The legacy of Nazism: Historical analogies and support for the far right

Abstract

This paper examined whether perceiving a historical analogy between former Nazi collaborator parties and contemporary far-right parties is negatively associated with support for the current far right. It was predicted that this relationship would be mediated by perceived immorality of far-right anti-Muslim standpoints. A survey study (Study 1) demonstrated that perceiving this historical analogy is negatively related to both electoral support for the far right and opposition to Muslim expressive rights. An experiment (Study 2) showed that drawing this historical analogy reduces both electoral support for the far right and opposition to Muslim expressive rights through perceived immorality of far-right anti-Muslim standpoints. The implications of historical analogies for people's attitudes toward present-day social and political issues are discussed.

Keywords::

• Historical analogy
• Nazism
• Morality perceptions
• Support far right
• Anti-Muslim attitudes

European far-right parties have become increasingly popular over the last decades (Lubbers, Gijsberts, & Scheepers, 2002). Ethnic nationalism is their central message (Rydgren, 2005, 2007) and some far-right parties have been publicly accused of racism and the spread of xenophobia (see Erk, 2005; Wood & Finlay, 2008). Recently, prominent politicians have pointed at similarities between the ideology of the far right and that of Nazi collaborator parties during World War II (WWII). For instance, in 2010, a Dutch minister stated about the views of the Dutch far-right party [Party for Freedom (PVV)]:

It seems as if we have forgotten what we promised after the war: that this may never happen again. And I can assure you that if you make the effort to read the book about the surge of the NSB1 … that you will come across shocking similarities [italics added]. (Baas, 2010)

By drawing an analogy between the Dutch Nazi collaborator party National Socialist Movement (Nationaal Socialistische Beweging, NSB) and the present far-right party PVV, the minister mobilizes the past to demoralize far-right parties. Far-right party politicians and supporters consider this historical analogy false and offensive, and see it as an unfair strategy to demonize their party.

The public debate about this historical analogy offers an interesting research opportunity to investigate how historical analogies shape support for current political parties. Although various studies in political science have examined how political elites make use of historical analogies in the realm of foreign policy making (for an overview see Mintz & DeRouen, 2010, pp. 103–114), the importance of historical analogies in informing lay people's political attitudes has received less attention (Bougher, 2012). In this paper, we examine whether a historical analogy to political parties of the past affects people's support for political parties in the present. More specifically, we investigate whether drawing a historical analogy between past Nazi parties' ideology and current far-right ideology is negatively associated with support for the far right.

In addition, we propose that this relationship can be explained by perceived (im)morality of far-right parties' anti-immigrant standpoints. Research on the moral foundations of politics has proposed that people's position toward controversial ideological issues, such as anti-immigration policies, may be grounded in different moral perceptions of these issues (e.g., Graham, Haidt, & Nosek, 2009). This line of research has mainly focused on liberals and conservatives' different moral perceptions of socio-political issues (for an overview see Haidt & Kesebir, 2010). Not much attention has been devoted to how perceptions of (im)morality can be changed in order to influence people's support for political parties. This research adds to the existing literature by studying how people's moral perceptions of political viewpoints (i.e., current far-right anti-immigrant viewpoints) can be altered by drawing a historical analogy to a political ideology that is consensually seen as immoral (i.e., Nazism), and by studying how such altered moral perceptions shape support for present political parties.

Historical analogical transfer

Nations typically define themselves by making reference to important events in national history. Collective memories of such events are transmitted through interpersonal (e.g., living grandparents) and institutional communications (e.g., education) (Paez & Liu, 2009). Although there is generally broad consensus about which events are important in national history (see Liu et al., 2005), their relevance for the present is often contested (Reicher & Hopkins, 2001). For instance, even though WWII and the Holocaust are widely considered to be important events in world history (Liu et al., 2005), their relation to the current socio-political context is disputed. It is likely that important events in national history only inform current socio-political dynamics when people feel that these contexts are to some extent comparable (Liu & Hilton, 2005). When two situations are perceived to be comparable, a process of analogical transfer occurs, which involves using knowledge about a relatively well-understood situation to make inferences about a more ambiguous one (e.g., Gick & Holyoak, 1983). Specifically, in cases of historical analogical transfer, knowledge of past events is used to make sense of current situations.

Previous empirical work on historical analogy has mainly been conducted in the USA, and has predominantly focused on how analogies to past foreign wars influence people's support for interventions in a new foreign war (Gilovich, 1981; Schuman & Rieger, 1992; Spellman & Holyoak, 1992). For instance, one study focused on how the analogies of WWII and the Vietnam War to the Gulf War predict support for military action in the Gulf war (Schuman & Rieger, 1992). Americans who considered the WWII analogy most adequate were more supportive of military action than Americans who found the Vietnam analogy most adequate.

To our knowledge, no studies have examined whether historical analogies to past political ideologies can affect concrete support for current political parties and their standpoints. This is a relevant question in relation to the growing success of far-right parties in various Western European countries and the ongoing debate about whether these parties bear ideological similarities to Nazi collaborator parties during WWII. We investigate whether drawing a historical analogy to Nazism affects support for far-right parties and their anti-immigrant standpoints, and whether a historical analogical transfer of perceived immorality explains this relationship.

Perceived immorality

In the Western world, Nazism and the Holocaust have become the dominant symbolic representations of evil (Alexander, 2002). As such, most people have developed a stable negative moral judgment of Nazi ideology: “When we hear the word Nazi we have a strong and immediate negative response” (Sears, 2001, p. 22). Moral judgment of others is an essential feature of social life (Gray, Young, & Waitz, 2012). People are particularly sensitive to moral violations committed by others against others, such as the atrocities committed to Jews by the Nazis (Rozin, Lowery, & Ebert, 1994). Research has demonstrated that when thinking about the Holocaust, people show facial expressions of disgust, which are typical reactions to moral violations (Rozin et al., 1994). Research has also shown that groups that are perceived as immoral are evaluated negatively (Phalet & Poppe, 1997). Furthermore, it has been suggested that when a group is perceived as immoral, its beliefs and behaviors are rejected (Crandall & Beasley, 2001).

On the basis of these findings, we reason that when an analogy is drawn between Nazism and the anti-immigrant ideology of current far-right parties, a process of historical analogical transfer will occur, by which people transfer their negative moral judgment of past Nazi parties to current far-right parties. This negative moral judgment may, in turn, reduce support for the far right.

Socio-political and historical context

This research was conducted in Flanders (Belgium) and the Netherlands. In both countries, there was a nationalist party that actively collaborated with the Nazis during WWII. The Flemish National Union (Vlaams Nationaal Verbond, VNV) and the Dutch NSB partially adopted the ideology of the Nazi party and supported the persecution and deportation of Jews (De Wever, 1994; Te Slaa & Klijn, 2010). In addition, both Flanders and the Netherlands currently have a far-right party that is well known for its anti-immigrant standpoints. The Flemish far-right party, Flemish interest [Vlaams Belang (VB)], won 12% of the total votes during the last federal election in 2010; the Dutch far-right party, PVV, gained 10% of the votes during the last national elections in 2012.

Within the literature on current far-right ideology, similarities as well as differences with Nazism have been identified. Both ideologies can be considered to share an emphasis on nativism (i.e., ethnic nationalism) (e.g., Swyngedouw, 1995; Zuquete, 2008), which combines elements of xenophobia and nationalism, and is defined as “an ideology, which holds that states should be inhabited exclusively by members of the native group and that nonnative elements are fundamentally threatening to the homogeneous nation-state” (Mudde, 2007, p. 19). Although the nativism of current far-right parties is less aggressive and expansive than that of former Nazi parties, it also emphasizes a longing for ethnic purity and homogeneity (Rydgren, 2007).

Current far-right parties, such as VB and PVV, have strong anti-immigration standpoints, primarily in relation to Muslim immigrants (Partij voor de Vrijheid, 2010; Vlaams Belang, 2009). Muslim immigrants are constructed as a threat to national welfare and as culturally deviant (Van Acker, 2012; Verkuyten, 2013). As a consequence, visible manifestations of Muslim culture and religion in public life (e.g., wearing of headscarves, building of mosques) are strongly opposed by these far-right parties (Betz & Meret, 2009). Several researchers contend that the ways in which Muslim immigrants are nowadays portrayed by far-right parties are similar to how Jews during the 1930s were portrayed by Nazi (collaborator) parties (Swyngedouw, 1995; Van de Valk, 2012). For instance, Swyngedouw (1995) argued that, similar to Jews in Belgium during the 1930s, Muslim immigrants are nowadays portrayed as causing economic harm and being culturally inferior.

Overview of the present research

We conducted two studies to examine the prediction that perceiving a historical analogy between past Nazi (collaborator) party ideology and current far-right ideology would be negatively related to voting tendencies for present-day far-right parties as well as support for their anti-Muslim standpoints. In order to assess support for far-right anti-Muslim standpoints, we measured the extent to which participants agree with far-right parties' opposition against Muslim rights to publicly confirm and express their identity (i.e., opposition to Muslim expressive rights), such as wearing a headscarf and building mosques.

Study 1 was a survey study conducted in two different subsamples. Study 1A was conducted in the Flemish part of Belgium and Study 1B was designed to replicate these findings in a Dutch sample. Study 2 used an experimental design, in which the historical analogy between Nazism and current far-right anti-immigrant ideology was induced through framing. In addition, Study 2 was designed to test the mediating role of perceived immorality of current far-right anti-Muslim standpoints. We tested whether drawing an analogy to Nazism affects support for the current far right through a process of historical analogical transfer, by which people transfer their negative moral judgment of past Nazi parties to current far-right parties.

Study 1

Method

Participants

Belgian (Study 1A) and Dutch (Study 1B) social science undergraduate students participated in this study during a collective research session in return for course credit. In Study 1A, the sample consisted of 97 participants (79.4% female), and there were no Muslim or Jewish participants. Ages ranged from 17 to 23 [M = 18.41, standard deviation (SD) = .97]. In Study 1B (N = 46), data from one Jewish participant were excluded from further analyses. This resulted in a sample of 45 participants (68.9% female). Ages ranged from 18 to 29 (M = 19.71, SD = 2.15). Participants in both samples were located toward the left end of the political spectrum. In Study 1A, 33% of the participants were most likely to vote for a center-left party in the next elections, 21.6% were most likely to vote for a center-right party, 14.4% were most likely to vote for a right-wing party, and 30.9% had no clear preference. In Study 1B, 40% of the participants were most likely to vote for a left party in the next elections, 24.4% for a center-left party, 13.3% for a center party, 8.9% for a center-right party, 2.2% for a right-wing party, and 11.1% had no clear preference.

Procedure

Participants first read a text of approximately 150 words listing some facts about WWII and the Nazi regime in order to introduce them to the topic of the survey. Participants then proceeded to the measures.

Measures

All items were rated on a scale ranging from 1 (strongly disagree) to 7 (strongly agree), unless otherwise indicated.

Perceived historical analogy. The extent to which participants perceived an analogy between the ideologies of former Nazi and present far-right parties was measured with the following three items: “The ideology of parties during WWII who collaborated with the Nazis is not comparable with the ideology of current anti-immigrant parties” (reverse-scored); “The standpoints of current anti-immigration parties concerning the protection of the own culture and identity bear resemblances with those of the Belgian/Dutch Nazi collaborator party during WWII,” and “The standpoints of current anti-immigration parties concerning the protection of the Belgian/Dutch economy and its employees bear resemblances with those of the Belgian Nazi collaborator party during WWII.” α for the three-item scale was .70 in Study 1A and .71 in Study 1B.

Opposition to Muslim expressive rights. Based on previous research in the Netherlands, four items were used to measure opposition to Muslim expressive rights (e.g., Smeekes, Verkuyten, & Poppe, 2011, 2012). Two sample items were “Muslims in Belgium/Netherlands should have the right to practice and express their religion in the public sphere” and “Muslims in Belgium/Netherlands should be allowed to wear the headscarf wherever they go.” All items were reverse scored and therefore a higher score on this scale indicates stronger opposition to Muslim expressive rights. α for the four-item scale was .82 in Study 1A and .73 in Study 1B.

Electoral support for the far right. Participants were asked how likely they were to vote for the Flemish/Dutch far-right party (i.e., VB in Belgium; PVV in the Netherlands) in the next national elections on a scale ranging from 1 (very small chance) to 7 (very high chance). In both samples, this question was embedded in a list of other existing political parties in Flanders and the Netherlands.

Results

To test our prediction that perceived historical analogy would be negatively related to electoral support for the far right and opposition to Muslim expressive rights, we specified the path model (using AMOS 18.0 software) illustrated in Figure 1. The model is saturated and hence the fit is perfect. Means, SDs, and bivariate correlations between all variables in the model are reported in Table 1.

Figure 1 Path model of Study 1, with perceived historical analogy as the independent variable and electoral support for the far right and opposition to Muslim expressive rights as dependent variables. Path-coefficients are standardized estimates, which are displayed in black for Study 1A and in gray for Study 1B. Correlations between dependent variables are standardized. R² = explained variance. *p < .05, **p < .01.

Table 1 Means, SDs and bivariate correlations for Study 1A (under the diagonal) and Study 1B (above the diagonal)

	1	2	3	M	SD
1. Perceived historical analogy	–	− .30*	− .39**	4.47	1.16
2. Electoral support far right	− .30**	–	.33*	1.71	1.34
3. Opposition to Muslim rights	− .21*	.34***	–	3.39	1.16
M	4.06	1.87	3.98
SD	.97	1.46	1.22

***p < .001, **p < .01, *p < .05.

sss

Figure 1 shows the standardized paths, explained variance (R²), and correlation between the dependent variables. As expected, in both samples, perceived historical analogy was negatively associated with electoral support for the far right and opposition to Muslim expressive rights (see Figure 1).

Study 2

The results of Study 1 provide support for our hypothesis that perceiving a historical analogy between the ideology of past Nazi collaborator and current far-right parties is negatively related to opposition to Muslim expressive rights and to the intention to vote for far-right parties. Yet, because of the correlational nature of the study, we cannot draw causal conclusions. It is possible that participants who are less supportive of current far-right ideology are more inclined to perceive a historical analogy, rather than the other way around. In order to strengthen our argument that historical analogies can alter people's support for far-right parties and their anti-Muslim standpoints, Study 2 used an experimental design in which the historical analogy between past Nazi (collaborator) party ideology and present far-right ideology was manipulated.

In addition, while Study 1 showed that there is a negative association between perceived historical analogy and support for current far-right parties, it does not shed light on the underlying process. In Study 2 we examine the idea that drawing a historical analogy to Nazi (collaborator) party ideology reduces support for the far right as well as opposition against Muslim expressive rights, because it enhances perceptions of immorality of current far-right parties' anti-Muslim standpoints. Moreover, in order to strengthen the external validity of the findings, Study 2 was conducted among a sample that was more equal in terms of gender composition and that was more right-wing oriented.

Method

Participants, design, and procedure

Participants were 100 Belgian students from a bachelor's program in secondary school teacher training in Flemish Belgium. We excluded Muslim and Jewish participants from further analyses. The remaining sample included 92 participants (51.1% female). All questionnaires were completed during regular class hours and in the presence of a teacher and a researcher. Ages ranged between 17 and 30 (M = 19.59, SD = 3.57). One participant was 45 years old. The majority of the participants were located toward the right end of the political spectrum. Specifically, 26.1% of the participants indicated that they were most likely to vote for a center-left party in the next elections, 7.6% indicated that they were most likely to vote for a center-right party, 46.7% indicated that they were most likely to vote for a right-wing party, and 19.6% showed no clear preference.

Historical analogy manipulation

Participants were randomly assigned to one of two framing conditions (historical analogy vs. control). As the representation of current far-right parties' anti-immigrant ideology as similar to Nazi ideology is highly contested and has strong political connotations, we developed a historical analogy manipulation that did not explicitly state this analogy. Instead, we choose to highlight two viewpoints of past Nazi collaborator parties that are well known as “owned” by current far-right parties. According to the literature on analogical reasoning, people will see an analogy when they find a mapping, or set of appropriate correspondences between the elements of the source (Nazi collaborator parties' ideology) and the target (current far-right ideology) (e.g., Spellman & Holyoak, 1992). Thus, in the historical analogy condition, we offered participants a mapping by only presenting them with ideological elements of Nazism that are also present in current far-right anti-immigrant discourse (i.e., cultural threat, economic threat) (Swyngedouw, 1995; Van de Valk, 2012),2 without explicitly telling them that these elements correspond. In the control condition, participants read a factual text about the Nazi regime and the Belgian collaborator party (VNV), but no information was given about either Nazi or VNV's ideology (see Appendix for exact wordings).

Measures

The measures were presented to participants after the frames in the order that follows. Responses to the items were recorded on 7-point rating scales ranging from 1 (strongly disagree) to 7 (strongly agree), unless otherwise indicated.

Manipulation check. To check whether the manipulation worked as intended, we used the same perceived historical analogy items as in Study 1 plus one additional item (α = .76). This additional item was “The ideas of the Belgian Nazi collaborator party during WWII about Jewish people are not comparable to the ideas of contemporary anti-immigration parties about Muslims” (reverse-scored).

Perceived immorality. The extent to which participants perceived current far-right anti-Muslim standpoints as immoral was assessed with the following three items (α = .88): “The attitude of current anti-immigrant parties towards Muslims is disgusting,” “The viewpoints of current anti-immigrant parties about Muslims are morally objectionable,” and “The standpoints of current anti-immigrant parties about Muslims are discriminating.”

Opposition to Muslim expressive rights. Opposition to Muslim expressive rights was measured with four items (α = .65) that were identical to Study 1, except that one item (about Islamic holidays) was replaced by the item: “Muslims in Belgium should have the right to found Islamic schools.” Items were again recoded, so that a higher score indicates more opposition.

Electoral support for the far right. Identical to Study 1A, participants were asked to indicate how likely they were to vote for the far-right party in Belgium (Flemish Interest), on a scale ranging from 1 (very small chance) to 7 (very high chance), and this question was again embedded in a list of other political parties in Belgium.

Results

Manipulation check

One-way ANOVA showed that there was a main effect of the analogy manipulation on perceived historical analogy, F (1, 90) = 4.52, MSE = 1.02, p = .036. Thus, even though our manipulation did not explicitly state the historical analogy, participants in the historical analogy condition did perceive more similarities between past Nazi collaborator party ideology and current far-right anti-immigrant ideology (M = 4.25, SD = .92) than participants in the control condition (M = 3.79, SD = 1.14).

Path analyses

We hypothesized that the historical analogy manipulation would have a negative effect on electoral support for the far right and opposition against Muslim expressive rights, and that these effects could be explained by enhanced perceptions of immorality of far-right anti-Muslim standpoints. To test these predictions, we specified the path models (using AMOS 18.0 software) illustrated in Figure 2. One path model includes perceived immorality as a mediator, the other does not. Both models are saturated and hence their fit is perfect.

As can be seen in Figure 2, in the model that did not include perceived immorality, the historical analogy manipulation had a significant negative effect on opposition to Muslim rights, and a marginally significant (p = .059) negative effect on electoral support for the far right (see Table 2 for mean differences between the experimental conditions for all continuous variables).3 When perceived immorality was included in the path model as a mediator, the direct effects of the manipulation on both dependent variables were no longer significant. Importantly, in this model, the historical analogy manipulation had a significant effect on perceived immorality. Perceived immorality, in turn, had a significant effect on both dependent variables, suggesting mediation. In addition, a trimmed mediation model that did not include the direct paths from the historical analogy manipulation to the dependent variables was not reliably different from the saturated model (Δχ²(2, N = 90) = 1.29, p = .525) and had a very good model fit, χ² (2) = 1.29, p = .526; comparative fit index (CFI) = 1.00, goodness of fit index (GFI) = .99, and root mean square error of approximation (RMSEA) = .00.4

Figure 2 Path models of Study 2, with historical analogy manipulation as the independent variable and electoral support for the far right and opposition to Muslim expressive rights as dependent variables, and perceived immorality as the mediator. Path-coefficients are standardized estimates, and path-coefficients in brackets reflect direct effects. Correlations between dependent variables are standardized. R² = explained variance. †p < .06, *p < .05, **p < .01, ***p < .001.

Table 2 Means and SDs by historical analogy and control condition, Study 2

	Historical analogy condition	Control condition
1. Perceived immorality	4.46(1.06)^a	3.92(1.39)^b
2. Electoral support far right	2.31(1.66)^a	3.50(2.10)^b
3. Opposition to Muslim expressive rights	4.15(1.02)^a	4.59(1.20)^b

Different superscripts in each row indicate significant differences at p < .05 (one-sided).

Analysis of indirect effects

To test more directly for mediation, we calculated the indirect effects of the historical analogy manipulation on support for the far right and opposition to Muslim expressive rights, via perceived immorality, using bootstrapping procedures. This involved generating 5000 random bootstrap samples with replacement from the data-set (N = 92) and testing the model with these samples. The analysis revealed a significant indirect effect of the historical analogy manipulation on (a) electoral support for the far right, with a point estimate of − .28 and a 95% bias-corrected bootstrap confidence interval (BBCI) of − .556 to − .013, and on (b) opposition to Muslim rights, with a point estimate of − .10 and a 95% BBCI of − .230 to − .010. These results support our prediction that exposing people to a historical analogy between past Nazi and present far-right ideology reduces electoral support for the current far right as well as opposition to Muslim expressive rights via perceived immorality of current far-right anti-Muslim standpoints^**

General discussion

This research investigated whether and how historical analogies affect people's support for political parties and their standpoints. We focused on a historical analogy that is fiercely debated in various Western European countries, namely, between the ideologies of former Nazi (collaborator) and current far-right (anti-immigrant) parties. We predicted that perceiving a historical analogy to Nazism would be negatively related to electoral support for the far right and their anti-Muslim standpoints. This prediction was confirmed in two studies, using samples from different countries, and using survey and experimental designs. Furthermore, based on previous theorizing on analogical reasoning, we proposed that historical analogies are effective because they transfer knowledge from past situations to make inferences about present situations, a process which we labeled historical analogical transfer. The results of Study 2 provide empirical support for this process by showing that the perceived immorality of Nazi ideology transferred to current far-right ideology when these ideologies were linked to each other through a historical analogy. These findings extend previous research which has shown that historical analogies can influence people's support for political actions (Gilovich, 1981; Schuman & Rieger, 1992; Spellman & Holyoak, 1992), by explicitly studying the transfer of moral reactions and how this relates to support for political parties and their standpoints. Our findings also extend research on the moral foundations of politics which has demonstrated that moral perceptions of socio-political issues depend on individuals' political preference (e.g., Graham et al., 2009), by showing that perceptions of immorality can be altered by drawing historical analogies.

Limitations

There are several limitations of this study that can be addressed in future research. First, one might question whether our findings are specific to the Dutch and Flemish (Belgian) context or rather have broader relevance. Far-right parties are popular in many European countries, and similar discourses on the historical analogy with Nazism have been observed, for example, in France, Great Britain, and Germany (see Rees, Allpress, & Brown, 2013; Wood & Finlay, 2008). Future studies could examine whether this historical analogy affects support for the far right in these countries as well.

Furthermore, the direction of causation warrants discussion. Even though this research provides causal evidence that drawing a historical analogy to Nazism reduced electoral support for far-right parties and their anti-immigrant standpoints, the opposite direction of influence might also hold. It may well be that people who support the far right are less likely to perceive a historical analogy. This would imply that those who support the far right are less affected by an analogy frame as used in Study 2. Speaking against this argument, the sample that we used in Study 2 was rather right-wing oriented and yet the framing sorted a clear effect. However, the proportion of participants that actually supported the far-right party in this study was relatively small. Prospective work may try to replicate the present findings among a sample of the far-right electorate.

Lessons from the past: The use of historical analogies

In this research, we focused on one specific historical analogy (i.e., between Nazi collaborator parties' ideology and current far-right parties' ideology) and one type of reaction to this analogy (i.e., moral judgment). However, the process of historical analogical transfer may occur for a variety ideas, situations, groups, and persons. For instance, historical analogies might play a role in understanding how characteristics of past politicians and public figures are applied to current politicians and other public figures. For example, journalists regularly drew a historical analogy between Martin Luther King and Barack Obama when reporting on Obama's presidential campaign, and thus ascribed characteristics of Martin Luther King to Barack Obama (Berkowitz & Raaii, 2010). Historical analogies might also play an important role in understanding how collective emotions, in response to historical wrongdoings of out-groups in the past, such as guilt or shame, inform attitudes toward out-groups in the present. Feelings of guilt and shame, in response to historical wrongdoings, may transfer to minority groups in the present when an analogy is drawn between the treatment of out-groups in the past and present (cf. Rees et al., 2013). In the context of this study, feelings of guilt that Belgian and Dutch people might feel for the mistreatment of Jews in the past may transfer to Muslims when an analogy is drawn between the stigmatization of Jews in the past and Muslims in the present.

Even though historical analogies can in principle be drawn between a wide variety of past and present ideas, situations, and persons, one may wonder about the limits in the extent to which people will apply or accept historical analogies. In order for people to apply or accept historical analogies, they need to be able to see a set of correspondences between a source analog in the past and a target analog in the present (cf. Spellman & Holyoak, 1992). This implies that historical analogies are bounded by reality constraints. In relation to this research, this means that it will be easier to draw an analogy between Nazi (collaborator) parties and contemporary far-right parties than between Nazi (collaborator) parties and contemporary left parties. In addition, historical analogies are specific to the source and target analog involved. Thus, even when one would (try to) draw an analogy between Nazi (collaborator) parties and contemporary left parties, one would need to highlight a different set of corresponding elements than when drawing an analogy between Nazi (collaborator) parties and contemporary far-right parties.

Furthermore, there are multiple ways in which historical analogies can be manipulated and made salient for people. Previous studies have used manipulations that explicitly mentioned historical analogies, and made participants write about the correspondences (e.g., Spellman & Holyoak, 1992, Experiment 1). Although this is the most straightforward way to manipulate historical analogies, we chose a more subtle manipulation as the analogy between past Nazi and present far-right parties' ideology is highly contested. Specifically, we manipulated the historical analogy between past Nazi and present far-right parties by drawing attention to a set of ideological viewpoints of Nazi collaborator parties that are well known as “owned” by current far-right parties without explicitly mentioning the analogy. Hence, we presented information solely about the past and yet people perceived similarities between the past and the present. This is an interesting finding, as it indicates that historical analogies can be triggered when people are merely presented with elements from the past. Yet, there are probably even more ways in which historical analogies can be manipulated. Future research could investigate and compare different ways of manipulating historical analogies.

In conclusion, this research shows that historical analogies have implications for how people position themselves toward present social developments. Many countries in Europe witness a substantial electoral success of far-right parties and relatively strong support for their anti-immigrant standpoints (Lubbers et al., 2002; Mudde, 2007). As a means to combat this development, politicians and intellectuals have drawn historical analogies to Nazism and the collaborator parties that adopted Nazi ideology. Nazism and the atrocities committed to Jewish people by Nazis during WWII are considered the ultimate evil, and they are part of a moral universalism that this “must never happen again” (Alexander, 2002). Our findings show that the legacy of Nazism indeed serves an important symbolic function in mobilizing opposition against far-right parties and their anti-immigrant standpoints.

Notes

^* These authors contributed equally.

¹ The NSB was a Dutch fascist and national socialist party that collaborated with the Nazis during WWII.

² We selected these viewpoints on the basis of Swyngedouw (1995) and after a thorough analysis of propaganda materials and speeches from VNV party officials and Flemish interest party officials. These materials are available from the authors upon request.

³ We also examined whether the historical analogy manipulation had an effect on electoral support for other existing political parties in Belgium, but this was not the case (ps>.341).

⁴ In order to further demonstrate that the observed effects of our subtle historical analogy manipulation were driven by a perceived historical analogy, we tested a double-mediation model that included perceived historical analogy (i.e., manipulation check) as an additional mediator between the historical analogy manipulation and perceived immorality. We tested an indirect-only model, which means that we did not include any direct effects of the perceived historical analogy manipulation on perceived immorality and on the dependent variables nor did we include direct effects of perceived historical analogy on the dependent variables. This indirect-only model had a good fit to the data, χ²(5) = 2.49, p = .778, CFI = 1.00, GFI = .99, and RMSEA = .00. Furthermore, the model fit could not be improved by adding direct paths from the historical analogy manipulation to the dependent variables (Δχ2(2, N = 90) = 1.29, p = .525) or to perceived immorality (Δχ2(1, N = 90) = 1.11, p = .292) nor by adding direct paths from perceived historical analogy to the dependent variables (Δχ2(2, N = 90) = 0.10, p = .951). Examining the path coefficients showed that the historical analogy manipulation had a significant positive effect on perceived historical analogy (β = .22, p = .016), and perceived historical analogy subsequently predicted perceived immorality (β = .46, p < .001). Perceived immorality, in turn, predicted opposition against Muslim rights and electoral support for the far right. Bootstrapping revealed that the historical analogy manipulation had a significant indirect effect on perceived immorality via perceived historical analogy, with a point estimate of .13 and a BBCI of .017–.305. In addition, the total indirect effects of the manipulation on opposition (with a point estimate of − .10 and a BBCI of − .244 to − .007) and electoral support (with a point estimate of − .28 and a BBCI of − .559 to − .009) were also significant. These results indicate that the effect of the historical analogy manipulation on perceived immorality is mediated by perceived historical analogy, and that perceived immorality subsequently decreased support for far-right parties.

Appendix

Historical analogy condition

During WWII Germany was ruled by the Nazi regime, with Hitler as dictator. This regime sought to establish a German world empire. Their slogan was: “One people, one empire, one leader.”

One of the most important characteristics of Nazi ideology was their battle for the protection of their people and fatherland. The Nazi regime believed in the creation of a homogeneous nation state that would consist entirely of people with the same cultural and religious background. Other cultures and religions were considered as a threat for the identity and unity of the native people.

In Flanders there was a party that openly collaborated with the Nazis and that adhered to their ideology: The Flemish National Union (VNV). This party had one prominent leader Staf De Clerck. During WWII, many Flemings were member of the VNV. In addition, a considerable proportion of the Flemish population supported the party. The VNV and its members helped the Nazis with tracing and deporting Belgian Jews to the concentration camps.

The most important viewpoints of the VNV were:

• Protection of the Flemish economy and Flemish working class against fortune seekers and profiteers from abroad and from other parts of the country. Flemish workers should have priority on the labor market.

• Striving for the maintenance of the original Flemish culture, with Christian heritage and fighting against the interference of other cultures. Foreign cultures were considered as incompatible with the Flemish culture and foreigners could not be real Flemings.

Control condition

This year 66 years have passed since the end of WWII. The people who have witnessed this war are now very old.

WWII was a merger of several military conflicts that were fought all around the world between 1939 and 1945. During WWII, Germany was ruled by the Nazi regime, with Hitler as dictator. This regime sought to establish a German world empire. Their slogan was: “One people, one empire, one leader.” The start of WWII was on 3 September 1939 when the UK and France declared war on Germany.

In Flanders there was a party that openly collaborated with the Nazis and that adhered to their ideology: the Flemish National Alliance (VNV). This party had one prominent leader Staf De Clerck. The VNV and its members helped the Nazis with tracing and deporting Belgian Jews to the concentration camps.

It is remarkable that Belgium does not have a central museum in which the events of WWII are told in a broader context. This is remarkable since WWII is the biggest conflict in world history.