Helping the Persecuted. Heuristics and Perspectives (exemplified by the Holocaust)

Helping and violence

In contexts of collective violence there have never been only exclusion1, abuse, rape, torture and murder. Almost at any time and in every constellation – be it in camps, on death marches or in torture chambers – there have also been acts that reduced the suffering of the persecuted or even helped them to escape persecution if only for some time. The quality of these acts of assistance, especially their duration and intensity, success and motivation, varied strongly; as do, by the way, the intensity with which people participated in persecutions and the reasons or else causes leading to these types of actions. Accordingly, studies on helping are indispensable and, in the best cases, not dealing with help as an isolated subject but as part of an integrated research on violence that as such is as comprehensive as possible. Help and violence belong, after all, to the same field of action or, more precisely: to historical constellations in which people either help those persecuted or kill them. Both modes of action can be found in contexts of events that, from an analytic point of view, are called genocide, massacre or – as mentioned above – instances of collective violence.

Collective violence

Every blow and every kick is an individual action. Violence becomes collective - no matter how tautological that may sound at first – when it occurs in complexes of actions and meanings, as, for instance, in a war or in dictatorial regimes. This first approximation simultaneously establishes a second tautology: In these contexts, violence is always directed against entities (nations; ethnic or political groups etc.) constructed by discourse and performance. In other words, collective violence presupposes forms differentiation between "we" and "they". Violence only becomes possible, allowed, imaginable because agents act or can act as members of a group or are treated as such. Accordingly, violence is collective if it is aimed against members of a collective – regardless of how that collective may be composed. It may also either be predominantly one-directional (e. g. genocide) or mutual (e. g. war). These observations establish the frame within which helping must be defined.

Helping – a definition

As captured in the title, heuristics or else perspectives assumed to discuss a specific type of action are focused upon here. Consequently, actions and agents are not identical and, accordingly, perpetrators, bystanders or rescuers will not be mentioned, but rather killing or helping. It will also not be of interest who, for instance, acts prosocially; instead the genesis of such deeds is brought to the fore: How is an action that may defined as prosocial caused? However, such an approach might be somewhat irritating. While perpetrators and helpers – if the words are used in their conventional sense – may be clearly related either to those to whom something was done or to those who were helped, such a relation may not be instantly ascribed when the focus lies with the action of helping. It is possible to help while executing violence2 or while those who have executed violence are trying to cope with possible negative psychological consequences. At the same time, violent action can be useful for those oppressed, persecuted or for those who require help in this sense. To help is, therefore, not necessarily the opposite of violence and may be (for our purpose) defined as follows:

Helping (in the context of violent action) refers to actions that (are meant to) alleviate the suffering of one (or more) persecuted person(s) caused by restrictions of their options to act.

By no means does this definition solve all definitional problems, but on the one hand, it sufficiently describes the research area in question while, on the other, it sensitizes for the disadvantages of an all too self-evident adoption of the conventional equation between agents and their actions. It has to be added that the aforementioned alleviation of suffering does not have to be the only or even essential intention, but that it is an essential consequence related to the action (and expected by the helper). 

Heuristics

In order to investigate actions in general and helping in particular, a rich arsenal of theories, methodologies and empirical approaches of completely different disciplinary and epistemological provenience are at hand. Some of these will be discussed here not so much as regards their results, but rather their ability to facilitate an understanding of the research subject.

What raises or reduces the likelihood that people help other people? Especially psychological and socio-psychological research since the 1960s has worked out in by now classic studies a set of factors for prosocial (i. e. explicitly helping) behavior in the broader sense of the word. Violence as a frame, for instance, when one person is threatened by others, is only of minor importance in these studies. Usually the experiment manipulated a factor that was likely to influence the probability that one person would help another. Often these were diffuse emergency situations which, for example, were suggested by noises in an adjacent room (Shotland 1985) or highly specific scenarios, for example, when the discovery of an unstamped, seemingly lost letter represented the starting point of the study (Hansson/Slade/Slade 1978). The reactions of the participants were observed, if and how they reacted to the noises and what they did with the letter. Taking this as a basis, influencing factors were deduced and these again provided the basis for extensive theorization. Especially these experimental settings investigate average behavior: A factor considered relevant because it is indicated by hypotheses is varied and subsequently the effect of this variation is measured. If, after some validating calculations, it reaches a certain value, it is considered significant.

One result of these studies was that the recognition of an individual's need for help is a necessary precondition for every helping action. This applies to the cognitive as well as the affective level: If this clarity is missing, processes of evaluation are initiated. The agents attempt to gather additional information, often by observing third parties. This process of orientation slows down the decision to act and this postponement may, in turn, also become part of the process of evaluating a situation: If other people do not act, it might possibly not be necessary. There might even be the risk to lose social recognition because the act to help – such the imaginable fear – might appear ridiculous. If more people are present, the need to evaluate grows so that help is offered less often and to a smaller degree (Latané/Nida 1981). The agents are usually not aware of these effects. Accordingly, subsequent interrogations of the participants of one of the experiments considered this aspect to be negligible (Darley/Latané 1968).

Be that as it may, as a matter of fact it is never the case that all participants help, even if they are alone, and there are always those who act even if conditions impede on their intentions. This means that the investigated variable, in this case the number of people present, only influences the behavior of some of the agents. Consequently, there must be more factors that influence actions but were not captured by the experimental designs.

By pointing this out, both the strengths as well as the weaknesses of such approaches become apparent: They reveal sets of factors that are relevant for the analysis of real, usually historical cases which conventional historical studies do not account for. In other words, they may not deliver explanations applicable to all cases, yet they sensitize for the imaginable. Applied to the example, these might be phenomena such as pluralistic ignorance or the diffusion of responsibility, thus two explanatory concepts for the described phenomenon (cf. Latané/Darley 1970). Secondly, the studies suggest that the described processes of evaluation should be scrutinized on a higher level of abstraction, i. e. that the question should be considered which prompts and possibilities to act agents might have had in specific historical settings. This aspect far exceeds any investigation of helping behavior. Simultaneously it must be considered – and in this respect this heuristic is only partially helpful – that in the world beyond the laboratory it is not always necessary to recognize situations in which help is required in order to become someone who helps; or, what is even more important, to help someone does not necessarily presuppose the wish to help as a central motive of action. This might be the case if a service is paid for or if assistance is perceived to be a social obligation toward a third party

Furthermore, the recognition of a corresponding situation is not a solely cognitive process but also emotionally coded and, in addition, related to evaluations. It is not only necessary to understand that someone requires help, this must also be accepted. The evaluation does not only occur as part of the aforementioned – often communicate – evaluation that takes place next to or else within a specific situation. Moreover, the agents are assessed: For example, they might be held responsible for their situation because their drinking might have caused their fall; or – in case they are Jews – they might be perceived as a threat or an enemy and as such they are – similar to, for instance, a homeless person lying on the ground – only to a limited degree part of the group towards which the agents feel socially and therewith morally obliged and which affords them to observe certain mutually valid rules and, thus a definition by Helen Feins – toward whom obligations are owed, to whom rules apply, and whose injuries call for [amends]" (Fein 1979, 4). For this reason, a situation may, in principle, be recognized as an emergency, yet help is not warranted because – among other things – the person seeking help might not seem unequivocally "eligible" to receive support.

Attitude and situation

Here, the focus is on the impact of attitudes as factors that motivate actions. Taking a closer look, for instance on the relatively well researched helping behavior during the Holocaust, it might be asked if people with anti-Semitic attitudes in principle do not help Jews. Empirical research shows that such a universal statement is not tenable; they do help, yet this depends on further factors and they probably help less often than people who hold different political attitudes. However, attitudes are definitely relevant for behavior, only in other ways as possibly anticipated because they are a lot more than doctrines tied to political ideologies. Accordingly, the attitude to help people in emergencies on principle (for instance because "it is the right thing to do") may supersede the political or racist opposition, depending on which other encouraging or impeding factors play into the situation or are part of it. Meta-studies judge the relation between specific attitudes and actual actions as tenuous (Wicker 1969; Piliavin/Charng 1990). Critically of the results of these studies, Bernd Six (1992) argues that they lack complexity. Yet this criticism does not apply because the studies showed just this, namely that attitudes are not a sufficient determinant of behavior and, accordingly, further contextualization is required. This complexity is accounted for by studies that combine experimental and qualitative research. For instance, Klaus Wahl, Christiane Tramitz and Jörg Blumenthal (2001) investigated how individuals behaved towards strangers. In an experimentally produced waiting room situation autochthone German participants met strangers of a recognizably German or foreign origin. The aim of the study was to observe their behavior towards these people. Then, the results of the data collection were compared to statements uttered by the participants that had been collected independently of the experiment. It could be shown that an agent who in the past had behaved violently against foreigners in the waiting-room-situation conversed in a friendly, curious and interested way with a person that was easily recognizable as non-autochthon German (cf. Wahl/Tramitz/Blumenthal 2001, 166f.). Observations such as these indicate the relevance of situational factors. In the case mentioned, this may apply as much for the setting "waiting room" as for the execution of violence. Depending on the social constellation one and the same individual can thus expose modes of behavior that seem to contradict each other.

In this context, it makes sense that apparently one of the strongest factors triggering help in the case of the persecution of the Jews was the direct address, i. e. a situational factor (cf. Varese/Yaish 2000). It is, by the way, exactly this social interaction that is not accounted for in many of the classic experiments on prosocial behavior. In a study, Frederic Varese and Meir Yaish investigate if and how to address someone correlates with a disposition to help, which is ultimately a character trait. Their calculations, which are based on the interpretation of interviews from the 1980s and which were carried out in the context of the search for an "altruistic personality", confirm such a connection: Whenever a prosocial orientation and the plea for help occur together, the probability grows that actions reducing the pressure of persecution are executed (cf. Varese/Yaish 2005).    

Address

However, not only the persecuted themselves ask for support. Frequently, although this up until now could only be quantified for very specific historical contexts, mediations occur in various respects. This happens directly when a person that is not persecuted asks another to help those persecuted. Already in these cases, fundamental differences of motivation may exist. Those addressed may help because they consider persecution wrong and, therewith, feel obliged to help for political and/or moral reasons. They may also help because of a sense of social obligation they feel towards the inquirer or the group or institution he or she represents. Accordingly, they would not, to stick with the example, help persecuted Jews but an appreciated person. Consequently, the political direction of the action can (!) be, as the example of the study by Wahl et. al shows, disconnected from the action itself. In other words, another appreciated person asking for support in undertaking an anti-Semitic action may likewise receive it.

Furthermore, helping consists of a multitude of individual actions that, case-by-case, are condensed to meaningful narrations on the complex of "helping" indeed only retrospectively. Just like someone distributing the furniture of deported Jews might possibly not be aware of the fact that he participates in genocide, helpful actions may exist that, at the moment of their execution, were not interpreted in the context of the resistance to such a crime. Consequently, helping cannot be investigated as a unified practice; rather it consists of separate practices which, if seen in combination, become what may be understood as helping in the context of collective violence. Individual actions such as providing information or food, transporting people or allowing them to stay overnight are the immediate product of social interaction. For the most part it is done what is conventionally done when being asked for help.

A further context of motivation is related to services that the agents offer in the knowledge of what they are contributing to, yet do so primarily or also with the aim to gain resources. As Marten Düring was able to show with his network analytic approach, it is exactly these interrelated, completely differently motivated actions that are of central importance for the success of complex deeds of helping comprising a number of agents (cf. Düring 2012; 2013). An engaged "hero" acting out of actual conviction is as much needed as someone whose contribution is bought and as all those whose motivation lies somewhere in the continuum between these two extremes. Exactly the same constellation structurally applies to persecution in the context of collective violence, albeit under a different moral presupposition. This is a first indicator as to why helping should be an important topic in research on violence: There are structural correspondences between constellations of motivations, i. e. the ways in which action contexts are motivated. This argument is exactly not stated with the purpose to qualify the moral evaluations of actions in contexts of violence. On the contrary, it is meant to shed light on dynamics of actions, as for example on the fact that people tend to do something when they are asked for it – unless a particular effort or unequivocal "warning signs" prevent them from executing such deeds. This becomes as much apparent when the – in retrospect often almost impossible to answer – question "why people in the past could do such things" is posed as in the arguably most famous experiment of social psychology in which Stanley Milgram explicated the prevalent impact of social obligations even if weighed up against fundamental convictions. 

Perpetrators who help, helpers who abuse

Next to the universe of possibilities to help thus outlined more modes of behavior can be observed that deliver more reasons why helping and helper are to be differentiated. Perpetrators are people who actively participate in oppression or persecution. However, wherever violence can be observed, be it in families of in concentration camps, there are agents who carry out actions that contradict the intentions of the practice of persecution. Motivations of such help are diverse forms of bribery (these may well include sexual favors), but also blackmailing. In addition, there are authentic affective decisions that almost on principle are brought about because of a particular social closeness. It is due to social interaction that the victim is – at least partially and often temporarily – once more included in the universe of mutual obligations even if only because emotions such as empathy come into effect.

Abusive behavior also unfolds within the social dynamic of interactions between people. Bob Moore, for example, in the context of the holocaust reports diverse cases of sexual abuse of women and children in the Netherlands and Belgium by those who accommodated them (Moore 2010, 326f.). In the same historical context, there are a few (which lies in the nature of such things) hints that those seeking shelter were killed. Not much is known about the motives; it is assumed – amongst other things – that the helpers thought the risk was getting too high and that they wanted to rid themselves of all traces making them suspects; this would then accordingly be the result of a processual shift of criteria of evaluation of an action context. Concerning the annexed or occupied Poland, Barbara Engelking argues in her study (according to a review published in German while the study so far has not been translated from Polish) that there were only very few Poles who did not demand payment in return for offering a hiding place and meals, while a large number of the helpers only hid and fed the Jews if they could profit by it. It was not uncommon that the hiding place and the food were revoked when Jews were no longer able to pay. A number of Jews that cannot be ascertained apparently were killed by their Polish "helpers", quite often with the motivation to seize their belongings (Rossolinski-Liebe 2012).

Geneses of actions

Actions are the result of previous actions. They do not follow any logic in the sense that specific conditions and intentions produce predictable modes of behavior. Rather, developments and shifts should be described that only to a certain degree are caused by intentions. This also applies to the social world that is constantly interpreted in cognitive and emotive processes of appropriation which are then translated into options to act: "Our perception is directed not towards the properties of the world as such, but rather towards our being able to use in practice in the context of our actions that which we perceive" (Joas 1992, 158). Thus practices of a performative appropriation of the world come into focus. Performance influences motivations; it might even produce them, as the example of Auguste Leißner illustrates, who, in the context of the national-socialist persecution of Jews, had to be urged by someone close to her to accommodate a Jewish girl who she then, over the period of several years, presented as her relative so that the girl's life was saved. Susanne Beer describes such processes of expansion with the example of another woman, Helene Jacobs, whose initiation into helping began unexpectedly, almost profanely (Beer 2010): In 1938 Jacobs refused a policeman who wanted to speak to her much appreciated Jewish boss, Hermann Barschall, access at a garden gate. In this specific historical context she was able to recognize a threat in this act. Beer shows that this action, from the point of view of the loyal young woman, was implied by the situation because she felt closely related to the Barschalls on an affective level (ibid. 107). This first successful act of help performatively confirms that it is possible to "resist a policeman's intention" so that she "practically experienced a scope of action that she could put to use in the course of her further assistance" (ibid. 108; translation JH). A different experience in this situation might not have encouraged a career as a helper. Once more, the example reveals structural parallels to an opposite action: The members of German police battalions also learned that it is possible to round up all the Jews living in a town and to kill them (Welzer 2005, 124f.). Previously they might have learned that it is possible, if not appreciated, to insult Jews.

These reconstructions of geneses of actions that consider practices and therewith performance are by no means reconstructions of a teleological phenomenological nature, but may constitute one methodological starting point for the further development of research on helping in contexts of violence. A second approach takes the allegedly opposite side as its starting point and devotes itself to structural aspects. One approach is the already mentioned historical network analysis research that shows why help is successful or threatened independently of individual actions. This research on performance and structure provides ample opportunities to be related to already existing action-theoretical concepts such as Clemens Kroneberg's pragmatic approach that investigates the relationship between personality features and environmental conditions: "The (non-)existence of financial resources, the number of rooms, the participation in resistance, the existence of many neighbors and the perceived risk strongly influenced the decision unless the agents had a particularly strong prosocial orientation and were not confronted with a request to help. If the two conditions mentioned last were fulfilled, however, none of the incentive variables influenced the decision to help" (Kroneberg 2011, 302; translation JH). These statistically proven findings in many respects require elaboration and framing so that an exclusive decision for one of the diverse approaches is not relevant. On the contrary, what is needed are integrative approaches that so far have not been devised.

Doers and contributors

"Leaders such as John Weidner and Joop Westerweel were men of strong beliefs and unswerving principles, men who inspired others, attracting them to the enterprise. They were people with a knack for obtaining the resources necessary to carry out their plans successfully" (Fogelman 1994, 211). This characterization neither tells us anything about the nature of their beliefs or principles nor about the resultant plans and actions. Eva Fogelman's emotively heroized depiction refers to men of the Dutch resistance who, amongst other things, helped Jews in hiding in the occupied Netherlands. Despite of its pathos, the quote helps to illustrate two further important aspects relevant to the geneses of actions in the context of collective violence. First, the qualities of the agents thus described are morally neutral. This is inextricably related to the second quality, namely that they indicate a central element for dynamics of actions that may be evaluated differently depending on the onlooker's perspective: So far helping was, on the one hand, considered to be the result of deeds which at the outset were little and isolated and subsequently became self-efficient while, on the other, it was the consequence of structural conditions. Personal dispositions were of minor importance and only came into play at the margins of the discussion, for instance when the limited scope of findings made in experimental settings was pointed out. All this is still valid. Yet, at the same time, even (or especially) those approaches abstracting from such individual actions as, for instance, the historical network analysis point out the central importance of a few agents who, as was illustrated in the example of the help networks for Jews in the Third Reich, were at the center of the networks. These people for the most part were not addressed or even implored to help. They seem to be mostly intrinsically motivated agents who, much more than the helpers they and others recruited, were able to expose deviant behavior. Accordingly, their actions must be explained and investigated differently; in their cases, the individual is indeed more important than the action. An imaginable heuristic to approach these people might be the comparison with other agents who act against officially sanctioned norms, such as pioneers or innovators. A resultant action-theoretically oriented hypothesis might then assume that such people orient themselves considerably less than the average towards the behavior of other agents in the same situation when interpreting a situation/engaging in evaluations that will initiate actions. In an experimental context, they thus will not allow themselves to be influenced by other people's actions. Actually, social psychology of all sciences provides the corresponding hints, for example in the case of aiders at an accident whose locus of control is apparently stronger than that of people who do not help. The aiders assume that they themselves and not the others or "the conditions" are responsible for the run of events (Bierhoff, Klein & Kramp 1991, 271ff.). Even Jan Allyn Piliavin and Hong-Wen Charng, who consider the relation between an altruistic outlook and the corresponding actions to be very tenuous, also cautiously point out that some consistent features could be observed that might support the assumption: "People high in self-esteem, high in competence, high in internal locus of control, low in need for approval, and high in moral development, appear to be more likely to engage in prosocial behavior" (Piliavin & Charng 1990, 31).

Historical-societal-cultural framework

 A comprehensive understanding of the action to help cannot ignore the historical and therewith political and especially cultural framework in which these actions unfold. This includes, on the one hand, the specific logic of certain frameworks such as war (Welzer 2013) or of practices such as pogroms (Bergmann 2002) or torture (Görling 2013). On the other hand, statistical analyses and careful historical reconstructions point out limiting or else encouraging conditions for the dynamics of helping and executing violence. Thus Marnix Croes and Peter Tammes have calculated which conditions influenced the Jews' probability to survive in the Nazi era, e. g.:
•    the more Catholics in a particular region, the higher the probability to survive
•    the more pro-German policemen in a particular region, the lower the probability to survive
•    the more polarized the society in a particular region, the lower the probability to survive
•    the more converted Jews in a particular region the higher the probability to survive
•    the later the time of deportations in a region, the lower the probability to survive (Croes 2006, 484)
And eventually: "More obedience to authority corresponded to a higher percentage of survivors instead of a lower one" (ibid. 488f.).
Results such as these require further theorization and empirical research since they may deliver facts but no explanations. One such theoretical contextualization for the last aspect "obedience" might discuss that authority is neither principally good nor bad, but rather generates specific social conditions, as, for instance, a comparatively homogenous behavior. This in turn may have positive as well as negative consequences for the persecuted, depending on the orientation in the individual historical case.

Ethan Holland analyzes from a politological point of view which options are open to states that want to protect their Jewish citizens, assuming the corresponding political intent exists or political considerations make it appear useful. He refers to the example of Bulgaria to illustrate that the collaboration with Germany opened up chances to negotiate in order to effect, for instance, postponements or exceptions (Hollander 2008)3. The issue of the decision-making process, i. e. why a majority of a collective ready and capable of action decided to help all persecuted Jews, is investigated by Andrew Buckser who picks up the example of Denmark. His interpretation holds that many Danes considered the Danish Jews to be a symbol of their own national independence (Buckser 2001, 22). Starting from a widely popular comprehension of what it means to be Danish, namely to be essentially culturally independent, the rescue of the Jews, who, from the point of view of the Danes, were persecuted exactly because of their cultural features, became an acid test of Danish identity construction: "The moment that the Danes permitted the Jews to be persecuted, they would have effectively abandoned that idea, and with it their claims to independence. In a very real sense, they would have given up their national existence" (ibid.).

Action and field

Regardless of how the various approaches are estimated, they serve as a point of departure for the development of further heuristics and perspectives. As was shown, there are a number of approaches to the research field of helping – experimental approaches, action- and social-theoretical conceptualizations of actions, (collective)-biographical perspectives, historical network-analysis research, statistics, cultural analysis and politological points of view – whose possibilities and limits can only be outlined in this article. Only their combination offers the chance to develop a comprehensive understanding of the action to help. Currently, the relations between performance and structure and those between doers and contributors seem to be of vital importance. In the field of collective executions of violence, however, both dynamics are as relevant for the investigation of helping as for the inquiry into persecuting action. From an emotional and moral point of view these indicators of the structural correspondence between helping and damaging behavior may appear to be an imposition – yet they are appropriate. Moreover they are important because of the extremely large range of aspects/moments that initiate helping behavior. Apart from those who indeed act because of political insight, i. e. civil courage, for most people it is not clear how they will act before they are confronted with the relevant contexts of action. This also applies to the opposite presuppositions, i. e. for those who rape and murder.
Accordingly, the integrative approach to helping in two ways promotes a more thorough understanding of violence. On the one hand, different methodological approaches, even epistemes, must be combined; the presented approaches may be used just as well for the analysis of violent action. On the other hand, any analysis separating the actions of helping and persecuting (as representing the context of the deed) would fall short of enabling a comprehensive understanding because – experimental settings aside – they are interrelated by agents and contexts of action – as for instance war – and influence each other.

 Bibliography

Beer, Susanne, « Helene Jacobs und “die anderen Deutschen”. Zur Rekonstruktion von Hilfeverhalten für Juden im Nationalsozialismus », in Brigitta Schmidt-Lauber et Gudrun Schwibbe (dir.), Alterität. Erzählen vom Anderssein. Göttingen, Schmerse Media, 2010, p. 85-109.

Bergmann, Werner, « Pogrome », in Wilhelm Heitmeyer et John Hagan (dir.). Internationales Handbuch der Gewaltforschung. Wiesbaden, Westdeutscher Verlag, 2002, p. 441-460.

Bierhoff, Hans-Werner, Klein, Renate et Kramp, Peter, « Evidence for the Altruistic Personality from Data on Accident Research », in Journal of Personality 59:2, Juin 1991, p. 263-280.

Buckser, Andrew, « Rescue and Cultural Context during the Holocaust: Grundtvigian Nationalism and the Rescue of the Danish Jews », in Shofar, 19, 2 (2001).

Christ, Michaela, Die Dynamik des Tötens. Die Ermordung der Juden in Berditschew. Ukraine 1941-1944. Francfort, Fischer, 2011.

Croes, Marnix, « The Holocaust in the Netherlands and the Rate of Jewish Survival », in Holocaust and Genocide Studies, V20 N3, Hiver 2006, p. 474–499.

Darley, John M. et Latané, Bibb, « Bystander Intervention in Emergencies: Diffusion of Responsibility », in Journal of Personality and Social Psychology, 8, 4 (1968), p. 377-383.

Düring, Marten, « The dynamics of helping behavior for Jewish refugees during the Second World War: The role of brokerage chains », in Gamper, Markus, Reschke, Linda et Düring, Marten, Knoten und Kanten III. Soziale Netzwerkanalyse in den Geschichts- und Politikwissenschaften. Bielefeld, De Gruyter, 2013.

Düring, Marten, Verdeckte soziale Netzwerke im Nationalsozialismus. Die Entstehung und Arbeitsweise von Berliner Hilfsnetzwerken für verfolgte Juden. Thèse non publiée. Université Johannes Gutenberg Mayence, 2012.

Elias, Norbert, Was ist Soziologie? Munich, Weinheim, 1970 [Qu’est-ce que la sociologie, trad. T. Hoffmann, Aix-en-Provence, Pandora, 1981]

Esser, Hartmut, Soziologie. Spezielle Grundlagen. Band 6: Sinn und Kultur. Francfort/New York, Campus, 2001 [en anglais : « The Rationality of Everyday Behavior : A Rational Choice Reconstruction of the Theory of Action by Alfred Schütz » in: Rationality and Society 1993, 5, p. 7.]

Fein, Helen, Accounting for Genocide, New York, Free Press, 1979.

Fogelman, Eva. Conscience and Courage: Rescuers of Jews During the Holocaust. New York, Toronto 1994. 

Görling, Reinhold, « Folter », in Christian Gudehus et Michaela Christ (éd.). Gewalt. Ein interdisziplinäres Handbuch, Stuttgart, J.B. Metzler, 2013.

Hansson, Robert O., Slade, Kenneth M. et Slade, Pamela S., « Urban-Rural Differences in Responsiveness to an Altruistic Model », in Journal of Personality and Social Psychology, 46, 1 (1978), p. 99-105.

Hollander, Ethan J., « The Final Solution in Bulgaria and Romania: A Comparative Perspective », in East European Politics and Societies, vol. 22, no. 2 (2008), p. 203-248.

Houwink Ten Cate, Johannes, « Mangelnde Solidarität gegenüber Juden in den besetzten niederländischen Gebieten? » in Wolfgang Benz & Juliane Wetzel (éd.), Solidarität und Hilfe für Juden während der NS-Zeit: Regionalstudien 3. Danemark, Hollande, Portugal, Espagne, Hongrie, Biélorussie. Berlin, Metropolverlag, 1999.

Joas, Hans. The Creativity of Action. Cambridge UK 1992.  

Joas, Hans et Knöbl, Wolfgang, Sozialtheorie. Zwanzig einführende Vorlesungen, Francfort, Suhrkamp, 2011.

Jonas, Kai J., Boos, Margarete et Brandstätter, Veronika (dir.): Zivilcourage trainieren! Theorie und Praxis, Göttingen et al., Hogrefe Verlag, 2007.

Kaci, Nadia, Laissées pour mortes. Le Lynchage des femmes de Hassi Messaoud, , Max Milo, 2010.

Kroneberg, Clemens, Die Erklärung sozialen Handelns. Grundlagen und Anwendung einer integrativen Theorie, Wiesbaden, VS Verlag für Sozialwissenschaften, 2011.

Latané, Bibb et Nida, Steve, « Ten Years of Research on Group Size and Helping », in Psychological Bulletin, 89, 2 (1981), p. 308-324.

Latané, Bibb et Darley, John M., The Unresponsive Bystander: Why Doesn't He Help ? New York, Appleton-Century Crofts, 1970.

Löb, Ladislaus, Dealing with Satan: Rezso Kasztner's Daring Rescue Mission, New York, Random House, 2008.

Lüdtke, Alf, Eigen-Sinn. Fabrikalltag, Arbeitererfahrungen und Politik vom Kaiserreich bis in den Faschismus, Hambourg, Ergebnisse, 1993.

Moore, Bob, Survivors. Jewish Self-Help and Rescue in Nazi-Occupied Western Europe, Oxford, Oxford University Press, 2010.

Mühlenberg, Jutta, Das SS-Helferinnenkorps. Ausbildung, Einsatz und Entnazifizierung der weiblichen Angehörigen der Waffen-SS 1942-1949. Hambourg, Hamburger Edition, 2012.

Piliavin, Jan Allyn et Charng, Hong-Wen, « Altruism: A Review of Recent Theory and Research », in Annual Review of Sociology, 16 (1990), p. 27-65.

Popitz, Heinrich, Wege der Kreativität, Tübingen, Mohr Siebeck, 1997.

Saerens, Lieven, « Die Hilfe für Juden in Belgien », in Wolfgang Benz & Juliane Wetzel (éd.). Solidarität und Hilfe für Juden während der NS-Zeit. Regionalstudien 4. Slowakei, Bulgarien, Serbien, Kroatien mit Bosnien und Herzegowina, Belgien Italien, Berlin, Metropolverlag, 2004, p. 193-280

Sémelin, Jacques, Persécutions et entraides dans la France occupée. Comment 75% des Juifs de France ont échappé à la mort. , Les Arènes, 2013.

Shotland, R. Lance, « When Bystanders Just Stand By », in Psychology Today, juin (1985), p. 50-55.

Rossolinski-Liebe, Grzegorz, « Rezension : Engelking, Barbara: Jest taki piekny słoneczny dzien´ . . . . Losy Z˙ ydów szukaja˛cych ratunku na wsi polskiej 1942-1945 ». Varsovie 2011, sur H-Soz-u-Kult 18.04.2012.

Six, Bernd, « Neuere Entwicklungen und Trends in der Einstellungs-Verhaltens-Forschung », in Erich H. Witte (éd.), Einstellung und Verhalten. Beiträge des 7. Hamburger Symposions zur Methodologie der Sozialpsychologie. Brunswick, Technische Universität, 1992, p. 13-33.

Varese, Frederico et Yaish, Meir, « Resolute Heroes: The Rescue of Jews During the Nazi Occupation of Europe », in European Journal of Sociology 46 (2005), p. 153-168.

Varese, Frederico et Yaish, Meir, « The Importance of Being Asked: The Rescue of Jews in Nazi Europe », in Rationality and Society 12 (2000), p. 307-334.

Wahl, Klaus, Tramitz, Christiane et Blumenthal, Jörg, Fremdenfeindlichkeit. Auf den Spuren extremer Emotionen. Opladen, Leske + Budrich, 2001.

Welzer, Harald, « Krieg », in Christian Gudehus & Michaela Christ (éd.), Gewalt. Ein interdisziplinäres Handbuch. Stuttgart, Metzler, 2013.

Welzer, Harald, Täter: Wie aus ganz normalen Menschen Massenmörder werden. Francfort, S. Fischer, 2005.

Wicker, Allan W., « Attitudes vs. Actions: The Relationship of Verbal and Overt Behavioural Responses to Attitude Object », in Journal of Social Issue 4 (1969), p. 41-78.

Wiswede, Günter, Sozialpsychologie-Lexikon. Munich, Vienne, Oldenbourg Wissenschaftverlag, 2004.

• 1. Essential parts of these considerations were developed in exchange with Susanne Beer and Martin Düring in the context of a research project generously promoted by the Volkswagen Stiftung. Accordingly, I would like to thank these people and institutions as well as Elissa Mailänder (Sciences Po, ) and the Franco-German University that organized and financed a workshop on the issue which eventually resulted in this publication. Parts of the article are based on Gudehus, Christian. Helfen. In: Gudehus, Christian & Christ, Michaela (Eds.). Gewalt. Ein interdisziplinäres Handbuch. J.B.Metzler: Stuttgart 2013.
• 2. The title of the SS-Helferinnenkorps (Corps of female helpers of the SS) already combines the seemingly contradictory attributes "member of the SS" and "helpers" (Mühlenberg 2012).
• 3. The ethical questions related to this permeate all cases of help based on collaboration. The conflict situation can be impressively illustrated by the case of Rezso Kasztner who managed to ransom a large group of Jews and later on was charged for this in the context of conflicts in domestic politics and who eventually was killed by an assassin in Israel (Löb 2010).