Reicher
Reicher
Abstract
This paper investigates the arguments used in public documents to mobilise Bulgarians against the
deportation of Jews in World War II. We focus on the key documents relating to the first wave of
mobilisation in 1940–1941 as provided by Todorov in The Fragility of Goodness (2001). We
demonstrate that these documents are based on three types of argument. The first, category inclusion,
treats the Jews as part of a common ingroup rather than as constituting a separate outgroup. The
second, category norms, proposes that help for those under attack is a core aspect of ingroup identity.
The third, category interest, suggests that the ingroup will be harmed if Jews are persecuted. In each
case, the predominant category on which arguments are based is national identity (i.e. ‘we
Bulgarians . . . ’). This analysis is used to validate and extend a social identity of model of helping.
The theoretical and practical implications of such an approach are considered in the discussion.
Copyright # 2006 John Wiley & Sons, Ltd.
INTRODUCTION
‘The road to Auschwitz’ wrote Ian Kershaw, ‘was built with hate but paved with indifference’
(Kershaw, 1984, p. 277). Although fanatics and ideologues may have formed the Nazi Party, framed its
policies and committed its worst excesses, they were only able to do so with the compliance—or at
least without the resistance—of the mass of ordinary Germans. As Abel (1938/1986) has shown, anti-
semitism may not have been a central factor even for the mass of Nazi Party members, but they
certainly had to condone the anti-semitism of the organisation. Equally, even the regime’s own
intelligence reports indicated that most Germans were disgusted by Kristallnacht—the great pogrom
*Correspondence to: Stephen Reicher, School of Psychology, University of St Andrews, St Andrews, Fife, KY16 9JU, UK.
E-mail: [email protected]
of 9 November 1938. But the population still accepted discriminatory race laws (Koonz, 2003). As
Koonz puts it, again quoting from the SS Security Service reports: ‘persecution [of Jews] produces no
enthusiasm among most people. But on the other hand . . . Although people despise its extreme forms,
racial propaganda leaves its traces’ (2003, p. 218). That is, few actively tried to make life miserable for
Jewish people, but most avoided contact with them, accepted that their participation in society should
be curtailed and hesitated to object at overt acts of repression.
Koonz (2003) stresses that ‘ordinary Germans’ had both knowledge and choice in these matters.
Even if they didn’t know the exact details of the persecution, that was because they chose not to
enquire too closely—and they certainly knew enough to be aware that their Jewish fellow-citizens
were desperately in need of aid and support. Apart from those few individuals who were politically
suspect, most of these Germans had considerable latitude in deciding where and when to comply with
the regime. Browning (1993) even shows that those soldiers directly ordered to murder Jews had the
option to refuse: it might have dented their career prospects, but it didn’t lead to persecution, let alone
death.
The point can be made more widely. Throughout the Nazi empire, people retained some autonomy
in how they responded to the holocaust. As Gross (2001) observes in his account of the destruction of
the Jewish community in Jedwabne, Poland by their non-Jewish neighbours: ‘a number of those actors
could have made different choices, with the result that many more European Jews could have survived
the war’ (p. 12). The implication, as encapsulated in the title of Hilberg’s seminal text (Perpetrators,
Victims, Bystanders; 1993), is that we cannot fully understand the holocaust by focusing on the Nazis
and the Jews alone. We must always consider the moderating role of bystanders. How they chose to act
made a critical difference.
The significance of these choices is underscored by considering those cases where people did help
Jewish populations. In recent years, increasing attention has been devoted to such cases (e.g. Cesarini
& Levine, 2002; Hilberg, 1993; Marrus, 1987). Consider the case of Poland, which is often thought of
as most complicit with the holocaust. After the liquidation of the Warsaw Ghetto a leading official of
the World Jewish Congress reflected on how such a thing could have happened without any local
opposition. He concluded that it was due to the Poles’ blind hatred for Jews which made them co-
responsible for the slaughter (Hilberg, 1993). Yet even here there were two organisations devoted to
helping the Jews. One was a civilian Council for Aid to Jews (Zegota), the other was organised by the
Armia Krajowa (an underground resistance movement attached to the Polish Government in Exile
in London).
In Romania, Jews may have been expelled from the occupied territories (Northern Bukovina and
Bessarabia which were lost to the Soviet Union in 1940 and recovered in 1941) but the German request
to deport the Jews of ‘Old’ Romania was refused. In Denmark and Finland the Jews survived
unscathed—the Danes arranged for the entire Jewish population to be transported to safety in Sweden.
However, perhaps the most remarkable example of rescue occurred in Bulgaria where a series of
petitions, individual letters of protest and public demonstrations against anti-semitic measures
ultimately prevented the deportation of the indigenous Jewish population. Various commentators
have pointed out that this was the only case where Jews largely survived within a country that was in
the pro-German camp and where, at the end of the war, there were more Jews living than before it had
started (Arendt, 1990; Ben-Yakov, 1990; Cohen & Assa, 1977; Genov & Baeva, 2003). As one
survivor wrote in a letter to the historian of the holocaust, Martin Gilbert, this was ‘the miracle of the
Jewish people’ (Gilbert, 1985).
Todorov (1999, 2001) has compiled a selection of the key contemporary texts relating to the
Bulgarian rescue. Using these public documents, this paper will consider the bases on which authors
sought to mobilise the population into a mass opposition. How did they argue against the oppression of
Jews? What was the basis on which they appealed to their audience to oppose anti-semitic measures?
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
Saving Bulgaria’s Jews 51
First, though, let us consider existing understandings of the factors which determined how bystanders
responded to the holocaust.
In recent years holocaust scholarship has turned to questions such as who helped the Jews, why they
helped the Jews and when they were effective. Much of this work is concerned with the structural and
political conditions of rescue. Clearly, such factors as the nature of German control, the structure of
government in different countries (and sometimes between different parts of the same country), the
point in the war when deportations were proposed and the prospects of German defeat at that time,
were all of critical importance in terms of affecting the choices that were available to people, the
possibilities of saving Jews and the costs of so doing (Geras, 1995; Hilberg, 1993; Marrus, 1987).
However, there is also a sizeable psychological literature which is concerned with the factors which
made individuals liable to intervene in the holocaust (Baron, 1985/1986; Bauer, 1989; Marrus, 1989;
Monroe, 1990, 1995, 1996, 2003; Oliner & Oliner, 1992; Tec, 1986). Not surprisingly, such a large
literature contains a host of findings, many of which are contradictory. There is no clear consensus
concerning the relationship of social class, gender, political affiliation or religion to rescue. However,
there is agreement that, for religious rescuers, what counted was the particular moral content of their
beliefs. This relates to the one finding on which nearly all the research concurs. It is that rescuers
shared a worldview which Oliner and Oliner (1992) term ‘extensivity’. That is to say, they included
victims as part of a common moral community with themselves and saw an obligation to help all
members of that community who were in need. Most commonly, they saw this moral community as
including all human beings and saw it as self-evident that anyone in need deserved their help whether
they were relatives, friends, Jews or indeed Nazis. However, this was not always the case. Sometimes
the subgroup might be more restricted (say, to members of the same nation) and sometimes it might
even extend beyond the human race to include all living creatures (Monroe, 2003).
In the main, such a moral outlook has been treated as a stable character trait. This is certainly true of
the Oliners who were seeking to isolate ‘the altruistic personality’ (which is the title of their 1992 text).
However, as Monroe (2003) argues, there are a number of difficulties with such an approach, not least
that many seemingly reprobate characters showed conspicuous virtue during the holocaust (Oskar
Schindler being a notable example) and many seemingly virtuous characters showed a singular
disinclination to help. In response, Monroe (2003) argues that moral values affect action only to the
extent that they are incorporated as part of identity. She further argues that identity is multi-faceted and
that different aspects of identity may become salient in different circumstances. In this way, context
can come to affect moral stance and the likelihood of rescue.
Monroe’s argument explicitly draws on a number of social psychological sources, notably social
identity theory (Tajfel, 1978, 1982; Tajfel & Turner, 1979). Moreover, her account of holocaust rescue
echoes developments in the social psychology of helping. A number of models have begun to
incorporate the idea that categorising victims as ingroup members is part of the intervention process.
Thus, recent formulations of the ‘arousal: cost-reward’ model (Piliavin, Rodin, & Piliavin, 1969;
Piliavin, Dovidio, Gaertner, & Clark, 1981) suggest that a sense of ‘we-ness’ leads to feelings of
greater closeness and responsibility for the welfare of others which in turn increases both arousal and
the costs of not helping (while decreasing the costs of helping) and these in combination make
intervention more likely (Dovidio, Piliavin, Schroeder, & Clark, 1991). Equally, a reformulation of
Batson’s empathy–altruism model (Batson, 1987, 1991) by Cialdini and his colleagues suggests that
we will be more empathic and hence more helpful to others when we see them and ourselves as both
included as part of an extended self-concept (Cialdini, Brown, Lewis, Luce, & Neuberg, 1997).
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
52 Stephen Reicher et al.
Most directly, work in the social identity tradition proposes that group relations are critical to the
way in which we respond to the plight of others. Indeed, it is built into the premises of this tradition
that we will be more concerned with the fate of ingroup members and more likely to respond to the
needs of those who share a common group membership with ourselves (Tajfel, 1978). Two strands of
research in particular have developed this idea. Levine, Reicher and colleagues (Hopkins, Reicher,
Harrison, Levine, & Cassidy, 2004; Levine, 1999; Levine, Cassidy, Brazier, & Reicher, 2002; Levine,
Prosser, Evans, & Reicher, 2005; Reicher, 1996a) address the helping literature. They argue that we
are more likely to aid those who we categorise as ingroup members and therefore who we aid will vary
as a function of how we identify ourselves. Thus, in one study (Levine et al., 2005) supporters of
Manchester United football club witnessed a person falling over and hurting themselves. That person
was wearing either a Manchester United shirt, a Liverpool shirt (a rival team) or a non-football t-shirt.
When participants were led to categorise themselves as Manchester United fans they helped only when
the ‘victim’ was wearing a Manchester United shirt. However, when they were led to define
themselves in terms of the more extensive category ‘football fan’, they helped both when the ‘victim’
was wearing a Manchester or a Liverpool shirt.
The implication of the study is that more inclusive self-categorisations will increase the scope of
helping behaviour. This accords with the philosopher Richard Rorty’s discussion of ‘solidarity’ (Rorty,
1989) where he suggests that it is incumbent upon us always to try and extend our ‘we-communities’
so as to appreciate the pain of others and eschew cruelty towards them (note that solidarity can either
be used to denote a ‘fellow feeling’ that may lead to actions that benefit others or to the actions
themselves, we shall use it in the latter sense and hence it constitutes a broader behavioural category of
which helping and rescue form more specific instances). The difference is that Rorty considers it
impossible for this community to embrace all of humanity whereas, from a social identity perspective,
there is no reason why one cannot define oneself as human (v non-human) or even more inclusively
(say as a living creature v the non-living).
A second strand of research is concerned with volunteerism—that is to say, the conditions under
which people will give time to organisations devoted to social care. A number of studies have shown
that people are more likely to volunteer when they identify themselves as belonging to the same group
as the recipients of care. Thus Simon, Sturmer and Steffens (2000) have shown that the more gay
people identify themselves as such the more willing they are to engage in AIDS volunteerism (given
that in Germany, where the study was carried out, the largest subgroup among people living with AIDS
are gay people). Similarly, Sturmer and Kampmeier (2003) have shown that the more people identify
with their local community, the more likely they are to volunteer for the local fire brigade. The
importance of this work is that it shows that social identity processes do not simply relate to individual
acts of face to face helping. They are also important to participation in collective movements which
aim to alleviate the plight of others (see Simon, 2004).
However, it is important to recognise that the drawing of ingroup–outgroup divisions is only one
part of the argument. For social identity theory, behaviour can never be derived simply from the fact of
group membership. Rather one must take into account the content of the specific identity that is made
salient. That is to say, our actions depend upon the beliefs and norms of the groups with which
we identify (Jetten, Spears, & Manstead, 1996; Reicher & Hopkins, 2001a, b; Turner, 1999). So, while
we may be generally inclined to help ingroup members, that does not mean we will always neglect
outgroup members. Whether we do so or not depends upon what our group norms have to say about
our obligations to others. For certain groups, there may be strong injunctions to ‘look after our own’
while, for others, the need to attend to those in need, irrespective of whether they are members of our
group or not, may define who ‘we’ are (Hopkins et al., 2004; Levine, 1999). Indeed in certain cases,
pro-social behaviour such as charity and helping may be the dimension along which we differentiate
our group from others (cf. Jetten et al., 1996; Reicher, 2004).
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
Saving Bulgaria’s Jews 53
Analytically, then, we need to distinguish between helping based on ingroup inclusion and helping
based on ingroup norms. In general, the focus of the former is on who the victim is (‘I helped her
because she was one of us’) whereas the focus of the latter is on who the helper is (‘I helped her
because we are concerned with everybody’s welfare’). To put it slightly differently, this argument
suggests that there are two dimensions of social identity which impact upon helping and social
solidarity. The first relates to the category boundaries. The second concerns the meanings associated
with group membership. If these dimensions are so important for helping—and if helping can be
affected by working along these dimensions—then the question of how category boundaries and
category meanings are defined becomes of crucial importance practically as well as theoretically.
Traditionally, the definition of social identity has been seen primarily in perceptual terms—as an
intra-psychic and cognitive process. Thus categories are computed in such a way as to reflect the
existing organisation of social reality (Oakes, Haslam, & Turner, 1994). This is also the perspective
taken by those, such as Monroe (2003) who have used social identity concepts in order to explain
the phenomenon of rescue. By contrast, those who have developed the social identity approach to
helping behaviour have highlighted the rhetorical dimension of category definition (although it should
be stressed that this is more a matter of emphasis than disagreement and that there is increasing
recognition that both dimensions need to be addressed—cf. Haslam, 2001). That is, they argue that,
precisely because of the importance of social identity for social action, those who seek to shape
collective movements will do so by arguing over category definitions. Such people can be described as
‘entrepreneurs of identity’ who seek to make their projects a reflection of the social identities of their
audience. To put it more conceptually, social identities are not simply perceptions about the world as it
is now but arguments intended to mobilise people to create the world as it should be in the future
(Reicher & Hopkins, 1996a,b, 2001a, 2004; Reicher, Hopkins, & Condor, 1997a,b).
In the case of helping, then, we might expect that those who wish to create social movements in
favour of intervention might do so, firstly, by construing social categories in such a way that victims
and potential helpers form a single ingroup and/or, secondly, by construing norms in such a way that
humanitarian action is a central tenet of the group. The implication is that helping is neither a fixed
function of personality nor automatically invoked by context. Rather it is something that can be
actively created through argument. Helping is not something individuals come to alone through
internal processes. It is something that can be publicly mobilised.
Such a perspective has the potential to make an important contribution to our understanding of large
scale social phenomena such as rescue during the holocaust. Indeed it points to a limitation of the
existing literature. This research tends to concentrate on individual acts such as harbouring Jews from
the Nazis. Heroic and important as these acts were, it is arguable that social movements which
involved large numbers of people in opposition to deportation (even if the commitment of each
individual was much lower) were more significant in saving Jewish people from the ‘final solution’. In
Denmark and, over a more extended period, Bulgaria, such movements stopped deportation in its
tracks rather than sheltered a limited number of Jews from the consequences.
It is therefore critical to understand how it is possible to mobilise significant portions of a
population—including those who have not been socialised into what Monroe (2003) calls a ‘virtue
ethic’ of seeing oneself as having an obligation to humanity as a whole—against genocide. To the
extent that a rhetorical social identity perspective can help explain such mobilisations then it may not
only contribute to an understanding of rescue during the holocaust but also of how to avoid atrocities in
the future. The aim of this paper, as we have already indicated, is to contribute towards such an
understanding by analysing the means by which Bulgarians were mobilised to oppose anti-semitic
measures during the Second World War.
To be absolutely explicit, our focus is on the appeals that were made to the population in public
documents and not on the response of the population itself. The value of studying such documents (as
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
54 Stephen Reicher et al.
opposed to private letters, diaries and so on) is precisely that they were intended for and available to
the population at large, moreover they are still available for analysis today (see Cohen & Assa, 1977;
Todorov, 1999, 2001). In these texts we have before us what Bulgarians had before them during the
war. We can therefore use them to discover the grounds on which Bulgarians were asked to oppose the
holocaust. Our specific concern is with the way in which the identity of this population was defined in
the text, whether and how these identity definitions were used to argue against the oppression of Jews,
and how widespread such argument were within these texts. More precisely, we will analyse the way in
which: (a) Jews were either included or excluded from the ingroup category; (b) opposition to anti-
semitism was linked to ingroup norms.
METHOD
Historical Context
In order to interpret the texts we are about to analyse, it is first necessary to understand something of
the context in which they were written. But first, a proviso is necessary. In his entry on Bulgaria in the
Encyclopedia of the Holocaust (Gutman, 1990), Ben Yakov notes that accounts of Jewish survival are
coloured by ideological bias, with differing sources wishing to allocate credit to different actors. Yet,
despite these differences, which mean that sources must be handled with care, there is a considerable
consensus as to what happened and who was involved in the rescue. Accordingly, the following
account is based on what is agreed between a number of texts (Arendt, 1970; Bar-Zohar, 1998;
Ben-Yakov, 1990; Boyadjieff, 1989; Chary, 1970; Cohen & Assa, 1977; Genov & Baeva, 2003;
Gilbert, 1985; Todorov, 2001). Only where there is disagreement between the texts, or claims are made
in one that are uncorroborated by others, will specific citations be provided.
Bulgaria was part of the Ottoman empire until 1878 when, with Russian support, it won independence.
In World War I, Bulgaria was an ally of Germany and was punished at the Paris Peace Conference of 1919
by losing approximately 10% of its territories and being burdened by heavy reparation payments. This led
to a prolonged period of instability which was only resolved in 1936 when King Boris III took complete
control of the country and exercised what was called ‘controlled democracy’. That is, the National
Assembly (Subranie) continued to exist and even contained some opposition members, both social
democratic and communist—although in 1942 the Bulgarian Communist Party embarked on a course of
armed struggle against the government and, from 1943 the left social democrats joined the Communists in
an anti-government ‘Fatherland Front’. However, from 1936 most members of the Subranie were
supporters of the government which was appointed and controlled by the King.
One of the major priorities of the regime was to regain the confiscated territories and rid itself of the
reparations. Accordingly, Bulgaria formed close ties with Germany both politically and economically.
Indeed by 1939, 68% of foreign trade was with Germany. However, at the outbreak of war Bulgaria
declared neutrality and was courted by both Stalin and Hitler. In particular, pressure mounted from
Germany to join the Three Partite Pact (Germany, Italy, Japan) signed in September 1940. This was
seen as the best way to regain what Bulgarians saw as their ‘rightful territories’. Accordingly the
country became what Genov and Baeva (2003) call ‘a passive and unwilling ally of Germany’.
However one of the prices for Hitler’s assistance was the introduction of anti-semitic legislation. Some
50 000 Jews lived in Bulgaria at the time, of whom some 25 000 lived in the capital, Sofia. They were
primarily workers, petty tradesmen and pedlars. Few were bankers, businessmen or professionals.
In mid-October a ‘Law for the Defence of the Nation’ was introduced by the Minister of the
Interior, Petur Gabrovski. It was based on the Nazi Nuremberg Race Laws, except that ‘Jews’ were
defined by religion, not ‘blood’. Restrictions were imposed upon residence, property ownership and
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
Saving Bulgaria’s Jews 55
the right to practise certain professions. The Jewish community campaigned vigorously against the bill
and gained widespread support from organisations and individuals both outside and inside parlia-
ment—notably the Bulgarian Orthodox Church which represented some 87.5% of the population,
including many converts from Judaism. This was reflected in a series of open letters, petitions,
individual letters of protest and declarations of position (many of these were displayed in an exhibition
in Sofia and are reproduced in Cohen & Assa, 1977). These in turn mobilised public sympathy and, as
Ben-Yakov (1990) states, the majority of public opinion was against the bill. However, it was
supported by the members of the pro-Government majority in parliament and accordingly passed into
law on 21 January 1941.
Even before the law was passed, however, Bulgaria’s relationship to Germany underwent a
fundamental change. Following Italy’s defeat by Greece in December 1940, Hitler felt it necessary
to move his troops through Bulgaria to assist Mussolini. King Boris III faced a choice between
occupation or alliance and opted for the latter. On 20 January 1941 his Council of Ministers approved
the German passage and on 1 March 1941 the Prime Minister formally signed a treaty of adherence to
the Axis Powers. On the same day Southern Dobroudja was returned to Bulgaria and shortly
afterwards it gained Thrace, Macedonia and parts of Eastern Serbia, thus realising the longstanding
nationalist dream of a ‘Greater Bulgaria’.
Over the following two years the conditions for Jews gradually deteriorated. A critical point was
reached with the arrival in Sofia of Adolf Eichmann’s special envoy, the SS Officer Theodor
Dannecker. On 22 February 1943, he signed an agreement with Aleksander Belev, the Bulgarian
Commissioner for Jewish Questions, to deport 20 000 Jews ‘as a first step’ (Ben-Yakov, 1990).
Originally the text referred explicitly to Jews from Thrace and Macedonia, but these words were struck
out. In fact, the deportation of Jews from these occupied territories went ahead between 20 and 29
March 1943. In total 11 343 people were taken to Auschwitz and Treblinka. Twelve survived.
Belev, however, was still short of his target figure and hence targeted the population of ‘old
Bulgaria’. He chose to start with the Jews of Kyustendil, a town near the old border with Serbia. The
town sent a delegation to Sofia to oppose the deportation. It was joined by Dimitar Peshev, Vice-
President of the Subranie and member for the town. He organised a letter of protest signed by 42 other
parliamentary representatives of the majority party. Although government pressure forced 13 to
withdraw their signatures and although Peshev himself was subjected to a vote of no confidence and
forced out of his post as Vice-President, the deportation was suspended and Belev resigned.
Shortly afterwards, a new plan was devised which involved the expulsion of Jews from Sofia,
pending their deportation from the country. The plan was published on 22 May 1943 and on 24 May a
demonstration of several thousand Jewish and non-Jewish Bulgarians marched to the royal palace in
protest. The event has been described as second only to the Warsaw ghetto uprising as an act of
resistance to the holocaust (Genov & Baeva, 2003) and it was at least partially successful. Although
19 153 Jews were driven from the capital, this was the climax of their persecution. Henceforth the
King categorically refused any deportations from Bulgaria.
In August 1943, the King died shortly after visiting Hitler. This led to persistent rumours that he had
been poisoned for defying the extermination policy. Since the Crown Prince was a minor, three regents
were appointed and a new government was formed. By October, Sofia’s Jews were allowed to return.
In August 1944 the ‘Law for the Defence of the Nation’ was rescinded. The next month, the regime
collapsed and the Soviet army occupied the country.
Texts
As we have outlined, the persecution of Bulgarian Jews falls into three periods: the passage of the
‘Law for the Defence of the Nation’ (October 1940–January 1941), the deportation of Jews to the
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
56 Stephen Reicher et al.
extermination camps (March–April 1943) and the deportation of Jews from Sofia (May 1943). Our
analysis will rely principally on the documents provided by Todorov relating to the first phase. This is
partly for practical and partly for principled reasons.
First, for reasons of space it is not possible to analyse all of the documents, but at the same time, the
first phase gives a sufficient number of public documents—as opposed to diaries, confidential reports
and minutes of closed meetings—to provide an adequate corpus. The number is seven as opposed to
four and two for phases two and three respectively.
Second, the documents in the first phase are clearest with respect both to the intended audience and
the intended outcome. Although they might have been addressed to specified figures the documents
were public documents that were intended to reach and influence the population as a whole. Open
letters were reported in the media, speeches to parliament were reproduced in the parliamentary record
and reported on both newspapers and radio stations. The authors’ aim was specifically to induce the
widest possible opposition to the new law and thereby put pressure on parliament to reject it. Knowing
who was being addressed and to what end, we can therefore analyse whether and how the documents
constitute identities in order to try and achieve these ends. In the second and third phases this is less
straightforward because there is more variety in the targets of the appeals and mobilization goals.
Thus, some texts (e.g. Peshev’s letter signed by majority party parliamentarians) are aimed specifically
within the party to demonstrate that deportations would harm its ability to govern. Other texts (notably
those by Communists supporting the ‘Fatherland Front’) are aimed to mobilise people to join the fight
against the regime as a whole.
The texts chosen by Todorov (1999/2001) to represent the large number of documents protesting
against the ‘Law for the Defence of the Nation’ are as follows:
Document 1: Statement by the Bulgarian Writers Union to the Prime Minister and the Chairman of
the National Assembly: 22 October 1940
Document 2: Statement by the Governing Board of the Bulgarian Lawyers Union to the Chairman of
the National Assembly: 30 October 1940
Document 3: Open letter from Christo Punev to the National Assembly Deputies (also signed by a
range of writers, professionals and politicians including three former ministers): not
dated
Document 4: Statement by the Holy Synod of the Bulgarian Orthodox Church to the Prime Minister:
15 November 1940
Document 5: Open letter from Dimo Kazasov (a journalist and politician affiliated to the non-
Communist opposition) to the Prime Minister: not dated
Document 6: Speech by Petko Stainov (an academic theoretician of jurisprudence and leader of the
non-Communist opposition) in the National Assembly: 19 November 1940
Document 7: Speech by Todor Polyakov (Communist opposition member) in the National Assembly:
20 December 1940.
There are two reasons to accept Todorov’s selection of these documents. The first is that they cover
the various categories of public opposition to the ‘Law for the Defence of the Nation’, including
professional organisations (documents 1 and 2), prominent ‘opinion leaders’ (documents 3 and 5), the
Bulgarian Orthodox Church (document 4), and the twin strands of parliamentary opposition, social
democratic and communist (documents 6 and 7). The second reason is that all the documents are
singled out in other sources as being of significance. Thus Ben-Yakov’s entry in the Encyclopedia of
the Holocaust explicitly mentions documents 1, 2 and 4, and indirectly refers to 3, 5, 6 and 7. Likewise,
Genov and Baeva (2003) explicitly mention documents 1, 2, 4, 6 and 7 while indirectly referring to 3
and 5. Cohen and Assa (1977) refer to documents 1, 2, 4, 6 and 7 and quote extensively from
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
Saving Bulgaria’s Jews 57
Kazasov’s letter (Document 5). The letter is also cited in Kasazov’s entry as one of ‘The Righteous
Among the Nations’ at Jerusalem’s Yad Vashem Holocaust Monument.
The originals of these documents can be found in the Bulgarian Historical state archives. The texts
of these documents (which will be coded as D1–D7 in the analysis) can be found in English in Todorov
(2001) pp. 45–69. They can be found in French in Todorov (1999) pp. 59–87. There are obvious
problems in analysing texts in translation (see Eco, 2004; Fairclough, 1992), especially when
undertaking close analyses of the deployment of social categories. These problems are exacerbated
by the fact that the French translation was undertaken by Todorov himself whereas the English version
was translated from the French by Arthur Denner. Thus the English text which we present in the paper
involves a double process of translation, and this must be kept in mind when evaluating the analysis.
However, as a minimum safeguard, once the analysis was completed using the English text, it was
checked against the French text by the first author in order to see if this would lead to any changes. It
did not.
Analytic Method
Our analysis of the documents is based upon SAGA (Structural Analysis of Group Arguments) as
outlined by Reicher and Sani (1998) and illustrated by Sani and Reicher (2000). SAGA is not a method
in the sense of an invariant set of procedures. It is better understood as a general means of investigating
the issues that arise from a stance which sees categories as constructed, debated and contested but
which also sees there to be systematic consequences which arise from category constructions. On the
one hand, then, it is necessary to look in detail at the process of contestation and the specific category
definitions which arise within it. This is a qualitative exercise involving ‘thick description’ (Geertz,
1973) of texts. On the other hand, it is necessary to show how constructions, or relations between
constructions, are predictably associated with given outcomes. This is a task of synthesis which is
amenable to summary description and quantitative analysis. The particular forms taken by these
different phases of the analysis depend upon the precise issue that is under investigation.
In the present case, we wish to investigate whether our texts contain definitions of either category
boundaries or category content which imply either that the Bulgarian Jews are part of a common
ingroup with the general audience (category inclusion arguments) and/or that the audience belongs to a
category for whom helping people irrespective of who they are and to what group they belong
(category norm arguments).
This requires us, in the first stage of analysis, to provide a detailed qualitative account of all of the
arguments used in the documents, paying particular (but not exclusive) attention to: (a) the way in
which the audience and the Jews are categorised, and more particularly to cases where the audience
and the Jewish population are included in a common category; (b) the ways in which any norms are
associated with the audience category, and more particularly where norms are characterised in such a
way as to require intervention (for details of how arguments are identified, see Reicher & Sani, 1998).
As we will show, a third broad form of category argument emerged from this stage of the analysis,
namely (c) considerations of the way in which deportations would impact on audience category
interests.
In a second stage of the analysis, we provide a systematic inventory of which arguments are used in
which documents, with a focus on how many of the documents employ each of argument types: (a)
category inclusion; (b) category norms; and (c) category interests. We also show how many employ at
least one of these three types.
All four authors undertook both stages of the analyses, reading all the texts, classifying the
arguments and assigning argument types to individual texts. Any differences of coding were resolved
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
58 Stephen Reicher et al.
through discussion. Where it was agreed that there are ambiguities in the material, these are indicated
in the text.
ANALYSIS
Our analysis is divided into two parts corresponding to the two stages of analysis. In the first part we
present the full range of category argument that were used: that is, (a) arguments relating to category
inclusion; (b) arguments relating to category norms; (c) arguments relating to category interests. In the
second part of the analysis we will summarize the frequency with which these argument types were
used within the texts.
Extract 1 (D1): ‘the bill’s objective is to deprive a Bulgarian national minority of its civil
rights . . . . Our legislature must not approve a law that will enslave one part of Bulgaria’s citizens,
and leave a black page in our modern history.’
Extract 2 (D3): ‘Have you not walked by the children of Yuchbunar [a working class quarter of
Sofia with a large Jewish population] on the streets of the capital? Little children and students,
have you not seen them, famished, jaundiced, wasted and ragged, marching alongside Bulgarian
children on Cyril and Methodius’ Day? Have you not heard them, their voices hoarse, singing O
Dobroudja [a patriotic song calling for the return of the town of Dobroudja which was ceded to
Romania under the 1913 Treaty of Bucharest] and all the other songs of our nation’s spirit at the
top of their weak little lungs.’
Extract 3 (D7): ‘I will refute the claim that the Jewish minority threatens the nation, and,
consequently, that measures must be taken to defend it . . . Bulgaria’s Jews . . . speak and think in
Bulgarian, have fashioned their style of thinking and their feelings after Botev, Vazov, Pencho,
Slaveikov, Yavorov [Bulgarian cultural icons] etc. They sing Bulgarian songs and tell Bulgarian
stories. Their private selves are modelled on ours—in the street, on our playing fields, at school,
in the barracks, in workshops and factories, in the mountains and the fields, our sufferings are their
sufferings, our joys their joys too.’
Extract 4 (D6): ‘And so, gentlemen, we come to the bill’s second clause, which sanctions a
number of important restriction to which Bulgarian citizens of Jewish origin are to be subject . . .
In singling out a group of people in order to assign them a particular status, and in restricting their
basic rights, this bill, in Article 15, Paragraph A, relies, as I said, on a sui generis racism, one
based on birth and blood. I do not subscribe to racial theories. Racial purity is a fairy tale. I do not
believe in fairy tales and I am not about to draw conclusions of inequality amongst our citizens on
the basis of an ill-founded theory of racism and racial purity, no matter how it is presented here.
The term ‘‘pure race’’ is a fiction. Who among us, knowing the history of this land, can say ‘I am
racially pure?’’’
These four extracts are presented at the start of our analysis not because they are the only passages
relating to category inclusion, but because, alone and in combination, they allow us to make a number
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
Saving Bulgaria’s Jews 59
of important points. First, and most fundamentally, all of the extracts presuppose a national framework
and include the Jews as part of the national ingroup. While our focus is more on what category
constructions are used rather than the detailed discursive means by which they are accomplished (see
Edwards, 1996; Potter, 1996), it is worth noting that this framework is sometimes made explicit
through the use of terms such as ‘nation’, ‘country’ or ‘Bulgaria’ but frequently invoked implicitly.
Take the term ‘Our legislature’ from extract 1. Insofar as this is discussing the legislature that is
considering the ‘Law for the Protection of the Nation’ it refers to the Subranie or National Assembly.
Hence the term ‘our’ invokes a national positioning. Another example of this is found in extract 3.
After stressing how Jewish culture and thought is Bulgarian, the text continues ‘Their private selves
are modelled on ours . . . our sufferings are their sufferings, our joys their joys too’. This only makes
sense if ‘ours’ and ‘our’ refers to Bulgaria and hence the audience are presupposed to have a national
identity. As Billig (1995) argues such deixis of small words is central to the way in which nationhood
becomes ‘banal’: an unnoticed, everyday, but for that, all the more crucial frame for viewing the world
and ourselves in the world. In other words, as we have already suggested, the most powerful way of
imposing category definitions may be not to argue for them but to take them for granted.
As with the imposition of the national category, so the place of Jews within this category is
sometimes taken for granted. This is true of extract 1 from document 1 in which the term ‘Jew’ never
appears. Rather, the target of the legislation is always referred to in national terms: as ‘a Bulgarian
national minority’ or else as ‘one part of Bulgaria’s citizens’. However, where categories are already a
matter of public debate it may be less effective to treat them as givens and a failure to argue may render
one more likely to succumb to the arguments of others. Although the use of a national framework to
debate a piece of national legislation may have been uncontroversial, the inclusion of Jews within that
framework certainly was. After all, the whole point of the legislation was to place restrictions upon a
supposed danger to the nation. In its introduction, the law stated ‘the Jews are an evil and a foreign
element among the Bulgarian people that acts against the State’ (cited in Ben-Yakov, 1990, p. 266).
The texts we are considering are counter-arguments to this claim. One aspect of that counter
argument is the assertion that Jews couldn’t be a threat even if they were alien. Hence they are
generally described as a minority and they are frequently portrayed as weak (‘famished, jaundiced,
wasted and ragged’ in the terms of extract 2). In several cases the argument is stated quite explicitly.
Indeed the passage that precedes this poignant description challenges the accusation that Jews are
speculators who damage the economy. The authors retort:
Extract 5 (D3): ‘the vast majority of Jews in Bulgaria are working class people: small grain
merchants, pushcart vendors, retail tradesmen, labourers and maids, all of them working for a
living and all of them going hungry.’
Another aspect of the counter argument is an explicit claim that, far from being alien, the Jews are
thoroughly Bulgarian. This is true of extracts 2 and 3. These extracts stress the commonality between
Jews and other Bulgarians. They invoke many of the conventional criteria of nationhood (common
territory, language, culture, loyalty—cf Anderson, 1983; Connor, 1994) to underscore the Bulgarian
identity of the Jewish population. Moreover, especially in extract 3, it is stressed that adherence to
these criteria is not merely a matter of public conformity that could mask an underlying difference:
the psychic worlds of Jews—their thoughts, their feelings, their private selves—are thoroughly
Bulgarian.
Extract 4 is even more explicit in its counter-argumentation. That is, it does not only use an
inclusive formulation at the outset (‘Bulgarian citizens of Jewish origin’) but then goes on to contest
the validity of exclusive (i.e. racial) formulations. These are dismissed as fairy tales and as fiction,
especially in Bulgaria where ‘knowing the history of this land, no-one can say ‘‘I am racially pure’’’.
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
60 Stephen Reicher et al.
One implication of this is that all Bulgarians are equally mixed and equal in being mixed. A second
implication is that, while the speaker knows about the national past, anyone who asserts exclusive
racial doctrines must be ignorant of Bulgaria—not fully Bulgarian themselves. This is to invoke a
further facet of the argument which arises in a number of documents.
As we have argued elsewhere (Reicher & Hopkins, 2001a), if category definitions are contested,
then it becomes necessary to establish why one’s own version, as opposed to others’, should hold sway.
There are various ways in which this is done. One, as in extract 2, is to invoke references to shared
cultural symbols of the category such as national songs. Another, as in extract 3, is to invoke cultural
icons such as national heroes and national poets. Such usage is given further authority by the use of
poetic language in the extract itself. What is more this language uses vivid illustrations that, in the
terms of social representations theory (e.g. Farr & Moscovici, 1984), render the construction more
familiar and accessible by making it concrete.
Another way of establishing one’s category definitions is by establishing one’s own credentials as a
person who has the right to speak for the category (or else disqualifying one’s rivals by challenging
their category credentials). The simplest and most powerful way of doing this is by establishing
oneself or ones organisation as a typical part of the ingroup (Reicher & Hopkins, 2001a, 2004), and it
is something that appears in all of the documents. It is what the writer is doing in extract 4 when he
writes about ‘knowing the history of this land’. It is also to be found in the first words of extract 1
which read: ‘Dear Prime Minister, we, the undersigned Bulgarian writers . . . ’. This is, of course, an
accurate description of the authors, but as discursive psychologists point out (Edwards & Potter, 1992;
Potter, 1996), no description is neutral. The authors could equally accurately have been described as
just ‘writers’ or as ‘intellectuals’. The term ‘Bulgarian’ serves to establish their ingroup prototypi-
cality. The point is even clearer in extract 6 below:
Extract 6 (D3): ‘I am just an ordinary Bulgarian citizen, and so if I speak to you today, it is
because I know in my heart that you too—sons and grandsons of those who died for freedom on
the gallows, on the hillsides and in the ravines of the Balkans, or on the executioner’s block at
Batak [site of Turkish repression of the Bulgarian insurrection of 1876]—have not forgotten the
oath they swore with their last breaths: Let us protect humanity and freedom!’
Here the author is entirely explicit in defining himself as an ‘ordinary’ Bulgarian citizen. Were he to
be categorised as exceptional (say, in terms of his position as an MP or his profession as a journalist),
he would be open to the charge that his views are also exceptional and differ from those of the
population in general. By being ‘ordinary’ the author makes himself similar to his fellow Bulgarians
and knowledgeable about what it means to be Bulgarian.
As for the identity of the audience, the issue is more nuanced. At one level the author presents them
as unambiguously Bulgarian—indeed as descendants of national martyrs. However, if they are
Bulgarian and live up to their ancestors, they will share the views of the author and protect the Jewish
population. If they fail to do so, they will be denying their heritage. They will be worse than foreigners,
they will be renegades. In sum, then, these texts do not simply create a common ingroup including
victims and helpers, but a tripartite relationship of inclusion between the author, the Jews and the
audience. It is this which allows the author to speak for the audience in including Jews amongst them.
But is inclusion always achieved through the use of national categories? Document 2, from the Union
of Bulgarian lawyers, does indeed employ national inclusion, but it also contains a passage in which
professional inclusion is stressed. It refers to ‘our colleagues of Jewish extraction’ and states that:
Extract 7 (D2): ‘As a group, they have always been upstanding members of our order and have
always assumed their professional and moral duties as lawyers. Some, both in the past and still
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
Saving Bulgaria’s Jews 61
today, have served as members of the governing bodies of our organisations and institutions and
have carried out the responsibilities entrusted to them with diligence and dignity’
This could be read as a form of inclusion which invokes the aid of non-Jewish lawyers for their
Jewish colleagues. However, given that the document is not addressed to lawyers but rather is
addressed from lawyers to the National Assembly, it might better be read as a concrete example of how
Jews are part of Bulgarian national life and national institutions and that, far from a law being
necessary to defend the nation against Jews, the Jews actively contribute to the national good. In the
context of the document as a whole, then, this apparent counter instance forms part of the pattern of
national inclusion.
Finally, consider document 4 which is addressed to the Prime Minister from the Holy Synod of the
Bulgarian Orthodox Church:
Extract 8 (D4): ‘the bill makes no distinction between Israelite Jews and those Jews who, though
of unbaptized parents, have personally adopted the Christian faith. It treats these Christians and
the Israelite Jews in the same way. The Christians of Jewish origin, who have personally adopted
the Christian religion 5, 10, 20 years ago or more, are, by their faith, by their religious and folk
customs, by their language and by their culture, naturally linked with the Bulgarian people; they
have severed their ties with the Jewish community and have assimilated into the Bulgarian people,
yet this bill forcibly separates them from our people by placing them in the same category as the
Israelite Jews.’
This extract differs from all those we have seen thus far by denying that all Jews are Bulgarians.
Only Christians are Bulgarians and hence only Jews who have converted to Christianity acquire the
language, the culture and the faith to be included in the Bulgarian nation. Note that the criteria of
nationhood are the same as in extracts 2 and 3. Note also that the appeal for solidarity is ultimately
grounded in the Bulgarian identity of converted Jews (i.e. save them not because they are Christian but
because, being Christian, they are Bulgarian). Hence, even if Christianity is the ostensible inclusive
category, in actual fact the argument is identical to those we have previously encountered in the sense
that Jews deserve solidarity as fellow Bulgarians. It differs in the sense that only some Jews (converts)
are Bulgarian and therefore only some deserve help on this basis—as document 4 goes on to state:
‘(t)hat is why our national Church is compelled to speak out in defence of its children who are bound
both to it and, by their faith, to the Bulgarian people.’ In answer to the question we posed above, then,
even where alternative categories are invoked (professional, religious), category inclusion tends to be
underpinned by the national category.
All this may seem to obscure a glaring omission. What of universalistic descriptions of the victims:
not as Jews or even as Bulgarians but as fellow human beings. The reason for this omission is simply
because such instances are all but absent in the data. There is only one clear example in the documents
we are considering. It comes from document 7: ‘As Gorky says: ‘‘Man: how proud that sounds’’. We no
longer consider it acceptable to inflict cruelty on animals, yet now we are about to reduce thousands of
innocent and law abiding people to the status of half-men . . . .’ Yet this is not to say that universalism is
lacking in the documents. Rather, it takes a somewhat different form as we shall show in the next section.
Whereas category inclusion is almost exclusively based on national categories, this is not true of the
use of category norms. In document 3, for instance, Christo Punev addresses members of the National
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
62 Stephen Reicher et al.
Assembly in the following terms: ‘As deputies, you have a legal and moral duty: to defend the
Constitution. This duty is unquestionable, as you know perfectly well, better than I.’ He concludes (just
after the passage reproduced in extract 5 that finishes with the words: ‘Let us protect humanity and
freedom!’):
Extract 9 (D3): ‘All Bulgarian citizens capable of bearing arms swore and still swear to defend
Bulgaria. And you, gentlemen, as deputies, you have sworn to do something else as well—to
defend the Constitution!’
In document 5, Dimo Kazosov addresses the Prime Minister and notes his various group memberships:
as head of the national educational system, as President of the Bulgarian Academy of Sciences, as
Professor at Bulgaria’s only university, as president of the Bulgarian Pen Club. He then continues:
Extract 10 (D5): ‘Everyone also has the right to expect you to show a heightened sensitivity to any
attempt to condemn defenceless citizens to a moral death, incite the young generation to shameful
violence, falsify historical facts, blacken the reputation of writers, political figures, scholars, and
soldiers, and question the loyalty of any and all who are proud enough not to think like your friends’
In other words, the Prime Minister’s various group memberships are invoked in order to represent
anti-semitic measures as anti-normative to each of them. An educationalist should not corrupt the
young, a scientist should not distort facts, a President of the Pen Club should not disparage his
members—and so on. In this way, the use of norms allows for a direct appeal to members of small but
influential groups to protect those less powerful than themselves. However, these extracts can also be
read on another level. That is, they are intended to put indirect pressure on political leaders by
mobilising the broader (national) population against them for supporting the proposed law. Any
leaders who attack ‘defenceless citizens’ reveal themselves to be unworthy of their position.
Thus the use of normative arguments relating to non-national categories is not in opposition to the
use of national categories. Indeed, the two were often complementary and, although not exclusive,
norm arguments based on nationhood were found in every document. Such arguments occurred at a
number of levels. To start with, the very use of nationhood is inherently normative (see Billig, 1995;
Taylor, 1989). That is to say, there are certain assumptions as to how a nation should be. Notably, it is a
sovereign and horizontal community (Anderson, 1983): one which is in charge of its own affairs and
within which all are equal as nationals. Unless these assumptions are explicitly contested they will be
presupposed and do not need continuous restatement. Correspondingly, it can be presupposed that
violation of these assumptions—either to divide the people or else to impose a foreign will upon it—
will be seen as an assault on the nation. It is on this basis that one can read the following statements as
accusations that the proposed legislation attacks the norms of nationhood. First, Christo Punev
sarcastically suggests to members of the National Assembly that:
Extract 11 (D3): ‘if you think our country should be destabilised during the critical days ahead,
when what we need more than ever before is peace and harmony among all citizens living under
our Bulgarian sky, then you will vote with the instigators and authors of this bill.’
Extract 12 (D7): ‘why is this law being proposed? That question troubles many of us, and so we
have to ask ourselves whether it just might be that it is being imposed on us by foreigners and by
foreign interests.’
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
Saving Bulgaria’s Jews 63
As well as invoking norms based upon nationhood in general, another form of argument is to invoke
specific norms relating to the Bulgarian nation in particular. We have already seen an instance of such a
norm argument in extract 5, in which it was stated that the invocation of martyrs to the nation was: ‘Let
us protect humanity and freedom!’. There are many similar examples where it is asserted that
Bulgarian culture is associated with a series of norms and values that impel them to help the oppressed,
the weak and the vulnerable. They have ‘traditions of religious tolerance and humanity, won at so
great a cost’ (document 1). Their cultural originality lies in their ‘humanity, justice and compassion
for all those who suffer’ (document 5). They are a people who are ‘tolerant and honourable’
(document 6) or else ‘tolerant, hospitable, good’ (document 7).
As in the case of category inclusion, these attributes are sometimes just stated and sometimes spelt
out and justified. Justification can be done formally, in relationship to core national texts, notably the
constitution. As Stainov states in his speech to the National Assembly (document 6): ‘the proposed
legislation violates our Constitution, particularly Article 57, which states that all Bulgarian citizens
are equal before the law. Equality before the law is one of the fundamental preconditions of Bulgarian
constitutional and public law.’ Not surprisingly, the Union of Bulgarian Lawyers makes similar points
in their statement (document 2) and they also cite article 57.
More usually, however, history is invoked. This is already hinted at in the reference in Document 1
to ‘traditions of religious tolerance and humanity, won at so great a cost.’ This invokes the long period
of foreign domination and the birth of modern Bulgaria in a war of independence against the Turks—a
war in which most national heroes such as Vasil Levski and Christo Botev gained their reputations. In
extract 13 this link between history and identity is elaborated:
Extract 13 (D3): ‘Now and then you [National Assembly members] will have heard what Bulgarians
are saying; in the cities and in the countryside they are all saying the same thing: ‘‘If only Levski and
Botev [heroes of the war of liberation against the Turks] were here today, they would make a whip
out of the rope that hanged the Apostle; they would chase us down and flog us to make us
understand their ideas and just what is this liberty in the name of which they died for Bulgaria.’’’
At one level, these are strongly universalistic arguments. Because of their own sufferings and their
own lust for freedom, Bulgarians respond to suffering irrespective of who experiences it and they
support everybody’s right to freedom. However, at another level, they are strongly particularistic.
Solidarity is a distinctive characteristic of the national ingroup. This is eloquently expressed by Dimo
Kazasov:
Extract 14 (D5): ‘When the Bulgarian people lose their sense of justice that they have nurtured
over the course of centuries and that is so much a part of their national identity, they will lose their
moral and spiritual uniqueness, their Slavic essence, their Bulgarian face.’
Likewise, the Synod of the Bulgarian Orthodox Church may only use category inclusion arguments
to defend converted Jews, but it uses general moral precepts to oppose persecution of all Jews. In
document 4 it is asserted that: ‘even in those of its clauses that target only Israelite Jews, the proposed
law contains measures that cannot be considered just or useful for the defence of the nation’
(document 4). Yet, at the end of the document it is made clear that these moral precepts are something
that attach specifically to the Bulgarian nation. Thus a change in the law is advocated as a means of
‘safeguarding our reputation as a freedom-loving, just and tolerant people.’ However, there are
important differences between this use of nationhood and that in extract 9. Whereas the latter
characterises ‘justice’ as something which distinguishes Slavs and Bulgarians from other national
groups, here no such distinction is made.
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
64 Stephen Reicher et al.
There are a number of similar instances. For instance, in document 3, Christo Punev writes: ‘Before
raising your hands to ratify a shameful and inhuman law, a law that goes against all civilized norms,
you should have a look at the history of our people, their struggle for freedom.’ In document 1, the
Writers Union refers to past sufferings of the Bulgarian people and then asks:
Extract 15 (D1): ‘Should we then imitate these atrocities and follow a similar and dangerous road
that will lead us to lose our place among the world’s free and civilized peoples?’
In these cases, the extracts can be read as saying that humanity is not something that separates
Bulgaria from other nations, but rather something that binds her to them. In other words, humanity is a
criterion which qualifies Bulgaria to belong to the world of nations. It is a norm of nationhood in
general rather than a norm of Bulgaria as a specific nation. In such usage, opposition to the law can be
seen to rest on norms that are universalistic rather than particularistic both in terms of who should be
helped and who should do the helping. But at the same time, even given such a reading, this is still a
universalism for which access comes through national identity: the audience are not positioned as
civilized individuals but as a civilized people. It is as Bulgarians that they belong to civilization and are
expected to act accordingly.
In one sense this takes us back to our earlier argument concerning the norms of nationhood, but in
another sense it moves us forward. For the expressed motive in these extracts is not only intrinsic (we
should adhere to our own norms) but also extrinsic (we are in danger of being punished by others).
Passing the law would lead to a loss of status and possible sanctions by others. Hence these extracts
introduce a third general type of argument: this is to do with the interests that would be served or that
would be threatened by the legislation.
As with inclusion and norm arguments, interest arguments were generally presented in categorical
rather than individual or general terms. The first and simplest argument was in fact a counter-
argument. Given that the legislation was labelled a ‘Law for the Defence of the Nation’, thus implying
that Jews had to be constrained because they threatened Bulgarian interests, it was frequently asserted
that there was no such threat and hence the law did not benefit the nation. This was done in a number of
ways: sometimes, as we have already seen in extracts 2 and 5, by implying or stating overtly that Jews
are too powerless to constitute a threat, sometimes (as in extract 7) by arguing that they are exemplary
and dutiful members of Bulgarian society and Bulgarian institutions.
However, as well as contesting the notion that legislation supports the national interest, a number of
arguments are used to suggest that it actively endangers Bulgarian interests. As ever, there are times
when it is simply asserted that the law will, in the words of the Bulgarian Writers Union (document 1):
‘be very harmful to our people.’ Others spell out some of these dangers. These include the danger of
destabilizing the country at a time when unity is crucial (see extract 11); the danger of weakening the
Bulgarian economy; and the danger of exposing Bulgarian minorities living in Thrace and Macedonia
to greater oppression:
Extract 16 (D2): ‘Our concern, our struggles to defend these oppressed minorities will lose much
of their judicial and moral foundations if we impose restrictions and arbitrary measures on a
national minority here at home’
This last example in particular underlines the categorical basis of such interest arguments. The
audience for this appeal were Bulgarians living in the country while the subjects of the appeal were
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
Saving Bulgaria’s Jews 65
Bulgarians living outside the country. If anything, defiance of the law made German intervention more
likely (see Genov & Baeva, 2003), hence, on an individual level, it threatened audience interests.
However, it is as nationals that the protection of unknown fellow Bulgarians furthers their collective
self interest.
There is another noteworthy aspect of this example. Whereas arguments about division and
economic damage suggest that persecution of Jews will directly harm Bulgaria, the fate of national
minorities introduces the idea that the harm will be mediated by the reaction of others to such a
measure. The most common and straightforward form of this argument is that oppression would
damage the international reputation of Bulgaria. It is encapsulated by Dimo Kazasov when he suggests
in Document 5 that the law would lead Bulgaria to: ‘lose a moral capital accumulated over many long
years.’ He then goes on to spell out the concrete consequences of such a loss:
Extract 17 (D5): ‘The war that is being waged here at home against the Jews will not pass
unnoticed there: it will inevitably put us morally at odds with public opinion in these various
countries, even as our national interests dictate that we maintain cultural as well as economic ties
with them. Powerful countries, rich in both material and cultural resources as are the Germans and
Italians, can permit themselves the luxury of such discord, but small countries like ours must avoid
it for their own good. We need all the friends, the compassion and the help we can get.’
Finally, if the law does not defend the national interest, if it is damaging to the national interest
in so many ways, then in whose interest is it? The answer was made explicit by Todor Polyakov in
extract 12 earlier: legislation serves (German) outgroup interests and its proponents if not puppets
of the outgroup are at least its dupes. At the risk of making a point that will already be obvious, the
ingroup/outgroup category in this case, as in every case we have encountered under ‘category
interest’ arguments, is a national category. The law is nearly always judged against whether it is
good or bad for Bulgaria and Bulgarians. The only clear exception comes in the statement of the
Holy Synod (Document 4). If all Jews are repressed for being Jewish they will have no choice but to
return to their religion and hence conversion will become impossible. Thus ‘such measures benefit
no one, neither the Church nor our people, whose interests would be better served if they and men of
other origins were united.’ But far from invoking a broader category—such as humanity —the
argument in fact invokes a narrower category beside the nation: the Bulgarian Orthodox religious
establishment.
Having analysed the arguments used in the documents, we are now in a position to summarise
our analysis by showing which arguments are used in which document. This is presented in Tables 1
and 2.
In Table 1 we summarise the arguments by providing a general description of each one. The
arguments are divided into the three types identified in Part 1 of the analysis: category inclusion
arguments, category norm arguments and category interest arguments. These in turn will be
subdivided in order to consider what type of categories the argument is based upon: national
categories, other categories (professional, political, class etc.) and universalistic categories (all human
beings).
In Table 2 we indicate which of these arguments is to be found in each of the seven documents. In
some cases there are ambiguities and these will be explained in the notes. These ambiguities generally
relate to points discussed at more length in Part 1 of the analysis.
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
66 Stephen Reicher et al.
National Jews are Bulgarians Bulgarians are civilized, Bulgaria will be harmed
tolerant and humane by anti-semitic measures
Other categorical Jews are fellow Politicians uphold the The Church will be unable
professionals constitution which demands to recruit Jewish converts
equality
Universal Jews are fellow Human beings should act in None found
human beings a civilized way
Document 1 þ þ ? (note 4) þ
Document 2 þ ? (note 2) þ þ þ þ
Document 3 þ þ þ ? (note 4) þ þ
Document 4 ? (note 1) ? (note 3) þ þ þ þ
Document 5 þ þ þ ? (note 4) þ ? (note 5)
Document 6 þ þ þ
Document 7 þ þ þ þ þ ? (note 4) þ
Key: þ: Argument appears in document; : Argument does not appear in argument; ?: Ambiguous case.
Note 1: In this document, the Holy Synod of the Bulgarian Orthodox Church includes Jewish converts to the Church as
Christian and hence as Bulgarian. Non-converts are excluded from the nation.
Note 2: The Union of Bulgarian lawyers praises Jews as exemplary members of the profession, however this could be read as an
example of their being exemplary Bulgarians.
Note 3: The Holy Synod includes Jewish converts as Christians, however this is used to argue that they are also Bulgarian.
Note 4: These are all instances where ‘civilisation’ is invoked and Bulgarians are characterised as ‘civilised’ but, in the absence
of explicit comparisons, this could be read as linking Bulgaria to a wider category of ‘civilised peoples’.
Note 5: Kasasov talks of the danger to young scientists if anti-semitism leads to Bulgaria being ostracised internationally,
however this can be read as an example of the overall danger to Bulgarian interests.
DISCUSSION
Three clear conclusions can be drawn from our summary of the use of arguments as illustrated in
Table 2. The first is that there is a rich diversity of category arguments contained in the appeals for
Jewish rescue. These do not just relate to category inclusion and category norms but include a third
type of argument not addressed previously: category interests. The second is that category arguments
are extremely widespread. All but one of the documents employs all three types of category argument
(category inclusion, category norm and category interest) and all of the documents use some form of
category argumentation. The third is that the predominant (but not exclusive) form of this
argumentation is based on national categories: Jews are Bulgarians; as Bulgarians we show solidarity
to others; Bulgarian interests are threatened by anti-semitic measures.
What are the implications of these findings for our understanding of the bases for which Jewish
rescue was argued (and which, more widely, may serve to promote social solidarity)? In order to
address this question, there are a number of inter-linked issues surrounding our analysis which need to
be considered. To what extent do our findings relate to the specific context we are dealing with—that is
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
Saving Bulgaria’s Jews 67
opposition to the proposed ‘Law for the Defence of the Nation’? Why do we find such a preponderance
of arguments based on national categories? What explains the difference between our findings and
previous analyses which stress the universalistic basis of rescue (Geras, 1995; Monroe, 1996, 2003)?
If we take Billig’s (1987) observation seriously that all positions are a statement in an argument,
then inevitably, the documents we have analysed are shaped by what they are arguing against—that is,
legislation based on the premise that Jews as a category are a threat to Bulgaria as a category. Such a
proposition invites (and receives) the retort ‘oh no they are not, and if anything it is the legislation that
threatens Bulgaria.’ That is, the specific context can easily be used to explain both the existence of
category arguments and the overwhelming preponderance of nationhood as the basis for such
arguments.
However, the first thing to say is that, if our findings relate to context, this is a context that, in its
general features, is extremely common not only in the Nazi holocaust, but also in other cases of
genocide. Nazi anti-semitism was premised upon characterising the Jews as ‘community aliens’ who
endangered the German nation (Koonz, 2003; Peukert, 1987). Similarly, the Rwandan genocide was
premised upon the notion that Tutsis were outsiders who dominated and polluted the country (Des
Forges, 1999). Solidarities always exist in response to exclusions and, since exclusions are so often
based upon nationhood, then our findings may well have wider applicability than the specific case
under consideration.
This argument is supported by considering some of the texts relating to later waves of repression in
Bulgaria. Thus, in response to the evacuation and internment of Sofia’s Jews in May 1943, a letter to
King Boris from a number of public figures asserts that:
‘In subjecting our innocent fellow citizens to this cruel and pitiless measure, not only are we
squandering a vast moral capital of which our generous and tolerant people had every right to be
proud, we are also harming Bulgaria’s reputation in the eyes of the world and compromising its
future national interests.’
In this single sentence we can see all three types of argument—category inclusion (‘our innocent
fellow citizens’), category norms (‘our generous and tolerant people’) and category interest (‘harming
Bulgaria’s reputation’). What is more, all three are unambiguously national. Thus the use of such
arguments is not dependent upon a law that invokes nationhood in its very title but rather relates to the
more general logic of exclusion. So, while the particular focus on national interest arguments in our
documents and the particular form these national interest arguments take may relate to the specific
context, we would argue that the more general use of national category arguments is of wider
significance.
Of course, this only raises a further question—why is the play of exclusion and inclusion centred
around nationhood? For us, the use of categories relates both to who one is trying to mobilise and also
to the mobilisation goal. In terms of the former, national categories appeal to an entire territorial
population rather than small sections of it—as would, say, categories of class, religion or even gender
(Reicher & Hopkins, 2001a). Thus using nationhood as a basis of exclusion has the advantage of
allowing advocates to appear as champions of the great majority and to mobilise the majority to
achieve their ends. More specifically, using nationhood as a ground for advocating solidarity has the
advantage of mobilising the great majority of the population against exclusion. If Jews are Bulgarians,
then the entire population should feel attacked by an attack on Jews. If Bulgarians are humanitarian,
then everybody is impelled to support Jewish victims.
In terms of the latter, the content of mobilisation, it is arguable that the specific histories of
particular nations and the general ideology of nationhood also lend an advantage to the use of national
categories in mobilising against persecution. Where a country has been formed through a struggle for
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
68 Stephen Reicher et al.
independence against a powerful and dominant other, then notions of ‘freedom’, of ‘justice’ and the
need to fight oppression will be highly available and highly charged. We are not suggesting that such
nations will necessarily eschew persecution or that nationhood cannot be used to argue for persecution
(it obviously can), but rather that when people want to challenge persecution, the writing of national
history will provide them rhetorical resources with which to do so. In this sense, history is not a
determinant of the present, but provides a ‘symbolic reserve’ (Reszler, 1992) which can be drawn upon
in building for the future (see also Reicher & Hopkins, 2001a; Reicher, Hopkins, & Condor, 1997b).
Thus we see extensive invocation of the Bulgarian independence struggle, and of its key figures, within
the documents we have been analysing.
In addition to Bulgaria’s specific history, the ideology of nationhood likewise provides powerful
rhetorical resources: if Jews can be included as nationals then any attempt to make them ‘second class
citizens’ violates the idea of a ‘horizontal community’. If a law can be characterised as arising out of
external pressure, then it can be rejected as violating the principle of ‘sovereignty’. If a measure can be
presented as ‘uncivilized’ then it threatens the right of one’s country to belong to the family of nations.
There is an important point here. The predominance of national categories should not be seen as
counterposed to universalism. Rather, in a world of nation states, it is through characterising what it
means to be a nation that one arrives at norms that aspire to encompass all of humanity.
But while there may be a series of advantages to the use of national categories in advocating
particular types of mobilisation, the thrust of our argument—and our claim to generality—is not to do
with the prevalence of any particular category, but rather in the relationship between the type of
category arguments that are used and the form of mobilisation that is advocated. That is, effective
arguments will be those which use categories best able: (a) to include all those one is seeking to
mobilise; and (b) to provide the resources with which to render normative the actions one is
advocating. Depending on the forms of action one seeks to organize, sometimes sub-national
categories will suffice. For instance, if one were appealing for help from a particular section of the
population—say the Trade Unions—one might argue that most Jews, like themselves, are poor
workers and an attack on them is an attack on the working classes. Equally, there are times when more
extended categories will be preferred since, at the human level, all victims are necessarily included in a
common ingroup.
Hence we do find examples of category arguments in our documents which are based on these more
restricted or more extended categories. In other contexts, involving different audiences and different
victims, it may well be that they would predominate over the national level of categorisation. To
reiterate the central point, the specific categories that are used to advocate help will always depend
upon context; the importance of construing relations of inclusion between helpers and victims, of
construing helping as normative for the audience ingroup and victim suffering as a threat to ingroup
interests—whatever categories are used to achieve them—obtains irrespective of context.
But still one must explain why, in addressing what is apparently the same phenomenon of rescue,
our findings concerning which categories are used to advocate intervention are so at odds with existing
research on the bases for acts of intervention: why do we find a preponderance of national rather than
universal positions? One part of the answer has to do with whether we are, in fact, looking at the same
phenomenon. As we argued in the introduction, we are dealing with the phenomenon of collective
mobilisation rather than individual acts of rescue. Our concern is with how the mass of ordinary people
can be induced to take a relatively small and safe step (but, collectively, one of massive significance)
rather than why relatively few exceptional people made major and highly dangerous leaps.
We have described the literature which suggests that individual rescuers tended to have a worldview
based on universalist categories which included any victim as a fellow human being and that this
worldview was often a product of a long socialisation. However, by this very token, it follows that
most people will not have such a worldview and will not view people in such terms. In the language of
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
Saving Bulgaria’s Jews 69
social identity (or rather, to be more accurate, self-categorisation theory), universalist categories will
simply not be available to them (Oakes, Turner, & Haslam, 1994). This being the case, an attempt to
mobilise people in universalist terms is unlikely to be successful. It is far more powerful to use such
potentially inclusive categories as will be available to all those in the target population, in terms of
which they will have been socialised and which they use regularly.
Clearly, nationhood meets these criteria. In a world of nation states, national categories suffuse our
everyday understanding, from the school books we read to the way the weather is described (Billig,
1995; Citron, 1989). Thus both mobilisers and the mobilised are far more likely to use these
categories, to be familiar with national references and to accept arguments based on nationhood. In
short, mobilisation is dependent upon the use of categories which are banal even if the ways they are
construed (i.e. who is included within them, the norms and interests ascribed to them) is highly
creative. The use of exceptional categories which shape the acts of exceptional people are therefore
less likely to be effective. For these reasons, we would suggest that the differences between our
findings and previous findings are indeed because we are looking at somewhat different phenomena
even if they may both be included under the rubric of ‘rescue’. The reason why we find few arguments
based on universalistic categories (and that, when we do, they are often arrived at through conceiving
the world as a system of nations) is not because people cannot relate to them or because, being more
extended, they are necessarily weaker (cf. Rorty, 1989) but simply because fewer people are socialised
into them in our divided world.
Let us conclude by considering the more general implications of our study for understanding and
promoting solidarity with the oppressed. The first thing is to acknowledge that this study alone can, at
best, establish only a prima facie case for the role of category construction in creating such solidarities.
What we know from the historical evidence is that there was a significant mass mobilisation against
the deportation of Jews from ‘old’ Bulgaria and that this mobilisation played an important part in
pressuring politicians and ultimately the King to halt the deportation. While, in the short term, the
movement against the ‘Law for the Protection of the Nation’ may have failed to stop the legislation
being passed, commentators from very different standpoints (Cohen & Assa, 1977; Todorov, 2001)
acknowledge that it was crucial in laying down the conditions of future success.
What we have shown in the present analysis is that arguments concerning category inclusion,
category norms and category interests were present, indeed prominent, in the documents that appealed
for opposition to the bill. Putting the two together, we can demonstrate the co-occurrence of certain
category arguments and anti-discrimination mobilisations (and it is also worth recalling that in the case
of the Jews of Macedonia and Thrace where these category arguments were absent and, notably, the
Jews were not seen as Bulgarian, there was no such mobilisation and over 11 000 people were deported
to the death camps). What we cannot demonstrate, however, is whether these arguments were effective
in inducing people to act (see Thompson, 1990). For that, one would have to address how and why
people responded to the appeals, but there is no systematic contemporaneous data on this matter and
post hoc accounts in these matters are notoriously problematic (Geras, 1995; Monroe, 2003). Our
present analysis is therefore necessary, but not sufficient in demonstrating that rescue hinged on the
ways in which bystanders were led to understand their own social identity.
A broader case concerning the actual importance of category arguments for rescue therefore
depends upon combining the findings of this study with other findings derived from different methods.
We have previously conducted experimental studies which manipulate category definitions and
demonstrate that these affect helping behaviour (Levine et al., 2002, 2005). These principally relate
to category inclusion such that the more widely the boundaries of the ingroup are drawn, the more
people are helped. Yet to show that that category constructions can affect helping does not demon-
strate that, in situations where helping becomes a life or death matter, they are used in order to promote
intervention. That is what we show in this study. Moreover, our analysis demonstrates the richness of
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
70 Stephen Reicher et al.
category argumentation which relates not only to inclusion, but to group norms and category interests
as well. The impact of such arguments merit further experimental investigation. There is a broader
methodological point here: in studying a major phenomenon such as helping we do not see there to be
a one way movement from the laboratory to the field, but a constant to-ing and fro-ing where each
domain of study enriches the other (cf. Blumer, 1969).
If this study has suggested a new facet to the social identity model of helping, it also suggests a
number of new facets to our understanding of solidarity and rescue. Most significantly, perhaps, we
need to complement the study of individual rescuers with a study of collective rescue and how it is
mobilised. A social identity account is of particular use in this enterprise, not only at a theoretical level
but also at a practical level. If we only consider helping in terms of long term socialisation then there is
nothing we can do in the short term when solidarity is urgently needed and when populations need to
be mobilised. However, we can always create new ways of defining familiar categories in order to
render solidarity a matter of supporting ‘our’ community, expressing ‘our’ identity or else defending
‘our’ interests.
We have previously shown that the architects of genocide have been all too skilful as ‘entrepreneurs
of identity’ who define categories and category relations such that the ingroup is imperilled by an
outgroup whose destruction then becomes an imperative of (collective) self-defence (Reicher, 1996b;
Reicher & Hopkins, 2004; see also Koonz, 2003). Architects of rescue need to become equally adept
entrepreneurs who construct more inclusive and more humane communities whose interests are served
by acts of rescue. A social identity account, we hope, will help to provide them with the tools for the
job.
REFERENCES
Abel, T. (1938/1986). Why Hitler came into power. Cambridge, MA: Harvard University Press.
Anderson, B. (1983). Imagined communities. London: Verso.
Arendt, H. (1970). Eichmann in Jerusalem: A report on the banality of evil. New York: Viking Press.
Baron, L. (1985/1986). The holocaust and human decency: A review of research on the rescue of Jews in occupied
Europe. Humboldt Journal of Social Relations, 13, 239–240.
Bar-Zohar, M. (1998). Beyond Hitler’s grasp: The heroic rescue of Bulgaria’s Jews. Avon, MA: Adams Media.
Batson, C. D. (1987). Prosocial motivation: Is it ever truly altruistic? Advanced Experimental Social Psychology,
20, 65–122.
Batson, C. D. (1991). The altruism question: Towards a social psychological answer. Hillsdale, NJ: Erlbaum.
Bauer, Y. (1989). Remembering for the future. Oxford: Pergamon.
Ben-Yakov, A. (1990). Bulgaria. In I. Gutman (Ed.), Encyclopedia of the Holocaust (Vol. 2). New York:
MacMillan.
Billig, M. (1987). Arguing and thinking. Cambridge: Cambridge University Press.
Billig, M. (1995). Banal nationalism. London: Sage.
Blumer, H. (1969). Symbolic interactionism. Englewood Cliffs, NJ: Prentice-Hall.
Boyadjieff, C. (1989). Saving the Bulgarian Jews in World Ward II. Singer Island: Free Bulgarian Center.
Browning, C. (1993). Ordinary men. New York: Harper Perennial.
Cesarini, D., & Levine, P. A. (2002). ‘Bystanders’ to the Holocaust. London: Frank Cass.
Chary, F. B. (1970). The Bulgarian Jews and the Final Solution. Pittsburgh: University of Pittsburgh Press.
Cialdini, R. B., Brown, S. L., Lewis, B. P., Luce, C., & Neuberg, S. L. (1997). Reinterpreting the empathy–
altruism Relationship: When one into one equals oneness. Journal of Personality and Social Psychology, 73,
481–494.
Citron, S. (1989). Le Mythe National. Paris: Editions Ouvrieres.
Cohen, A., & Assa, A. (1977). Saving the Jews in Bulgaria: 1941–1944. Sofia: State Publishing House
‘Septemvri’.
Connor, W. (1994). Ethnonationalism. Princeton, NJ: Princeton University Press.
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
Saving Bulgaria’s Jews 71
Des Forges, A. (1999). Rwanda: Leave none to tell the story. New York: Human Rights Watch.
Dovidio, J. F., Piliavin, J. A., Gaertner, S., Schroeder, D. A., & Clark, R. D. III. (1991). The arousal: Cost-reward
model and the process of intervention: A review of the evidence. In M. Clark (Ed.), Prosocial behavior. Review
of personality and social psychology (Vol. 12, pp. 86–118). Newbury Park, CA: Sage.
Eco, U. (2004). Mouse or Rat? Translation as negotiation. London: Phoenix.
Edwards, D. (1996). Discourse and cognition. London: Sage.
Edwards, D., & Potter, J. (1992). Discursive Psychology. London: Sage.
Fairclough, N. (1992). Discourse and social change. Cambridge: Polity Press.
Farr, R., & Moscovici, S. (1984). Social representations. Cambridge: Cambridge University Press.
Geertz, C. (1973). The interpretation of cultures. New York: Basic Books.
Genov, R., & Baeva, I. (2003). ‘Incomprehension of the nature of the race question’: Saving the Bulgarian Jews
from the Holocaust. In G. Halfdanarson (Ed.), Racial discrimination and ethnicity in European history. Pisa:
Edizioni Plus, Pisa University Press.
Geras, N. (1995). Solidarity in the conversation of humankind. London: Verso.
Gilbert, M. (1985). The Holocaust. London: Holt, Rinehart & Winston.
Gross, J. T. (2001). Neighbours: The Destruction of the Jewish Community in Jedwabne, Poland. Princeton, NJ:
Princeton University Press.
Gutman, I. (1990). Encyclopedia of the Holocaust (Vol. 2). New York: MacMillan.
Haslam, S. A. (2001). Psychology in organizations: The social identity approach. London: Sage.
Hilberg, R. (1993). Perpetrators, victims, bystanders: The Jewish catastrophe 1933–1945. New York:
HarperPerennial.
Hopkins, N., Reicher, S. D., Harrison, K., Levine, M., & Cassidy, C. (2004). Social identity, group norms and
outgroup helping. Paper presented to the Annual Conference of the International Society for Political
Psychology. Lund, Sweden.
Jetten, J., Spears, R., & Manstead, A. S. R. (1996). Ingroup norms and intergroup discrimination: Distinctive self-
categorization and social identity effects. Journal of Personality and Social Psychology, 71, 1222–1233.
Kershaw, I. (1984). Popular opinion and political dissent in the Third Reich. Oxford: Oxford University Press.
Koonz, C. (2003). The Nazi Conscience. Cambridge, MA: Belknap Press.
Levine, M. (1999). Rethnking bystander non-intervention: Social categorization and the evidence of witnesses at
the James Bulger trial. Human Relations, 52, 1–23.
Levine, M., Cassidy, C., Brazier, G., & Reicher, S. (2002). Self-categorisation and bystander non-intervention:
Two experimental studies. Journal of Applied Social Psychology, 7, 1452–1463.
Levine, M., Prosser, A., Evans, D., & Reicher, S. (2005). Identity and emergency intervention: How social group
membership and inclusiveness of group boundaries shapes helping behavior. Personality and Social Psychology
Bulletin, 31, 443–453.
Marrus, M. R. (1987). The Holocaust in history. Harmondsworth: Penguin.
Marrus, M. R. (1989). The Nazi Holocaust 5: Public Opinion and Relations to Jews in Nazi Europe. Wesport:
Meckler.
Monroe, K. R. (1990). Altruism and the theory of rational action: Rescuers of Jews in Nazi Europe. Ethics, 101,
117–118.
Monroe, K. R. (1995). The psychology of genocide: A review of the literature. Ethics and International Affairs, 9,
215–239.
Monroe, K. R. (1996). The heart of altruism. Princeton, NJ: Princeton University Press.
Monroe, K. R. (2003). The psychology of altruism. Paper presented to the annual meeting of the International
Society of Political Psychology, Boston.
Oakes, P., Haslam, S. A., & Turner, J. C. (1994). Stereotyping and social reality. Oxford: Blackwell.
Oliner, S. P., & Oliner, P. M. (1992). The altruistic personality: Rescuers of Jews in Nazi Europe. New York: Free
Press.
Peukert, D. J. K. (1987). Inside Nazi Germany. London: Batsford.
Piliavin, I. M., Rodin, J., & Piliavin, J. (1969). Good Samaritanism: An underground phenomenon? Journal of
Personality and Social Psychology, 13, 289–299.
Piliavin, J. A., Dovidio, J. F., Gaertner, S. L., & Clark, R. D. (1981). Emergency intervention. New York, NY:
Academic Press.
Potter, J. (1996). Representing reality. London: Sage.
Reicher, S. (1996a). The Battle of Westminster: Developing the social identity model of crowd behaviour in order
to deal with the initiation and development of collective conflict. European Journal of Social Psychology, 26,
115–134.
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)
72 Stephen Reicher et al.
Reicher, S. (1996b). The crowd century: Reconciling theoretical failure with practical success. British Journal of
Social Psychology, 35, 535–553.
Reicher, S. (2004). The context of social identity: Domination, resistance and change. Political Psychology, 25,
40–62.
Reicher, S., & Hopkins, N. (1996a). Constructing categories and mobilising masses: An Analysis of Thatcher’s
and Kinnock’s speeches on the British miner’s strike 1984–1985. European Journal of Social Psychology, 26,
353–371.
Reicher, S., & Hopkins, N. (1996b). Seeking influence through characterising self-categories: An analysis of anti-
abortionist rhetoric. British Journal of Social Psychology, 35, 297–312.
Reicher, S., & Hopkins, N. (2001a). Self and nation. London: Sage.
Reicher, S., & Hopkins, N. (2001b). Psychology and the end of history: A critique and a proposal for the
psychology of social categorisation. Political Psychology, 22, 383–407.
Reicher, S., & Hopkins, N. (2004). On the science of the art of leadership. In D. van Knippenberg, & M. Hogg
(Eds.), Identity, leadership and power. Oxford: Blackwell.
Reicher, S., Hopkins, N., & Condor, S. (1997a). Stereotype construction as a strategy of influence. In R. Spears,
P. Oakes, A. Haslam, & N. Ellemers (Eds.), Stereotyping and social identity. Oxford: Blackwell.
Reicher, S., Hopkins, N., & Condor, S. (1997b). The lost nation of psychology. In G. Barfoot (Ed.), Ethnic
stereotypes and national purity. DQR Studies in Literature.
Reicher, S., & Sani, F. (1998). Introducing SAGA: The Structural Analysis of Group Arguments. Group
Dynamics, 2, 267–285.
Reszler, A. (1992). L’Europe a la recherche de ses symboles. Les Temps Modernes, 550, 209–220.
Rorty, R. (1989). Contingency, irony, solidarity. Cambridge: Cambridge University Press.
Sani, F., & Reicher, S. (2000). Contested identities and schisms: Opposing the ordination of women as priests in
the Church of England. British Journal of Social Psychology, 39, 95–112.
Simon, B. (2004). Identity in modern society: A social psychological perspective. Oxford: Blackwell.
Simon, B., Sturmer, S., & Steffens, K. (2000). Helping individuals or group members? The role of individual and
collective identification in AIDS volunteerism. Personality and Social Psychology Bulletin, 26, 497–506.
Sturmer, S., & Kampmeier, C. (2003). Active citizenship: The role of community identification in community
volunteerism and local participation. Psychologica Belgica, 43, 103–122.
Tajfel, H. (1978). Differentiation between social groups. London: Academic Press.
Tajfel, H. (1982). Social identity and intergroup relations. Cambridge: Cambridge University Press.
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin, & S. Worchel
(Eds.), The social psychology of intergroup relations. Monterey, CA: Brooks-Cole.
Taylor, C. (1989). Sources of the self. Cambridge, MA: Harvard University Press.
Tec, N. (1986). When light shines in the darkness. Oxford: Oxford University Press.
Thompson, J. B. (1990). Ideology and modern culture. Oxford: Polity.
Todorov, T. (1999). La Fragilite du Bien. Paris: Albin Michel.
Todorov, T. (2001). The Fragility of Goodness. London: Weidenfeld & Nicolson.
Turner, J. C. (1999). Some current issues in research on social identity and self-categorisation theories. In N.
Ellemers, R. Spears, & B. Doojse (Eds.), Social identity. Oxford: Blackwell.
Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 36, 49–72 (2006)