ArticlePDF Available

Nie Wieder: Group-Based Emotions for In-Group Wrongdoing Affect Attitudes toward Unrelated Minorities



This article focuses on the effects of group-based emotions for in-group wrongdoing on attitudes towards seemingly unrelated groups. Two forms of shame are distinguished from one another and from guilt and linked to positive and negative attitudes towards an unrelated minority. In Study 1 (N = 203), Germans' feelings of moral shame—arising from the belief that the in-group's Nazi past violates an important moral value—are associated with increased support for Turks living in Germany. Image shame—arising from a threatened social image—is associated with increased social distance. In Study 2 (N = 301), Britons' emotions regarding atrocities committed by in-group members during the war in Iraq have similar links with attitudes towards Pakistani immigrants. We extend the findings of Study 1 by demonstrating that the effects are mediated by a sense of moral obligation and observed more strongly when the unrelated group is perceived as similar to the harmed group. Guilt was unrelated to any outcome variable across both studies. Theoretical and practical implications about the nature of group-based emotions and their potential for affecting wider intergroup relations are discussed.
Running head: NEVER AGAIN
Nie wieder: Group-based emotions for ingroup wrongdoing affect attitudes toward unrelated
Jonas H. Rees, Department of Psychology, University of Bielefeld, Germany;
Jesse A. Allpress and Rupert Brown, School of Psychology, University of Sussex, United
(forthcoming) Political Psychology
Author Note
This research was supported by a German Academic Exchange Service and a German
National Academic Foundation scholarship awarded to the first author.
The authors are grateful to Gerd Bohner and Rüdiger Schmidt for their permissions to collect
data for Study 1 in the University of Bielefeld and the Citizens Advice Bureau Bielefeld.
Parts of this paper were presented at the 16
General Meeting of the European Association for
Social Psychology, Stockholm, Sweden.
Address correspondence to Jonas Rees, Department of Psychology, University of Bielefeld,
P.O. Box 100131, 33501 Bielefeld, Germany. E-mail:
Running head: NEVER AGAIN
This paper focuses on the effects of group-based emotions for ingroup wrongdoing on
attitudes towards seemingly unrelated groups. Two forms of shame are distinguished from
one another and from guilt, and linked to positive and negative attitudes towards an unrelated
minority. In Study 1 (N = 203), Germans’ feelings of moral shame – arising from the belief
that the ingroup’s Nazi past violates an important moral value – are associated with increased
support for Turks living in Germany. Image shame – arising from a threatened social image
is associated with increased social distance. In Study 2 (N = 301), Britons’ emotions regarding
atrocities committed by ingroup members during the war in Iraq have similar links with
attitudes towards Pakistani immigrants. We extend the findings of Study 1 by demonstrating
that the effects are mediated by a sense of moral obligation and observed more strongly when
the unrelated group is perceived as similar to the harmed group. Guilt was unrelated to any
outcome variable across both studies. Theoretical and practical implications about the nature
of group-based emotions and their potential for affecting wider intergroup relations are
Keywords: Group-based emotions, shame, guilt, social distance, moral obligation
Running head: NEVER AGAIN
It is a frequently expressed view that collective remembrance of large scale atrocities,
such as the Holocaust perpetrated by Germans or war crimes committed by one group against
another, will prevent future transgressions from happening. For example, in a speech to the
German parliament on 27 January 2010, International Holocaust Remembrance Day, this idea
was encapsulated in what Shimon Peres, then President of Israel, called the “decisive lesson
[from the Holocaust]: ‘Never again’.” Similarly, Horst Köhler, then Federal President of
Germany, in a speech marking the 60th anniversary of the end of the Second World War in
2005, said: “We bear the responsibility to keep the memory of all this suffering and its causes
alive and we have to ensure that it never happens again.” In this paper we seek to examine the
effects of reminders of ingroup wrongdoing on wider contemporary intergroup relations. We
investigate the effects of such reminders in two quite different contexts: the Holocaust in
Germany, and the abuse of Iraqi civilians by British forces in Iraq.
We propose that collective remembrance of ingroup wrongdoing – past or present –
may play an important role in determining how perpetrator group members relate to
outgroups, even when these outgroups are unrelated to the original wrongdoing. If
remembrance of the ingroup’s wrongdoing does indeed influence how group members
confront unrelated outgroups, then this represents an important and hitherto neglected factor
influencing contemporary intergroup relations. Although we are proposing that remembrance
of ingroup transgressions plays an important role in preventing future wrongdoing, it is
possible that not all individuals respond equally to reminders of ingroup misdeeds. It might be
that, rather than reducing the tendency to offend in the future, some group members will react
negatively to such reminders, adopting a defensive and antagonistic stance. What drives an
individual to react in a certain manner? We propose that the type of emotional reaction a
person experiences in response to such reminders may play a crucial role in determining how
Running head: NEVER AGAIN
an individual responds to the ingroup’s misdeeds. We investigate this assertion in the present
Implicit in that famous phrase, never again (or “Nie wieder” in German), is the
assumption that people’s feelings and thoughts about harmful injustice done to one group will
be linked to their attitudes towards other unrelated groups. This should be true especially for
individuals directly or indirectly associated with that injustice who, because of their group
identifications, may continue to feel a sense of guilt or shame for their ingroup’s misdeeds
(Doosje, Branscombe, Spears, & Manstead, 1998). However, research on such collective or
group-based emotions has so far focused exclusively on the link between collective guilt and
shame and restitutional responses, such as willingness to apologise or make reparations to the
harmed outgroup (Branscombe & Doosje, 2004a; Brown, 2009; Iyer & Leach, 2008; Wohl,
Branscombe, & Klar, 2006).
To grasp the broader consequences of the remembrance of crimes perpetrated by the
ingroup, we need to understand not only the complexity of group-based emotions but also
their associations with people’s attitudes towards outgroups other than those originally
victimised. By doing so, the present paper is the first we know of to empirically study a
phenomenon that has long been assumed but is far from trivial: the link between group-based
emotions concerning harm done to one outgroup and attitudes towards another, unrelated
minority. We first propose an analytical distinction between two different forms of group-
based shame and guilt and provide evidence for it. We then show how these emotions are
differentially related to attitudes towards a minority outgroup not originally associated with
the transgression.
Interpersonal and group-based guilt and shame
Guilt and shame have often been conceptualised as separate emotions with different
phenomenologies and different effects (Gausel & Leach, 2011; Teroni & Deonna, 2008;
Tracy & Robins, 2006). An early contribution was by Lewis (1971) who suggested that, while
Running head: NEVER AGAIN
both emotions were self-focused and arose from some perceived moral violation, they differed
in the degree to which the self is implicated. She argued that guilt results from a focus on the
misdeed and how that act affected the other, and shame results from a feeling that the self as a
whole is morally flawed. Lewis therefore argued that guilt should provoke attempts to make
restitutions in some way, whilst shame will result in hiding, avoidance or even self-defensive
Others have made similar arguments about the sense of global worthlessness provoked
by shame and a more action-oriented sense of responsibility implied by guilt (e.g.,
Branscombe, Slugoski, & Kappen, 2004; Tangney & Dearing, 2002). Relatedly, some authors
have suggested that guilt is more other-focused than shame. Baumeister and his colleagues,
for instance, propose that guilt serves to strengthen and maintain social relationships by
motivating perpetrators to make restitutions to those they have wronged (Baumeister,
Stillwell, & Heatherington, 1994). Another distinction has consisted in underlining the more
private nature of guilt relative to shame’s more public character (Ausubel, 1955; Fontaine et
al., 2006; Gehm & Scherer, 1988; Smith, Webster, Parrott, & Eyre, 2002). It has also been
suggested that guilt stems from norm violations while shame derives more strongly from a
failure to live up to certain central values (Higgins, 1987; Lynd, 1956; Teroni & Deonna,
2008). In sum, shame is often characterised as self-defensive and having negative reactions
towards the victim, while guilt is usually conceptualised as an emotion with more pro-social
In some studies of interpersonal relations, shame has, indeed, been linked to self-
blame and avoidance responses (Miller & Tangney, 1994; Roseman, Wiest, & Swartz, 1994;
Smith et al., 2002; Tangney, Miller, Flicker, & Barlow, 1996), and guilt to pro-social
tendencies (Niedenthal, Tangney, & Gavanski, 1994; Tangney, 1991; Tangney et al., 1996;
Tangney, Wagner, Fletcher, & Gramzow, 1992). However, reparation is not always associated
more with guilt than with shame (Roseman et al., 1994; Tangney et al., 1996). Indeed, de
Running head: NEVER AGAIN
Hooge, Breugelmans, and Zeelenberg (2008) found that shame episodes motivated
relationship-enhancing interpersonal behaviour in situations relevant to the initial shame
episode. There is also evidence that guilt is sometimes linked to maladaptive outcomes (de
Hooge, Nelissen, Breugelmans, & Zeelenberg, 2011; Fedewa, Burns, & Gomez, 2005; Luby
et al., 2009; Meehan et al., 1996). So, not only can shame have adaptive, relationship-
enhancing effects at the interpersonal level, but also guilt can have negative effects.
Research on group-based emotions reveals similar ambiguities. Some research has
found that people’s feelings of guilt about misdeeds committed by their ingroup are, indeed,
associated with wishes to make reparations to the harmed outgroup (e.g., Allpress, Barlow,
Brown, & Louis, 2010; Allpress & Brown, 2011; Brown & Čehajić, 2008; Brown, González,
Zagefka, Manzi, & Čehajić, 2008; Doosje et al., 1998; Harvey & Oswald, 2000; Iyer, Leach,
& Crosby, 2003; McGarty et al., 2005; Swim & Miller, 1999). However, Iyer, Schmader, and
Lickel (2007) found that feelings of guilt about the invasion of Iraq did not predict American
and British students’ support for any reparative actions once shame and anger were controlled
for (see also, Leach, Iyer, & Pederson, 2006; Harth, Kessler, & Leach, 2008). Also, Allpress
and colleagues (2010) observed that, in the context of Australia’s apology to Aboriginal
Australians, shame was a stronger positive correlate of collective political action than guilt.
Research on the effects of group-based shame has provided mixed evidence. Lickel,
Schmader, Curtis, Scarnier, and Ames (2005) and Schmader and Lickel (2006) showed that
shame for others’ misdeeds was associated with a desire to distance oneself from both the
situation and those responsible for the wrongdoings (see also, Johns, Schmader, & Lickel,
2005; Iyer et al., 2007). However, other studies have observed positive associations between
group-based shame and reparation attitudes and other pro-social responses (Allpress et al.,
2010; Brown et al., 2008; Brown & Čehajić, 2008; Gausel, Leach, Vignoles, & Brown, 2010).
These findings pose something of a puzzle for the traditional view that guilt and shame should
have opposite consequences for social relations.
Running head: NEVER AGAIN
Moral and image shame. Elsewhere, we have argued that these inconsistencies may
be resolved by distinguishing between two different varieties of shame (Allpress & Brown,
2011; Allpress et al., 2010). There, and in the present paper, we follow the theoretical work of
Teroni and Deonna (2008) and Deonna, Rodogno, and Teroni (2011), who provide a novel
way of distinguishing between shame and guilt to help solve the inconsistencies in the shame
literature. Deonna and colleagues’ distinction sees shame as intimately connected to the
violation of self-relevant values and guilt as connected to the violation of normative
prohibitions. The significant advantage of this distinction is that it explains the occurrence of
shame in seemingly diverse and disparate situations. As shame is connected with values,
feelings of shame may arise in any circumstance in which an individual feels that the
behaviour or situation in question seriously undermines an important value. This
conceptualisation of shame implies that it can arise in relation to many different values, and it
is therefore unlikely that shame will always lead to the same type of response. Rather, the
motivational effects of shame will be determined by the value to which the feeling of shame is
connected in that particular situation (For a similar argument that, depending on context,
different outcomes may be connected to one single emotion, see Reifen Tagar, Federico, &
Halperin, 2011).
What might determine the direction of shame’s motivational effects? Shame is an
aversive experience, regardless of its origins. Given its aversive nature, people will be
motivated to think and act in ways that best reduce their levels of shame. In line with this
view, the effects of any particular form of shame can be seen as representing a ‘coping
strategy’, by which people seek to reduce the negative shame experience. Although an
individual may have a number of different coping strategies available to them, they are likely
to choose the strategy that they believe is most likely to reduce their shame. What determines
which strategy is likely to be most effective? The answer to this question depends on the
value to which shame is connected. As shame arises when a self-relevant value has been
Running head: NEVER AGAIN
seriously violated, an individual is most likely to reduce the feelings of shame by adopting
attitudes and behaviours that reduce the threat to this value and allow one or one’s own or the
group’s actions to be seen as living up to or consistent with this value. Which attitudes and
behaviours achieve these goals is likely to differ for different values, and therefore, types of
shame. It is for this reason that we predict that different forms of shame will have different
attitudinal and behavioural effects – individuals will strive to reduce these feelings in different
ways because different forms of shame are related to different values.
In previous work (Allpress & Brown, 2011; Allpress et al., 2010) we have proposed
that it is important to differentiate between two particular forms of shame: image shame and
moral shame. We make this distinction because we believe that the values of image and
morality are especially important in the context of intergroup relations. Feelings of shame that
arise from the sense that the ingroup’s social image has been seriously tarnished – which we
call image shame – should have different effects from shame that arises from the belief that
the ingroup’s actions violate an important moral value – which we call moral shame. For
image shame, because one’s concern is solely with the image and reputation of the group –
and not the well-being of the victim group – the easiest way to restore one’s valued social
image, and to reduce the shame that arises from the judgement of others, is likely to involve
avoidance and withdrawal from the critical gaze of others, in a hope that the issue will blow
over. Such a strategy can be successful considering the short attention span of the world’s
media and its average consumer. An individual in this situation might also, if they believe
such a strategy will reduce or avoid further external blame of the group, adopt a defensive
strategy of actively covering up the group’s misdeeds.
It is also possible that an individual experiencing image shame might view a strategy
of (disingenuous) support for apology and superficial acts of restitution as a useful way to
bolster the ingroup’s reputation amongst third parties. This strategy is less likely to occur,
however, because acts of apology and compensation usually commit the ingroup to further
Running head: NEVER AGAIN
and prolonged acts of restitution. As the concern in image shame lies with the restoration of
the ingroup’s image, and not with the well-being of the victim group, an individual
experiencing this form of shame is unlikely to support opening themselves and their group up
to further reparative commitments. The initial preference, therefore, is likely to reflect an
image-maintenance strategy characterised by avoidance and cover-up. We would expect that
it is only once this initial preference is shown to be ineffective that alternative strategies will
be adopted.
It is not as easy, however, to avoid and forget transgressions giving rise to moral
shame, because it is the immoral acts themselves – rather than the criticisms of others – that
have high self-importance. The coping strategy that is adopted in relation to moral shame will
therefore reflect the particular importance of one’s moral values. In this case, avoidance of the
issue is unlikely to restore one’s personal value of morality, and is therefore likely to be a less
fruitful strategy for reducing moral shame. In order for moral shame to be reduced, an
individual must once again be able to see themselves
(or their group) as acting in a manner
that is consistent with their moral values. For this reason, an individual is most likely to adopt
attitudes and behaviours that address the initial wrongdoing and restore the well-being of the
victim group. Such a strategy is likely to include a (genuine) support for apology and
compensations, and openness to discussing and addressing the original transgression.
The theoretical account of Deonna and colleagues also provides insight into the
motivational effects of guilt. Their approach suggests that because of guilt’s connection with
normative prohibitions, the reparation that is envisaged in guilt is likely to be limited to the
compensation for the harm done, rather than further reaching soul-searching and self-
Note, the self in moral shame is different from the self in Lewis’ (1971) popular conception of shame. Here, the
self is important because of the importance of values for how we see ourselves. In Deonna and colleagues’
conception of shame, the aspect of the self that is the focus of shame is related specifically to the value in
question. Lewis, on the other hand, sees the self-focus of shame as global and all-encompassing, and for this
reason predicts that shame will be overwhelming and will lead to withdrawal. It is the specific nature of the self-
focus in moral shame, we believe, that allows an individual to face their (group’s) immorality in a positive and
adaptive way.
Running head: NEVER AGAIN
questioning that moral shame invites. Because guilt arises from societal prohibitions, rather
than internalised moral convictions, the pro-social effects of guilt are likely to be limited. This
latter point is of importance for the distinction between moral shame and guilt. It is likely,
given their respective causal factors, that the effects of guilt will be much weaker than those
of moral shame. This is because the threat to one’s valued self-conception as moral is likely to
prove to be a more potent motivator of reparative behaviour than concern with normative
Evidence in support of these ideas was provided by Allpress and Brown (2011; see
also Allpress et al., 2010, Study 2). In three studies, Allpress and Brown (2011) demonstrated
that the two forms of shame could be empirically distinguished and were differently
correlated with orientations towards the outgroup: image shame had clear associations with
negative orientations towards the harmed outgroup whereas moral shame had associations
with positive outgroup orientations. In line with predictions, guilt was found to have less
consistent associations with either kind of orientation, being weakly associated with support
for apology and compensation in only one of the three studies.
The theoretical rationale of Gausel and Leach (2011) and findings of Gausel and
colleagues (2010) are also generally supportive of the ideas we present here, although these
authors advance a different conceptualisation of shame
. Focusing on contemporary
Norwegians’ feelings about the mistreatment of the Tater minority, Gausel and colleagues
were able to separate shame from feelings of inferiority and social rejection. The latter
emotion was positively correlated with tendencies to avoid or cover up the Tater historical
issue and negatively related to pro-social orientations towards the Tater. In contrast, shame
Gausel and colleagues propose that shame can only arise in response to the perception that one’s behaviour
represents a specific moral failure. They argue that shame cannot occur for other reasons and that shame reported
for any non-moral reason is not in fact a feeling of shame but some other emotion. They focus, in particular, on
responses to social criticism and claim that any feelings of shame that arise for this reason are not feelings of
shame but feelings of rejection. We address this issue elsewhere (XXXX), showing that not only is image shame
distinct from feelings of rejection, but that it is image shame, and not rejection, that is connected to negative
responses to ingroup wrongdoing.
Running head: NEVER AGAIN
revealed an exactly opposite pattern: a negative correlation with avoidance and cover-up, and
a positive association with pro-sociality. Guilt proved largely unrelated to either outcome
measure in these studies.
Building upon the theoretical work of Teroni and Deonna (2008) and Deonna and
colleagues (2011), as well as our own empirical substantiation of this distinction (Allpress &
Brown, 2011; Allpress et al., 2010), we seek to investigate how image shame, moral shame
and guilt influence individuals’ responses to unrelated minorities within their societies. In
order to investigate these influences and to ensure the generalisability of our findings, we
conducted two studies in two different countries: Germany and the United Kingdom. Study 1,
was conducted in Germany, and investigated how contemporary Germans’ feelings of shame
and guilt in relation to the Holocaust are linked to their attitudes towards contemporary
Turkish immigrants in Germany. In Study 2, we investigated the relations of Britons’
emotional responses to the abuse of Iraqi civilians by British soldiers during the most recent
invasion of Iraq with their attitudes towards contemporary Pakistani immigrants in the U.K.
There is one final issue to be raised before turning to the empirical contribution of this
paper. In many studies of intergroup relations, positive and negative orientations towards
outgroups are treated as two sides of the same coin. Typically, researchers will measure
ingroup bias, outgroup hostility and various other emotions and regard them all as equally
useful and essentially equivalent indices of intergroup attitudes (Brown, 2010). Yet, in the
past two decades, it has become apparent that indicators of a negative orientation towards
outgroups may not be the same as indicators of a less positive orientation towards that same
outgroup. Brewer (1999) has argued persuasively that outgroup “hate” should not be conflated
with ingroup “love”. And Mummendey and Otten (1998) have shown in a sustained
programme of research that there is some asymmetry between positive and negative attitudes
towards the outgroup (see also, Gardham & Brown, 2001). In view of these arguments, in the
first study to be presented here, we include both positive and negative indicators of outgroup
Running head: NEVER AGAIN
attitudes. Conveniently, this is also consistent with our predictions for the divergent effects of
image and moral shame.
The present research
The main aims of the present paper are to, a) contribute to our understanding of group-
based emotions by further substantiating the separability of guilt, moral shame and image
shame, b) demonstrate the hypothesised influence of group-based emotions about an incident
involving one particular outgroup (Jews in Study 1, Iraqis in Study 2) on attitudes towards
another current minority outgroup (Turks in Study 1, Pakistanis in Study 2), and c) identify
conditions under which the aforementioned generalising effects of group-based emotions on
intergroup attitudes occur or are amplified. We conducted the following research in Germany
– in the context of unthinkable crimes during the Holocaust in the past (Study 1) – and in the
United Kingdom – in the context of abuse of Iraqi prisoners during the recent invasion of Iraq
(Study 2). Both Germany and the UK show evidence of ongoing discrimination against ethnic
minorities (BBC, 2004; Department for Work and Pensions, 2009; Human Rights Watch,
1995; Wagner et al., 2008).
Study 1
The response of many Germans born after 1945 to the atrocities committed during the
Nazi era can be considered a paradigmatic example of collective guilt
and shame. Even
though these post-war generations of Germans could have no personal responsibility for the
Holocaust, the mere fact of their German identity is often enough to evoke such self-
conscious emotions (Branscombe & Doosje, 2004b; Doosje et al., 1998). Interestingly,
however, there is little psychological research into German group-based shame and guilt. In
one exception, Peetz, Gunn, and Wilson (2010) found that German participants who distanced
Note, however, that the German Kollektivschuld (collective guilt) is a highly emotionally laden term differing
in meaning from the technical way collective or group-based guilt is used here. Whereas it has been employed to
(ostensibly or effectively) accuse all Germans of being guilty of the Holocaust – in the contexts of
Historikerstreit (Knowlton & Cates, 1993) and particularly the Goldhagen controversy (Rensmann, 1999) – we
are referring to it to describe the experience of emotions merely because of one’s group-membership but without
personal involvement (Doosje et al., 1998).
Running head: NEVER AGAIN
the event further into the past experienced less guilt, as a defensive reaction to being reminded
of the Holocaust. Dresler-Hawke and Liu (2006) established positive links between young
Germans’ shame and their willingness to confront and accept responsibility for the Nazi past.
The authors’ conclusion that young Germans are struggling to reconcile their German identity
with feelings of shame regarding the Nazi era demonstrates the importance of this topic in
Germany to this day. Similarly, Rensmann (2004) in a review of qualitative and quantitative
data on collective guilt and national identification in Germany argued that how guilt
“influences collective identification in contemporary Germany touches on the very core of
German social identity” (p. 169). In sum, Germany is an ideal environment for the study of
group-based emotions. Interestingly, however, psychological research in the German context
has only recently begun to emerge (see Imhoff & Banse, 2009; Peetz et al., 2010;
Zimmermann, Abrams, Doosje, & Manstead, in press, Studies 4 and 5).
The Holocaust and contemporary debates about Turkish people in Germany
In the now culturally diverse contemporary Germany, it is not uncommon to hear
references to the Holocaust in seemingly unrelated debates, for example, about ethnic
minorities and with regard to immigration. Anecdotally, references to the Holocaust appear to
be used by social commentators in an attempt to add emotive force and support to their
argument. Interestingly, this tactic seems to be used by both supporters and opponents of
immigration, such that the Holocaust is portrayed by opponents as an unfair burden and
barrier to national pride that forces Germans to submit to a tide of immigration, and is
referenced by supporters as a reminder of how not to treat people and the dangers of
xenophobic rhetoric (see Cohen, 1998; Kulish, 2010; for a comprehensive overview of how
reference to the German past is instrumentalised and how anti-Semitism and other ideologies
are mingled in contemporary political debates, see Rensmann, 2005). The largest of these
minority groups living in Germany today are Turks with more than a fourth of the 8.2% of
non-German population in 2009 originating from Turkey and an even greater number of
Running head: NEVER AGAIN
“Germans with Turkish migration background” (German Federal Statistical Office, 2009).
Prejudice and discrimination, whether subtle or blatant, are still prevalent problems in
Germany today and Turkish immigrants are amongst the commonly targeted minority groups
(Wagner, Christ, & Pettigrew, 2008).
While it seems that references to the Holocaust are used by politicians and public
figures to support and add emotional content to their arguments, what is less clear is how such
allusions might consequently influence Germans’ attitudes and emotional reactions within
contemporary contexts, particularly with regard to contemporary minority groups. We set out
to test the idea that Germans’ feelings of shame or guilt for the Holocaust can be linked to
attitudes and feelings towards another outgroup, Turkish immigrants living in Germany today.
The effect of guilt and shame for the Holocaust on contemporary intergroup relations
With the Nazi past deeply embedded in the German cultural identity and essential
parts of the German national curriculum devoted to teaching and discussing the topic in
school, Germans cannot help but be aware of their past (Buruma, 2002; Knowlton & Cates,
1993; Peetz et al., 2010; Rensmann, 2004). Frequent reminders of the Holocaust in public
discourse or reference to a special burden Germans must never forget suggest that emotional
reactions to historical wrongdoing may – or are commonly assumed to – affect contemporary
intergroup relations. We argue that group-based emotions experienced in response to the
Holocaust can in fact influence attitudes towards a contemporary outgroup not connected to
the original event. We refer to this phenomenon of emotional reactions to one specific event,
in this case the Holocaust, affecting attitudes towards seemingly unrelated outgroups as
generalisation. We expect the specific emotions (moral shame, image shame and guilt) to
motivate similar responses to a contemporary minority group as they do to the victim group
itself: as such, individuals concerned about their ingroup’s flawed morality can be expected to
react more pro-socially towards another outgroup, borne, in part, of a sense of moral
obligation for past wrongs. Individuals concerned primarily about damage to the ingroup’s
Running head: NEVER AGAIN
image as a result of past wrongdoing, on the other hand, should be more prone to avoidance-
based, anti-social reactions to an unrelated minority group, possibly in an attempt to avoid
less powerful outgroup members who may threaten the power differential by evoking the
ingroup’s misdeeds, or possibly to repair their social identity by derogating others
(Branscombe, Schmitt, & Schiffhauer, 2007). Theoretically, group-based guilt should provoke
a limited pro-social response to the victim group, but because the concern with guilt is simply
with repairing the specific damage done to the victim group, it will not motivate attitudes or
behaviour towards unrelated outgroups.
Hypotheses. (1) Group-based emotions – guilt, moral shame and image shame – are
expected to be clearly separable empirically. (2) Moral shame with regard to the Holocaust is
predicted to be negatively linked to social distance from an unrelated outgroup, Turks living
in Germany today. Conversely, the same emotion is predicted to be positively related to
supportive attitudes towards Turkish people. Image shame is predicted to function in an
opposite fashion.
Based on Deonna and colleagues’ (2011) theoretical analysis and previous research
(Allpress & Brown, 2011; Allpress et al., 2010; Gausel et al., 2010), we did not expect guilt to
be related to attitudes towards Turks.
Two hundred and three German participants, ranging from 18 to 71 years of age (M =
27.5, SD = 9.8), were approached in either a university in a larger German city (N
= 146)
or in public spaces in the same city (N
= 57) to take part in a study of their opinion on
“Germany in past and present”. Of the full sample, 85 participants (42%) were female and the
modal highest formal degree was the German Abitur (general qualification for university
entrance) with a university degree aimed for or completed (90.1%). Naturally, the two sub-
samples differed on this latter variable (χ
(4) = 49.94, p < .01) with virtually all students
Running head: NEVER AGAIN
having completed further education, but fewer having done so in the general sample (66.7%
university degree attained for or completed; 7% Certificate of Secondary Education; 17.5%
General Certificate of Secondary Education; 7% advanced technical college entrance
Having agreed to take part, participants were informed about their right to withdraw
from the study without incurring any penalties and that anonymity was ensured at all times,
filled out the questionnaire and returned it to the investigator
. Upon completion of the
questionnaire, they were given the opportunity to ask any remaining questions about the study
and thoroughly debriefed. No financial rewards were offered. The study conformed fully to
APA/BPS ethical guidelines.
Guilt. Guilt experienced because of the Holocaust was assessed using a six-item scale
partially adapted from Peetz and colleagues (2010; e.g., “Even if I have done nothing bad, I
feel guilty for the behaviour of Germans towards Jews.”; α = .82).
Shame. The two facets of shame were measured by asking participants how they felt
with regard to the Holocaust. As a preliminary measure, moral shame was assessed with one
item (“I feel ashamed for the damage done to Jewish people by Germans.”) and image shame
was measured with three items (“I feel disgraced because the behaviour of Germans towards
Jewish people has created a bad image of Germany in the eyes of the world.”, “I feel ashamed
when I realise that other countries might think of Germany negatively because of the
Holocaust.” and “Sometimes it shames me how others can see Germany as immoral for our
involvement in the Holocaust.”; α = .90).
In an experimental part of this study, we intended to alter participants’ perception of the Holocaust as either a
closed or open chapter (vs. a control condition without manipulation) using a timeline procedure. This
manipulation did not affect participants’ group-based emotions or any of the variables of interest here.
Running head: NEVER AGAIN
Social Distance. An adapted scale from Wagner and colleagues (2008) was used to
measure social distance from Turks living in Germany (6 items, e.g., “I would have problems
moving into a district where many Turks live.”; α = .85).
Minority support. Conversely, supporting attitudes towards Turks in Germany were
assessed by three items (“I would approve of tax increases in order to support Turks living in
Germany.”, “I, personally, would like to be more involved in supporting Turks living in
Germany.”, “People with Turkish background living in Germany should be supported more
by us Germans.”; α = .78).
Control measures. Age and highest formal degree were assessed as potential control
All items were measured using a six-point scale ranging from “disagree strongly” to
“agree strongly” and presented in randomised order. An overview of the main scales, control
variables and their inter-correlations is given in Table 1.
Differences by sub-sample
As expected, age and highest formal degree were found to differ by sub-sample in
preliminary analyses. It was thus decided to control for age and highest formal degree where
appropriate. No other differences by sub-sample emerged.
The research questions will now be addressed in turn: first, the factorial structure of
the measures will be assessed, and then the hypothesised effects of each emotion on attitudes
towards Turks will be demonstrated.
Running head: NEVER AGAIN
Factorial Structure of Measures
The following confirmatory factor analyses (CFAs) and structural equation modelling
(SEM) were conducted using AMOS 18.0 for Windows. Missing values of the manifest
indicators were mean-substituted
Guilt, shame, support of and social distance from foreigners. To assess the
factorial structure of guilt and shame as well as the validity of the assumed distinction
between support of and social distance from Turks living in Germany, two sets of CFAs were
conducted: one to assess the factorial structure of the three emotions and one to assess the
structure of the two intergroup attitudes. The assumed tripartite structure of guilt, moral
shame and image shame fitted the data well, with items from each scale loading only on their
respective factors (see Table 2). Moreover, this three factor solution (Model 1) was superior
to three alternative models: Model 1a, a two factor model in which both types of shame were
collapsed into a single factor, separate from guilt, Model 1b, another two factor model
combining moral shame and guilt into one factor separate from image shame, and Model 1c,
in which all three emotions were combined into a single “negative emotions” factor
. The
predicted two-factor structure of intergroup attitudes, in which positive and negative
orientations were separated, was also found to fit the data well and to be superior to a single
“general intergroup attitude” factor, all χ
s, p < .01
(See Table 2; Burnham & Anderson,
2002; Hu & Bentler, 1999; MacCallum, Browne, & Sugawara, 1996).
Generalising effects of group-based emotions
The hypothesised effects of each emotion on attitudes towards Turks were then tested
by specifying a model predicting social distance from and support for Turks living in
The number of missing values was well below 1% overall and not exceeding 3% per variable. No systematic
pattern of missing values or correlation of missing values with any of the independent or dependent variables
were found and data was thus assumed to be missing completely at random.
As a rule of thumb, an AIC difference of < 2 indicates no meaningful discrepancy between models; a difference
between 4 and 7 indicates considerable evidence that the model with the lower AIC is better; and a difference of
> 10 indicates substantial support for the model with the lower AIC (Burnham & Anderson, 2002).
CFAs reported here were performed separately for emotions about the Holocaust (guilt, moral shame and
image shame) and attitudes towards Turks (social distance from and support of Turks). CFAs including all these
latent constructs together, however, yielded the same pattern of results.
Running head: NEVER AGAIN
Germany from feelings of guilt, moral shame and image shame regarding the Holocaust.
Predictors and outcome variables were allowed to covary but no single item was allowed to
cross-load. The suggested model fitted the data well, even when controlling for age and
highest formal degree, χ
(169) = 255.55, p < .001, CFI = .959, RMSEA = .050, PCLOSE >
.40, and explained correlations to within an average error of .071 as indicated by the
standardised root mean square residual (SRMR). Fit indices were thus well in line with the
cut-off criteria suggested in the literature (Hu & Bentler, 1999; MacCallum et al., 1996). The
model is presented in Figure 1.
A clear generalisation of moral and image shame but not of guilt on attitudes towards
Turks was observed: moral shame significantly and positively predicted supportive attitudes
(β = .29, p < .01) whereas image shame significantly and positively predicted social distance
from Turks (β = .35, p < .01). In addition, a negative path from moral shame to social distance
(β = -.18, p = .07) was very close to conventional levels of significance. A negative path from
image shame to supportive attitudes (β = -.12) and both paths from guilt were not significant
(ps .10). The overall explained variance in the dependent variables was moderate (R
= .17 and R
2Foreigner Support
= .13).
As hypothesised, group-based emotions about a historical injustice perpetrated by the
ingroup were found to be clearly separable into moral shame, image shame and guilt. These
emotions were then found to be differentially related to attitudes towards a contemporary
minority outgroup. German participants’ moral shame about the Holocaust was positively
associated with support for and negatively associated with social distance from Turks living in
Germany today. Image shame showed an opposite pattern of relationships. Such findings are
difficult to reconcile with some previous accounts of the distinction between guilt and shame
The model held even when controlling for participants’ self-reported political orientation (measured as one
item coded from 1 = “left” to 6 = “right”) which did not change the pattern of interrelations substantially.
Political orientation was consequently discarded as a potential control measure.
Running head: NEVER AGAIN
(e.g., Branscombe et al., 2004; Brown et al., 2008; Tangney & Dearing, 2002) but are
consistent with recent research that has sought to disentangle them more precisely (Allpress &
Brown, 2011; Allpress et al., 2010; Deonna et al., 2011; Gausel et al., 2010). Moreover, they
extend those subsequent analyses by showing that such emotions can influence, or generalise
to attitudes towards a seemingly unrelated outgroup. To our knowledge, studies of group-
based emotions have exclusively focused on how guilt and shame about harm done to one
outgroup affect attitudes and behaviour towards that same outgroup. We demonstrate,
however, that uncoupled from the emotion-eliciting event and outgroup, group-based
emotions can also be related to individuals’ attitudes towards other outgroups as well.
One possible explanation of this process would be a generalised tendency for people to
categorise in terms of group-memberships. This notion that attitudes towards one outgroup are
linked to attitudes towards another outgroup has found its way into the literature under the
label of Group-Focused Enmity (Zick et al., 2008). Zick and colleagues (2008) argue that
different types of prejudice should be “interrelated because they all mirror a generalized
devaluation of out-groups” (p. 364) and go on to find negative attitudes towards various
outgroups (e.g. xenophobia, islamophobia, anti-Semitism) to be substantially correlated in
large representative samples of German participants (see also Aosved, Long, & Voller,
It might then be that Germans’ attitudes towards Jews and Turks as two different
outgroups are similar, and as group-based emotions about the Holocaust are associated with
attitudes towards the former, they are equally correlated with the latter.
The application of Zick and colleagues’ ideas to the present findings imply that
feelings of shame should influence attitudes towards all outgroups equally. We do not believe
this to be the case. Rather, we suggest that shame feelings will most strongly influence
attitudes towards outgroups that, although not directly connected with the original victim
group, are perceived as similar to the harmed group. Furthermore, we propose that any
See also the literature on displaced aggression (Dollard et al., 1939; Marcus-Newhall, Pedersen, Carlson, &
Miller, 2000) for examples of how reactions elicited by one stimulus can generalise to another.
Running head: NEVER AGAIN
positive effects of emotions about ingroup wrongdoing on attitudes towards an unrelated
outgroup (i.e., the effects of moral shame) will be explained, at least in part, by the perception
that the ingroup has a particular moral obligation, an outstanding moral debt to repay as a
result of the ingroup’s original transgression. Thus, we are proposing that the association
between group-based emotions and attitudes towards an unrelated outgroup is driven
(mediated) by feelings of moral obligation, and is stronger (moderated) when the outgroup is
seen as similar to the originally harmed group. We sought to investigate these ideas in Study
Study 2
Study 2 was conducted in the context of a contemporary intergroup conflict, the war in
Iraq, in which reliable press reports and witness testimony agree that members of the ingroup
(British soldiers in this case) had consistently and systematically abused Iraqi prisoners under
their charge (Amnesty International, 2007; International Committee of the Red Cross, 2004).
Such crimes have given rise to considerable public outcry in Britain (e.g., BBC, 2004) and
thus provided a convenient and topical focus for our research. This research context provides
a strong test of the ideas developed in Study 1, as it represents a quite different situation. Not
only was the war still ongoing at the time of the study, but the issues surrounding the abuse of
Iraqi prisoners were yet to be resolved.
The aim of Study 2 was to extend the findings of Study 1, by demonstrating two
important underlying processes, mediation through feelings of moral obligation and
moderation via perceived similarity. An additional and important aspect of Study 2 related to
the refinement of the moral and image shame measures. The items used in Study 1
represented an early conceptualisation of these emotions and as such, we refined these items
in Study 2. In particular, we added two items, referring to both the damage done to the victim
group and to the immorality of the ingroup’s actions to the scale measuring moral shame, to
improve on the single item employed in Study 1. Similar adjustments were made to the image
Running head: NEVER AGAIN
shame scale to bring it more in line with our theoretical account, including direct reference to
damage to the ingroup’s reputation.
Hypotheses. (1) British people’s moral shame with regard to the Iraq war is predicted
to be negatively linked to social distance
from Pakistanis living in Britain. Image shame is
predicted to function inversely. (2) The effects of moral shame will be mediated by a sense of
moral obligation for human rights violations in Iraq. (3) This mediation will be moderated by
the perceived similarity between original outgroup (Iraqis) and target outgroup (Pakistanis),
such that the emotions will be more strongly related to attitudes towards Pakistanis when
Pakistanis are seen as similar to Iraqis.
Participants and Procedure
A community sample was recruited via an online survey. The survey was run through
a rewards-based online shopping network, in which users receive rewards points for
completing surveys, entering competitions and purchasing items through the system’s
website. Of those recruited, 301 participants (36% female) who self-identified as “British”
were included in the final analysis. The ages in the sample ranged from 19 to 81 (M = 47.9,
SD = 11.4). The study conformed fully to APA/BPS ethical guidelines.
The measures used in this study were similar to those used in Study 1, although the
moral shame and image shame scales were expanded, in line with Allpress and Brown (2011)
and refined in order to more accurately capture moral shame and image shame. The items
were as follows:
Guilt (“I feel guilty for the manner in which Iraqi people have been treated by
British.”, “Even if I have done nothing bad, I feel guilty for the behaviour of British towards
Iraqis.”, “I feel guilty for the bad living conditions of the Iraqi people.”; α = .92).
We focus here only on social distance from foreigners because the present study was part of a larger research
project in which support for Pakistani immigrants was not measured.
Running head: NEVER AGAIN
Moral shame (“Our treatment of Iraqi people makes me feel somewhat ashamed
about what it means to be British.”, “I feel ashamed to be British for the way we have treated
the Iraqi people.”, “I feel ashamed for the damage done to Iraqi people by Brits.”, “I feel
ashamed because Britain's actions with regard to Iraq have been immoral.”; α = .95).
Image shame (“I feel disgraced because the behaviour of Brits towards Iraqi people
has created a bad image of Britain in the eyes of the world.”, “I feel ashamed when I realise
that other countries might think of Britain negatively because of our involvement in Iraq.”,,
“To think how Britain is seen for its treatment of Iraqi people makes me feel ashamed.”, “I
feel humiliated when I think of how Britain is seen negatively by the rest of the world for how
it has treated the Iraqi people.”, “I feel ashamed because Britain has a damaged reputation.”; α
= .97).
Social distance was assessed with a scale parallel to that used in Study 1 (“I would
have problems moving into a district where many Pakistanis live.”, “Too many Pakistanis live
in Great Britain.”, “Pakistanis should be sent back to their home country if jobs become
scarce.”; α = .93).
Moral obligation was measured with three items (“Britain has an ethical obligation to
make up for its actions in Iraq.”, “I am concerned that British troops acted immorally in
Iraq.”, “Britain has absolutely no moral obligation to the Iraqi people (reversed).”; α = .77)
and perceived similarity with one item (“How similar are Pakistanis and Iraqis?”).
All items were assessed using a seven-point scale ranging from “strongly disagree” to
“strongly agree”. An overview of the measures and their inter-correlations is given in Table 3.
In preliminary analyses, the hypothesised factorial structure of the three group-based
emotions was, again, supported in CFAs (see Table 2). Alternative models fitted the data
significantly worse. Thus, closely replicating Study 1, these findings obtained in a different
Running head: NEVER AGAIN
intergroup context confirm that it is possible to differentiate moral shame, image shame and
Another set of CFAs was run to confirm that the hypothesised mediator, moral
obligation, was distinct from the independent and dependent variables. A five factor model
including the three group-based emotions, social distance from Pakistanis, and moral
obligation as separate factors, fit the data significantly better than any model in which the
latter had been combined into one factor with: moral shame (model 4a), image shame (model
4b), guilt (model 4c), or social distance (model 4d), all χ
s, p < .001, all AICs > 100.
Results of these analyses thus clearly supported the separability of the measures. Fit statistics
for the models can be found at the bottom of Table 2.
Generalisation of group-based emotions
A model predicting social distance from Pakistanis from feelings of guilt, moral shame
and image shame regarding British misdeeds in Iraq provided a good fit to the data (χ
(97) =
243.35, p < .01, CFI = .977, RMSEA = .071, SRMR = .024). Moral shame (β = -.52, p < .01)
was significantly and negatively, whereas image shame (β = .38, p < .05) was positively,
linked with social distance. The link of guilt (β = -.22, p = .12) with the dependent variable
was again not significant. These findings replicate those of Study 1 by showing that Britons’
emotional reactions to ingroup wrongdoing in Iraq are related to attitudes towards a seemingly
unrelated minority group, Pakistanis living in Britain.
An additional model was created in which moral obligation mediated the effects of the
three emotions. This model also fitted the data well, χ
(141) = 345.81, p < .01, CFI = .970,
RMSEA = .070, SRMR = .041. Bootstrapping analyses using 5,000 bootstrap re-samples and
bias-corrected 95% confidence intervals (CIs) showed that indirect effects from moral shame
(standardised indirect effect -.41; CI: -.77 to -.20) but not from guilt (-.11; CI: -.29 to .02) or
image shame (.08; CI: -.07 to .27) were significant at the α = .05 level (see Preacher, Rucker
& Hayes, 2007). This second analysis can be seen in Figure 2. We also tested the mediation
Running head: NEVER AGAIN
model without the direct paths from emotions to social distance. This nested model explained
the data equally well as the full model (χ
(3) = 5.29, p > .15). These analyses therefore
show that the effects of moral shame on social distance are mediated by a sense of moral
obligation for the ingroup’s wrongdoing.
Testing the moderated mediation hypothesis
As the reader will recall, we did not expect feelings of moral shame for ingroup
wrongdoing, and the accompanying sense of moral obligation, to influence attitudes towards
all outgroups equally. Rather, we predicted these feelings would specifically influence
attitudes towards outgroups that are perceived as similar to the original victim group in core
aspects. To test the hypothesised role of perceived similarity, we designed a moderated
mediation model according to Preacher and colleagues’ (2007) Model 3. We assumed that
while group-based emotions with regard to the original incident, especially moral shame,
should always be associated with moral obligation, the latter’s effect on social distance from
an unrelated outgroup would depend on perceived similarity of the two outgroups. Following
Kline and Dunn (2000) and to avoid multicollinearity and linear dependency of the constructs
in the model, the interaction term of moral obligation and similarity needed for this test was
computed as a the manifest product of the two centred variables (see also Aiken & West,
1991). The model fitted the data well (χ
(169) = 375.01, p < .01, CFI = .970, RMSEA =
.064, SRMR = .039). Both the interaction term and perceived similarity itself emerged as
meaningful predictors of social distance from Pakistanis (β = -.11 and β = .14, respectively,
both ps .05), thus qualifying for the theorised moderated mediation. The only direct path
remaining significant (i.e., not fully mediated) in the final model was that from image shame
to social distance (β = .31, p = .03) paralleling the mediation analysis reported above. Further
analyses confirmed that, as expected, the links from moral obligation to intergroup attitudes
were especially pronounced for individuals perceiving Pakistanis to be similar to the original
outgroup Iraqis: when the sample was split into a high and low similarity group (M ± 1 SD),
Running head: NEVER AGAIN
the path from moral obligation to social distance remained significant for the high similarity
group only (β
High Similarity
= -.43, p < .05) and was considerably higher than for the low
similarity group (β
Low Similarity
= -.28, p = .06).
Study 2 replicates and extends the findings from Study 1 and thus further supports the
idea that effects of group-based emotions in response to specific ingroup wrongdoing can be
linked to attitudes towards a seemingly unrelated outgroup, which is the target of
contemporary discrimination. The emotions of guilt, moral shame and image shame
concerning the British involvement in Iraq were, again, found to be clearly separable
constructs and showed the predicted links to attitudes towards Pakistanis living in the UK.
Furthermore, our analyses showed that the effects of moral shame were mediated by a sense
of moral obligation. These mediation effects were more pronounced if individuals judged the
two groups to be more similar.
There is an additional issue raised by Study 2 that deserves further discussion, namely
the high correlations among predictor variables, a possible indication of multicollinearity.
There are three points to note about this issue. Firstly, the findings of Study 2 were strikingly
consistent, not only with those of Study 1, but with the studies presented by Allpress and
Brown (2011), where the inter-correlations among emotion variables were lower (and
consistently fell between .50 and .70). Secondly, the relatively large sample size of the present
study offers some protection against the disruptive effects of multicollinearity (O’Brien,
2007). Thirdly, regression analyses (not reported here) show that the collinearity statistics
were all within normal ranges, all variance inflation factors < 5.30 and tolerances > .18. As
such, the evidence suggests that multicollinearity did not unduly influence the results of Study
General Discussion and Implications
Running head: NEVER AGAIN
The present paper supports and extends previous research on guilt and shame in a
number of ways. As expected, group-based shame was found to consist of the two facets of
moral shame and image shame and to be clearly separable from guilt. Both forms of shame
were then shown to be predictive of attitudes towards a seemingly unrelated outgroup – that
is, they generalised in a systematic way.
Structure of guilt and shame and their correlates
In line with Allpress and Brown’s (2011) claim of a distinction between guilt, moral
shame and image shame, data in both German and British contexts were found to be more
compatible with a three-factor than with a two- or one-factor structure. Not only has this
finding important theoretical implications in offering insights into the nature of group-based
emotions, it may also prove to be practically meaningful in that reactions motivated by
genuine feelings of moral shame could, in fact, be supportive of wider intergroup relations.
Behaviour motivated by image shame, on the other hand, may also trigger actively anti-social
responses such as wishing to maintain social distance from other, socially devalued
outgroups, possibly in reaction to social identity threat (Allpress et al., 2010; Branscombe et
al., 2007).
The present data, together with others (Allpress & Brown, 2011; Allpress et al., 2010),
challenge some thinking in the literature on group-based emotions by suggesting that, rather
than guilt being the important motivator of intergroup behaviour, the two forms of shame
investigated here are primarily driving reactions to reminders of past and current injustices:
just as group-membership can make us “guilty by association” (Doosje et al., 1998), so too
can it make us feel ashamed by association, and the latter feelings may be more potent.
Guilt showed consistently negligible relationships with the dependent measures in
both studies. Others, too, have found similarly weak or unstable associations between guilt
and various outcome measures (Allpress & Brown, 2011; Allpress et al., 2010; Gausel et al.,
2010; Harth et al., 2008; Iyer et al., 2007; Leach et al., 2006). Such findings are consistent
Running head: NEVER AGAIN
with the characterisation of guilt put forward by both Teroni and Deonna (2008) and Deonna
and colleagues (2011). These authors conceptualise guilt as motivating limited reparative acts
directly to the victim group, with the intention solely of reversing the harm done, but with
little concern for the greater morality of the ingroup or for the more general well-being of the
victim group. For this reason, we would not expect guilt to influence attitudes towards any
group other than the victim group. This is unlike moral shame, which motivates a degree of
soul-searching that is likely to influence wider attitudes and behaviours.
Our data suggest that image shame differs from other group-based emotions in core
aspects. Not only did it have negative effects on intergroup attitudes, Study 2 highlighted the
fact that its operating mode is different from that of moral shame. While the effects of moral
shame were fully mediated by the experience of moral obligation, image shame retained a
direct positive link with social distance. Together with recent theorising (Allpress & Brown,
2011; Deonna et al., 2011; Gausel & Leach, 2011; Teroni & Deonna, 2008), the present
results suggest that this image component might be the reason why research into group-based
emotions has misconceptualised shame in general as being solely linked to anti-social
reactions. Further research is needed to investigate the mediating and moderating processes
involved in the effects of image shame on attitudes towards outgroups.
Implications for wider intergroup relations
In this paper we investigated the novel notion that a link exists between group-based
emotions concerning one intergroup episode and attitudes towards another unrelated
outgroup. These findings show that emotional reactions for ingroup wrongdoing may have
important effects on wider intergroup relations. If perpetrator group members feel moral
shame for their group’s misdeeds then this is likely to have positive effects on the wider
treatment of minority groups. This finding provides support for the notion that collective
remembrance of large-scale historical atrocities such as the Holocaust may prevent future
transgressions from happening.
Running head: NEVER AGAIN
However, a risk is that shaming a group for their actions (and thus inducing image
shame in the perpetrator group) can have wider and more extensive negative effects than
originally thought. Allpress and Brown (2011), for instance, provide evidence that image
shame is associated with a host of negative outcomes connected with the initial shame-
eliciting situation and raise concerns that, for this reason, shaming can sometimes be an
unproductive strategy by which to encourage groups to address their misdeeds. Our present
findings provide additional cause for concern by showing that shaming may have further
implications for how the perpetrator group treats other minority groups within the wider
Future Research
We believe the present paper presents a fertile base for future research. By
demonstrating that the effects of shame with regard to harm done to one group are not
confined to attitudes towards that originally victimised group, we hope to broaden the scope
of research into group-based emotions. Although our analyses provide initial insights into the
conditions under which shame may influence attitudes towards unrelated outgroups, future
work should continue to test and document the mediating and moderating mechanisms,
particularly in relation to image shame. We have offered the possibility earlier that a desire to
maintain or bolster the ingroup’s status may account for some of the negative effects of image
shame. Further work may find it fruitful to experimentally manipulate important variables,
such as perceived similarity or moral obligation (Aquino & Reed, 2002; Leach, Ellemers, &
Barreto, 2007). Our laboratories are currently conducting studies investigating the
longitudinal effects of these emotions, the role of moral obligation, and the impact of
perceived victim status of the target outgroup, thus addressing one obvious shortcoming of the
present correlational findings – that they only allow for somewhat limited inferences about
Running head: NEVER AGAIN
The present research has implications for the role of reminders of historical or
contemporary injustice in debates about seemingly unrelated issues. Although such reminders
sometimes seem so startlingly out of place, our research suggests that their use by members of
both the left and right ends of the political spectrum may have meaningful effects on members
of the general population. Given the generalising effects of collective emotions, appealing to
the immorality of the group’s misdeeds may elicit moral shame and lead to an increase in
positive attitudes towards members of an unrelated outgroup. Appeals to Germans’ “historical
responsibility” – for example, as made by then German Federal President Richard von
Weizsäcker in his commemoration speech in 1985, and recently by Federal Minister for
Foreign Affairs Guido Westerwelle in a speech given on July 1, 2010 – fit into this line of
reasoning. Highlighting continuing external criticism of the ingroup, however, may trigger
feelings of image shame, which may, in turn, lead to an increase in distancing from other
outgroups. References to the shame that is “imposed on us by others” may be an example of
public figures using the evocation of feelings of image shame to advance (either consciously
or unconsciously) their own political ends in a seemingly unrelated domain such as
immigration policies. Moral obligation and its effects – positive and negative – are thus not
confined to only one outgroup.
Running head: NEVER AGAIN
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting
interactions. Thousand Oaks, CA: Sage.
Allpress, J. A., & Brown, R. (2011). Feeling ashamed versus being shamed: Two forms of
shame differentially predict positive and negative responses to ingroup wrongdoing.
Unpublished manuscript.
Allpress, J. A., Barlow, F. K., Brown, R., & Louis, W. R. (2010). Atoning for colonial
injustices: Group-based shame and guilt motivate support for reparation. International
Journal of Conflict and Violence, 4(1), 75-88.
Amnesty International Report: Iraq. (2007). Retrieved from
Aosved, A. C., Long, P. J., & Voller, E. K. (2009). Measuring sexism, racism, sexual
prejudice, ageism, classism, and religious intolerance: The intolerant schema measure.
Journal of Applied Social Psychology, 39(10), 2321-2354.
Aquino, K., & Reed, A. (2002). The self-importance of moral identity. Journal of Personality
and Social Psychology, 83(6), 1423-1440.
Ausubel, D. P. (1955). Relationships between shame and guilt in the socializing process.
Psychological Review, 62(5), 378-390.
Baumeister, R. F., Stillwell, A. M., & Heatherton, T. F. (1994). Guilt: An interpersonal
approach. Psychological Bulletin, 115(2), 243-267.
BBC Radio 4. (2004). ‘Shocking’ racism in jobs market. Monday 12 July, 13:46 GMT.
Available at
British Broadcasting Corporation. (2004). ‘Blair criticised over Iraq claims’. Retrieved from
Branscombe, N. R., & Doosje, B. (2004a). Collective guilt: International perspectives. New
York: Cambridge University Press.
Running head: NEVER AGAIN
Branscombe, N. R., & Doosje, B. (2004b). International perspectives on the experience of
collective guilt. In N. Branscombe & B. Doosje (Eds.), Collective guilt: International
perspectives (pp. 3-15). New York: Cambridge University Press.
Branscombe, N. R., Schmitt, M. T., & Schiffhauer, K. (2007). Racial attitudes in response to
thoughts of white privilege. European Journal of Social Psychology, 37, 203-215.
Branscombe, N. R., Slugoski, B., & Kappen, D. M. (2004). The measurement of collective
guilt: What it is and what it is not. In N. Branscombe & B. Doosje (Eds.), Collective
guilt: International perspectives (pp. 16-34). New York: Cambridge University Press.
Brewer, M.B. (1999). The psychology of prejudice: Ingroup love or outgroup hate? Journal of
Social Issues, 55, 429-444.
Brown, R. (2009). From both sides now: Perpetrator and victim responses to intergroup
transgressions. In S. Otten, K. Kassenberg, & T. Kessler (Eds.) Intergroup relations:
The role of motivation and emotion (pp. 182-198). Hove: Psychology Press.
Brown, R. (2010). Prejudice: Its social psychology (2nd ed.). NY: Wiley-Blackwell.
Brown, R., & Čehajić, S. (2008). Dealing with the past and facing the future: Mediators of the
effects of collective guilt and shame in Bosnia and Herzegovina. European Journal of
Social Psychology, 38(4), 669–684.
Brown, R., González, R., Zagefka, H., Manzi, J., & Čehajić, S. (2008). Nuestra culpa:
Collective guilt and shame as predictors of reparation for historical wrongdoing.
Journal of Personality and Social Psychology, 94, 75-90.
Burnham, K. P., and Anderson, D.R. (2002). Model selection and multimodel inference: A
practical information-theoretic approach, 2nd ed. New York: Springer-Verlag.
Buruma, I. (2002). The wages of guilt. Memories of war in Germany and Japan. London:
Phoenix Paperback.
Running head: NEVER AGAIN
Cohen, R. (1998). Anniversary sets Germans to quarreling on Holocaust. New York Times.
Retrieved from
de Hooge, I. E., Breugelmans, S. M., & Zeelenberg, M. (2008). Not so ugly after all: When
shame acts as a commitment device. Journal of Personality and Social Psychology,
95(4), 933-943.
de Hooge, I. E., Nelissen, R., Breugelmans, S. M., & Zeelenberg, M. (2011). What is moral
about guilt? Acting “prosocially” at the disadvantage of others. Journal of Personality
and Social Psychology, 100(3), 462-473.
Deonna, J. A., Rodogno, R., & Teroni, F. (2011). In defense of shame: The faces of an
emotion. New York City: Oxford University Press, USA.
Department for Work and Pensions. (2009). Research report 607: A test of racial
discrimination in recruitment practice. Available at
Dollard, J., Doob, L. W., Miller, N. E., Mowrer, O. H., & Sears, R. R. (1939). Frustration and
aggression. New Haven, CT: Yale University Press.
Doosje, B., Branscombe, N. R., Spears, R., & Manstead, A. S. R. (1998). Guilty by
association: When one's group has a negative history. Journal of Personality and
Social Psychology, 75(4), 872-886.
Dresler-Hawke, E., & Liu, J. H. (2006). Collective shame and the positioning of German
national identity. Psicología Política, 32, 131-153.
Fedewa, B., Burns, L., & Gomez, A. (2005). Positive and negative perfectionism and the
shame/guilt distinction: Adaptive and maladaptive characteristics. Personality and
Individual Differences, 38(7), 1609-1619.
Fontaine, J. R. J., Luyten, P., De Boeck, P., Corveleyn, J., Fernandez, M., Herrera, D., Ittze,
S, A., et al. (2006). Untying the Gordian knot of guilt and shame: The structure of
Running head: NEVER AGAIN
guilt and shame reactions based on situation and person variation in Belgium,
Hungary, and Peru. Journal of Cross-Cultural Psychology, 37(3), 273-292.
Gardham, K. & Brown, R. (2001). Two forms of intergroup discrimination with positive and
negative outcomes: Explaining the positive negative asymmetry effect. British Journal
of Social Psychology, 40(1), 23-34.
Gausel, N., & Leach, C. W. (2011). Concern for self-image and social-image in the
management of moral failure: Rethinking shame. European Journal of Social
Psychology, 41(4), 468-478.
Gausel, N., Leach, C. W., Vignoles, V., & Brown, R. (2010). Is shame really that bad?
Uncovering the pro-social potential of shame about in-group wrongdoing.
Unpublished manuscript.
Gehm, T. L., & Scherer, K. R. (1988). Relating situation evaluation to emotion
differentiation: Nonmetric analysis of cross-cultural questionnaire data. In K. R.
Scherer (Ed.), Facets of emotion: Recent research (pp. 61-78). Hillsdale, NJ: Erlbaum.
German Federal Statistical Office (2009). Statistical Yearbook 2009 for the Federal Republic
of Germany. Wiesbaden: Federal Statistical Office. Available online at
Harth, N. S., Kessler, T., & Leach, C. W. (2008). Advantaged group's emotional reactions to
intergroup inequality: The dynamics of pride, guilt, and sympathy. Personality and
Social Psychology Bulletin, 34(1), 115-129.
Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological
review, 94(3), 319-340.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure
analysis: Conventional criteria versus new alternatives. Structural Equation Modeling,
6(1), 1-55.
Running head: NEVER AGAIN
Human Rights Watch (1995). Germany for Germans. Xenophobia and racist violence in
Germany. New York: Human Rights Watch.
Imhoff, R. & Banse, R. (2009). Ongoing victim suffering increases prejudice: The case of
secondary antisemitism. Psychological Science, 20, 1443-1447.
International Committee of the Red Cross. (2004). Retrieved from
Iyer, A., & Leach, C. W. (2008). Emotion in inter-group relations. European Review of Social
Psychology, 19(1), 86-125.
Iyer, A., Leach, C. W., & Crosby, F. (2003). White guilt and racial compensation: The
benefits and limits of self-focus. Personality and Social Psychology Bulletin, 29(1),
Iyer, A., Schmader, T., & Lickel, B. (2007). Why individuals protest the perceived
transgressions of their country: The role of anger, shame and guilt. Personality and
Social Psychology Bulletin, 33(4), 572-587.
Johns, M., Schmader, T., & Lickel, B. (2005). Ashamed to be an American? The role of
identification in predicting vicarious shame for anti-Arab prejudice after 9–11. Self
and Identity, 4(4), 331-348.
Kline, T. J. B., & Dunn, B. (2000). Analysis of interaction terms in structural equation
models: A non-technical demonstration using the deviation score approach. Canadian
Journal of Behavioural Science, 32(2), 127-132.
Knowlton, J., & Cates, T. (1993). Forever in the shadow of Hitler? Original documents of the
Historikerstreit, the controversy concerning the singularity of the Holocaust. New
Jersey: Humanities Press International.
Kulish, N. (2010). German identity, long dormant, reasserts itself. New York Times. Retrieved
Running head: NEVER AGAIN
Leach, C. W., Iyer, A., & Pederson, A. (2006). Anger and guilt about ingroup advantage
explain the willingness for political action. Personality and Social Psychology
Bulletin, 32(9), 1232-1245.
Lewis, H. B. (1971). Shame and guilt in neurosis. New York: International Universities Press.
Lickel, B., Schmader, T., & Barquissau, M. (2004). The evocation of moral emotions in
intergroup contexts: The distinction between collective guilt and collective shame. In
N. Branscombe & B. Doosje (Eds.), Collective guilt: International perspectives (pp.
35-55). New York: Cambridge University Press.
Lickel, B., Schmader, T., Curtis, M., Scarnier, M., & Ames, D. R. (2005). Vicarious shame
and guilt. Group Processes and Intergroup Relations, 8(2), 145-57.
Luby, J., Belden, A., Sullivan, J., Hayen, R., McCadney, A., & Spitznagel, E. (2009). Shame
and guilt in preschool depression: Evidence for elevations in self-conscious emotions
in depression as early as age 3. Journal of Child Psychology and Psychiatry, 50(9),
Lynd, H. M. (1956). On shame and the search for identity. New York: Harcourt Brace
MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power analysis and
determination of sample size for covariance structure modeling. Psychological
Methods, 1(2), 130-149.
Marcus-Newhall, A., Pedersen, W. C., Carlson, M., & Miller, N. (2000). Displaced
aggression is alive and well: A meta-analytic review. Journal of Personality and
Social Psychology, 78(4), 670-689.
McGarty, C., Pedersen, A., Leach, C. W., Mansell, T., Waller, J., & Bliuc, A. M. (2005).
Group-based guilt as a predictor of commitment to apology. The British Journal of
Social Psychology, 44(4), 659-680.
Running head: NEVER AGAIN
Meehan, W., O'Connor, L. E., Berry, J. W., Weiss, J., Morrison, A., & Acampora, A. (1996).
Guilt, shame, and depression in clients in recovery from addiction. Journal of
Psychoactive Drugs, 28(2), 125-134.
Miller, R. S. & Tangney, J. P. (1994). Differentiating embarrassment and shame. Journal of
Social and Clinical Psychology, 13, 273-287.
Mummendey, A., & Otten, S. (1998). Positive-negative asymmetry in social discrimination.
In W. Stroebe, & M. Hewstone (Eds.), European Review of Social Psychology, 9, (pp.
107-143) Chichester: Wiley.
Niedenthal, P.M., Tangney, J., & Gavanski, I. (1994). "If only I weren't" versus "If only I
hadn't": Discriminating shame and guilt in counterfactual thinking. Journal of
Personality and Social Psychology, 67, 585-595.
O'Brien, R. M. (2007). A caution regarding rules of thumb for variance inflation factors,
Quality and Quantity, 41, 673-690.
Peetz, J., Gunn, G. R., & Wilson, A. E. (2010). Crimes of the past: Defensive temporal
distancing in the face of past in-group wrongdoing. Personality and Social Psychology
Bulletin, 36(5), 598-611.
Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Addressing moderated mediation
hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research,
42(1), 185-227.
Reemtsma, J. P. (2005). Schuld und Verantwortung [Guilt and responsibility]. taz-Journal
2005/1, 86-97.
Reifen Tagar, M., Federico, C. M., & Halperin, E. (2011). The positive effect of negative
emotions in protracted conflict: The case of anger. Journal of Experimental Social
Psychology, 47, 157-164.
Rensmann, L. (1999). Holocaust memory and mass media in contemporary Germany:
Reflections on the Goldhagen debate. Patterns of Prejudice, 33(1), 59-76.
Running head: NEVER AGAIN
Rensmann, L. (2004). Collective guilt, national identity, and political processes in
contemporary Germany. In N. Branscombe & B. Doosje (Eds.), Collective guilt:
International perspectives (pp. 169-192). New York: Cambridge University Press.
Rensmann, L. (2005). Demokratie und Judenbild. Antisemitismus in der politischen Kultur
der Bundesrepublik Deutschland [Democracy and the concept of the Jew. Anti-
Semitism in the Federal Republic of Germany’s political culture]. Wiesbaden: Verlag
für Sozialwissenschaften.
Roseman, I. J., Wiest, C., & Swartz, T. S. (1994). Phenomenology, behaviors, and goals
differentiate discrete emotions. Journal of Personality and Social Psychology, 67(2),
Smith, R. H., Webster, J. M., Parrott, W. G., & Eyre, H. L. (2002). The role of public
exposure in moral and nonmoral shame and guilt. Journal of Personality and Social
Psychology, 83(1), 138-159.
Swim, J. K., & Miller, D. L. (1999). White guilt: Its antecedents and consequences for
attitudes toward affirmative action. Personality and Social Psychology Bulletin, 25(4),
Tangney, J. P. (1991). Moral affect: The good, the bad, and the ugly. Journal of Personality
and Social Psychology, 61(4), 598-607.
Tangney, J. P., & Dearing, R. L. (2002). Shame and guilt. The Guilford Press.
Tangney, J. P., Miller, R. S., Flicker, L., & Barlow, D. H. (1996). Are shame, guilt, and
embarrassment distinct emotions? Journal of Personality and Social Psychology,
70(6), 1256-1269.
Tangney, J. P., Wagner, P., Fletcher, C., & Gramzow, R. (1992). Shamed into anger? The
relation of shame and guilt to anger and self-reported aggression. Journal of
Personality and Social Psychology, 62, 669-675.
Running head: NEVER AGAIN
Teroni, F., & Deonna, J. A. (2008). Differentiating shame from guilt. Consciousness and
Cognition, 17, 725-740.
Tracy, J. L., & Robins, R. W. (2006). Appraisal antecedents of shame and guilt: Support for a
theoretical model. Personality and Social Psychology Bulletin, 32, 1339-1351.
Wagner, U., Christ, O., & Pettigrew, T. F. (2008). Prejudice and group-related behavior in
Germany. Journal of Social Issues, 64(2), 403-416.
Wohl, M. J. A., Branscombe, N. R., & Klar, Y. (2006). Collective guilt: Emotional reactions
when one's group has done wrong or been wronged. European Review of Social
Psychology, 17(1), 1-37.
Zick, A., Wolf, C., Küpper, B., Davidov, E., Schmidt, P., & Heitmeyer, W. (2008). The
syndrome of group-focused enmity: The interrelation of prejudices tested with
multiple cross-sectional and panel data. Journal of Social Issues, 64(2), 363-383.
Zimmermann, A., Abrams, D., Doosje, B., & Manstead, A. (in press). Causal and moral
responsibility: Antecedents and consequences of group-based guilt. European Journal
of Social Psychology.
Running head: NEVER AGAIN
Tables and Figures
Table I
Overview of Study 1 main scales and their inter-correlations
Measure M (SD) (2) (3) (4) (5) (6) (7)
(1) Guilt 3.37 (0.97)
-.09 .28**
.10 -.06
(2) Moral Shame 3.59 (1.80)
-.12 .34**
(3) Image Shame 2.84 (1.42)
.11 -.04 -.01
(4) Social Distance 2.52 (1.04)
.01 -.26**
(5) Foreigner Support 3.65 (0.98)
.02 .07
(6) Age 27.5 (9.7)
(7) Highest Formal Degree
** p .01 * p .05
p .10.
Running head: NEVER AGAIN
Table II
Fit indices of models tested in confirmatory factor analyses in Studies 1 and 2
Study 1
Emotions about Holocaust
1) Three factor model 61.08 (31) .976 .069 109.08
1a) Two factor model 1
(moral shame combined with
image shame) 115.87 (32) .932 .114 161.87
1b) Two factor model 2
(moral shame combined with
guilt) 72.75 (32) .967 .079 118.75
1c) One factor model 322.13 (33) .766 .208 366.13
Attitudes towards Turks
2) Two factor model 31.97 (26) .992 .034 69.97
2a) One factor model 88.40 (27) .919 .106 124.40
Study 2
Emotions about Iraq war
3) Three factor model 167.17 (61) .981 .076 227.17
3a) Two factor model 1
(moral shame combined with
image shame) 365.01 (63) .945 .126 421.01
3b) Two factor model 2
(moral shame combined with
guilt) 342.05 (63) .950 .122 398.05
3c) One factor model 591.44 (64) .905 .166 645.44
Separate Mediator
4) Five factor model 345.81 (141) .970 .070 443.81
4a) Four factor model 1
(moral obligation combined
with moral shame) 472.68 (145) .952 .087 562.68
4b) Four factor model 2
(moral obligation combined
with image shame) 561.96 (145) .939 .098 651.96
4c) Four factor model 3
(moral obligation combined
with guilt) 525.11 (145) .945 .093 615.11
4d) Four factor model 4
(moral obligation combined
with social distance) 652.57 (145) .926 .108 742.57
Note. CFI = Comparative Fit Index, RMSEA = Root Mean Square Error of
Approximation, AIC = Akaike Information Criterion.
Running head: NEVER AGAIN
Table III
Overview of Study 2 main scales and their inter-correlations
Measure M (SD) (2) (3) (4) (5) (6)
(1) Guilt 3.61 (1.61)
.87 .86 -.35 .70 -.17
(2) Moral Shame
4.16 (1.61)
.90 -.37 .77 -.18
(3) Image Shame 4.03 (1.60)
-.28 .68 -.15
(4) Social Distance 3.84 (1.81)
-.52 .27
(5) Moral Obligation 4.69 (1.27)
(6) Similarity 3.56 (1.53)
Note. All correlations significant at p .01.
Running head: NEVER AGAIN
Figure captions
Figure 1. Structural equation model of generalisation of guilt and shame in Study 1.
Note. Controlled for age and highest formal degree. Indicators of latent constructs and their
error terms are not displayed for the sake of clarity.
** p .01 * p .01
p .07.
Figure 2. Mediated structural equation model of generalisation of guilt and shame in Study 2.
Note. Manifest indicators and their error terms are not displayed for the sake of clarity. Direct
(un-mediated) effects in parantheses.
** p .01 * p .05
p = .06.
Running head: NEVER AGAIN
Figure 1
Moral Shame
Image Shame
Social Distance
Foreigner Support
.65 -.77
= .17
= .13
Running head: NEVER AGAIN
Figure 2
Moral Shame
Image Shame Social Distance
= .29
Moral Obligation
= .60
.36* (.44*)
-.14 (-.66**)
-.13 (-.25)
... The effects of exposure to selective historical narratives are not restricted to the marginalized group whose histories are silenced or sanitized. For instance, moral shame with regard to the Holocaust led Germans to endorse more supportive attitudes toward an unrelated minority group: Turks living in Germany (Rees et al., 2013). Similarly, British participants who were reminded of in-group transgressions against Iraq then experienced moral shame and expressed more favorable attitudes toward Pakistanis, an unrelated minority group. ...
... Similarly, British participants who were reminded of in-group transgressions against Iraq then experienced moral shame and expressed more favorable attitudes toward Pakistanis, an unrelated minority group. Thus, reminders of historical injustices can promote positive relations with a wider range of marginalized groups and impact identity-relevant perception and action (Rees et al., 2013), including perception of present-day injustice and support or opposition to policies to address historical grievances (Sibley et al., 2008). ...
... Over time, such policies can make ethnic minoritygroup members feel respected and welcomed in the local and national space and subsequently increase their engagement with ethnic majority culture (Banks, 2017;Huo & Molina, 2006). Such policies can also benefit ethnic majority group members and promote positive relations with a wide range of marginalized groups (Rees et al., 2013). ...
The United States is a nation of immigrants with significant ethnic and racial diversity. Yet, American identity is associated with European-Americans and their cultural values, defining ethnic minorities as less American. Experiences of identity denial are associated with negative mental and physical health outcomes, as well as lower political and civic engagement. Perceptions of prototypical American-ness link to a wide range of social policy about language, affirmative action, and redistribution. A cultural psychological perspective analyzes the contexts that promote exclusive conceptions of American identity, and it focuses on individual people who make up these contexts. Policies that recognize minority-group cultures and acknowledge the historical injustices against them can promote inclusive conceptions of American identity.
... The primary lesson espoused by perpetrator groups who have accepted responsibility for the genocide seems to be "never again. " This lesson implies that the injustice committed in the past was illegitimate and has no place in present or future interactions and is adopted through educational programs, commemoration events, and speeches by political leaders (see discussion in Rees, Allpress, & Brown, 2013). Similar to the universal extension of the lesson among victim groups to other groups, the "never again" lesson may also affect attitudes among members of the perpetrator group toward other minority groups and not just the former victim group. ...
... Similar to the universal extension of the lesson among victim groups to other groups, the "never again" lesson may also affect attitudes among members of the perpetrator group toward other minority groups and not just the former victim group. For example, among a German community sample, moral shame about the in-group's actions during the Holocaust predicted support for an unrelated minority group, namely ethnic Turks in Germany (Rees et al., 2013). In other words, accepting responsibility for and feeling shame about the in-group's crimes appeared to translate into the universal lesson of "never again" and extend to caring for other minority groups as well, even if this lesson was only assessed indirectly in this study. ...
Full-text available
This chapter reviews research on how historical genocide continues to affect victim and perpetrator groups’ beliefs, emotions, and intergroup attitudes in the present. The authors organize their review around four central psychological processes that help in understanding why and how members of victim and perpetrator groups respond in such divergent ways: which psychological needs members of these groups have in light of the events (e.g., needs for meaning, agency, power, acknowledgment), how central the genocide is to their identity and how relevant it is seen to the present, how they perceive the scope of genocide and who is considered a victim, and the various lessons group members draw from the events. The authors also discuss possibilities for bridging these divergent responses and factors that complicate the picture such as when groups were both victims and perpetrators.
... Past acting and its consequences influence our relations with groups (such as minorities or other nations) in the present (Rees, Allpress, & Brown, 2013) and relate to current issues of international politics Spellman & Holyoak, 1992). refer to the influence of a group's representation of its history on current politics as charters. ...
Full-text available
This article explores the role negative history plays in political discourse on crisis management and how times of sociopolitical change in turn influence the strategies that can be employed to write a positive historical charter. Choosing Germany as a case study, we analyzed how political leaders negotiate Germany's narrative and political role during the European “refugee crisis” in speeches (n = 332) held between 2015 and 2018. Applying a combination of corpus‐based and qualitative narrative analysis, we found that the context of a crisis is used to attach new meaning to Germany's role in World War II. By focusing on the lessons learned from history and pointing out the parallels between the current crisis and sociopolitical developments that took place 80 years ago, Germany is presented as the ideal advocate for a free and united Europe, a narrative that legitimizes its advancingly dominant role within the EU and beyond. The analysis demonstrates how times of change can facilitate social creativity strategies for nations whose past is dominated by negative history, with implications for the theory of identity management.
... Hence, it is conceivable that these symbols of the Nazi era are still generating cognitive dissonance among those living in the surrounding areas. Research in psychology suggests that reminders of ingroup wrongdoing may prompt defensive reactions (Imhoff and Banse 2009;Rotella and Richeson 2013; but see Rees, Allpress, and Brown 2013). This mechanism would generate the same patterns uncovered here without requiring the transmission of attitudes across generations. ...
We explore the long-term political consequences of the Third Reich and show that current political intolerance, xenophobia, and voting for radical right-wing parties are associated with proximity to former Nazi concentration camps in Germany. This relationship is not explained by contemporary attitudes, the location of the camps, geographic sorting, the economic impact of the camps, or their current use. We argue that cognitive dissonance led those more directly exposed to Nazi institutions to conform with the belief system of the regime. These attitudes were then transmitted across generations. The evidence provided here contributes both to our understanding of the legacies of historical institutions and the sources of political intolerance.
... In exploring this issue Chapter 4 offered the first empirical comparison of the impact of reminders of ingroup wrongdoings compared with reminders of ingroup rightdoings regarding the majority's reconciliatory tendencies within the context of separatist conflict. The findings dicussed in Chapter 4 were also the first to verify that the majority can simultaneously claim the role of perpetrator and victim, whereas these constructs (i.e., sense of perpetratorhood and victimhood) were assumed to be mutually exclusive in the previous research (e.g., Doosje, Branscombe, Spears, & Manstead, 1998Peetz, Gunn, & Wilson, 2010;Rees, Allpress, & Brown, 2013;. Based on self-criticism literature (Brander & Hornsey, 2006;Hornsey, 2005;Hornsey & Imani, 2004) and related lines of research (e.g., Doosje et al., 2006;Zebel, Zimmermann, Viki, & Doosje, 2018), I predicted that the majority's reconciliatory stance could be more effectively enhanced by means of a reminder of ingroup wrongdoings rather than a reminder of ingroup rightdoings, given the more consequential role of the former in inducing a greater sense of perpetratorhood. ...
Full-text available
Separatist movements are a social phenomenon that spread in many countries across the world, threatening the stability of established political entities due to its violent and intractable nature. Deescalating conflict and establishing peace among the disputing parties is of paramount importance. Despite the pressing need, there is a notable dearth of social psychological research that has made reconciliation in such conflicts the focal point of its empirical investigation--and even less so pertaining to social-psychological interventions conducive to reconciliation. The aim of this dissertation was to address these gaps, offering an examination of the social-psychological factors and interventions that can promote reconciliation in separatist conflict. The empirical studies presented in this dissertation assume that separatist conflicts involve two disputing parties harbouring contrary points of view: the majority that resists separatism and the separatist group that advocates its demand for autonomy. I investigated how majority and separatist groups alike think and feel about their involvement in conflict. Uncovering these perspectives has helped me through this dissertation in gaining a better insight into the social- psychological factors that facilitate or hinder support for reconciliation among the majority and separatist groups. This dissertation also examines the effectiveness of social-psychological interventions in attenuating the majority’s defensive reactions to its harm against the separatist group and, in turn, in promoting reconciliation among members of this non-separatist group. Finally, this dissertation confirms that the nature and the dynamics of reconciliation are complex. In this regard, I interpret reconciliation as a multidimensional construct that bears within it attitudinal, affective, intentional, and behavioural components, and assess the relations among these components. This dissertation provides answers to several important research questions. Some limitations remain, however, raising new questions, and potential directions for future research have been outlined accordingly. I hope that by addressing these remaining gaps in our knowledge, follow-up studies will provide a more in-depth account concerning the underlying mechanisms of reconciliation in separatist conflict. These future studies can hopefully provide scholars and practitioners with further insight into social-psychological interventions that are effective in promoting reconciliation not only among members of the majority, which is the focus of this dissertation, but also among those of the separatist group.
Full-text available
Abstract Active white ignorance is accompanied by an epistemic and affective insensitivity that allows American white people to avoid the negative affect that might typically accompany harmdoing. Resisting active ignorance about racism and white supremacy, therefore, often gives rise to shame. Yet, thinkers have debated the value of shame for white people's antiracism. This article asserts that shame is an appropriate response for white people recognizing our culpability for and complicity in racist injustices and violence. However, the article exposes problems with philosophical accounts of white shame, and draws on recent psychological research to show that contextual factors actually determine whether shame can support white antiracism. The article proposes a role for shame in what José Medina calls an “ethics and epistemology of discomfort,” arguing that there are conditions under which shame may encourage the sustained self-interrogation, sensitivity, and humility required if white people are to contribute meaningfully to antiracist action.
Shame is an emotion that is the cornerstone of International Relations (IR) human rights scholarship but remains undertheorized from an explicitly emotional perspective. Given the dubious and unsettled efficacy of human rights “naming and shaming” campaigns, in this article, we outline the theoretical and methodological contours of a research agenda designed (1) to uncover the emotional content of naming and shaming and (2) to pay greater attention to how nonstate actors, especially human rights nongovernmental organizations (NGOs), evoke and experience shame, thus engaging in “emotional diplomacy.” Drawing on theories of emotions in IR and political psychology, we present a thicker account of shame by highlighting the individual and social origins of shame, discussing different varieties of shame, and by distinguishing between emotions that are often conflated with shame. We end with a discussion of the methodological tools suitable for pursuing this agenda, using examples of prominent human rights NGOs.
Existing studies on candidate evaluation have posited that racial cues would invoke negative attitudes toward outgroups, thus lowering support for minority candidates. However, recent studies have found that even implicit racial cues show no negative effect but actually work positively in favor of the minority candidates. In this study, I explore this puzzle by setting up a survey experiment that pairs an Asian candidate against competitors with varying racial backgrounds. Consistent with the existing evidence, I found that White voters tend to support an Asian candidate to a greater degree than a co‐ethnic, White competitor. However, departing from the previous studies that have explained this tendency as a reward for model minority, I argue that such a pattern is associated with reaffirming Whites’ ingroup identity in a racial hierarchy by compensating minorities. When the apparent racial hierarchy—White versus non‐White—is replaced with a minority‐only context, Whites no longer need to favor an Asian candidate and divide their support more evenly to the two minority candidates. I further show that this tendency is moderated by the intensity of their ingroup attitudes.
Während die größte Anzahl Geflüchteter seit dem Ende des Zweiten Weltkriegs Deutschland und Europa erreicht, sind die Reaktionen gespalten. Während ein großer Teil der Bevölkerung sich seit Sommer 2015 in bisher nicht gekannte Maße für die Aufnahme und Integration dieser Menschen engagiert, üben andere sich in Protest gegen die Ankunft eben jener Geflüchteter, erfahren rechtspopulistische Parteien neuen Zulauf und erreicht die Anzahl an Anschlägen auf Unterkünfte für Geflüchtete Rekordniveau. Im vorliegenden Beitrag interessieren uns die Wirkungen eines Fokus auf solche positiven oder negativen Beispiele auf die eigene Bereitschaft, sich solidarisch zu engagieren. Wir diskutieren die Literatur zu den Wirkmechanismen positiver sowie negativer Beispiele um dann die Ergebnisse einer eigenen im Herbst 2015 erhobenen Studie zu referieren.
Full-text available
Content analysis of 122 social psychology textbooks confirmed that displaced aggression received a surge of attention immediately following J. Dollard, L. W. Doob, N. E. Miller, O. H. Mowrer, and R. R. Sears (1939), but subsequent interest sharply declined. Contemporary texts give it little attention. By contrast, meta-analysis of the experimental literature confirms that it is a robust effect (mean effect size = +0.54). Additionally, moderator analyses showed that: (a) The more negative the setting in which the participant and target interacted, the greater the magnitude of displaced aggression; (b) in accord with N. E. Miller's (1948) stimulus generalization principle, the more similar the provocateur and target, the more displaced aggression; and (c) consistent with the contrast effect (L. Berkowitz & D. A. Knurek, 1969), the intensity of initial provocation is inversely related to the magnitude of displaced aggression.