noor et al., 2008;Staub & Pearlman, 2001), resisting the impulse to blame the former adversary (Wohl & Branscombe, 2005), and suppressing the desire for revenge . A more recent conceptualization of forgiveness implies that forgiveness is determined by multiple factors (Noor, 2016). ...
Article
Four correlational studies (N Albanians = 232, N Serbs = 129, N Bosniaks = 147, N Croats = 367) and one experimental study (N Bosniaks = 682), investigated the link between mass-mediated contact (i.e., information about former adversaries from the mass media) and forgiveness towards former adversaries in post-conflict societies. Specifically, we tested the association between positive and negative mass-mediated contact with one former adversary and forgiveness towards this former adversary (i.e., a primary transfer effect, Studies 1-4) and another former adversary (i.e., a secondary transfer effect, Studies 2-4). Positive mass-mediated contact with one former adversary was linked to greater forgiveness towards that former adversary and another former adversary. Intergroup trust mediated the primary transfer effect of positive mass-mediated contact, whereas generalization of trust and forgiveness from one former adversary to another mediated the secondary transfer effect. Our results underline the important role of positive but not negative mass-mediated contact with former adversaries in reconciliation.
... Specifically, for defensive HC, we should find a pattern that is typical for defensive reactions. Such a pattern should be characterized, for instance, by reduced collective guilt, attempts to downgrade the severity of the crimes committed, and avoidance of confrontation with the past (Bilewicz, 2016;Branscombe et al., 2004;Wohl & Branscombe, 2005). ...
... The 17 reconciliatory HC items were adapted and expanded from various measures on intergroup reconciliation (Brewer & Hayes, 2015;Cehajic et al., 2008;Hanke et al., 2013;Hewstone et al., 2004;e.g., "It is better not to open old wounds by talking about what happened in the past" by Brewer & Hayes, 2015 was transferred to "Collectively drawing a line under the period of national socialism can help to heal old wounds"). Finally, 15 items on defensive HC were taken from previous scales assessing the desire to close discussion on history (e.g., Allpress et al., 2014;Hagemann & Nathanson, 2015;Hanke et al., 2013;Imhoff et al., 2017;Sahdra & Ross, 2007;Wohl & Branscombe, 2005). Moreover, we added the 5-item HC scale by Imhoff (2010b). ...
... Noticeably, the pattern was almost identical to the one found in Study 1a, thus representing a typical pattern of defensive reactions. The only differences in the German sample were that SDO correlated significantly with all three subscales, whereas perceived pervasiveness of genocide-the perception that the committed crime is not unique but represents a pervasive phenomenon within human history (Wohl & Branscombe, 2005)-did not correlate with any of them. However, the overall correlation pattern of perceived pervasiveness also differed between samples (see SOM-B for a full overview of correlations), suggesting a context-specific variation in the meaning of that construct. ...
Article
Historical perpetrator groups seek to shield themselves from image threat by advocating for closing the discussion of their crimes. However, from a broader theoretical perspective, such demand for historical closure (HC) may also reflect willingness to reconcile with the victim group or to focus on the future rather than the past. In nine studies across four different contexts (Germany, United States, Italy, and Australia; N = 3405), we analyzed whether these three facets of HC (defensive, reconciliatory, and future-oriented) indeed substantially differ. Contrary to expectations, nomological network analyses suggested that all three facets reflect the same defensive desire (Studies 1a-2c) and are perceived as overall similar from a third-party perspective (Study 3). Finally, all three HC facets showed a positive trend toward costly avoidance of confrontation with the ingroup's perpetrator past (Studies 4a-c). We discuss implications for (and against) a more nuanced understanding of the demand for HC.
... In the context of intergroup conflict, forgiveness has been considered as an essential element for restoring intergroup relations and promoting reconciliation (e.g., Hewstone et al., 2006;Tavuchis, 1991;Van Tongeren et al., 2014). Extensive research showed that forgiveness is linked to a positive change in attitudes towards a rival group (Wohl & Branscombe, 2005) as well as peaceful coexistence McLernon et al., 2004;Noor et al., 2008). Moreover, forgiveness was associated with less feelings of social distance towards the former adversary (Čehajić et al., 2008), less feelings of injustice (Wenzel & Okimoto, 2015), and greater prosocial behaviour towards the former adversary (Wohl & Branscombe, 2005). ...
... Extensive research showed that forgiveness is linked to a positive change in attitudes towards a rival group (Wohl & Branscombe, 2005) as well as peaceful coexistence McLernon et al., 2004;Noor et al., 2008). Moreover, forgiveness was associated with less feelings of social distance towards the former adversary (Čehajić et al., 2008), less feelings of injustice (Wenzel & Okimoto, 2015), and greater prosocial behaviour towards the former adversary (Wohl & Branscombe, 2005). ...
... Forgiveness has been defined in terms of leaving transgressions behind and moving forward (Hewstone et al., 2006;Noor et al., 2008;Staub & Pearlman, 2001), resisting the impulse to blame the former adversary (Wohl & Branscombe, 2005), and suppressing the desire for revenge . A more recent conceptualization of forgiveness implies that forgiveness is determined by multiple factors (Noor, 2016). ...
... Specifically, for defensive HC, we should find a pattern which is typical for defensive reactions. Such a pattern should be characterized, for instance, by reduced collective guilt, attempts to downgrade the severity of the crimes committed, and avoidance of confrontation with the past (Bilewicz, 2016;Branscombe et al., 2004;Wohl & Branscombe, 2005). ...
... The 17 reconciliatory HC items were adapted and expanded from various measures on intergroup reconciliation (Brewer & Hayes, 2015;Cehajic et al., 2008;Hanke et al., 2013;Hewstone et al., 2004;e.g Hanke et al., 2013;Imhoff et al., 2017;Wohl & Branscombe, 2005). ...
... Noticeably, the pattern was almost identical to the one found in Study 1a, thus representing a typical pattern of defensive reactions. The only differences to the German sample were that SDO correlated significantly with all three subscales, whereas perceived pervasiveness of genocide -the perception that the committed crime is not unique but represents a pervasive phenomenon within human history (Wohl & Branscombe, 2005)-did not correlate with any of them. However, the overall correlation pattern of perceived pervasiveness also differed between samples (see SOM-B for a full overview of correlations), suggesting a context-specific variation in the meaning of that construct. ...
Preprint
Full-text available
Historical perpetrator groups seek to shield themselves from image threat by advocating for closing the discussion of their crimes. However, from a broader theoretical perspective, such demand for historical closure (HC) may also reflect willingness to reconcile with the victim group or to focus on the future rather than the past. In nine studies across four different contexts (Germany, US, Italy, Australia; N = 3405), we analyzed whether these three facets of HC (defensive, reconciliatory and future-oriented) indeed substantially differ. Contrary to expectations, nomological network analyses suggested that all three facets reflect the same defensive desire (Studies 1a–2c) and are perceived as overall similar from a third-party perspective (Study 3). Finally, all three HC facets showed a positive trend toward costly avoidance of confrontation with the ingroup’s perpetrator past (Studies 4a–c). We discuss implications for (and against) a more nuanced understanding of the demand for HC.
... Cehajic et al. (2008) argue that the reason behind the negative relationship between ingroup identification and forgiveness in conflict contexts may be that forgiveness as an act may be seen as disloyal by ingroup members. On the other hand, some other studies also show that ingroup identification, particularly with higher-order groups such as national identity (Baysu & Duman, 2016) or even more inclusive humanity identity (Greenaway et al., 2011;Noor et al., 2010;Wohl & Branscombe, 2005; but see Ünal et al., 2022), can be a facilitator of forgiveness. ...
Article
People's religious identity is often the central identity in many ethnopolitical conflicts. These identities in conflict contexts may be associated with how people see conflict and their willingness to forgive the outgroup members for their wrongdoings in the past. Study 1 (N = 287) tested how religious group identification in the Northern Irish context predicted forgiveness through the endorsement of dominant conflict narratives (i.e., terrorism and independence narratives) among Protestants and Catholics. We also tested how group membership may moderate these relationships. The results showed that among Protestants, higher Protestant identification predicted less forgiveness through higher endorsement of the terrorism narrative and less endorsement of the independence narrative. Among Catholics, on the other hand, higher Catholic identification predicted stronger endorsement of the independence narrative, and in turn, less forgiveness. Study 2 (N = 526) aimed to replicate the findings of Study 1 with a larger sample and extend them by testing the role of an alternative conflict narrative (i.e., the Northern Irish identity narrative). The results were largely replicated for the independence and terrorism narratives, and the Northern Irish identity narrative was associated with higher forgiveness across both groups. We discuss the results in terms of how ingroup identities and conflict narratives can become both facilitators of and barriers to peacebuilding in post-conflict societies.
... À l'inverse, la résolution de conflit intergroupe passe par la paix déclarée, mais aussi par l'augmentation de l'humanité attribuée à l'autre groupe. Il est effectivement nécessaire de percevoir l'humanité chez les autres pour améliorer les relations intergroupes, obtenir la paix et adopter des comportements prosociaux plus positifs (Tutu, 1999;Wohl & Branscombe, 2005). ...
Thesis
Réduire la stigmatisation des personnes en situation de handicap est un enjeu de société important mais complexe, qui nécessite des interventions efficaces. Cette question de la stigmatisation préoccupe également les entreprises. Pour favoriser l'accès à l'emploi des personnes en situation de handicap et leur maintien en emploi, il semble nécessaire de réduire leur stigmatisation. Les travaux de recherche sur les relations intergroupes des cinquante dernières années ont montré que les personnes en contact avec des membres exogroupes, réduisent les stéréotypes et préjugés à leur égard, dès les premiers contacts, et face à des groupes variés. Par la suite, d'autres études se sont intéressées au contact indirect comme le contact via une vidéo ou le contact imaginé. Ces contacts indirects seraient également efficaces pour réduire les préjugés face à un exogroupe, mais parfois dans une moindre mesure. Pour renforcer les études sur le contact intergroupe comme moyen de réduire les préjugés face au handicap, nos travaux se sont intéressés aux contacts imaginé, vidéo et incarné, avec comme objectif d'adresser certaines limites existantes dans ce champ de recherche. La thèse principale défendue ici stipule que les effets du contact imaginé, relevant principalement de l'élaboration et de l'imagination, se maintiendraient à long terme. De plus, outre la réduction des préjugés par un changement de perspective cognitive (i.e. contact vidéo et imaginé), le changement de perspective corporelle serait également un moyen efficace de réduire les préjugés (i.e. contact incarné). Ces interventions sont efficaces pour réduire les préjugés face au handicap chez des salariés d'entreprise, et donc généralisables hors du milieu sanitaire et médico-social. Pour soutenir cette thèse, les effets du contact imaginé ont été étudiés en lien avec le niveau d'élaboration cognitive de la tâche d'imagination, afin de comprendre la place de ce processus dans la diminution des attitudes négatives face à la maladie mentale (étude 1). Les effets du contact imaginé ont ensuite été étudiés à moyen terme (étude 2) puis, afin d'en généraliser au maximum les effets, à plus long terme et chez des salariés (étude 3). Le contact vidéo a également été étudié à court, moyen et long terme chez des salariés (étude 3). Par la suite, une action de sensibilisation proposée en entreprise, visant à permettre aux salariés de changer de perspective à un niveau corporel, en réalisant un atelier de cuisine en situation de handicap, a été testée (étude 4). Enfin, de façon plus exploratoire et fondamentale, le contact incarné a été étudié par la mise en situation de handicap dans une tâche d'immobilisation, chez des étudiants (étude 5).Nos résultats ont permis de renforcer la validité du contact imaginé, comme relevant principalement de l'imagination et de l'élaboration dans la tâche, et de fait pas d'un effet de demande expérimentale. Les effets du contact imaginé sur la réduction de la stigmatisation de la maladie mentale ont également été montrés à long terme, sur plus de six mois, et ont été généralisés à des salariés d'une grande entreprise du secteur de l'énergie (i.e. hors du secteur sanitaire et médico-social). Toutefois, le contact imaginé n'a pas permis de réduire les biais implicites face à la maladie mentale dans notre étude et les effets du contact vidéo semblent faibles. En outre, le contact incarné semble une piste prometteuse pour réduire les attitudes implicites face au handicap. Cette thèse CIFRE avait également pour objectif de proposer le contenu d'un outil de formation en ligne, à destination de tous les salariés de l'entreprise (GRDF). Ce cahier des charges a été rédigé au regard des résultats de ces travaux, et plus largement des avancées scientifiques dans le champ des relations intergroupes. Finalement, l'ensemble de ces résultats est discuté et des pistes de recherches futures sont proposées.
... The literature reviewed above highlights that a group's traumatic history may elicit overly defensive reactions to present conflicts in subsequent generations of both victim and perpetrator groups, especially when this history is viewed in monolithic and simplistic terms that are predicated on threat. Subsequent generations of victims tend to become more defensive when they see the trauma as evidence that the world is against them (Bar-Tal & Antebi, 1992a, 1992b or does not share in their suffering (Vollhardt, 2013;Wohl & Branscombe, 2005), when they essentialize the perpetrator group and view them as inherently evil (Campbell & Vollhardt, 2014), or when they perceive current threats as indistinguishable from past traumas (Canetti et al., 2018;Lifton, 2005). Subsequent generations of perpetrators tend to become more defensive when the trauma, posing an unbearable threat to the moral image and identity of the group, feeds into defensive representations of history (Hirschberger, Kende, & Weinstein, 2016). ...
Article
Full-text available
Collective memories of trauma can have profound impact on the affected individuals and communities. In the context of intergroup conflict, in the present article, we propose a novel theoretical framework to understand the long-term impact of historical trauma on contemporary intergroup relations from both victim and perpetrator perspectives. Integrating past research on intergroup conflict and the biopsychosocial model of threat and challenge, we argue that people appraise their group’s past victimization and perpetration differently, either as a threat or as a challenge. Shaped by contextual factors and individual differences, these differential appraisals will subsequently influence how group members respond to contemporary intergroup conflict, with both adaptive and maladaptive consequences. This model contributes to unifying the previous research that has shown diverse effects of historical trauma on present-day intergroup dynamics. We present preliminary empirical evidence in support of the framework and discuss its theoretical and practical implications.
Article
In the aftermath of collective violence, reconciliation is supposed to ensure that violent conflict does not re-erupt after official treaties have brought it to a halt. This requires attention to the psychological processes that were shaped by the group’s role as victims or perpetrators of violence, as well as to the sociopolitical context in the aftermath of violence. Based on central conceptualisations of reconciliation in political and social psychology, we review five broad categories of psychological processes that obstruct or facilitate reconciliation in the aftermath of collective violence: Identity and identity threats, collective memories of collective victimhood, acknowledgement versus denial of collective victimisation, emotions, and justice. We discuss these processes from the perspective of victim and perpetrator groups, and review examples of interventions that utilise these processes with the aim of increasing willingness for reconciliation and positive outcomes of reconciliation in the aftermath of collective violence in different contexts.
Article
Understanding the underpinnings of climate justice, especially justice in carbon allocation, is paramount for international cooperation in coping with climate change. Previous work has attempted to promote carbon allocation justice based on a utilitarian theory of justice, but it has backfired. Adopting the per capita approach in defining justice suggested by a Rawlsian theory of justice, the current research addresses the psychosocial processes underlying justice in carbon allocation. Inspired by the social identity approach, we propose that whether people choose to behave justly in carbon allocation originates, in part, from their perceptions about whether an outgroup shares similar humanness with the ingroup. We conducted four studies (N = 1,326) to test this assumption. The results indicated that humanization increased the allocation of carbon credits to an outgroup (Study 1), increased the amount of carbon credits contributed and decreased free-riding behavior (Studies 2 & 3), and reduced carbon emissions and over-emitting behavior (Study 4) in carbon allocation; the effects were mediated by an expanded scope of justice (Studies 1–4). By identifying the roles of humanization and the scope of justice in carbon allocation justice, this research provides a psychosocial framework for understanding climate justice, which has implications for guiding the priorities of policymakers.
Article
Full-text available
In this article, we attempt to distinguish between the properties of moderator and mediator variables at a number of levels. First, we seek to make theorists and researchers aware of the importance of not using the terms moderator and mediator interchangeably by carefully elaborating, both conceptually and strategically, the many ways in which moderators and mediators differ. We then go beyond this largely pedagogical function and delineate the conceptual and strategic implications of making use of such distinctions with regard to a wide range of phenomena, including control and stress, attitudes, and personality traits. We also provide a specific compendium of analytic procedures appropriate for making the most effective use of the moderator and mediator distinction, both separately and in terms of a broader causal system that includes both moderators and mediators. (46 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Chapter
Emotion can result from interpreting group actions as reflecting on the self due to an association between the two. This volume considers the nature of collective guilt, the antecedent conditions necessary for it to be experienced, how it can be measured, as well as how collective guilt differs from other group based emotions. Research from Australia, Canada, Germany, Israel, the Netherlands, Northern Ireland, and the USA addresses critical questions concerning the who, when, and why of the experience of collective guilt. The political implications of collective guilt and forgiveness for the past are considered, and how those might depend on the national context. How collective guilt can be harnessed and used to create a more peaceful future for groups with a history of violence between then is emphasized.
Chapter
Emotion can result from interpreting group actions as reflecting on the self due to an association between the two. This volume considers the nature of collective guilt, the antecedent conditions necessary for it to be experienced, how it can be measured, as well